Bis(tridentate) Polypyridine Ruthenium(II) Complexes with Push-Pull Character Synthesis, Understanding and Application Dissertation Zur Erlangung des Grades „Doktor der Naturwissenschaften“ Im Promotionsfach Chemie am Fachbereich Chemie, Pharmazie und Geowissenschaften der Johannes Gutenberg-Universität Mainz Christoph Kreitner geboren in Wiesbaden Mainz, 2016 Die vorliegende Arbeit wurde in der Zeit von Januar 2013 bis Juni 2016 am Institut für Anorganische Chemie und Analytische Chemie der Johannes Gutenberg-Universität Mainz unter Anleitung von angefertigt. Mainz, Juni 2016 Dekan: 1. Berichterstatter: 2. Berichterstatter: Tag der mündlichen Prüfung: 11. Juli 2016 Zusammenfassung Die genaue Kenntnis der elektronischen Eigenschaften einer Klasse von Übergangsmetallkomplexes ist für Chemiker und Materialwissenschaften von großem Interesse, da es das Einstellen der Eigenschaften ermöglicht und die Entwicklung passender Anwendungen erleichtert. In der hier vorgestellten Arbeit wird die Synthese und Charakterisierung neuer push-pull-substituierter bis(tridentater) Rutheniumkomplexe diskutiert. In ersten Abschnitt wird ein zweikerniger amid-verbrückter Bis(terpyridin)ruthenium-Komplex mit einer hohen elektronischen Symmetrie trotz der intrinsischen strukturellen Asymmetrie diskutiert. Im gemischt- valenten Zustand wird keine Metall-Metall-Wechselwirkung beobachtet. Das ungepaarte Elektron ist vollständig auf einem der beiden Rutheniumatome lokalisiert, weil die spintragenden Orbitale und die Grenzorbitale des die elektronische Kopplung vermittelnden Brückenliganden energetisch zu weit auseinanderliegen. Im photoangeregten Zustand jedoch ist der Brückenligand einfach reduziert. Dies ermöglicht eine elektronische Wechselwirkung zwischen den Metallzentren, sodass bei Raumtemperatur zwei Valenztautomere anhand ihrer dualen Emission gleichzeitig beobachtet werden können. Weiterhin wurden cyclometallierte Ru(N^N^N)(N^C^N)]+-Komplexe mit Substituenten entweder nur am cyclometallierenden oder an beiden Liganden synthetisiert und untersucht. Ihre Absorptionseigenschaften wurden durch eine Kombination von spektroskopischen und theoretischen Methoden studiert, die gezeigt haben, dass die niederenergetischen Absorptionsbanden aus Ru(N^N^N) und Ru(N^C^N)-Übergängen in ähnlichen Anteilen zusammengesetzt sind. Alle untersuchten Komplexe sind bei Raumtemperatur schwach emissiv ausgehend von einem 3MLCT-Zustand (Metal-zu-Ligand-Charge-Transfer), der allerdings durch einen wohlbekannten metallzentrierten und einen zuvor unerkannten Ligand-zu-Ligand-Charge- Transfer-Zustand effizient depopuliert wird. Dies wurde durch temperaturabhängige Messung der Quantenausbeute sowie unterstützende dichtefunktionaltheoretische Rechnungen weiter untermauert. Darüber hinaus wurden cyclometallierte Komplexe mit [Ru(N^N)2(N^C)]+- und [Ru(N^N^N)(N^N^C)]+- Koordinationsumgebung auf theoretischer Ebene untersucht und Gemeinsamkeiten und Unterschiede in den Mechanismen der Depopulation der angeregten Zustände herausgestellt. In Analogie zum zweikernigen Bis(terpyridin)ruthenium-Komplex wurde ein strukturell ähnlicher Komplex mit einem zweifach cyclometallierenden amidverknüpften Brückenliganden hergestellt und untersucht. Die veränderten Energien der Grenzorbitale des Brückenliganden ermöglichen hier eine elektronische Wechselwirkung zwischen den Metallzentren, sodass im gemischtvalenten Zustand eine intensive Intervalenz-Charge-Transfer-Bande im Nahinfrarot-Bereich des elektromagnetischen Spektrums beobachtet werden kann. Im photoangeregten Zustand wird darüber hinaus duale Emission beobachtet. Diese resultiert aus zwei elektronisch ungekoppelten 3MLCT-Zuständen, die auf den äußeren Terpyridinliganden lokalisiert sind. Die Distanz zwischen diesen beiden angeregten Zuständen ist zu groß, um die Einstellung eines thermischen Gleichgewichts via Energietransfer zu erlauben. Erst durch Abkühlen und Einfrieren der Lösung werden die strahlungslosen Zerfallsprozesse ausreichend verlangsamt, dass eine Equilibrierung stattfindet. Unter Berücksichtigung dieser Erkenntnisse wurden cyclometallierte Polypyridinruthenium-Komplexe mit Triarylamin-Substituenten entwickelt und ihre Eignung als Sensibilisatoren in Farbstoffsolarzellen untersucht. Diese Komplexe sind im gemischtvalenten Zustand valenzdelokalisiert, was nach der Ladungsinjektion einen mesomeren Ladungstransport weg von der Halbleiteroberläche ermöglicht. Allerdings resultiert aus der Delokalisation auch eine messbare Resonanzstabilisierung des gemischtvalenten Zustands. Dies führte zu einer erschwerten Regeneration des Farbstoffs in mehreren der gewählten Farbstoff/Elektrolyt-Kombinationen. Als Folge davon wird die Effizienz der Solarzellen mit dem Referenzfarbstoff N719 in Gegenwart von Iodid/Triiodid als Elektrolyt von keinem der entwickelten cyclometallierten Farbstoffe erreicht. Bei der Verwendung kationischer Polypyridincobalt-Elektrolyte hingegen zeigen N719 und die cyclometallierten Farbstoffe eine ähnliche Performance, aber die Effizienzen sind insgesamt deutlich geringer als mit Iodid/Triiodid. Abstract The profound understanding of the electronic properties of a class of transition metal complexes is of interest to chemists and material scientists as it allows the tuning of their properties and the development of suitable applications. In the work presented herein, the synthesis and characterization of novel push-pull substituted bis(tridentate) ruthenium complexes is presented. In the first section, a dinuclear amide-bridged bis(terpyridine) ruthenium complex with a high electronic symmetry despite the intrinsic structural asymmetry is studied. No metal-metal-interaction is detected in the mixed valent state with the odd electron being entirely localized at one of the two ruthenium centers. This is because the spin carrying orbitals and the mediating orbitals of the bridging ligand are energetically fairly separated. In the photo-excited state, on the other hand, the bridging ligand is formally reduced by one electron. This enables electronic coupling between the two metal centers, so that two valence tautomers are detected simultaneously at room temperature by their dual emission. Further, cyclometalated Ru(N^N^N)(N^C^N)]+ complexes with substituents on either the cyclometalating ligand or both ligands have been synthesized and investigated. Their absorption properties were studied using a combination of spectroscopic and theoretical methods showing that the low-energy absorption bands are composed of Ru(N^N^N) and Ru(N^C^N) transitions to a similar extent in all cases. All complexes are very weakly emissive at room temperature from a 3MLCT (metal-to-ligand charge transfer) state, that is efficiently depopulated via a well-known metal-centered excited state and a previously unrecognized ligand-to-ligand charge transfer state. This was evidenced by temperature-dependent quantum yield measurements and supplementary density functional theory calculations. Additionally, [Ru(N^N)2(N^C)]+ and [Ru(N^N^N)(N^N^C)]+ complexes were studied on a theoretical basis highlighting common features and differences in the excited state depopulation mechanics of the different classes of complexes. In analogy to the dinuclear bis(terpyridine) ruthenium complex, a structurally related complex with an amide-linked biscyclometalating bridging ligand was synthesized and studied. The altered bridge’s frontier orbitals result in electronic coupling between the metal centers in the mixed-valent state as evidenced from an intense intervalence charge transfer band in near infrared region of the electromagnetic spectrum. In the photo-excited state, dual emission is observed at room temperature from two electronically uncoupled 3MLCT states localized at peripheral terpyridine ligands. The distance between the emissive states is too large to allow for a thermally equilibrating energy transfer to occur. Only upon freezing the solution, the non-radiative decay processes are retarded sufficiently to allow for equilibration. Considering these findings, cyclometalated polypyridine ruthenium complexes bearing triarylamine substituents were devised for use as sensitizers in dye-sensitized solar cells. These complexes are substantially valence-delocalized in the mixed-valent state allowing for mesomeric charge delocalization away from the semiconductor surface after charge injection. However, this delocalization results in a measurable resonance stabilization of the mixed-valent state that hampers dye regeneration by the electrolyte in several of the employed dye/electrolyte combinations. As a consequence, the efficiency of solar cells employing benchmark sensitizer N719 are unmatched by the developed cyclometalated polypyridine ruthenium dyes when combined with iodide/triiodie as electrolyte. Using cationic polypyridine cobalt electrolytes, N719 and the cyclometalated dyes exhibit similar performances, but the overall efficiencies are lower than with iodide/triiodide. TABLE OF CONTENTS 1 Introduction 1 1.1 Redox and Photochemistry of Bis(terpyridine) Ruthenium(II) Amino Acids and Their Amide Conjugates – from Understanding to Applications 3 1.2 Excited State Decay Mechanisms in Polypyridine Ruthenium Complexes 29 1.2.1 Phosphorescence 29 1.2.2 Non-radiative Decay 30 1.2.3 Other Excited State Decay Channels 33 1.3 Mixed Valence and Optical Electron Transfer 36 1.4 Cyclometalation 39 1.5 Dye-Sensitized Solar Cell 42 2 Aim of the Work 47 3 Results and Discussion 49 3.1 Dual Emission and Excited-State Mixed-Valence in a Quasi-Symmetric Dinuclear Ru−Ru Complex 53 3.2 Understanding the Excited State Behavior of Cyclometalated Bis(tridentate)ruthenium(II) Complexes: A Combined Experimental and Theoretical Study 69 3.3 The Photochemistry of Mono- and Dinuclear Cyclometalated Bis(tridentate)ruthenium(II) Complexes: Dual Excited State Deactivation and Dual Emission 89 3.4 Strongly Coupled Cyclometalated Ruthenium Triarylamine Chromophores as Sensitizers for DSSCs 109 3.5 Excited State Decay of Cyclometalated Polypyridine Ruthenium Complexes: Insight from Theory and Experiment 125 3.6 [Cr(ddpd)2]3+: A Molecular, Water-Soluble, Highly NIR-Emissive Ruby Analogue 155 4 Summary and Outlook 161 5 References 165 6 Appendix 171 6.1 Supporting Information To 1.1: Redox and Photochemistry of Bis(terpyridine) Ruthenium(II) Amino Acids and Their Amide Conjugates – from Understanding to Applications 171 6.2 Supporting Information To 3.1: Dual Emission and Excited-State Mixed-Valence in a Quasi-Symmetric Dinuclear Ru−Ru Complex 175 6.3 Supporting Information to 3.2: Understanding the Excited State Behavior of Cyclometalated Bis(tridentate)ruthenium(II) Complexes: A Combined Experimental and Theoretical Study 184 6.4 Supporting Information to 3.3: The Photochemistry of Mono- and Dinuclear Cyclometalated Bis(tridentate)ruthenium(II) Complexes: Dual Excited State Deactivation and Dual Emission 209 6.5 Supporting Information to 3.4: Strongly Coupled Cyclometalated Ruthenium Triarylamine Chromophores as Sensitizers for DSSCs 240 6.6 Supporting Information to 3.5: Excited State Decay of Cyclometalated Polypyridine Ruthenium Complexes: Insight from Theory and Experiment 259 6.7 Supporting Information to 3.6: [Cr(ddpd) 3+2] : A Molecular, Water-Soluble, Highly NIR-Emissive Ruby Analogue 281 7 Acknowledgments 311 8 Curriculum Vitae 313 8.1 List of Publications 315 8.2 Conference Contributions 316 ABBREVIATIONS AND PHYSICAL QUANTITIES bpy 2,2'-bipyridine 𝜷𝑬𝒏𝑻 attenuation factor for energy transfer 𝒄 speed of light CS charge-separated dcbpy 4,4’-dicarboxy-2,2’-bipyridine ddpd N,N′-dimethyl-N,N′-dipyridine-2-ylpyridine-2,6-diamine 𝚫𝑬 energy difference 𝚫𝑬𝒂 activation energy 𝚫𝑮‡ Gibbs free energy of activation DFT density functional theory DSSC Dye-sensitized solar cell 𝑬 energy/potential 𝑬𝑴 Stokes shift 𝑬𝒐𝒙 oxidation potential 𝑬𝒓𝒆𝒅 reduction potential 𝑬𝟎𝟎 energy gap between ground and first excited state 𝒆 elementary charge 𝝐𝒎𝒂𝒙 exctintion coefficient ES excited state 𝑭 Faraday constant 𝝓 quantum yield 𝒇𝒇 fill factor FTO fluorine-doped tin oxide 𝚪 measure of electronic delocalization in mixed-valent systems 𝜼 solvent's refractive index 𝜼 quantum efficiency ℎ ℏ reduced Planck constant ℏ = 2𝜋 𝑯𝒂𝒃 electronic coupling matrix element ?̂?𝑺𝑶𝑪 spin-orbit coupling Hamiltonian 𝑰 current / current density 𝑰𝑷𝑪𝑬 incident-photon-to-current conversion efficiency IR infrared ISC intersystem crossing IVCT intervalence charge transfer 𝑱𝑫 Dexter overlap integral 𝑱𝑭 Förster resonance integral 𝒌 rate constant 𝜿 orientational factor 𝒌𝑩 Boltzmann constant 𝑲𝒄 comproportionation constant 𝜿𝒆𝒍 electronic transmission factor 𝒌𝑬𝑻 electron transfer rate 𝒌𝑬𝒏𝑻 energy transfer rate 𝒌𝒏𝒓 non-radiative decay constant 𝒌𝒒 quenching rate 𝒌𝒓 radiative decay constant 𝑲𝑺𝑽 Stern-Volmer constant 𝝀 reorganization energy 𝝀𝒔 solvent's reorganizational energy LL‘CT ligand-to-ligand charge transfer LUMO lowest unoccupied molecular orbital MC metal centered MLCT metal-to-ligand charge transfer NIR near infrared ?̃? frequency in wavenumbers 𝝂𝑵 average nuclear frequency factor ?̃?𝟏/𝟐 full width at half maximum Oh octahedral point group pbpyH 6-phenyl-2,2'-bipyridine 𝑷 power PES potential energy surface ppyH 2-phenylpyridine 𝑸 reaction coordinate [𝑸] concentration of quencher 𝒓 distance 𝑹 ideal gas constant S sensitizer 𝑺𝑴 Huang-Rhys factor 𝝉 lifetime t2g symmetry label tctpy 4,4’,4’’-tricarboxy-2,2’;6’,2’’-terpyridine tpy 2,2';6',2''-terypridine 𝑼 voltage UV ultra violett VIS visible 𝝌 nuclear wavefunction 𝝎 frequency | 1 1 INTRODUCTION Polypyridine complexes of ruthenium have fascinated inorganic chemists for the past 60 years. With the discovery of their luminescent properties in the late 1950s1 a lot of effort has been put into understanding the origin of the emission and the underlying mechanisms controlling its efficiency.2–8 Furthermore, a variety of other photophysical phenomena such as energy transfer,9 photoinduced electron transfer10–12 and mixed-valency13–15 have been studied using oligonuclear polypyridine ruthenium complexes.16 Soon thereafter, the potential of this class of complexes for interesting applications was discovered. The long excited state lifetimes and high excited state reduction potentials allowed for usage in the field of photocatalysis, particularly as light-harvesting sensitizer in water and carbon dioxide reduction.17–20 Additionally, the typically intense color and emission of polypyridine complexes allows for colorimetric and luminescent sensing applications, respectively.21 However, most of this research was essentially academic with little economic driving force. This changed in the early 1990s, with the ground-breaking discovery of the dye-sensitized solar cell (DSSC),22 when a materials science oriented community turned their attention to polypyridine ruthenium complexes. With an increasing need for renewable energy resources, the DSSC was considered a sustainable and cheap alternative to the conventional silicon-based solar cell.23–25 Ruthenium-based dyes are easily accessible from a synthetic point of view and their electronic properties are tunable in a wide range based on the ligand properties which made them a prime target for researchers all over the world.26,27 In the following, a still on-going hunt for more and more efficient dyes began, particularly focused around ruthenium, although other transition metals proved suitable in this context as well.28–31 In the periphery of this research, not only the light-harvesting properties of polypyridine ruthenium complexes were exploited, but also their luminescence, leading to the development of ruthenium-based light-emitting electrochemical cells as well.32,33 Among the studied systems, ruthenium complexes bearing tridentate ligands provide a particularly interesting class.11,34–36 Their meridional coordination sphere suppresses the occurrence of stereoisomers even in the presence of several differing functional groups or in oligonuclear systems. Additionally, the facile introduction of functional groups allows for an individual tailoring of the molecular frontier orbitals. This was taken advantage of by several research groups installing electron-donating and -accepting functionalities in the same complex on opposing ligands.11,37 Such push-pull substituted systems are nowadays well-understood and provide sophisticated insight into the molecular electronics. However, an electronic push-pull environment cannot only be introduced by peripheral substituents but also by altering the nature of the ligands themselves. A prominent example is the class of cyclometalated polypyridine ruthenium complexes, which, despite known since the late 1980s, has received more interest just in the past decade.38–40 This was mainly due to the discovery of their astonishingly good performance as sensitizers in DSSCs in 2007. 41,42 In a cyclometalated polypyridine complex, one of the nitrogen atoms of the metal’s coordination sphere is replaced 2 | 1 INTRODUCTION by an isoelectronic carbon anion. This leads to an interesting electronic situation: Since polypyridine ligands are good π-acceptors and cyclometalation typically yields a strong σ- and π- donating ligand, cyclometalated polypyridine complexes are inherently strong push-pull systems.43,44 This intrinsic electronic directionality has made them a high-priority research target for DSSC sensitizers.45,46 However, the luminescence of these systems is generally less intensively studied and thus, the understanding of their photophysical properties is not as elaborate as for conventional polypyridine ruthenium complexes. Hence, one of the major purposes of this dissertation is the elucidation of the electronic properties of cyclometalated polypyridine ruthenium complexes. In the following introduction, the current state of research on bis(tridentate) polypyridine ruthenium complexes with push-pull substitution will be summarized in section 1.1 in terms of a Microreview article written by Aaron Breivogel, Christoph Kreitner and Katja Heinze, which was published in the European Journal of Inorganic Chemistry in 2014. This article already covers to some extent the general photophysical properties of polypyridine ruthenium complexes governing their excited state deactivation. However, the excited state decay mechanisms in polypyridine ruthenium complexes will be further elaborated in section 1.2 as those are a key component of this dissertation. This will be followed by an introduction of the concept of mixed valence and optical electron transfer in section 1.3. Cyclometalation reactions in general and cyclometalated polypyridine complexes in particular will be discussed in section 1.4. The introduction will be rounded off by a description of the working principle of the dye-sensitized solar cell and an illustrative selection of current state-of-the-art sensitizers in section 1.5. Section 1.1 | 3 1.1 REDOX AND PHOTOCHEMISTRY OF BIS(TERPYRIDINE) RUTHENIUM(II) AMINO ACIDS AND THEIR AMIDE CONJUGATES – FROM UNDERSTANDING TO APPLICATIONS Aaron Breivogel, Christoph Kreitner and Katja Heinze Eur. J. Inorg. Chem. 2014, 2014, 5468–5490. Ruthenium(II) amino acid [Ru(4′-tpy-COOH)(4′-tpy- NH2)]2+ interacts with photons, electrons, and/or protons, to lead to phosphorescence, oxidative and reductive chemistry, acid/base chemistry, proton-coupled electron transfer, photoinduced reductive and oxidative electron transfer, excited- state proton transfer, and energy transfer. Applications include light-emitting electrochemical and dye-sensitized solar cells. Author contributions The ruthenium complexes were synthesized and characterized by Aaron Breivogel. The manuscript for this invited microreview was written by Aaron Breivogel (40 %), Christoph Kreitner (40 %) and Katja Heinze (20 %). Aaron Breivogel used parts of it for the introduction of his dissertation. Supporting Information for this article is found at pp. 171. For full Supporting Information, refer to http://onlinelibrary.wiley.com/store/10.1002/ejic.201402466/asset/supinfo/ejic_201402466_s m_miscellaneous_information.pdf?v=1&s=d36c160e970e957a0a90f0bedd7c19747c8b4daa. „Reprinted with permission from Breivogel, A.; Kreitner, C.; Heinze, K. Chem. Eur. J. 2014, 2014, 5468–5490. Copyright 2016 Jon Wiley and Sons.” 4 | 1 INTRODUCTION Section 1.1 | 5 MICROREVIEW DOI:10.1002/ejic.201402466 Redox and Photochemistry of Bis(terpyridine) ruthenium(II) Amino Acids and Their Amide Conjugates – from Understanding to Applications Aaron Breivogel,[a] Christoph Kreitner,[a,b] and Katja Heinze*[a] COVER PICTURE Keywords: Electron transfer / Energy transfer / Luminescence / Photochemistry / Redox chemistry / Ruthenium / Tridentate ligands The push-pull-substituted bis(terpyridine)ruthenium(II) pansion of the chelate ring. Furthermore, the chemically or- amino acid [Ru(4-tpy-COOH)(4-tpy-NH )]2+ ([5]2+2 ; tpy = thogonal functional groups enable the incorporation of this 2,2;6,2-terpyridine) with carboxylic acid and amino sub- metallo amino acid into peptide architectures in a highly se- stituents features exceptional chemical and photophysical lective manner, even by solid-phase peptide synthesis proto- properties. Its interaction with photons, electrons, and/or pro- cols. Amide-linked conjugates with other metal complexes tons results in room-temperature phosphorescence, revers- [(terpyridine)ruthenium(II), ferrocene, (bipyridine)rheni- ible oxidative and reductive redox chemistry, reversible acid/ um(I), (bipyridine)platinum(II)], organic chromophores, or base chemistry, proton-coupled electron transfer, photoin- ZnO nanoparticles underscore the versatile synthetic, redox, duced reductive and oxidative electron transfer, excited-state and photochemistry of this building block. First real-world proton transfer and energy transfer reactions. These proper- applications of [5]2+ and its derivatives include light-emitting ties can be fine-tuned by variations of the bis(terpyridine) electrochemical cells (LECs) and dye-sensitized solar cells amino acid motif, namely extension of the π system and ex- (DSSCs). 1. Introduction and Scope of the Review gand charge transfer; Figure 1a). From this state quantita- tive intersystem crossing (ISC) into the 3MLCT state oc- Polypyridine complexes of ruthenium(II) are a unique class of complexes with unprecedented photophysical, chemical, and electrochemical properties.[1] There is a pleth- ora of applications of polypyridine complexes of ruth- enium(II) such as dye-sensitized solar cells (DSSCs),[2–4] light-emitting devices,[5–8] anticancer and photodynamic therapy,[9–11] sensing of ions[12–14] and small neutral mo- lecules,[15–17] energy transfer,[18,19] electron transfer and mixed valency,[20–25] triplet–triplet annihilation upconver- sion,[26–29] and molecular data storage.[30–32] Catalytic appli- cations comprise important photocatalytic reactions such as water oxidation,[33,34] generation of H [35–39]2, reduction of CO ,[37–39]2 and photocatalysis of organic redox reac- tions.[38,40–44] Tris(2,2-bipyridine)ruthenium(II), [Ru- (bpy) ]2+3 , is one of the most prominent ruthenium(II) com- plexes. The absorption of a 450 nm photon by [Ru(bpy) 2+3] populates an excited 1MLCT state (MLCT = metal-to-li- [a] Institute of Inorganic Chemistry and Analytical Chemistry, Johannes Gutenberg University of Mainz, Duesbergweg 10-14, 55128 Mainz, Germany E-mail: katja.heinze@uni-mainz.de http://www.ak-heinze.chemie.uni-mainz.de/ 2+ [b] Graduate School “Materials Science in Mainz”, Figure 1. Qualitative state diagrams of (a) [Ru(bpy)3] and (b) 2+ 2+ Staudinger Weg 19, 55128 Mainz, Germany [Ru(tpy)2] [1] (MLCT = metal-to-ligand charge transfer, ISC = Supporting information for this article is available on the intersystem crossing, MC = metal-centered, GS = ground state, bpy WWW under http://dx.doi.org/10.1002/ejic.201402466. = 2,2-bipyridine, tpy = 2,2;6,2-terpyridine). Eur. J. Inorg. Chem. 2014, 5468–5490 5468 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 6 | 1 INTRODUCTION www.eurjic.org MICROREVIEW curs.[1] The long excited-state lifetime (τ ≈ 1 μs) of the chelate effect of tridentate ligands as compared with that of 3MLCT state at room temperature in solution renders bidentate ligands is favorable in terms of complex sta- [Ru(bpy)3]2+ exceptionally suitable as a photoredox catalyst bility.[53] However, despite the similar absorption character- (Table 1).[38,45,46] The 3MLCT state is emissive with a high istics and redox potentials of [Ru(bpy) ]2+3 and [1]2+, the phosphorescence quantum yield (Φ ≈ 10%), which favors excited state properties differ significantly (Table 1). Unfor- applications in light-emitting devices as luminescent sensors tunately, [1]2+ has a dramatically reduced lifetime of the or as imaging agents.[46] lowest excited 3MLCT state (τ ≈ 0.1–0.2 ns) and emission The properties of [Ru(bpy) ]2+3 can easily be tailored by quantum yield (Φ  0.0007%; Table 1). The underlying modifications of the bpy ligand. However, the intrinsic Δ, reason for the poor excited-state photophysical properties Λ chirality of [Ru(bpy)3]2+ is a serious drawback when more is an effective radiationless deactivation process via 3MC than one bpy ligand is substituted or when more than one states [from the electron configuration (t )52g (eg*)1], which [Ru(bpy) ]2+3 -type complexes are combined to form di- or are thermally populated from 3MLCT states (MC = metal- oligonuclear complexes, because diastereomeric complexes centered; Figure 1b).[54] (e.g. rac-Δ,Δ/Λ,Λ and meso-Δ,Λ) have to be separated or The three bpy ligands in [Ru(bpy) 2+3] create a coordina- avoided by complicated synthetic procedures.[47–49] It is ob- tion sphere that enables better metal–ligand orbital overlap vious that interaction with chiral molecules, such as DNA than that in [Ru(tpy) ]2+2 with two constrained tpy ligands. or proteins, will even modify the individual properties of Δ, As the overlap between the nitrogen lone pairs of pyridine Λ enantiomers, and any interaction with chiral biomolec- and the eg* orbitals of Ru is higher in [Ru(bpy) ]2+3 , the ules is complicated when using racemates, for example as ligand-field splitting is stronger in [Ru(bpy) ]2+3 , which in- anticancer drugs.[48d] duces less accessible 3MC states with a larger 3MLCT–3MC Bis(tridentate) meridional coordination as in [Ru(tpy)2]2+ energy gap compared with that of [Ru(tpy) ]2+2 (Fig- ([1]2+, Figure 2) avoids the formation of diastereomers even ure 1).[54] In order to improve ground-state and especially in the case of heteroleptic [Ru(tpy-R1)(tpy-R2)]2+[50] and di- excited-state photophysical properties of bis(terpyridine)- nuclear complexes (tpy-R1, tpy-R2 = 4-substituted ruthenium(II) complexes, extensive efforts have been made 2,2;6,2-terpyridine).[25,51,52] Furthermore, the stronger in the last two decades.[55] Long-lived and highly emissive Aaron Breivogel received his diploma in chemistry at the Johannes Gutenberg University of Mainz, Germany, in 2009, and he is currently finishing his Ph.D. Thesis in the research group of Prof. Dr. Katja Heinze in inorganic chemistry. During his studies he spent one semester (2007/2008) at the University of Valencia, Spain, in the Department of Analyti- cal Chemistry in the group of Prof. Dr. Miguel de la Guardia working on the quantitative determination of glycolic acid in cosmetics by online liquid chromatography and Fourier transform infrared spectroscopy. Currently he is working on the synthesis of bis(tridentate) complexes of ruthenium(II) and their applications in dye-sensitized solar cells and light- emitting electrochemical cells. He received a grant from the “International Research Training Group (IRTG) 1404 – Self-organized Materials for Optoelectronics” funded by the Deutsche Forschungsgemeinschaft (2011–2013) and served as IRTG student speaker for 18 months. During his Ph.D. Thesis he went for a six-month research stay to the group of Prof. Dr. Kookheon Char at the Seoul National University in Seoul, Republic of Korea. His research interests focus on the application of novel ruthenium complexes in dye-sensitized solar cells and light-emitting electrochemical cells. He has co-authored nine refereed papers. Christoph Kreitner received his diploma in chemistry at the Johannes Gutenberg University of Mainz, Germany, in 2012, and he is currently doing his Ph.D. Thesis in the research group of Prof. Dr. Katja Heinze at the Johannes Gutenberg University of Mainz, Germany. His research focuses on the design and synthesis of new cyclometalated ruthenium com- plexes and their applications in dye-sensitized solar cells and photocatalytic systems. During his studies he spent one semester at the University of Toronto, Toronto, Canada (2010/2011), working in the Department of Inorganic Chemistry in the group of Prof. Dr. Douglas W. Stephan on frustrated Lewis pairs and their reactions with lactones and lactide. For his Ph.D. Thesis he received a scholarship from the Graduate School of Excellence “Materials Science in Mainz” (2013). Recently he was elected a junior member of the Gutenberg Academy of the Johannes Gutenberg University of Mainz, Germany (2014). His research interests aim at the understanding of electron transfer and optical excitation mechanisms in molecular systems and the possibility of utilizing sunlight in chemical syntheses. Katja Heinze is professor of organometallic and bioinorganic chemistry at the Johannes Gutenberg University of Mainz, Germany. After receiving a diploma degree (1995) and a Ph.D. degree (1998) from the Ruprecht Karls University of Heidelberg, Germany (G. Huttner), she went for a postdoctoral stay to the University of Zurich, Zurich, Switzerland (1999). She was appointed Privatdozent in 2004 at the University of Heidelberg, Germany, and Full Professor in 2008 at the Johannes Gutenberg University of Mainz, Germany. She received the Lieseberg award of the Faculty of Chemistry and Earth Sciences, University of Heidelberg (2002), a Heisenberg fellowship from the Deutsche Forschungsgemeinschaft (2004), and a Hengstberger Award of the University of Heidelberg (2007). Currently she serves as vice spokesperson of the International Research Training Group 1404 – Self-organized Materials for Optoelectronics and as vice chair of the Institute of Inorganic and Analytical Chemistry, University of Mainz. Since 2011 she has been serving as a member of the International Advisory Board of Organometallics. Her key research interests comprise functional complex systems based on coordination and organometallic compounds with special emphasis on molecular wires, light-harvesting systems, bistable systems, switches, and sensors as well as on (biomimetic) catalysts. She has authored more than 100 international refereed papers. Eur. J. Inorg. Chem. 2014, 5468–5490 5469 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Section 1.1 | 7 www.eurjic.org MICROREVIEW Table 1. Photophysical and electrochemical properties of ruthenium(II) polypyridine complexes in CH3CN at 295 K. See Figures 2, 11, 12, 13, 15, 17, and 18 for compound numbering. Complex Absorption Emission Electrochemistry λmax [nm] λ Φ τ E (RuII/RuIII) E red [a]max 1/2 1/2 (ε [m–1 cm–1]) [nm] [%] [ns] [V][a] [V] [Ru(bpy) ]2+ [46,57]3 452 (13000) 615 9.4 1100 +0.89 –1.73 [1]2+ [50,54b] 474 (10400) 629  0.0007[b] 0.1–0.2[c] +0.92 –1.67 [2]2+ [8] 485 (18100) 667 0.041[d] 32 +0.96[e] –1.27[e] [3]2+ [58] 479 (19500) 637[f] 0.02[f] – +0.91 –1.69 [4]2+ [59,60] 502 (19100) 734 0.27[d] 34 +0.68 –1.54 [5]2+ [59,60] 501 (20700) 739 0.18[d] 26 (92 %), 4 (8 %) +0.66 –1.60 [6]2+ [61,62] 492 (22100) 677 0.059 22 +0.90 –1.46 [7]2+ [50] 490 (16800) 706 0.07[b] 50 +0.92 –1.53 [8]2+ [63] 490 (24000) 660 – – +0.87 –1.58 [9]2+ [64] 493 (29300)[g] 660 0.029[h] 1.1 +0.63 –1.43 [10]2+ [50] 487 (26200) 715 0.006[b] 1.0 +0.90 –1.66 [11]2+ [60] 502 (35600) 707 0.053[d] 23 – – [12]2+ [60] 501 (24700) 659 0.053[d] 21 (3 %), 3 (97 %) – – [13]2+ [60] 498 (26600) 664 0.030[d] 23 (96 %), 2 (4 %) – – [14a]2+ [65] 495 (31300) 713 0.13[b] 200 +0.95 –1.32 [14b]2+ [65] 506 (42000) 705 0.17[b] 231 +0.99 –1.29 [15a]2+ [66] 511 (44800) 698 0.76[d] 580 – – [15b]2+ [66] 500 (25000) 710 0.45[d,i] 2500 +0.94 –1.49 [16]2+ [67] 463 (10000) 643[j] 17.3[d,j] 385[j] +0.60 –1.95 [17]2+ [67] 473 (10000) 694[j] 2.6[d,j] 7900[j] +0.58 –1.88 [18]2+ [68,69] 491 (14000) 700[j] 3.2[j,h] 3000[j] +0.71 –1.73 [19]2+ [70] 553 (10000) 693[j] 11.2[j,h] 5500[j] +0.82 –1.52 [20]2+ [71] 522 (6425) 608 30[k] 3300 +1.11 –1.36 [21]2+ [56] 517 (7500) 729 0.45[d] 722[i] +0.81 –1.47 [22]2+ [56] 539 (6360) 744 1.1[d] 841[i] +0.92 –1.25 [23]2+ [56] 525 (8230) 762 0.042[d] 149[i] +0.64 –1.50 [24]2+ [56] 546 (7810) 788 0.052[d] 136[i] +0.73 –1.32 [25]2+ [72] 517 (8110) 732 0.068[d] – +0.81 –1.50[l] [26]2+ [72] 537 (6600) 743 0.059[d] – +0.87 –1.37[l] [27]2+ [72] 544 (6690) 771 0.067[d] – +0.71 –1.53[l] [33a]2+ [18] 490 (22290) 670 0.093[d] 19 (5 %), 4 (95 %) +0.90 –1.34 [33b]2+ [18] 490 (26180) 667 0.082[d] 18 (9 %), 3 (91 %) +0.90 –1.35 [33c]2+ [18] 490 (25820) 665 0.13[d] 19 (1 %), 3 (10 %), 0.5 (89 %) +0.92 –1.70 [33d]2+ [18] 491 (25830) 666 0.16[d] 22 (9 %), 4 (91 %) +0.91 –1.45 [33e]2+ [18] 490 (20590) 667 0.11[d] 19 (13 %), 4 (87 %) +0.87 –1.25 [34]2+ [18] 492 (25750) 668 0.14[d] 17 (7 %), 3 (93 %) +0.92 –1.16 [35]2+ [18] 490 (24950) 665 0.076[d] 15 (2%), 4 (40 %), 0.4 (57 %) +0.91 –1.23 [36]2+ 490 (17100) 668 0.12[d] – +0.89 –1.15 [37]2+ 490 (16500) 667 0.080[d] – +0.88 –1.14 [38]4+ [25] 522 (50620) 750 0.24[d] 22 (99 %), 2 (1 %) +0.68, +0.91 –1.48 [39]4+ [25] 496 (46550) 692 0.21[d] 22 +0.80, +0.90 –1.49 [40]4+ [62] 504 (63000) 684 0.032 24 (71 %), 44 (29 %) +0.91 (2e) –1.49 (2e) [41]4+ [73] 487 (50800) – – – +0.85 (2e) –1.75 [421]4+[74] 494 (ca. 60000) – – – +0.86 (2e) –1.57 [431]4+ [51] 499 (63000) – – – +0.87 (2e) –1.58 [44]2+ [59] 496 (26200) 704[m] 0.011 – +0.65[l,n] –1.50 [45]2+ [59] 502 (19300) 739[m]  0.01 – +0.90[l,o] –1.61 [46]2+ [59] 496 (25300) 704[m]  0.01 – +0.90[l,p] –1.58 [470]2+ [75a,75b] 478 (15000) – – – +1.01[l,q] –1.60 [480]2+ [75a,75b] 482 (15000) – – – +0.98[l,r] –1.62 [471]2+ [75c] 489 (28700) 699[v] 0.040[k,v] 260[v] +0.94[s,t] –1.55[s] [481]2+ [75c] 505 (28000) 697[v] 0.084[k,v] 260[v] +1.00[s,u] –1.52[s] [49]2+ [77] 493 (24400) 673 0.17 16 +0.92 –1.57 [50]2+ [78] 502 (23000) 735 0.20 29 +0.68 –1.51 [51]2+ [77] 492 (26000) 674 0.16 – +0.90 –1.48 [52]2+ [77] 500 (23500) 727 0.15 – +0.68 –1.53 [53]2+ [77] 496 (27800) 678 0.14 – +0.92 –1.53 [54]2+ [77] 494 (27700) 674 0.19 – +0.91 –1.48 [55]2+ [77] 502 (21800) 739 0.11 – +0.69 –1.47 [57]2+ [78] 494 (26600) 671 0.12 18 +0.93 –1.30, –1.47 [58]2+ [78] 504 (22900) 739 0.12 22 +0.69 –1.45, –1.69 [a] Vs. ferrocene/ferrocenium. [b] Recalculated from used value Φ = 0.028[79] for [Ru(bpy)3]Cl2 in H2O to the updated [Ru(bpy)3]Cl2 standard Φ = 0.040 in H O.[57]2 [c] Various solvents. [d] Recalculated from used value Φ = 0.062[79] for [Ru(bpy)3]Cl2 in CH3CN to the updated [Ru(bpy)3]Cl2 standard Φ = 0.094 in CH CN.[57]3 [e] No solvent given. [f] In DMF vs. 9,10-diphenylanthracene as reference. [g] In MeOH. [h] Recalculated from used value Φ = 0.059 for [Ru(bpy)3]Cl2 in CH3CN to the updated [Ru(bpy)3]Cl2 standard Φ = 0.094 in CH CN.[57]3 [i] In PrCN. [j] In EtOH/MeOH (4:1, v/v). [k] As given in the reference. [l] Irreversible. [m] In acetone. [n] E II III1/2(Fe /Fe ) = 0.24 V. [o] E (FeII/FeIII1/2 ) = –0.03 V. [p] E II III1/2(Fe /Fe ) = –0.03, 0.26 V. [q] E II III1/2(Fe /Fe ) = 0.18 V. [r] E II III1/2(Fe /Fe ) = 0.17 V (2 e). [s] In dichloromethane. [t] E1/2(FeII/FeIII) = 0.15 V. [u] E II III1/2(Fe /Fe ) = 0.24 V (2e). [v] In H2O/CH3CN (4:1, v/v). Eur. J. Inorg. Chem. 2014, 5468–5490 5470 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 8 | 1 INTRODUCTION www.eurjic.org MICROREVIEW and higher extinction coefficients compared with those of [Ru(bpy) 2+3] (λmax = 452 nm; εmax = 13000 m–1 cm–1) and [1]2+ (λmax = 474 nm; εmax = 10400 m–1 cm–1), which is bene- ficial for efficient light harvesting in DSSCs, especially in the low-energy part of the electromagnetic spectrum (Table 1).[8] In this overview we will discuss the chemical, photo- chemical, and redox properties of heteroleptic bis(terpyrid- ine)ruthenium(II) complexes with two orthogonal func- tional groups, especially with R1 = COOH and R2 = NH2 (ruthenium amino acid [5]2+, Figure 2),[59] and related com- plexes. These multifunctional complexes enable electron- transfer (RuII/III, tpy/tpy·–) and proton-transfer studies (tpy- COOH/tpy-COO–; tpy-NH +3 /tpy-NH2/tpy-NH–) as well as selective functionalization at the C- or at the N-terminus, or at both termini, to give amide-linked conjugates, nano- composites, and functionalized materials with potential ap- plications in photophysics, photochemistry, and materials science. 2. Physicochemical Properties of Bis(tridentate) Ruthenium(II) Polypyridine Complexes The physicochemical properties of bis(tridentate) ruth- enium(II) polypyridine complexes will be illustrated by ex- amples of donor–acceptor-substituted bis(terpyridine) com- plexes [4]2+ and [5]2+ (Figure 2) prepared in our group.[59] Heteroleptic complexes such as [4]2+ are readily synthesized by a stepwise approach.[50,80,81] RuCl3 is treated with the first tpy ligand (tpy-R1, e.g. tpy-COOEt) to form the ruth- enium(III) complex Ru(tpy-R1)Cl3, which is typically easily isolated by precipitation. In the second step, the chlorido ligands are replaced by the second tpy ligand (tpy-R2, e.g. tpy-NH2) under reducing conditions (e.g. N-ethylmorph- oline), often assisted by microwave irradiation. This two- step procedure results in the desired heteroleptic complex [RuII(tpy-R1)(tpy-R2)]2+ with meridional coordination of the two chelate ligands without formation of homoleptic complexes [Ru(tpy-R1) ]2+ and [Ru(tpy-R2) ]2+.[59]2 2 Typi- Figure 2. Bis(terpyridine) complexes of ruthenium(II) relevant for cally, these dications are isolated as hexafluorophosphate this review. salts. The ester group of [4]2+ is straightforwardly hy- drolyzed to the carboxylic acid [5]2+ by refluxing [4]2+ in excited states in bis(tridentate) complexes of ruthenium(II) 20% sulfuric acid. The harsh reaction conditions already can be obtained by the introduction of substituents in the underline the high thermal and chemical stability of bis(tri- 4-position of the tpy ligand.[8,50] Electron-withdrawing dentate) complexes such as [4]2+ and [5]2+.[59] The two func- substituents stabilize the 3MLCT state with respect to the tional groups of amino acid [5]2+ offer the possibility of 3MC state, while electron-donating substituents destabilize orthogonal functionalization either at the C-terminus or at the 3MC state. Both effects increase the 3MLCT–3MC en- the N-terminus. ergy gap and diminish radiationless deactivation via the Ruthenium amino acid [5]2+ itself displays unique acid– 3MC state.[50,56] In 4-substituted [Ru(tpy)(tpy-COOEt)]2+ base properties. The amino function of [5]2+ is a poor nu- ([2]2+, Figure 2) prepared by Bolink, the electron-with- cleophile and a weak base {[5 + H]3+: pKa  0 (estimated drawing character of the ester substituent leads to a less by UV/Vis spectroscopy in concentrated sulfuric acid)[59] accessible 3MC state, a significantly longer lifetime of the and pKa ≈ 1.6 (estimated electrochemically in nitric acid); 3MLCT state, and a higher emission quantum yield (τ = Ph-NH +3 : pKa = 4.9[82]}, because the lone pair is substan- 32 ns, Φ = 0.04%) compared with that of unsubstituted tially delocalized into the aromatic ring and towards the [1]2+ (Table 1).[8] Furthermore, [2]2+ has a redshifted ab- electron-withdrawing metal cation to give an essentially sorption maximum (λ = 485 nm; ε = 18100 m–1 cm–1) planar C–NH unit.[59,61]max max 2 On the other hand, the acidity of Eur. J. Inorg. Chem. 2014, 5468–5490 5471 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Section 1.1 | 9 www.eurjic.org MICROREVIEW the COOH group (pK = 2.7[59]a ) is increased in comparison to those of organic aromatic acids (benzoic acid: pKa = 4.2,[83] isonicotinic acid: pKa = 4.9[84]) because of the elec- tron-withdrawing effect of the pyridine backbone and the coordinated positively charged metal ion. In contrast to ali- phatic amino acids, such as natural α-amino acids, no zwit- terionic character is observed for [5]2+, which again is based on the low basicity of the amino group.[59] Species [5]2+ is dominant in aqueous solutions at pH 2 and can be proton- ated at the amino function to afford [5 + H]3+ at lower pH values. Deprotonation of [5]2+ first leads to the amino carboxylate [5 – H]+ and then to the neutral species [5 – 2H], where both the COOH and the NH2 group are deprotonated. During purification and precipitation of 2+ [5]2+ careful pH control is imperative, as otherwise mixtures Figure 3. Pourbaix diagram of [5] recorded in 0.5 m HNO3/H2O of species with different degrees of protonation [5 + H]3+/ titrated with NaOH. E (vs. SCE) – 0.16 V ≈ E (vs. ferrocene/ferro-cenium);[87] estimated pK 2+ + 2+ a values are denoted by vertical dotted [5] or [5 – H] /[5] are obtained. For synthetic applica- lines. tions, the amine deprotonation step [5 – H]+  [5 – 2H] needs to be carried out in rigorously dried, aprotic media respectively. Similarly, for [5 – H]2+ Mulliken spin densities by employing strong bases such as the phosphazene at Ru and N amount to 0.78 and 0.10, respectively. Thus, base P1-tBu [tert-butylimino-tris(dimethylamino)phos- when [5]3+ is deprotonated to [5 – H]2+ at the carboxylic phorane[85]]. acid group, the redox process remains essentially ruth- In the presence of nitrate ions, dicationic amino acid enium-centered and reversible. The second deprotonation [5]2+ is water soluble, which also enables pH-dependent occurs at the amino group. The resulting complex, [5 – electrochemical measurements. The Pourbaix diagram of 2H]+, features Mulliken spin densities of 0.36 and 0.51 on [5]2+ compiles the redox potentials (E1/2) of the RuII/RuIII Ru and the amino nitrogen atom, respectively (Figure 4). couple at different pH values (Figure 3). Below pH 1.6 the Thus, the deprotonated amino group is considerably en- potential amounts to E1/2 = 0.79 V (vs. SCE). Between pH gaged in the oxidation process, which renders the oxidation 1.6 and 2.7 the redox potential drops to 0.74 V, which is irreversible. In summary, at pH  9.2 [5]2+ and [5– H]+ are because of the deprotonation of the carboxylic acid group, reversibly oxidized at the ruthenium center on the electro- and the potential is hence assigned to the [5]2+/[5 – H]2+ chemical time scale, while at higher pH the (deprotonated) redox couple.[59] For a typical proton-coupled electron amino group is irreversibly oxidized. transfer (PCET), the expected slope of the redox potential is given by –(m/n) 59 mV per pH unit (m = number of transferred protons, n = number of transferred electrons; Nernstian behavior; 298 K).[86] The experimental slope amounts to approximately 4610 mV per pH unit suggest- ing that the electron transfer is indeed coupled to proton transfer in this pH region. The drop of the redox potential between pH 1.6 and 2.7 is easily rationalized, as the COO– group is a weaker electron acceptor than the COOH group. Thus, [5 – H]+ is oxidized at lower potential than [5]2+. From pH 2.7 to 9.2, the redox potential is pH-independent with E1/2 = 0.74 V. When the pH is above 9.2, the redox potential again drops as a result of deprotonation of the amino group, which is converted into an even stronger elec- tron donor by deprotonation.[59] At pH values below 9.2, the RuII/III oxidation process is reversible on the electro- chemical timescale, while the process becomes irreversible Figure 4. DFT-calculated spin densities of ruthenium(III) com- at pH  9.2. plexes [5]3+, [5 – H]2+, and [5 – 2H]+ (B3LYP/LANL2DZ, The irreversibility of the [5 – 2H] oxidation has been ra- IEFPCM, H2O; contour value 0.01; CH hydrogen atoms omitted;Mulliken spin densities at indicated atoms in parentheses). tionalized by density functional theory (DFT) calculations. The DFT-calculated spin densities of the ruthenium(III) In aprotic acetonitrile the fully reversible simple one-elec- complexes [5]3+, [5 – H]2+, and [5 – 2H]+ are depicted in tron RuII/RuIII oxidation of amino acid ester [4]2+ is ob- Figure 4. The spin density in [5]3+ is located at the ruth- served at E = 0.68 V vs. ferrocene/ferrocenium.[59]1/2 The enium center with a small contribution at the amino nitro- EPR spectrum of [4]3+ prepared by chemical oxidation of gen atom. The Mulliken spin densities on Ru and on the [4]2+ with ceric ammonium nitrate shows a rhombic signal amino nitrogen atom are calculated as 0.76 and 0.11, with g1,2,3 = 2.347, 2.178, 1.843 in frozen solution, which Eur. J. Inorg. Chem. 2014, 5468–5490 5472 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 10 | 1 INTRODUCTION www.eurjic.org MICROREVIEW is characteristic for RuIII.[56,77] The reversible one-electron tible to ligand loss and ligand degradation under these oxi- reduction of [4]2+ to [4]+ is located at the tpy-COOEt li- dative conditions than the [4]2+/[4]3+ couple.[72,88] Under gand.[56,59] The EPR spectrum of the N-acetylated analogue acidic conditions (0.5 m TFA), the re-reduction of [5]3+  [6]+ (Figure 2) prepared by reduction of [6]2+ with decame- [5]2+ in the spectroelectrochemical experiments is only com- thylcobaltocene shows a rhombic signal with g1,2,3 = 2.005, plete to approximately 90% and features different isosbestic 1.989, 1.955 and small g anisotropy in frozen solution.[62] points as compared with the oxidation [5]2+  [5]3+. We This is in full accord with a ligand-centered reduction with suggest that, under acidic conditions irreversible substitu- some ruthenium admixture.[56] The DFT-calculated spin tion of even a tpy ligand is feasible via the six-coordinate densities of [4]3+ and [4]+ confirm that oxidation occurs at ruthenium(III) aqua complex [Ru(κ2-tpy-R1)(κ3-tpy- the ruthenium center ([4]3+: Mulliken spin density on Ru = R2)(H2O)]3+, which tautomerizes to the six-coordinate hy- 0.76, Figure 5a) with some admixture from the amino nitro- droxido pyridinium complex [Ru(H-κ2-tpy-R1)(κ3-tpy- gen atom (Mulliken spin density on N = 0.11), while re- R2)(OH)]3+ (Figure 5c). Finally, protonated [H tpy-R1]3+3 duction is essentially localized at the tpy-COOEt ligand can be irreversibly released after further protonation to ([4]+: Mulliken spin density on Ru = 0.10, Figure 5b). Due di(pyridinium) and tri(pyridinium) under acidic conditions. to the meridional coordination of the tpy ligands involved, these two redox processes occur in orthogonal π systems (Figure 5a, b). Figure 5. DFT-calculated spin densities (B3LYP/LANL2DZ, IEFPCM, CH3CN; contour value 0.01; CH hydrogen atoms omit- ted; Mulliken spin densities at indicated atoms in parentheses) of (a) [4]3+, (b) [4]+, and (c) [5]3+  H2O with κ2-tpy-NH2 (IEFPCM, H2O). The stability of the ruthenium(III) complex [4]3+ in the presence of water and potentially coordinating hydroxide ions has been probed by spectroelectrochemical measure- ments.[72] Indeed, ruthenium(II) complex [4]2+ is reversibly oxidized to ruthenium(III) complex [4]3+ by gradually in- creasing the potential. By reversing the potential, [4]3+ is then quantitatively reduced back to [4]2+ (Figure 6). Upon Figure 6. UV/Vis spectra during (a) electrochemical oxidation of 2+ 3+ oxidation, the MLCT band of [4]2+ at λ = 502 nm disap- [4] (E = 600  1000 mV) and (b) back reduction of [4] (E = 1000  600 mV) in an optically transparent thin layer electrochem- pears while a LMCT band of [4]3+ at λ = 729 nm rises ical (OTTLE) cell in 10–3 m NaOH and 0.1 m [nBu4N](PF6) in (LMCT = ligand-to-metal charge transfer). Seven isosbestic CH3CN/H2O (98:2, v%). Black arrows indicate isosbestic points.[72] points are observed and confirm the clean conversion of [4]2+ to [4]3+ (Figure 6). Back reduction fully restores the initial spectrum, and the same isosbestic points are ob- The UV/Vis absorption spectrum of [4]2+ features ππ* served, which clearly demonstrates the stability of ruth- transitions below λ = 400 nm and a characteristic MLCT enium(II/III) complexes [4]2+ and [4]3+ in the presence of band at λmax = 502 nm (ε –1 –1max = 19100 m cm ; Figures 6 water and coordinating OH– ions on this time scale and 7).[59] Time-dependent DFT calculations (B3LYP/ (hours).[72] In contrast, the famous N719 sensitizer as tetra- LANL2DZ, IEFPCM, CH3CN) confirm the ligand-cen- ethyl ester, cis-[Ru{{bpy(COOEt)2}}2(NCS)2], featuring bi- tered character of the ππ* transitions at λ  400 nm and dentate 4,4-diethylcarboxy-2,2-bipyridine ligands and the essentially MLCT character of the transitions at monodentate isothiocyanato ligands, is much more suscep- λ  400 nm. Relevant Kohn–Sham frontier molecular or- Eur. J. Inorg. Chem. 2014, 5468–5490 5473 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Section 1.1 | 11 www.eurjic.org MICROREVIEW bitals of [4]2+ are depicted in Figure 8. Five frontier orbit- als, HOMO-2 to LUMO+1, participate in transitions with λ  400 nm. HOMO to HOMO-2 correspond to the t2g or- bitals in Oh symmetry and are mainly ruthenium-centered with a small contribution from the amino nitrogen atom in the HOMO. The contribution of the amino group to the MLCT transition is experimentally verified by resonance Raman experiments involving excitations of the MLCT transitions (λ = 458–514 nm).[61] The LUMO and LUMO+1 are located on the acceptor-functionalized tpy ligand. Again, resonance Raman spectroscopy corroborates the participation of the respective ligand, namely the tpy- COOEt ligand by the enhancement of the characteristic 1726 cm–1 ester stretching mode.[61] Thus, the 500 nm ab- sorption band is best described by a mixed 1MLCT/1LLCT character (LLCT = ligand-to-ligand charge transfer). For brevity we will refer to this mixed charge transfer transition in the following as MLCT only. The high-energy orbitals LUMO+9 and LUMO+13 corresponding to the eg* orbit- als (in Oh symmetry) do not participate in the 1MLCT ab- sorption process (Figure 8). Figure 8. DFT-calculated Kohn–Sham frontier molecular orbitals (contour value 0.06 a. u.) of [4]2+ (B3LYP/LANL2DZ, IEFPCM, CH3CN; CH hydrogen atoms omitted). Figure 7. UV/Vis absorption spectrum (black) and emission spec- trum (red) of [4](PF6)2 in CH3CN at 295 K. After population of the initial 1MLCT state, ISC and Figure 9. Transient absorption spectra of [4]2+ (pulse λexc = 400 nm, [56] vibrational relaxation occur, forming the emissive 3MLCT 2600 nJ) in PrCN at 295 K. state (Figure 1). Excited-state dynamics on the picosecond timescale of [4]2+ has been experimentally probed by tran- lifetime of τ = 34 ns and a quantum yield of Φ = 0.27 % sient absorption measurements.[56] When [4]2+ is excited at (Figure 7).[61] In frozen butyronitrile at 77 K, the emission λexc = 400 nm, its transient absorption spectra show an in- energy shifts to higher values (λ = 702 nm).[56] Notably, stant ground-state bleach at λ = 509 nm together with a amino acid [5]2+ shows a second time constant of τ = 4 ns photoinduced absorption at λ = 565–960 nm (Figure 9). (8%) for the excited-state decay, which might be ascribed to The latter absorption nicely fits to a LMCT absorption of proton quenching of the MLCT state by a nearby acid.[89] a RuIII-based excited state (cf. Figure 6 for [4]3+).[56] The Indeed, ground-state aggregation of carboxylic acids is very region of the photoinduced absorption reveals a fast pro- common,[90] and in the MLCT excited state the basicity of cess with a time constant of 7.2 ps at 295 K before reaching the formally reduced tpy-COOH ligand should be in- the thermalized 3MLCT state. These dynamics might be as- creased, which suggests a proton transfer to the tpy-CO sociated with localization of the excited electron on the tpy- group of the excited complex from a ground-state acid (ex- COOEt ligand, ISC, or vibrational relaxation within cited-state proton transfer, ESPT[89]). 1/3MLCT states.[56] In essence, the 3MLCT state is popu- The emitting 3MLCT state as well as the deactivating lated and thermally equilibrated in the picosecond time 3MC state were successfully modeled by DFT calcula- scale. tions.[56] The 3MC state of push-pull-substituted [4]2+ is cal- At 295 K in fluid solution, [4]2+ shows phosphorescence culated to be higher in energy by 26.8 kJmol–1 relative to from this 3MLCT state at λmax = 734 nm with a 3MLCT its 3MLCT state (Figure 10). On the other hand, the corre- Eur. J. Inorg. Chem. 2014, 5468–5490 5474 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 12 | 1 INTRODUCTION www.eurjic.org MICROREVIEW sponding triplet states of the parent complex, [1]2+, are efficient (ε = 19100 m–1max cm–1).[59] In the electronically found to be essentially isoenergetic.[56] While the geometry similar push-pull-substituted [Ru(tpy-OH)(tpy-SO Me)]2+2 of the 3MLCT state of [4]2+ differs only slightly from the complex [7]2+ (Figure 2, Table 1), the combination of an 1GS geometry, the 3MC geometry is significantly distorted OH donor with a SO2Me acceptor group enables similar with respect to the 1GS and 3MLCT states. For example, excited-state properties (τ = 50 ns; Φ = 0.07%; Table 1).[50] the bond lengths between Ru and the central N atom of Thiophene or triarylamine substituents, for example in tpy-COOEt are 1.99, 2.04, and 2.17 Å for the 1GS, 3MLCT, [8]2+[63] and [9]2+,[64] also exert an electron-donating effect and 3MC states, respectively. The N–C–C–N dihedral as seen in the bathochromically shifted MLCT absorption angles of the tpy-COOEt ligand amount to 0°, 0°, and 11° maxima with respect to that of [1]2+ (Figure 2, Table 1). for the 1GS, 3MLCT, and 3MC states, respectively, which Typically, an enlargement of the chromophore system shows that the tpy-COOEt ligand has lost its planarity in stabilizes the 3MLCT state relative to the 3MC state. In- the 3MC state. The spin density of the 3MLCT state is deed, in [Ru(tpy-Ph) 2+2] ([10]2+, Figure 2), featuring phenyl shared between the ruthenium center and the tpy-COOEt groups appended to the 4-positions of the tpy ligands, the ligand, while for the 3MC state the spin density is confined 3MLCT lifetime and quantum yield (τ = 1 ns; Φ = 0.006%) to the ruthenium center, as expected for a ligand-field ex- of [10]2+ are substantially enhanced as compared with those cited state based on the (t )5(e *)12g g electron configuration of [1]2+, but they are still far from useful (Table 1).[50] The (Figure 10). The connecting transition state (3TS) between combination of donor–acceptor functionalization and an the 3MLCT and the 3MC state of [4]2+ is calculated to have enlarged chromophore system has been realized in extended an energy 31.7 kJmol–1 higher than the 3MLCT state, while amino acid derivatives [11]2+–[13]2+ featuring para-phenyl- the 3TS of [1]2+ is only 7.2 kJmol–1 higher than its 3MLCT ene spacers between the 4-substituents and the tpy ligands state, which explains its rapid radiationless excited-state de- (Figure 2).[60] However, this combination does not lead to a cay. Regarding energy, spin density, and geometry, the 3TS synergetic effect: [11]2+ features a lower 3MLCT lifetime transition states strongly resemble the 3MC rather than the and a lower quantum yield (τ = 23 ns; Φ = 0.053 %) com- 3MLCT states (Figure 10).[56] pared with those of the phenylene-free analogue [4]2+ (Table 1). Compounds [12]2+ and [13]2+ also exhibit excited- state properties that are inferior to those of the phenylene- free analogue [5]2+ (Figure 2, Table 1).[59,60] The ring planes of the phenylene ring and the central pyridyl ring of a tpy ligand are far from co-planar. Thus, the reduced π-conjuga- tion mitigates the positive effect of donor–acceptor substi- tution.[60] Coplanarity in extended tpy ligands has been achieved by replacing para-phenylene with pyrimidine spa- cers, for example in complexes [14a]2+ and [14b]2+ (Fig- ure 2).[65] In these complexes the enhanced π-conjugation efficiently stabilizes the 3MLCT state with respect to the 3MC state, giving long excited-state lifetimes up to τ = 231 ns and quantum yields up to Φ = 0.17 % (Table 1).[65] Figure 10. DFT-calculated geometries, relative energies, and spin Long excited-state lifetimes can also be achieved by the densities of triplet minima (3MLCT and 3MC) and the transition so-called multichromophore approach: additional ap- state (3TS) of [4]2+ (B3LYP/LANL2DZ, IEFPCM, CH3CN; con- pended chromophores, such as pyrene or anthracene, can tour value 0.015; energies in kJmol–1; CH hydrogen atoms omitted; possess triplet intraligand excited states (3IL) with energies Mulliken spin densities at indicated atoms in parentheses).[56] similar to the 3MLCT state. In this case, a triplet–triplet equilibrium between these triplet states is feasible.[11] As de- activation from the triplet state of the organic chromophore 3 3. Strategies toward Long-Lived and Highly is spin-forbidden and hence slow, the IL state of the or- Emissive Excited States ganic chromophore acts as an excited state reservoir for the emitting 3MLCT state and such a 3MLCT/3IL equilibrium With respect to ester complex [2]2+ (τ = 32 ns; Φ = can significantly prolong the phosphorescence.[91,92] For in- 0.04 %; Table 1; Figure 2),[8] the electron-donating NH 2+ 2+2 stance, complexes [15a] and [15b] (Figure 2), featuring group in the bis-4-substituted amino acid ester [4]2+ desta- pyrene units as organic triplet reservoirs, indeed reach life- bilizes the 3MC state compared with the 3MLCT state, times of τ = 580 and 2500 ns, respectively.[66] which leads to a further improvement of the excited-state A pronounced push-pull situation is also present in carb- properties (τ = 34 ns; Φ = 0.27%; Table 1; Figure 2).[59] The ene complexes [16]2+ and [17]2+ (Figure 11) prepared by push-pull substitution of [4]2+ induces an even smaller Schubert and Berlinguette.[67] The combination of an elec- HOMO–LUMO energy gap (Figure 8), which significantly tron-accepting tpy ligand with the strongly σ-donating shifts the MLCT absorption maximum to lower energy CNC bis(carbene) ligand 2,6-bis(3-methyl-1,2,3-triazol-4- (λmax = 502 nm). In addition, the NH2 group in [4]2+ en- yl-5-idene)pyridine results in exceptionally long excited- larges the chromophore system and raises the extinction co- state lifetimes and quantum yields up to τ = 7900 ns and Φ Eur. J. Inorg. Chem. 2014, 5468–5490 5475 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Section 1.1 | 13 www.eurjic.org MICROREVIEW = 17.3% (Table 1).[67] Further N-heterocyclic carbene do- idine]. Compound [20]2+ has a very long-lived 3MLCT state nor ligands as well as cyclometalating ligands and their (τ = 3300 ns) and the highest room-temperature quantum ruthenium complexes will not be discussed here in more yield in fluid solution (Φ = 30%) reported to date among detail, and the reader is referred to recent literature.[4,93–97] bis(tridentate) ruthenium(II) complexes (Table 1).[71] Although these excellent photophysical properties are highly beneficial, one drawback is associated with large che- late rings. Six-membered chelate rings such as those found in [18]2+ and [19]2+ enable highly flexible coordination, and this can lead to undesired stereoisomers, namely mer, cis- fac, and trans-fac isomers.[99] Mixed-ligand complex [Ru(tpy-COOEt)(ddpd)]2+ ([21]2+, Figure 11) prepared by our group overcomes the problem of fac/mer-stereoisomers by using the combination of a five-membered tpy and a six-membered ddpd chelate ligand (ddpd = N,N-dimethyl- N,N-dipyridin-2-ylpyridine-2,6-diamine).[56,100] Complex [21]2+ can be prepared in a stepwise manner by first intro- ducing the tpy ligand to obtain ruthenium(III) complex Ru(tpy-COOEt)Cl3, which is converted into the final, pure meridional stereoisomer [21]2+ by treatment with the ddpd ligand. The tpy ligand only coordinates in a planar meridio- nal fashion, thus forcing the ddpd ligand to adopt the me- ridional coordination as well. The bite angles of the ddpd ligand are 88°, similar to those of dqp and dcpp. Despite the presence of only a single ddpd ligand with 88° bite angles and a tpy ligand with only 79° angles, complex [21]2+ achieves a quantum yield of Φ = 0.45% and a re- markably long 3MLCT lifetime of τ = 722 ns at room tem- perature in solution (Table 1).[100] The synergy of the elec- tron-accepting tpy ligand and the electron-donating ddpd ligand containing NCH3 groups ([21]2+) creates a push-pull Figure 11. Bis(tridentate) complexes of ruthenium(II) [16]2+–[27]2+. situation and shifts the maximum absorption wavelength further to lower energy (λmax = 517 nm) relative to that of A further successful strategy to improve the excited-state [2]2+ lacking the NCH3 groups (λmax = 485 nm) (Figure 2, properties in bis(tridentate) ruthenium(II) complexes is the Table 1). In push-pull complex [22]2+, the two outer pyr- optimization of N–Ru–N bite angles. All complexes [1]2+– idine rings of the tpy ligand are additionally functionalized [17]2+ feature five-membered chelate rings and N–Ru–N by electron-withdrawing ester groups. The maximum ab- bite angles of around 79°. The carbene chelate ligands in sorption wavelength of [22]2+ (Figure 11) is further red- [16]2+ and [17]2+ have even smaller C–Ru–N bite angles of shifted (λmax = 539 nm), and even higher values for the 77°.[54c,56,59,67] Bite angles of 90° maximize the orbital over- 3MLCT lifetime (τ = 841 ns) and the quantum yield (Φ = lap between the pyridine nitrogen lone pairs and the d or- 1.1%) are obtained (Table 1).[56] bitals of the central metal with eg symmetry and hence in- However, the increase of 3MLCT lifetimes and quantum crease the antibonding character of the eg* orbitals and the yields by push-pull substitution tuning has a limit. A pro- ligand-field splitting. This stronger ligand-field splitting nounced push-pull situation not only increases the energy augments the energy difference between 3MLCT and 3MC gap between 3MLCT and 3MC states but simultaneously states, which again hampers radiationless deactivation via lowers the gap between the 3MLCT state and the ground the latter state (Figure 1).[54c] Hammarström et al. intro- state 1GS. According to the energy gap law, a small duced bite angles of 90° by using six-membered chelates in 3MLCT–1GS energy difference induces fast radiationless [Ru(dqp) ]2+ ([18]2+2 , Figure 11), which led to high values deactivation into the ground state.[79,101–103] Such a strong for the room-temperature lifetime of the excited state and push-pull situation exists in complexes [23]2+ and [24]2+ emission quantum yield [τ = 3000 ns, Φ = 3.2 %, dqp = 2,6- (Figure 11), which feature an additional NH2 group on the di(quinolin-8-yl)pyridine].[68,69,98] Homoleptic complex electron-donating ddpd ligand relative to their NH2-free [Ru(EtOOC-dqp) ]2+2 ([19]2+, Figure 11) with electron-with- counterparts [21]2+ and [22]2+, respectively. The push-pull drawing substituents features even higher values (τ = character of [23]2+ and [24]2+ manifests itself by the red- 5500 ns; Φ = 11.2%, Table 1).[70] Ruben et al. introduced shifted absorption maxima and by the lower RuII/RuIII re- [Ru(dcpp) 2+2] ([20]2+, Figure 11), featuring six-membered dox potentials (ca. 0.2 V) as compared with those of [21]2+ chelates and 90° bite angles by formal insertion of carbonyl and [22]2+, respectively (Table 1).[56] Indeed, [23]2+ and spacers between the pyridine rings of the parent [Ru- [24]2+ are among the best red absorbers (and red emitters) (tpy)2]2+ complex [dcpp = 2,6-di(2-carboxypyridyl)pyr- collected in Table 1. However, the NH2 group dramatically Eur. J. Inorg. Chem. 2014, 5468–5490 5476 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 14 | 1 INTRODUCTION www.eurjic.org MICROREVIEW reduces the 3MLCT lifetimes of [23]2+ and [24]2+ by a factor Amino acid [5]2+ (Figure 2) has even been employed as of about five to six, and quantum yields are reduced by an building block in solid-phase peptide synthesis (SPPS) by even higher factor compared with those of the NH2-free our group.[18] The general synthetic strategy is depicted in analogues (Table 1). The faster excited state deactivation in Figure 12. Initially, a polymer TentaGel S resin[109,110] [23]2+ and [24]2+ is induced by high-energy oscillators, in equipped with a Wang linker[111] (compound 28) is func- this case predominantly by the N–H modes (multiphonon tionalized with Cl-Gly-Fmoc (Gly = glycine, Fmoc = fluor- deactivation). This is shown by (NH2  ND2) deuteration enylmethoxycarbonyl). The Fmoc protecting group is re- experiments with deuterated complexes [23D]2+ and [24D]2+ moved by piperidine to obtain the free NH2 group (com- featuring higher quantum yields than those of their parent pound 29). After activation of the carboxylic acid group of complexes [23]2+ and [24]2+, respectively.[56] In essence, the [5]2+ by PyBOP, activated [5]2+ is connected to the solid above-mentioned small 3MLCT–1GS energy gap, together support through an amide bond to obtain the immobilized with high-energy oscillators, is responsible for radiationless complex [30]2+. The weak nucleophilicity of the aromatic deactivation of the 3MLCT state in [23]2+ and [24]2+. NH2 group of [5]2+ renders an amino protection group un- Regarding thermal and photochemical stability, bis(tri- necessary.[18,61] However, for the activation of the NH2 dentate) complexes should outperform tris(bidentate) com- group of [30]2+, strong coupling reagents such as acid chlor- plexes. The benchmark complex, [Ru(bpy)3]2+, has been re- ides (e.g. Cl-Gly-Fmoc) are necessary in order to form an ported to be photolabile in the presence of coordinating amide bond such as that in [31]2+. Monitoring of SPPS re- anions such as Cl–, Br–, I–, SCN–, or NO –3 . In its excited actions with charged ruthenium amino acids is advan- state, [Ru(bpy) ]2+3 is prone to photoinduced ligand substi- tageously achieved by treating a small portion of the resin tution, photooxygenation, and photoracemization.[104–108] In contrast, bis(tridentate) ruthenium(II) complexes such as [4]2+, [18]2+, and [21]2+ feature significantly higher photo- stabilities under constant irradiation (λ = 400–500 nm) as compared with that of [Ru(bpy)3]2+.[70,72] The enhanced photostability of complexes with tridentate ligands is obvi- ously highly favorable for photo applications of all kinds, and some aspects will be discussed in the following sections. 4. Conjugates of the Ruthenium(II) Amino Acids 4.1 Amide Bond Formation and Solid-Phase Peptide Synthesis Amide bond formation at the C-terminus of amino acid [5]2+ is straightforward by using standard amide coupling reagents such as HOBt/DCC, PfpOH/DCC, PyBOP, or HATU/NEt3 {HOBt = 1-hydroxybenzotriazole, DCC=N,N-dicyclohexylcarbodiimide, PfpOH = penta- fluorophenol, PyBOP = benzotriazole-1-yl-oxy-tris- pyrrolidino-phosphonium hexafluorophosphate, HATU = 1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]- pyridinium 3-oxide hexafluorophosphate}.[25,77,78] The choice of coupling reagent hereby solely depends on the nu- cleophilicity of the chosen amine and eventual protective groups to be employed. Even under the strongest coupling conditions (HATU) no amide coupling of [5]2+ with itself is observed, which underlines once again the extremely poor reactivity of the NH2 group.[77] The amino group can be transformed into a much better nucleophile by deproton- ation with a strong base (see above), so that amide-bond formation becomes feasible. Then, even comparably mild coupling conditions, for example PfpOH/DCC, are suf- ficient to overcome the Coulomb repulsion between two positively charged complex fragments to yield dinuclear quadruply charged species (see below).[25,62] Coupling of or- ganic acids to [5]2+ is successful by using acid chlorides or Figure 12. Solid-phase peptide synthesis including ruthenium(II)amino acid [5]2+ (Fmoc = fluorenylmethoxycarbonyl, PyBOP = anhydrides without further activation of the NH2 group of benzotriazole-1-yl-oxy-tris-pyrrolidino-phosphonium hexafluoro- [5]2+.[18,25,61,77,78] phosphate, TFA = trifluoroacetic acid).[18] Eur. J. Inorg. Chem. 2014, 5468–5490 5477 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Section 1.1 | 15 www.eurjic.org MICROREVIEW with TFA, drying the filtered solution, re-dissolving in efficiently quenched (Figure 14b).[18] The excitation emis- CH3CN and recording ESI mass spectra of the released cat- sion matrix of [35]2+ shows that excitation at λexc = 470– ionic intermediates). After deprotection by piperidine, the 500 nm results in emission at λem = 670 nm, which perfectly aliphatic NH2 group of [31]2+ readily forms amide bonds corresponds to the phosphorescence of the coumarin-free with carboxylic acids R–COOH of all kind by PyBOP or reference [37]2+ (Figure 14a). Furthermore irradiation of acid chloride activation to give peptides [32]2+. Several (ac- [35]2+ at a wavelength of λexc = 390–460 nm, which corre- tivated) acids R–COOH have been attached as terminal sponds to the coumarin excitation, leads to ruthenium- functional group, for example CH3COOH, coumarin-3- based phosphorescence (Figure 14b). For the longer bis(gly- carboxylic acids, anthracene-2-carboxylic acid, or 9-fluor- cine) bridge ([36]2+, n = 2, Figure 13), excitation of the cou- enylmethoxycarbonyl chloride.[18] Finally, metallo peptides marin at λexc = 390–460 nm essentially leads to fluorescence [33]2+ are released from the solid phase by treatment with of the organic chromophore at λmax = 465 nm, while phos- TFA. phorescence from the [Ru(tpy) ]2+2 moiety is practically not observed (Figure 14c and Supporting Information). Hence energy transfer is inefficient in [36]2+ (n = 2) but efficient in [34]2+ and [35]2+ (n = 0, 1). Interestingly, all peptides [33]2+– [35]2+ with a terminal COOH group possess an additional τ 4.2 Energy Transfer between Organic Dyes and = 2–4 ns emission component similar to that of amino acid Bis(terpyridine)ruthenium(II) Polypyridine complexes of ruthenium(II), especially [Ru(bpy) 2+3] -type complexes, are suitable compounds for directional energy transfer and can act as donors and ac- ceptors for photoinduced energy transfer.[112] Bis(tridentate) complexes of ruthenium(II) have been incorporated into en- ergy-transfer systems by using ethynyl spacers between a [Ru(tpy) ]2+ donor and thiophene units as acceptors.[113]2 Donor–acceptor-substituted complex [5]2+ has been con- nected to organic chromophores with amide bridges by means of SPPS (Figure 12) with a different number of gly- cine amino acids in between (Figure 13).[18] A single glycine unit or none ([34]2+: n = 0; [35]2+: n = 1) enables coplanarity of the coumarin dye and the adjacent tpy ligand, which is favorable for efficient energy transfer by the Dexter mecha- nism.[18] With these short bridges, excitation of the couma- rin ([35]2+, n = 1, λexc = 422 nm) induces energy transfer to the [Ru(tpy) ]2+2 moiety, while the coumarin fluorescence is Figure 14. Excitation emission matrices of (a) reference complex Figure 13. Energy transfer in peptides of [5]2+, coumarin dyes [37]2+, (b) [35]2+ (n = 1), and (c) [36]2+ (n = 2). The color bar [34]2+–[36]2+, and reference complex [37]2+.[18] indicates the emission intensity (* = 2λexc). Eur. J. Inorg. Chem. 2014, 5468–5490 5478 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 16 | 1 INTRODUCTION www.eurjic.org MICROREVIEW [5]2+ (Table 1). Again, we ascribe this additional pathway of 3MLCT decay in carboxylic acids to ESPT, namely to proton transfer from an acid to the excited complex.[18,89] 4.3 Mixed-Valence and Mixed-Metal Complexes As described in Section 4.1, amide coupling reactions with building block [5]2+ can also be used to selectively build dinuclear complexes containing two ruthenium(II) centers without formation of rac/meso diastereomers (Fig- ure 15, [38]4+–[40]4+).[25,62] The electrochemical properties of these systems are very similar to those of their constitu- ent mononuclear systems. For example, dinuclear [38]4+ (Figure 15) can be considered to be built up from mononu- clear complexes [4]2+ and [6]2+ (Figure 2). The redox poten- tials of the RuII/RuIII couples in [38]4+ are E1/2 = 0.68 V at the N-terminal ruthenium center and E1/2 = 0.91 V at the C-terminus,[25] and thus they are essentially unperturbed in comparison with mononuclear species [4]2+ and [6]2+ (E1/2 = 0.68 and 0.85 V, respectively; Table 1).[59,61] On the other hand, the absorption spectrum of [38]4+ (Table 1) cannot be regarded as a simple superposition of the absorption spec- tra of [4]2+ and [6]2+. A bathochromic and hyperchromic shift of the MLCT band from around λmax = 500/490 nm (ε ≈ 20000 m–1 cm–1 for mononuclear complexes [4]2+max and [6]2+) to λmax = 522 nm (εmax = 50600 m–1 cm–1)[25] is ob- served relative to the mononuclear complexes due to an ex- tension and stronger push-pull substitution of the chromo- phores. Emission of [38]4+ occurs from the lowest-energy 3MLCT state (λmax = 750 nm, Φ = 0.24 %, τ = 22 ns, Table 1), which is located on the N-terminal side of the di- nuclear system, because of the electron-donating effect of the amino group and the strongly electron-withdrawing ef- fect of the RuII complex attached to the C-terminus. This is why the emission energy and quantum yield of [38]4+ better match those of NH 2+2-substituted [5] than those of NHAc- substituted [6]2+ (Table 1). Obviously, the excited-state en- ergy is efficiently transferred to the lowest emitting 3MLCT state irrespective of the initial excitation locus. Similar ob- servations are made for the phenylene-extended dinuclear amide [39]4+ (Figure 15), although the bathochromic and hyperchromic effects are less pronounced because of the “phenylene dilution” (see above).[25] Similar to acids [33]2+– [35]2+ and to amino acid [5]2+ itself, acid [38]4+ shows a second decay component with τ = 2 ns ascribed to ESPT.[18,89] One-electron oxidation of these dinuclear complexes Figure 15. Dinuclear amide-linked bis(terpyridine)ruthenium(II) II III [25,51,52,62,73,74]yields mixed-valent Ru Ru complexes [38]5+ and complexes and related complexes. [39]5+.[25] No photoinduced intervalence charge transfer (IVCT) from RuII to RuIII is observed by UV/Vis/NIR spec- All these facts result in valence-localized mixed-valent sys- troscopy. Theoretical results also indicate no electronic in- tems [38]5+ and [39]5+ of Robin–Day class I.[25,114] teraction between the metal centers in [38]5+ and [39]5+. In order to maximize the redox symmetry of [38]4+, the This is attributed to the substantial electronic difference of capping functionalities were adjusted to give dinuclear the individual complex moieties and the resulting high re- amide-linked complex [40]4+ (Figure 15) with ethyl carbox- dox asymmetry within the complexes, the large distance be- ylato and acetamido functional groups.[62] While the redox tween the ruthenium centers of about 13 and 18 Å for asymmetry indeed essentially vanishes in [40]4+ and both [38]5+ and [39]5+, respectively, and the twisting in the bridge. ruthenium(II) ions are oxidized at E1/2 = 0.91 V, still no Eur. J. Inorg. Chem. 2014, 5468–5490 5479 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Section 1.1 | 17 www.eurjic.org MICROREVIEW electronic communication is observed after oxidation of temperature in solution and has an emission band signifi- [40]4+ to mixed-valent RuIIRuIII cation [40]5+ with ceric am- cantly broadened in comparison to that of [6]2+ and other monium nitrate in aqueous sulfuric acid solution. Theoreti- mononuclear complexes. Indeed, this broad band can be cal data on [40]5+ again indicate an entirely valence-local- perfectly approximated by two emission bands with λmax = ized mixed-valent complex. According to paramagnetic 675 nm (71%) and λmax = 706 nm (29%).[62] Notably, NMR spectroscopy studies, the first oxidation yields the [40]4+ also shows two decay components with τ1 = 24 ns RuIIRuIII valence isomer that has the RuIII center located at (71%) and τ2 = 44 ns (29%). These components are ob- the N-terminus. The observation of electronically essentially served independent of the presence of additional water, uncoupled ruthenium centers in these dinuclear mixed-val- chloride, or dioxygen. Hence, this biexponential excited de- ence complexes with two terpyridine ligands is consistent cay is an intrinsic property of [40]4+ and is assigned to the with data on similar systems reported in the literature: dinu- dual emissions of the two different chromophores in [40]4+. clear amide [41]4+ and its mixed-valence congener [41]5+ re- As the emission band shape and position are independent ported by Colbran et al.,[73] in which two para-phenylene of the excitation energy (λexc = 450–550 nm), these two linkers connect the chromophores with the amide bridge, chromophores are involved in a rapid excited-state equili- contain two electronically isolated ruthenium ions with bration at room temperature (triplet–triplet energy trans- identical properties (Figure 15). This can be attributed to fer). the even larger Ru–Ru distance compared with that in In addition to the rich optical and redox chemistry of [39]4+. A similar picture was obtained for the alternating the dinuclear complexes, acid–base reactivity arises from leucine bis(terpyridine)ruthenium(II) peptides [42n](2n+2)+ (n the two amide functionalities in [40]4+, similar to [5]2+ (see = 1–4)[74] with two chromophores separated by two para- above). The electron-withdrawing effects of the complex phenylene groups and a leucine spacer. Extinction coeffi- fragments strongly polarize the N–H bonds and thus these cients of this series of compounds are proportional to the protons become substantially acidic. This is especially pro- number (n) of chromophores present per molecule with un- nounced for the central amide unit flanked by two doubly shifted absorption maxima. The observation of a single re- cationic charges. The combination of two oxidation and two dox wave for all RuII/RuIII couples additionally underlines deprotonation steps yields eight conceivable species starting the electronic independence of the complex fragments. The from [40]4+. These are summarized in the 33 square parent dinuclear complex, [43n]4+ (n = 0),[51,52] with directly scheme in Figure 16. While the successive deprotonation linked (back-to-back) terpyridine ligands also exhibits only steps [40]4+  [40 – H]3+  [40 – 2H]2+ could be thor- a single redox wave for the two RuII/RuIII couples. Upon oughly examined experimentally (horizontal reactions), pre- oxidation to [43n]5+ (n = 0), however, a weak IVCT band is parative oxidation of deprotonated species (vertical reac- observed in the NIR region, which is indicative of an elec- tions), ([40 – H]3+  [40 – H]4+  [40 – H]5+) was success- tronic interaction of the ruthenium centers. Extension of ful neither by chemical nor by electrochemical means be- the bridge by one or two para-phenylene linkers reduces the cause of the high RuII/RuIII potential. NMR spectroscopy electronic coupling parameter in the complexes [43n]5+ (n reveals that the bridging amide proton of [40]4+ is readily = 1, 2),[51] but despite the significant twisting of the para- deprotonated in MeCN/H2O mixtures by employing NEt3 phenylene linkers against each other the communication as base, which leads to [40 – H]3+. The less acidic terminal does not vanish completely. We suggest that the combina- amide proton requires the much stronger phosphazene base tion of an amide linker with terpyridine ligands in [38]4+– P1-tBu[85] in dry acetonitrile and gives [40 – 2H]2+ (Fig- [42]4+ suppresses an electronic coupling of the ruthenium ure 16).[62] For each deprotonation step, a unique set of six ions. The lack of interaction is attributed to a mismatch of isosbestic points is observed in the UV/Vis spectra indica- the orbital energies of the bridge and the metal centers but tive of two successive reversible processes. The 1MLCT not to unsuitable frontier orbital symmetries.[114–118] band of [40]4+ is substantially shifted by 22 nm to λmax = One-electron reduction of [40]4+ to [40]3+ by cobaltocene 526 nm during the first [40]4+  [40 – H]3+ deprotonation affords a ligand mixed-valent complex in which the un- and further to λ = 533 nm during the second [40 – H]3+max paired electron is localized on the carboxy terpyridine of  [40 – 2H]2+ step. This is certainly related to the strong the bridging ligand according to DFT calculations and electron-donating character of deprotonated amide moieties paramagnetic NMR spectroscopy studies. EPR spectro- that increases the push-pull character of the respective scopic studies on [40]3+ reveal a rhombic signal pattern [Ru(tpy)2]2+ chromophores in [40 – H]3+ and in [40 – (g1,2,3 = 2.006, 1.989, 1.958) in frozen solution, closely re- 2H]2+. sembling the spectrum of the singly reduced complex [6]+ Using ferrocene carboxylic acid or aminoferrocene and (see above) and suggesting a valence-localized radical at the [5]2+ as coupling partners affords heterodinuclear RuII/FeII bridging ligand.[62] complexes [44]2+ or [45]2+, respectively, which differ in the The optical properties of dinuclear compound [40]4+ are site of ferrocene attachment at the ruthenium amino acid related to those of mononuclear [6]2+ carrying the same (Figure 17).[59] Heterotrinuclear FeII/RuII/FeII complex functional groups.[61,62] A hyperchromic effect on the [46]2+ is obtained by treatment of [5]2+ (with HOBt/DCC MLCT absorption band stronger than that for the [5]2+/ activation) with aminoferrocene followed by reaction with [38]4+ pair is observed, while the bathochromic shift is less an activated ferrocene carboxylic acid without using any pronounced (Table 1). Compound [40]4+ is emissive at room amine protection groups (Figure 17).[59] The MLCT ab- Eur. J. Inorg. Chem. 2014, 5468–5490 5480 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 18 | 1 INTRODUCTION www.eurjic.org MICROREVIEW Figure 16. 3 3 square scheme of multifunctional [40]4+ comprising two RuII/RuIII oxidations and two NH deprotonations.[62] sorption bands of [44]2+ and [46]2+ appear at λmax = lower potential than that of the Fc-CO moiety (E1/2 = 496 nm, somewhat shifted to lower energy as compared +0.25 V).[119,120] The first reduction of the ferrocenyl com- with that of [6]2+ (λmax = 492 nm), possibly because of the plexes is located at the carboxy-substituted tpy ligand (see slightly stronger electron-donating effect of ferrocenyl com- above) differing only slightly between tpy-COOEt com- pared with methyl. The MLCT band of [45]2+ is shifted to pound [44]2+ and tpy-CONHFc compounds [45]2+ and λmax = 502 nm, similar to that of amino acid ester [4]2+ [46]2+ (Table 1). (Table 1). In addition, difference spectra reveal the presence Ferrocene is known to deactivate electronically excited of a d(FcCO)π*(tpyNH) MLCT band at λmax = 493 nm states either by electron transfer from ferrocene to the ex- in [44]2+ and in [46]2+, similar to those in amide-free ferro- cited chromophore or by energy transfer from the chromo- cenyl complexes [Ru(tpy)(tpy-Fc)]2+ [470]2+ and [Ru(tpy- phore to give the ferrocene triplet state (Figure 17).[76] All Fc) ]2+ [480]2+2 featuring extra d(Fe)π*(tpy) CT bands at three RuII/FeII conjugates [44]2+–[46]2+ are emissive at room λ [75a,75b]max = 515 and 526 nm, respectively. The difference temperature with emission wavelength maxima of 739 nm in the d(Fe)π*(tpy) CT energies between [470]2+/[480]2+ for [45]2+ bearing an amino group and 704 nm for the on the one hand and [44]2+/[46]2+ on the other hand is at- amide-functionalized conjugates [44]2+ and [46]2+ similar to tributed to the electron-withdrawing effect of the carboxyl the ferrocene-free counterparts [4]2+ and [6]2+ (Table 1). group at the ferrocene in the latter complexes, lowering the While the emission quantum yield of [44]2+ is in the range ferrocene HOMO energy and hence increasing the of those of other bis(terpyridine)ruthenium(II) complexes d(FcCO)π*(tpyNH) absorption band energy. A corre- (Table 1), the phosphorescence of complexes [45]2+ and sponding inverse d(FcNH)π*(tpyCO) MLCT band in [46]2+ is strongly quenched. Obviously, this phenomenon is [45]2+ is obviously of low intensity and not observed.[59] related to the presence of a C-terminal easy-to-oxidize Cyclic voltammograms (Table 1) of the ferrocenyl com- ferrocenyl substituent while an N-terminal Fc-CO substitu- pounds reveal two (for [44]2+and [45]2+) and three (for ent has only a marginal effect on the quantum yield [46]2+) oxidation waves, corresponding to FeII/FeIII (revers- (Table 1). With the Rehm–Weller equation, ΔGET = E1/2(Fc/ ible) and RuII/RuIII redox couples (irreversible as a result Fc+) – E1/2(tpy–/tpy) – e 20 /[4πε0ε(solvent)r] – E ,[121]00 experi- of the large positive charge accumulation and deposition on mental redox potentials and emission energies can be used the electrode).[59] As expected from substituent effects, the to estimate that photoinduced electron transfer to the oxidation of the FcNH moiety (E1/2 = –0.03 V) occurs at 3MLCT excited ruthenium moiety is more exergonic from Eur. J. Inorg. Chem. 2014, 5468–5490 5481 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Section 1.1 | 19 www.eurjic.org MICROREVIEW Intermolecular, concentration-dependent phosphores- cence quenching is also found for [6]2+ and ferrocene (E1/2 = 0.0 V), ferrocenecarboxylic acid methyl ester (E1/2 = 0.27 V[87]) and ferrocene-1,1-dicarboxylic acid dimethyl es- ter (E1/2 = 0.49 V[87]). For the first two ferrocenes reductive quenching is exergonic (ΔGET  0), while it is estimated slightly endergonic for the latter (ΔGET  0). The Stern– Volmer constants[124] of KSV = 244(5), 162(6), 148(2) L mol–1 reflect this trend predicted by Marcus theory.[62] However, considerable quenching is even observed for fer- rocene-1,1-dicarboxylic acid dimethyl ester, and this ex- cited state deactivation is largely attributed to triplet–triplet energy transfer to give the ferrocene triplet state. Hence, the organometallic ferrocene seems to act both as electron do- nor as well as energy acceptor towards excited bis(terpyrid- ine)ruthenium(II) chromophores, in an intramolecular ([44]2+–[46]2+) as well as intermolecular fashion ([6]2+ + fer- rocene). To broaden the scope of the bis(terpyridine)ruth- enium(II) chromophores in heterometallic assemblies, 2,2- bipyridine units have been attached either at the C- or at the N-terminus by using 4- or 5-amino-2,2-bipyridine and 2,2-bipyridine-4- or -5-carboxylic acid chloride and [4]2+ or [5]2+ as coupling partners, respectively, to obtain metallo ligands [49]2+–[52]2+ with bipyridine coordination sites (Figure 18).[77,78] The attachment of an additional bpy unit Figure 17. Ferrocene conjugates [44]2+–[46]2+ of [5]2+[59] and refer- ence compounds [47n]2+ and [48n]2+.[59,75] the NH-substituted ferrocene in [45]2+ and [46]2+ (ΔGET  0) than that from the CO-substituted ferrocene in [44]2+ by the difference in Fc/Fc+ oxidation potentials (ca. 0.27 eV, Table 1). Hence, reductive quenching seems to be a feasible deactivation pathway in [45]2+ and [46]2+ in ad- dition to triplet–triplet energy transfer to give ferrocene triplet states similar to amide-linked porphyrin–ferrocenyl dyads[122,123] and alkynyl-bridged ferrocenyl ruthenium(II) complexes [471]2+/[481]2+.[75c] Unfortunately, a meaningful comparison with the series [1]2+/[470]2+/[480]2+ is impossible because of the very low quantum yields and excited-state lifetimes of [1]2+/[470]2+ and [480]2+ in fluid solution (as dis- cussed above).[75b] Interestingly, excited-state mixed-valent FeIII/RuII species seem to be accessible either from direct excitation of the d(FcCO)π*(tpyNH) MLCT in [44]2+ and [46]2+ or through the ruthenium-based MLCT followed by reductive photoinduced electron transfer from FcNH in Figure 18. 2,2-Bipyridine conjugates [49]2+–[52]2+ of [5]2+ and [45]2+ and [46]2+. their heterometallic complexes.[77,78] Eur. J. Inorg. Chem. 2014, 5468–5490 5482 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 20 | 1 INTRODUCTION www.eurjic.org MICROREVIEW hardly affects the UV/Vis absorption properties of [49]2+– [52]2+ with respect to the parent bis(terpyridine)ruth- enium(II) complexes [5]2+ and [6]2+, only the π–π* transi- tions in the UV range are intensified as a result of the pres- ence of the bpy. Electrochemical studies reveal a reversible RuII/RuIII re- dox process at E1/2 = 0.68 V for amines [50]2+ and [52]2+ and at E1/2 = 0.90–0.92 V for esters [49]2+ and [51]2+, matching the data of [4]2+–[6]2+ (see above, Table 1). Additionally, several reduction waves are observed between E1/2 = –1.48 and E1/2 = –2.5 V. The first reductions are as- cribed to ligand-centered processes localized on the tpy-CO unit. This is supported by DFT-calculated spin densities and EPR spectroscopic signatures of radical species [49]+–[52]+ prepared by chemical reduction. All four metallo ligands [49]2+–[52]2+ are emissive at room temperature with emission wavelengths in typical regions (Table 1). Amino-substituted complexes [50]2+/[52]2+ emit at λmax = 735/727 nm, while amide-functionalized counterparts [49]2+/[51]2+ emit at λmax = 673/674 nm. Their emission quantum yields are very sim- ilar and in the range Φ = 0.15–0.20%.[77,78] The optical and redox data of bpy derivatives [49]2+– [52]2+ suggest that these complexes can be considered as chromophore- and redox-switch-functionalized metallo li- gands. Hence, typical bpy coordination chemistry is pos- sible with these phosphorescent, redox-active metallo li- gands. First studies were devoted to the coordination of these metallo ligands to rhenium(I) and platinum(II) com- plex fragments. Treating [49]2+, [51]2+, and [52]2+ with Re(CO)5Cl affords heterodinuclear complexes [53]2+–[55]2+ with Re(CO)3Cl fragments coordinated to the 2,2-bpy unit.[77] These Ru–Re complexes exhibit a new shoulder in their UV/Vis absorption spectra at around λmax = 350– 370 nm attributed to the 1MLCT transitions of the bipyr- idine rhenium(I) unit. A new oxidation wave is observed in the cyclic voltammograms of [53]2+–[55]2+ at E1/2 = 0.98 V, which is assigned to the ReI/ReII couple. The reduction po- tentials are essentially unaffected by coordination of the ad- ditional Re(CO)3Cl fragment. Phosphorescence is observed for all three heterodinuclear complexes with wavelengths and emission band shapes very similar to those of the par- ent metallo ligands. As emissive 3MLCT(Ru/pyridine) and 3MLCT(Re/pyridine) states are typically located in a com- Figure 19. DFT-calculated spin densities (B3LYP/LANL2DZ, parable energy range, a distinction based solely on energy IEFPCM, CH3CN; contour value 0.01; irrelevant CH hydrogen 3 2+ data is arguable; however, the striking resemblance of the atoms omitted) of the MLCT states of [53] and its contact ionpair {[53](PF6)}+ (a), and the radical [53]+ and its contact ion pair emission band shapes of mixed-metal complexes [53]2+, {[53](PF6)} (b). (c) The molecular structure of [51]2+ showing the [54]2+, and [55]2+ and the corresponding metallo ligands anion-amide hydrogen bonding interaction of the contact ion pair [49]2+, [51]2+, and [52]2+ strongly suggests a 3MLCT(Ru/ {[51](PF )}+6 in the solid state (CH hydrogen atoms omitted).[77] tpy)-based phosphorescence. In all cases, triplet–triplet en- ergy transfer from 3MLCT(Re/bpy) states to the slightly fa- While the first oxidation of [53]2+–[55]2+ to [53]3+–[55]3+ vored bis(terpyridine)ruthenium moiety is thermodynami- is unambiguously ruthenium-centered, the assignment of cally feasible. Indeed, DFT studies suggest a localization the first reduction site (tpy or bpy) is not straightforward. of the lowest-energy 3MLCT state on ruthenium and the Hence, the Re(bpy)(CO)3 moiety was experimentally in- carboxy-substituted terpyridine ligand for [54]2+ and [55]2+. terrogated by means of IR spectroscopic analysis. Indeed, In contrast to experimental results, DFT calculations lo- the CO stretching vibrations are only marginally affected calize the lowest 3MLCT state at the Re(bpy) site in [53]2+ by chemical reduction of [53]2+–[55]2+ to [53]+–[55]+, in (Figure 19a).[77] An explanation for this unexpected finding contrast to the significant shifts observed for the will be given below. Re(bpy)(CO) Cl/[Re(bpy)(CO) Cl]–3 3 redox couple resulting Eur. J. Inorg. Chem. 2014, 5468–5490 5483 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Section 1.1 | 21 www.eurjic.org MICROREVIEW from an increased Re-to-CO π-backbonding. This finding density from the bridging bpy-CO to the terminal tpy-CO rules out a significant localization of the odd electron at the ligand both in radical [53]+ and in the triplet excited state Re(bpy)(CO)3 site in [53]2+–[55]2+ and localizes the un- of [53]2+. This is an important finding, as Re(bpy)(L)3X paired electron in the proximity of the ruthenium center, complexes and chromophore-appended derivatives, for ex- namely at the tpy-CO ligand.[77] ample [56]3+ in Figure 20, have been reported to photocata- Interestingly, DFT calculations for radical [55]+ fully lytically reduce CO to CO via [Re(bpy)]· active sites.[125,126]2 agree with experimental findings, while for radical [53]+ the spin density is calculated at the Re(bpy) site (Figure 19b) and for [54]+ the reduction site even depends on the orienta- tion of the bpy.[77] Hence, the [Re(bpy-CO)]+ and [Ru(tpy- CO)]2+ moieties feature a similar electron affinity according to the calculations, and small perturbations, such as the po- sition of the counterions, might favor one valence isomer over the other. Indeed, the hexafluorophosphate counter- ions – neglected in all discussions so far – might play a decisive role. This is clearly seen already in the solid-state structures of metallo ligands [51](PF6)2 and [49](PF6)2· HPF6·2H2O, featuring contact ion pairs through NH···FPF5 hydrogen bonds (Figure 19c) and solvent-sepa- rated ion pairs through NH···OH2···FPF5 hydrogen bonds, respectively.[77] Such hydrogen bonds of the counterions to the central amide units might be even more important in heterobimetallic complexes because of the additional polar- ization by the second metal center. Indeed, for the [49]- (PF6)2/[53](PF6)2 pair, significant shifts of distinct proton resonances at the tpy-NH unit and the bpy-CO unit are observed (NH: Δδ = 2.28 ppm; CH: Δδ = 0.35–0.71 ppm; Figure 18, top; relevant hydrogen atoms highlighted in red) in addition to the expected coordination shift induced by the rhenium atom. These additional shifts are attributed to Figure 20. Heterometallic Ru/M reference compounds (M = Re, [126,127] the effect of CH···FPF5 and NH···FPF5 hydrogen bonds Pt). persisting even in solution. In DFT geometry optimizations of the contact ion pair {[53](PF +6)} , the explicit inclusion Metallation of metallo ligands [49]2+ and [50]2+ with of a (PF )–6 counterion in this binding pocket confirms PtCl2(dmso)2 gives heterodinuclear complexes [57]2+ and hydrogen bonds to these very CH and NH groups (shown [58]2+ in which the PtII(bpy)Cl2 fragment is positioned in red in Figure 18). In the radical contact ion pair either at the N- or at the C-terminus of amino acid [5]2+.[78] {[53](PF6)} including the counterion, the spin density is Upon coordination of the PtCl2 fragment, new absorption now shifted from the [Re(bpy-CO)·] unit to the terminal bands appear at λmax = 408 nm ([57]2+) and at λmax = [Ru(tpy-CO)·] unit according to the DFT calculations (Fig- 364 nm ([58]2+), assigned to platinum to bpy-CO and plati- ure 19b). The same holds for the localization of the lowest- num to bpy-NH MLCT bands [1MLCT(Pt/bpy)], respec- energy 3MLCT state, which is shifted from the [Re(bpy- tively.[78] CO)] to the [Ru(tpy-CO)] site by including the counterion In addition to the expected tpy-CO reductions at E1/2 = (Figure 19a). Hence, for these positively charged bimetallic –1.47 and –1.45 V and RuII/RuIII oxidation waves at E1/2 = complexes, the explicit inclusion of the counterion is essen- +0.93 and +0.69 V (Table 1), the cyclic voltammograms of tial to correctly reproduce the experimental data.[77] Con- [57]2+ and [58]2+ reveal additional reduction waves at E1/2 versely, the experimental exchange of the coordinating = –1.30 and –1.69 V, respectively. Both EPR spectroscopic (PF – – + +6) counterion by a non-coordinating (BPh4) counter- and DFT studies on radicals [57] and [58] confirm that ion leads to [Re(bpy-CO)·]-centered radicals as shown by the odd electron is localized on the bpy-CO unit in [57]+ the characteristic IR pattern of the [Re(bpy)·(CO)3Cl]– unit but on the tpy-CO unit in [58]+ (Figure 21). Specifically, the in [53]+ prepared by chemical reduction of [53](BPh4)2.[77] EPR signal pattern of [57]+ can only be explained properly Hence, the equilibrium between the [(EtOOC-tpy)·Ru(tpy- by taking a significant superhyperfine coupling to the plati- NHCO-bpy)Re(CO) Cl]+ and [(EtOOC-tpy)Ru(tpy- num nucleus (1953 Pt; I = 1/2; 33.8% natural abundance) into NHCO-bpy)·Re(CO)3Cl]+ valence isomers of [53]+, as well account [|A 1951,2,3( Pt)| = 68, 68, 23 G]. Furthermore, the g as the triplet-excited-state equilibrium between the values for [57]+ (g1,2,3 = 2.0290, 2.0049, 1.9400; Figure 21a) 3MLCT(Ru/tpy) and 3MLCT(Re/bpy) states of [53]2+, is are close to those found for genuine [Pt(bpy)Cl ]–2 radicals shifted by the coordinating properties of the counterion. [g1,2,3 = 2.0380, 2.0110, 1.9380; |A 1951,2,3( Pt)| = 61, 86, 22 G] Simply speaking, coordination of the counterion to the and distinct from those for ruthenium(II)-coordinated tpy- bridging amide pushes the partial negative charge and spin CO radicals (Figure 21b; and see above).[78] Eur. J. Inorg. Chem. 2014, 5468–5490 5484 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 22 | 1 INTRODUCTION www.eurjic.org MICROREVIEW quenchers (Table 1). Indeed, the excited states of metallo ligands [49]2+–[52]2+ as well as heterobimetallic complexes [53]2+–[58]2+ are quenched by triethanolamine (TEOA), which can, in principle, act as electron donor or proton ac- ceptor {E ·+ [125,128]1/2(TEOA/TEOA ) = 0.19 V, pKa[H- (TEOA)]+ = 7.76[81]/7.74[129]} towards the MLCT excited states. The Stern–Volmer constants K [124]SV of [49]2+–[58]2+ are not strictly correlated to the driving force for photoin- duced electron transfer, ΔGET, although some reasonable re- lations are obvious, for example the susceptibility of the ester-substituted complexes towards quenching is always higher than that of the amino-substituted complexes ([49]2+  [50]2+, [51]2+  [52]2+, [57]2+  [58]2+, [54]2+  [55]2+), which matches the ease of reduction and hence the driving force ΔG .[77,78]ET However, proton transfer to the amine, es- pecially when pre-coordinated to the amide unit through NH···N(CH2CH2OH)3 hydrogen bonds, might be a further feasible pathway (ESPT[89]). Indeed, excitation of the esters should yield the [(EtOOC-tpy)·RuIII(tpy-NHCO-bpy)- ML ]2+n state which should feature a polarized and more acidic NHCO group as a result of the RuIII center. In fact, the strongly polarized amide in the highly charged bis(ru- thenium) complex [40]4+ is already deprotonated in the ground state by triethylamine (see above). Furthermore, for the Ru–Pt complexes, non-linear Stern–Volmer plots[124] have been obtained, which suggests a pre-coordination of TEOA, possibly to the polarized amide unit (static quench- ing). Hence, TEOA might compete with the (PF –6) counter- ions for the “binding pocket” as described above (Fig- ure 19). This pre-coordination followed by ESPT is even re- sponsible for some photoinduced hydrolysis of [57]2+ to amine [4]2+ and acid Pt(bpy-COOH)Cl2 in the presence of water/TEOA.[78] Photocatalytic reduction of CO2 or H [37–39]2O by Ru–Re and Ru–Pt complexes [53]2+–[58]2+ to CO or H2 by using TEOA as sacrificial reductant, as reported by Ishitani and Figure 21. DFT-calculated spin densities (B3LYP/LANL2DZ, Sakai for several ruthenium/bipyridine-based bimetallic IEFPCM, CH CN; contour value 0.01; CH hydrogen atoms omit- complexes (Figure 20),[126,127]3 were unsuccessful. Even, ted), X-band EPR spectra (77 K, CH3CN, 9.42 GHz), and simula- complexes [53](BPh4)2 (Figure 19b) and [57](PF6)2 (Fig- tions of (a) [57]+ and (b) [58]+.[78] ure 21a), with a favorable bpy-centered reduction site, are catalytically incompetent. In the Ru–Re case, [53](BPh4)2 In spite of the presence of platinum, phosphorescence is this failure is ascribed to the insufficient reduction potential still observed for both Ru–Pt complexes [57]2+ and [58]2+ of the one-electron-reduced species [53]+ to reduce the in- with essentially unperturbed band shapes and band max- termediate Re–CO adduct.[77]2 For Ru–Pt-based proton re- ima, albeit with lower quantum yield. Hence, the lowest- duction, for example by [59]2+ or [60]2+ (Figure 20), charge- energy emissive states are best described as 3MLCT(Ru/ separated states RuIII(bpy-CONH-phen·)PtCl2 or RuIII- tpy), while the 3MLCT(Pt/bpy) states are possibly higher (phen-NHCO-bpy·)PtCl2 are believed to be responsible for in energy. However, they might be thermally accessible by photohydrogen production.[127] The corresponding RuIII- triplet–triplet energy transfer, which decreases the phospho- (tpy-NHCO-bpy·)PtCl2 charge-separated state of [57]2+, rescence quantum yield. Conversely, irradiation in the however, is thermodynamically uphill because of the com- 1MLCT(Pt/bpy) absorption band results in emission from parably low-energy 3MLCT state of the push-pull bis(terpy- the 3MLCT(Ru/tpy) state, which also suggests the accessi- ridine)ruthenium chromophore {[57]2+: λmax = 671 nm; bility of an intramolecular triplet–triplet energy transfer [Ru(bpy) ]2+3 : λmax = 615 nm; Table 1}. Furthermore, pathway. In this respect, Ru–Pt complexes [57]2+ and ESPT[89] and subsequent chemical reactions might ad- [58]2+ behave similarly to the ferrocenyl-appended ruth- ditionally impede photocatalytic reduction of carbon diox- enium amino acids [44]2+–[46]2+. ide or protons. Hence, the push/pull functional groups fa- The lifetime of the excited states of derivatives of [5]2+ is vorable for long excited-state lifetimes and phosphorescence long enough to enable bimolecular reactions with external also pave the way for undesired side reactions with sub- Eur. J. Inorg. Chem. 2014, 5468–5490 5485 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Section 1.1 | 23 www.eurjic.org MICROREVIEW strates in the excited state, especially ESPT to the reduced COOH groups in [63]2m+ anchor block-copolymer [63]2m+ tpy-CO unit (see above) or ESPT from the polarized tpy- to ZnO nanorods in a multipoint fashion to give the stable NH unit. Other photocatalytic applications circumventing non-aggregated nanocomposite [63]2m+@ZnO.[130] In con- coordinating and basic/acidic substrates and reductants are trast to [63]2m+, excitation of the [Ru(tpy) 2+2] chromophore currently under investigation. Photoinduced electron trans- in [63]2m+@ZnO by irradiation into its absorption band fer between electronically excited complexes based on [5]2+ (λmax = 498 nm, λexc = 488 nm) does not lead to phospho- and non-molecular acceptors (interfacial electron transfer) rescence. Instead, excited electrons are injected into the will be discussed in the next section. ZnO nanorods, and electron holes are generated in the tri- phenylamine-containing polymer block (Figure 22). Kelvin probe force microscopic (KPFM) studies of [63]2m+@ZnO 4.4 Photoinduced Electron Transfer at Interfaces reveal a significant difference of the surface potential of the In order to probe light-induced interfacial charge separa- polymer-coated nanorod in the dark and under irradiation tion on the nanoscale, bifunctional amino acid [5]2+ has into the MLCT band, which is assigned to oxidative been incorporated as chromophore in a donor–chromo- quenching of the excited chromophore, that is, charge injec- phore–acceptor nanocomposite.[130] A block-copolymer tion into ZnO and hence positive charging of the polymer with triphenylamine units in one block was used as electron (Figure 23).[130] Such semiconductor/chromophore/con- donor and the second block was equipped with the [5]2+ ducting polymer architectures are of particular interest for chromophore. ZnO nanorods were employed as electron ac- solid-state dye-sensitized solar cells (ssDSSCs).[131] ceptor. Experimentally, [5]2+ was converted into the ruth- enium-containing tripeptide [61]2+ by SPPS (Figure 12),[18] and the amino group of [61]2+ was attached to the Pfp- activated carboxylic acid of block-copolymer 62 through an amide bond to give [63]2m+ (Figure 22). The remaining Figure 23. Surface potential maps of [63]2m+@ZnO obtained by KPFM (a) in the dark and (b) under irradiation with λexc = 488 nm.[130] 5. Applications 5.1 Dye-Sensitized Solar Cells COOH-substituted complexes [5]2+, [25]2+, [26]2+, [27]2+, as well as 2,2-bipyridine-substituted complexes [49]2+ and [50]2+ (Figures 2, 11, 18, and 24) have been adsorbed onto nanostructured TiO2 and employed as sensitizers in stan- dard DSSCs with an I–/I –3 redox electrolyte.[72] All com- plexes feature absorption spectra similar to that of the stan- dard ruthenium(II) sensitizer N719 with [27]2+, and they even have a somewhat stronger absorption in the NIR re- gion. Optical inspection of the loaded FTO/TiO2 electrode already reveals that the 2,2-bipyridine anchors in [49]2+ and [50]2+ are inferior to carboxylate linkers (Figure 24a), likely because these large functional groups require more space and hence the dye loading with [49]2+ and [50]2+ is significantly reduced (1.0–2.6 10–8 molcm–2) relative to those with the other sensitizers [5]2+, [25]2+, [26]2+, [27]2+, and N719 (5.6–11 10–8 molcm–2).[72] In spite of the sim- ilar loadings, cells with [5]2+, [25]2+, [26]2+, and [27]2+ de- liver only low cell-power conversion efficiencies (η = 0.13– 0.26%) relative to those with N719 (η = 5.03 %). This has Figure 22. Assembly of donor–chromophore–acceptor nanoc- omposite [63]2m+@ZnO by coupling of tripeptide [61]2+ to block- been traced back to the poor short-circuit current, the large copolymer 62 (n ≈ 50, m ≈ 10) and coating of ZnO nanorods with dark current, and the high electron Ti(e–)/I2 recombination functionalized block-copolymer [63]2m+.[130] rate of the cells based on [5]2+, [25]2+, [26]2+, and [27]2+. Eur. J. Inorg. Chem. 2014, 5468–5490 5486 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 24 | 1 INTRODUCTION www.eurjic.org MICROREVIEW This might be due to the twofold positive charge of the somewhat enhances the external quantum efficiencies as a complexes, which increases the I –3 /I2 concentration near the result of diminished excited-state deactivation by radiation- TiO2 electrode as a result of electrostatic interactions (ion less processes.[134] pairing and hydrogen bonding, see above) favoring electron recombination.[132] In contrast, N719 is twofold negatively charged, which prevents fast recombination with the elec- trolyte.[72] The positive charge of bis(tridentate) complexes of ruthenium(II) such as [5]2+, [25]2+, [26]2+, and [27]2+ might be reduced in the future by introduction of negatively charged chelating ligands. Indeed, ruthenium(II) complexes with pyrazolato or cyclometalating ligands have been re- Figure 25. Photographs of LECs with ITO/PEDOT:PSS/ruth- [134] ported to give DSSCs with high power conversion efficienc- enium(II) dye/Ag structure. ies up to η = 10.7%.[96,133] Another approach would be to employ other electrolytes instead of the classical I–/I –3 cou- ple, such as positively charged metal complexes, for example 2+/3+ [63] Conclusions[Co{4,4-(tBu)2bpy}3] . The use of amino acids as building blocks of well-defined arrays is certainly one of the most successful ideas nature has come up with. The versatile heteroleptic, push-pull-sub- stituted ruthenium(II) amino acid [5]2+ constitutes an im- portant member of a growing class of metallo amino acids, such as 1,1-ferrocene amino acid[119,137] or biferrocene amino acid,[138] based on ferrocenes or metallo porphyrin amino acids based on porphyrins.[122,123,139] Such metallo amino acids expand our pool of useful building blocks with specialized properties not covered by organic amino acids. Figure 24. (a) Photographs of TiO on FTO impregnated with the Ruthenium(II) amino acid [5] 2+ is readily prepared from 2 indicated adsorbed ruthenium sensitizers and (b) photographs of RuCl3 and the 4-substituted terpyridine ligands 4-tpy- the corresponding DSSCs with FTO/TiO2/sensitizer/electrolyte/Pt/ COOEt and 4-tpy-NH2 in isomerically pure form (tpy = FTO structure (FTO = fluorine-doped tin oxide).[72] 2,2;6,2-terpyridine).[59] Its exceptionally rich redox, acid/ base, and photochemical properties are now well under- stood, namely metal-based oxidation, ligand-centered re- 5.2 Light-Emitting Electrochemical Cells duction, ground-state protonation and deprotonation, pro- ton-coupled electron transfer, excited-state dynamics, phos- Charged bis(tridentate) complexes [4]2+, [21]2+, and phorescence, oxidative and reductive quenching, excited- [22]2+ have been utilized as emitters in light-emitting elec- state proton transfer, and triplet–triplet energy transfer. trochemical cells (LECs).[134] In comparison to organic As is characteristic for amino acids, the orthogonal reac- light-emitting diodes (OLEDs), LECs feature a much sim- tivity at the C- and N-terminal sites of [5]2+ enables highly pler device structure and are hence less difficult to pre- selective transformations and precise incorporation in pare.[5–8] A device composition of ITO/PEDOT:PSS/ruth- larger peptide architectures. Even solid-phase peptide syn- enium(II) complex/Ag has been used [ITO = indium tin ox- thesis protocols are applicable with only minor modifica- ide, PEDOT = 3,4-ethylenedioxythiophene, PSS = poly- tions.[18,140] (styrenesulfonate), Figure 25]. The thickness of the PEDOT/ The special electrochemical and optical properties of PSS and ruthenium complex layers were characterized by push-pull-substituted complex [5]2+ are highly useful for ap- atomic force microscopy. Upon applying moderate poten- plications in photochemical, photophysical, and redox tials, emission up to a maximum emission wavelength of chemical contexts. Therefore, the versatile building block 722–755 nm is achieved. The CIE coordinates[135] of the [5]2+ has been successfully incorporated into molecular electroluminescence of [22]2+ are x = 0.731 and y = 0.269, and nanoscale energy- and electron-transfer sys- which corresponds to a deep red emission. To the best of tems[18,25,59,77,78,130] as well as in first applications of low- our knowledge, the observed electroluminescence features cost lighting devices (LECs)[134] and solar energy conver- the lowest emission energy for LECs containing bis(tri- sion (DSSCs).[72] Future applications will involve sensing dentate) ruthenium(II) complexes so far.[7,8,136] In fact, of small molecules that can switch the phosphorescence of most of the emission occurs in the near infrared, invisible suitable sensors based on [5]2+. to the human eye.[134] For such a low emission energy, the In terms of excited-state properties (emission quantum energy gap law[79,101–103] predicts enhanced radiationless de- yield, excited state lifetime), a successful tpy ligand varia- activation of the excited state, which explains the compara- tion is the expansion of the small N–Ru–N bite angle by tively small external quantum efficiencies. However, diluting formally inserting a N–CH3 fragment between the chelating the ruthenium(II) complexes in poly(methyl methacrylate) pyridines of tpy to give ddpd-based amino acid derivatives Eur. J. Inorg. Chem. 2014, 5468–5490 5487 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Section 1.1 | 25 www.eurjic.org MICROREVIEW [23]2+, [24]2+, and [27]2+ (ddpd = N,N-dimethyl-N,N-di- [18] K. Heinze, K. Hempel, Chem. Eur. J. 2009, 15, 1346–1358. pyridin-2-ylpyridine-2,6-diamine).[56,100] In photophysical [19] R. Horvath, J. Lombard, J.-C. Leprêtre, M.-N. Collomb, A. Deronizer, J. Chauvin, K. C. Gordon, Dalton Trans. 2013, 42, respects, extension of the tpy ligands by phenylene groups 16527–16537. at the 4-positions ([11]2+–[13]2+) is rather ineffective [4-(4- [20] S. Roeser, M. Z. Ertem, C. Cady, R. Lomoth, J. Benet-Buch- NH2–C6H4)-tpy, 4-(4-ROOC–C6H4)-tpy],[60] and other li- holz, L. Hammarström, B. Sarkar, W. Kaim, C. J. Cramer, A. gand extensions might be envisaged in future work. Llobet, Inorg. Chem. 2012, 51, 320–327. In essence, the present report summarizes the rich chem- [21] L.-Z. Sui, W.-W. Yang, C.-J. Yao, H.-Y. Xie, Y.-W. Zhong, In- 2+ org. Chem. 2012, 51, 1590–1598.istry of the ruthenium(II) amino acid [5] and its deriva- [22] O. S. Wenger, Chem. Soc. Rev. 2012, 41, 3772–3779. tives, demonstrating that this research is an interdisciplinary [23] H.-J. Nie, X. Chen, C.-J. Yao, Y.-W. Zhong, G. R. Hutchison, field comprising ligand design, coordination chemistry, J. Yao, Chem. Eur. J. 2012, 18, 14497–14509. peptide chemistry, redox chemistry, photochemistry, and [24] C.-J. Yao, Y.-W. Zhong, J. Yao, Inorg. Chem. 2013, 52, 4040– materials science. It is hoped that further variations and 4045.[25] A. Breivogel, K. Hempel, K. Heinze, Inorg. Chim. Acta 2011, optimizations of the basic design concept will further ex- 374, 152–162. pand the utility of metallo amino acids in general and up- [26] P. Ceroni, Chem. Eur. J. 2011, 17, 9560–9564. grade the useful redox and optical properties of [5]2+ in par- [27] T. N. Singh-Rachford, F. N. Castellano, Coord. Chem. Rev. ticular while maintaining its high chemical flexibility. In our 2010, 254, 2560–2573. opinion there is still a lot of room for improvement and [28] W. Wu, S. Ji, W. Wu, J. Shao, H. Guo, T. D. James, J. Zhao,Chem. Eur. J. 2012, 18, 4953–4964. development in this highly interdisciplinary field. [29] P. Hanczyc, B. Norden, M. Samoc, Dalton Trans. 2012, 41, Supporting Information (see footnote on the first page of this arti- 3123–3125. cle): Synthesis and analytical data of tetrapeptides [36](PF ) and [30] P. Ceroni, G. Bergamini, V. Balzani, Angew. Chem. Int. Ed.6 2 [37](PF ) . 2009, 48, 8516–8518; Angew. Chem. 2009, 121, 8668–8670.6 2 [31] J. Lee, H. Chang, S. Kim, G. S. Bang, H. Lee, Angew. Chem. Int. Ed. 2009, 48, 8501–8504; Angew. Chem. 2009, 121, 8653– Acknowledgments 8656. [32] K. Terada, K. Kanaizuka, V. M. Iyer, M. Sannodo, S. Saito, K. Kobayashi, M.-a. Haga, Angew. Chem. Int. Ed. 2011, 50, This work was supported by the International Research Training 6287–6291; Angew. Chem. 2011, 123, 6411–6415. Group “Self Organized Materials for Optoelectronics” funded by [33] R. H. Crabtree, Energy Production and Storage: Inorganic the Deutsche Forschungsgemeinschaft (DFG) (IRTG 1404). Finan- Chemical Strategies for a Warming World, Wiley-VCH, cial support through the Excellence Initiative (DFG/GSC 266) is Weinheim, 2010. acknowledged by C. K. [34] X. Liu, F. Wang, Coord. Chem. Rev. 2012, 256, 1115–1136. [35] W. T. Eckenhoff, R. Eisenberg, Dalton Trans. 2012, 41, 13004– [1] S. Campagna, F. Puntoriero, F. Nastasi, G. Bergamini, V. Balz- 13021. ani, Top. Curr. Chem. 2007, 280, 117–214. [36] T. Stoll, M. Gennari, I. Serrano, J. Fortage, J. Chauvin, F. Odo- [2] B. O’Regan, M. Grätzel, Nature 1991, 353, 737–740. bel, M. Rebarz, O. Poizat, M. Sliwa, A. Deronizer, M.-N. Col- [3] J.-F. Yin, M. Velayudham, D. Bhattacharya, H.-C. Lin, K.-L. lomb, Chem. Eur. J. 2013, 19, 782–792. Lu, Coord. Chem. Rev. 2012, 256, 3008–3035. [37] M. Schulz, M. Karnahl, M. Schwalbe, J. G. Vos, Coord. Chem. [4] P. G. Bomben, K. C. D. Robson, B. D. Koivisto, C. P. Berling- Rev. 2012, 256, 1682–1705. uette, Coord. Chem. Rev. 2012, 256, 1438–1450. [38] A. Inagaki, M. Akita, Coord. Chem. Rev. 2010, 254, 1220–1239. [5] R. D. Costa, E. Ortí, H. J. Bolink, F. Monti, G. Accorsi, N. [39] Z. Chen, P. Kang, M.-T. Zhang, T. J. Meyer, Chem. Commun. Armaroli, Angew. Chem. Int. Ed. 2012, 51, 8178–8211; Angew. 2014, 50, 335–337. Chem. 2012, 124, 8300–8334. [40] S. Ohzu, T. Ishizuka, Y. Hirai, S. Fukuzumi, T. Kojima, Chem. [6] P. Dreyse, B. Loeb, M. Soto-Arriaza, D. Tordera, E. Ortí, J. J. Eur. J. 2013, 19, 1563–1567. Serrano-Pérez, H. J. Bolink, Dalton Trans. 2013, 42, 15502– [41] H. Nitadori, T. Takahashi, A. Inagaki, M. Akita, Inorg. Chem. 15513. 2012, 51, 51–62. [7] H. J. Bolink, E. Coronado, R. D. Costa, P. Gaviña, E. Ortí, S. [42] P. Guillo, O. Hamelin, P. Batat, G. Jonusauskas, N. D. McClen- Tatay, Inorg. Chem. 2009, 48, 3907–3909. aghan, S. Ménage, Inorg. Chem. 2012, 51, 2222–2230. [8] H. J. Bolink, L. Cappelli, E. Coronado, P. Gaviña, Inorg. Chem. [43] F. Li, M. Yu, Y. Jiang, F. Huang, Y. Li, B. Zhang, L. Sun, 2005, 44, 5966–5968. Chem. Commun. 2011, 47, 8949–8951. [9] L. Marcélis, J. Ghesquière, K. Garnir, A. K.-D. Mesmaeker, C. [44] D. Chao, W.-F. Fu, Dalton Trans. 2014, 43, 306–310. Moucheron, Coord. Chem. Rev. 2012, 256, 1569–1582. [45] C. K. Prier, D. A. Rankic, D. W. C. MacMillan, Chem. Rev. [10] M. R. Gill, J. A. Thomas, Chem. Soc. Rev. 2012, 41, 3179–3192. 2013, 113, 5322–5363. [11] H. Komatsu, K. Yoshihara, H. Yamada, Y. Kimura, A. Son, [46] A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser, A. S.-i. Nishimoto, K. Tanabe, Chem. Eur. J. 2013, 19, 1971–1977. von Zelewsky, Coord. Chem. Rev. 1988, 84, 85–277. [12] D. Maity, C. Bhaumik, D. Mondal, S. Baitalik, Dalton Trans. [47] A. Reynal, E. Palomares, Eur. J. Inorg. Chem. 2011, 4509–4526. 2014, 43, 1829–1845. [48] a) C. Fu, M. Wenzel, E. Treutlein, K. Harms, E. Meggers, In- [13] Z.-B. Zheng, Z.-M. Duan, Y.-Y. Ma, K.-Z. Wang, Inorg. Chem. org. Chem. 2012, 51, 10004–10011; b) E. Meggers, Chem. Eur. 2013, 52, 2306–2316. J. 2010, 16, 752–758; c) E. Meggers, Eur. J. Inorg. Chem. 2011, [14] M. Li, Q. Liang, M. Zheng, C. Fang, S. Peng, M. Zhao, Dalton 2911–2926; d) T. Joshi, V. Pierroz, C. Mari, L. Gemperle, S. Trans. 2013, 42, 13509–13515. Ferrari, G. Gasser, Angew. Chem. Int. Ed. 2014, 53, 2960–2963; [15] S. A. Poteet, F. M. MacDonnell, Dalton Trans. 2013, 42, Angew. Chem. 2014, 126, 3004–3007. 13305–13307. [49] a) A. C. Lees, C. J. Kleverlaan, C. A. Bignozzi, J. G. Vos, Inorg. [16] H. Chen, X. Li, Y. Wu, W. Gao, R. Bai, Dalton Trans. 2012, Chem. 2001, 40, 5343–5349; b) D. M. D’Alessandro, F. R. 41, 13292–13297. Keene, S. D. Bergman, M. Kol, Dalton Trans. 2005, 332–337; [17] R. Zhang, Z. Ye, G. Wang, W. Zhang, J. Yuan, Chem. Eur. J. c) S. Bodige, A. S. Torres, D. J. Maloney, D. Tate, G. R. Kinsel, 2010, 16, 6884–6891. A. K. Walker, F. M. MacDonnell, J. Am. Chem. Soc. 1997, 119, Eur. J. Inorg. Chem. 2014, 5468–5490 5488 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 26 | 1 INTRODUCTION www.eurjic.org MICROREVIEW 10364–10369; d) E. Baggaley, M. R. Gill, N. H. Green, D. Tur- [73] G. D. Storrier, S. B. Colbran, Inorg. Chim. Acta 1999, 284, 76– ton, I. V. Sazanovich, S. W. Botchway, C. Smythe, J. W. Hay- 84. cock, J. A. Weinstein, J. A. Thomas, Angew. Chem. Int. Ed. [74] T.-a. Okamura, T. Iwamura, S. Seno, H. Yamamoto, N. Uey- 2014, 53, 3367–3371; Angew. Chem. 2014, 126, 3435–3439. ama, J. Am. Chem. Soc. 2004, 126, 15972–15973. [50] M. Maestri, N. Armaroli, V. Balzani, E. C. Constable, [75] a) B. Farlow, T. A. Nile, J. L. Walsh, A. T. McPhail, Polyhedron A. M. W. Cargill Thompson, Inorg. Chem. 1995, 34, 2759– 1993, 12, 2891–2894; b) K. Hutchison, J. C. Morris, T. A. Nile, 2767. J. L. Walsh, D. W. Thompson, J. D. Petersen, J. R. Schoonover, [51] J.-P. Collin, P. Lainé, J.-P. Launay, J.-P. Sauvage, A. Sour, J. Inorg. Chem. 1999, 38, 2516–2523; c) U. Siemeling, J. Chem. Soc., Chem. Commun. 1993, 434–435. Vor der Brüggen, U. Vorfeld, B. Neumann, A. Stammler, H.- [52] E. C. Constable, M. D. Ward, J. Chem. Soc., Dalton Trans. G. Stammler, A. Brockhinke, R. Plessow, P. Zanello, F. Laschi, 1990, 1405–1409. F. Fabrizi de Biani, M. Fontani, S. Steenken, M. Stapper, G. [53] a) R. Hogg, R. G. Wilkins, J. Chem. Soc. 1962, 341–350; b) Gurzadyan, Chem. Eur. J. 2003, 9, 2819–2833. R. H. Holyer, C. D. Hubbard, S. F. A. Kettle, R. G. Wilkins, [76] a) A. Farmilo, F. Wilkinson, Chem. Phys. Lett. 1975, 34, 575– Inorg. Chem. 1966, 5, 622–625. 580; b) M. S. Wrighton, L. Pdungsap, D. L. Morse, J. Phys. [54] a) J.-P. Sauvage, J.-P. Collin, J.-C. Chambron, S. Guillerez, C. Chem. 1975, 79, 66–71; c) R. Ciasson, E. J. Lee, X. Zhao, M. S. Coudret, V. Balzani, F. Barigelletti, L. D. Cola, L. Flamigni, Wrighton, J. Phys. Chem. 1993, 97, 2596–2601. Chem. Rev. 1994, 94, 993–1019; b) Y. Liu, R. Hammitt, D. A. [77] J. Dietrich, U. Thorenz, C. Förster, K. Heinze, Inorg. Chem. Lutterman, R. P. Thummel, C. Turro, Inorg. Chem. 2007, 46, 2013, 52, 1248–1264. 6011–6021; c) L. Hammarström, O. Johansson, Coord. Chem. [78] J. Dietrich, A. Wünsche von Leupoldt, M. Grabolle, U. Resch- Rev. 2010, 254, 2546–2559. Genger, K. Heinze, Eur. J. Inorg. Chem. 2013, 3009–3019. [55] a) E. A. Medlycott, G. S. Hanan, Coord. Chem. Rev. 2006, 250, [79] J. V. Caspar, T. J. Meyer, J. Am. Chem. Soc. 1983, 105, 5583– 1763–1782; b) E. A. Medlycott, G. S. Hanan, Chem. Soc. Rev. 5590. 2005, 34, 133–142. [80] H. Hofmeier, U. S. Schubert, Chem. Soc. Rev. 2004, 33, 373– [56] A. Breivogel, M. Meister, C. Förster, F. Laquai, K. Heinze, 399. Chem. Eur. J. 2013, 19, 13745–13760. [81] U. S. Schubert, H. Hofmeier, G. R. Newkome, Modern Terpyri- [57] K. Suzuki, A. Kobayashi, S. Kaneko, K. Takehira, T. Yoshi- dine Chemistry, Wiley-VCH, Weinheim, 2006. hara, H. Ishida, Y. Shiina, S. Oishi, S. Tobita, Phys. Chem. [82] D. Xiong, Z. Li, H. Wang, J. Wang, Green Chem. 2013, 15, Chem. Phys. 2009, 11, 9850–9860. 1941–1948. [58] a) B. D. Koivisto, K. C. D. Robson, C. P. Berlinguette, Inorg. [83] a) A. Habibi-Yangjeh, M. Danandeh-Jenagharad, M. Noosh- Chem. 2009, 48, 9644–9652; b) V. Duprez, M. Biancardo, F. C. yar, Bull. Korean Chem. Soc. 2005, 26, 2007–2016; b) Y. Kobay- Krebs, Sol. Energy Mater. Sol. Cells 2007, 91, 230–237. ashi, T. Komatsu, M. Sumi, S. Numajiri, M. Miyamoto, D. [59] K. Heinze, K. Hempel, M. Beckmann, Eur. J. Inorg. Chem. Kobayashi, K. Sugibayashi, Y. Morimoto, Eur. J. Pharm. Sci. 2006, 2040–2050. 2004, 21, 471–477. [60] K. Heinze, K. Hempel, A. Breivogel, Z. Anorg. Allg. Chem. [84] B. García, S. Ibeas, J. M. Leal, J. Phys. Org. Chem. 1996, 9, 2009, 635, 2541–2549. 593–597. [61] K. Heinze, K. Hempel, S. Tschierlei, M. Schmitt, J. Popp, S. [85] R. Schwesinger, J. Willaredt, H. Schlemper, M. Keller, D. Rau, Eur. J. Inorg. Chem. 2009, 3119–3126. Schmitt, H. Fritz, Chem. Ber. 1994, 127, 2435–2454. [62] C. Kreitner, K. Heinze, to be submitted. [86] M. D. Kärkäs, T. Åkermark, H. Chen, J. Sun, B. Åkermark, [63] S. Caramori, J. Husson, M. Beley, C. A. Bignozzi, R. Argazzi, Angew. Chem. Int. Ed. 2013, 52, 4189–4193; Angew. Chem. P. C. Gros, Chem. Eur. J. 2010, 16, 2611–2618. 2013, 125, 4283–4287. [64] K. C. D. Robson, B. D. Koivisto, T. J. Gordon, T. [87] N. G. Connelly, W. E. Geiger, Chem. Rev. 1996, 96, 877–910. Baumgartner, C. P. Berlinguette, Inorg. Chem. 2010, 49, 5335– [88] a) G. Wolfbauer, A. M. Bond, D. R. MacFarlane, Inorg. Chem. 5337. 1999, 38, 3836–3846; b) S. Kämper, A. Paretzki, J. Fiedler, S. [65] a) Y.-Q. Fang, N. J. Taylor, G. S. Hanan, F. Loiseau, R. Passal- Záliš, W. Kaim, Inorg. Chem. 2012, 51, 2097–2104. acqua, S. Campagna, H. Nierengarten, A. V. Dorsselaer, J. Am. [89] a) L. G. Arnaut, S. J. Formosinho, J. Photochem. Photobiol. A: Chem. Soc. 2002, 124, 7912–7913; b) Y.-Q. Fang, N. J. Taylor, Chem. 1993, 75, 1–20; b) C. Turro, S. H. Bossmann, Y. Jenkins, F. Laverdière, G. S. Hanan, F. Loiseau, F. Nastasi, S. Cam- J. K. Barton, N. J. Turro, J. Am. Chem. Soc. 1995, 117, 9026– pagna, H. Nierengarten, E. Leize-Wagner, A. V. Dorsselaer, In- 9032. org. Chem. 2007, 46, 2854–2863. [90] G. R. Desiraju, Angew. Chem. Int. Ed. Engl. 1995, 34, 2311– [66] a) A. Harriman, M. Hissler, A. Khatyr, R. Ziessel, Chem. Com- 2327; Angew. Chem. 1995, 107, 2541–2558. mun. 1999, 8, 735–736; b) M. Hissler, A. Harriman, A. Khatyr, [91] S. Ji, W. Wu, W. Wu, H. Guo, J. Zhao, Angew. Chem. Int. Ed. R. Ziessel, Chem. Eur. J. 1999, 5, 3366–3381; c) A. C. Bennis- 2011, 50, 1626–1629; Angew. Chem. 2011, 123, 1664–1667. ton, A. Harriman, D. J. Lawrie, A. Mayeux, Phys. Chem. [92] a) J. Wang, G. S. Hanan, F. Loiseau, S. Campagna, Chem. Chem. Phys. 2004, 6, 51–57. Commun. 2004, 2068–2069; b) J. Wang, Y.-Q. Fang, L. Bourget- [67] D. G. Brown, N. Sanguantrakun, B. Schulze, U. S. Schubert, Merle, M. I. J. Polson, G. S. Hanan, A. Juris, F. Loiseau, S. C. P. Berlinguette, J. Am. Chem. Soc. 2012, 134, 12354–12357. Campagna, Chem. Eur. J. 2006, 12, 8539–8548; c) E. A. Medly- [68] M. Abrahamsson, M. Jäger, T. Österman, L. Eriksson, P. cott, G. S. Hanan, F. Loiseau, S. Campagna, Chem. Eur. J. Persson, H.-C. Becker, O. Johansson, L. Hammarström, J. Am. 2007, 13, 2837–2846. Chem. Soc. 2006, 128, 12616–12617. [93] Y. Chi, P.-T. Chou, Chem. Soc. Rev. 2010, 39, 638–655. [69] G. A. Parada, L. A. Fredin, M.-P. Santoni, M. Jäger, R. Lom- [94] V. Friese, S. Nag, J. Wang, M.-P. Santoni, A. Rodrigue-Witchel, oth, L. Hammarström, O. Johansson, P. Persson, S. Ott, Inorg. G. S. Hanan, F. Schaper, Eur. J. Inorg. Chem. 2011, 39–44. Chem. 2013, 52, 5128–5137. [95] W. Ghattas, H. Müller-Bunz, M. Albrecht, Organometallics [70] M. Abrahamsson, M. Jäger, R. J. Kumar, T. Österman, P. 2010, 29, 6782–6789. Persson, H.-C. Becker, O. Johansson, L. Hammarström, J. Am. [96] K. C. D. Robson, B. D. Koivisto, A. Yella, B. Sporinova, M. K. Chem. Soc. 2008, 130, 15533–15542. Nazeeruddin, T. Baumgartner, M. Grätzel, C. P. Berlinguette, [71] F. Schramm, V. Meded, H. Fliegl, K. Fink, O. Fuhr, Z. Qu, W. Inorg. Chem. 2011, 50, 5494–5508. Klopper, S. Finn, T. E. Keyes, M. Ruben, Inorg. Chem. 2009, [97] a) F. Barigelletti, B. Ventura, J.-P. Collin, R. Kayhanian, P. Ga- 48, 5677–5684. viña, J.-P. Sauvage, Eur. J. Inorg. Chem. 2000, 113–119; b) S. H. [72] A. Breivogel, S. Wooh, J. Dietrich, T. Y. Kim, Y. S. Kang, K. Wadman, J. M. Kroon, K. Bakker, R. W. A. Havenith, G. P. M. Char, K. Heinze, Eur. J. Inorg. Chem. 2014, 2720–2734. van Klink, G. van Koten, Organometallics 2010, 29, 1569–1579; Eur. J. Inorg. Chem. 2014, 5468–5490 5489 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Section 1.1 | 27 www.eurjic.org MICROREVIEW c) K. C. D. Robson, B. D. Koivisto, C. P. Berlinguette, Inorg. 36, 920–921; c) H. Ozawa, K. Sakai, Chem. Commun. 2011, Chem. 2012, 51, 1501–1507; d) B. Schulze, D. Escudero, C. Fri- 47, 2227–2242. ebe, R. Siebert, H. Görls, S. Sinn, M. Thomas, S. Mai, J. Popp, [128] J. Hawecker, J.-M. Lehn, R. Ziessel, Helv. Chim. Acta 1986, B. Dietzek, L. González, U. S. Schubert, Chem. Eur. J. 2012, 69, 1990–2012. 18, 4010–4025; e) S. Sinn, B. Schulze, C. Friebe, D. G. Brown, [129] M. R. Simond, K. Ballerat-Busserolles, Y. Coulier, L. Rodier, M. Jäger, E. Altuntaş, J. Kübel, O. Guntner, C. P. Berlinguette, J.-Y. Coxam, J. Solution Chem. 2012, 41, 130–142. B. Dietzek, U. S. Schubert, Inorg. Chem. 2014, 53, 2083–2095. [130] L. zur Borg, A. L. Domanski, A. Breivogel, M. Bürger, R. [98] T. Österman, M. Abrahamsson, H.-C. Becker, L. Ham- Berger, K. Heinze, R. Zentel, J. Mater. Chem. C 2013, 1, marström, P. Persson, J. Phys. Chem. A 2012, 116, 1041–1050. 1223–1230. [99] M. Jäger, R. J. Kumar, H. Görls, J. Bergquist, O. Johansson, [131] a) M. Matsumoto, H. Miyazaki, K. Matsuhiro, Y. Kuma- Inorg. Chem. 2009, 48, 3228–3238. shiro, Y. Takaoka, Solid State Ionics 1996, 89, 263–267; b) [100] A. Breivogel, C. Förster, K. Heinze, Inorg. Chem. 2010, 49, A. F. Nogueira, J. R. Durrant, M. A. De Paoli, Adv. Mater. 7052–7056. 2001, 13, 826–830. [101] T. J. Meyer, Pure Appl. Chem. 1986, 58, 1193–1206. [132] a) B. C. O’Regan, K. Walley, M. Juozapavicius, A. Anderson, [102] J. V. Caspar, E. M. Kober, B. P. Sullivan, T. J. Meyer, J. Am. F. Matar, T. Ghaddar, S. M. Zakeeruddin, C. Klein, J. R. Chem. Soc. 1982, 104, 630–632. Durrant, J. Am. Chem. Soc. 2009, 131, 3541–3548; b) B. C. [103] J. V. Caspar, T. J. Meyer, J. Phys. Chem. 1983, 87, 952–957. O’Regan, I. López-Duarte, M. V. Martínez-Díaz, A. Forneli, [104] R. J. Watts, J. Chem. Educ. 1983, 60, 834–842. J. Albero, A. Morandeira, E. Palomares, T. Torres, J. R. Durr- [105] B. Durham, J. V. Caspar, J. K. Nagle, T. J. Meyer, J. Am. ant, J. Am. Chem. Soc. 2008, 130, 2906–2907; c) J. M. Gard- Chem. Soc. 1982, 104, 4803–4810. ner, J. M. Giaimuccio, G. J. Meyer, J. Am. Chem. Soc. 2008, [106] Z. J. Fuller, W. D. Bare, K. A. Kneas, W.-Y. Xu, J. N. Demas, 130, 17252–17253. B. A. DeGraff, Anal. Chem. 2003, 75, 2670–2677. [133] a) C.-C. Chou, K.-L. Wu, Y. Chi, W.-P. Hu, S. J. Yu, G.-H. [107] W. M. Wallace, P. E. Hoggard, Inorg. Chem. 1979, 18, 2934– Lee, C.-L. Lin, P.-T. Chou, Angew. Chem. Int. Ed. 2011, 50, 2935. 2054–2058; Angew. Chem. 2011, 123, 2102–2106; b) C.-C. [108] T. E. Keyes, E. Gicquel, L. Guerin, R. J. Forster, V. Hultgren, Chou, F.-C. Hu, H.-H. Yeh, H.-P. Wu, Y. Chi, J. N. Clifford, A. M. Bond, A. G. Wedd, Inorg. Chem. 2003, 42, 7897–7905. E. Palomares, S.-H. Liu, P.-T. Chou, G.-H. Lee, Angew. [109] E. Bayer, Angew. Chem. Int. Ed. Engl. 1991, 30, 113–129; An- Chem. Int. Ed. 2014, 53, 178–183; Angew. Chem. 2014, 126, gew. Chem. 1991, 103, 117–133. 182–187; c) K.-L. Wu, C.-H. Li, Y. Chi, J. N. Clifford, L. [110] M. Grøtli, C. H. Gotfredsen, J. Rademann, J. Buchardt, A. J. Cabau, E. Palomares, Y.-M. Cheng, H.-A. Pan, P.-T. Chou, Clark, J. Ø. Duus, M. Meldal, J. Comb. Chem. 2000, 2, 108– J. Am. Chem. Soc. 2012, 134, 7488–7496. 119. [134] A. Breivogel, M. Park, D. Lee, S. Klassen, A. Kühnle, C. Lee, [111] W. C. Chan, P. D. White (Eds.), Fmoc Solid Phase Peptide K. Char, K. Heinze, Eur. J. Inorg. Chem. 2014, 288–295. Synthesis Oxford University Press, Oxford, 2000. [135] International Commission on Illumination (CIE), 1931. [112] A. Barbieri, B. Ventura, R. Ziessel, Coord. Chem. Rev. 2012, [136] W. Y. Ng, X. Gong, W. K. Chan, Chem. Mater. 1999, 11, 256, 1732–1741. 1165–1170. [113] A. Barbieri, B. Ventura, F. Barigelletti, A. D. Nicola, M. Que- [137] a) J. R. Butler, S. C. Quayle, J. Organomet. Chem. 1998, 552, sada, R. Ziessel, Inorg. Chem. 2004, 43, 7359–7368. 63–68; b) T. Okamura, K. Sakauye, N. Ueyama, A. Naka- [114] a) M. B. Robin, P. Day, Adv. Inorg. Chem. 1968, 10, 247–422; mura, Inorg. Chem. 1998, 37, 6731–6736; c) K. Heinze, M. b) N. S. Hush, Prog. Inorg. Chem. 1967, 8, 391–444. Schlenker, Eur. J. Inorg. Chem. 2004, 2974–2988; d) K. [115] N. Hush, Coord. Chem. Rev. 1985, 64, 135–157. Heinze, M. Schlenker, Eur. J. Inorg. Chem. 2005, 66–71; e) K. [116] J.-P. Launay, Chem. Soc. Rev. 2001, 30, 386–397. Heinze, M. Beckmann, Eur. J. Inorg. Chem. 2005, 3450–3457; [117] C. Chiorboli, M. T. Indelli, F. Scandola, Top. Curr. Chem. f) K. Heinze, U. Wild, M. Beckmann, Eur. J. Inorg. Chem. 2005, 257, 63–102. 2007, 617–623; g) M. Čakić Semenčić, D. Siebler, K. Heinze, [118] W. Kaim, G. K. Lahiri, Angew. Chem. Int. Ed. 2007, 46, 1778– V. Rapić, Organometallics 2009, 28, 2028–2037; h) D. Siebler, 1796; Angew. Chem. 2007, 119, 1808–1828. C. Förster, K. Heinze, Eur. J. Inorg. Chem. 2010, 523–527; i) [119] K. Heinze, D. Siebler, Z. Anorg. Allg. Chem. 2007, 633, 2223– M. Čakić Semenčić, K. Heinze, C. Förster, V. Rapić, Eur. J. 2233. Inorg. Chem. 2010, 1089–1097; j) C. Förster, D. Siebler, K. [120] S. Lu, V. V. Strelets, M. F. Ryan, W. J. Pietro, A. B. P. Lever, Heinze, Acta Crystallogr., Sect. C 2010, 66, m235–m237; k) Inorg. Chem. 1996, 35, 1013–1023. D. Siebler, M. Linseis, T. Gasi, L. M. Carrella, R. F. Winter, [121] A. Weller, Z. Phys. Chem. (Muenchen Ger.) 1982, 133, 93–98. C. Förster, K. Heinze, Chem. Eur. J. 2011, 17, 4540–4551; l) [122] J. Melomedov, J. R. Ochsmann, M. Meister, F. Laquai, K. D. Siebler, C. Förster, K. Heinze, Dalton Trans. 2011, 40, Heinze, Eur. J. Inorg. Chem. 2014, 1984–2001. 3558–3575; m) C. Förster, M. Kovačević, L. Barišić, V. Rapić, [123] J. Melomedov, J. R. Ochsmann, M. Meister, F. Laquai, K. K. Heinze, Organometallics 2012, 31, 3683–3694. Heinze, Eur. J. Inorg. Chem. 2014, 2902–2915. [138] D. Siebler, C. Förster, T. Gasi, K. Heinze, Organometallics [124] a) O. Stern, M. Volmer, Phys. Z. 1919, 20, 183–188; b) J. R. 2011, 30, 313–327. Lakowicz, Principles of Fluorescence Spectroscopy, 3rd edi- [139] a) S. L. Gould, G. Kodis, R. E. Palacios, L. de la Garza, A. tion, Springer Science+Business Media, LLC, New York, Brune, D. Gust, T. A. Moore, A. L. Moore, J. Phys. Chem. B 2006. 2004, 108, 10566–10580; b) K. Heinze, A. Reinhart, Dalton [125] J. Hawecker, J.-M. Lehn, R. Ziessel, J. Chem. Soc., Chem. Trans. 2008, 469–480; c) J. Melomedov, A. Wünsche von Leu- Commun. 1983, 536–538. poldt, M. Meister, F. Laquai, K. Heinze, Dalton Trans. 2013, [126] a) H. Takeda, O. Ishitani, Coord. Chem. Rev. 2010, 254, 346– 42, 9727–9739. 354; b) Y. Tamaki, K. Watanabe, K. Koike, H. Inoue, T. Mor- [140] K. Heinze, M. Beckmann, K. Hempel, Chem. Eur. J. 2008, imoto, O. Ishitani, Faraday Discuss. 2012, 155, 115–127. 14, 9468–9480. [127] a) H. Ozawa, M.-a. Haga, K. Sakai, J. Am. Chem. Soc. 2006, Received: May 26, 2014 128, 4926–4927; b) H. Ozawa, K. Sakai, Chem. Lett. 2007, Published Online: August 15, 2014 Eur. J. Inorg. Chem. 2014, 5468–5490 5490 © 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 28 | 1 INTRODUCTION Section 1.2 | 29 1.2 EXCITED STATE DECAY MECHANISMS IN POLYPYRIDINE RUTHENIUM COMPLEXES Polypyridine ligands typically are strong π-accepting ligands providing the lowest unoccupied molecular orbital (LUMO) in polypyridine ruthenium complexes. At the same time, the metal’s d orbitals of the t2g set (in idealized Oh symmetry) are good π-donors. Hence, upon irradiation into the low-energy absorption band (at around 500 nm) in the visible range of the electromagnetic spectrum, a singlet metal-to-ligand charge transfer (1MLCT) state is populated in such complexes.47 The spin-orbit coupling caused by the heavy ruthenium atom leads to quantitative intersystem crossing (ISC) to the triplet potential surface populating a 3MLCT state.48,49 This state typically is long-lived (τ > 1 ns) and has several decay pathways. These will be discussed in the following. 1.2.1 Phosphorescence As an emissive excited state deactivation from the 3MLCT state to the singlet ground state (1GS) has to occur with spin inversion, phosphorescence is a spin-forbidden process. However, the presence of a heavy element lifts this restriction to some extent due to spin-orbit coupling. Physically, phosphorescence is described via the Einstein coefficient for spontaneous emission. In the Franck-Condon approximation, the decay rate constant for spontaneous emission is given by the following expression:50,51 3 8𝜋 2𝜂3 2 𝑘𝑟( MLCT→ 1GS) = |𝑴 2 3𝑻(𝑄0)| ∑?̃? 𝜈 ∫ |𝜒 ∗ 1 ′𝜒 3𝑀𝐿𝐶𝑇,𝜈′′| (1.1) 3𝜖0ℏ 𝐺𝑆,𝜈 Hereby, 𝜂 is the solvent’s refractive index, 𝑴𝑻(Q0) is the transition dipole moment for the 3MLCT → 1GS transition at the 3MLCT geometry Q0 and 𝜈 is the emission energy (in cm −1). 𝜒 1𝐺𝑆,𝜈′ and 𝜒 3𝑀𝐿𝐶𝑇,𝜈′′ are the nuclear wavefunctions of the ground and excited state with quantum numbers 2 𝜈′ and 𝜈′′, respectively. The overlap integral ∫ |𝜒∗1 ′𝜒 3𝑀𝐿𝐶𝑇,𝜈′′| between these vibrational 𝐺𝑆,𝜈 wavefunctions is referred to as the Franck-Condon factor. Hence, the rate for spontaneous emission is directly proportional to the third power of the emission energy 𝜈. 〈 3 2 𝑀𝐿𝐶𝑇|?̂? 𝑴 (𝑄 ) = ∑ |∑ 𝑆𝑂𝐶 |𝑆𝑚〉 𝑻 0 𝑗∈𝑥,𝑦,𝑧 𝑚 𝑴 (𝑄 )| (1.2) 𝐸(𝑆𝑚)−𝐸( 3𝑀𝐿𝐶𝑇) 𝑺𝒎,𝒋 0 In first-order perturbation theory, the 3MLCT → 1GS transition dipole moment 𝑴𝑻(𝑄0) is dominated by the strength of the spin-orbit coupling (〈 3𝑀𝐿𝐶𝑇|?̂?𝑆𝑂𝐶|𝑆𝑚〉) and the singlet excited state (𝑆𝑚) → 1GS transition dipole moments 𝑴𝑺 (𝑄0) of energetically close-lying singlet states 𝒎,𝒋 (equation 1.2).51 These singlet states are typically 1MLCT states with relatively high transition dipole moments as the respective transitions are not symmetry-restricted. The Franck-Condon factors are closely related to the distortion of the excited state with respect to the singlet ground state, but they do not affect the overall emission rate (or intensity). They only provide a weighing of the vibrational wavefunctions of the ground state: a larger distortion of the excited state with respect to the ground state yields an emission spectrum with a 30 | 1 INTRODUCTION pronounced vibronic progression whereas emission from an undistorted excited state is free of progressions. For polypyridine complexes of ruthenium and osmium, the rate constants for phosphorescence are typically in the range of 104 – 106 s−1.6,7,35,52 A simple dependence between structure and rate of phosphorescence, however, cannot be drawn making the selective manipulation of the latter a difficult task. 1.2.2 Non-radiative Decay The excited 3MLCT state of polypyridine ruthenium complexes can also evolve into the singlet ground state without emission of a photon. Two major relaxation pathways have been described theoretically which are referred to as the weak and the strong coupling limit (Figure 1.1).53 They differ in the displacement of the excited state (ES) potential energy surface (PES) with respect to the ground state along the reaction coordinate, which comprises all geometric changes in the GS → ES transition. In the weak coupling limit, the displacement of the ES surface is small leading to two parabolic potentials stacked vertically above one another (Figure 1.1 a).53 In the strong coupling limit, on the other hand, the displacement of the ES is considerably larger. This yields a surface crossing point between the potential surfaces of the two states in the vicinity of the minimum of the upper state (Figure 1.1 b).53 The consequences of the two limits for the excited state decay mechanisms will be discussed in the following. Figure 1.1 Schematic potential energy surfaces in a) the weak coupling limit and b) the strong coupling limit. Weak coupling limit In the weak coupling limit (Figure 1.1 a), the transition probability for the ES → GS transition is given by the following expression:7,51,53,54 𝑛 2𝜋𝐻2 1 𝑀𝑎𝑏 𝑆 (Δ𝐸−𝑛 ℏ𝜔 −𝜆 ) 2 𝑘𝑛𝑟 = ( ) (4𝜋𝜆𝑠𝑘𝐵𝑇) − 2∑∞ [ 𝑀𝑛 exp(−𝑆𝑀) exp (− 𝑀 𝑀 𝑠 )] (1.3) ℏ 𝑀 𝑛𝑀! 4𝜆𝑠𝑘𝐵𝑇 Section 1.2 | 31 Here, 𝐻𝑎𝑏 is the electronic coupling matrix element for the ES → GS transition, 𝜆𝑠 is the solvent’s reorganizational energy and 𝑆𝑀 is the Huang-Rhys factor. This factor is a measure of the geometrical distortion between the equilibrium geometries of ground and excited state in terms of dimensionless fractional displacements along the complex’s normal modes. 𝜔𝑀 and 𝑛𝑀 are the frequency and quantum number of high-frequency intraligand vibrational modes. Δ𝐸 is approximately corresponds the energy difference between the ground and excited state. Non-radiative decay in the weak coupling limit is a two-step process. Firstly, tunneling from the electronically excited state into the vibrationally excited ground state occurs (horizontal transition in the Jablonski diagram), followed by thermal cooling under emission of IR radiation (heat; vertical transition in the Jablonski diagram). From equation 1.3 follows, that predominantly C−H, N−H and O−H vibrations (𝜔𝑀 = 3000 – 3400 cm −1) contribute to the non-radiative decay in the weak coupling limit, as they require overtones of significantly lower quantum numbers 𝑛𝑀 to be in resonance with the excited state (Δ𝐸 = 10000 – 20000 cm−1) than intraligand C−C and C=C vibrations (𝜔𝑀 = 1200 – 1600 cm −1).53 Additionally, the displacement of the excited state with respect to the ground state plays a crucial role in promoting non-radiative decay. The dependence of the rate of non-radiative decay on the energy gap Δ𝐸 between ground and excited state has led to the often found reference “energy gap law”.7,53,55 In a series of structurally related complexes, in which variations of 𝐻𝑎𝑏 and 𝑆𝑚 are small, the natural logarithm of the decay rate is proportional to −Δ𝐸.7,55 Hence, in the weak coupling regime, non-radiative decay can be suppressed by raising the emission energy or by reducing the excited state distortion. In order to lower the energy of the high-frequency vibrations, deuteration of ligands and solvent is a viable, but synthetically challenging tool (𝜔𝐶−𝐻 ≈ 3000 cm −1, 𝜔 −1 56,57𝐶−𝐷 ≈ 2200 cm ). Strong coupling limit In the strong coupling limit (Figure 1.1 b), the transition probability is given by the following expression:53 1 𝑘 𝑇 2𝜋 2 Δ𝐸 𝑘 = ( 𝐵 )𝐻2 ( ) exp (− 𝑎𝑛𝑟 𝑎𝑏 3 ) (1.4) ℏ 𝐸𝑀(𝑘𝐵𝑇) 𝑘𝐵𝑇 Again, 𝑘𝑛𝑟 depends on the electronic coupling matrix element 𝐻𝑎𝑏. 𝐸𝑀 is the Stokes shift between excitation and emission energy (of states with the same multiplicity, Figure 1.1 b). However, in this case, 𝑘𝑛𝑟 is temperature-dependent with an Arrhenius-like activation term involving the energy difference between the minimum of the excited state and the surface crossing point. Hence, an increasing temperature yields faster non-radiative decay in the strong coupling regime. As the activation barrier Δ𝐸𝑎 depends on the excited state distortion, non-radiative decay can become barrier-free and very fast in the strong coupling limit. 32 | 1 INTRODUCTION Figure 1.2 Potential energy profile of [Ru(bpy)3]2+. The green arrow indicates excitation, the orange arrow shows emissive relaxation, and the red arrows highlight the thermal population of the 3MC state followed by thermally activated surface crossing and ISC to the singlet ground state. An intermediate case of weak and strong coupling limit was found for [Ru(bpy)3]2+ (bpy = 2,2’- bipyridine). Here, the emissive 3MLCT state is accompanied by a metal-centered (3MC) ligand-field excited state of very similar energy.4,6,8,58 While the 3MLCT is only weakly displaced with respect to the 1GS geometry corresponding to the weak coupling limit, the 3MC state is substantially distorted and strongly coupled to the singlet PES (Figure 1.2). Both states are connected via a low- energy transition state allowing the population of the latter from the former and vice versa.6 Hence, after excitation into a 1MLCT state and intersystem crossing (ISC) into the 3MLCT state, excited state decay can occur via emission ℎ𝜈′, tunneling to high-lying vibrationally excited singlet states (weak coupling limit) or thermal population of the 3MC state followed by surface crossing and ISC to the singlet ground state (strong coupling limit). The lifetime 𝜏0 of an excited state is determined by the rate constants of all processes that depopulate this state, which typically is the radiative decay and several non-radiative decays:59,60 1 1 𝜏0 = = (1.5) 𝑘0 𝑘𝑟+∑𝑘𝑛𝑟,𝑖 The quantum efficiency 𝜙 of the emission process thus is described as follows:59 𝑘𝑟 𝑘 𝜙 = = 𝑟 = 𝑘 𝜏 (1.6) 𝑘0 𝑘𝑟+∑𝑘 𝑟 0 𝑛𝑟,𝑖 In the case of [Ru(bpy)3]2+, the decay via the 3MC state is associated with an activation barrier. As a consequence, the excited state lifetime (and the quantum yield) become temperature- dependent:6,8 −1 Δ𝐸 −1 𝜏0(𝑇) = [𝑘𝑟 + 𝑘𝑛𝑟,𝑀𝐿𝐶𝑇→𝐺𝑆 + 𝑘𝑛𝑟,𝑀𝐿𝐶𝑇→𝑀𝐶(𝑇)] = [𝑘 0 𝑎 1 + 𝑘2 exp (− )] (1.7) 𝑅𝑇 Section 1.2 | 33 With 𝑘1 = 𝑘𝑟 + 𝑘𝑛𝑟,𝑀𝐿𝐶𝑇→𝐺𝑆 describing the temperature-independent emissive and 3MLCT−1GS Δ𝐸 tunneling processes and 𝑘02 exp (− 𝑎) describing the 3MLCT−3MC transition with Δ𝐸𝑎 being the 𝑅𝑇 activation barrier of the 3MLCT−3MC transition. In fact, the temperature profile of the excited state lifetime of [Ru(bpy) 2+3] is more accurately described by two thermally activated processes with different activation barriers. The second activated process (𝐸 ≈ 100 cm−1𝑎 ) is associated with a second emissive MLCT state of higher singlet character.2,3,8,58 1.2.3 Other Excited State Decay Channels In addition to luminescence and non-radiative decay, which are both unimolecular deactivation channels, an excited state can also be quenched in a bimolecular fashion when its lifetime is sufficiently long (> 1 ns). These quenching processes either involve energy transfer to a molecule with an energetically suitable excited state or electron transfer from or to the excited molecule by an oxidant or reductant with suited redox potentials. Both processes will be discussed briefly from a mechanistic point of view. Figure 1.3 Schematic representation of a) energy transfer between singlet excited and ground state molecules via Förster (blue) and Dexter energy transfer (red) and b) oxidative (red) and reductive (blue) electron transfer quenching of an excited singlet state. Energy Transfer In general, energy transfer can only occur when the excited states of a donor and an acceptor are in resonance, i.e. when the energy that the excited donor molecule emits is suitable to excite the acceptor molecule into one of its excited states. Quantum mechanically, the energy transfer rate is composed of two contributions accounting for through-space and through-bond interaction between the donor and acceptor molecule, respectively. Depending on which component dominates, energy transfer is referred to as Förster-type (through-space) or Dexter-type (through- bond, Figure 1.3 a).60–62 For a Förster-type through-space energy transfer, the rate constant is given by the following equation:60,61 𝜙 𝜅2 𝑘𝐸𝑛𝑇 = 8.8 ∙ 10 −25 0 6 4 𝐽𝐹 (1.8) 𝜏0𝑟 𝜂 𝜙0 and 𝜏0 are hereby the quantum yield and lifetime of the donor molecule in absence of an energy acceptor, 𝜅 is an orientation factor taking into account the transition dipole moments of 34 | 1 INTRODUCTION donor and acceptor molecule, 𝑟 corresponds to the donor-acceptor distance and 𝜂 is the solvent’s refractive index. The factor 𝐽𝐹 is known as Förster resonance integral and describes the spectral overlap of the donor’s normalized emission spectrum and the acceptor’s normalized absorption spectrum. The distance dependence 𝑟−6 results from that fact that this energy transfer effectively arises from the interaction of two transition dipoles. Typically, Förster resonance energy transfer (FRET) is encountered between different molecules provided all involved states have the same multiplicity. Dexter-type energy transfer comes into play when the involved states differ in their multiplicity. As it is mediated either by direct orbital overlap of donor and acceptor or by the frontier orbitals of a bridge, its efficiency decreases exponentially with the donor-acceptor distance:60,62 2𝜋 𝑘𝐸𝑛𝑇 = (𝐻𝑎𝑏) 2 𝐽𝐷exp[−𝛽𝐸𝑛𝑇(𝑟 − 𝑟0)] (1.9) ℏ For Dexter energy transfer, the electronic coupling matrix element 𝐻𝑎𝑏 between the donor and acceptor states as well as the Dexter overlap integral 𝐽𝐷 are crucial. The spectral overlap, however, does not nearly need to be as large as for Förster energy transfer. This is because formally, Dexter energy transfer is a double electron transfer between donor and acceptor and does not require the resonant transfer of a photon (Figure 1.3 a). Instead, the attenuation factor 𝛽𝐸𝑛𝑇 takes the tunneling barriers for both electron transfer processes into account. Electron Transfer The theory of thermal electron transfer reactions has been extensively studied by Marcus63–65 and the so-called Marcus theory was extended to photochemical electron transfer by Hush.66,67 The rate for thermal electron transfer is given by:60,68 Δ𝐺‡ 𝑘𝐸𝑇 = 𝜈𝑁𝜅𝑒𝑙 exp [− ] (1.10) 𝑅𝑇 Here, 𝜈𝑁 is the average nuclear frequency factor, 𝜅𝑒𝑙 is the electronic transmission factor and Δ𝐺 ‡ corresponds to the activation energy required for electron transfer to occur. Generally, before an electron is transferred between a donor and an acceptor, their molecular geometries have to adjust in order to allow for barrier-free electron transfer. The average nuclear frequency factor gives effective frequency with which geometries appropriate for electron transfer are reached through molecular vibrations. The transmission coefficient is linked to the probability of electron transfer in the transition region. Marcus showed that the potential surfaces can be considered harmonic yielding a quadratic dependence of the activation barrier Δ𝐺‡ on the thermodynamic driving force Δ𝐺 600: ‡ 𝜆 Δ𝐺 2 Δ𝐺 = (1 + 0) (1.11) 4 𝜆 Here, 𝜆 is the nuclear reorganizational energy. From equations 1.10 and 1.11 follows, that the electron transfer rate reaches a maximum when Δ𝐺0 = −𝜆. This means that the rate will decrease when the electron transfer becomes too exergonic (Marcus-inverted region). In the endergonic and slightly exergonic range, however, the electron transfer rate increases with increasing driving force (Marcus-normal region). Section 1.2 | 35 Typically, electron transfer reactions in the Marcus-inverted region involve photo-excited molecules, since both their oxidative and reductive strength are substantially increased in the excited state (Figure 1.3 b). According to the Rehm-Weller equation, the excited state redox potentials 𝐸∗ can be estimated from the ground state redox potentials 𝐸𝑜𝑥/𝐸𝑟𝑒𝑑 and the energy gap 𝐸 between the excited and ground state as follows:6900 𝐸∗𝑜𝑥 = 𝐸𝑜𝑥 − 𝐸00 (1.12) 𝐸∗𝑟𝑒𝑑 = 𝐸𝑟𝑒𝑑 + 𝐸00 (1.13) Spectroscopically, both electron and energy transfer quenching of excited states by an external quencher can be observed and quantified. This is typically done on the basis of the Stern-Volmer equation 1.14. It relates the dependence of the emission intensity on the concentration of an external quencher to a bimolecular quenching constant 𝑘𝑞, which incorporates both, the diffusion rate and the rate of electron/energy transfer:70 𝜙 0 = 1 + 𝐾𝑆𝑉[𝑄] = 1 + 𝑘𝑞𝜏0[𝑄] (1.14) 𝜙([𝑄]) With 𝜙0 and 𝜙([𝑄]) being the emission quantum yields in the absence and presence of a quencher of concentration [𝑄], 𝐾𝑆𝑉 being the Stern-Volmer constant and 𝜏0 being the emitter’s lifetime in the absence of a quencher. 36 | 1 INTRODUCTION 1.3 MIXED VALENCE AND OPTICAL ELECTRON TRANSFER The expression “mixed valence” is usually utilized in the context of compounds that contain the same atom or fragment in two (or more) different oxidation states. These atoms or fragments may or may not interact depending on their respective electronic environments and the distance between them leading to interesting spectroscopic phenomena that are not observed for the independent fragments. [(NH3)5Ru(µ-pz)Ru(NH 5+3)5] (pz = pyrazine) constitutes a textbook example of such a mixed-valent complex formally containing two ruthenium atoms with an averaged oxidation state of +2.5. Depending on the spectroscopic method, however, the complex’s properties appear to reflect the presence of Ru2+/Ru3+ or Ru2.5+/Ru2.5+. To classify mixed-valent compounds, Robin and Day introduced a classification based on the degree of interaction between the two fragments and the delocalization of the odd electron.71,72 Hereby, Robin-Day class I describes entirely valence- localized complexes, while class III refers to a completely delocalized system. The intermediate class II consists of complexes with predominantly localized valences, but with measurable electronic coupling.72,73 As expected, electron transfer can occur between the two redox centers in different oxidation states in Robin-Day class II complexes. Starting from a system A+−L−A’, with A and A’ being the redox-active sites and L being a bridging ligand, electron transfer yields the valence tautomer A−L−A’+. This electron transfer can be driven thermally or optically. Figure 1.4 a) shows a potential energy diagram including the potential curves of the tautomers A+−L−A’ and A−L−A’+. In this picture, the A−L−A’+ configuration is an electronically excited state at the A+−L−A’ geometry. Hence, optical excitation with a suitable energy, typically in the near infrared region of the electromagnetic spectrum, can promote electron transfer from the reduced to the oxidized fragment resulting in an intervalence charge transfer (IVCT) absorption.66,73,74 Figure 1.4 Potential energy surfaces of a) a class II mixed-valent system and b) a delocalized class III system. Section 1.3 | 37 The electronic coupling 𝐻𝑎𝑏 between the redox centers in class II complexes can be deduced from energy and shape of the IVCT band:66 2.05∙10−2√𝜖𝑚𝑎𝑥?̃?𝑚𝑎𝑥?̃?1/2 𝐻𝑎𝑏 = (1.15) 𝑅 With the maximum extinction coefficient of the IVCT band 𝜖 −1𝑚𝑎𝑥 in M cm −1, the absorption maximum 𝜈 in cm−1 and the full width at half maximum 𝜈 in cm−1𝑚𝑎𝑥 1/2 . 𝑅 describes the distance between the redox centers in Å. In Robin-Day class III systems (Figure 1.4 b), no IVCT band is observed. Instead, optical excitation in the near infrared yields a charge-resonance transition that is associated with a substantially asymmetric absorption band broadened towards higher energies. The energy of this transition is directly correlated with the electronic coupling:66 ?̃? 𝐻 = 𝑚𝑎𝑥𝑎𝑏 (1.16) 2 In an ideal mixed-valent system of class II, the theoretically expected full width at half maximum 𝜈1/2 depends on the energy of the absorption maximum 𝜈 73 𝑚𝑎𝑥: 𝜈1/2,𝑡ℎ𝑒𝑜𝑟 = √2310𝜈𝑚𝑎𝑥 (1.17) This allows the introduction of a parameter Γ that allows to assign complexes to the different classes:73 Γ = 1 − (𝜈1/2,𝑜𝑏𝑠𝑣𝑑/𝜈1/2,𝑡ℎ𝑒𝑜𝑟) (1.18) For Γ = 0, the complex belongs to the very weakly coupled class I/II regime, at a value of 0.5 it is at the class II/III transition. Beyond that, a complex is considered entirely valence-delocalized. Besides the IVCT band, also the splitting of the half-wave potentials Δ𝐸 of consecutive redox processes with potentials 𝐸1 and 𝐸2 in the series A−L−A’ → [A−L−A’] + → [A−L−A’]2+ is often consulted to evaluate the interaction between the redox centers. Frequently, the comproportionation constant is employed instead of the potential splitting:14,15,75 2 ' +𝑐 ([A-L-A ] ) 𝐾𝑐 = 2+ (1.19) 𝑐([A-L-A'])∙𝑐([A-L-A'] ) 𝑧𝐹Δ𝐸 𝐾𝑐 = exp (− ) with Δ𝐸 = 𝐸 − 𝐸 (1.20) 𝑅𝑇 2 1 However, the detection of a sizable splitting Δ𝐸 cannot unambiguously be traced back to a large electronic coupling, as illustrated by equation 1.21:76 Δ𝐸 = Δ𝐸𝑠𝑡𝑎𝑡 + Δ𝐸𝑖𝑛𝑑 + Δ𝐸𝑒𝑥 + Δ𝐸𝑒𝑙 + Δ𝐸𝑟𝑒𝑠 (1.21) Here, Δ𝐸𝑠𝑡𝑎𝑡 is the statistical contribution, which amounts to 36 mV in a system with two identical complex halves. The inductive contribution Δ𝐸𝑖𝑛𝑑 accounts for the influence that a redox process on one redox-active site has on all intramolecular bond strengths and thus on the redox potential 38 | 1 INTRODUCTION of the second redox center. Δ𝐸𝑒𝑥 accounts for the magnetic exchange contribution. The electrostatic contribution arising from the coulombic repulsion as a consequence of the accumulation of charges in a single molecule is taken into account with Δ𝐸𝑒𝑙. Only the resonance contribution Δ𝐸𝑟𝑒𝑠 actually arises from the partial delocalization of the odd electron in the mixed- valent state but it is not accessible individually in electrochemical experiments. This is why the half-wave splitting Δ𝐸 should not be used to estimate the electronic coupling in a mixed-valent system.76 Section 1.4 | 39 1.4 CYCLOMETALATION The concept of “cyclometation” was introduced by Trofimenko in 197377 and refers to reactions of transition metal complexes, in which one of the ligands undergoes a metalation step yielding a chelate ring that contains a carbon-metal σ-bond.78 Rather common are reactions at ortho- positions of phenyl rings attached to a ligand leading to the less common term “ortho-metalation”. Products of such cyclometalation reactions can be regarded as having a coordinating nitrogen formally exchanged by an isoelectronic carbon anion. The cyclometalation step is usually accomplished under relatively mild conditions without the usage of highly reactive starting materials such as aryl lithium or Grignard reagents. Particularly in the last few decades, cyclometalation reactions have emerged as one of the most studied fields in organometallic chemistry since the products can show interesting luminescent properties or be suited as sensitizers in the dye-sensitized solar cell (DSSC, see section 1.5).46,79,80 Despite that, the cyclometalation reaction itself still has some “black box” character to it, as the actual reaction mechanism is often unresolved.78,79 Figure 1.5 Schematic representation of the cyclometalation of a C−H bond in the vicinity of a donor atom D. Most commonly, aromatic C−H bonds in the proximity of a donor atom D are metalated. The accepted mechanism is sketched in Figure 1.5.79 It involves coordination of the donor atom to the metal followed by oxidative addition of the pre-organized C−H bond yielding the metal in a twofold oxidized state as well as a carbanionic and a hydride ligand. The complex is deprotonated in the following giving the product with the metal in its original oxidation state. However, this mechanism is unlikely for cyclometalation reactions involving ruthenium(III) starting materials such as RuCl3(tpy) (tpy = 2,2’;6’,2’’-terpyridine), as the metal center would be forced to go through a RuV state. Instead, it is plausible that in the pre-organized state with an electropositive metal center in the vicinity, deprotonation occurs before or simultaneous to the M−C bond formation step circumventing the high oxidation state. The most frequently encountered metallacycle that is formed in these reactions is five-membered and entirely conjugated in the organic backbone. This probably results from a maximized degree of preorganization in these systems. Furthermore, cyclometalation of D^D^(C−H) ligands occurs much more readily than of D^(C−H)^D ligands which can be traced back to the chelating effect of the D^D anchor prior to cyclometalation. In recent years, particularly cyclometalated polypyridine complexes of d6, d8 and d10 transition metal cations have gained much interest for their luminescent properties.79–82 Additionally, cyclometalated polypyridine ruthenium(II) complexes have emerged as a new class of very promising class of DSSC sensitizers besides the benchmark thiocyanate-containing ruthenium dyes 40 | 1 INTRODUCTION (see section 1.5).42,46 A few exemplary complexes are shown in Figure 1.6 along with the synthetic protocol of the respective cyclometalation step. Figure 1.6 Exemplary cyclometalation reactions yielding the prototype complexes [Ru(bpy)2(ppy)]+ 39, [Ru(tpy)(pbpy)]+ 83,84, [Ir(ppy) 852]2Cl2 and [Ir(bpy)(ppy) ]+ 862 (ppyH = 2-phenylpyridine, pbpyH = 6-phenyl-2,2’-bipyridine). Comparing the spectroscopic, electrochemical and electronic properties of cyclometalated polypyridine ruthenium complexes to those of their non-cyclomelatated counterparts, a few general trends can be established:45,79,87,88  Cyclometalation reduces the overall charge of the complex by one unit per cyclometalating site. This greatly affects the solubility of cyclometalated complexes.  Cyclometalating ligands are less prone to nucleophilic ligand substitution under basic or neutral conditions but can be substituted in the presence of strong acids.  The high σ-donor strength of cyclometalating ligands increases the ligand field splitting and pushes metal-centered excited states to higher energies.  Due to the overall increase of electron density at the metal center, it is oxidized at substantially lower potentials than its non-cyclometalated pyridine analogue. Reduction of the remaining polypyridines on the other hand requires only slightly higher potentials. This is associated with an overall shift of the frontier orbitals to higher energies. Section 1.4 | 41  Additionally, cyclometalation reduces the local symmetry around the metal center (e.g. RuN5C instead of RuN6) which results in the metal’s d orbitals to be shifted apart energetically. This is accompanied by a substantially broadened absorption spectrum as the number of allowed optical transitions is drastically increased and their transition energies are spread over a large portion of the visible range of the electromagnetic spectrum. In summary, cyclometalation is a powerful tool to manipulate the electronic properties of a given complex. However, the synthesis can be challenging as the functional group tolerance of the cyclometalation step is limited.78 42 | 1 INTRODUCTION 1.5 DYE-SENSITIZED SOLAR CELL The dye-sensitized solar sell (DSSC) was first introduced by O’Regan and Grätzel in 1991.22 Typically, it is composed of a light-harvesting dye, often referred to as sensitizer S, coated onto a nanoporous semiconducting carrier material such as TiO2. The semiconductor itself is mounted on a transparent conductor like fluorine-doped tin oxide (FTO) as anode. As cathode, usually silver or platinum is employed. The gap between the electrodes is filled with an electrolyte containing a redox shuttle. The working principle of a DSSC is depicted in Figure 1.7 a). Upon irradiation with visible light, the sensitizer is promoted into a photo-excited state S*. This state has sufficient reducing power to inject an electron into the conducting band of the semiconductor (1). The injected electrons are collected at the anode and transferred into the electric circuit (2). At the sensitizer/electrolyte interface, the oxidized sensitizer S+ is reduced to its original state S by the surrounding redox mediator (3). After diffusing to the cathode, the redox mediator is regenerated at the electrode surface closing the electric circuit (4). However, not only productive steps are possible within a DSSC. Several parasitic processes reducing the overall performance of a DSSC are known. The photo-excited sensitizer can evolve into its ground state without electron injection (5). Additionally, as the TiO2 surface is in contact with the sensitizer and the electrolyte, recombination processes between injected electrons and oxidized dye (6) or redox mediator (7) molecules can occur. Finally, the photo-excited dye can recombine with oxidized redox mediator molecules (8). For an efficient performance of a DSSC, the rate constants of the productive steps 1 – 4 must surpass those of the recombination processes 5 – 8.23,46 Figure 1.7 a) Schematic working principle of the DSSC. Black arrows indicate productive electron transfer steps, red arrows show parasitic recombination processes. b) Current density-voltage plot of a typical DSSC including short circuit and maximum power current Isc and Imp, open circuit and maximum power voltage Uoc and Ump as well as maximum output power (area of the grey rectangle). The efficiency 𝜂 of a DSSC is measured in terms of the output power 𝑃𝑜𝑢𝑡 with respect to the input power 𝑃𝑖𝑛 and is usually given under standardized conditions with normalized solar light irradiation (Air Mass 1.5, 𝑃𝑖𝑛 = 1000 W m −2). The output power obtainable from a DSSC is given by Section 1.5 | 43 the product of photovoltage 𝑈 and current density 𝐼, as displayed in the current density-voltage plot in Figure 1.7 b, maximizing at the maximum power point: 𝑃 𝜂 = 𝑜𝑢𝑡 𝑈 = 𝑚𝑝 𝐼𝑚𝑝 (1.22) 𝑃𝑖𝑛 𝑃𝑖𝑛 The fill factor ff and the wavelength-dependent incident photon-to-current conversion efficiency (IPCE) are two further quantities characterizing the performance of a DSSC. The fill factor describes the shape of the current density-voltage curve and is given by: 𝑈𝑚𝑝𝐼𝑚𝑝 𝑓𝑓 = (1.23) 𝑈𝑜𝑐𝐼𝑠𝑐 The IPCE is determined as the ratio between the number of electrons collected per number of photons irradiated at a given wavelength:23,46 𝐼𝑠𝑐(𝜆) 1 ℎ𝑐 𝐼𝑃𝐶𝐸(𝜆) = (1.24) 𝑃𝑖𝑛(𝜆) 𝜆 𝑒 Despite the sophisticated definition of all relevant parameters describing the performance of a DSSC, a standard reference dye is usually measured alongside the studied compounds as cell setup, solvents, additives or even impurities in any of the components vary from laboratory to laboratory making the obtained results usually difficult to compare.23,46 In order to optimize the performance of a DSSC, all its components have to be adjusted to each other. Considering the amount of sensitizers, electrolytes, additives, co-adsorbents etc. known to date, finding the perfect DSSC configuration is a nearly impossible task especially because many of the key step are interfacial reactions which are not too well understood. However, it is possible to establish criteria that make a certain component suitable for a DSSC application. In the following, these criteria will be discussed for the sensitizer. First and foremost, a sensitizer needs to be capable of effectively harvesting solar light with a wavelength shorter than 900 nm to ensure maximum usage of the incident energy.23,25,46 Additionally, it must provide one or more anchoring groups allowing for tight and irreversible binding to the semiconductor surface and to ensure efficient electron injection into the conduction band.46 This is usually accomplished by introduction of −CO2H or −PO3H groups that are known to bind firmly to TiO2. Moreover, the molecular frontier orbitals must be geometrically suited to allow efficient charge injection into TiO2 and charge recombination with the electrolyte.42 This means, that the lowest unoccupied orbital (LUMO) should be found in proximity to the anchor groups on the anchoring ligand, while the highest occupied orbital (HOMO) points away from the semiconductor surface. Additionally, the energy of the LUMO must be higher than the Fermi edge of the semiconductor and the HOMO energy lower than that of the redox-active electrolyte to make sure electron transfer is thermodynamically feasible. Last, but not least, sensitizers must be thermally, photochemically, and electrochemically stable under the conditions present in the DSSC to avoid decomposition and loss of efficiency over time. For ruthenium-based dyes, anchoring to the TiO2 surface is typically accomplished by 4,4’- dicarboxy-2,2’-bipyridine (dcbpy) or 4,4’,4’’-tricarboxy-2,2’;6’,2’’-terpyridine (tctpy) ligands that also provide LUMOs with suited geometry and energy. On the opposite site facing away from the semiconductor surface, thiocyanato ligands seemed to be most suited for a long time. In this 44 | 1 INTRODUCTION category, the sensitizers with the best DSSC performances are N71989 and the so-called black dye25 (Figure 1.8 and Table 1.1) reaching external efficiencies of more than 10 %. In recent years, cyclometalated ruthenium complexes have entered the field. In these complexes, the cyclometalating ligand substitutes the thiocyanato ligands circumventing decomposition reactions arising from the presence of monodentate ligands to a great extent. YE05 is the most prominent representative of this class with an efficiency of 10.1 % (Figure 1.8 and Table 1.1).42 Also outside the field of ruthenium-based dyes, some remarkable compounds have been developed. The highest performance with an entirely metal-free sensitizer was reached with C219 containing triarylamine and thiophene subunits (𝜂 = 10.1 %).90 However, the record holding class of sensitizers is that of zinc porphyrins. There, SM315 is at the top of the list with an efficiency of 13.0 %91 closely followed by YD2-o-C8 (𝜂 = 12.3 %).92 All these sensitizers are depicted in Figure 1.8 and their performance parameters are summarized in Table 1.1. Figure 1.8 Selected state-of-the-art DSSC sensitizers with efficiencies greater than 10 % reported in the literature (N71989, black dye25, YE0542, C21990, YD02-o-C892, SM31591). Section 1.5 | 45 Table 1.1 Performance parameters of selected state-of-the-art DSSC sensitizers reported in the literature (N71989, black dye25, YE0542, C21990, YD02-o-C892, SM31591). 𝑉𝑜𝑐 (V) 𝐼𝑠𝑐 / mA cm −2 𝑓𝑓 𝜂 / % N71989 0.846 17.73 0.72 11.2 black dye25 0.720 20.53 0.70 10.4 YE0542 0.800 17.00 0.74 10.1 C21990 0.770 17.94 0.73 10.1 YD02-o-C892 0.935 17.66 0.74 12.3 SM31591 0.910 18.10 0.78 13.0 | 47 2 AIM OF THE WORK The synthesis and study of mixed-valent complexes has intrigued chemists for a long time as it provides insight into the underlying physicochemical principles. Particularly interesting in this context are structurally asymmetric, but electronically symmetric complexes. Such systems were previously studied in the research group of Prof. Katja Heinze by means of the amide-bridged bis(terpyridine)ruthenium complexes [I a]5+ and [I b]5+ as well as several amide-bridged ferrocene oligomers [II]+ (Figure 2.1).93–96 While electronic coupling between the metal centers could be detected for the oligoferrocenes, no measurable electronic communication was found for the ruthenium complexes. This was mainly attributed to the bridge’s frontier orbitals being energetically too separated from the metal dπ orbitals which leads to high tunneling barriers for thermal and optical electron transfer (see section 3.1).96 Figure 2.1 Chemical structures of exemplary literature-known mixed-valent systems [I a]5+, [I b]5+ and [II]+ developed by the Heinze group)93–96 as well as key objective structures of this work [1]3+ and [2]+. However, preliminary density functional theoretical results suggested that these tunneling barriers can be lowered using a bis(cyclometalating) bridging ligand as in [1]3+ as this shifts the bridge’s highest occupied orbital into the energy range of the metal dπ orbitals. A primary objective of this work is the synthesis and characterization of a dinuclear bis(cyclometalated) polypyridine ruthenium complex as shown in Figure 2.1. This requires the development of suitable synthon such as [2]+ that allows the formation of an amide bond between the two complex halves as the final step. While the introduction of carboxy groups on cyclometalating ligands is well-established,87,88,97 no primary or secondary amine or amide functional groups have been used on ligands involved in cyclometalation reactions to date. 48 | 2 AIM OF THE WORK Hence, a robust synthetic protocol will be elaborated that enables the cyclometalation step in high yields and provides access to a cyclometalated bis(tridentate) ruthenium complex bearing a free −NH2 group. In contrast to the cyclometalated complexes of iridium, all literature-known cyclometalated polypyridine ruthenium complexes have been shown to be only very weakly emissive at room temperature.97–100 Accomplishing the synthesis of −NH2 carrying mononuclear complexes also enlarges the pool of accessible functional groups to strongly electron donating ones and allows to study the dependence of the emission on the electronics of the substituents in more detail. By a combination of theoretical and spectroscopic techniques, the excited state properties of cyclometalated bis(tridentate) ruthenium complexes will be studied in close detail particularly investigating the dependence of occurring non-emissive deactivation channels on the functional groups. Additionally, the excited state properties of the known complex [I b]5+ will be studied in more detail. Ultimately, as cyclometalated polypyridine ruthenium complexes have been shown to be suited for sensitizing TiO2, the impact of an amino group on the cyclometalating ligand will be studied. The redox-activity of the –NR2 group might help stabilize the complex after charge injection and suppress parasitic charge recombination processes.101 | 49 3 RESULTS AND DISCUSSION All findings of this dissertation have been published/submitted as scientific articles in/to peer- reviewed chemistry journals. These articles will be reprinted in the following with permission of the respective publishers. The synthesis and characterization of the dinuclear bis(terpyridine)ruthenium(II) complex [I b]4+ with high electronic symmetry despite an intrinsic structural asymmetry is presented in section 3.1 “Dual Emission and Excited-State Mixed-Valence in a Quasi-Symmetric Dinuclear Ru−Ru Complex”. While the study of the ground state properties of [I b]4+ and [I b]5+ has already been carried out as part of the diploma thesis of Christoph Kreitner, the dual emission of photo-excited [I b]4+ was discovered during the time of this dissertation. The unusual dual emission was traced back to two energetically close-lying and thermally equilibrated excited states, one being a 3MLCT state and the second a triplet charge-separated (3CS) state. The electronic implications of this observation are discussed. In section 3.2, “Understanding the Excited State Behavior of Cyclometalated Bis(tridentate)ruthenium(II) Complexes: A Combined Experimental and Theoretical Study”, the synthesis and characterization of the two cyclometalated complexes [Ru(dpb-NHAc)(tpy-COOEt)]+ and [Ru(dpb-COOEt)(tpy-NHAc)]+ is presented. The electronic transitions responsible for the visisble range absorption bands of the UV-Vis spectrum are studied using a combination of resonance Raman spectroscopy and theoretical methods. Additionally, an attempt at understanding the different emissive properties of the two complexes is undertaken based on variable-temperature measurements of the emission quantum yields which are supported by density functional theory (DFT) calculations. The synthesis of the target molecules [1]2+ and [2]+ is presented in section 3.3 along with several other substituted [Ru(dpb-R)(tpy)]+ complexes (R = NHAc, NH2, COOEt and COOH, respectively). The article “The Photochemistry of Mono- and Dinuclear Cyclometalated Bis(tridentate)ruthenium(II) Complexes: Dual Excited State Deactivation and Dual Emission” focusses on the elucidation of the excited state deactivation processes on the basis of temperature-dependent emission quantum yield measurements and highlights how two parasitic triplet states flank the emissive 3MLCT state to cause efficient emission quenching at room temperature in all complexes of the [Ru(dpb)(tpy)]+ type. Furthermore, the dinuclear complex [1]3+ was found to exhibit electronic communication between the metal centers in the mixed-valent state, while no coupling was found in the photo-excited state of [1]2+. The degree of electronic coupling in the mixed-valent state is estimated and the marked difference between mixed-valent [1]3+ and photo-excited state of [1]2+ is explained. The article “Strongly Coupled Cyclometalated Ruthenium Triarylamine Chromophores as Sensitizers for DSSCs” in section 3.4 demonstrates the usage of [Ru(dpb-NR2)(tpy(-COO)3)]2- (R = aryl) complexes as sensitizers in TiO2-based DSSCs in combination with standard iodide/triiodide and novel polypyridine cobalt-based electrolytes and traces the trends in the solar cell efficiencies back to insufficiencies in the dye regeneration rates after charge injection into the semiconductor. 50 | 3 RESULTS AND DISCUSSION This publication was developed in a joint effort of Christoph Kreitner, who provided the ruthenium complexes, and Andreas Mengel, who synthesized the cobalt electrolytes and constructed and evaluated the solar cells. Section 3.5, “Excited State Decay of Cyclometalated Polypyridine Ruthenium Complexes: Insight from Theory and Experiment” presents a perspective article following an invitation by the publishers of the journal Dalton Transactions of the Royal Society of Chemistry. It summarizes the current state of research in the field of luminescent cyclometalated ruthenium(II) complexes discussing common features and differences in the non-radiative decay between complexes with [Ru(N^N)2(N^C)]+, [Ru(N^N^N)(N^C^N)]+ and [Ru(N^N^N)(N^N^C)]+ coordination cages. It recapitulates the findings presented in sections 3.2 and 3.3 and puts them into the context of other publications involving cyclometalated [Ru(N^N^N)(N^C^N)]+ complexes. Finally, predictions about the temperature-dependence of [Ru(N^N)2(N^C)]+ and [Ru(N^N^N)(N^N^C)]+ complexes are made along with suggestions how to improve the quantum yields of such complexes in general. Not only second and third row transition metal complexes are known for their manifold luminescent properties. Also first row complexes have been shown to emit at room temperature, but such observations are rarer. The article “[Cr(ddpd) 3+2] : A Molecular, Water-Soluble, Highly NIR- Emissive Ruby Analogue” in section 3.6 presents an example of a polypyridine chromium complex with an intense near-infrared emission. The synthetic and spectroscopic work shown in this publication has been carried out by Sven Otto, Christoph Kreitner assisted in the setup and evaluation of the DFT calculations particularly evolving around the accurate description of the doublet and quartet excited states. Comments on the DFT methodology Since all presented findings rely heavily on the quality of the results obtained from quantum chemical calculations, a few thoughts on density functional theory and the choice of the level of theory will be collected here. The employed level of theory varies slightly between the DFT calculations presented in the different publications. Although there is no right or wrong level of theory, when it comes to DFT, several key components evolved during these studies as being crucial to ensure sufficiently accurate results. In general, B3LYP serves well in the description of transition metal complexes.102 While geometry optimizations are typically less dependent on the choice of functional, particularly charge transfer processes are very sensitive to the amount of Hartree-Fock exchange considered in the respective functional.103 This is why GGA functionals such as BPE or BP fail to describe the electronic transitions of polypyridine ruthenium complexes despite the proper description of their geometries (section 3.2). Interestingly, calculations on cyclometalated polypyridine ruthenium complexes using the popular range-separated CAM- B3LYP104 functional yielded fairly inaccurate transitions judging from the vertical excitation energies and the orbital parentage of the respective transition. However, this is not true for all sorts of transition metal complexes.105 The best description of the electronic transitions was obtained employing B3LYP and PBE0.106 B3LYP was used in most calculations as the relative energies of different electronic states were found to be in better agreement with the experimental | 51 data than with PBE0 (section 3.2). Concerning the choice of basis set, ideally it is as large as possible. For geometry optimizations and single point energies, a split-valence double-ξ basis set including polarization functions on all non-hydrogen atoms is usually sufficient and manageable in terms of computational cost.107–109 For the calculation of electronic properties (electronic transitions, EPR parameters etc.), it is highly advised to use larger basis sets of triple-ξ quality.107– 109 To further improve the description of long-range charge-separated states such as 3MLCT or 3LL’CT states, an inclusion of a continuum solvation model is helpful and important, as it stabilizes the charge separation to some extent.103,110 Additionally, relativistic effects must not be neglected for the heavy transition metal atoms. Inclusion of such can be done using the effective core potential approach111, or, more accurately, with the zeroth order regular approximation.112 This typically also requires a refinement of the integration grids, as basis functions can become fairly steep in the proximity of a heavy atom. Interestingly, the incorporation of dispersion interactions in the calculation of the electronic properties of cyclometalated polypyridine ruthenium complexes does not seem to be necessary. In fact, geometries optimized with the D3BJ dispersion correction113 differed more strongly from the crystal structure than those without dispersion correction (section 3.2). The individual contributions of all authors are expressed in more detail at the beginning of the section of the respective article. 52 | 3 RESULTS AND DISCUSSION Section 3.1 | 53 3.1 DUAL EMISSION AND EXCITED-STATE MIXED-VALENCE IN A QUASI- SYMMETRIC DINUCLEAR RU−RU COMPLEX Christoph Kreitner, Markus Grabolle, Ute Resch-Genger and Katja Heinze Inorg. Chem. 2014, 53, 12947–12961. The bimetallic dipeptide [(EtOOC-tpy)Ru(tpy- NHCO-tpy)Ru(tpy-NHCOCH )]4+3 shows dual emission in fluid solution at room temperature arising from two equilibrating triplet states centered on the C- and N-terminal ruthenium sites. The RuIIand RuIII centers in these triplet valence isomers are electronically coupled through a bridging radical anion. In contrast the mixed-valent complex [(EtOOC-tpy)Ru(tpy-NHCO- tpy)Ru(tpy-NHCOCH3)]5+ with a neutral bridging ligand is valence-localized. The equilibrating excited triplet states can be reduced by PhNMe2 via a N···HN hydrogen bridge. Author Contributions The synthesis of complexes 14+ and 22+ and characterization of the ground state of 22+, 14+ and 15+ has been conducted by Christoph Kreitner during his diploma thesis. However, DFT calculations and emission studies of 14+ discovering the dual emission and excited-state mixed-valence have been performed by Christoph Kreitner during his PhD thesis. The excited state lifetime measurements which have been carried out by Markus Grabolle and Ute Resch-Genger at the Bundesanstalt für Materialforschung und -prüfung (BAM) in Berlin, Germany. Hence, the key findings of the manuscript were aquired during the PhD thesis. The manuscript was written by Christoph Kreitner (90 %) and Katja Heinze (10 %). Supporting Information for this article is found at pp. 175 (excluding Cartesian Coordinates of DFT-optimized geometries). For full Supporting Information, refer to http://pubs.acs.org/doi/suppl/10.1021/ic5020362. „Reprinted with permission from Kreitner, C.; Grabolle, M.; Resch-Genger, U.; Heinze, K. Inorg. Chem. 2014, 53, 12947–12961. Copyright 2013 American Chemical Society.” 54 | 3 RESULTS AND DISCUSSION Article pubs.acs.org/IC Dual Emission and Excited-State Mixed-Valence in a Quasi- Symmetric Dinuclear Ru−Ru Complex Christoph Kreitner,†,‡ Markus Grabolle,§ Ute Resch-Genger,§ and Katja Heinze*,† †Institute of Inorganic and Analytical Chemistry, Johannes Gutenberg-University of Mainz, Duesbergweg 10−14, 55128 Mainz, Germany ‡Graduate School Materials Science in Mainz, Staudingerweg 9, 55128 Mainz, Germany §Federal Institute for Materials Research and Testing (BAM), Division 1.5, Richard-Willstaẗter-Str. 11, 12489 Berlin, Germany *S Supporting Information ABSTRACT: The synthesis and characterization of the new dinuclear dipeptide [(EtOOC-tpy)Ru(tpy-NHCO-tpy)Ru(tpy- NHCOCH )]4+ 4+3 3 of the bis(terpyridine)ruthenium amino acid [(HOOC-tpy)Ru(tpy-NH )]2+2 1 2+ are described, and the properties of the dipeptide are compared to those of the mononuclear complex [(EtOOC-tpy)Ru(tpy-NHCOCH 2+3)] 42+ carrying the same functional groups. 34+ is designed to serve a high electronic similarity of the two ruthenium sites despite the intrinsic asymmetry arising from the amide bridge. This is confirmed via UV−vis absorption and NMR spectros- copy as well as cyclic voltammetry. 42+ and 34+ are emissive at room temperature, as expected. Moreover, 34+ exhibits dual emission from two different triplet states with different energies and lifetimes at room temperature. This is ascribed to the presence of a unique thermal equilibrium between coexisting [RuII(tpy- NHCO-tpy·−)RuIII] and [RuIII(tpy-NHCO-tpy·−)RuII] states leading to an unprecedented excited-state RuIIRuIII mixed-valent system via the radical anion bridge tpy-NHCO-tpy·−. The mixed-valent cation 35+, on the other hand, shows no measurable interaction of the RuIIRuIII centers via the neutral bridge tpy-NHCO-tpy (Robin−Day class I). Reduction of 34+ to the radical cation 33+ by decamethylcobaltocene is bridge-centered as evidenced by rapid-freeze electron paramagnetic resonance spectroscopy. Interestingly, all attempts to observe 33+ via NMR and UV−vis absorption spectroscopy only led to the detection of the diamagnetic complex 3-H3+ in which the bridging amide is deprotonated. Hence 3-H3+ (and 4-H+) appear to reduce protons to dihydrogen. The ease of single and double deprotonation of 42+ and 34+ to 4-H+, 3-H3+, and 3−2H2+ was demonstrated using a strong base and was studied using NMR and UV−vis absorption spectroscopies. The equilibrating excited triplet states of 34+ are reductively quenched by N,N-dimethylaniline assisted by hydrogen bonding to the bridging amide. ■ INTRODUCTION The lowest of these 3MLCT excited states is emissive at room The controlled assembly of multinuclear metal complexes temperature and exhibits a reasonably long lifetime (Φ = 0.095,16,18 incorporating electrochemically and photochemically active τ = 855 μs at 298 K in CH3CN). Because of the use of moieties is of great interest for the fundamental understanding chelating ligands this complex has a fairly high thermal and19,20 of energy and electron transfer1−3 on a molecular level and the chemical stability. 2+ modeling of natural photosynthesis4 as well as for the design of The use of [Ru(tpy)2] (tpy = 2,2′:6′,2″-terpyridine)2+ molecular wires5,6 and switches,7,8 photocatalysts,9−11 and instead of [Ru(bpy)3] leads to structurally similar com-21,22 information storage devices.12 (Polypyridine)ruthenium(II) plexes, but these compounds have been far less studied and complexes, especially the archetype compound [Ru(bpy) ]2+ applied in photochemical setups than their bpy analogues. This3 (bpy = 2,2′-bipyridine), have found wide application in such is due to low emission quantum yields and short excited-state arrays due to their high stability and outstanding photochemical lifetimes at room temperature in fluid solution because the3 3 properties. Further applications of this class of compounds MLCT states can undergo thermal depopulation via MC comprise photosensitizers in dye-sensitized solar cells13 and states followed by vibrational relaxation and ISC to the ground1,23−25 emitters in light-emitting electrochemical cells.14 state. This hampers the use of these complexes in the [Ru(bpy) 2+3] features unique optical and electrochemical fields of photoelectron or energy transfer. Several attempts have properties.15,16 The energetically low-lying π* orbitals of the been carried out to increase emission lifetimes and quantum heteroaromatic ligands allow for a metal-to-ligand charge transfer (1MLCT) excitation upon irradiation. Rapid inter- Received: August 21, 2014 system crossing (ISC) leads to population of 3MLCT states.17 Published: November 20, 2014 © 2014 American Chemical Society 12947 dx.doi.org/10.1021/ic5020362 | Inorg. Chem. 2014, 53, 12947−12961 Section 3.1 | 55 Inorganic C hemistry Article yields of bis(tridentate)ruthenium(II) complexes. Increasing We had previously reported the unprotected dipeptide [(R- the bite angle (N−Ru−N) within the ligands raises the energy tpy)Ru(tpy-NHCO-tpy)Ru(tpy-R′)]4+, R = COOH, R′ = NH2 of the 3MC states through better overlap of ligand and metal 24+.57 Its mixed-valent state 25+ features two electronically orbitals thus shifting the thermal population in favor of the uncoupled ruthenium moieties due to differing local redox 3MLCT states.26−30 Functionalization of the parent [Ru- potentials leading to an intrinsic electronic asymmetry. No (tpy) 2+2] in the 4′ position with push−pull substituents has a evidence of photochemical electron transfer from the Ru II similar effect: electron-withdrawing substituents lower the moiety onto the RuIII species was found (Robin−Day class energy of the 3MLCT states, while electron-donating groups I). In contrast to 12+, 24+ containing both a carboxylic acid and increase the energy of the 3MC states.24,25,31 Emission can be an amino group has not been explored in terms of acid−base intensified by several orders of magnitude via these approaches. chemistry, although interesting properties can arise from the A major advantage of [Ru(tpy)2] 2+ over [Ru(bpy) ]2+3 for combination of redox and acid−base active centers in a single functionalization in the ligand backbone is the lack of a molecule (e.g., proton-coupled electron transfer). 58−60 stereocenter in the former. This is important for the In this work, we present an intrinsically structurally development of multinuclear assemblies as it simplifies asymmetric but, in terms of local redox potentials, highly synthesis and purification significantly. It becomes evident symmetric derivative of 24+, with R = COOEt and R′ = considering the stereogenic character of [Ru(bpy) ]2+. Its D NHCOCH (34+3 3 3 ). Its unique electronic and optical properties symmetry leads to enantiomers in the parent complex (Δ, Λ). are studied in detail and are compared to a closely related Complexes of the type [Ru(bpy)(bpy-R)(bpy-R′)]2+, with bpy- mononuclear derivative of the ruthenium amino acid of the R and bpy-R′ carrying di erent functional groups, result in a form [(R-tpy)Ru(tpy-R′)]2+ff (R = COOEt, R′ = NHCOCH3, mixture of diastereomers that requires sophisticated methods to 42+) with the same terminal functional groups as 34+. The be separated or avoided.32−35 Employing [Ru(bpy) ]2+3 in extent of electronic coupling between the redox moieties is dinuclear systems gives rise to three stereoisomers (ΔΔ, meso- evaluated in the neutral, singly oxidized, and singly reduced ΔΛ, and ΛΛ) that can only be circumvented under certain states as well as in the excited state. conditions.33 Typically, aliphatic and aromatic amides exhibit only weakly 61 This is why we employed donor- and acceptor-functionalized acidic behavior (pKa ≈ 18−22 in dimethyl sulfoxide). tpy ligands to develop emissive complexes of the type [Ru(tpy- However, inserting amide bonds in between charged R)(tpy-R′)]2+.31,36,37 Using the functional groups R = COOH polypyridine ruthenium(II) complexes leads to a substantial and R′ = NH2 gives rise to the metallo amino acid 12+31 in polarization of the amide and an acidification of the N−H4+ which the metal is placed in one line with the functional groups bond, which is why the acid−base chemistry of 3 is thus maximizing the ligands’ electronic effects. Amino acid investigated as well. building blocks of this type allow the synthesis of oligopeptides in which ruthenium takes a unique position by enhancing the ■ EXPERIMENTAL SECTION electronic communication between the building blocks,31,38−40 General Procedures. Chemicals were obtained from commercial which is not observed when the metal is placed in a side chain 41−43 suppliers and were used without further purification. Bis(terpyridine)-of the peptide structure. ruthenium(II) complexes [(HOOC-tpy)Ru(tpy-NH2)](PF6)2 In the work presented herein, we demonstrate the synthesis 1(PF6)2, [(EtOOC-tpy)Ru(tpy-NHCOCH3)](PF6)2 4(PF6)2, and and characterization of a protected dinuclear dipeptide [(R- [(EtOOC-tpy)Ru(tpy-NH2)](PF6)2 5(PF6)2 were synthesized accord- tpy)Ru(tpy-NHCO-tpy)Ru(tpy-R′)]4+ of the ruthenium amino ing to literature-known procedures.31,36,37 Air- or moisture-sensitive acid 12+. Dinuclear mixed-valent ruthenium complexes have reactions and compounds were handled in dried glassware under an been widely studied in terms of electronic interaction between inert gas atmosphere (argon, quality 4.6). Acetonitrile was refluxed the metal centers. Symmetric complexes, especially the Creutz− over CaH2 and distilled under argon prior to use in these reactions. IR 5+ spectra were recorded on a BioRad Excalibur FTS 3000 spectrometerTaube ion [(NH3)5Ru-(μ-pz)-Ru(NH3)5] (pz = pyrazine) as 44 using cesium iodide disks. UV−vis absorption spectra were recordedprototype, with identical electronic environments around − on a Varian Cary 5000 spectrometer in 1 cm cuvettes. Emissionboth metal sites have been extensively examined,45 47 and the spectra were recorded on a Varian Cary Eclipse spectrometer. theoretical background is well-understood.48−50 The strength of Quantum yields were determined by comparing the areas under the the through-bond electronic interaction is dominated by the emission spectra recorded for solutions of the samples and a reference distance between the redox centers, as well as the planarity and with matching absorbances on an energy scale (ϕ([Ru(bpy)3]Cl2) = 18 appropriate symmetry of the bridging ligand.51 Additionally, the 0.094 in deaerated CH3CN). Experimental uncertainty is estimated frontier orbitals of the bridge need to be in a similar energy to be 15%. Luminescence decay curves of the samples in acetonitrile range as the involved metal orbitals for the interaction to be were measured under ambient conditions or under inert atmosphere 45,47,51 − by time-correlated single-photon counting (TCSPC) at 22 °C undersignificant. A classification into three classes (Robin magic-angle conditions with an Edinburgh Instruments lifetime Day) distinguishes the degree of communication between the spectrometer (FLS 920) equipped with a supercontinuum laser redox centers, with class I being ascribed to noninteracting and (SC400-PP, Fianium) in combination with a double monochromator, class III ascribed to strongly coupled systems.52−54 a MCP-PMT (R3809U-50, Hamamatsu), and a TCSPC module Directional electronic coupling through asymmetric bridging (TCC 900). The instrument response time was 200 ps; the repetition ligands has not been studied in great detail mainly because of rate was 5 MHz. Sample excitation was at 504 and 492 nm, and the difficulty to generate distinct asymmetric structures55 that fluorescence decays were measured at 684 and 690 nm for 3(PF6)4 meet the basic requirements for electronic interaction and 4(PF6)2, respectively. Decay times were obtained from single- or (planarity, su ciently short distances). Electron transfer in biexponential fits using the spectrometer software. Electrosprayffi ionization (ESI+) and high-resolution (HR) ESI+ mass spectra were natural systems, on the other hand, always occurs directionally 56 recorded on a Micromass QTof Ultima API mass spectrometer withwith small driving forces. This is why systematic synthesis and analyte solutions in acetonitrile. ESI+ mass spectra are reported giving investigation of structurally asymmetric but nearly redox- the m/z ratio and relative intensity of the most intense peak of the symmetric mixed-valent systems is of general interest. typical ruthenium isotope pattern, while HR ESI+ numbers are given 12948 dx.doi.org/10.1021/ic5020362 | Inorg. Chem. 2014, 53, 12947−12961 56 | 3 RESULTS AND DISCUSSION Inorga nic Chemistry Article for the lowest m/z ratio in a given ruthenium isotope pattern. triturated by addition of a solution of NH4PF6 (297 mg) in water (70 Elemental analyses were performed by the microanalytical laboratory mL). The product was collected via filtration, washed with small of the chemical institutes of the University of Mainz. NMR spectra amounts of water and diethyl ether, and dried under reduced pressure were obtained with a Bruker Avance II 400 spectrometer at 400.31 to give [(C6F5OOC-tpy)Ru(tpy-NHCOCH3)](PF6)2 7(PF6)2 as red (1H), 100.66 (13C), and 376.60 MHz (19F) at 25 °C. Chemical shifts δ powder. Yield: 217.5 mg (0.193 mmol, 93%). Anal. Calcd for [ppm] are reported with respect to residual solvent signals as internal C39H24F17N7O3P2Ru (1124.6)·2H2O: C, 40.36; H, 2.43; N, 8.45. standards (1H, 13C) or external standards (19F): CD3CN δ( 1H) = 1.94 Found: C, 40.24; H, 2.21; N, 8.61%. MS (ESI+): m/z (%) = 417.6 ppm, δ(13C) = 1.32 and 118.26 ppm,62 CFCl3 δ( 19F) = 0.00 ppm. (15) [M-2PF 2+6] , 834.1 (3) [M-2PF + + 6−H] , 980.1 (100) [M-PF6] , Electrochemical experiments were performed with a BioLogic SP-50 1542.6 (3) [3M-2PF ]2+6 , 2103.6 (3) [4M-2PF 2+ 6] . HR-MS (ESI +, m/ voltammetric analyzer using platinum wire working and counter z): calcd. for C39H24F11N7O3PRu [M-PF6] +: 974.0579; found: electrodes and a 0.01 M Ag/AgNO3 reference electrode. Measure- 974.0562. 1H NMR (CD3CN): δ = 9.54 (s, 1H, NH), 9.38 (s, 2H, ments were carried out at a scan rate of 100 mV s−1 for cyclic H2), 9.01 (s, 2H, H2′), 8.70 (d, 3J = 8 Hz, 2H, H5 3HH ), 8.42 (d, JHH = 8 voltammetry experiments and at 10 mV s−1 for square-wave Hz, 2H, H5′), 7.99 (t, 3J = 8 Hz, 2H, H6HH ), 7.93 (t, 3JHH = 8 Hz, 2H, voltammetry experiments using 0.1 M [nBu4N][PF6] as supporting H 6′), 7.51 (d, 3J 8HH = 6 Hz, 2H, H ), 7.32−7.26 (m, 4H, H7, H8′), electrolyte in acetonitrile. Potentials are given relative to the 7.17−7.12 (m, 2H, H7′), 2.39 (s, 3H, CH ). 133 C{1H} NMR ferrocene/ferrocenium couple (0.40 V vs SCE,63 E1/2 = 0.90 ± 5 (CD3CN): δ = 172.4 (s, NHCOCH3), 160.6 (s, COOC6F5), 157.8 mV under the given conditions). Electron paramagnetic resonance (2s, C4, C4′), 157.0 (s, C3), 154.8 (s, C3′), 152.8 (s, C8′), 152.3 (s, C8), (EPR) spectra were recorded on a Miniscope MS 300 X-band CW 147.3 (s, C1′), 138.7 (s, C6), 138.4 (s, C6′), 131.6 (s, C1), 128.3 (s, spectrometer (Magnettech GmbH, Germany). Values of g are C7′), 127.6 (s, C7), 125.1 (s, C5′), 124.8 (s, C5), 123.4 (s, C2), 113.5 referenced to Mn2+ in ZnS as external standard (g = 2.118, 2.066, (s, C2′), 24.0 (s, NHCOCH3), (carbon signals of C6F5 not 2.027, 1.986, 1.946). Simulations were performed with the EasySpin observed).74 19F NMR (CD3CN): δ = −73.3 (d, 1JFP = 707 Hz, program package.64 12F, PF6), −154.8 (d, 3JFF = 17 Hz, 2F, o-F), −159.0 (t, 3JFF = 21 Hz, Density functional theoretical (DFT) calculations were carried out 1F, p-F), −163.8 (dd, 3JFF = 21, 17 Hz, 2F, m-F). using the Gaussian09/DFT series of programs65 employing B3LYP as Synthesis of 3(PF6)4. [(C6F5OOC-tpy)Ru(tpy-NHCOCH3)](PF6)2 functional.66 The choice of functional was made due to the vast 7(PF6)2 (59.5 mg, 0.053 mmol) and [(EtOOC-tpy)Ru(tpy-NH2)]- abundance of publications using this functional in calculations on (PF6)2 5(PF6)2 (50 mg, 0.053 mmol) were each dissolved separately in transition metal compounds. Previously published theoretical results acetonitrile (10 mL) under an atmosphere of argon and left to stand on mono- and oligonuclear donor−acceptor functionalized [Ru(tpy)2] overnight over activated molecular sieves (3 Å) to remove crystal complexes were in good agreement with the experimental water. The solution of 5(PF6)2 then was added to a solution of data.29,39,40,57 The LANL2DZ implementation of Gaussian09 was phosphazene base tert-butylimino-tris(dimethylamino)phosphorane used as basis set for all atoms. It comprises Dunning/Huzinaga’s D95 (25.8 mg, 0.110 mmol) in absolute acetonitrile (5 mL) and stirred V valence double-ξ basis without polarization functions for hydrogen, for 45 min followed by the addition of the solution of 7(PF6)2. After it carbon, nitrogen, and oxygen67 and a Los Alamos effective core was stirred at room temperature for 4 h, the reaction was quenched by potential approach plus valence double-ξ basis for ruthenium.68−70 the addition of a few drops of acetic acid and concentrated under This rather small basis set combination was chosen to manage the reduced pressure to 5 mL. The product was precipitated by addition of computational effort of the large systems under study. To account for NH4PF6 (423 mg) and water (80 mL) and collected via filtration. The solvent effects a polarized continuum model modeling acetonitrile crude product was recrystallized from an ethanol/acetone mixture (20 solution was used (IEFPCM, acetonitrile).71−73 Explicit counterions mL, 3:1) and dried under reduced pressure to give [(EtOOC- and/or solvent molecules were not taken into account. All structures tpy)Ru(tpy-NHCO-tpy)Ru(tpy-NHCOCH3)](PF6)4 3(PF6)4 as red were characterized as local minima of the potential energy surface by powder. Yield: 78.2 mg (0.042 mmol, 79%). Anal. Calcd for vibrational analysis (Nimag = 0). No symmetry constraints were C66H50F24N14O4P4Ru2 (1885.2)·4H2O: C, 40.50; H, 2.99; N, 10.02. imposed on the molecular geometries. Found: C, 40.61; H, 2.95; N, 9.78%. MS (ESI+): m/z (%) = 309.0 (5) Synthetic Procedures. Synthesis of 6(PF6) . 57 2 [(HOOC-tpy)Ru- [M-4PF6-Et-Ac] 4+, 435.1 (15) [M-4PF6−H]3+, 725.1 (29) [M-3PF6− (tpy-NH )](PF ) 1(PF ) 312 6 2 6 2 (339 mg, 0.370 mmol) was suspended in H]2+, 777.1 (6) [M-2PF −Ac]2+6 , 798.1 (100) [M-2PF 2+6] , 1741.3 (15) acetyl chloride (25 mL) and refluxed for 2 h giving a dark red solution. [M-PF6] +. HR-MS (ESI+, m/z): calcd. for C66H50F12N14O4P2Ru2 [M- Acetyl chloride was distilled from this, and the residual solid was 2PF ]2+6 : 792.0788; found: 792.0782. 1H NMR (CD3CN): δ = 10.42 dissolved in acetonitrile. The crude product was triturated by addition (s, 1H, tpy-CONH-tpy), 9.49 (s, 1H, CONHCH3), 9.36 (s, 2H, H 2′), of excess diethyl ether and collected via filtration. It was dissolved 9.34 (s, 2H, H2″), 9.23 (s, 2H, H2), 9.01 (s, 2H, H2‴), 8.75 (d, 3JHH = again in boiling water (250 mL) to cleave the mixed anhydride formed 8 Hz, 2H, H5″), 8.70 (d, 3JHH = 8 Hz, 2H, H5), 8.54 (d, 3JHH = 8 Hz, in the reaction of the carboxyl group with acetic anhydride and 2H, H5′), 8.44 (d, 3JHH = 8 Hz, 2H, H5‴), 8.11−7.89 (m, 8H, H6, H6′, precipitated after addition of a solution of NH4PF6 (250 mg) in water H6″, H6‴), 7.62−7.50 (m, 4H, H8, H8″), 7.44−7.33 (m, 4H, H8′, (1 mL). The precipitate was collected, washed with water, and dried H8‴), 7.33−7.26 (m, 4H, H7, H7″), 7.24−7.14 (m, 4H, H7′, H7‴) 4.67 under reduced pressure to give [(HOOC-tpy)Ru(tpy-NHCOCH3)]- (q, 2H, 3JHH = 7 Hz, OCH2CH3), 2.39 (s, 3H, NHCOCH3), 1.59 (t, (PF6)2 6(PF6)2 as a red powder. Yield: 330 mg (0.350 mmol, 95%). 3H, 3J 13 1HH = 7 Hz, OCH2CH3). C{ H} NMR (CD3CN): δ = 171.6 (s, Anal. Calcd for C33H25F12N7O3P2Ru (958.6)·4H2O: C, 38.46; H, 3.23; NHCOCH3), 165.3 (s, tpy-CONH-tpy), 165.1 (s, tpy-COOEt), 158.8, N, 9.51. Found: C, 38.63; H, 3.13; N, 9.68%. Mass spectrometry (MS) 158.8, 158.7, 158.7, (s, C4, C4′, C4″, C4‴), 157.5 (s, C3″), 157.4 (s, (ESI+): m/z (%) = 334.6 (30) [M-2PF ]2+6 , 814.1 (100) [M-PF + 6] , C3), 156.3 (s, C3′), 156.0 (s, C3‴), 153.9, 153.8 (s, C8′, C8‴), 153.5, 1293.1 (3) [3M-2PF ]2+, 1772.6 (3) [4M-2PF ]2+. HR-MS (ESI+, m/ 153.4 (s, C86 6 , C8″), 148.2 (s, C1‴), 147.4 (s, C1′), 140.0 (s, C1″), 139.5, z): calcd. for C +33H25F6N7O3PRu [M-PF6] : 808.0737; found: 139.4 (s, C6′, C6‴), 139.2, 139.2 (s, C6, C6″), 137.5 (s, C1), 129.0 (s, 808.0732. 1H NMR (CD3CN): δ = 9.42 (s, 1H, NH), 9.17 (s, 2H, C7″), 128.9 (s, C7), 128.7 (s, C7′), 128.6 (s, C7‴), 125.9 (s, C5), 125.8 H2), 8.94 (s, 2H, H2′), 8.62 (d, 3J 5HH = 8 Hz, 2H, H ), 8.37 (d, 3J 5HH = 8 (s, C ″), 125.6 (s, C5′), 125.5 (s, C5‴), 123.8 (s, C2), 122.7 (s, C2″), Hz, 2H, H5′), 7.97−7.82 (m, 4H, H6, H6′), 7.46−7.41 (m, 2H, H8), 115.2 (s, C2′), 114.0 (s, C2‴), 63.9 (s, OCH2CH3), 24.9 (s, 7.28−7.15 (m, 4H, H7, H8′), 7.12−7.06 (m, 2H, H7′), 2.35 (s, 3H, NHCOCH3), 14.7 (s, OCH2CH3). CH3). General Procedure for Removal of Crystal Water from the Synthesis of 7(PF6)2. [(HOOC-tpy)Ru(tpy-NHCOCH3)](PF6)2 Complexes 4(PF6)2 and 3(PF6)4. The complex (100 mg) was 6(PF6)2 (200 mg, 0.209 mmol) was dissolved in absolute acetonitrile suspended in chlorotrimethylsilane (5 mL) in an atmosphere of dry (15 mL), and pentafluorophenol (46.2 mg, 0.251 mmol) and N,N′- argon and left to stand for 15−20 min. After removal of excess silane diisopropylcarbodiimide (31.7 mg, 0.251 mmol) were added. After it and the formed siloxane under reduced pressure, the complex was stirred at room temperature for 90 min, the reaction mixture was dissolved in absolute acetonitrile (5 mL) and dried under reduced concentrated to 5 mL under reduced pressure, and the product was pressure again to remove residual acid. NMR analysis showed slight 12949 dx.doi.org/10.1021/ic5020362 | Inorg. Chem. 2014, 53, 12947−12961 Section 3.1 | 57 Inorganic C hemistry Article downfield shifts of amide proton resonances indicative of traces of dissolved in acetonitrile over 3 Å molecular sieves prior to remaining hydrochloric acid that could not be removed with this setting up the reaction. During the coupling reaction a striking method. color change from red to purple was observed, which is attributed to the deprotonation of the generated dinuclear ■ RESULTS AND DISCUSSION species (vide infra). Neither cleavage of the terminal ester nor Synthesis of Dinuclear Amide. The dinuclear ruthenium the amine function was observed under the given water-free complex 34+ is extremely challenging to synthesize via amide conditions (by NMR and ESI-MS). coupling in a classical fashion because of the poor reactivity of The synthesis of the corresponding protected mononuclear the pyridylamine coordinated to the electron-withdrawing complex with identical capping functionalities 42+ was carried ruthenium(II), which is further augmented by the Coulombic out via a literature-known procedure in good yields (Scheme repulsion of the doubly charged mononuclear precursors. The 2).57 pyridylamine can be viewed as an iminium-like structure ( Characterization of Mono- and Dinuclear Amides. The NH +2 ) with rather acidic properties that can be deprotonated successful formation of the pentafluorophenylesters of 62+ and using strong bases.19,31,36 Another possibility to acylate the 12+ is easily evidenced in the 1H NMR spectra of 72+ and 82+ amino group is by employing acid chlorides at elevated because the resonances of the protons H2 are shifted downfield temperatures.57 by ∼0.15 ppm. This is attributed to the stronger electron- The synthesis of the dinuclear dipeptide 34+ was effected in a withdrawing effect of the OPfp group compared to the free four-step synthesis starting from the ethyl ester of the carboxylic acid or its ethyl ester (see Schemes 1 and 2 for atom ruthenium amino acid 52+. The first step was acidic cleavage 31 numbering). The remainder of the 1H NMR spectra is rather of the ester to the amino acid 12+. Subsequent acetylation of unaffected from carboxylic acid activation. For example, in 72+ the amino function with acetyl chloride leads to N-acetyl amido 2+ 57 the amide proton resonates at 9.54 ppm, and proton H 2′ acid 6 in a yield of 95%. For the amide coupling of the two 2+ 2+ 4+ resonates at 9.01 ppm; for 8 2+, the protons of NH2 and H 2′ are building blocks 5 and 6 to 3 suitable conditions needed to found at 6.04 and 8.00 ppm, respectively, which does not differ be established. A broad range of typical conditions for amide 75 signi 2+ 2+ ficantly from the parent compounds 1 and 6 couplings is known, most of which employ active esters in (Supporting Information, Figures S3 and S6). 13C NMR different forms such as 1-hydroxybenzotriazolyl esters (OBt esters),76−78 1-hydroxy-7-azabenzotriazolyl esters (OAt es- chemical shifts (Supporting Information, Figures S4 and S7) ters),79 penta uorophenyl esters (OPfp esters),80,81 and p- are easily assigned via 1H13C correlated techniques (except for fl nitrophenylesters.82 Intermediate activation can be achieved the C6F5 carbon signals, which are not detected under the given19 using acid chlorides83 or N,N′-dicyclohexylcarbodiimide measurement settings). F NMR spectroscopy confirms the (DCC) adducts84 as active species. More advanced and forcing presence of a perfluorinated phenyl ring as well as of two PF6 activation procedures use aminium or phosphonium salt based counterions at typical chemical shifts (Supporting Information, coupling reagents.77,78,85 We have recently shown that Figures S5 and S8). ESI + mass spectra confirm the integrity of the OPfp esters 72+ and 82+intermediate activation of ruthenium amino acids and coupling since only signals of intact cations to amino-functionalized ferrocenes, ruthenium complexes, and with no or one counterions are observed. 4+ bipyridines can be achieved using HOBt/DCC,31 PyBOP,57 As expected the NMR spectra of the dinuclear species 3 are and HATU,39 respectively, when a strong base, typically a more complicated (Figure 1 and Supporting Information, phosphazene base (P1-tBu) is present (PyBOP = benzotriazol- Figures S9 and S10). The intended high electronic similarity of 1-yl-oxytripyrrolidinophosphonium hexafluorophosphate; both complex subunits leads to a multitude of overlapping or HATU = 1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo- close-lying resonances in both the 1H and the 13C NMR [4,5-b]pyridinium 3-oxide hexafluorophosphate). The latter spectra. Nevertheless the success of the amide coupling reaction conditions even distinguish between aromatic amines and the is most easily evidenced by the downfield shift of proton H2′ by pyridylamine present in the ruthenium amino acid 12+, so that 39 ∼1.3 ppm now resonating at 9.36 ppm because the influence ofprotection of the amino group of the complex is obsolete. the electron-donating amino group is lost. Four sets of signals In this work the active ester is isolated and purified to consisting of one singlet, two doublets, and two doublets of provide a well-defined starting material for the following amide doublets (ignoring 4J contributions) are expected with coupling. Pentafluorophenol (PfpOH) is used in combination with N,N′-diisopropylcarbodiimide to activate the acid. The intensities of 1:1:1:1 originating from the four different4+ corresponding urea formed during the reaction is soluble in the terpyridine moieties present in 3 . Especially the four singlets2 2 2 2 water/acetonitrile mixture of the aqueous workup and thus is (protons H , H ′, H ″, and H ‴) are sufficiently separated and4+ easily separated from the insoluble OPfp ester 72+. The confirm the successful formation of 3 (Figure 1). Significant procedure is generalizable and also applicable to the amino acid downfield shifts of the proton resonances of the bridging ligand4+ 12+ affording the OPfp ester 82+. This active ester does not are observed due to the enhanced positive charge of 3 and the exhibit any reactivity toward the free pyridylamino group stronger electron-withdrawing effect affecting particularly 2 2 present in the compound itself but rapidly reacts with aliphatic protons H ′ and H ″. The high charge also affects the amide amines such as tert-butylamine giving the corresponding amide proton of the bridging amide: its resonance appears at 10.42 92+ (for experimental procedures and 1H and 13C NMR spectra ppm and is shifted by 0.93 ppm compared to the terminal see Supporting Information, Figures S1 and S2). amide proton. In the high-field region of the spectrum the The dried active ester 72+ readily reacts with the water-free expected singlet of the acetyl group and quartet/triplet pattern amino ester 52+ after deprotonation of its amino function with of the ethyl ester group are observed at 2.39 and 4.67/1.59 ppm P1-tBu at ambient conditions in reasonable reaction times (4 with correct integral ratios, respectively. Despite the over- h). Removal of residual crystal water in the starting materials lapping of several signals full assignment of all 1H and 13C was accomplished via storage of the respective compounds resonances was possible using 1H13C correlation spectroscopy. 12950 dx.doi.org/10.1021/ic5020362 | Inorg. Chem. 2014, 53, 12947−12961 58 | 3 RESULTS AND DISCUSSION Inorga nic Chemistry Article Scheme 1. Pentafluorophenylester (OPfp Ester) Formation 1589 cm−1. The PF −6 counterions are responsible for a broad of Ruthenium Amino Acid 1(PF6)2 and Its Acetyl Amide intense band at 840 cm −1. 6(PF6)2 Leading to 7(PF6)2 and 8(PF6)2 and Subsequent Spectroscopic Properties of Mono- and Dinuclear Amidation of 8(PF ) with tert-Butylamine to 9(PF ) a6 2 6 2 Amides. Both the mono- and the dinuclear bis(terpyridine)- ruthenium(II) complexes 42+ and 34+ exhibit a characteristic 1MLCT transition in the UV−vis/NIR (NIR = near-infrared) absorption spectrum at ∼500 nm (Figure 2). For the mononuclear complex 42+ this band is located at 492 nm in good agreement with wavelengths observed for similar bis(terpyridine)ruthenium(II) complexes carrying amido- and carboxylic acid functionalities.31,37,57 For the dinuclear system 34+ this band is significantly shifted bathochromically to 504 nm. This shift can be attributed to the enhanced push−pull situation caused by the additional charge-carrying complex fragment on the one hand and to the enlarged conjugated aromatic π system on the other, both lowering the energy difference between the highest occupied and the lowest unoccupied orbital (HOMO−LUMO gap). This also affects the extinction coefficient of the 1MLCT band of 34+. The band is shifted hyperchromically and cannot be described as a simple superposition of two similar but independent bis(terpyridine)- ruthenium-based chromophores. The intraligand π−π* tran- sitions in the UV region of the absorption spectra of 42+ and 34+, on the other hand, are very similar in shape and position to those of 34+ being roughly twice as intense as those of 42+, which is in very good agreement with the doubled number of terpyridine ligands present in 34+. DFT calculations employing B3LYP as functional and LANL2DZ as basis set with acetonitrile as solvent in a polarized continuum model (IEFPCM) support the spectro- scopic observations and assignments. The visible region is dominated mainly by two transitions, one originating from RuII → tpy-CO transitions showing up at 490 nm for 34+ and at 476 nm for 42+, consistent with the trend of the experimental 1MLCT absorption maxima. The other band is based on transitions from RuII into the more electron-rich tpy-NH ligands and is consequently found at higher energies (435 nm for 34+, 431 nm for 42+). This is in good agreement with the observed high-energy shoulders in the 1MLCT bands for both compounds. Both mononuclear 42+ as well as dinuclear 34+ are emissive at room temperature in fluid solution with emission quantum yields in the range of other bis(terpyridine)ruthenium(II) aAtom numbering for NMR assignment included. amino acid derivatives (Table 1, Figure 2). The quantum yield of dinuclear 34+ is lower than that of 42+ by a factor of 2. This might be attributed to the presence of a strongly polarized ESI+ mass spectra further confirm the formation of 34+. Peaks amide proton in the bridging ligand, which could allow for a attributable to [M-PF +6] , [M-2PF6] 2+, [M-3PF −H]2+6 , and [M- more efficient radiationless deactivation pathway. The emission 4PF6−H]3+ dominate the mass spectrum. Presumably the energy of 34+ is shifted slightly bathochromically with respect to proton that is lost is the bridging amide proton since its acidity 42+ matching the trend in the absorption spectra, which again is substantially increased due to the neighboring positively supports the assumption of a smaller HOMO−LUMO gap in charged complex subunits (vide infra). Lower intensity signals 34+. of fragments lacking the acetyl and/or ethyl groups are While the asymmetric shape of the emission band of 42+ at observed as well. Since no other evidence for cleavage of the room temperature is typical for a ruthenium-based emission, terminal amide and/or ester could be found, this fragmentation the band shape of 34+ is significantly different: it is more is believed to occur just during desolvation in the aerosol or symmetric and has a plateaulike maximum with nearly during the ionization process. IR spectroscopy also reveals the unchanged emission intensity over a range of 20 nm (Figure integrity of the dinuclear complex 34+. The NH and OH 2). On the other hand, the low-temperature emission spectra of stretching vibrations from the amide groups and residual water 42+ and 34+ in a solid nPrCN matrix have essentially the same show up at 3407 and 3649 cm−1. The ester and amide I shape with maxima of 657 and 660 nm, respectively, and a carbonyl stretching vibrations appear as overlapping bands pronounced shoulder at ∼720 nm originating from a vibronic between 1723 and 1691 cm−1. Additionally the amide groups progression (Figure 3). We attribute this unusual room- show typical NH deformation bands (amide II) at 1604 and temperature emission behavior to the coexistence of two 12951 dx.doi.org/10.1021/ic5020362 | Inorg. Chem. 2014, 53, 12947−12961 Section 3.1 | 59 Inorganic C hemistry Article Scheme 2. Amide Coupling of Amino Acid Ester 5(PF6)2 and Acetyl Amido Acid Pentafluorophenyl (OPfp) Ester 7(PF6)2 Giving Dinuclear Complex 3(PF6)4 and Acylation of 5(PF6)2 Leading to the Reference Compound 4(PF a 6)2 aAtom numbering of 3(PF6)4 for NMR assignment included. Figure 1. 1H NMR spectrum of 3(PF6)4 in CD3CN (lower), aromatic region (upper). emissive triplet states in 34+. This is in good agreement with the emission lifetimes of both complexes at room temperature. While 42+ exhibits an essentially monoexponential excited-state decay (the second component with 5% relative intensity is likely due to a strongly emissive but otherwise elusive impurity), the excited-state decay of 34+ is clearly biexponen- tial.86 The room-temperature emission spectrum of the dinuclear Figure 2. Experimental UV−vis absorption and normalized emission complex 34+ can be fit by a simple superposition of two bands spectra of 34+ (upper) and 42+ (lower) at room temperature in mimicking the emission band shape of a mononuclear complex. deaerated CH3CN including oscillator strengths of computed optical This was accomplished using the emission spectrum of 42+ transitions (time-dependent DFT: B3LYP, LANL2DZ, IEFPCM: twice at appropriate energies (676 and 705 nm, see Figure 4). CH3CN). The quality of this fit using weighing fractions of 71:29 for the two components, as indicated by the different measured Furthermore, we were interested in the dependence of the emission lifetimes (Table 1), compared to the emission shape of the emission spectrum of 34+ and hence the ratio of spectrum of 34+ is remarkable. This allows us to assign the the emitting states as a function of the excitation wavelength 676 nm emission to τ = 24 ns (71%) and the low energy (Supporting Information, Figure S11). The emission intensity emission (705 nm) to τ = 44 ns (29%). follows that of the absorption spectrum, and the band shape is 12952 dx.doi.org/10.1021/ic5020362 | Inorg. Chem. 2014, 53, 12947−12961 60 | 3 RESULTS AND DISCUSSION Inorga nic Chemistry Article Table 1. Absorption and Emission Properties in Deaerated CH3CN at Room Temperature and Emission Properties of 3 4+ and 42+ in Deaerated nPrCN at 77 K λ 1max ( MLCT) (ε) λEm. (λExc.) at 298 K λEm. (λExc.) at 77 K Φa τb (contribution) 42+ 492 (22 100) 678 (492) 657 (499) 5.9 × 10−4 21 (95); 58 (5) 34+ 504 (63 000) 684 (504) 660 (507) 3.2 × 10−4 24 (71); 44 (29) aQuantum yields Φ are determined at room temperature and given as fraction of emitted photons per absorbed photons. bEmission lifetimes τ were determined at the respective emission maxima (λmax/nm; ε/M −1 cm−1; λEm./nm; λExc./nm; τ/ns, contribution/%). be expected also at low temperatures. Consequently, the two emissive states likely involve RuII → bridge-tpy-CO triplet states of both RuII sites due to the spatial proximity of the involved centers. Although the real electronic situation certainly is more complicated the simplified one-electron orbital representation in Scheme 3 helps to illustrate the processes leading to the observed dual emission. Four different excited states involving the bridging ligand and the two ruthenium centers of 34+ can be thought of according to this diagram. In this simple picture the four conceivable triplet states can be regarded as a RuIIIRuII Figure 3. Normalized emission spectra of 34+ (red line) and 42+ (black mixed-valent system with a radical anion as bridging ligand. line) at 77 K in butyronitrile. Scheme 3. Schematic Illustration of the Four 3MLCT and 3CS Excited States of 34+ Involving the Bridging Liganda Figure 4. Normalized emission spectrum of 34+ at room temperature in deaerated CH3CN (red line), emission spectrum of 4 2+ (dashed lines) shifted to λmax = 676 nm (contribution: 71%) and 705 nm (contribution: 29%), and their sum (solid black line). The blue vertical line indicates the detection wavelength of the emission lifetime measurements. independent from the irradiation energy. As the relative abundances of the two emissive species are obviously independent from λexc, the two excited states are in thermal equilibrium in 34+. As at 77 K only single emission is observed, both emissive states must be connected via a reaction path on the triplet hypersurface with a very low activation barrier to allow for thermal equilibration at room temperature and at 77 K state prior to emission.87 To allow for a rapid thermal exchange between the two relevant excited states even at low temperatures the transition between the two emissive states may only involve minor geometric changes. As an exciton transfer between the spatially separated 3MLCT states involving RuIII(tpy·−-CONH) and RuIII(tpy·−-COOEt) requires the reorganization of various bond lengths the activiation barrier between such two states is expected to be high. Accordingly, this process is unlikely to occur rapidly at 77 K in a frozen matrix, and these two states are ruled out as an origin for the dual emission. If such 3MLCT states with a large activation barrier in between were involved in aThe two electron configurations marked in red are most likely those the emission process dual rather than a single emission would involved in the room temperature emission of 34+. 12953 dx.doi.org/10.1021/ic5020362 | Inorg. Chem. 2014, 53, 12947−12961 Section 3.1 | 61 Inorganic C hemistry Article Two of these triplet states include an odd electron on the tpy- absorption spectra (Figure 5) a two-step process is observed NH fragment of the bridge with one having the character of an with two independent sets of isosbestic points, namely, at 512, MLCT state (3MLCT2) and the other one having charge- separated state character (3CS2). Because of the electron- donating effect of the −NH functionality their energies are substantially higher than those of the other two states involving the OC-tpy·− moiety (3CS1 and 3MLCT1). Consequently only the latter are relevant for excited-state emission according to Kasha’s rule.87 Electron transfer between the two ruthenium centers connects the two excited states 3CS1 and 3MLCT1, which are thus valence-isomeric states. Since 3CS1 features a larger distance between the sites of the excited electron and the RuIII center, recombination/relaxation to the ground state might be slower. This fits to the assignment of a larger lifetime for 3CS (τ = 44 ns) as compared to that of 31 MLCT1 (τ = 24 ns). The lowest-energy excited triplet state of 34+ was modeled by DFT calculations (B3LYP, LANL2DZ). Its spin density is localized on the bridging OC-tpy and the adjacent RuIII center (Supporting Information, Figure S12), which agrees with studies on the site of the first oxidation (experimental and theoretical, vide infra) and reduction (theoretical, vide infra). Hence triplet 34+ is described as an excited-state mixed-valent system ([RuII(tpy-NHCO-tpy·−)RuIII]/[RuIII(tpy-NHCO- Figure 5. UV−vis absorption spectra of 34+ in dry CH3CN upon tpy·−)RuII]) of Robin−Day class II exhibiting substantial titration with a solution of phosphazene base P1-tBu in CH3CN 3+ electronic coupling after optical population of a RuII-bridge·−- (upper) 0 equiv → 1 equiv leading to 3-H , (lower) 1 equiv → 2.52+ RuIII state. equiv leading to 3−2H . Arrows indicate most dominant spectral Dual emission of polypyridine ruthenium(II) complexes has changes. been observed very rarely. In mononuclear heteroleptic complexes it usually only arises with electronically very different 432, 343, 318, 297, 240, and 225 nm for the first step and at π-accepting ligands such as in [Ru(bpy) (phen-4-R)]2+2 (phen = 519, 425, 326, 315, 306, 284, 239, and 223 nm for the second 1,10-phenanthroline, R = phenylalkynyl)88 allowing for two one. This observation is straightforwardly interpreted as the 3MLCT states with a high activation barrier in between so that stepwise deprotonation of the complex with the first proton both excited states emit simultaneously at room temperature abstraction occurring at the strongly polarized bridging amide and at 77 K.89−91 Alternatively the presence of 3MLCT states as giving 3-H3+ and the second one at the terminal NHCOCH3 well as intraligand CT states (3ILCT) can be responsible for amide generating 3−2H2+. Notably, the second deprotonation dual emission in bis(tridentate)ruthenium(II) complexes.92 In is not accessible in the presence of water. The bathochromic dinuclear ruthenium(II) complexes dual emission has also been shift of the 1MLCT absorption band from 504 nm in 34+ to 533 observed previously based on two 3MLCT states involving nm in 3−2H2+ upon deprotonation is reflected by a color either phen or bpy as accepting ligands.93 To the best of our change from red to purple (vide supra) and can be traced back knowledge no similar observation of dual emission of dinuclear to changes in the geometry of the bridging ligand. As suggested complexes originating from two RuIII-tpy·− triplet states by DFT calculations (B3LYP, LANL2DZ, IEFPCM: acetoni- involving the bridging ligand has been reported before. For a trile; vide infra) the tpy-NHCO-tpy bridge planarizes with series of dinuclear bis(terpyridine)ruthenium(II) complexes, dihedral angles at the bridging amide of ∼0° after however, with either back-to-back or para-phenylene linkage deprotonation. This allows for a stronger π conjugation within ([(4′-tolyl-tpy)Ru(tpy-(C6H4)n-tpy)Ru(tpy-4′-tolyl)]4+ (n = 0, the bridge leading to an enlargement of the chromophore and a 1, 2) it was shown that partial charge delocalization within the lowering of the ligand-based LUMO energies. Additionally, the excited triplet state is responsible for a substantial extension of donor strength of the N-substituted terpyridine of the bridge is the luminescence lifetime (up to τ = 570 ns, n = 0) along with a increased raising the energy of the ruthenium-based HOMO. bathochromic shift of the emission just as observed in the case Support for a stepwise deprotonation mechanism is also of 34+.94 obtained from 1H NMR spectroscopy (Figure 6). Upon Acid−Base Chemistry of the Dinuclear Amide 34+. addition of 1 equiv of P -tBu to a solution of 34+1 in CD3CN While the acid−base chemistry of various derivatives of the the resonance of the bridging amide proton at 10.5 ppm mononuclear bis(terpyridine)ruthenium(II) amino acid has disappears, and several other resonances are shifted significantly been previously discussed,19,31 new reactivity arises from the with respect to the spectrum of 34+. Major changes are dinuclear complex 34+ with two amides and a 4+ charge. The observed for the resonances of the bridging ligand tpy-NHCO- strongly polarizing effect of the 2-fold positively charged tpy with the resonances of the tpy-NH fragment being shifted complex fragments on the bridging amide renders its proton downfield, while the resonances of the tpy-CO fragment are significantly more acidic so that it can be readily abstracted found further upfield. This can be explained considering the using mild bases such as aliphatic tertiary amines in H2O/ stronger electron-donating effect of tpy-N − compared to tpy- CH3CN mixtures. The two possible NH deprotonation NH increasing the electron density in this terpyridine. On the reactions of 34+ have been studied via NMR and UV−vis other side, the lowered dihedral angle (from −25° to 0°) absorption spectroscopy employing the strong phosphazene between the carbonyl group and the proximal terpyridine base P1-tBu under water-free conditions. In the UV−vis increases the overlap of the π orbitals of these two fragments 12954 dx.doi.org/10.1021/ic5020362 | Inorg. Chem. 2014, 53, 12947−12961 62 | 3 RESULTS AND DISCUSSION Inorga nic Chemistry Article Figure S13). The slightly higher oxidation potential of 34+ compared to that of 42+ may be attributed to the unfavorable charge accumulation in 34+ (double oxidation affords a 6-fold positive charge). Interestingly, no separation of the oxidation waves of the two ruthenium centers is observed indicating no or only weak interaction between the metal sites using [nBu4N][PF6] as electrolyte 95−97 although potential differences are poor measures of electronic coupling.98−101 Additionally 42+ exhibits four one-electron reduction waves, with the first two being reversible and the second two being quasireversible when examined individually. These are attributed to tpy/tpy− and tpy−/tpy2− reductions starting with the acceptor-substituted tpy-COOEt ligand. In contrast, 34+/2+ shows only one reversible reduction wave, which accounts for a transfer of two electrons (referenced against internal ferrocene). It occurs essentially at the same potential as the first reduction of the mononuclear system 42+/+ (−1.49 V vs −1.46 V) and therefore is attributed to tpy/tpy− reductions of both tpy-CO ligands. The second 1 4+ reduction is a quasireversible two-electron reduction (3 2+/0). All Figure 6. H NMR spectra of 3 in CD3CN (upper), after addition of 3+ further reductions overlap significantly so that a clear separation1 equiv of phosphazene base P1-tBu (center; 3-H ) and after addition of 2 equiv of phosphazene base P1-tBu (lower; 3−2H2+). Arrows into individual reduction waves is impossible. indicate shifts upon deprotonation. The irreversible peak at ca. −0.5 V is a common feature of the cyclic voltammograms (CVs) of both complexes under resulting in a stronger −M e ect of the carbonyl group. Upon study. It only arises in the CVs after reducing the respectiveff addition of a second equivalent of base the resonance of the compounds at potentials below −2.0 V (quasi-reversible). terminal amide proton disappears, and all aromatic signals are Hence, it arises from the reoxidation of follow-up products of shifted upfield, except for those of the terminal tpy-COOEt the reduced or doubly reduced state (vide infra for detailed ligand, which remain essentially unaltered. This is in agreement discussion). with an overall increase of the electron density within the As can be seen from the oxidation potentials of 4 2+ and 34+ complex upon deprotonation of the terminal amide. The (Table 2), a strong oxidant is required to perform the oxidation II III pronounced acidity of the bridging amide will be relevant for of Ru → Ru . Only few chemical redox reagents such as Ce IV the ground- and excited-state redox potentials of 34+ as well. in acidic aqueous solution (E1/2 = 1.3 V in HClO4, 0.88 V in 63 Redox Properties of Mono- and Dinuclear Amides. H2O) and the tris(2,4-dibromophenyl)aminium radical 102 The cyclic voltammograms of 34+ and 42+ in CH CN have a cation in acetonitrile (E1/2 = 1.14 V) 63 are capable to do 3 very similar shape (Figure 7, Table 2). Both complexes show a so in a clean fashion. Reproducible UV−vis spectroscopic examination of the oxidation of 42+ and 34+ to 43+ and 36+, respectively, was only possible in 0.5 M H2SO4(aq) employing excess Ce(SO4)2 as oxidant (Figure 8). A set of six isosbestic points is observed for the oxidation of 34+ to 36+ at 556, 421, 338, 298, 285, and 267 nm, indicative of a clean transformation without accumulation of 35+. Even though an excess of oxidant is used, no additional band at ∼400−420 nm is observed originating from remaining CeIV, which is obviously consumed entirely immediately after addition. The 1MLCT absorption band of 34+ disappears completely, while a new broad and weak band appears with a maximum at 574 nm and a shoulder at ∼720 nm. The disappearance of the 1MLCT band indicates the complete oxidation of both RuII centers to RuIII under these conditions. The new band is consequently ascribed to a 1LMCT transition from the donor-substituted tpy-NH ligand to RuIII. Its intensity is rather low compared to, for example, the Figure 7. Cyclic voltammograms of 1 mM 42+ (upper) and 34+ (lower) 1LMCT of [(HOOC-tpy)RuIII(tpy-NH 3+2)] due to the less- in CH3CN with 0.1 M [nBu4][NPF6] as supporting electrolyte + pronounced donor effect of −NHAc as compared to that ofreferenced against the FcH/FcH couple. The first oxidation and −NH .19,57 Performing the oxidation of 42+ to 43+ under the reduction waves are shown individually. 2 same conditions proved to be difficult since on the time scale of recording of the UV−vis absorption spectrum (minutes) after reversible oxidation wave at ∼0.9 V versus FcH/FcH+. For 42+ partial oxidation with CeIV substantial decomposition of the this wave represents the one-electron oxidation of RuII to RuIII product was observed (absence of isosbestic points, loss of at 0.85 V. In 34+ both RuII centers are oxidized virtually at the intensity). Only by addition of 10 equiv of oxidant followed by same potential, leading to a two-electron oxidation wave at 0.91 rapid measurement a reproducible spectrum of 43+ could be V (referenced against 2 equiv of ferrocene as internal standard obtained (Figure 8, lower). It resembles that of the fully in the square-wave voltammogram, Supporting Information, oxidized dinuclear complex 36+ (1LMCT band, maxima at 590 12955 dx.doi.org/10.1021/ic5020362 | Inorg. Chem. 2014, 53, 12947−12961 Section 3.1 | 63 Inorganic C hemistry Article Table 2. Ground- and Excited-State Electrochemical Properties of 42+ and 34+ in 0.1 M [nBu4][NPF6]/CH3CN at Room Temperaturea Eox (Ru II/RuIII) Ered,1 (tpy/tpy −) Ered,2 (tpy/tpy −) E* (RuII/RuIII box ) Er*ed,1 (tpy/tpy −)c 42+ 0.85 (68) −1.46 (73) −1.86 −1.04 0.43 34+ 0.91 (84, 2e−) −1.49 (81, 2e−) −1.78 (2e−) −0.97 0.39 aThe peak-to-peak separations ΔEpp of the first oxidation and reduction waves are given in parentheses (E, V vs FcH/FcH+ (E1/2 (FcH/FcH+) = 0.40 V vs SCE), ΔE b cpp, mV). E*ox = Eox − E00. Er*ed = Ered + E00. E00 determined from emission spectra at 77 K. Figure 8. (upper) UV−vis absorption spectra of 34+ in 0.5 M H2SO4(aq) upon titration with a solution of Ce(SO4)2 in 0.5 M H2SO4(aq) (0 equiv → approximately 8 equiv). Arrows indicate spectral changes. (lower) UV−vis absorption spectra of 42+ and 43+ for comparison, obtained under the same conditions. Dashed lines indicate spectra of 34+ and 42+, and bold lines show RuIII complexes 36+ and 43+. and 739 nm) once again underlining the chemical similarity of 42+ and 34+. The mixed-valent RuII−RuIII species 35+ is obtained in a statistical mixture with 34+ and 36+ due to facile disproportio- nation (2 RuIIRuIII ⇌ RuIIRuII + RuIIIRuIII; statistical ratio of 1:2:1 for 34+:35+:36+). During the oxidation of 34+ to 36+, no band attributable to an intervalence charge-transfer (IVCT) transition is observed in the NIR region of the spectrum up to Figure 9. DFT (B3LYP, LANL2DZ, IEFPCM: acetonitrile) optimized 1350 nm (solvent absorption beyond 1350 nm prevented geometric structures of 34+, 35+, 36+, and 3-H3+ (upper to lower), recording at longer wavelengths). Oxidation of 34+ in including tpy-NHCO-tpy dihedral angles (deg), Ru−Ru distance (Å), acetonitrile with substoichiometric amounts of tris(2,4- and calculated spin densities of 35+ (doublet) and 36+ (triplet). dibromophenyl)aminium hexachloroantimonate did not show Contour value: 0.01, CH hydrogen atoms are omitted. the appearance of a new IVCT band in the range between 1000 and 3000 nm. This is in agreement with results from cyclic cation 35+ with the C-terminal ruthenium center being oxidized voltammetry and allows the interpretation of 35+ as a valence- in repeated attempts suggesting that the RuIIRuIII species is localized mixed-valent cation without observable electronic lower in energy. interaction between the ruthenium centers in different To further probe the hypothesis of noninteracting ruthenium oxidation states (Robin−Day class I).45,48,52,101,103 centers and to localize the electron-hole oxidation, experiments This interpretation is in accordance with DFT calculations were performed employing paramagnetic 1H NMR spectros- (B3LYP, LANL2DZ, IEFPCM: CH3CN) of the mixed-valent copy. 3 4+ was titrated with substoichiometric amounts of 35+ and the RuIIIRuIII 36+ complex (Figure 9). Spin density tris(2,4-dibromophenyl)aminium hexachloroantimonate as ox- calculations performed on 35+ localize the unpaired electron on idant in deuterated acetonitrile (Figure 10). Paramagnetic line the N-terminal ruthenium atom. Upon oxidation to the broadening and upfield shifts are observed only for certain RuIIIRuIII species 36+ spin density is found on both metal proton resonances, namely, those assigned to the N-terminal centers. Time-dependent calculations on 35+ performed on the bis(terpyridine)ruthenium(II) fragment. Especially pronounced same level of theory predict no intensity for IVCT transitions of is the shift of the resonances of H2″, H2‴, and the bridging NH any kind in the NIR spectral region. Geometry optimizations (highlighted with blue boxes in Figure 10), but also the proton failed to a ord the valence-tautomeric mixed-valent RuIIff RuIII resonances of H5″, H6″, H7″, H8″, H5‴, H6‴, H7‴, and H8‴ 12956 dx.doi.org/10.1021/ic5020362 | Inorg. Chem. 2014, 53, 12947−12961 64 | 3 RESULTS AND DISCUSSION Inorga nic Chemistry Article Figure 10. 1H NMR spectra of 34+ in CD3CN upon partial oxidation to 3 5+ with substoichiometric amounts of tris(2,4-dibromophenyl)aminium hexachloroantimonate (resonances of the corresponding amine in red frames). Blue frames highlight most significant spectral changes. respond to the partial oxidation of 34+. At higher concentrations greater and two lower than ge. This is in agreement with an of oxidant (>0.2 equiv) substantial broadening of the proton unpaired electron in the proximity of a low-spin RuII center. signals of the C-terminal complex fragment also becomes visible Interestingly, the signal occurring at the highest field is split by because then the concentration of the RuIIIRuIII complex 36+ a hyperfine coupling to one nitrogen atom giving a 1:1:1 triplet becomes spectroscopically significant due to disproportiona- (A (14N) = 15−18 G) suggesting that the unpaired electron is tion. The observation of the N-terminal bis(terpyridine)- significantly localized on one of the coordinating nitrogen ruthenium(II) fragment being the site of the first oxidation atoms. This is in agreement with the large g anisotropy of ∼Δg agrees with the theoretical results discussed above (Figure 9). = 0.05 and the substantial superhyperfine coupling to While a clean chemical oxidation of 34+ and 42+ is challenging ruthenium required to fit the spectrum (see Table 3), which to accomplish, the ligand-based reductions can easily be carried out using an acetonitrile solution of decamethylcobaltocene Table 3. The g Values and Hyperfine and Superhyperfine (E = −1.91 V).63 The ligand-centered radicals generated Coupling Constants A of the Unpaired Electron in 4+ and 33+1/2 upon addition of 0.9 equiv of CoCp* are examined using EPR Obtained by Simulation of the Experimental Spectra2 spectroscopy after rapid-freezing to 77 K (Figure 11). The EPR Recorded in Dry CH3CN at 77 K Using EasySpin spectra of the singly reduced species 33+ and 4+ are strikingly g Δga A (99,1011,2,3 1,2,3 Ru)b A 14 b1,2,3 ( N) similar. Both show a rhombic signal pattern with one g value 4+ 2.0045, 1.9885 1.9550 0.0495 2, 10, 24 1, 3, 18 33+ 2.0057, 1.9892, 1.9580 0.0477 2, 8, 15 3, 2.5, 15 aΔg = g1 − g . b3 (A, G). is in the range of other nitrogen-based radicals coordinated to a ruthenium(II) ion.39,40 DFT calculations (B3LYP, LANL2DZ, IEFPCM: acetonitrile) support this interpretation: the spin density of the dinuclear complex 33+ is calculated to be spread mainly on the central pyridyl ring of the bridging tpy-CO ligand with a minor contribution from the coordinated ruthenium center (Supporting Information, Figure S12). Interestingly, when measuring UV−vis absorption spectra while carrying out the reduction of 42+ and 34+ with up to 4 equiv of CoCp*2 a clean transition with isosbestic points very similar to those observed upon deprotonation is obtained (Supporting Information, Figure S14). Furthermore, the spectra after addition of an excess of reductant (2 equiv for 2+ 4+ Figure 11. X-band EPR spectra of 33+ (black) and 4+ (red) in dry 4 , 4 equiv for 3 ) resemble those of the deprotonated species CH3CN at 77 K after reduction with 0.9 equiv of CoCp*2 including 4-H + and 3-H3+ (see Figure 5). This cannot be interpreted as simulations. stepwise reductions of the respective complexes via 42+ → 4+ → 12957 dx.doi.org/10.1021/ic5020362 | Inorg. Chem. 2014, 53, 12947−12961 Section 3.1 | 65 Inorganic C hemistry Article 40 and 34+ → 33+ → 32+ → 3+ → 30, respectively, since this Table 4. Excited-State Stern−Volmer Quenching Constants should give several sets of isosbestic points in the UV−vis K a of 42+SV and 34+ by Various Quenchers, Fraction f b of absorption spectra. However, these observations can be easily Complex Accessible for Quencher, Bimolecular Quenching explained by a follow-up reaction after the initial reduction to Rate Constants k ,c and Quenching Fractions η dq q by 0.1 M 4+ and 33+ (rapid-freeze EPR), namely, the irreversible Quencher in CH3CN at Room Temperature reduction of protons to H2. Alternatively, a direct reduction 2+ 4+ of protons by decamethylcobaltocene yielding dihydrogen is 4 3 63 plausible.104 The proton source could be residual crystal water FcH (E1/2 = 0.00 V) Ksv( f) 246 (100) 432 (100) generating OH− 10 10 , which deprotonates the amides. Excess of kq 1.17 × 10 1.80 × 10 reductant is required due to varying amounts of water present ηq 0.96 0.98 2+ 4+ 125in 4 and 3 (see Experimental Section). Spectral changes of FcCOOMe (E1/2 = 0.30 V) Ksv( f) 160 (100) 9 similar shape have been observed previously with dinuclear kq 7.62 × 10 amide conjugates in our group upon addition of decamethylco- ηq 0.94 126 baltocene as reductant.39,40 The results obtained in the current Fc(COOMe)2 (E1/2 = 0.50 V) Ksv( f) 147 (100) 399 (100) 9 10 study suggest that also in those cases the bridging amide is kq 7.00 × 10 1.66 × 10 deprotonated in the presence of H2O (UV−vis) after initial ηq 0.94 0.98117 reduction of the complexes (rapid-freeze EPR). Ph-NMe2 (E1/2 = 0.39 V) Ksv( f) 1.8 (100) 14.4 (57) The same process, namely, deprotonation of 34+ 7 , is observed kq 8.6 × 10 6.00 × 10 8 when monitoring the addition of CoCp*2 via NMR spectros- ηq 15 34 a −1 b c −1 −1 d copy (slow time scale). No paramagnetic signal broadening KSV, M . f, %. kq, M s = KSV/τ. ηq, % = f Ksv[Q](1 + −1 appears upon addition of reductant to a solution of 34+ Ksv[Q]) . (Supporting Information, Figure S15). This would have been indicative of the presence of a radical anion especially because cally slightly uphill by 70 mV as in the reaction of 42+ with the expected line broadening of the proton resonances is larger Fc(COOMe)2 very efficient quenching of the emission of 4 2+ is for a ligand-based radical as compared to a metal-based radical still observed. This cannot be accounted for solely with a (Figure 10). The discrepancy between EPR results, on the one contribution from a reductive electron-transfer step from the hand, where the unpaired electron originating from a complex- ferrocene to the ruthenium complex. An additional feasible path based reduction can be observed, and NMR and UV−vis for radiationless deactivation is an energy transfer from the absorption spectroscopy, on the other hand, which reveal 3MLCT state of the Ru complex populating the nonemissive follow-up products of this initial reduction, is ascribed to the triplet excited state of the respective ferrocene deriva- different time scales of the respective experiments: because of tive.31,112−114 While stronger electron-withdrawing substituents the apparent instability of the radical formed in the presence of on ferrocene lower its reduction potential, they also stabilize its residual H2O, rapid freezing of the solution of 3 4+ a few seconds triplet state facilitating energy transfer. It is worth noting that after addition of CoCp*2 allows detection of an EPR signal, for both the mono- and the dinuclear complexes 42+ and 34+ while the NMR and UV−vis absorption measurements are very rapid quenching with formal bimolecular rate constants carried out several minutes after the addition of reductant, close to the diffusion limit (k = 1.9 × 1010, 298 K, CH CN)1153 allowing follow-up reactions to occur prior to measurement. is observed with all ferrocene derivatives without any detectable CV experiments further support this interpretation: scanning static quenching due to preorganization phenomena of the two just the potential range of the first ligand-based reduction of 34+ components in their respective ground states. This emphasizes delivers a reversible redox wave. Scanning the full solvent that a significant contribution of the excited-state quenching by window requires enough time to allow for further reactions of ferrocene originates from energy transfer. the complex after reduction. The reoxidation of the follow-up The choice of amines as electron sources for reductive species then occurs at ca. −0.5 V for both complexes 42+ and electron transfer quenching of the complexes 42+ and 34+ is 34+ shifted by approximately 1 V to more positive values limited due to the facile deprotonation of the amide protons (Figure 7). This lends further support to a reaction sequence (vide supra). Using N,N-dimethylaniline, which is not 34+ → 33+ → 3-H3+ + 1/2H2. sufficiently basic to abstract protons from the bridge of 34+ The excited-state redox potentials of the complexes 42+ and (pK = 5.1;116s substantiated by UV−vis absorption spectros- 34+ were calculated from Eox* = Eox − E00 and Ered* = Ered + E00 copy) as electron source (E1/2 = 0.39 V vs FcH/FcH+), 117 it is (Table 2).105,106 As expected the complexes become stronger possible to record Stern−Volmer plots for both complexes 42+ reductants and oxidants in the 3MLCT excited state. To probe and 34+ (Figure 12). While its quenching efficiency with respect the excited-state properties of 42+ and 34+ Stern−Volmer plots to 42+ is weak (2 orders of magnitude lower than that for with various ferrocene derivatives and amines as potential weak Fc(COOMe)2; Table 4) it is increased by almost 1 order of electron donors were recorded (Table 4).107 Employing magnitude in the 34+/amine pair. This cannot be explained just ferrocene, ferrocenecarboxylic acid methyl ester, and 1,1′- by the marginally increased driving force for the electron ferrocenedicarboxylic acid dimethyl ester as electron donors for transfer by 40 mV (Table 2). Additionally a curve bent the reductive quenching of the 3MLCT state of 42+, linear downward toward the x-axis is obtained when (I0/I − 1) is Stern−Volmer plots are obtained that show a clear dependence plotted against cquencher indicating a precoordination of the of the quenching rate kq from the redox potential of the quencher to the emissive species (Figure 12). An appropriate corresponding quencher (Supporting Information, Figure plot employing I0/(I0 − I) = ( f Ksv[Q])−1 + f−1 gives the S16).108−111 Lowering the driving force for the electron- fraction f of the emissive species actively taking part in the transfer step reduces the efficiency of the reductive quenching bimolecular quenching process as well as the Stern−Volmer significantly as expected from Marcus theory. Interestingly even constant Ksv (Supporting Information, Figure S17). 118 We if the electron-transfer step is estimated to be thermodynami- ascribe the substantial quenching fraction to an association of 12958 dx.doi.org/10.1021/ic5020362 | Inorg. Chem. 2014, 53, 12947−12961 66 | 3 RESULTS AND DISCUSSION Inorga nic Chemistry Article which facilitates reductive electron transfer from PhNMe2 to the excited complex 34+. Further studies will be conducted to elucidate this process, namely, whether 34+ can act as an electro- or photocatalyst for the reduction of protons from water and which role the proton of the bridging amide plays in such a process. ■ ASSOCIATED CONTENT *S Supporting Information Experimental procedures for the syntheses of 8(PF6)2 and 9(PF ) ; 1H and 136 2 C NMR spectra of 7(PF6)2, 8(PF6)2, and 9(PF 196)2; F NMR spectra of 7(PF6)2 and 8(PF6)2; CH-HSQC Figure 12. Stern−Volmer plots of the mono- and dinuclear complexes and CH-HMBC spectra of 3(PF6)4; emission spectra of 42+ (blue) and 34+ (red) employing N,N-dimethylaniline as quencher. 3(PF6)4 at different excitation wavelengths; figures of DFT- Plots were obtained using complex concentrations of c = 2 × 10−5 mol optimized geometries of 33+ and triplet 34+; UV−vis absorption L−1. spectra of 34+ and 42+ upon titration with CoCp*2; 1H NMR spectra of 34+ upon deprotonation; Stern−Volmer plots of 34+ the PhNMe2 nitrogen atom to the polarized proton of the and 42+ with different ferrocene derivatives; Cartesian bridging amide via a strong hydrogen bond. This facilitates coordinates of DFT-optimized geometries of 34+, 35+, 36+, inner-sphere reductive electron transfer into one of the two triplet 34+, and 33+. This material is available free of charge via excited states of the dinuclear complex as illustrated in Scheme the Internet at http://pubs.acs.org. 4. Similar observations of precoordination, especially via hydrogen bonds facilitating electron transfer from/to excited AUTHOR INFORMATION states, have been documented in the literature.119−124 ■ Corresponding Author Scheme 4. Reductive Electron Transfer from PhNMe2 to the *Fax: +49613127277. E-mail: katja.heinze@uni-mainz.de. Two Excited Triplet States of 34+ Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Funding This work was financially supported by the Deutsche Forschungsgemeinschaft (GSC 266, Materials Science in Mainz, scholarship for C.K.). Notes The authors declare no competing financial interest. ■ REFERENCES (1) Sauvage, J. P.; Collin, J. P.; Chambron, J. C.; Guillerez, S.; Coudret, C.; Balzani, V.; Barigelletti, F.; Cola, L.; de Flamigni, L. Chem. Rev. 1994, 94, 993−1019. (2) D’Alessandro, D. M.; Keene, F. R. New J. Chem. 2006, 30, 228− 237. (3) Chiorboli, C.; Indelli, M. T.; Scandola, F. Top. Curr. Chem. 2005, 257, 63−102. (4) Sun, L.; Hammarström, L.; Åkermark, B.; Styring, S. Chem. Soc. ■ CONCLUSION Rev. 2001, 30, 36−49.(5) Barigelletti, F.; Flamigni, L. Chem. Soc. Rev. 2000, 29, 1−12. The new dinuclear bis(terpyridine)ruthenium(II) complex 34+ (6) Flores-Torres, S.; Hutchison, G. R.; Soltzberg, L. J.; Abruña, H. with remarkable electronic symmetry despite an asymmetric D. J. Am. Chem. Soc. 2006, 128, 1513−1522. bridge was synthesized and fully characterized. No electronic (7) Akasaka, T.; Inoue, H.; Kuwabara, M.; Mutai, T.; Otsuki, J.; Araki, coupling is observed in the mixed-valent state 35+. However, the K. Dalton Trans. 2003, 815−821. similarity of the chemical environments of the ruthenium (8) Siebler, D.; Linseis, M.; Gasi, T.; Carrella, L. M.; Winter, R. F.; centers in 34+ enables a thermal electron transfer between RuII Förster, C.; Heinze, K. Chem.Eur. J. 2011, 17, 4540−4551. and RuIII in the triplet excited states of 34+. The two valence (9) Ozawa, H.; Haga, M.; Sakai, K. J. Am. Chem. Soc. 2006, 128, tautomers are observed via emission spectroscopy (dual 4926−4927. emission) and excited-state lifetime measurements. (10) Elvington, M.; Brown, J.; Arachchige, S. M.; Brewer, K. J. J. Am. Reduction of 34+ with CoCp* initially yields the radical Chem. Soc. 2007, 129, 10644−10645.2 (11) Takeda, H.; Koike, K.; Inoue, H.; Ishitani, O. J. Am. Chem. Soc. localized on the bridge. This radical further reacts to finally give 2008, 130, 2023−2031. the deprotonated complex 3-H3+ and presumably H . 3-H3+2 (12) Kumar, A.; Chhatwal, M.; Mondal, P. C.; Singh, V.; Singh, A. K.; and 3−2H2+ are prepared directly from strong bases and 34+. Cristaldi, D. A.; Gupta, R. D.; Gulino, A. Chem. Commun. 2014, 50, PhNMe2 is not basic enough to deprotonate 3 4+ but 3783. coordinates to the bridging NH group via a hydrogen bond, (13) O’Regan, B.; Graẗzel, M. Nature 1991, 353, 737−740. 12959 dx.doi.org/10.1021/ic5020362 | Inorg. Chem. 2014, 53, 12947−12961 Section 3.1 | 67 Inorganic C hemistry Article (14) Bolink, H. J.; Cappelli, L.; Coronado, E.; Gaviña, P. Inorg. Chem. (52) Robin, M. B.; Day, P. Adv. Inorg. Chem. 1968, 10, 247−422. 2005, 44, 5966−5968. (53) Brunschwig, B. S.; Creutz, C.; Sutin, N. Chem. Soc. Rev. 2002, (15) Kalyanasundaram, K. Coord. Chem. Rev. 1982, 46, 159−244. 31, 168−184. (16) Caspar, J. V.; Meyer, T. J. J. Am. Chem. Soc. 1983, 105, 5583− (54) D’Alessandro, D. M.; Topley, A. C.; Davies, M. S.; Keene, F. R. 5590. Chem.Eur. J. 2006, 12, 4873−4884. (17) Meyer, T. J. Pure Appl. Chem. 1986, 58, 1193−1206. (55) Crutchley, R. J. Angew. Chem. 2005, 117, 6610−6612; Angew. (18) Suzuki, K.; Kobayashi, A.; Kaneko, S.; Takehira, K.; Yoshihara, Chem., Int. Ed. 2005, 44, 6452−6454. T.; Ishida, H.; Shiina, Y.; Oishi, S.; Tobita, S. Phys. Chem. Chem. Phys. (56) Moser, C. C.; Keske, J. M.; Warncke, K.; Farid, R. S.; Dutton, P. 2009, 11, 9850−9860. L. Nature 1992, 355, 796−802. (19) Breivogel, A.; Kreitner, C.; Heinze, K. Eur. J. Inorg. Chem. 2014, (57) Breivogel, A.; Hempel, K.; Heinze, K. Inorg. Chim. Acta 2011, 2014, 10.1002/ejic.201402466. 374, 152−162. (20) Breivogel, A.; Wooh, S.; Dietrich, J.; Kim, T. Y.; Kang, Y. S.; (58) Sortino, S.; Petralia, S.; Di Bella, S. J. Am. Chem. Soc. 2003, 125, Char, K.; Heinze, K. Eur. J. Inorg. Chem. 2014, 2720−2734. 5610−5611. (21) Biner, M.; Buergi, H. B.; Ludi, A.; Roehr, C. J. Am. Chem. Soc. (59) Weinberg, D. R.; Gagliardi, C. J.; Hull, J. F.; Murphy, C. F.; 1992, 114, 5197−5203. Kent, C. A.; Westlake, B. C.; Paul, A.; Ess, D. H.; McCafferty, D. G.; (22) Lashgari, K.; Kritikos, M.; Norrestam, R.; Norrby, T. Acta Meyer, T. J. Chem. Rev. 2012, 112, 4016−4093. Crystallogr., Sect. C: Cryst. Struct. Commun. 1999, 55, 64−67. (60) Neidlinger, A.; Ksenofontov, V.; Heinze, K. Organometallics (23) Calvert, J. M.; Caspar, J. V.; Binstead, R. A.; Westmoreland, T. 2013, 32, 5955−5965. D.; Meyer, T. J. J. Am. Chem. Soc. 1982, 104, 6620−6627. (61) Bordwell, F. G.; Ji, G. Z. J. Am. Chem. Soc. 1991, 113, 8398− (24) Maestri, M.; Armaroli, N.; Balzani, V.; Constable, E. C.; 8401. Thompson, A. M. W. C. Inorg. Chem. 1995, 34, 2759−2767. (62) Fulmer, G. R.; Miller, A. J. M.; Sherden, N. H.; Gottlieb, H. E.; (25) Medlycott, E. A.; Hanan, G. S. Chem. Soc. Rev. 2005, 34, 133. Nudelman, A.; Stoltz, B. M.; Bercaw, J. E.; Goldberg, K. I. (26) Abrahamsson, M.; Jag̈er, M.; Österman, T.; Eriksson, L.; Organometallics 2010, 29, 2176−2179. Persson, P.; Becker, H.-C.; Johansson, O.; Hammarström, L. J. Am. (63) Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877−910. Chem. Soc. 2006, 128, 12616−12617. (64) Stoll, S.; Schweiger, A. J. Magn. Reson. 2006, 178, 42−55. (27) Schramm, F.; Meded, V.; Fliegl, H.; Fink, K.; Fuhr, O.; Qu, Z.; (65) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Klopper, W.; Finn, S.; Keyes, T. E.; Ruben, M. Inorg. Chem. 2009, 48, Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, 5677−5684. B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. (28) Breivogel, A.; Förster, C.; Heinze, K. Inorg. Chem. 2010, 49, P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; 7052−7056. Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, (29) Breivogel, A.; Meister, M.; Förster, C.; Laquai, F.; Heinze, K. T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. J. A.; Chem.Eur. J. 2013, 19, 13745−13760. Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, (30) Brown, D. G.; Sanguantrakun, N.; Schulze, B.; Schubert, U. S.; K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Berlinguette, C. P. J. Am. Chem. Soc. 2012, 134, 12354−12357. Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, (31) Heinze, K.; Hempel, K.; Beckmann, M. Eur. J. Inorg. Chem. N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; 2006, 2040−2050. Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; (32) Reynal, A.; Palomares, E. Eur. J. Inorg. Chem. 2011, 4509−4526. Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; (33) Sun, P.; Krishnan, A.; Yadav, A.; Singh, S.; MacDonnell, F. M.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Armstrong, D. W. Inorg. Chem. 2007, 46, 10312−10320. Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, Ö.; (34) Gong, L.; Mulcahy, S. P.; Harms, K.; Meggers, E. J. Am. Chem. Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian09, Soc. 2009, 131, 9602−9603. Rev. A.02; Gaussian, Inc.: Wallingford, CT, 2009. (35) Meggers, E. Chem.Eur. J. 2010, 16, 752−758. (66) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (36) Heinze, K.; Hempel, K.; Tschierlei, S.; Schmitt, M.; Popp, J.; (67) Dunning, T. H.; Hay, P. J. in Methods of Electronic Structure Rau, S. Eur. J. Inorg. Chem. 2009, 3119−3126. Theory; Schaefer, H. F., Ed.; Springer: Boston, MA, 1977. (37) Heinze, K.; Hempel, K.; Breivogel, A. Z. Anorg. Allg. Chem. (68) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270. 2009, 635, 2541−2549. (69) Wadt, W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284. (38) Heinze, K.; Hempel, K. Chem.Eur. J. 2009, 15, 1346−1358. (70) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299. (39) Dietrich, J.; Thorenz, U.; Förster, C.; Heinze, K. Inorg. Chem. (71) Miertus,̌ S.; Scrocco, E.; Tomasi, J. Chem. Phys. 1981, 55, 117− 2013, 52, 1248−1264. 129. (40) Dietrich, J.; Wünsche von Leupoldt, A.; Grabolle, M.; Resch- (72) Miertus,̌ S.; Tomasi, J. Chem. Phys. 1982, 65, 239−245. Genger, U.; Heinze, K. Eur. J. Inorg. Chem. 2013, 3009−3019. (73) Pascual-ahuir, J. L.; Silla, E.; Tunon, I. J. Comput. Chem. 1994, (41) Hurley, D. J.; Roppe, J. R.; Tor, Y. Chem. Commun. 1999, 993− 15, 1127−1138. 994. (74) Rosales, A.; Muñoz-Bascoń, J.; Loṕez-Sańchez, C.; Álvarez- (42) Bishop, B. M.; McCafferty, D. G.; Erickson, B. W. Tetrahedron Corral, M.; Muñoz-Dorado, M.; Rodríguez-García, I.; Oltra, J. E. J. Org. 2000, 56, 4629−4638. Chem. 2012, 77, 4171−4176. (43) Kise, K. J.; Bowler, B. E. Inorg. Chem. 2002, 41, 379−386. (75) Han, S.-Y.; Kim, Y.-A. Tetrahedron 2004, 60, 2447−2467. (44) Creutz, C.; Taube, H. J. Am. Chem. Soc. 1969, 91, 3988−3989. (76) König, W.; Geiger, R. Chem. Ber. 1970, 103, 788−798. (45) Kaim, W.; Klein, A.; Glöckle, M. Acc. Chem. Res. 2000, 33, 755− (77) Coste, J.; Le-Nguyen, D.; Castro, B. Tetrahedron Lett. 1990, 31, 763. 205−208. (46) Kaim, W.; Sarkar, B. Coord. Chem. Rev. 2007, 251, 584−594. (78) Freŕot, E.; Coste, J.; Pantaloni, A.; Dufour, M.-N.; Jouin, P. (47) Kaim, W.; Lahiri, G. K. Angew. Chem. 2007, 119, 1808−1828; Tetrahedron 1991, 47, 259−270. Angew. Chem., Int. Ed. 2007, 46, 1778−1796. (79) Carpino, L. A. J. Am. Chem. Soc. 1993, 115, 4397−4398. (48) Hush, N. Electrochim. Acta 1968, 13, 1005−1023. (80) Kisfaludy, L.; Schön, I. Synthesis 1983, 1983, 325−327. (49) Fuerholz, U.; Buergi, H. B.; Wagner, F. E.; Stebler, A.; Ammeter, (81) Atherton, E.; Cameron, L. R.; Sheppard, R. C. Tetrahedron J. H.; Krausz, E.; Clark, R. J. H.; Stead, M. J.; Ludi, A. J. Am. Chem. Soc. 1988, 44, 843−857. 1984, 106, 121−123. (82) Gagnon, P.; Huang, X.; Therrien, E.; Keillor, J. W. Tetrahedron (50) Zhang, L. T.; Ko, J.; Ondrechen, M. J. J. Am. Chem. Soc. 1987, Lett. 2002, 43, 7717−7719. 109, 1666−1671. (83) Beyermann, M.; Bienert, M.; Niedrich, H.; Carpino, L. A.; (51) Hush, N. Coord. Chem. Rev. 1985, 64, 135−157. Sadat-Aalaee, D. J. Org. Chem. 1990, 55, 721−728. 12960 dx.doi.org/10.1021/ic5020362 | Inorg. Chem. 2014, 53, 12947−12961 68 | 3 RESULTS AND DISCUSSION Inorga nic Chemistry Article (84) Neises, B.; Steglich, W. Angew. Chem. 1978, 90, 556−557; (115) McCleskey, T. M.; Winkler, J. R.; Gray, H. B. J. Am. Chem. Soc. Angew. Chem., Int. Ed. 1978, 17, 522−524. 1992, 114, 6935−6937. (85) Albericio, F.; Bofill, J. M.; El-Faham, A.; Kates, S. A. J. Org. (116) Burk, P.; Ruasse, M.-F.; Kaljurand, I.; Lilleorg, R.; Murumaa, Chem. 1998, 63, 9678−9683. A.; Mishima, M.; Koppel, I.; Koppel, I. A.; Leito, I. J. Phys. Org. Chem. (86) First interpretations of this observation were based on the 2013, 26, 171−181. presence of two different solvates of 34+, one potentially an H2O and (117) Andrew, T. L.; Swager, T. M. J. Org. Chem. 2011, 76, 2976− the other one a CH3CN solvate (via hydrogen bonds to the NH 2993. groups). Another interpretation included the simultaneous presence of (118) Principles of Fluorescence Spectroscopy; Lakowicz, J. R., Ed.; contact ion pairs and solvent-separated ion pairs. These contact ion Springer: Boston, MA, 1983. pairs were proposed as the complex cation 34+ and a strongly (119) Tyson, D. S.; Luman, C. R.; Zhou, X.; Castellano, F. N. Inorg. hydrogen-bonded PF −6 counterion in the “binding pocket” of the Chem. 2001, 40, 4063−4071. bridging amide functionality as observed for other dinuclear amide- (120) Zhao, G.-J.; Liu, J.-Y.; Zhou, L.-C.; Han, K.-L. J. Phys. Chem. B bridged complexes in our group.39,40 To verify these interpretations 2007, 111, 8940−8945. lifetime measurements with varying concentrations of H O as good (121) Huynh, M. H. V.; Meyer, T. J. Chem. Rev. 2007, 107, 5004−2 solvent and Cl− ions as competing counterion were performed. These 5064. lifetime experiments showed that the ratio of the two emissive species (122) Wenger, O. S. Chem.Eur. J. 2011, 17, 11692−11702. and their emission lifetimes are essentially independent from these (123) Nishino, T.; Hayashi, N.; Bui, P. T. J. Am. Chem. Soc. 2013, parameters excluding an interpretation of this kind. 135, 4592−4595. (87) Kasha, M. Discuss. Faraday Soc. 1950, 9, 14−19. (124) Wenger, O. S. Acc. Chem. Res. 2013, 46, 1517−1526. (88) Glazer, E. C.; Magde, D.; Tor, Y. J. Am. Chem. Soc. 2007, 129, (125) Abasq, M.-L.; Saidi, M.; Burgot, J.-L.; Darchen, A. J. Organomet. 8544−8551. Chem. 2009, 694, 36−42. (89) Chen, Y.; Zhou, X.; Wei, X.-H.; Yu, B.-L.; Chao, H.; Ji, L.-N. (126) Blankespoor, R. L. Inorg. Chem. 1985, 24, 1126−1128. Inorg. Chem. Commun. 2010, 13, 1018−1020. (90) Siebert, R.; Schlütter, F.; Winter, A.; Presselt, M.; Görls, H.; Schubert, U. S.; Dietzek, B.; Popp, J. Cent. Eur. J. Chem. 2011, 9, 990− 999. (91) Sen, R.; Koner, S.; Bhattacharjee, A.; Kusz, J.; Miyashita, Y.; Okamoto, K.-I. Dalton Trans. 2011, 40, 6952. (92) Song, L.-q.; Feng, J.; Wang, X.-s.; Yu, J.-h.; Hou, Y.-j.; Xie, P.-h.; Zhang, B.-w.; Xiang, J.-f.; Ai, X.-c.; Zhang, J.-p. Inorg. Chem. 2003, 42, 3393−3395. (93) Glazer, E. C.; Magde, D.; Tor, Y. J. Am. Chem. Soc. 2005, 127, 4190−4192. (94) Hammarström, L.; Barigelletti, F.; Flamigni, L.; Indelli, M. T.; Armaroli, N.; Calogero, G.; Guardigli, M.; Sour, A.; Collin, J.-P.; Sauvage, J.-P. J. Phys. Chem. A 1997, 101, 9061−9069. (95) Creutz, C. Prog. Inorg. Chem. 1983, 30, 1−73. (96) Richardson, D. E.; Taube, H. Coord. Chem. Rev. 1984, 60, 107− 129. (97) Crutchley, R. J. Adv. Inorg. Chem. 1994, 41, 273−325. (98) LeSuer, R. J.; Buttolph, C.; Geiger, W. E. Anal. Chem. 2004, 76, 6395−6401. (99) LeSuer, R. J.; Geiger, W. E. Angew. Chem. 2000, 112, 254−256; Angew. Chem., Int. Ed. 2000, 39, 248−250. (100) D’Alessandro, D. M.; Keene, F. R. Dalton Trans. 2004, 3950. (101) Winter, R. F. Organometallics 2014, 33, 4517−4536. (102) Yueh, W.; Bauld, N. L. J. Am. Chem. Soc. 1995, 117, 5671− 5676. (103) Hush, N. S. Prog. Inorg. Chem. 1967, 8, 391−444. (104) Koelle, U.; Infelta, P. P.; Graetzel, M. Inorg. Chem. 1988, 27, 879−883. (105) Rehm, D.; Weller, A. Ber. Bunsen-Ges. 1969, 73, 834−839. (106) Bock, C. R.; Connor, J. A.; Gutierrez, A. R.; Meyer, T. J.; Whitten, D. G.; Sullivan, B. P.; Nagle, J. K. J. Am. Chem. Soc. 1979, 101, 4815−4824. (107) Stern, O.; Volmer, M. Phys. Z. 1919, 20, 183−188. (108) Marcus, R. A. J. Chem. Phys. 1956, 24, 979−989. (109) Marcus, R. A. J. Chem. Phys. 1956, 24, 966−978. (110) Marcus, R. A. J. Chem. Phys. 1965, 43, 2654−2657. (111) Marcus, R. A. J. Chem. Phys. 1965, 43, 679−701. (112) Farlow, B.; Nile, T. A.; Walsh, J. L.; McPhail, A. T. Polyhedron 1993, 12, 2891−2894. (113) Hutchison, K.; Morris, J. C.; Nile, T. A.; Walsh, J. L.; Thompson, D. W.; Petersen, J. D.; Schoonover, J. R. Inorg. Chem. 1999, 38, 2516−2523. (114) Siemeling, U.; Vor der Brüggen, J.; Vorfeld, U.; Neumann, B.; Stammler, A.; Stammler, H.-G.; Brockhinke, A.; Plessow, R.; Zanello, P.; Laschi, F.; Fabrizi de Biani, F.; Fontani, M.; Steenken, S.; Stapper, M.; Gurzadyan, G. Chem.Eur. J. 2003, 9, 2819−2833. 12961 dx.doi.org/10.1021/ic5020362 | Inorg. Chem. 2014, 53, 12947−12961 Section 3.2 | 69 3.2 UNDERSTANDING THE EXCITED STATE BEHAVIOR OF CYCLOMETALATED BIS(TRIDENTATE)RUTHENIUM(II) COMPLEXES: A COMBINED EXPERIMENTAL AND THEORETICAL STUDY Christoph Kreitner, Elisa Erdmann, Wolfram W. Seidel and Katja Heinze Inorg. Chem. 2015, 54, 11088–11104. The visible absorption bands of the isomers [Ru(dpb-NHCOMe)(tpy-COOEt)]+ 1+ and [Ru(dpb-COOEt)(tpy-NHCOMe)]+ 2+(dpbH = 1,3- dipyridin-2-ylbenzene, tpy = 2,2′;6,2″- terpyridine) arise from mixed Ru → tpy/Ru → dpb MLCT excitations according to resonance Raman spectroscopy and DFT calculations. 2+ is phosphorescent (3MLCT state), while 1+ is nonemissive. Partial deactivation of the 3MLCT state of 2+ occurs via 3MC states 11 kJ mol−1 higher in energy, while the3MLCT state of 1+ is deactivated via a lower-lying 3LLCT state. Author Contributions The synthesis and characterization of the ruthenium complexes as well as all DFT calculations were performed by Christoph Kreitner. The resonance Raman spectroscopic studies were carried out by Elisa Erdmann and Wolfram Seidel at the University of Rostock, Germany. The manuscript was written by Christoph Kreitner (90 %) and Katja Heinze (10 %). Supporting Information for this article is found at pp. 184 (excluding Cartesian Coordinates of DFT-optimized geometries). For full Supporting Information, refer to http://pubs.acs.org/doi/suppl/10.1021/ acs.inorgchem.5b01151. Correction A correction to this article was published as C. Kreitner, E. Erdmann, W. W. Seidel and K. Heinze, Inorg. Chem., 2015, 54, 12046 rectifying a mistake in the evaluation of the redox potentials of 1+. This correction is printed after the original article. “This is an unofficial adaptation of an article that appeared in an ACS publication. ACS has not endorsed the content of this adaptation or the context of its use. Copyright 2015 American Chemical Society.” 70 | 3 RESULTS AND DISCUSSION This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes. Article pubs.acs.org/IC Understanding the Excited State Behavior of Cyclometalated Bis(tridentate)ruthenium(II) Complexes: A Combined Experimental and Theoretical Study Christoph Kreitner,†,‡ Elisa Erdmann,§ Wolfram W. Seidel,§ and Katja Heinze*,† †Institute of Inorganic and Analytical Chemistry, Johannes Gutenberg-University of Mainz, Duesbergweg 10-14, 55128 Mainz, Germany ‡Graduate School Materials Science in Mainz, Staudingerweg 9, 55128 Mainz, Germany §Institute of Chemistry, University of Rostock, Albert-Einstein-Straße 3a, 18059 Rostock, Germany *S Supporting Information ABSTRACT: The synthesis and characterization of the donor−acceptor substituted cyclometalated ruthenium(II) polypyridine complex isomers [Ru(dpb-NHCOMe)(tpy- COOEt)](PF6) 1(PF6) and [Ru(dpb-COOEt)(tpy- NHCOMe)](PF6) 2(PF6) (dpbH = 1,3-dipyridin-2-ylbenzene, tpy = 2,2′;6,2″-terpyridine) with inverted functional group pattern are described. A combination of resonance Raman spectroscopic and computational techniques shows that all intense visible range absorption bands arise from mixed Ru → tpy/Ru → dpb metal-to-ligand charge transfer (MLCT) excitations. 2(PF6) is weakly phosphorescent at room temperature in fluid solution and strongly emissive at 77 K in solid butyronitrile matrix, which is typical for ruthenium(II) polypyridine complexes. Density functional theory calculations revealed that the weak emission of 2(PF6) arises from a 3MLCT state that is efficiently thermally depopulated via metal-centered (3MC) excited states. The activation barrier for the deactivation process was estimated experimentally from variable-temperature emission spectroscopic measurements as 11 kJ mol−1. In contrast, 1(PF6) is nonemissive at room temperature in fluid solution and at 77 K in solid butyronitrile matrix. Examination of the electronic excited states of 1(PF6) revealed a ligand-to-ligand charge-transfer ( 3LL′CT) as lowest-energy triplet state due to the very strong push−pull effect across the metal center. Because of the orthogonality of the participating ligands, emission from the 3LL′CT is symmetry-forbidden. Hence, in this type of complex a stronger push−pull effect does not increase the phosphorescence quantum yields but completely quenches the emission. ■ INTRODUCTION has a long excited-state lifetime of 855 ns (in acetonitrile) due5 Polypyridine complexes of ruthenium have been studied to the spin-forbidden character of the luminescence. 1 A qualitatively similar picture of the excited-state mecha-extensively in the last 50 years. Especially, the photophysics 2+ nisms is gained for the meridionally coordinated 12 tridentate and photochemistry of their prototype [Ru(bipy)3] (bipy = ′ 2−5 analogue of [Ru(bipy) ] 2+, namely, [Ru(tpy) ]2+ (tpy = 2,2 -bipyridine are well understood. The visible range of the 3 22,2′;6′,2″-terpyridine). MLCT absorption occurs at 474 nm absorption spectrum is dominated by an intense metal-to-ligand slightly bathochromically and hyperchromically shifted (ε = charge transfer (MLCT) absorption from ruthenium d-orbitals 16 100 M−1 cm−1)13 due to the larger accepting π*-orbitals of into the low-lying antibonding π*-orbitals of the bipy ligands3,6 the terpyridine ligands. Upon ISC again 3MLCT states are with an absorption maximum at 452 nm and an extinction populated.14,15 In contrast to the bipy counterpart an efficient coefficient of 14 600 M−1 cm−1.4 The UV region is dominated deactivation pathway is available for this emissive 3MLCT state: by π−π* transitions within the aromatic ligands.4 Following because of the smaller N−Ru−N bite angles in this tpy complex Kasha’s rule7 rapid vibrational relaxation and internal compared to the parent bipy complex the orbital overlap of the conversion occur upon optical excitation populating the pyridine nitrogen lone pairs with the ruthenium d orbitals of lowest-energy 1MLCT state. From this state nearly quantitative the eg set (in idealized Oh symmetry) is lowered. The loss in intersystem crossing (ISC)8,9 onto the triplet hypersurface ligand-field splitting shifts d−d excited states (3MC states, MC occurs, which leads to population of the lowest-energy 3MLCT state.10 This state is highly phosphorescent10 at room Received: May 27, 2015 temperature (λem = 621 nm, ϕ = 0.095 in acetontrile) 11 and Published: August 6, 2015 © 2015 American Chemical Society 11088 DOI: 10.1021/acs.inorgchem.5b01151 Inorg. Chem. 2015, 54, 11088−11104 72 | 3 RESULTS AND DISCUSSION Inorga nic Chemistry Article Scheme 1. Bis(tridentate)ruthenium(II) and Iridium(III) Polypyridine Complexes a aBy Constable (A2+),15 van Koten (B2+, I+, J+, K+),20 Heinze (C2+, D2+, E2+),21−23 Hammarström (F2+),24 Ruben (G2+),25 Williams (H2+),26 and from this work (1+ and 2+). = metal-centered) into the energy regime of the 3MLCT states. 2-ylpyridine-2,6-diamine, ddpd).22 The increased σ-donor These MC states are thermally populated at room temperature strength of ddpd compared to tpy sufficiently separates the and lead to an efficient emission quenching in [Ru(tpy) ]2+ 32 MC states in the heteroleptic complex [Ru(ddpd)(tpy- (λ = 629 nm, ϕ < 1 × 10−5).16em At 77 K, the available thermal COOEt)]2+ D2+ (Scheme 1) from the 3MLCT states to allow energy does not suffice to overcome the activation barrier for [Ru(ddpd)(tpy)]2+ to be emissive (ϕ = 0.0045) despite its population of the 3MC states, and an intense emission is much lower emission energy (λem = 729 nm). 22,23 Gradually regained (ϕ = 0.48).13 tuning the vertical 3MLCT → 1GS transition energy within a Just as this undesirable side effect also the major advantage of series of structurally similar [Ru(ddpd)(tpy)]2+ complexes by bis(terpyridine)ruthenium(II) complexes over their bipyridine variation of appended functional groups decreases the emission analogues arises from their coordination geometry. The C2v quantum yield with decreasing emission energy following the symmetry of the core structure of this complex12 prevents the energy gap law as pointed out by Meyer and co-workers.30−32 formation of diastereomers even when heteroleptic complexes Similarly, Hammarström and co-workers used di(quinolin-8- bearing ligands with different functional groups are formed. yl)pyridine (dqp) as tridentate ligand forming six-membered Syntheses of similarly substituted bipyridine complexes usually chelate rings with ruthenium as metal center.24,33 The give mixtures of diastereomers that require elaborate methods homoleptic complex [Ru(dqp) ]2+ F2+2 (Scheme 1) is to be puri ed17fi or to be circumvented.18,19 phosphorescent at room temperature (ϕ = 0.02) with a Several successful approaches improve the emissive behavior remarkably long excited-state lifetime of 3.0 μs. Ruben and co- of bis(tridentate)ruthenium complexes by influencing the workers employed the carbonyl analogue of the ddpd ligand, energies of the relevant excited states.27 Introducing π- 2,6-di(2-carboxypyridyl)pyridine (dcpp) as chelating ligand accepting functional groups (−SO2R, −COOR, Scheme 1, with N−Ru−N bite angles of 90°. The homoleptic complex A2+, B2+) in the ligand backbone (typically in 4′-position) [Ru(dcpp)2]2+ G2+ (Scheme 1) exhibits an extraordinary high lowers the energy of the 3MLCT states while leaving the energy room-temperature emission quantum yield of 0.30 with a long of the 3MC states unaltered. This hinders the thermal excited-state lifetime of 3.3 μs.25 deactivation process to some extent and increases both Cyclometalation34,35 (i.e., isoelectronic substitution of a excited-state lifetimes and quantum yields of such compounds nitrogen atom for a carbanion in the coordination sphere (ϕ ≈ 1−5 × 10−4).15,28 Alternatively, introducing σ-donating around the metal) is discussed as another option for raising the functional groups in the ligand’s periphery directly influences 3MC states since the strong σ-donating effect of the anionic the energy of the 3MC states.15 They are shifted to higher carbon greatly increases the ligand field splitting.20 While for energies with respect to the 3MLCT states again hampering iridium(III) a large variety of highly phosphorescent cyclo- thermal depopulation of the latter. Combining both approaches metalated complexes are known,36−39 most cyclometalated yields excited-state lifetimes of up to 50 ns (Scheme 1, A2+)15 ruthenium(II) complexes are barely emissive at room temper- and quantum yields of up to 0.003 (Scheme 1, C2+)21,29 but ature.40−42 For tris(bidentate)iridium(III) complexes with always at the cost of a lowered excited-state energy.27 cyclometalating ligands of the type [Ir(bipy)n(ppy)3−n] n+ (n = The 3MC states are even more efficiently shifted to higher 1, 2; ppyH = 2-phenylpyridine) the excited-state mechanisms energies by widening the N−Ru−N bite angles. This is that are responsible for the efficient luminescence are well- achieved upon introduction of N−CH in between the pyridine understood.433 The emissive excited state of IrIII complexes is a rings of the terpyridine ligand (N,N′-dimethyl-N,N′-dipyridin- linear combination of a mixed 3MLCT/3LL′CT state (LL′CT = 11089 DOI: 10.1021/acs.inorgchem.5b01151 Inorg. Chem. 2015, 54, 11088−11104 Section 3.2 | 73 Inorganic C hemistry Article ligand-to-ligand charge-transfer state) and an energetically very uncertainty is estimated to be 15%. FD+ mass spectra were recorded similar ligand-centered (3LC) excited state due to the very high on a FD Finnigan MAT95 spectrometer. Electrospray ionization + ligand field splitting of IrIII combined with the ppy ligands. This (ESI ) and high-resolution ESI + mass spectra were recorded on a state is well-separated from other states so that emission Micromass QTof Ultima API mass spectrometer with analyte solutions becomes very efficient.26,44−46 Since most cyclometalated in acetonitrile. Elemental analyses were performed by the micro-analytical laboratory of the chemical institutes of the University of ruthenium complexes are essentially nonemissive at room Mainz. NMR spectra were obtained with a Bruker Avance II 400 temperature in solution it is much more difficult to obtain a spectrometer at 400.31 (1H) and 100.66 (13C) at 25 °C. Chemical profound understanding of the excited-state mechanisms in shifts δ [parts per million] are reported with respect to residual solvent these systems. Berlinguette and co-workers showed that the signals as internal standards (1H, 13C): CD3CN δ(1H) = 1.94 ppm, energy gap law is obeyed in complexes of type [Ru(bipy)2- δ( 13C) = 1.32 and 118.26 ppm.54 Electrochemical experiments were (ppy)]+ demonstrating that direct ISC onto the singlet performed with a BioLogic SP-50 voltammetric analyzer at a sample hypersurface followed by vibrational cooling is the dominant concentration of 1 × 10 −3 M using platinum wire working and counter deactivation pathway.41,42 electrodes and a 0.01 M Ag/AgNO3 reference electrode. Measure- Van Koten and co-worker recently discussed [Ru(tpy- ments were performed at a scan rate of 100 mV s −1 for cyclic ′ + ′ voltammetry experiments and at 10 mV s −1 for square-wave R)(pbpy-R )] (pbpyH = 6-phenyl-2,2 -bipyridine) and [Ru- + voltammetry experiments using 0.1 M [ nBu4N][PF6] as supporting (tpy-R)(dpb-R′)] (dpbH = 2,6-di(pyrid-2′-yl)benzene) com- electrolyte in acetonitrile. Potentials are given relative to the plexes and their application in dye-sensitized solar cells.20,47,48 ferrocene/ferrocenium couple (0.40 V vs standard calomel electrode Electron-accepting anchor groups (COOR) were appended at (SCE), E1/2 = 0.90 ± 5 mV under the given conditions). 55 EPR spectra either one of the two ligands or at both resulting in a series of were recorded on a Miniscope MS 300 X-band CW spectrometer weakly or nonemissive tridentate complexes (Scheme 1, I+, J+, (Magnettech GmbH, Germany). Values of g are referenced against and K+ 2+ ). A structurally similar weakly emitting iridium(III) Mn in ZnS as external standard (g = 2.118, 2.066, 2.027, 1.986, complex (Scheme 1, H2+) was synthesized by Williams and co- 1.946). Simulations were performed with the EasySpin program56 workers.26 Despite the fact that the energy gap law is obeyed package. A Horiba LabRAM HR Raman microscope was used for rRmeasurements with an object lens (10× NA 0.25) from Olympus. within these complex series luminescence quenching is Samples were optically excited with a red laser (633 nm, 17 mW, discussed to arise from thermal depopulation of the very low- HeNe-laser), green laser (532 nm, 50 mW, air-cooled frequency- lying 3MLCT states via 3MC states. However, the latter should doubled Nd:YAG-solid state laser), or blue laser (473 nm, 20 mW, air- be high in energy due to the strong σ-donor strength of the cooled solid-state laser). Samples were measured in acetonitrile cyclometalating ligand.20 This apparent discrepancy will be (Chemsolute, for HPLC) solution in capillary tubes (80 × 1.5 mm, addressed in this paper. Marienfeld-Superior). In this study we present an extension of our previous work Density functional theory (DFT) calculations were performed using the ORCA program package (version 3.0.2).57on tridentate polypyridine ruthenium complexes bearing both Tight convergence electron-donating amino and electron-withdrawing carboxylic criteria were chosen for all calculations (keywords TightSCF and 2+ 21,27 TightOpt, convergence criteria for the SCF part: energy change 1.0 ×acid functionalities (such as A , Scheme 1) into the field of 10−8 Eh, 1 − El. energy change 1.0 × 10−5 Eh, orbital gradient 1.0 ×cyclometalated complexes and elucidate the excited-state 10−5, orbital rotation angle 1.0 × 10−5, DIIS error 5.0 × 10−7; for deactivation mechanisms of these complexes in detail. In the geometry optimizations: energy change: 1.0 × 10−6 Eh, maximum isomeric [Ru(dpb-R′)(tpy-R)]+ complexes (1+: R = COOEt, R′ gradient 1.0 × 10−4 Eh/bohr, root-mean-square (RMS) gradient 3.0 × = NHCOMe; 2+: R = NHCOMe, R′ = COOEt) the position of 10−5 Eh/bohr, maximum displacement 1.0 × 10−3 bohr, RMS the functional groups with respect to the site of cyclometalation displacement 6.0 × 10−4 bohr). All calculations employ the resolution should have a strong impact on their electronic structure and of identity (Split-RI-J) approach for the coulomb term in combination excited-state ordering. The ground- and excited-state electronic with the chain-of-spheres approximation for the exchange term structures as well as excited-state dynamics are elucidated by a (COSX) where Hartree−Fock exchange is required. 58,59 Geometry − optimizations were performed using the GGA functional PBE 60,61 in combination of UV vis, electron paramagnetic resonance combination with Ahlrichs’ split-valence double-ξ basis set def2-SV(P) (EPR), resonance Raman (rR), and emission spectroscopies for all atoms except ruthenium, which comprises polarization functions and theoretical techniques to provide a better understanding of for all non-hydrogen atoms.62,63 For ruthenium a Stuttgart−Dresden the unexplained low-emission efficiencies in cyclometalated effective core potential (ECP, def2-SD) was combined with Ahlrich’s bis(tridentate)ruthenium(II) complexes. def2-TZVP basis set for the valence electrons.64,65 To account for solvent effects a conductor-like screening model (COSMO) modeling ■ EXPERIMENTAL SECTION acetonitrile was used in all calculations except for excited-stategradients.66 This proved to be particularly important for time- General Procedures. Chemicals were obtained from commercial dependent DFT calculations (TD-DFT) where gas phase calculations suppliers and used without further purification. Air- or moisture- lead to a substantial underestimation of excitation energies.67 The sensitive reactions were performed in dried glassware in an inert gas optimized geometries were confirmed to be local minima on the atmosphere (argon, quality 4.6). Acetonitrile and dichloromethane respective potential energy surface by subsequent numerical frequency were refluxed over CaH2 and distilled under argon prior to use. analysis (Nimag = 0). Toluene and xylenes were refluxed over sodium and distilled prior to Calculation of EPR parameters and TD-DFT calculations were use. Palladium precatalyst [Pd] 492 and the ligand precursors 1-bromo- performed based on the PBE/def2-SV(P)/ECP(def2-TZVP) opti- 3,5-dipyridin-2-ylbenzene LA,50 4′-chloro-2,2′:6′,2″-terpyridine LB,51 mized geometry of the respective complex employing the triple-ξ basis 4′-amino-2,2′:6′,2″-terpyridine LC,52 and ethyl 3,5-dibromobenzoate set def2-TVZP and several functionals with varying amounts of HF LD53 were synthesized following literature-known procedures. UV−vis exchange:68 PBE (0%), TPSSh (10%),69 B3LYP (20%),70 PBE0 spectra were recorded on a Varian Cary 5000 spectrometer in 1 cm (25%),71 and CAM-B3LYP (19−65%).72 The Douglas−Kroll−Hess cuvettes. Emission spectra were recorded on a Varian Cary Eclipse (DKH) relativistic approximation73−76 was used to describe relativistic spectrometer. Quantum yields were determined by comparing the effects in these calculations. The DKH keyword in ORCA automati- areas under the emission spectra on an energy scale recorded for cally invokes adjusted basis sets (TZV_DKH).77 At least 50 vertical solutions of the samples and a reference with matching absorbances transitions were calculated in TD-DFT calculations. The electron g (ϕ([Ru(bipy)3]Cl2) = 0.094 in deaerated CH3CN). 11 Experimental value and hyperfine coupling constants of the unpaired electron to the 11090 DOI: 10.1021/acs.inorgchem.5b01151 Inorg. Chem. 2015, 54, 11088−11104 74 | 3 RESULTS AND DISCUSSION Inorga nic Chemistry Article ruthenium atom and all atoms coordinated to ruthenium were removed under reduced pressure to give 0.95 g of crude N-acetyl-3,5- determined in EPR calculations. On the basis of the optimized 1GS di(pyridin-2-yl)aniline. After column chromatography on neutral Alox molecular geometries and the associated Hessian matrices excited-state (Brockmann II, 3% water (w/w), solvent ethyl acetate) the product gradients were calculated for the lowest 10 excitations at the B3LYP/ was obtained as colorless powder. Yield: 848 mg (2.90 mmol, 78%). def2-TZVP/DKH level of theory to generate excited-state displace- Anal. Calcd C18H15N3O (289.33): C, 74.72; H, 5.23; N, 14.52. Found: ments for the rR spectra simulation. The advanced spectra analysis C, 74.93; H, 4.98; N, 14.39%. MS(FD+): m/z (%) = 289.2 (100) tool provided with the ORCA program package (orca_asa)78,79 was [M]+, 579.4 (2) [2M+H]+. 1H NMR (CD2Cl2): δ [ppm] = 8.67 (ddd, employed to fit the absorption spectra of 1(PF6) and 2(PF6) and to 3JHH = 5 Hz, 4J 8 4 9HH = 2 Hz, 1 Hz, 2H, H ), 8.38 (t, JHH = 1 Hz, 1H, H ) simulate rR spectra. The Gibbs free energy was used to compare 8.27 (d, 4JHH = 1 Hz, 2H, H 2), 8.25 (s, 1H, NH), 7.79 (ddd, 3JHH = 8 relative energies of the different triplet states of the complexes under Hz, 4J 5 3 4 6HH = 1 Hz, 2H, H ), 7.74 (vtd, JHH = 8, JHH = 2 Hz, 2H, H ), study. Explicit counterions and/or solvent molecules were not taken 7.24 (ddd, 3JHH = 8 Hz, 5 Hz, 4JHH = 1.2 Hz, 2H, H 7), 2.12 (s, 3H into account in all cases. To reduce the computational cost methyl CH 13 13). C{ H} NMR (CD2Cl2): δ [ppm] = 169.3 (s, C 10), 157.0 (s, instead of ethyl groups were used throughout all calculations at the C4), 150.1 (s, C8), 140.9 (s, C3), 140.0 (s, C1), 137.4 (s, C6), 123.1 (s, ester moiety. C7), 121.3 (s, C9), 121.1 (s, C5), 119.2 (s, C2), 24.9 (s, C11). Synthesis of N-Acetyl-4′-amino-2,2′:6′,2″-terpyridine L2. Proce- Scheme 2a dure (a). XantPhos (123 mg, 213 μmol, 9 mol %) and Pd precatalyst [Pd]2 (39 mg, 53 mmol, 4.5 mol % based on Pd) were dissolved under argon in 15 mL abs. xylenes and left to stand. After 10 min 4′-chloro- 2,2′:6′,2″-terpyridine LB (616 mg, 2.30 mmol, 1 equiv), acetamide (151 mg, 2.56 mmol, 1.1 equiv), and sodium tert-butanolate (246 mg, 2.56 mmol, 1.1 equiv) dissolved in additional 15 mL of abs. xylenes were added, and the mixture was heated to reflux for 20 h. Drying of all reagents prior to use is necessary since the resulting amide is prone to hydrolysis under the given reaction conditions in the presence of traces of water. The same workup routine as for N-acetyl-3,5- di(pyridin-2-yl)aniline was followed. Column chromatography on neutral Alox (Brockmann II, 3% water (w/w), solvent gradient ethyl acetate/hexanes 1:3 → 3:1) afforded pure N-acetyl-4′-amino- 2,2′:6′,2″-terpyridine as off-white powder. Yield: 508 mg (1.75 mmol, 76%). Anal. Calcd C17H14N4O (290.32): C, 70.33; H, 4.86; N, 19.30. Found: C, 69.92; H, 4.81; N, 19.02%. MS(FD+): m/z (%) = 290.2 (100) [M+], 313.1 (10) [M + Na]+, 603.3 (2) [2M+Na]+. 1H NMR (0.5 mL of CD2Cl2 + 0.1 mL of deuterated dimethyl sulfoxide): δ [ppm] = 10.23 (s, 1H, NH), 8.59 (s, 2H, H2), 8.58−8.53 (m, 2H, H8), 8.48 (d, 3JHH = 8 Hz, 2H, H 5), 7.76 (vtd, 3J 4HH = 8 Hz, JHH = 2 Hz, 2H, H6), 7.25 (ddd, 3J 4 7HH = 8 Hz, 5 Hz, JHH = 1 Hz, 2H, H ), 2.06 (s, 3H, CH ). 133 C{ 1H} NMR (CD2Cl2): δ [ppm] = 170.0 (s, C 10), 156.3 (s, C4), 156.1 (s, C3), 149.0 (s, C8), 148.3 (s, C1), 137.0 (s, C6), 124.1 (s, C7), 121.0 (s, C5), 110.6 (s, C2), 24.5 (s, C11). Procedure (b). To a solution of 4′-amino-2,2′:6′,2″-terpyridine LC (2.22 g, 8.94 mmol, 1 equiv) in dichloromethane (30 mL) was added a solution of acetyl chloride (10 mL, exc.) in dichloromethane (30 mL) dropwise over a period of 15 min. The resulting mixture was heated to reflux for 3 h. A slightly yellow precipitate formed during the heating. After the mixture cooled to room temperature, the solvent and the acetyl chloride were removed under reduced pressure. The remaining aBuchwald−Hartwig amination of 1-bromo-3,5-di(pyridin-2-yl)- solid was dissolved in a mixture of water and tetrahydrofuran (1:1, 100 benzene LA and 4′-chloro-2,2′:6′,2″-terpyridine LB with acetamide mL), and the pH was adjusted to 8 using aqueous sodium bicarbonate yielding N-acetyl-3,5-di-(pyridin-2-yl)aniline L1 and N-acetyl-4′- solution. The resulting colorless precipitate was collected via filtration amino-2,2′:6′,2″-terpyridine L2. Atom numbering for NMR assign- yielding microanalytically pure N-acetyl-4′-amino-2,2′:6′,2″-terpyri- ment is included. dine L2 (1.09 g, 3.75 mmol). The aqueous phase was further extracted with dichloromethane (3 × 50 mL). The combined organic phases were dried over magnesium sulfate and evaporated to dryness. The Synthesis of N-Acetyl-3,5-di(pyridin-2-yl)aniline L1. 4,5-Bis- crude product was purified via column chromatography on neutral (diphenylphosphino)-9,9-dimethylxanthene (XantPhos, 64 mg, 111 Alox (Brockmann II, 3% water (w/w), solvent gradient ethyl acetate/ μL, 3 mol %) and Pd precatalyst [Pd]2 (14 mg, 19 μmol, 1.1 mol % hexanes 1:3 → 3:1) to give a second fraction of pure product as off- based on Pd) were dissolved under argon in 15 mL of abs. toluene and white solid (1.00 g, 3.44 mmol). Yield: 2.09 g (7.20 mmol, 81%). The left to stand. After 10 min 1-bromo-3,5-di(pyridin-2-yl)benzene LA 1H NMR spectra of both fractions match those obtained from (1.15 g, 3.70 mmol, 1 equiv), acetamide (273 mg, 4.62 mmol, 1.25 procedure a). equiv), and sodium tert-butanolate (444 mg, 4.62 mmol, 1.25 equiv), Synthesis of RuCl3(R-tpy), R = COOC2H5, NHCOCH3. A standard dissolved in 15 mL of abs. toluene, were added, and the mixture was procedure was followed for the synthesis of the RuCl3(R-tpy) heated to re ux for 8 h. After the mixture cooled to room temperature, precursors:29,52fl Ruthenium(III) chloride hydrate (36% Ru (w/w); R the solvent was removed under reduced pressure, and the remaining = COOC2H5: 1.46 g, 5.20 mmol, 1.3 equiv; R = NHCOCH3: 566 mg, solid was dissolved in concentrated hydrochloric acid (20 mL), water 2.01 mmol, 1.3 equiv) was dissolved in ethanol (50 mL), and the (20 mL), and dichloromethane (50 mL). The phases were separated, respective terpyridine (R = COOC2H5: 1.21 g, 3.96 mmol, 1 equiv; R and the aqueous phase was extracted twice with dichloromethane (2 × = NHCOCH3: 450 mg, 1.55 mmol, 1 equiv) was added. The resulting 50 mL). The aqueous phase was neutralized with dilute aqueous mixture was heated to reflux for 3 h during which time the product sodium hydroxide solution (pH = 9) followed by extraction with precipitated as a red solid. It was filtered off and washed thoroughly dichloromethane (3 × 50 mL). The organic fractions of the second with ethanol to remove residual RuCl3. The product was dried under extraction were dried over magnesium sulfate, and the solvent was reduced pressure and used without further purification. Yield: R = 11091 DOI: 10.1021/acs.inorgchem.5b01151 Inorg. Chem. 2015, 54, 11088−11104 Section 3.2 | 75 Inorganic C hemistry Article Scheme 3. Synthetic Procedurea aStarting from RuCl3 leading to the heteroleptic cyclometalated ruthenium complex isomers 1(PF6) and 2(PF6) as well as the amino complex 3(PF6). Numbering for NMR assignments is included. COOC2H5: 1.94 g (3.78 mmol, 96%). R = NHCOCH3: 730 mg (1.47 (CD3CN): δ [ppm] = 215.8 (s, C 9), 169.8 (s, C13), 169.3 (s, C4B), mmol, 95%). Because of the poor solubility of RuCl3(tpy-R) no 166.0 (s, C 10), 159.8 (s, C4A), 155.2 (s, C8A), 154.3 (s, C3A), 153.1 (s, characterization was performed. C8B), 141.9 (s, C3B), 136.9 (s, C5B), 136.2 (s, C5A), 134.4 (s, C1B), Synthesis of [Ru(dpb-NHCOCH3)(tpy-COOC2H5)](PF6) 1(PF6). 132.5 (s, C 1A), 127.6 (s, C7A), 124.8 (s, C5A), 122.9 (s, C2A), 122.6 (s, RuCl3(tpy-COOC2H5) (100 mg, 0.195 mmol, 1 equiv) was suspended C 7B), 120.8 (s, C5B), 117.9 (s, C2B), 63.3 (s, C11), 24.4 (s, C14), 14.7 (s, under argon in 20 mL of abs. acetone, and silver tetra uoroborate C12fl ). IR (KBr disk): λ−1 [cm−1] = 3435 (crystal water), 1723 (C (110 mg, 0.566 mmol, 2.9 equiv) was added. The resulting reaction Oester), 1711 (COamide), 1600 (CC), 1518 (amide II), 845 (P−F). mixture was heated to reflux for 2 h in the dark. After the mixture UV−vis (MeCN): λ 3 −1 −1max (ε) [nm (1 × 10 M cm )] = 241 (49.8), cooled to room temperature, the dark brown solution was filtered 282 (62.1), 319 (29.8), 378 (14.1), 418 (shoulder, 9.7), 506 (17.3), through a syringe filter to remove precipitated silver chloride prior to 555 (13.7). evaporation of the solvent. The dark, oily residue was dissolved in abs. Synthesis of [Ru(dpb-COOC2H5)(tpy-NHCOCH3)](PF6) 2(PF6). nBuOH (20 mL), and CH3CONH-dpbH L1 (68 mg, 0.234 mmol 1.2 RuCl3(tpy-NHCOCH3) (100 mg, 0.201 mmol, 1 equiv) was equiv) was added. The resulting dark brown to purple solution was suspended in 20 mL of abs. acetone, and silver tetrafluoroborate heated to reflux for 16 h giving an intensely colored purple solution. (113 mg, 0.583 mmol, 2.9 equiv) was added. The resulting reaction After removal of the solvent under reduced pressure the remaining mixture was heated to reflux for 2 h in the dark. After the mixture solid was dissolved in acetonitrile (5 mL), and a solution of cooled to room temperature, the dark brown solution was filtered ammonium hexafluorophosphate (125 mg, 0.78 mmol, 4 equiv) in through a syringe filter to remove precipitated silver chloride prior to water (1 mL) was added. Addition of more water (∼80 mL) resulted evaporation of the solvent. The dark, oily residue was dissolved in abs. in the precipitation of the crude product, which was filtered o . nff BuOH (20 mL), and C2H5OOC-dpbH L3 (73 mg, 0.241 mmol, 1.2 Column chromatography on silica gel (solvent gradient chloroform → equiv) was added. The resulting dark brown to purple solution was chloroform/methanol 7:1, after a yellow impurity was eluted) afforded heated to reflux for 16 h giving an intensely colored red solution. After pure [Ru(dpb-NHCOCH3)(tpy-COOC2H5)](PF6) as dark purple removal of the solvent under reduced pressure the remaining solid was solid. Yield: 114 mg (0.136 mmol, 70%) Anal. Calcd for dissolved in acetonitrile (5 mL), and a solution of ammonium C36H29F6N6O3PRu (839.7)·1.5H2O: C, 49.89; H, 3.72; N, 9.70. hexafluorophosphate (130 mg, 0.80 mmol, 4 equiv) in water (1 mL) Found: C, 50.01; H, 3.50; N, 9.53%. MS(ESI+): m/z (%) = 347.6 (1) was added. Addition of more water (∼80 mL) resulted in the [M-PF ]2+6 , 695.1 (100) [M-PF ] + 6 , 1535.3 (3) [2M-PF6] +. HR- precipitation of the crude product, which subsequently was filtered off. MS(ESI+, m/z): Calcd for C36H29N6O3Ru [M-PF + 6] : 695.1345; Column chromatography on neutral Alox (Brockmann II, 3% water Found: 695.1336. 1H NMR (CD3CN): δ [ppm] = 9.21 (s, 2H, (w/w), solvent gradient chloroform → chloroform/methanol 50:1, H2A), 8.69 (s, 1H, NH), 8.57 (d, 3JHH = 8 Hz, 2H, H 5A), 8.45 (s, 2H, after a yellow impurity was eluted) afforded pure [Ru(dpb-COOEt)- H2B), 8.06 (d, 3JHH = 8 Hz, 2H, H 5B), 7.71 (vtd, 3JHH = 8 Hz, 4JHH = 1 (tpy-NHCOCH3)](PF6) as dark solid. Yield: 87 mg (0.104 mmol, Hz, 2H, H6A), 7.58 (vtd, 3J = 8 Hz, 4HH JHH = 1 Hz, 2H, H 6B), 7.18 (d, 52%) A second fraction consisting of [Ru(dpb-COOEt)(tpy-NH2)]- 3JHH = 5 Hz, 2H, H 8A), 7.04−6.95 (m, 2H, H7A), 6.93 (d, 3JHH = 5 Hz, (PF6) 3(PF6) was isolated as well (40 mg, 0.050 mmol, 25%). Anal. 2H, H8B), 6.58 (vt, 3JHH = 6 Hz, 2H, H 7B), 4.63 (q, 3JHH = 7 Hz, 2H, Calcd for C36H29F6N6O3PRu (839.7): C, 51.49; H, 3.48; N, 10.01. H11), 2.23 (s, 3H, H14), 1.57 (t, J = 7 Hz, 3H, H12). 13C{1H} NMR Found: C, 51.46; H, 3.30; N, 9.73%. MS(ESI+): m/z (%) = 347.6 (1) 11092 DOI: 10.1021/acs.inorgchem.5b01151 Inorg. Chem. 2015, 54, 11088−11104 76 | 3 RESULTS AND DISCUSSION Inorga nic Chemistry Article Figure 1. 1H NMR spectra of 1(PF6) (upper, blue) and 2(PF6) (lower, red) at room temperature in CD3CN (for atom numbering see Scheme 3). [M-PF ]2+6 , 695.1 (100) [M-PF ] + 6 , 1535.3 (6) [2M-PF6] +. HR- rigorous protective gas conditions, and only 2.9 equiv of MS(ESI+, m/z): Calcd for C +36H29N6O3Ru [M-PF6] : 695.1345; Ag[BF4] (instead of 3.6 equiv as usually found in the Found: 695.1342. 1H NMR (CD3CN): δ [ppm] = 10.42 (s, 1H, 2A 2B 3 literature) 20,81 were employed to prevent undesired side NH), 9.12 (s, 2H, H ), 8.83 (s, 2H, H ), 8.30 (d, JHH = 8 Hz, 2H, H5A), 8.25 (d, 3 5B − reactions. Under these conditions, we obtained the complexesJHH = 8 Hz, 2H, H ), 7.70 7.60 (m, 4H, H6A, H6B), 7.18 (d, 3J = 5 Hz, 2H, H8B), 7.02 (d, 3J = 5 Hz, 2H, H8A), 6.91− 1(PF6) and 2(PF6) in yields of 70% and 52%, respectively,HH HH 6.82 (m, 2H, H7A), 6.76−7.66 (m, 2H, H7B), 4.51 (q, 3J = 7 Hz, 2H, besides small quantities of a side product with a hydrolyzedHH H11), 2.36 (s, 3H, H14), 1.51 (t, 3J 12 13 1 acetyl amino group ([Ru(dpb-COOEt)(tpy-NHHH = 7 Hz, 3H, H ). C{ H} NMR 2)](PF6) (CD3CN): δ [ppm] = 233.7 (s, C 9B), 171.3 (s, C13), 169.1 (s, C4B), 3(PF6); see Scheme 3 and Supporting Information). 168.7 (s, C10), 159.8 (s, C4A), 155.6 (s, C8A), 153.7 (s, C3A), 152.8 (s, Characterization of the Isomers. Since both complexes C8B), 145.7 (s, C1A), 143.4 (s, C3B), 136.5 (s, C6A), 136.4 (s, C6B), 1(PF6) and 2(PF6) are constitutional isomers they share their 127.3 (s, C7A), 124.5 (s, C5A), 124.5 (s, C2B), 123.0 (s, C7B), 122.7 (s, 1B 5B 2A 11 14 elemental composition and show essentially identical massC ), 120.8 (s, C ), 112.8 (s, C ), 61.5 (s, C ), 24.9 (s, C ), 15.0 (s, 12 −1 −1   spectra and isotope patterns (Supporting Information FigureC ). IR (KBr disk): λ [cm ] = 1695 (C Oester,amide), 1600 (C C), 1514 (amide II), 844 (P−F). UV−vis (MeCN): λ S8). The most prominent differences are observed in the NMRmax (ε) [nm (1 × 103 M−1 cm−1)] = 242 (49.2), 282 (69.4), 317 (32.4), 351 (16.9), 428 spectra (Figure 1, Supporting Information Figures S9−S18). All (8.7), 502 (15.2), 544 (shoulder, 11.8). (NMR and mass spectrometric tpy aromatic proton resonances (2A, 5A, 6A, 7A, and 8A) data of 3(PF6) can be found in the Supporting Information.) appear systematically further downfield in the 1H NMR spectrum (Figure 1, Scheme 3) compared to corresponding ■ RESULTS AND DISCUSSION resonances of the dpb protons due to the electron-deficiency of Syntheses of Ru Complexes. Details of the ligand the tpy ligand. Proton resonances of the functional groups syntheses can be found in the Experimental Section (L1, L2) (NHCOMe and COOEt) show the same trend. In 1 +, with the and in the Supporting Information (L3). Several different tpy-COOEt ligand, the CH2 and CH3 resonances of the ethyl experimental protocols have been reported for the synthesis of group are found at 4.63 and 1.57 ppm, respectively, whereas in + heteroleptic ruthenium complexes with terpyridine and the regioisomer 2 the corresponding protons of the dpb- dipyridylbenzene ligands. Complexation can be performed in COOEt ligand appear at 4.51 and 1.51 ppm, respectively. water/methanol solution starting from RuCl (tpy-R) at Similarly, the resonance of the acetyl CH3 protons is found at3 + + elevated temperatures and in the presence of a tertiary amine 2.23 ppm in 1 and at 2.36 ppm in 2 , underlining the as sacrificial reductant.41 This path resembles the microwave- electronic influence of the different ligands. This effect is most assisted synthesis that we employed to obtain heteroleptic pronounced for the amide NH proton due to its proximity to bis(terpyridine)ruthenium(II) complexes in high yields.21,27 the aromatic system in both complexes. The NH resonance of For cyclometalated complexes, this procedure gives practicable tpy-NHCOMe in 2+ is found at very low field (10.42 ppm), yields when the coordinating carbon atom is located at one of while in 1+, the NH resonance of dpb-NHCOMe is found at the peripheral aromatic rings of a multidentate ligand.20,41 A 8.69 ppm. Similar trends are also observed in the 13C NMR more robust protocol was presented in 1991 by Sauvage and spectra (Supporting Information Figures S10 and S15). The co-workers in the first report on the parent cyclometalated resonance of the carbon atom C9B involved in the metal− ruthenium(II) complex [Ru(dpb)(ttpy)]+ (ttpy = 4′-tolylter- carbon bond is found at 215.8 ppm in 1+ with the dpb- pyridine):80 RuCl3(ttpy) is activated via chloride abstraction NHCOMe ligand and at 233.7 ppm in 2 + with the dpb-COOEt with silver tetrafluoroborate. Upon addition of dpbH the ligand. desired cyclometalated complex forms readily at elevated Although the final complexation step requires harsh reaction temperatures in high yields.20,81 We successfully adapted this conditions and long reaction times, the reaction proceeds in protocol for the synthesis of the complexes presented herein good yields without significant side reactions. The structural (Scheme 3). Since acetylated amino groups are prone to integrity of the complexes 1+ and 2+ with all functional groups hydrolysis and oxidation, all reactions were performed under is further confirmed by ESI mass spectrometry (Supporting 11093 DOI: 10.1021/acs.inorgchem.5b01151 Inorg. Chem. 2015, 54, 11088−11104 Section 3.2 | 77 Inorganic C hemistry Article Information Figure S8). No significant mass peaks indicating 104 M−1 cm−1 appears to be a characteristic marker for the cleavage of either the ester or the amide are detected. cyclometalation in the central position of the tridentate ligand Interestingly, all mass spectra show a weak peak that can be and is well-reproduced via theoretical calculations (vide infra). assigned to [M−PF ]2+6 at m/z = 347.6. This is likely explained The visible region of the absorption spectra of 1(PF6) and by the typically rather low oxidation potential of these electron- 2(PF6) is dominated by four absorption bands (Tables 1 and rich cyclometalated (polypyridine)ruthenium(II) complexes 2). The strongest absorption occurs at 506 nm in 1(PF6) and at (vide infra). IR spectroscopy confirms the presence of both 502 nm in 2(PF6) and is best described as 1MLCT transition an ester group and a primary amide by characteristic bands for  involving the pyridine rings of both ligands as acceptors (videester C O and amide I stretching vibrations at 1723 and 1711 infra).20 In complex 1(PF ), the donor and acceptor effect of cm−1 for 1+ (DFT: 1733 and 1701 cm−1 6 , respectively). For 2+, the respective ligand and functional group reinforce one the IR stretching vibrations of the ester and amide carbonyl another. An additional absorption maximum appears at the low- function are observed at 1697 cm−1. Indeed, DFT calculations energy side of this MLCT transition at 555 nm. On the predict essentially identical vibrational frequencies for both contrary, 2(PF6) only shows a weak shoulder in this region groups at ∼1718−1719 cm−1 for 2+ reflecting the different (Figure 2). The overall bathochromic shift of the visible light electronic character of tpy and dpb (Supporting Information absorption features of 1(PF6) compared to 2(PF6) is Figure S21). The NH deformation bands are found at ∼1580− −1 accompanied by an increase in absorption intensity. Both1600 cm for both isomers along with coupled C−C observations are best explained by the increased push−pull vibrations within the aromatic backbone (see rR spectra). effect arising from the functional groups, which lowers the The presence of the PF −6 counterion is revealed by broad and −1 highest occupied molecular orbital−lowest unoccupied molec-intense P−F stretching bands at 840 cm . ular orbital (HOMO−LUMO) gap (cf. MO diagram, UV−vis Spectroscopic Properties of the Cyclometa- + + Supporting Information Figure S22) and increases thelated Isomers 1 and 2 . The experimental UV−vis spectra transition dipole moments for the 1MLCT transitions in of the two complexes 1(PF6) and 2(PF6) are depicted in Figure 1(PF6). In 2(PF6) on the other hand, the donor strength of dpb2, and relevant spectroscopic data are summarized in the and the acceptor strength of tpy are both partially canceled by the substituents leading to a larger HOMO−LUMO gap and weaker 1MLCT absorptions. As a consequence, 1(PF6) appears black in the solid state and deep purple in acetonitrile solution, while 2(PF6) is dark red in the solid state and in solution. This observation is in accordance with previous results for similar complexes.20 Both compounds show two more absorption bands between 340 and 450 nm with similar intensities as the 1MLCT absorptions. In the literature the origins of these bands are consistently discussed as MLCT transitions involving the cyclometalating ligand.20,41,81 However, this interpretation contradicts the observed hypsochromic shift of this absorption from 378 nm (1(PF6)) to 351 nm (2(PF6)) upon exchanging the electron-donating N-acetyl amino group for an electron- accepting carboxy group. A more consistent explanation of these absorptions is gained from TD-DFT calculations and rR experiments. Theoretical calculations were performed using geometries optimized at the PBE/def2-SV(P) level of theory with an effective core potential at ruthenium (def2-SD, def2-TZVP). This level of theory for the geometry optimization was chosen based on data obtained by screening multiple functionals and basis sets and by comparison to geometrical parameters obtained from crystal structures of three structurally related compounds ([Ru(dpb-COOMe)(tpy)](PF6), [Ru(dpb)(tpy- Figure 2. Experimental absorption spectra of 1(PF6) (upper, blue) and COOEt)](PF6), and [Ru(pbpy-COOMe)(tpy)](PF6)). 20 This 2(PF6) (lower, red) in acetonitrile at room temperature; c = 2 × 10 −5 evaluation showed only a marginal dependence of the mol l−1 and calculated UV−vis spectra from TD-DFT calculations geometrical parameters on the size of basis set allowing for (B3LYP, black). The calculated spectra are shifted by 1000 cm−1 to usage of a rather small and computationally efficient basis set. lower energies to match calculated and experimental π−π* absorption The variation of the structural parameters with the choice of energies. functional was larger, but still, qualitatively similar results were obtained with all functionals under study (BP, PBE, BLYP, Experimental Section. Both complexes show absorption TPSS, TPSSh, M06L, B3LYP, PBE0). Remarkably, within the features of very similar shape, intensity, and energy between mean deviation the hybrid functionals PBE0 and B3LYP 200 and 325 nm. This is because of the identical ligand yielded identical optimized geometries compared to the backbone of both complexes and is characteristic for [Ru- corresponding GGA functionals PBE and BLYP. Hence, the (dpb)(tpy)]+-type complexes.20,81 Especially the sharp absorp- more economic GGA functional PBE was preferred over hybrid tion band at ∼320 nm with an extinction coefficient of ∼3.0 × functionals for geometry optimizations. 11094 DOI: 10.1021/acs.inorgchem.5b01151 Inorg. Chem. 2015, 54, 11088−11104 78 | 3 RESULTS AND DISCUSSION Inorga nic Chemistry Article Table 1. Spectral Decomposition of the Absorption Bands of 1(PF6) in Acetonitrile Solution in the Range from 9000 to 29 000 cm−1 Using the Advanced Spectral Analysis Tool of ORCA (orca_asa)a experimental data theoretical data state λ−1, cm−1 λ, nm f ε , M−1 −1osc max cm state λ −1, cm−1 fosc assignment 1 12 935 773 1.39 × 10−3 145 Ru→tpy 3MLCT 2 15 438 648 1.15 × 10−2 714 2 15 095 5.5 × 10−3 Ru→tpy 1MLCT 3 16 197 617 1.48 × 10−2 1940 4 17 711 565 0.102 11 300 4 18 194 0.030 Ru→tpy 5 19 150 0.040 1MLCT 5 18 694 535 6.99 × 10−3 1570 6 19 687 508 0.107 11 700 6 20 914 0.091 Ru→dpb 1MLCT 7 20 910 478 0.124 6200 7 21 962 0.213 Ru→tpy 1MLCT 8 21 399 467 4.13 × 10−3 690 8 21 843 2.6 × 10−3 Ru→dpb 1MLCT 9 23 871 419 5.38 × 10−2 5290 10 25 166 397 7.88 × 10−3 1210 11 26 500 377 0.234 14 700 aEleven Gaussian bands were required to reproduce the shape of the absorption profile, and λ−1, fosc, and εmax are obtained from the respective fitted bands. Theoretical data are obtained from TD-DFT calculations (B3LYP) and assigned to the experimental bands based on their energy and oscillator strengths. Table 2. Spectral Decomposition of the Absorption Bands of 2(PF6) in Acetonitrile Solution in the Range from 9000 to 29 000 cm−1 Using the Advanced Spectral Analysis Tool of ORCA (orca_asa)a experimental data theoretical data state λ−1, cm−1 λ, nm f ε , M−1 cm−1 state λ−1, cm−1osc max fosc assignment 1 15 479 646 1.29 × 10−3 208 Ru→tpy 3MLCT 2 17 345 577 1.49 × 10−3 331 4 18 332 0.014 Ru→tpy 1MLCT 3 17 455 573 3.25 × 10−2 2530 5 18 593 0.060 Ru→tpy 1MLCT 4 19 445 514 0.216 13 000 7 21 675 0.170 Ru→dpb 1MLCT 8 21 681 0.096 5 20 269 493 1.54 × 10−2 2460 6 21 472 466 9.72 × 10−3 1430 7 23 244 430 7.42 × 10−2 6240 11 24 120 0.111 Ru→tpy,dpb 1MLCT 8 24 575 407 1.39 × 10−4 56 9 25 035 399 1.12 × 10−2 1480 16 26 366 0.026 Ru→tpy 1MLCT 10 28 059 356 1.16 × 10−2 1250 11 29 114 343 0.504 16 300 aEleven Gaussian bands were required to reproduce the shape of the absorption profile, and λ−1, fosc, and εmax are obtained from the respective fitted bands. Theoretical data are obtained from TD-DFT calculations (B3LYP) and assigned to the experimental bands based on their energy and oscillator strengths. Vertical excitations were generated within the TD-DFT match reasonably well in the UV region, when the entire formalism with a triple-ξ basis set (def2-TZVP) and the theoretical spectrum is shifted by 1000 cm−1 to lower energies Douglas−Kroll−Hess relativistic approximation in combination (Figure 2). In the visible region, however, the agreement is with functionals of varying Hartree−Fock (HF) exchange somewhat lower for both 1(PF6) and 2(PF6). This is mostly (PBE, 0%; TPPSh, 10%; B3LYP, 20%; PBE0, 25%). Addition- because of the weakness of TD-DFT in the description of ally, the range-separated CAM-B3LYP functional was em- charge transfer excitations.68,82,83 Difference density plots of the ployed. While PBE and CAM-B3LYP both gave unsatisfactory 10 lowest excitations and all further significant transitions ( f > results, TPSSh, B3LYP, and PBE0 performed equally well in 0.01) are shown for 1+ and 2+ in the Supporting Information, TD-DFT calculations of 1(PF6) and 2(PF6) compared to the Tables S1 and S2. In both complexes the lowest-energy corresponding experimental data (see Supporting Information, transition is a well-defined HOMO−LUMO excitation with no Figures S23 and S24). A systematic increase of all transition significant admixture of other orbitals. The LUMO in both energies and transition probabilities (oscillator strengths) was complexes is a π*-orbital located at the tpy ligand, while the observed with increasing HF exchange (TPSSh < B3LYP < HOMO is essentially a π-orbital of the dpb ligand. The lowest- PBE0). energy excitation can formally be regarded as a LL′CT In the following the TD-DFT calculations using B3LYP as transition although ruthenium d-orbitals of the t2g set are functional and the corresponding spectra generated with admixed into the frontier orbitals (mixed LL′CT and MLCT orca_mapspc (line width 1500 cm−1) will be discussed and character). To differentiate between these transitions and the correlated with the experimental absorption spectra of 1(PF6) MLCT transitions these states will be labeled LL′CT. Since and 2(PF6). The theoretical and experimental UV−vis spectra both ligands are perpendicular to each other so are the 11095 DOI: 10.1021/acs.inorgchem.5b01151 Inorg. Chem. 2015, 54, 11088−11104 Section 3.2 | 79 Inorganic C hemistry Article contributing d-orbitals, which renders the metal contribution negligible. As a consequence, these HOMO−LUMO tran- sitions for 1+ and 2+ are symmetry-forbidden excitations due to the vanishing overlap integral (oscillator strengths of 1.1 × 10−5 and 7.1 × 10−5 for 1+ and 2+, respectively) and do not contribute to the absorption spectrum. The corresponding 3LL′CT states, however, might play a significant role for the excited-state behavior of 1(PF6) and 2(PF6), as discussed below. The higher-energy excitations predicted in the visible region are 1MLCT transitions from metal d-orbitals onto the ligands. Interestingly, these excitations do not only involve tpy π*- orbitals. Already in the range above 400 nm dpb accepting orbitals play a major role for the absorption profile. The UV transitions in the range between 400 and 320 nm also consist of 1MLCT transitions onto both ligands. A distinct separation of Ru → tpy and Ru → dpb MLCT transitions, with the former being responsible for the low-energy absorption band between 600 and 450 nm and the latter yielding the UV band between 320 and 430 nm, is not valid. To be able to match the experimentally obtained spectrum with the theoretical data spectral decompositions of the visible range of the absorption spectra were performed. This is of particular interest considering the predicted low intensity of the LL′CT transitions in 1(PF6) and 2(PF6). The fit data are summarized in Tables 1 and 2 for 1(PF6) and 2(PF6), respectively. Despite the high quality of the fit (see Supporting Information, Figure S25) the low-intensity LL′CT absorptions in the low-energy edge of the absorption spectrum were not detected in either case. An upper limit of the oscillator strengths of these transitions is estimated as f ≤ 1 × 10−4, which is in agreement with the computational data. The decom- position of the absorption bands supports the qualitative discussion of the spectra above. A plausible assignment of TD- DFT excitations to the most intense bands was possible based on similarities in oscillator strengths and transition energies underlining that the computational method gives a reasonable estimate of the absorption spectrum. The lowest-energy excitation observed within the spectral decomposition (12 935 and 15 479 cm−1 for 1(PF6) and 2(PF6), respectively) could not be assigned to any calculated vertical singlet excitation. We ascribe these to 3MLCT absorptions that become partially spin-allowed due to spin−orbit coupling in the presence of ruthenium.4,5 A complete assignment of all observed bands is of course out of reach at the presented level of theory and will generally be very difficult based on the complexity of the absorption characteristics of 1(PF6) and Figure 3. (a) Resonance Raman spectra of 1(PF ) in acetonitrile 2(PF6). 6 + + solution (298 K) in the range of 1400−1700 cm −1 at different Resonance Raman Studies on 1 and 2 . To further excitation wavelengths. (b) Experimental (blue) and DFT-calculated support this interpretation of the absorption characteristics of (black, line width = 10 cm−1) rR spectra of 1(PF6) at 473, 532, and 1(PF6) and 2(PF6) in the visible region rR spectroscopic 633 nm excitation wavelength. Asterisks indicate Raman bands of studies in acetonitrile solution were performed. This technique CH3CN. has proven to be useful just recently in the elucidation of the charge redistribution upon optical excitation in bis- spectra obviously directly depends on the extinction coefficient (terpyridine)ruthenium(II) complexes.84,85 The rR spectra of at the given irradiation wavelength so that a maximum in rR 1(PF6) and 2(PF6) with excitation at 473, 532, and 633 nm are intensity is expected in the range of 530−470 nm for both shown in Figures 3 and 4, respectively. Since even the idealized complexes. On the other hand, the rR intensity depends on core symmetry of these complexes (C2v) is rather low and the whether a given vibrational mode contributes to the geo- number of atoms is high (N = 76), a multitude of Raman metrical reorganization associated with the given optical bands, many of them overlapping, is observed in the rR spectra. transition at the Franck−Condon point. Remarkably, the Qualitatively, the spectra appear very similar at the different carbonyl stretching vibrations of both the ester and the excitation wavelengths with the only differences lying in the amide functionality only play a subordinate role at all excitation intensities of the bands. On one hand, the intensity of the rR wavelengths for both 1(PF6) and 2(PF6). Since the common 11096 DOI: 10.1021/acs.inorgchem.5b01151 Inorg. Chem. 2015, 54, 11088−11104 80 | 3 RESULTS AND DISCUSSION Inorga nic Chemistry Article stretching bands are observed at any excitation wavelength (cf. IR spectrum, vide supra). Since the MLCT transitions involve π*-orbitals of the aromatic ligands the corresponding CC valence vibrations should be excited and give intense rR responses (see Figures 3a and 4a). They are assigned based on DFT-calculated ground- state vibrational frequencies. The bands at 1600 cm−1 for 1(PF6) and 2(PF6) are assigned to the symmetric valence vibrations (local A1 symmetry) of the aromatic rings. While for 1(PF6), all these vibrations are very close in energy (DFT: between 1597 and 1605 cm−1) and overlap in the rR spectra at all excitation wavelengths, the symmetric vibration of the carboxy-substituted phenyl ring of 2(PF6) is shifted by 20 cm −1 to lower energy yielding a well-resolved band in the rR spectra of 2(PF ) at 1581 cm−16 (calculated at 1584 cm −1) that is not present in 1(PF6). The significant participation of the phenyl vibration in the rR spectra of 2(PF6) indicates substantial Ru→ dpb character of the MLCT absorption band at 532 nm and at 473 nm. This corroborates the findings of TD-DFT calculations that Ru→dpb excitations are present even at energies below 450 nm. Since the corresponding vibration of the phenyl moiety of 1(PF6) overlaps with those of the pyridine rings a similar conclusion cannot be drawn for 1(PF6) although the width of the band suggests participation of all six totally symmetric aromatic vibrations. Similar behavior is observed for the antisymmetric (local B2 symmetry) CC valence vibrations at ∼1525 cm−1. While for 1(PF6) these vibrations overlap yielding one broad Raman band, a distinct shoulder at 1514 cm−1 appears for 2(PF6), which is assigned to the phenyl moiety again underlining the mixed Ru→tpy/Ru→dpb MLCT character of the absorption band at 473 and 533 nm. The intensity of the bands between 1400 and 1550 cm−1 increases substantially upon increasing λexc from 473 to 532 nm. Ground-state vibrational frequencies and Raman intensities provide no straightforward explanation for that. The independent mode-displaced harmonic oscillator (IMDHO) model was employed, which assumes harmonic ground- and excited-state potential energy surfaces (PES) and no frequency changes upon excitation. Additionally, the excited-state PES are considered as displaced with respect to the ground-state PES along certain (or all) normal modes. More evolved theoretical approaches have been employed previously in the description of rR spectra of large molecules,86 but these require much more computational time and are limited to a small number (2−3) of electronically excited states that can be considered in the calculations. Figure 4. (a) Resonance Raman spectra of 2(PF6) in acetonitrile At least three prerequisites must be met to yield a solution (298 K) in the range of 1400−1700 cm−1 at different qualitatively good description of the rR behavior of a given excitation wavelengths. (b) Experimental (red) and DFT-calculated compound: First, a high quality of the normal mode (black, line width = 10 cm−1) rR spectra of 2(PF6) at 473, 532, and displacements is crucial for a reasonable description of the rR 633 nm excitation wavelength. Asterisks indicate Raman bands of spectra since they determine the intensity of the corresponding CH3CN. Raman bands. These can be computed within the above- mentioned theoretical model from excited-state gradient calculations.78,87 Second, the vibrational frequencies obtained description of the intense visible range absorption band is that from DFT calculations must correspond well to the of a MLCT transition onto the terpyridine ligand the experimentally observed ones since these define the normal terpyridine ester CO vibration at 1723 cm−1 in 1(PF6) modes and have a large impact on the displacement parameters. should be visible for all excitation wavelengths under study. At Third, the character of the calculated electronic excited states λexc = 633 nm this is indeed true, but at higher energies (λexc = must match that of the actual transitions. This is the most 532 nm) the contribution of this vibration diminishes until challenging part, especially for charge transfer processes, since disappearance at λexc = 473 nm. At shorter wavelengths, Ru→ DFT has its weakness in describing such excitations. 68,82,83 dpb MLCT excitations become increasingly relevant for the All calculations were performed based on the B3LYP/def2- absorption characteristics. For 2(PF6), no distinct CO SV(P)/DKH/COSMO(acetonitrile) optimized geometry of 1+ 11097 DOI: 10.1021/acs.inorgchem.5b01151 Inorg. Chem. 2015, 54, 11088−11104 Section 3.2 | 81 Inorganic C hemistry Article and 2+. Vibrational frequencies were obtained at the same level prevail (excitations 4 and 5), the absorption at 533 nm is of theory. Since ORCA does not support a solvation model for dominated by excitation 5 (Ru→tpy MLCT; see Table 2 and excited-state gradients, the 10 lowest vertical excitations and Supporting Information Table S2). At λexc = 473 nm Ru→dpb gradients were generated in the gas phase at the B3LYP/def2- transitions come into play (state 7). Again, charge transfer TZVP/DKH level of theory. These 10 excitations describe the processes into the electron-rich dpb ligand occur at significantly spectral range under study sufficiently well. The orca_asa lower energies than expected. In essence absorption bands software was then used to simulate the first-order rR spectra between 550 and 450 nm consist of MLCT absorptions from (higher-order Raman bands were considered but did not ruthenium onto both ligands in both complexes 1(PF6) and improve the quality of the simulations). A homogeneous line 2(PF6). broadening of 1200 cm−1 was assumed for all 10 excitations. Emission Properties of 1+ and 2+. Cyclometalated The vibrational frequencies of both compounds were uniformly polypyridine complexes of ruthenium usually exhibit only scaled by a factor of 0.967. This factor yields a maximum of weak emission. The carboxy-substituted complex [Ru(dpb)- agreement between the experimental and simulated spectra. (tpy-COOR)]+ is nonemissive at room temperature, while its Neither the different MLCT optical excitations nor the regioisomer [Ru(dpb-COOR)(tpy)]+ shows weak room- molecular vibrations are energetically separated. Consequently, temperature emission.20 Similarly, [Ru(pbpy)(tpy-COOR)]+ the rR spectrum at a given excitation wavelength is a is nonemissive at room temperature.20 superposition of rR profiles of the individual electronic 2(PF6) emits at room temperature at 751 nm with an excitations weighted by their contribution to the absorption emission quantum yield of 1.4 × 10−5 (Supporting Information spectrum at that wavelength. At the same time the individual Figure S26). At 77 K, a much more intense emission is Raman bands are a superposition of multiple vibrational modes. observed with an emission maximum at 716 nm and a band Despite the large number of approximations and assumptions shape typical for a ruthenium-based emission arising from a the experimental rR spectra of 1(PF6) are remarkably well- single vibronic progression (see Supporting Information Figure reproduced by these simulations. This allows further con- S27).27,88 firmation of the character of respective absorption bands by The temperature dependence of the phosphorescence of assigning optical transitions to the respective absorption 2(PF6) is shown in Figure 5. The emission intensity rapidly energies. The shape of the rR spectrum of 1(PF6) at 633 nm is dominated by the most intense low-energy optical transitions 4 and 5 (see Table 1 and Supporting Information Table S1), which are Ru→tpy MLCT transitions. At λexc = 532 nm, the character of the involved absorptions changes and so does the rR spectrum. The range of 1000−1300 cm−1 (in-plane deformation vibrations of the ligand backbones) is very characteristic for these changes, and a good agreement between simulation and experiment is obtained. The rR spectrum at λexc = 532 nm is dominated by the optical excitations 6 and 7 (Ru→ tpy and Ru→dpb; see Table 1 and Supporting Information Table S1). Consequently, the absorption region around the MLCT absorption maximum is composed of both Ru→tpy and Ru→dpb MLCT transitions. This is in contrast to the widely accepted picture of this band as exclusively arising from Ru→ tpy transitions. The quality of the rR simulation decreases Figure 5. Emission quantum yield of 2(PF6) in fluid butyronitrile somewhat at λ = 473 nm. This is most likely because, at this solution in the temperature range between 160 and 300 K (lower toexc upper). (inset) Plot of ln(ϕ) vs T−1 and the corresponding linear fit wavelength, absorption bands of the second MLCT absorption curve based on ln(ϕ) = ln(kr/k′0) + ΔE′/R·1/T (see text for(between 450 and 320 nm) contribute already. To keep the explanation). computational effort manageable these electronic transitions were neglected and thus are missing in the simulation. Consequently, the number of predicted rR bands is lower increases upon lowering the temperature. This behavior is easily than experimentally observed. It is worth noting that the understood following the argumentation of van Houten and 2 3 calculations also give an explanation for the missing resonant Watts and later Meyer and co-workers. The lifetime of the increase of the carbonyl stretching vibrations: For almost all emissive 3MLCT state depends on the rates of radiative and major optical excitations the CO fragment is located in a nonradiative decay, kr and knr. Additionally, irreversible thermal nodal plane of the involved orbitals and thus is only marginally depopulation of the emissive 3MLCT states via 3MC states is a affected by the rR effect. relevant nonradiative relaxation pathway in (polypyridine) 2(PF6) gives a qualitatively very similar picture although the ruthenium complexes. Because of the irreversibility of this overall agreement between experiment and simulation is process it can be accounted for by a third rate constant k′0 and slightly lower (Figure 4). Especially, the B2 symmetric CC an Arrhenius-like activation barrier ΔE′. As the quantum yield valence vibrations at 1525 cm−1 seem to be missing in the is proportional to the lifetime of the emissive 3MLCT state and simulation as they are calculated at higher energy at 1556 cm−1. the rate constant for radiative decay, 3 the following relationship At 632 nm excitation wavelength essentially off-resonance between ϕ and T is obtained: excitation is achieved leading to a spectrum with a low signal- ϕ = k /[k + k + k′0 ·exp(−ΔE′/RT)] to-noise ratio. Still, the absorption characteristics at the r r nr different wavelengths are identical. While at 632 nm low- This equation gives a nonlinear relationship between ln(ϕ) energy MLCT transitions from ruthenium to the tpy ligand and T−1, as has been shown by Meyer and co-workers.3 In the 11098 DOI: 10.1021/acs.inorgchem.5b01151 Inorg. Chem. 2015, 54, 11088−11104 82 | 3 RESULTS AND DISCUSSION Inorga nic Chemistry Article Figure 6. Jablonski diagrams and electronic spin densities of B3LYP-optimized triplet states of 1+ and 2+ (contour plots at 0.001 isosurface value). 1MLCT and 1LL′CT energies are obtained from TD-DFT calculations. 3LL′CT energy of 1+ is obtained as energy difference from relaxed singlet and triplet geometries from DFT calculations. The 3MLCT energy of 2+ is determined experimentally from the E00 emission at 77 K. Other triplet-state energies are obtained from B3LYP geometry optimizations. present case, however (see inset of Figure 5), a linear C−C−Nperipheral dihedral angles of ∼9°). This nicely illustrates relationship between ln(ϕ) and T−1 is obtained in the the dissociative character of this excited state that has temperature range from 180 to 300 K. This can be explained previously been illustrated for a series of other bis(tridentate) assuming an efficient irreversible excited-state deactivation via ruthenium complexes23 and that is responsible for the intrinsic low-lying 3MC states with a small barrier ΔE′. With this photochemical reactivity of (polypyridine)ruthenium com- assumption and at su ciently high temperatures k and k plexes.2,3,15ffi r nr The 3MC−3MLCT energy difference is determined become negligible with respect to the exponential term as 9.4 kJ mol−1 and 19.0 kJ mol−1 (B3LYP and PBE0, associated with the rate constant k′0, and ln(ϕ) indeed depends respectively). Even though this energy difference is not directly linearly on T−1. From the linear regression, the thermal related to the experimentally determined activation barrier ΔE′ activation barrier for the 3MLCT−3MC surface crossing was of 11.4 kJ mol−1 it serves as a lower limit to the latter. The determined to be ΔE′ = 11.4 ± 0.5 kJ mol−1, which is just a calculation using B3LYP as functional seems to give a good fourth of the activation barrier found for [Ru(bipy) ]2+3 (ΔE′ = estimate to the 3MLCT−3MC energy difference, while PBE0 42.6 kJ mol−1).3 slightly overestimates this energy gap. To get a clearer picture of the involved excited-states, DFT The 3MLCT−3MC energy difference calculated for 2+ is calculations were performed. Using B3LYP or PBE0 as significantly lower than that obtained for [Ru(tpy-COOH)(tpy- functional, one 3MLCT and one 3MC state could be localized NH )]2+ at a similar level of theory (26.8 kJ mol−1).232 This for 2+ (see Figure 6). The geometry of the 3MLCT state is might be attributed to the fact that the strong σ-donating effect essentially unaltered compared to the 1GS geometry. Because of dpb is partially diminished by the electron-accepting ester of the dipolar character of this excited state the electron- functionality in 4-position and the tpy ligand is a weaker deficient ruthenium atom is slightly shifted by 3 pm toward the electron acceptor. More importantly though, cyclometalation of tpy ligand, while the dpb−tpy distance remains unaffected the central phenyl ring only raises one of the three 3MC states corresponding to a simple motion of ruthenium toward tpy in that are responsible for the excited-state deactivation, while the the fixed N5C coordination sphere. A much stronger distortion other 3MC states remain low in energy and are efficiently with respect to the 1GS geometry is observed for the 3MC state populated at room temperature.20 since an antibonding metal orbital is occupied. The Ru−Ntpy In contrast to 2(PF6), 1(PF6) is nonemissive both at room bond lengths are substantially elongated by ∼20 pm, and the temperature and at 77 K. This cannot be accounted for with a peripheral pyridine rings of the tpy ligand are significantly thermally activated deactivation mechanism of the excited state twisted out of the plane of the central pyridine ring (Ncentral− unless the activation barrier is close to zero. Hence another 11099 DOI: 10.1021/acs.inorgchem.5b01151 Inorg. Chem. 2015, 54, 11088−11104 Section 3.2 | 83 Inorganic C hemistry Article mechanism must be responsible for the excited-state deactivation. As was shown above by TD-DFT calculations, symmetry-forbidden LL′CT transitions exist at the low-energy edge of the absorption spectra of 1(PF6) and 2(PF6). DFT calculations yielded a corresponding 3LL′CT excited state for 1+ (see Figure 6) as well as 3MLCT and 3MC states with spin distributions similar to those of 2+. Remarkably, the complex core is essentially undistorted for the 3LL′CT state of 1+ although the Ru−Cdpb bond is slightly shortened by 3 pm and the central Ru−Ntpy is elongated by 6 pm corresponding to the movement of ruthenium toward the dpb ligand within a fixed ligand framework. The 3MLCT state of 1+, however, is slightly distorted compared to the 1GS geometry of 1+: The central pyridine ring of the tpy ligand is somewhat shifted out of the plane perpendicular to the dpb ligand. Again, the ruthenium atom is closer to the tpy ligand because of the dipolar character of this excited state. The dissociative character of the 3MC state is also found for 1+ with characteristically elongated Ru−Ntpy bond lengths and a significant distortion of the peripheral pyridine rings away from the metal center. For Figure 7. Cyclic voltammograms of 1(PF6) (upper, blue) and 2(PF6) emissive 2+, the 3LL′CT state could not be found. However, the (lower, red) (c = 1 mM) in 0.1 M acetonitrile solution of corresponding 1LL′CT absorption is calculated to be ∼3800 [nBu4N][PF6] at 298 K. Potentials are referenced against the FcH/ cm−1 (45 kJ mol−1) higher in energy as compared to that of 1+ FcH + couple (E1/2 = 0.40 V vs SCE). (Figure 6). Hence, we suggest that the 3LL′CT state does not + + play a significant role for the excited-state dynamics of 2+. Table 3. Electrochemical Data of 1 , 2 , and + n The order of these three states gives a straightforward [(dpb)Ru(tpy)] (1 mM) in 0.1 M [ Bu4N][PF6] Electrolytea explanation to the nonemissive behavior of 1(PF6). The lowest- Solution lying triplet excited state is the 3LL′CT state (Figure 6). Similar Eox,1, V E1 ′ ox,2 , V Ered,1, V Ered,2, V Ered,3, V to the corresponding LL CT transition in the absorption (vide + supra), a 3LL′ 1 1 0.02 0.77 −1.82 −2.25 −2.42CT→ GS emission process is symmetry- 2+ 0.14 1.28 −1.96 −2.29 forbidden due to the orthogonality of the two ligands. The [Ru(dpb)(tpy)]+b 0.12 1.36 −1.95 only available deactivation pathway is via radiationless ISC into a + the ground state followed by vibrational relaxation. The Potentials are given in volts and referenced against the FcH/FcHb 3MLCT state that could evolve into the ground state radiatively couple (E1/2 = 0.40 V vs SCE). Values taken from the literature. 20 is ∼10 kJ mol−1 higher in energy than the 3LL′CT state and is only very inefficiently populated at room temperature. to the RuII/RuIII redox couple with contributions from the Apart from the 3LL′CT state described by the strongly highest occupied π-orbital of the cyclometalating ligand electron-accepting tpy and the electron-donating dpb ligand, (HOMO of 1+ and 2+, Supporting Information Figure S22) the 3MLCT−3MC separation is calculated to be substantially as evidenced from Mulliken spin population analysis and spin larger in 1+ than in 2+ also due to the stronger push−pull density plots of 12+ and 22+ (Supporting Information Figures substitution in 1+ that stabilizes the tpy-based LUMO while at S30 and S31). The origin of the second oxidation process, the same time destabilizing the metal-centered excited states. however, is less clear. DFT calculations suggest a mixed In summary, the introduction of a carbon atom in the oxidation of the metal center and cyclometalating ligand as coordination sphere of ruthenium indeed increases the primary step (see Supporting Information Figure S32), but the 3MLCT−3MC energy gap sufficiently to render [Ru(dpb)- irreversibility of this process points to follow-up reactions, so (tpy)]+ and 2+ emissive at room temperature, while [Ru- that the IIfinal oxidation product remains unidentified. The Ru / (tpy) ]2+2 is silent. This effect is reinforced by attaching Ru III oxidation is shifted by 0.6−0.7 V to lower potentials additional donor and acceptor functionalities in the ligand compared to bis(terpyridine)ruthenium complexes bearing the periphery that further increase this 3MLCT−3MC energy gap. same functional groups.89 This illustrates the strong σ-donor Unfortunately, this push−pull approach suffers from the character of the cyclometalating ligand that greatly increases the concomitant formation of a very low-lying 3LL′CT dark state electron density at the metal center. The NHCOMe group at when the donor−acceptor substitution becomes too strong. the cyclometalating ligand in 1(PF6) indeed leads to a further Electrochemical Properties of 1(PF6) and 2(PF6). The shift of the ruthenium-based oxidation by 0.10 V to lower spatial orientation and symmetry of the frontier orbitals were values as compared to [Ru(tpy)(dpb)]+, while the COOEt further experimentally probed by electrochemical and EPR substitution of the dpb ligand in 2(PF6) slightly increases this experiments. The cyclic voltammograms of the complexes redox potential by 0.02 V. 1(PF6) and 2(PF6) in acetonitrile using 0.1 M [ nBu4N][PF6] as In the range accessible for reduction (up to −2.5 V vs FcH/ supporting electrolyte are shown in Figure 7 and Supporting FcH+) three quasireversible or irreversible redox waves are Information Figures S28 and S29. Both complexes show a found for 1(PF6), while for 2(PF6) only two reduction waves reversible oxidation wave at a potential of 0.02 V vs FcH/FcH+ are detected. These are assigned to ligand-centered reductions (1(PF6)) and 0.14 V vs FcH/FcH + (2(PF6); see Table 3). with the first one centered on the tpy ligand. The localization of Additionally, an irreversible oxidation occurs for both this redox process on the ligand leads to a much stronger complexes at higher potentials. The first oxidation is ascribed dependence of the corresponding redox potential on the tpy 11100 DOI: 10.1021/acs.inorgchem.5b01151 Inorg. Chem. 2015, 54, 11088−11104 84 | 3 RESULTS AND DISCUSSION Inorga nic Chemistry Article Figure 8. DFT-calculated spin densities (B3LYP/def2-TZVP/DKH/COSMO(acetonitrile), contour value: 0.01) of 1 (blue, left) and 2 (red, right) and experimental X-band EPR spectra (ν ≈ 9.4 GHz) obtained from frozen acetonitrile solutions of 1 and 2 (c = 5 mM) generated in situ with Co(Cp*)2. CH hydrogen atoms are omitted for clarity. Table 4. Electron Paramagnetic Resonance Spectroscopic Parameters of 12+, 22+, 1, and 2 Determined Experimentally (in Frozen Acetonitrile Solution at 77 K) and Theoretically (B3LYP, def2-TZVP, DKH, COSMO(Acetonitrile))a A (99,1011,2,3 Ru)/MHz (A 14 iso/ A1,2,3( N,1)/MHz (Aiso/ A ( 14 1,2,3 N,2)/MHz (Ab c d d d iso / g1,2,3 giso Δg MHz) MHz) MHz) 12+ expt. 2.238, 2.176, 2.045 2.153 0.193 84, 140, 112 (112) calcd. 2.312, 2.172, 2.020 2.168 0.292 88, 162, 91 (114) 22+ calcd. 2.586, 2.427, 2.021 2.345 0.565 94, 187, 103 (128) 1 expt. 2.0008, 1.9921, 1.9594, 1.9841 0.0414 28, 36, 73 (46) 17, 10, 45 (24) calcd. 2.0147, 2.0001, 1.9414 1.9854 0.0733 41, 41, 63 (48) 6, 5, 48 (19)e 2 expt. 2.0008, 1.9998, 1.9685, 1.9897 0.0323 8, 8, 8 (8) 3, 7, 31 (14) 3, 6, 39 (16) calcd. 2.0034, 2.0009, 1.9963 2.0002 0.0071 11, 8, 6 (8) 0, 0, 29 (10)f 2, 1, 46 (16)f aFor theoretically determined hyper ne coupling constants A(14N), only the largest values (A (14N) > 10 MHz) are given. bfi iso giso = (g1 + g2 + g3)/3. cΔg = g − g d e f1 3. Aiso = (A1 + A2 + A3)/3. Hyperfine coupling to the central pyridine nitrogen atom of the tpy ligand. Hyperfine coupling to one peripheral pyridine nitrogen atom of the tpy ligand. functional groups than in the case of the metal-centered acetonitrile, −1.94 V vs FcH/FcH+ in dichloromethane). To oxidation. Changing the functional group on the tpy ligand study the nature of the oxidized and reduced species EPR from ethyl carboxy (1(PF6)) to N-acetyl amino (2(PF6)) shifts spectra were recorded. Solutions were prepared at a 5 mM the first reduction potential by 0.14 V to more negative values sample concentration with 0.9 equiv of the respective oxidant (cf. MO diagrams in Supporting Information Figure S22). or reductant. While both complexes show EPR signals at 77 K The oxidation steps in the potential range of −1.0 to −0.5 V after being reduced to 1 and 2 (see Figure 8), only 12+ is for 1(PF6) following the irreversible reduction processes are detected via X-band EPR spectroscopy at 77 K (see Supporting similar to those observed for bis(terpyridine)ruthenium(II) Information Figure S30). At room temperature in dichloro- complexes bearing amide functional groups and might be methane solution all samples were EPR-silent. associated with reduction of the NH proton at the terpyridine The EPR signal of 12+ (Supporting Information Figure S30; moiety to hydrogen.88 Table 4) is highly anisotropic (Δg = 0.193) and very broad Electron Paramagnetic Resonance Studies on Redox indicating a strong contribution of metal d-orbitals to the spin Products. As evidenced from the cyclic voltammograms, both density. Hyperfine couplings are not resolved in the spectrum, complexes can be oxidized to 12+ and 22+ using tris(4- but a good estimate of the coupling constants of the electronic bromophenyl)aminium hexachloridoantimonate as oxidant spin to the nuclear spin of ruthenium (99Ru and 101Ru: I = 5/2, (0.67 V vs FcH/FcH+ in acetonitrile, 0.70 V vs FcH/FcH+ in natural abundance: 30%) is obtained by simulations. This dichloromethane) and reduced to 1 and 2 using decamethylco- coupling (A (99,1011,2,3 Ru) = 84, 140, 112 MHz) is large baltocene Co(Cp*)2 as reductant (−1.91 V vs FcH/FcH+ in underlining the strong metal contribution to the radical 11101 DOI: 10.1021/acs.inorgchem.5b01151 Inorg. Chem. 2015, 54, 11088−11104 Section 3.2 | 85 Inorganic C hemistry Article character. Theoretical g tensor and hyperfine coupling range absorption spectrum is dominated by MLCT transitions parameter calculations on the DFT wave function generated to the electron-poor terpyridine (Ru → tpy) as well as the at the B3LYP/def2-TZVP/DKH/COSMO(acetonitrile) level electron-rich dipyridylphenyl ligand (Ru → dpb), which was of theory are in excellent agreement with the experimentally evidenced by a combined DFT and rR spectroscopic approach. determined quantities. 22+ is EPR-silent at 77 K and at room Theoretical calculations additionally suggest a symmetry- temperature, but theory predicts similar g values and hyperfine forbidden and hence experimentally undetected 1LL′CT as coupling constants to ruthenium as observed for 12+ (Table 4). lowest spin-allowed optical transition in the red spectral range. EPR spectra of 1 and 2 obtained upon reduction of the The first oxidation is metal-centered in both complexes 1(PF6) respective cations (Figure 8) are substantially sharper and and 2(PF6) with substantial contribution from the central better-resolved. The g-value anisotropy (Δg = 0.0414 for 1, Δg phenyl ring of the dpb ligand, which corresponds to the ground = 0.0323 for 2) is reduced by a factor of 5 compared to the EPR state HOMO in both cases. The reduction is tpy-centered with spectrum of 12+ indicating a significantly stronger ligand-based the unpaired electron localized on the central pyridine ring in character of the radical. Hyperfine couplings to the ruthenium 1(PF6), while it is delocalized over both peripheral pyridine center and one or two nitrogen atoms for 1 and 2, respectively, rings of tpy in 2(PF6). Reduction of 2 + to 2 reverses the order are well-resolved and were determined by simulations of the of the unoccupied orbitals LUMO and LUMO+1 as they are experimental spectra (see Table 4). The observed spectra are close in energy in 2+ resulting in a characteristic fingerprint in easily explained with the reduction occurring at the tpy ligand the respective EPR spectra. (cf. MO diagram in Supporting Information Figure S22). Spin While both isomers have similar absorption and electro- density calculations for the neutral complexes 1 and 2 explain chemical characteristics they differ fundamentally in their the occurrence of a single nitrogen hyperfine coupling in 1, excited-state and emission behavior. 1(PF6) is nonemissive while 2 shows couplings to two chemically different nitrogen both at room temperature and at 77 K, while 2(PF6) shows a atoms. The unpaired electron in 1 is essentially localized at the very weak emission at room temperature and a much stronger central pyridine ring of the tpy−COOEt ligand with its highest luminescence at 77 K. Temperature-dependent emission coefficient at the nitrogen p−π orbital leading to strong spectroscopy revealed that a very low activation barrier of ca. anisotropic superhyperfine coupling to that nitrogen nucleus, 11 kJ mol−1 for the thermal deactivation of the emissive which leads to a distinctive triplet splitting of the g3 signal 3MLCT state via a 3MC state is responsible for the measurable, (Table 4). In 2, however, the unpaired electron is delocalized but low, emission quantum yield at room temperature. over the two peripheral pyridine rings of tpy−NHCOMe in 2 For 1(PF6) a completely different picture emerges. The with a nodal plane orthogonal to the ligand plane containing stronger push−pull substitution substantially raises the 3MC the metal center. Consequently, the EPR signal is much less states in energy, which should lead to an increase in emission well-resolved especially because the superhyperfine coupling intensity compared to 2(PF6). Indeed, DFT calculations find constants to the two peripheral nitrogen nuclei differ (Figure the 3MC states high in energy. Hence, these do not contribute 8). This renders unambiguous determination of all super- to the efficient excited-state deactivation in 1(PF6). Instead, an hyperfine and hyperfine coupling constants (except A3 of the unrecognized 3LL′CT state was found to be lower in energy two nitrogen atoms) in 2 rather challenging, so these are than the 3MLCT state in 1(PF 36). The LL′CT state undergoes estimated from line width and broadening (Table 4). The radiationless deactivation as the radiative relaxation is electron-donating effect of the N-acetyl amino group attached symmetry-forbidden (dark state). to the terpyridine ligand in 2 increases the electron density at In essence, cyclometalation using dpb ligands shifts the 3MC the central pyridine ring and consequently varies the character state above the 3MLCT for 1+ compared to [Ru(tpy) ]2+2 thus and symmetry of the singly occupied molecular orbital reducing the emission quenching via thermal depopulation (SOMO). It resembles the LUMO+1 of 2+ (see Supporting through 3MC states. At the same time it generates low-lying Information, Figure S22), whereas the SOMO of 1 is similar to 3LL′CT states that evolve radiationless into the ground state the LUMO of 1+ (DFT: B3LYP/def2-TZVP/DKH/COSMO- due to the symmetry-forbidden character of the transition (acetonitrile)). The superhyperfine coupling constants to imposed by the orthogonality of the ligands. This quenching via nitrogen are smaller for 2 than for 1 because of this spin a 3LL′CT state is dominant for push−pull substituted delocalization over two pyridine rings (spin dilution). cyclometalated bis(tridentate)ruthenium(II) complexes of Furthermore, the Ru−Nterminal distances are larger than the ruthenium, and the underlying mechanisms should be trans- Ru−N 26central distances leading to a reduction of the spin−orbit ferable to iridium(III) as well, where similar nonemissive coupling affecting the unpaired electron and consequently a behavior has been observed. lowering of the ruthenium hyperfine coupling in 2 and the g- Despite the fact that [Ru(dpb)(tpy)]+ complexes are value anisotropy. Since the amide bridge is rigid and rotation nonemissive or only weakly emissive at room temperature, about the Namide−Cterpyridine bond is slow at the EPR time scale the charge-separation at the Franck−Condon point and the (especially in rigid matrix) the spin density is asymmetric, high reducing potential of the excited state, both induced by the which explains the slight differences in coupling constants to cyclometalation, render these complexes promising candidates the two peripheral coordinating nitrogen atoms. as sensitizers in photoredox applications. ■ CONCLUSION ■ ASSOCIATED CONTENT The key properties of the cyclometalated bis(tridentate)- *S Supporting Information ruthenium(II) complexes [Ru(dpb-NHCOMe)(tpy- Experimental procedure for the synthesis of L3; character- COOEt)]+ 1+ and [Ru(dpb-COOEt)(tpy-NHCOMe)]+ 2+ ization of 3(PF6); discussion of the synthetic procedures of L1, were revealed by introduction of electron-donating and L2 and L3; FD mass spectra, 1H and 13C NMR spectra of L1, electron-accepting functional groups in the ligand periphery L2 and L3; ESI mass spectra, 1H, 13C, COSY, HSQC, and of [Ru(tpy)(dpb)]+ complexes. For both isomers the visible- HMBC NMR spectra of 1(PF6) and 2(PF6); 1H and 13C NMR 11102 DOI: 10.1021/acs.inorgchem.5b01151 Inorg. Chem. 2015, 54, 11088−11104 86 | 3 RESULTS AND DISCUSSION Inorga nic Chemistry Article spectra of 3(PF6); solid-state IR spectra and room-temperature (19) Meggers, E. Chem. - Eur. J. 2010, 16, 752−758. emission spectra of 1(PF6) and 2(PF ); decomposition of the (20) Wadman, S. H.; Lutz, M.; Tooke, D. M.; Spek, A. L.; Hartl, F.;6 absorption spectra of 1(PF6) and 2(PF6); MO diagrams of Havenith, R. W. A.; van Klink, G. P. M.; van Koten, G. Inorg. Chem. 1(PF ) and 2(PF ); TD-DFT excitations of 1(PF ) and 2(PF ) 2009, 48, 1887−1900.6 6 6 6 with di erence density plots; cyclic and square wave (21) Heinze, K.; Hempel, K.; Beckmann, M. Eur. J. Inorg. Chem.ff voltammograms of 1(PF6) and 2(PF6); EPR spectrum of 1 2+; 2006, 2006, 2040−2050. 2+ 3+ 3+ (22) Breivogel, A.; Förster, C.; Heinze, K. Inorg. Chem. 2010, 49,DFT spin-density plots of 2 , 2 and 2 ; Cartesian + + 2+ 2+ 2+ 3+ 7052−7056.coordinates of geometry-optimized 1 , 1 , 2 , 1 , 2 , 1 , (23) Breivogel, A.; Meister, M.; Förster, C.; Laquai, F.; Heinze, K. 23+, 1, and 2, as well as of the triplet states 3LL′CT, 3MLCT, Chem. - Eur. J. 2013, 19, 13745−13760. and 3MC of 1+ and 3MLCT and 3MC of 2+. The Supporting (24) Abrahamsson, M.; Jag̈er, M.; Österman, T.; Eriksson, L.; Information is available free of charge on the ACS Publications Persson, P.; Becker, H.-C.; Johansson, O.; Hammarström, L. J. Am. website at DOI: 10.1021/acs.inorgchem.5b01151. Chem. Soc. 2006, 128, 12616−12617. (25) Schramm, F.; Meded, V.; Fliegl, H.; Fink, K.; Fuhr, O.; Qu, Z.; ■ AUTHOR INFORMATION Klopper, W.; Finn, S.; Keyes, T. E.; Ruben, M. Inorg. Chem. 2009, 48,5677−5684. Corresponding Author (26) Wilkinson, A. J.; Puschmann, H.; Howard, J. A. K.; Foster, C. E.; *Fax: +49613127277. E-mail: katja.heinze@uni-mainz.de. Williams, J. A. G. Inorg. Chem. 2006, 45, 8685−8699. Author Contributions (27) Breivogel, A.; Kreitner, C.; Heinze, K. Eur. J. Inorg. Chem. 2014, The manuscript was written through contributions of all 2014, 5468−5490. authors; rR spectra were measured by E.E. and W.W.S. All (28) Bolink, H. J.; Cappelli, L.; Coronado, E.; Gaviña, P. Inorg. Chem. authors have given approval to the final version of the 2005, 44, 5966−5968. manuscript. (29) Heinze, K.; Hempel, K.; Breivogel, A. Z. Anorg. Allg. Chem. 2009, 635, 2541−2549. Notes (30) Englman, R.; Jortner, J. Mol. Phys. 1970, 18, 145−164. The authors declare no competing financial interest. (31) Caspar, J. V.; Kober, E. M.; Sullivan, B. P.; Meyer, T. J. J. Am. Chem. Soc. 1982, 104, 630−632. ■ ACKNOWLEDGMENTS (32) Caspar, J. V.; Meyer, T. J. J. Phys. Chem. 1983, 87, 952−957. Parts of this research were conducted using the supercomputer (33) Parada, G. A.; Fredin, L. A.; Santoni, M.-P.; Jag̈er, M.; Lomoth, Mogon and advisory services o ered by Johannes Gutenberg R.; Hammarström, L.; Johansson, O.; Persson, P.; Ott, S. Inorg. Chem.ff 2013, 52, 5128−5137. Univ. Mainz (www.hpc.uni-mainz.de), which is a member of (34) Constable, E. C.; Holmes, J. M. J. Organomet. Chem. 1986, 301, the AHRP and the Gauss Alliance e.V. This work was 203−208. financially supported by the Deutsche Forschungsgemeinschaft (35) Beley, M.; Collin, J. P.; Sauvage, J. P. Inorg. Chem. 1993, 32, (GSC 266, Materials Science in Mainz, scholarship for C.K.). 4539−4543. (36) King, K. A.; Spellane, P. J.; Watts, R. J. J. Am. Chem. Soc. 1985, ■ REFERENCES 107, 1431−1432. (1) Brandt, W. W.; Dwyer, F. P.; Gyarfas, E. D. Chem. Rev. 1954, 54, (37) King, K. A.; Watts, R. J. J. Am. Chem. Soc. 1987, 109, 1589− 959−1017. 1590. (2) van Houten, J.; Watts, R. J. Inorg. Chem. 1978, 17, 3381−3385. (38) Lamansky, S.; Djurovich, P.; Murphy, D.; Abdel-Razzaq, F.; Lee, (3) Durham, B.; Caspar, J. V.; Nagle, J. K.; Meyer, T. J. J. Am. Chem. H.-E.; Adachi, C.; Burrows, P. E.; Forrest, S. R.; Thompson, M. E. J. Soc. 1982, 104, 4803−4810. Am. Chem. Soc. 2001, 123, 4304−4312. (4) Kalyanasundaram, K. Coord. Chem. Rev. 1982, 46, 159−244. (39) Lowry, M. S.; Bernhard, S. Chem. - Eur. J. 2006, 12, 7970−7977. (5) Caspar, J. V.; Meyer, T. J. J. Am. Chem. Soc. 1983, 105, 5583− (40) Constable, E. C.; Housecroft, C. E. Polyhedron 1990, 9, 1939− 5590. 1947. (6) Meyer, T. J. Pure Appl. Chem. 1986, 58, 1193−1206. (41) Bomben, P. G.; Robson, K. C. D.; Sedach, P. A.; Berlinguette, C. (7) Kasha, M. Discuss. Faraday Soc. 1950, 9, 14−19. P. Inorg. Chem. 2009, 48, 9631−9643. (8) Demas, J. N.; Taylor, D. G. Inorg. Chem. 1979, 18, 3177−3179. (42) Bomben, P. G.; Koivisto, B. D.; Berlinguette, C. P. Inorg. Chem. (9) Yoon, S.; Kukura, P.; Stuart, C. M.; Mathies, R. A. Mol. Phys. 2010, 49, 4960−4971. 2006, 104, 1275−1282. (43) Chirdon, D. N.; McCusker, C. E.; Castellano, F. N.; Bernhard, (10) Lytle, F. E.; Hercules, D. M. J. Am. Chem. Soc. 1969, 91, 253− S. Inorg. Chem. 2013, 52, 8795−8804. 257. (44) Colombo, M. G.; Hauser, A.; Guedel, H. U. Inorg. Chem. 1993, (11) Suzuki, K.; Kobayashi, A.; Kaneko, S.; Takehira, K.; Yoshihara, 32, 3088−3092. T.; Ishida, H.; Shiina, Y.; Oishi, S.; Tobita, S. Phys. Chem. Chem. Phys. (45) Hay, P. J. J. Phys. Chem. A 2002, 106, 1634−1641. 2009, 11, 9850−9860. (46) Flamigni, L.; Barbieri, A.; Sabatini, C.; Ventura, B.; Barigelletti, (12) Lashgari, K.; Kritikos, M.; Norrestam, R.; Norrby, T. Acta F. Top. Curr. Chem. 2007, 281, 143−203. Crystallogr., Sect. C: Cryst. Struct. Commun. 1999, 55, 64−67. (47) Wadman, S. H.; Kroon, J. M.; Bakker, K.; Lutz, M.; Spek, A. L.; (13) Stone, M. L.; Crosby, G. A. Chem. Phys. Lett. 1981, 79, 169− van Klink, G. P. M.; van Koten, G. Chem. Commun. 2007, 1907−1909. 173. (48) Wadman, S. H.; Kroon, J. M.; Bakker, K.; Havenith, R. W. A.; (14) Winkler, J. R.; Netzel, T. L.; Creutz, C.; Sutin, N. J. Am. Chem. van Klink, G. P. M.; van Koten, G. Organometallics 2010, 29, 1569− Soc. 1987, 109, 2381−2392. 1579. (15) Maestri, M.; Armaroli, N.; Balzani, V.; Constable, E. C.; (49) Bruno, N. C.; Tudge, M. T.; Buchwald, S. L. Chem. Sci. 2013, 4, Thompson, A. M. W. C. Inorg. Chem. 1995, 34, 2759−2767. 916−920. (16) Amini, A.; Harriman, A.; Mayeux, A. Phys. Chem. Chem. Phys. (50) Chavarot, M.; Pikramenou, Z. Tetrahedron Lett. 1999, 40, 2004, 6, 1157−1164. 6865−6868. (17) Sun, P.; Krishnan, A.; Yadav, A.; Singh, S.; MacDonnell, F. M.; (51) Constable, E. C.; Ward, M. D. J. Chem. Soc., Dalton Trans. 1990, Armstrong, D. W. Inorg. Chem. 2007, 46, 10312−10320. 1405−1409. (18) Gong, L.; Mulcahy, S. P.; Harms, K.; Meggers, E. J. Am. Chem. (52) Fallahpour, R.-A. Eur. J. Inorg. Chem. 1998, 1998, 1205−1207. Soc. 2009, 131, 9602−9603. (53) Suda, K.; Akagi, K. Macromolecules 2011, 44, 9473−9488. 11103 DOI: 10.1021/acs.inorgchem.5b01151 Inorg. Chem. 2015, 54, 11088−11104 Section 3.2 | 87 Inorganic C hemistry Article (54) Fulmer, G. R.; Miller, A. J. M.; Sherden, N. H.; Gottlieb, H. E.; Nudelman, A.; Stoltz, B. M.; Bercaw, J. E.; Goldberg, K. I. Organometallics 2010, 29, 2176−2179. (55) Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877−910. (56) Stoll, S.; Schweiger, A. J. Magn. Reson. 2006, 178, 42−55. (57) Neese, F. WIREs Comput. Mol. Sci. 2012, 2, 73−78. (58) Neese, F.; Wennmohs, F.; Hansen, A.; Becker, U. Chem. Phys. 2009, 356, 98−109. (59) Izsaḱ, R.; Neese, F. J. Chem. Phys. 2011, 135, 144105. (60) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77, 3865−3868. (61) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1997, 78, 1396. (62) Schaf̈er, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97, 2571−2577. (63) Schaf̈er, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100, 5829−5835. (64) Andrae, D.; Haüßermann, U.; Dolg, M.; Stoll, H.; Preuß, H. Theoret. Chim. Acta 1990, 77, 123−141. (65) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7, 3297−3305. (66) Sinnecker, S.; Rajendran, A.; Klamt, A.; Diedenhofen, M.; Neese, F. J. Phys. Chem. A 2006, 110, 2235−2245. (67) Vlcěk, A.; Zaĺis,̌ S. Coord. Chem. Rev. 2007, 251, 258−287. (68) Magyar, R. J.; Tretiak, S. J. Chem. Theory Comput. 2007, 3, 976− 987. (69) Tao, J.; Perdew, J. P.; Staroverov, V. N.; Scuseria, G. E. Phys. Rev. Lett. 2003, 91, 146401. (70) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652. (71) Adamo, C.; Barone, V. J. Chem. Phys. 1999, 110, 6158−6170. (72) Yanai, T.; Tew, D. P.; Handy, N. C. Chem. Phys. Lett. 2004, 393, 51−57. (73) Douglas, M.; Kroll, N. M. Ann. Phys. 1974, 82, 89−155. (74) Hess, B. Phys. Rev. A: At., Mol., Opt. Phys. 1985, 32, 756−763. (75) Hess, B. Phys. Rev. A: At., Mol., Opt. Phys. 1986, 33, 3742−3748. (76) Jansen, G.; Hess, B. Phys. Rev. A: At., Mol., Opt. Phys. 1989, 39, 6016−6017. (77) Pantazis, D. A.; Chen, X.-Y.; Landis, C. R.; Neese, F. J. Chem. Theory Comput. 2008, 4, 908−919. (78) Petrenko, T.; Neese, F. J. Chem. Phys. 2007, 127, 164319. (79) Petrenko, T.; Neese, F. J. Chem. Phys. 2012, 137, 234107. (80) Beley, M.; Collin, J. P.; Louis, R.; Metz, B.; Sauvage, J. P. J. Am. Chem. Soc. 1991, 113, 8521−8522. (81) Koivisto, B. D.; Robson, K. C. D.; Berlinguette, C. P. Inorg. Chem. 2009, 48, 9644−9652. (82) Dreuw, A.; Weisman, J. L.; Head-Gordon, M. J. Chem. Phys. 2003, 119, 2943. (83) Tozer, D. J. J. Chem. Phys. 2003, 119, 12697−12699. (84) Heinze, K.; Hempel, K.; Tschierlei, S.; Schmitt, M.; Popp, J.; Rau, S. Eur. J. Inorg. Chem. 2009, 2009, 3119−3126. (85) Preiß, J.; Jag̈er, M.; Rau, S.; Dietzek, B.; Popp, J.; Martínez, T.; Presselt, M. ChemPhysChem 2015, 16, 1395−1404. (86) Egidi, F.; Bloino, J.; Cappelli, C.; Barone, V. J. Chem. Theory Comput. 2014, 10, 346−363. (87) Heller, E. J.; Sundberg, R.; Tannor, D. J. Phys. Chem. 1982, 86, 1822−1833. (88) Kreitner, C.; Grabolle, M.; Resch-Genger, U.; Heinze, K. Inorg. Chem. 2014, 53, 12947−12961. (89) Breivogel, A.; Hempel, K.; Heinze, K. Inorg. Chim. Acta 2011, 374, 152−162. 11104 DOI: 10.1021/acs.inorgchem.5b01151 Inorg. Chem. 2015, 54, 11088−11104 This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes. Addition/Correction pubs.acs.org/IC Correction to “Understanding the Excited State Behavior of Cyclometalated Bis(tridentate)ruthenium(II) Complexes: A Combined Experimental and Theoretical Study” Christoph Kreitner, Elisa Erdmann, Wolfram W. Seidel, and Katja Heinze* Inorg. Chem. 2015, 54 (23), 11088−11104. DOI: 10.1021/acs.inorgchem.5b01151 Pages 11100 and 11101. The electrochemical data of compound first reduction potential by 0.14 V to more negative values 1+ were erroneously referenced against the decamethylferrocene/ (cf. MO diagrams in Supporting Information Figure S22).” decamethylferrocenium couple in dichloromethane instead of should by replaced by acetonitrile. This results in erroneously reported redox potentials “Changing the functional group on the tpy ligand from ethyl for 1+ by 0.11 V.1 carboxy (1(PF6)) to N-acetyl amino (2(PF6)) shifts the The corrected first row of Table 3 should read as follows: first reduction potential by 0.25 V to more negative values (cf. MO diagrams in Supporting Information Figure S22).” Eox,1, V Eox,2, V Ered,1, V Ered,2, V Ered,3, V We regret the mistake, which does not affect the key find- 1+ 0.13 0.88 −1.71 −2.14 −2.31 ings reported in the paper, and sincerely apologize for any inconvenience. Figure 7 should be replaced by the following: ■ REFERENCES (1) Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877−910. The sentence “Both complexes show a reversible oxidation wave at a potential of 0.02 V vs FcH/FcH+ (1(PF6)) and 0.14 V vs FcH/FcH+ (2(PF6)) (see Table 3).” should be replaced by “Both complexes show a reversible oxidation wave at a potential of 0.13 V vs FcH/FcH+ (1(PF6)) and 0.14 V vs FcH/FcH+ (2(PF6)) (see Table 3).” The sentence “The NHCOMe group at the cyclometalating ligand in 1(PF6) indeed leads to a further shift of the ruthenium-based oxidation by 0.10 V to lower values as compared to [Ru(tpy)(dpb)]+, while the COOEt substitution of the dpb ligand in 2(PF6) slightly increases this redox potential by 0.02 V.” should be deleted. The sentence “Changing the functional group on the tpy ligand from ethyl carboxy (1(PF6)) to N-acetyl amino (2(PF6)) shifts the Published: December 4, 2015 © 2015 American Chemical Society 12046 DOI: 10.1021/acs.inorgchem.5b02701 Inorg. Chem. 2015, 54, 12046−12046 Section 3.3 | 89 3.3 THE PHOTOCHEMISTRY OF MONO- AND DINUCLEAR CYCLOMETALATED BIS(TRIDENTATE)RUTHENIUM(II) COMPLEXES: DUAL EXCITED STATE DEACTIVATION AND DUAL EMISSION Christoph Kreitner and Katja Heinze Dalton Trans. 2016, 45, 5640–5658. A low-energy 3LL′CT state efficiently depopulates the emissive 3MLCT state in cyclometalated [Ru(dpb-R)(tpy)]+ complexes (dpbH = 1,3-di(2-pyridyl)benzene, tpy = 2,2′;6′,2′′-terpyridine). Author Contributions The synthesis and characterization of the ruthenium complexes as well as all DFT calculations were performed by Christoph Kreitner. The manuscript was written by Christoph Kreitner (90 %) and Katja Heinze (10 %). Supporting Information for this article is found at pp. 209 (excluding Cartesian Coordinates of DFT-optimized geometries). For full Supporting Information, refer to http://www.rsc.org/suppdata/c6/dt/c6dt00384b/ c6dt00384b1.pdf. „Kreitner, C.; Heinze, K. Dalton Trans. 2016, 45, 5640–5658. - Published by The Royal Society of Chemistry.” 90 | 3 RESULTS AND DISCUSSION Dalton Transactions View Article Online PAPER View Journal | View Issue The photochemistry of mono- and dinuclear cyclometalated bis(tridentate)ruthenium(II) Cite this: Dalton Trans., 2016, 45, 5640 complexes: dual excited state deactivation and dual emission† Christoph Kreitnera,b and Katja Heinze*a The synthesis and characterization of a series of weakly emissive mononuclear cyclometalated [Ru(dpb-R) (tpy)]+ complexes with functional groups R of varying electron-donating characters at the dpb ligand are described (dpbH = 1,3-di(2-pyridyl)benzene, tpy = 2,2’;6’,2’’-terpyridine, 1+: R = NHCOMe, 2+: R = NH2, 3+: R = COOEt, 4+: R = COOH). Steady-state emission spectroscopy in the temperature range between 298 K and 77 K revealed a previously unrecognized excited state deactivation pathway via low-lying triplet ligand-to-ligand (3LL’CT) charge transfer states in addition to the well-known pathway via 3MC states. Thermal activation barriers for depopulation of the emissive metal-to-ligand charge transfer (3MLCT) states via the 3MC (metal-centered) and 3LL’CT states were determined experimentally for complexes 1+ and 3+. The experimental results were further corroborated by calculating the respective 3MLCT–3LL’CT and 3MLCT–3MC transition states and their energies with density functional theoretical methods. The R substituent modifies the energy difference between the 3MLCT and 3LL’CT states and the corresponding activation barrier but leaves the analogous 3MLCT/3MC energetics essentially untouched. Additionally, the dinuclear complex [(tpy)Ru(dpb-NHCO-dpb)Ru(tpy)]2+, 62+, containing a biscyclometalating bridge was devised. Despite the asymmetric nature induced by the amide bridge, the mixed-valent cation 63+ is ascribed to Robin–Day class II with a broad and intense intervalence charge-transfer (IVCT) absorption Received 27th January 2016, (λmax = 1165 nm). Upon optical excitation, the Ru II/RuII complex 62+ exhibits dual emission in liquid solu- Accepted 16th February 2016 tion from two independently emitting 3MLCT states localized at the two remote [Ru(tpy)] fragments. No DOI: 10.1039/c6dt00384b equilibration via Dexter energy transfer is possible due to their large distance and short excited state www.rsc.org/dalton lifetimes. 2+ Introduction The prototype of this class of complexes is [Ru(bpy)3] (bpy = 2,2′-bipyridine), whose photophysical properties have Polypyridine complexes of ruthenium(II) have been known and been extensively studied and are well understood. Under studied for the past sixty years.1,2 Although the fundamentals visible light irradiation, excitation into a singlet metal-to- of their photo- and electrochemical properties are well ligand charge transfer (1MLCT) state occurs (λmax = 452 nm, understood,3–7 research efforts have not diminished over the εmax = 14.6 M −1 cm−1).1,3 This state undergoes rapid and quan- last few years mainly due to a widespread potential for appli- titative intersystem crossing onto the triplet hypersurface18,19 cations for this class of metal complexes. These vary from populating a long-lived 3MLCT state that is phosphorescent at photoredox catalysis,8–12 over light sensitization in dye-sensi- room temperature (λem = 621 nm, ϕ = 0.095, τ = 855 ns in tized solar cells,13 and sensing applications in biological14,15 MeCN).20,21 Upon cooling, both, emission quantum yield and and chemical16 contexts to optoelectronics.17 excited state lifetime, increase drastically. Using lifetime measurements at varying temperatures, T. J. Meyer and co- workers showed that this temperature dependence is due to a aInstitute of Inorganic and Analytical Chemistry, Johannes Gutenberg University, thermally accessible d–d excited state (metal centered, 3MC) Duesbergweg 10-14, D-55128 Mainz, Germany. E-mail: katja.heinze@uni-mainz.de that rapidly undergoes vibrational relaxation into the ground bGraduate School Materials Science in Mainz, Staudingerweg 9, D-55128 Mainz, state (1GS).20,22 Additionally, this excited state is dissociative in Germany 2+ †Electronic supplementary information (ESI) available. See DOI: 10.1039/ nature and enables [Ru(bpy)3] to undergo photosubstitution 20,22,23 c6dt00384b reactions. 5640 | D alton Trans., 2016, 45, 5640–5658 This journal is © The Royal Society of Chemistry 2016 Open Access Article. Published on 29 February 2016. Downloaded on 20/05/2016 14:13:44. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Section 3.3 | 91 View Article Online Dalton Transactions Paper To suppress these reactions and also to circumvent the states.40 Since the involved ligands are orthogonal to one chiral nature of [Ru(bpy)3] 2+ stronger chelating, tridentate another in the meridional coordination geometry, so are the ligands such as tpy (tpy = 2,2′;6′,2″-terpyridine) were intro- spin-carrying orbitals. Hence, emission from such 3LL′CT duced in bis(tridentate)ruthenium( ) complexes.5,24II Their mer- states is symmetry-forbidden and leads to efficient radiation- idional coordination geometry25 allows the functionalization less deactivation of the excited state. of the ligand periphery without resulting in stereoisomers. To further study this phenomenon and to elaborate a A major drawback of these complexes compared to their bpy general view, the work presented herein devised four cyclo- counterparts is the almost complete lack of emission at room metalated ruthenium complexes [Ru(dpb-R)(tpy)]+ with varying temperature (ϕ = 5 × 10−6).5,26 Due to the weaker ligand field substituents at the 5-position of the dpb ligand (R = NHCOMe, caused by the smaller bite angle of the terpyridine ligand 1+; R = NH , 2+2 ; R = COOEt, 3 +; R = COOH, 4+). Using these, it (N–Ru–N ≈ 79°), the emissive 3MLCT states of [Ru(tpy)2]2+ are is possible to systematically study the impact of varying push– very efficiently depopulated via low-lying 3MC states.24 Upon pull strengths across the metal center on the ground and cooling, thermal depopulation of the emissive state is retarded excited state properties of these cyclometalated complexes. By yielding bright luminescence at 77 K (λem = 599 nm, ϕ = 0.48, employing steady-state emission spectroscopy, we demonstrate τ = 110 µs in MeOH/EtOH).27 that the occurrence of low-energy 3LL′CT states is a common Various attempts have been made to regain room tempera- theme in cyclometalated bis(tridentate)ruthenium complexes ture luminescence from bis(tridentate)ruthenium(II) com- providing a second excited state deactivation pathway in plexes. By introducing an electron-donating functional group addition to the well-known pathway mediated by 3MC states. on one of the terpyridine ligands, the energy of the 3MC state Additionally, the presence of free amino and carboxylic acid is increased with respect to the 3MLCT state energy rather groups allows the straightforward formation of a dinuclear selectively.5 Similarly, electron-accepting functionalities lower complex with an amide-linked biscyclometalating bridging the 3MLCT state energy.5 Combining these two approaches in ligand ([(tpy)Ru(dpb-NHCO-dpb)Ru(tpy)]2+ (62+) that we syn- a push–pull system, the activation barrier for depopulation of thesized and studied as well. Dinuclear bisruthenium com- the emissive 3MLCT state is increased. As a result, room temp- plexes received wide interest since the discovery of the mixed- erature quantum yields of up to 0.003 and excited state life- valent Creutz–Taube ion, [(NH ) Ru(µ-pz)Ru(NH ) ]5+3 5 3 5 (pz = times of 50 ns are achieved.5,28–31 Since the coordination mode pyrazine).41–43 The ruthenium oxidation states within this of the tpy ligand with five-membered chelate rings only mixed-valent complex cannot be assigned unambiguously. allows for rather constrained geometries around the metal Depending on the spectroscopic method either 2+/3+ or 2.5+/ center with small bite angles, several research groups focussed 2.5+ is obtained.44–47 Dinuclear mixed-valent complexes are on expanding the ligand backbone to increase the overlap assigned to three different groups based on Robin’s and Day’s between the ruthenium d orbitals of the eg set and the nitro- classification. 48 Systems with entirely localized valences and gen lone pairs. This yields an enlarged ligand field splitting no electronic coupling between the redox centers in the mixed- and thus makes 3MC states as deactivation pathway thermally valent state are referred to as Robin–Day class I, and systems less accessible at room temperature.32–35 Following this with entirely delocalized valences are assigned as class III. concept, bis(tridentate)ruthenium(II) complexes were syn- Class II describes valence-localized complexes with measur- thesized with optical properties comparable to those of able electronic interactions between the redox sites. The [Ru(bpy) 2+3] (ϕ = 0.30, τ = 3.3 µs). 33 theoretical basis for an accurate physicochemical treatment of In a very similar approach, by introduction of very strong Robin–Day class II complexes was laid by Hush49–51 describing σ-donors in the coordination sphere, the ligand field splitting the photochemical electron transfer occurring between the can be increased compared to [Ru(tpy)2] 2+. Conceptually, this donor and acceptor sites [Mn+ − M(n+1)+ → M(n+1)+ − Mn+]. This was shown by Berlinguette and Schubert using N-heterocyclic process yields an intervalence charge transfer (IVCT) absorp- carbene containing tridentate ligands (C^N^C coordination tion that is typically observed in the Near Infrared (NIR) region mode) with quantum yields of 0.11 and excited state lifetimes of the electronic absorption spectrum of a Robin–Day class II of up to 8 µs.36 Disappointingly, attempts using 1,3-di(2- compound. According to Marcus–Hush theory, this IVCT band pyridyl)benzene (dpbH), deprotonated in the 2-position of the is correlated with the electronic coupling parameter Vab central benzene ring, as a strong cyclometalating σ-donor between the redox centers calculated as: Vab = 2.06 × 10−2 ligand in conjunction with tpy as a π-accepting ligand ν̃ ·ε ·ν̃ )1/2r−1max max 1/2 with the absorption maximum ν̃max in ([Ru(dpb)(tpy)]+), gave only very weakly emissive cm−1, the extinction coe cient ε at ν̃ in M−1 cm−1ffi max max , complexes.30,37–40 This was originally ascribed to a very small the full width at half maximum ν̃ in cm−11/2 and the donor– activation barrier for thermal depopulation of the emissive acceptor distance r in Å.47,52 3MLCT state via low-lying 3MC states since the cyclometalation Several amide-bridged dinuclear bis(terpyridine)ruthenium(II) at the central position of the dpb ligand merely shifts one of complexes and their mixed-valent counterparts have been the 3MC states to a higher energy.30 Recently, we have described in the literature.53–55 While the back-to-back linked suggested that the introduction of a very strong push–pull (n = 0) or phenylene-extended (n = 1–2) dinuclear bis(terpyri- arrangement across the metal center additionally gives rise to dine)ruthenium(II) complexes [(ttpy)Ru(tpy-(1,4-C6H4)n-tpy)Ru- low-lying triplet ligand-to-ligand charge transfer (3LL′CT) (ttpy)]4+ (ttpy = 4′-tolylterpyridine, 1,4-C6H4 = para-phenylene) This journal is © The Royal Society of Chemistry 2016 Dalton Trans., 2016, 45, 5640–5658 | 5641 Open Access Article. Published on 29 February 2016. Downloaded on 20/05/2016 14:13:44. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 92 | 3 RESULTS AND DISCUSSION View Article Online Paper Dalton Transactions exhibit electronic coupling of the metal centers in the mixed- valent state to a small extent (n = 0: Vab = 0.047 eV, n = 1: Vab = 0.030 eV, n = 2: V = 0.022 eV),56,57ab the introduction of an amide bridge seems to reduce the molecular and redox-chemi- cal symmetry enough to prevent the electronic interactions entirely.54,55 In the cyclometalated analogue of the dinuclear back-to-back linked bis(terpyridine)ruthenium complex, [(ttpy) Ru(dpb-dpb)Ru(tpy)]2+, on the other hand, the metal–metal interaction is increased to V = 0.127 eV.58,59ab This increase was attributed to an energy shift of the bridge’s frontier orbitals to better match those of the metal centers.59,60 In this work, we study to what extent the insertion of an NHCO group into the bridge reduces the electronic coupling between the metal centers in the mixed-valent state 63+ and the interaction of the triplet excited states of 62+. Results and discussion Synthesis and characterization of mono- and dinuclear complexes The synthesis (Scheme 1) of the target mononuclear complexes was carried out following a previously described synthetic route starting from RuCl (tpy).40,583 In the first step, this pre- cursor is activated by chloride abstraction using silver tetra- fluoroborate. The resulting solvent complex intermediate was subsequently treated with the respective dipyridylbenzene ligand L1 or L2 40 to give the amide or ester substituted [Ru(dpb-R)(tpy)]+ complexes 1(PF6) and 3(PF6) in good yields. Cleavage of the functional groups for the liberation of free amine or carboxylic acid was achieved in aqueous methanolic solutions using sodium hydroxide as a base and hydrazine as a reductant to prevent oxidative decomposition. This hydro- lysis protocol gives comparable yields to the hydrolysis of structurally related ruthenium complexes by trimethylamine employed by Berlinguette and coworkers.61 In order to accomplish the coupling reaction between the free acid and the amine moieties of 2+ and 4+ to the dinuclear complex 62+, activation of the acid component is necessary. This was achieved similarly to a previously employed tech- nique using N-hydroxybenzotriazole (HOBt) and N,N′-diiso- Scheme 1 Synthesis of the mononuclear complexes 1(PF6)–5(PF6) and propylcarbodiimide (DIC).55,62 Compared to the amide the dinuclear complex 6(PF6)2 from RuCl3(tpy). Atom numbering for coupling reaction between bis(terpyridine)ruthenium(II) amino NMR assignment is included. acid derivatives described previously,54,55 the coupling had to be performed at elevated temperatures, possibly attributed to the reduced acidity of the amino functionality and relatively weak nucleophilicity of the OBt ester compared to other active appears at 4.24 ppm indicating the presence of a free amino esters. group. Similarly, ester saponification (3(PF6) → 4(PF6)) yields a All complexes were characterized using 1D- and 2D-NMR loss of the characteristic CH2 and CH3 proton resonances of techniques (ESI, Fig. S1–S14†) as well as ESI and high- the ethyl group while essentially leaving the aromatic region of resolution ESI mass spectrometry (ESI, Fig. S15†). The purity the 1H NMR spectrum unaffected. For the hydroxybenzotri- of all compounds under study was confirmed by elemental azole ester 5(PF6), the resonances of the dipyridylbenzene analyses. Successful amide cleavage (1(PF6) → 2(PF6)) is proven ligand, predominantly those in the 2B-position, are shifted to by the disappearance of the NH (8.62 ppm) and CH3 a lower field. This is in agreement with the formation of a (2.23 ppm) resonances in the proton NMR spectrum of 2(PF6). more electron-deficient species that is activated towards Simultaneously, a new significantly broadened resonance nucleophilic attack. 5642 | D alton Trans., 2016, 45, 5640–5658 This journal is © The Royal Society of Chemistry 2016 Open Access Article. Published on 29 February 2016. Downloaded on 20/05/2016 14:13:44. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Section 3.3 | 93 View Article Online Dalton Transactions Paper Interestingly, the functional group attached to the dpb ligand strongly affects the 13C chemical shift of the coordinat- ing carbon atom. While this resonance is found at 239.5 ppm in complex 5(PF6) with the strongly electron-withdrawing COOBt substituent, it is shifted upfield to 233 ppm in com- plexes 3(PF6) and 4(PF6) with COOEt and COOH functional groups. In the N-substituted complexes, it is found at even lower chemical shifts, namely at 217.2 ppm for 1(PF6) and at 208.9 ppm for 2(PF6). This also reflects the electrochemistry at the ruthenium center (vide infra). Evidence for the success of the amide coupling between 2(PF6) and 5(PF6) is gained from the 1H NMR spectrum of 6(PF6)2. The proton resonance at low field (9.63 ppm) with an integral of a single proton indicates the presence of an amide bridge. Additionally, all aromatic signal sets occur four times in a 1 : 1 : 1 : 1 ratio. Although the resonances of the two terpyr- idine ligands are distinguishable due to the different substitu- ents at the remote dpb ligands, an unambiguous assignment to one of the two capping ligands is impossible. The ESI mass spectrum, which shows the required peaks at m/z = 586.6 for 62+ and at 1318.3 for 6(PF +6) with isotope patterns characteri- stic for a complex containing two ruthenium atoms, gives additional support to the successful formation of the dinuclear complex. IR spectroscopy further confirms all structures under study Fig. 1 Cyclic voltammograms of 1(PF6)–4(PF6) and 6(PF6)2 in MeCN (ESI, Fig. S16†). All the complexes exhibit an intense IR absorp- with 0.1 mol l−1 [nBu4N][PF6] as the supporting electrolyte. tion at 843 cm−1 arising from P–F stretching vibrations within the PF −6 counterion. The amino-substituted complex 2 + shows a broad, intense absorption at 3420 cm−1 arising from N–H stretching vibrations of the NH2 group. The amide containing Density functional theory (DFT) calculations further illus- complexes 1+ and 62+ exhibit a broad absorption band at trate and enlighten these experimental findings. We have around 3220–3230 cm−1 ascribed to the N–H stretch along shown previously40 that the B3LYP functional63 along with a with intense CvO vibrations at 1650 cm−1 and 1635 cm−1, split-valence double-ξ basis set and polarization functions on respectively. Similar CvO vibrations are observed for the all non-hydrogen atoms (def2-SV(P))64–66 provides reasonable carboxy-substituted complexes 3+ and 4+, with that of the ester access to the electronic properties of the numerous charge occurring at 1695 cm−1 and that of the carboxylic acid at transfer states of the complexes under study when combined 1665 cm−1. Additionally, the carboxylic acid 4+ exhibits a broad with the ZORA relativistic approximation67 and a continuum absorption at 3440 cm−1 (O–H stretch) along with absorptions solvent model (COSMO).68 in the range between 3000 and 2300 cm−1 typical for carboxylic Indeed, DFT calculations for the singlet ground states of acids. the respective cationic complexes nicely reproduce the depen- dence of the energy of the HOMO from the substitution Electrochemical properties of complexes 1(PF6)–4(PF6) and pattern (Fig. 2). Additionally, the shape of the HOMO parallels 6(PF6)2 that of the doublet spin density of the RuIII complexes 12+–42+ The cyclic voltammograms of the complexes 1(PF6)–4(PF6) and (ESI, Table S2†) supporting the fact that oxidation occurs on 6(PF6)2 are depicted in Fig. 1 and the respective electrochemi- both the metal site and the dpb ligand. At substantially higher cal data are summarized in Table 1. For all mononuclear com- potentials, a second, irreversible oxidation is observed. It is plexes, 1(PF6)–4(PF6), a single reversible oxidation is observed again assigned to a mixed metal/dpb ligand oxidation yielding in the range between −0.2 V and 0.28 V versus the ferrocene/ a [Ru(dpb)]3+ state as suggested previously by DFT calculations ferrocenium redox couple. It is ascribed to the RuII/RuIII for analogous complexes.40 The dependence of this second oxi- couple. The electrochemical data of the ethyl ester-substituted dation from the substitution pattern is even more pronounced complex 3(PF6) agree well with those of the methyl ester so that its potential ranges from 0.35 V for amine-substituted reported in the literature.30 With increasing electron-accepting 2(PF6) to 1.49 V for ester-substituted 3(PF6). properties of the respective substituent, this redox process is All four mononuclear complexes exhibit one reversible and shifted to higher potentials by almost 500 mV. This suggests a several unresolved irreversible reductions. According to DFT strong contribution of the dpb ligand to the highest occupied calculations on the ground and the one-electron reduced molecular orbital (HOMO) of these complexes. states of 1+–4+ (10–40), the first reduction is ascribed to a This journal is © The Royal Society of Chemistry 2016 Dalton Trans., 2016, 45, 5640–5658 | 5643 Open Access Article. Published on 29 February 2016. Downloaded on 20/05/2016 14:13:44. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 94 | 3 RESULTS AND DISCUSSION View Article Online Paper Dalton Transactions Table 1 Electrochemical data of complexes 1(PF6)–4(PF6) and 6(PF6)2, obtained from 0.1 mol l −1 [nBu4N][PF6] containing acetonitrile solution. Potentials are referenced against the FcH/FcH+ couple. Energy differences EHOMO − ELUMO are obtained from DFT calculations (see the MO diagram in Fig. 2) Ered,2/ Eox,1 − ELUMO − Eox,1/V Eox,2/V Eox,3/V Ered,1/V V Ered,1/V EHOMO/eV 1(PF6) 0.06 ([Ru–dpb]/ 0.86 ([Ru–dpb] +/ — −1.93 (tpy/tpy−)a −2.54c 1.99 2.72 [Ru–dpb]+)a [Ru–dpb]2+)b 2(PF6) −0.20 ([Ru–dpb]/ 0.35 ([Ru–dpb]+/ — −1.95 (tpy/tpy−)a −2.48c 1.75 2.48 [Ru–dpb]+)a [Ru–dpb]2+)a 3(PF6) 0.28 ([Ru–dpb]/ 1.49 ([Ru–dpb] +/ — −1.87 (tpy/tpy−)a −2.40c 2.15 2.98 [Ru–dpb]+)a [Ru–dpb]2+)b 4(PF6) 0.28 ([Ru–dpb]/ 1.42 ([Ru–dpb] +/ — −1.86 (tpy/tpy−)a −2.50c 2.14 2.99 [Ru–dpb]+)a [Ru–dpb]2+)b 6(PF ) 0.05 ([Ru–Ru]/ 0.29 ([Ru–Ru]+/ 1.58 ([Ru–Ru]2+6 2 / −1.85 (2 e−, tpy/tpy−)a −2.51c 1.90 2.64 [Ru–Ru]+)a [Ru–Ru]2+)a [Ru–Ru]3+)b a Reversible, E1/2 given. b Irreversible, anodic peak potential given. c Irreversible, cathodic peak potential given. Fig. 2 Molecular orbital energy diagram of complexes 1(PF6)–4(PF6) and 6(PF6)2 obtained from DFT calculations (B3LYP, def2-SV(P), COSMO (acetonitrile), ZORA). Frontier orbitals are depicted exemplary for 3(PF6) since the shape of the respective orbital varies only marginally among the mononuclear complexes (see also ESI, Table S1†). Hydrogen atoms are omitted for clarity. tpy-centered reduction (ESI, Table S2†). The COOH-substituted to a minor inductive effect. Consequently, the first reduction complex 4(PF6) shows a stripping peak upon reoxidation fol- occurs at very similar potentials for all four complexes in the lowing the first reduction. We ascribe this phenomenon to pre- range between −1.86 (COOH-substituted 4(PF6)) and −1.95 V cipitation of the neutral complex moiety [RuII(dpb−-COOH) (NH2-substituted 2(PF6)) spanning just 90 mV. Accordingly, the (tpy−)] 40 on the electrode surface and subsequent redissolu- HOMO–LUMO gap, which is closely correlated to the differ- tion after reoxidation to 4+.55 ence of the redox potentials of the first reduction and oxi- Due to the orthogonal mer-coordination of the two triden- dation, varies considerably in the order 4+ ≈ 3+ > 1+ > 2+ tate ligands, the electronic influence of the different functional (Table 1). This trend is in excellent agreement with DFT calcu- groups attached to the dpb ligand on the tpy ligand is reduced lations (Fig. 2 and Table 1). 5644 | D alton Trans., 2016, 45, 5640–5658 This journal is © The Royal Society of Chemistry 2016 Open Access Article. Published on 29 February 2016. Downloaded on 20/05/2016 14:13:44. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Section 3.3 | 95 View Article Online Dalton Transactions Paper All complexes exhibit follow-up oxidation peaks in the range between −1 and −0.5 V once reduction has been carried out beyond −2 V. We had observed such behaviour previously both in mono- and dinuclear bis(terpyridine)ruthenium(II) complexes and cyclometalated ruthenium complexes bearing amide functionalities. We had suggested that these follow-up oxidations are associated with species that are formed after reduction of the substantially acidified amide NH proton (hydrogen formation).55 The observation of similar processes in complexes such as 3(PF6) and 4(PF6) lacking NH functional- ities contradicts this hypothesis. In fact, such more or less pro- nounced follow-up reoxidation peaks can be found for a large variety of tpy containing complexes of different metals such as chromium,69 manganese,70 and ruthenium,35 once a sufficient number of reduction events have taken place at the tpy unit. The triplet spin densities of the twofold reduced complexes 11−–41− do not provide further hints on possible follow-up reactions (ESI, Table S2†). Compared to the respective 1GS structures, their geometries are undistorted with a spin density homogeneously distributed over all three pyridine rings of the terpyridine ligand. For the dinuclear complex 6(PF6)2, cyclovoltammetric Fig. 3 UV-Vis absorption spectra of (a) 1(PF6)–4(PF6) and (b) 6(PF6)2 in−5 −1 studies reveal a single reversible two-electron reduction, as evi- dry acetonitrile solution at room temperature (c = 2 × 10 mol l ). denced from square-wave voltammetry, followed by an intense stripping peak. Again, this stripping peak arises from precipi- tation of the large uncharged complex 60 on the electrode UV-Vis spectroscopic properties of complexes 1(PF6)–4(PF6) surface and redissolution after reoxidation to 62+. The first, and 6(PF6)2 unsplit reduction processes are ascribed to tpy-centered The absorption spectra of all mononuclear complexes (Fig. 3) reductions occurring at both terminal ligands of the bimetallic exhibit very similar features. Besides intense transitions in the complex as evidenced from DFT calculated triplet spin UV region attributed to π–π* transitions within the ligands, densities of 60 (ESI, Table S3 and Fig. S17†). Additionally, four discernible absorption bands are observed in the visible two reversible oxidation processes at 0.05 and 0.29 V, respecti- range between 350 and 650 nm. DFT calculations30,39 and reso- vely, are observed. Based on the redox potentials of the mono- nance Raman spectroscopy studies40 suggest that such bands nuclear complexes, these can be ascribed to a primary characteristic for cyclometalated ruthenium complexes con- oxidation of the NH-substituted complex fragment followed taining polypyridine and N^C, N^C^N or N^N^C ligands arise by oxidation of the CO-substituted moiety. Interestingly, from metal-to-ligand charge transfer transitions (1MLCT) invol- the difference of the two oxidation potentials is slightly ving both the polypyridine and the cyclometalating ligand as increased by 20 mV compared to that of the mononuclear electron accepting sites. complexes 1+ and 3+ (240 versus 220 mV, Table 1). This might As the visible-range absorption bands are governed by be due to spatial charge accumulation or to a weak electronic 1MLCT transitions involving both ligands, variation of the communication between the two complex fragments in the functional group on the cyclometalating ligand greatly affects mixed-valent state 63+. Missing shifts of the electrochemical the position of the low-energy absorption maximum (Table 2). potentials of asymmetric dinuclear complexes compared to While the ester- or acid-substituted complexes 3+ and 4+ similar mononuclear complexes or negligible splittings exhibit absorption maxima at 493 nm, the respective between the RuIIRuII/RuIIRuIII and the RuIIRuIII/RuIIIRuIII oxi- maximum of amide-substituted 1+ is observed at 509 nm and dation potentials in symmetrical complexes have already been that of the amine complex 2+ is found at 550 nm (Fig. 3). This observed with other bimetallic bis(tridentate)ruthenium trend is in good agreement with the HOMO–LUMO gap complexes.53–55,57 Some of these were accompanied by a weak (Table 1 and Fig. 2) in this series of complexes. In contrast, electronic interaction between the Ru centers while others DFT calculations reveal that the most intense Ru → tpy MLCT showed no metal–metal interaction. These examples illustrate transitions (HOMO−1 (dxz) → LUMO) are not responsible for that a clear conclusion as to whether electronic communi- the observed trend since they appear at very similar energies cation occurs between the metal centers of the complex frag- for all four complexes (transition 5 in ESI, Tables S4–S7†). This ments is impossible purely based on these electrochemical is easily understood based on a closer examination of the orbi- data.71 UV-Vis spectroscopy studies on the mixed-valent tals of the complexes 1+–4+ involved in this transition (Fig. 2 species 63+ will provide deeper insight into that matter and ESI, Table S1†): the symmetry of the LUMO (tpy) only (vide infra). allows for constructive interference with the dxz orbital of the This journal is © The Royal Society of Chemistry 2016 Dalton Trans., 2016, 45, 5640–5658 | 5645 Open Access Article. Published on 29 February 2016. Downloaded on 20/05/2016 14:13:44. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 96 | 3 RESULTS AND DISCUSSION View Article Online Paper Dalton Transactions Table 2 Experimental UV-Vis absorption and emission data of the mononuclear complexes 1(PF6), 2(PF6), 3(PF6), and 4(PF6) as well as the dinuclear complex 6(PF6)2. Absorption and emission data are obtained from (deaerated) acetonitrile solution, and low-temperature emission data are recorded in butyronitrile. Excitation wavelengths are given in parentheses where wavelength dependence of the emission maximum was observed, otherwise λexc = 500 nm λem/nm λem/nm λem/nm ϕ at λmax/nm (ε/10 3 M−1 cm−1) at 298 K at 155 K at 77 K 298 K 1(PF6) 533 (11.9, sh), 509 (12.5), 419 (7.9), 373 (7.9), 315 (34.8) 800 798 736 8 × 10 −6 2(PF6) 550 (12.9), 519 (12.9, sh), 417 (9.1), 379 (9.4), 316 (35.8) 780 768 731 <2 × 10 −6 3(PF6) 529 (9.9, sh), 493 (12.3), 428 (7.6), 343 (13.5), 315 (35.3) 744 738 708 14 × 10 −6 4(PF6) 529 (10.2, sh), 493 (12.7), 429 (7.8), 343 (13.3), 315 (35.6) 744 738 709 15 × 10 −6 6(PF6)2 530 (23.4, sh), 504 (25.1), 422 (14.6), 356 (27.4), 315 (68.7) 756 (480), 772 (560) 746 736 9 × 10 −6 metal (HOMO−1). On the other hand, both, HOMO−1 and LUMO are perpendicular to the Ru dyz orbital and the dpb π-orbital which strongly contribute to the HOMO. Hence, the dpb functional group’s impact on the involved orbitals is again reduced to a minor inductive effect explaining the weak dependence of the Ru → tpy MLCT transitions on the dpb substituent. The strong bathochromic shift of the experimental absorp- tion maximum accompanying the more electron-donating N-acetyl amino and amino substituents at the dpb ligand in fact arises from symmetry-allowed dyz(Ru) → dpb MLCT tran- sitions. Especially the HOMO → LUMO+2 transition plays a key role within the absorption characteristics (transition 6 in Tables S4–S7†). These transitions are calculated at 486 (1+), Fig. 4 Visible range of the absorption spectra of 1+ (blue), 3+ (red) and + 2+ + +507 (2 ), and 456 nm (3+ and 4+), respectively, and they nicely 6 (green) in dry acetonitrile solution as well as superposition (1 + 3 ) reproduce the trends within the absorption maxima of the (black, dashed line) and difference spectra (3 + − 1+) (black, solid line). respective complexes (Table 2). This fully confirms that the two main 1MLCT transitions in the visible range of the electronic spectrum, namely dxz(Ru) → tpy and dyz(Ru) → dpb, are elec- To probe the metal–metal interaction in the mixed-valent tronically decoupled for simple symmetry reasons. state 63+, careful in situ chemical oxidation of 6(PF6)2 in aceto- At first sight, the absorption spectrum of the dinuclear nitrile solution was carried out using (NH4)2[Ce(NO3)6] as an complex 62+ resembles the absorption spectra of the carboxy- oxidant (E ≈ 0.8–0.9 V).72 Its oxidation potential is high substituted mononuclear complexes 3+ and 4+ with roughly enough to allow for a stepwise double oxidation of 62+ to the doubled extinction coefficients due to its dinuclear nature bis(ruthenium(III)) complex 64+. Absorption spectra (Fig. 5) (Fig. 3). A closer inspection reveals that the spectrum of the were recorded each time after addition of 0.25 equivalents of dinuclear complex is much better reproduced by a 1 : 1 super- the oxidant. A broad, symmetrical absorption band appears in position of the absorption spectra of the ester- and the amide- the near infrared (NIR) region of the absorption spectrum substituted mononuclear complexes 1+ and 3+ (Fig. 4). This upon addition of 0 → 1 equivalents of the oxidant with an suggests that the dinuclear compound 62+ consists of two absorption maximum at 1165 nm (8585 cm−1, ε = 2620 M−1max essentially non-interacting bis(tridentate)ruthenium(II) frag- cm−1, full width at half maximum ν̃ −11/2 = 6020 cm ). Simul- ments connected via an amide bond. Indeed, this is under- taneously, a second, significantly sharper band appears in the lined by time-dependent DFT calculations which reveal that all red region (maximum at 716 nm). A set of isosbestic points is charge transfer excitations >400 nm between the two complex observed for the oxidation of 62+ to 63+ at 233, 326, 335, 486, fragments have negligible oscillator strengths and should play and 619 nm indicating a clean reaction without side products. no role in the observed absorption features (Table S8†). Upon addition of more oxidant (1 → 1.5 eq.), a new set of iso- Similar observations have previously been made for other sbestic points is observed at 273, 325, 335, 638, and 810 nm. amide-linked dinuclear ruthenium(II) complexes with triden- Hence, the reaction 62+ → 63+ → 64+ occurs stepwise as tate ligands.53–55 Since in this study the visible absorption- expected from the separation of the first and second oxidation spectroscopic fingerprints of the two subunits are more dis- waves in the cyclic voltammogram of 62+. Simultaneously, the tinct than in the literature-known bimetallic examples, the intensity of the NIR band decreases while the band in the red superimposed nature of the absorption bands of 62+ is more region rises further. Interestingly, upon addition of more obvious. In principle, the two [Ru(dpb)(tpy)]+ subunits are oxidant (1.5 → 2 eq.), the isosbestic points are lost and a new essentially uncoupled in the RuIIRuII state. absorption band appears at around 940 nm (ESI, Fig. S18†). 5646 | D alton Trans., 2016, 45, 5640–5658 This journal is © The Royal Society of Chemistry 2016 Open Access Article. Published on 29 February 2016. Downloaded on 20/05/2016 14:13:44. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Section 3.3 | 97 View Article Online Dalton Transactions Paper ling parameter, the calculated Vab value of 580 cm −1 represents a lower limit. This coupling in 63+ is roughly half as strong as in the back-to-back linked symmetrical dinuclear complex [(ttpy)Ru(dpb-dpb)Ru(ttpy)]3+ (ttpy = 4′-tolylterpyridine).59 We attribute this weakening to the redox asymmetry introduced by the amide bridge in 63+. Simultaneously, the NHCO group increases the donor–acceptor distance and reduces the orbital overlap between the two complex moieties. Clearly, the mixed- valent complex 63+ has to be assigned to the Robin–Day class II with localized valencies and a moderate electronic coupling between the complex subunits.48 The activation barrier for thermal electron transfer can be calculated when the strength of electronic coupling and the energy difference ΔG0 of the two valence isomers are known.52 The latter can be estimated based on the difference in redox potentials of the RuII/RuIII couple of the two complex subunits. Since for 62+, this differ- ence is shifted towards larger values due to charge accumu- Fig. 5 UV-Vis-NIR absorption spectra of 62+ in acetonitrile solution lation and the resonance stabilization of the mixed-valent upon addition (a) of 0 → 1 equivalents of (NH 3+4)2[Ce(NO3)6] as an oxidant species 6 , we used the difference in RuII/RuIII redox poten- and (b) of 1 → 1.5 equivalents of (NH4)2[Ce(NO3)6] as an oxidant. Spectra tials of the mononuclear complexes 1+ and 3+ to estimate ΔG0 are recorded after addition of 0.25 equivalents each time. as 0.22 eV (1775 cm−1, 21 kJ mol−1).52 This yields an activation barrier of the electron transfer from [(tpy)RuIII(dpb-NHCO- dpb)RuII(tpy)]3+ to [(tpy)RuII(dpb-NHCO-dpb)RuIII(tpy)]3+ of We ascribe this to the decomposition of the highly charged 2190 cm−1 (26 kJ mol−1). complex 64+ on the timescale of the measurement (about The electronic coupling in 63+ is in contrast to the amide 45 minutes). bridged dinuclear ruthenium complex [(EtOOC-tpy)RuII(tpy- The fact, that the NIR band is only present in the mixed- NHCO-tpy)RuIII(tpy-NHCOMe)]5+.55 Based on a simple mole- valent state 63+, allows for the conclusion that it arises from an cular orbital consideration, the electronic coupling occurs via intervalence charge transfer (IVCT) process between the two a superexchange mechanism involving the bridge’s frontier metal centers RuII → RuIII. The absorption band in the red orbitals.77,78 In the bis(terpyridine)ruthenium system, these spectral region on the other hand is ascribed to ligand-to- are well separated in energy from the donor and acceptor orbi- metal (LMCT) transitions in the newly formed RuIII fragment tals at the metal centers. Thus, the tunnel barrier for electron (dpb → Ru). This is supported by TD-DFT calculations which transfer is much higher than in 63+ leading to no detectable predict such a symmetry allowed IVCT transition (dyz(Ru) + electronic interaction in the former. In contrast, the mediating dpb-CO → dyz(Ru) + dpb-NH) to occur at a wavelength of bridge orbitals of 63+ are already mixed into the ground state 1395 nm and LMCT excitations at around 630 nm for 63+ (ESI, donor and acceptor orbitals of the metal centers, significantly Table S9†). Additionally, the absorption spectra of the mono- increasing the electronic coupling in 63+.55 Obviously, cyclo- nuclear complexes 1+ and 3+ exhibit very similar LMCT bands metalating bridging ligands enable electronic communication in the range between 600 and 800 nm upon oxidation under in mixed-valent RuII/RuIII complexes.59,60,75,76 the same conditions (ESI, Fig. S19†). In particular, the excel- lent agreement between the LMCT maximum of 12+ (720 nm) and 63+ (716 nm) underlines that the first oxidation of 62+ Emission spectroscopy and triplet excited states of complexes occurs at the N-substituted [Ru(dpb)(tpy)]+ fragment. 1(PF6)–4(PF6) Although a straight-forward Hush analysis of the band All four mononuclear complexes 1(PF6)–4(PF6) exhibit very shape and energy of the IVCT band is formally not correct due weak room temperature emission in the red spectral range to the energy difference ΔG0 of the two valence isomers [(tpy) (Fig. 6 and Table 2). The carboxy-substituted complexes 3+ and RuII(dpb-NHCO-dpb)RuIII(tpy)]3+ and [(tpy)RuIII(dpb-NHCO- 4+ show the highest energy emission along with the highest dpb)RuII(tpy)]3+, the latter being the lower energy isomer, we phosphorescence quantum yield. Both are in excellent agree- analysed the IVCT band to obtain a rough estimate of the elec- ment with the values for the methyl ester complex reported by tronic coupling parameter Vab (ESI, Fig. S20†). 49,73 As the van Koten and coworkers.30 Interestingly, the phosphorescence donor–acceptor distance rMM, the Ru–Ru distance of 13.1 Å of these complexes is not quenched by oxygen present during (from DFT calculation) was taken into account despite the fact the measurement. This is attributed to very short excited state that the involved orbitals are substantially delocalized towards lifetimes in the picosecond range that are too short for bimole- the cyclometalated bridging ligand, thus rendering the cular quenching processes by triplet oxygen to occur. Indeed, e 59,60,74–76ffective charge transfer distance smaller. As using a attempts to measure the luminescence lifetimes by time- too large value for rMM will underestimate the electronic coup- correlated single photon counting failed underlining that the This journal is © The Royal Society of Chemistry 2016 Dalton Trans., 2016, 45, 5640–5658 | 5647 Open Access Article. Published on 29 February 2016. Downloaded on 20/05/2016 14:13:44. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 98 | 3 RESULTS AND DISCUSSION View Article Online Paper Dalton Transactions stood from a simple consideration of the HOMO–LUMO gap of the respective complexes (Table 1). In order to gain a deeper understanding of the excited state properties of the respective complexes, DFT calculations on the excited triplet states were performed. The symmetry allowed emission of (polypyridine)ruthenium(II) complexes arises from a low-energy 3MLCT state. It exhibits spin density both at the metal site and the π-accepting polypyridine ligand. In fact, in cyclometalated complexes of the type [Ru(dpb-R1) (tpy-R2)],+ the LUMO of the terpyridine is always involved in the 3MLCT emissive state as well.40 Consequently, geometry optimizations were performed on the triplet states of all com- plexes under study yielding the respective 3MLCT states (Fig. 7). Despite the fact that these states are distorted to some extent compared to the singlet ground states (1GS) (vide infra), it is obvious from the respective spin densities that the 3MLCT states are composed ofHOMO−1 (dxz(Ru)) as the electron donor and LUMO πtpy* as the electron acceptor. Similar to the previous discussion concerning the 1MLCT excitations (vide supra), this orbital parentage of the 3MLCT state results in rather similar 3MLCT-1GS energy gaps despite the strongly varying HOMO (dyz + πdpb)–LUMO gaps. Insight into excited state deactivation pathways can be gained from temperature dependent measurements of excited state lifetimes or quantum yields. Seminal work by T. J. Meyer and co-workers22 revealed a metal-centered 3MC state as a ther- mally accessible state in [Ru(bpy)3] 2+. This state depopulates 3 Fig. 6 Normalized steady-state emission spectra of 1(PF6)–4(PF ) (λ the emissive MLCT state and substantially shortens its life-6 exc = 500 nm) (a) at room temperature in degassed acetonitrile solution, (b) time at room temperature. In strongly push–pull substituted at 155 K in liquid butyronitrile solution and (c) at 77 K in a frozen butyro- cyclometalated complexes such as [Ru(dpb-NHCOMe)(tpy- nitrile matrix. COOEt)],+ a second pathway via a low-energy ligand-to-ligand (dpb → tpy) charge transfer (3LL′CT) state is accessible that prevents emission entirely.40 Temperature-dependent steady-state emission spectra were excited state lifetimes at room temperature are well below one recorded for complexes 1(PF6) and 3(PF6) in butyronitrile solu- nanosecond. tion in the temperature range between 298 K and 155 K The shape of the emission band of the two COOR-substituted (Fig. 8). Due to the low quantum yield of complex 2(PF6) and compounds 3+ and 4+ is very similar to that of many other the spectroscopic similarity of 3(PF6) and 4(PF6), 2(PF6) and (polypyridine)ruthenium(II) complexes with a vibronic pro- 4(PF6) were not considered in this variable temperature (VT) gression resulting in a typical low-energy shoulder.2,27,31,40 The emission study. Interestingly, the VT emission plots ln(ϕ) vs. emission band shape of the N-substituted complexes 1+ and 2+ T−1 obtained for complexes 1(PF6) and 3(PF6) differ qualitat- on the other hand is different. Spectral decomposition in sep- ively from those of [Ru(bpy)3](PF6)2 (ESI, Fig. S22 and S23†) arate Gaussian shaped bands (ESI, Fig. S21†) suggests that 0–1 and the structurally related complex [Ru(dpb-COOEt)(tpy- and especially 0–2 transitions dominate in these complexes at NHCOMe)](PF ).406 The shape of the curves clearly is not linear room temperature. The 0–0 transition, which typically is quite as has been found for [Ru(dpb-COOEt)(tpy-NHCOMe)]+.40 strong in other [Ru(dpb)(tpy)]+-complexes at room tempera- Meyer’s equation22,23 which assumes a single thermally acti- ture, apparently is of less relevance in complexes with dpb- vated deactivation pathway (3MC) for the emissive 3MLCT state NHR ligands (ESI, Fig. S21†). Consequently, in a solid butyro- fails to reproduce the shape of the VT emission plots of 1+ and nitrile matrix at 77 K, a more pronounced hypsochromic shift 3+ as well, while it perfectly fits the VT emission plot of is observed for complexes 1+ (1085 cm−1) and 2+ (860 cm−1) [Ru(bpy) 2+3] (ESI, Fig. S22 and S23†). than for 3+ and 4+ (580 cm−1). At 77 K in frozen butyronitrile A rational explanation for this behaviour was found upon solution, the carboxy-substituted complexes 3+ and 4+ emit at extended DFT examination of the triplet potential energy a wavelength of 708–709 nm, while the amido- and amino-sub- surface. Besides the emissive 3MLCT state, two additional low- stituted complexes 1+ and 2+ emit at 736 and 731 nm, respect- energy triplet states could be localized as local minima for all ively. The similarity in the emission energy of the latter two four complexes 1+–4+. These are assigned as 3MC states with a complexes is remarkable and not straight-forwardly under- spin density essentially found on the metal site and as 3LL′CT 5648 | D alton Trans., 2016, 45, 5640–5658 This journal is © The Royal Society of Chemistry 2016 Open Access Article. Published on 29 February 2016. Downloaded on 20/05/2016 14:13:44. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Section 3.3 | 99 View Article Online Dalton Transactions Paper Fig. 7 Jablonski diagram of the triplet states of complexes 1+–4+ including DFT spin density plots (B3LYP, def2-SV(P), COSMO (acetonitrile), ZORA; contour value: 0.01). 3MLCT energies are given as experimental 0–0 emission energies, 3LL’CT and 3MC energies are calculated based on DFT derived Gibbs free energies relative to the respective 3MLCT energy and given in kJ mol−1. Hydrogen atoms are omitted for clarity. states in which the tpy ligand can be formally regarded as rings of the dpb unit is shifted away from the metal center to singly reduced while the Ru–dpb moiety carries an electron some extent yielding a long Ru–N bond of 219 pm while the hole.40 This latter low-energy state was considered responsible trans Ru–N bond is shortened to 209 pm (from 212 pm in the for the lack of emission from the strongly donor–acceptor sub- 1GS of 1+ and 2+). A similar shift of the metal center towards stituted complex [Ru(dpb-NHCOMe)(tpy-COOEt)]+ because the the tpy ligand as observed for 3+ and 4+ is also found for 1+ orthogonality of the orbitals of the electron and hole prevents and 2+. This difference in the geometry of the 3MLCT states the radiative recombination from the 3LL′CT state.40 between the NHR- and COOR-substituted complexes might Remarkably, all triplet states exhibit characteristic distor- explain the dominance of 0–2 transitions in the emission tions compared to the geometry of the respective singlet spectra of 1+ and 2+ as it corresponds to a larger horizontal ground state structures with a strong resemblance between the offset on the 1GS-3MLCT reaction coordinate. COOR-substituted complex on one side and the NHR-substi- In the 3LL′CT states again a clear distinction is found tuted complex on the other (Fig. 7). In the 3MLCT states of between the geometries of complexes 1+ and 2+ on one side complexes 3+ and 4+, the arrangement of the ligand periphery and 3+ and 4+ on the other. The 3LL′CT geometries of com- is essentially unaltered while the metal center is shifted plexes 1+ and 2+ appear essentially undistorted compared to towards the tpy ligand. The Ru–Ntpy central bond length is the 1GS structures with a slight elongation of the central Ru– shortened by 2 pm (1GS: 204 pm, 3MLCT: 206 pm for 3+ and Ntpy bond by about 4 pm.40 A similar shift is observed in the 4+) while the Ru–Cdpb bond is elongated by 4 pm in both cases 3LL′CT states of complexes 3+ and 4+. Yet, in their 3LL′CT geo- (1GS: 195 pm, 3MLCT: 199 pm for 3+ and 4+). This is in agree- metries, the tpy ligand is twisted by about 8° out of the plane ment with an increased coulombic interaction between the for- perpendicular to the dpb ligand. mally oxidized Ru and reduced tpy ligands upon population of The 3MC states of all four complexes appear structurally the 3MLCT state and has been described before for other poly- similar with immensely elongated bond lengths between Ru pyridine ruthenium complexes.35,40 All Ru–N bonds involving and the tpy nitrogen atoms (central Ru–Ntpy bonds: 225–227 the four peripheral pyridines are nearly unaffected with pm, peripheral Ru–Ntpy bonds: 235–237 pm). This distortion is similar bond lengths between 210 pm and 212 pm in all cases. accompanied by a tilt of the peripheral pyridine rings com- This is in stark contrast to the geometry of the 3MLCT state of pared to the central one within the tpy unit by 9–11°. The dpb both 1+ and 2+. Here, the ligand periphery is substantially dis- ligand on the other side is undistorted with typical Ru–dpb torted compared to the 1GS geometry: the central pyridine ring bond lengths (central Ru–Cdpb bonds: 195–196 pm, peripheral of the tpy unit is offset from the plane perpendicular to the Ru–Ndpb bonds: 215–217 pm). dpb ligand with a central Ntpy–Ru–Cdpb bond angle of just Based on DFT calculated Gibbs free energies, the 3LL′CT 167°. At the same time, one of the two peripheral pyridine state is the triplet ground state of the NHR-substituted com- This journal is © The Royal Society of Chemistry 2016 Dalton Trans., 2016, 45, 5640–5658 | 5649 Open Access Article. Published on 29 February 2016. Downloaded on 20/05/2016 14:13:44. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 100 | 3 RESULTS AND DISCUSSION View Article Online Paper Dalton Transactions non-emissive state at low energy gives rise to an additional excited state deactivation pathway.40 This is important for the interpretation of the temperature dependence of the emission spectra of 1(PF6) and 3(PF6) (Fig. 8). In fact, a second exponen- tial term needs to be taken into account, compared to Meyer’s original equation which accounts for a single depopulating state.22,23 Including a second state yields the following equation (for derivation see the ESI†): lnðϕÞ ¼ lnðkrÞ  ln½kr þ knr þ k1 expðΔE1=RTÞ þ k2 expðΔE2=RTÞ: The rate constants kr and knr describe the radiative and non-radiative decays (3MLCT → 1GS), ΔE1 corresponds to the activation barrier for surface crossing from the 3MLCT to the 3MC state (ΔE = ΔG‡1 1) and k1 is the rate constant for this surface crossing at infinite temperature as shown by Meyer.22 An analogous equation was previously used by Balzani and coworkers to describe the photodynamics of complexes of the [Ru(bpy) ]2+3 family. 79,80 In these cases, ΔE1 corresponds to the barrier for the thermally activated 3MLCT → 3MC surface crossing while ΔE2 (typically <1 kJ mol−1) is interpreted as the energy separation between multiple close-lying 3MLCT states split by spin–orbit coupling.80–82 In the present study, ΔE2 can be interpreted either as the energy difference ΔG0 of the 3MLCT and 3LL′CT states in thermal equilibrium or the activation barrier ΔG‡2 for the surface crossing from the 3MLCT to the 3LL′CT state (see the ESI† for in-depth elaboration). This depends on the relative rate constants for the reverse internal conversion 3LL′CT → 3MLCT and the non-radiative intersystem crossing (ISC) to the Fig. 8 Variable-temperature emission plots of ln(ϕ) vs. T−1 for com- ground state (3LL′CT → 1GS). Upon cooling of solutions of all plexes (a) 1(PF6) and (b) 3(PF6) including fit curves (dashed lines; see the four mononuclear complexes, even complexes 1+ and 2+, in text for fit function and parameters). The insets show emission spectra 3 in the range between 298 K and 155 K. which the emissive MLCT state is not the triplet ground state, the emission intensity increases. This corroborates that 3MLCT and 3LL′CT cannot be in thermal equilibrium at least in complexes 1+ and 2+. For complexes 3+ and 4+, this con- plexes 1+ and 2+ (Fig. 7) followed by the 3MLCT and 3MC clusion cannot be drawn purely based on the temperature states. This order is identical to that of the strongly donor– dependence of the emission quantum yield, since both, the acceptor substituted complex [Ru(dpb-NHCOMe)(tpy- energy difference of the 3MLCT and 3LL′CT states and the acti- COOEt)]+.40 In complexes 3+ and 4+, in which the donor vation barrier ΔG‡2, are positive. Based on the DFT calculated strength of the dpb ligand is weakened by the COOR substitu- energies of the activation barriers for the 3MLCT–3LL′CT ents, the order of 3MLCT and 3LL′CT is inverted. Increasing surface crossing and the experimentally determined ΔE2 the push–pull substitution of a given heteroleptic [Ru(dpb) values (vide infra), however, it is plausible, that also for com- (tpy)]+ complex will lower the energy of a donor–acceptor plexes 3+ and 4+, the surface crossing into the 3LL′CT state is charge-separated state, here the 3LL′CT state, relative to the irreversible and followed by rapid relaxation into the singlet other excited states. Remarkably, the trends of the geometrical ground state. Consequently, ΔE2 is identified in analogy to features of the various states can be related to their relative ΔE1 = ΔG‡1 as the activation barrier ΔG‡2 for the thermal energies. While for 1+ and 2+, the 3LL′CT state is the least dis- depopulation of the 3MLCT via the 3LL′CT states. torted compared to the 1GS geometries, for 3+ and 4+ this is Based on the very similar 0–0 emission energies which true for the 3MLCT state instead. should give similar rate constants for the non-radiative decay Considering the relative energies of the various triplet (3MLCT → 1GS), the large differences in the phosphorescence states, it is apparent that the emissive 3MLCT state is flanked quantum yields of the four complexes 1+–4+ are quite un- by two non-emissive states (3MC and 3LL′CT) for all four com- expected (Table 2).83–85 Yet, combining the ln(ϕ) vs. T−1 plots plexes. Both are thermally accessible, instead of just a single with the relative energies of the involved states as determined state (3MC) as in [Ru(bpy)3](PF6)2. The presence of a second by DFT provides an explanation. For complexes 3+ and 4+ the 5650 | D alton Trans., 2016, 45, 5640–5658 This journal is © The Royal Society of Chemistry 2016 Open Access Article. Published on 29 February 2016. Downloaded on 20/05/2016 14:13:44. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Section 3.3 | 101 View Article Online Dalton Transactions Paper relatively high quantum yield is associated with the emissive 3MLCT state being the triplet ground state. For 1+ and 2+ on the other hand, the non-emissive 3LL′CT state becomes the triplet ground state giving rise to a deactivation pathway with a potentially very low activation barrier ΔG‡1. Due to the small experimentally accessible temperature range, the fit using the biexponential equation given above is overparametrized. Consequently, quantitative results have to be considered very carefully. For 1+, activation barriers of ΔG‡1 = 21.7 and ΔG‡2 = 2.1 kJ mol−1 are obtained from the fit, while for 3+, the acti- vation barriers are ΔG‡1 = 23.1 and ΔG‡2 = 6.2 kJ mol−1. Based on the calculated energies of the various triplet states (Fig. 7), it is reasonable to assume that the higher activation barriers ΔG‡1 of >20 kJ mol−1 are associated with the deactivation via the 3MC state (vide infra). ΔG‡2 is very similar for both the NHR- and COOR-substituted complex types, corroborating that the substitution pattern at the cyclometalating ligand only has a marginal effect on the ligand field splitting in the [Ru(dpb- R)(tpy)]+ type of complexes. The second activation barrier ΔG‡2 of 1+ is only one third of that of 3+. Hence, thermal de- activation via 3LL′CT states is significantly accelerated by the presence of an electron donating substituent at the dpb ligand explaining the substantially lower quantum yield of the former. Substituents at the tpy ligand on the other hand are Fig. 9 Profile of the triplet hypersurface of (a) 1+ and (b) 3+ obtained expected to influence both activation barriers but especially from DFT calculations (B3LYP, def2-SV(P), ZORA, COSMO (acetonitrile)). ΔG‡1 between the 3MLCT and 3MC states since the Gibbs free energies are given in kJ mol−1 relative to the emissive 3MLCT substituents at the tpy ligand significantly impact the 3MLCT state (G −1 MLCT = 0 kJ mol ). Spin densities of the transition states (TS) are energy. given at a contour value of 0.01. Hydrogen atoms are omitted for clarity. To gain a better understanding of the excited state pro- cesses, we performed DFT based geometry optimizations to find the transition states connecting the 3MLCT and the 3MC description of the excited state deactivation processes can be states on one side and the 3MLCT and 3LL CT states on the obtained for these cyclometalated complexes.′ other. All four transition states could be localized successfully In summary, dpb ligands in bis(tridentate)ruthenium(II)3 and their nature confirmed by the presence of a single negative complexes indeed induce high-energy MC states, but give rise vibrational frequency representing the reaction coordinate of to low-energy 3LL′CT states. As for both states, emission is the respective transition (Fig. 9). Subsequent spin density cal- symmetry-forbidden, both contribute to the rapid excited state culations further confirmed the nature of the localized states deactivation observed for these types of complexes. The combi- as the desired transition states. For both complexes, 1+ and 3+, nation of two [Ru(dpb)(tpy)] + emitters is discussed in the next the spin density of the 3MLCT–3LL′CT transition state shows chapter. contributions of both ligands and, predominantly, the metal center. Remarkably, the spin carrying orbital at the metal Emission spectroscopy and triplet excited states of complex center neither corresponds to the dxy orbital as in the 3MLCT 6(PF6)2 state nor to the dyz orbital as in the 3LL′CT state but is a linear For the dinuclear complex 62+, a broadened emission spectrum combination of both. This further underlines the transition is obtained at room temperature compared to the formally state character of the localized state. Similarly, the constituting mononuclear complexes 1+ and 3+ (Fig. 10). 3MLCT–3MC transition states of 1+ and 3+ show a substantial Additionally, the position of the emission maximum is depen- amount of spin density at the metal center (1.46 electrons dent on the excitation wavelengths and shifts from 756 nm based on Mulliken’s spin population analysis). But instead of upon excitation at 480 nm to 772 nm when being irradiated at the nitrogen lone pairs, a tpy π*-orbital (LUMO of 1+ and 3+) 560 nm (Fig. 10 and Table 2). Measurement at 155 K in butyro- contributes to this transition state. The DFT calculated tran- nitrile yields substantially sharpened emission spectra with an sition state energies ΔG‡ ‡ ‡1(DFT) and ΔG2(DFT) and G1(exp.) and emission maximum at 746 nm and a pronounced shoulder at ΔG‡2(exp.) extracted from the fits of the ln(ϕ) vs. T−1 plots 800 nm. The intensity of this shoulder increases upon increas- show remarkable agreement with deviation as small as ing the excitation wavelength from 480 to 560 nm. The blue- ±2 kJ mol−1. This suggests that despite the narrow temperature shift of the emission maximum (180 cm−1) upon freezing the range of the VT measurement and their very low quantum butyronitrile solution of 62+ is much smaller than that of all yields and short excited state lifetimes, a very reasonable mononuclear complexes under study. This journal is © The Royal Society of Chemistry 2016 Dalton Trans., 2016, 45, 5640–5658 | 5651 Open Access Article. Published on 29 February 2016. Downloaded on 20/05/2016 14:13:44. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 102 | 3 RESULTS AND DISCUSSION View Article Online Paper Dalton Transactions electron transfer from RuII to RuIII and from tpy− to tpy (Dexter energy transfer), these states could interconvert.90,91 Apparently, due to the large distance between the two [Ru(tpy)] moieties (rRuRu = 13 Å, rtpy,tpy ≈ 20 Å), Dexter energy transfer, whose rate constant decays exponentially with distance, is rather slow between the complex subunits. All other radiative and non-radiative relaxation pathways of triplet 62+ are extre- mely fast (below 1 ns as evidenced from time-resolved emis- sion spectroscopy). Consequently, in fluid solution, emission occurs faster than thermal equilibration between the two emis- sive 3MLCT states. If equilibration was faster than emission, the 3[(tpy−)RuIII(dpb-NHCO-dpb)RuII(tpy)]2+ state would be favoured over 3[(tpy)RuII(dpb-NHCO-dpb)RuIII(tpy−)]2+ thermo- dynamically and would yield single emission at around 800 nm, but this is not observed. Given that the two 3MLCT states are not in thermal equili- brium, it should be possible to selectively populate one or the other excited state by irradiation into one of the two complex subunits. Since in the absorption spectrum of 62+ the absorp- tion bands of the two fragments [(tpy)Ru(dpb-NHR)] and [(ROC-dpb)Ru(tpy)] overlap substantially, it is not possible to excite them with 100% selectivity (Fig. 4). But by changing the excitation wavelength it is possible to gradually tune the ratio at which the two building blocks are excited. The difference spectrum of the two mononuclear complexes 1+ and 3+ carry- ing similar functional groups as the two subunits of 62+ (Fig. 4) gives an idea where a maximum difference of absorp- Fig. 10 Normalized steady-state emission spectra of 6(PF ) at varying tion can be expected between the NH- and CO-substituted6 2 excitation wavelengths (a) at room temperature in degassed acetonitrile [Ru(dpb)(tpy)]+ complex subunits. This difference spectrum solution, (b) at 155 K in liquid butyronitrile solution and (c) at 77 K in a reveals a maximized and minimized absorption of the COR- frozen butyronitrile matrix. substituted complex at around 480 and 560 nm, respectively. This is in excellent agreement with the above mentioned minimum and maximum of the shoulder at 800 nm in the A wavelength dependence of the emission maximum from emission spectrum of 62+ at 155 K. Additionally, the difference the excitation energy is very atypical for polypyridineruthe- spectrum of the emission spectra recorded at 155 K with λexc = nium(II) complexes. We ascribe this behaviour to two indepen- 480 and 560 nm reveals a band with a maximum at 800 nm dent emission processes in solution involving the two complex (ESI, Fig. S25†) that resembles the emission band of 1+ at that subunits of 62+. In fact, similar dual emission processes have temperature (Fig. 6b). These observations strongly support that previously been invoked to explain the emission wavelength dual emission occurs from two uncoupled 3MLCT excited dependence from the excitation energy.86–89 Unfortunately, states of the dinuclear complex 62+ in solution. attempts to measure the excited state lifetimes by time- The origin of this dual emission process is markedly correlated single photon counting failed in this case due to different than that observed for the structurally similar amide- the very rapid excited state decay of 62+. Hence, no evidence for bridged dinuclear complex [(EtOOC-tpy)Ru(tpy-NHCO-tpy) a biexponential character of the excited state decay, which Ru(tpy-NHCOMe)]4+.55 In the latter, the involved emissive would support the presence of a dual emission mechanism, states are sufficiently long-lived and at a significantly shorter could be obtained. distance to allow for thermal equilibration prior to emission. However, a reasonable explanation for the dual emission of Since the emissive states [(EtOOC-tpy)RuII(tpy-NHCO-tpy−) 62+ can be given based on its absorption characteristics. As RuIII(tpy-NHCOMe)]4+ and [(EtOOC-tpy)RuIII(tpy-NHCO-tpy−) shown above, the visible range of the absorption spectrum of RuII(tpy-NHCOMe)]4+ are very similar in energy, occupation 62+ is composed of 1MLCT excitations localized on one of the between the two is Boltzmann distributed leading to dual two complex halves. Upon intersystem crossing and vibrational emission at room temperature. relaxation, dxz(Ru) → tpy 3MLCT states are populated. Two Interestingly, upon freezing of the butyronitrile solution of such triplet excited states are conceivable, namely 3[(tpy−) 62+, single emission is observed arising from the NHR-substi- RuIII(dpb-NHCO-dpb)RuII(tpy)]2+ and 3[(tpy)RuII(dpb-NHCO- tuted subunit as judged from the position of the emission dpb)RuIII(tpy−)]2+, with the triplet spin density localized on maximum as well as the independence of the emission band opposing [Ru(tpy)] fragments (ESI, Fig. S24†). Via double shape from the excitation wavelength (Fig. 10c). This loss of 5652 | D alton Trans., 2016, 45, 5640–5658 This journal is © The Royal Society of Chemistry 2016 Open Access Article. Published on 29 February 2016. Downloaded on 20/05/2016 14:13:44. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Section 3.3 | 103 View Article Online Dalton Transactions Paper dual emission can be traced back to the change in the rate couple (0.40 V vs. SCE, E1/2 = 0.09 ± 5 mV under the given constants involved with the excited state decay. Upon freezing conditions).72 the solvent matrix around a given luminescent dye, both non- radiative vibrational relaxation and emissive decay are slowed Density functional theory calculations down substantially. This is because they are typically DFT calculations were carried out using the ORCA program accompanied by geometrical rearrangements of the dye and package (version 3.0.2).97 Tight convergence criteria were the environment and such rearrangements are much more chosen for all calculations (keywords TightSCF and TightOpt, difficult in a rigid solvent cage. The rate for intramolecular convergence criteria for the SCF part: energy change 1.0 × 10−8 Dexter energy transfer on the other hand is not significantly Eh, 1-El. energy change 1.0 × 10 −5 Eh, orbital gradient 1.0 × diminished upon freezing of the solvent.92,93 Consequently, in 10−5, orbital rotation angle 1.0 × 10−5, DIIS error 5.0 × 10−7; for frozen solution, the two 3MLCT states of 62+ equilibrate ther- geometry optimizations: energy change: 1.0 × 10−6 Eh, max. mally prior to emission from the lower-energy 3[(tpy−) gradient 1.0 × 10−4 Eh per bohr, RMS gradient 3.0 × 10 −5 Eh RuIII(dpb-NHCO-dpb)RuII(tpy)]2+ state following Kasha’s rule.94 per bohr, max. displacement 1.0 × 10−3 bohr, RMS displace- ment 6.0 × 10−4 bohr). All calculations employ the resolution of identity (Split-RI-J) approach for the coulomb term in com- Experimental bination with the chain-of-spheres approximation for the exchange term (COSX).98,99 All calculations were performed General procedures using the hybrid functional B3LYP63 in combination with Ahl- Chemicals were obtained from commercial suppliers and used richs’ split-valence double-ξ basis set def2-SV(P) which com- without further purification. Air- or moisture-sensitive reac- prises polarization functions for all non-hydrogen atoms.64,65 tions were performed in dried glassware under an inert gas Relativistic effects were calculated at the zeroth order regular atmosphere (argon, quality 4.6). Acetonitrile was refluxed over approximation (ZORA) niveau.67 The ZORA keyword automati- CaH2 and distilled under argon prior to use. The ligands cally invokes relativistically adjusted basis sets. 100 To account N-acetyl-3,5-dipyrid-2′-ylaniline L1 40 and ethyl 3,5-dipyrid-2′- for solvent effects, a conductor-like screening model (COSMO) ylbenzoate L2 40 as well as RuCl 953(tpy) were synthesized fol- modelling acetonitrile was used in all calculations. 68 TD-DFT lowing the literature-known procedures. Infrared spectra were calculations with at least 50 vertical transitions were carried recorded on a Varian Excalibur Series 3100 FT-IR spectrometer out based on the def2-SV(P) optimized geometry of the respect- using KBr disks. IR absorption band intensities are classified ive complex. Explicit counterions and/or solvent molecules as s (strong), m (medium) and w (weak). UV/Vis spectra were were not taken into account in all cases. To reduce the compu- recorded on a Varian Cary 5000 spectrometer in 1 cm cuvettes. tational cost, methyl instead of ethyl groups at the ester moiety Emission spectra were recorded on a Varian Cary Eclipse were used throughout all calculations. spectrometer. Quantum yields were determined by comparing Synthesis of [Ru(dpb-NHCOCH3)(tpy)](PF6) 1(PF6). the areas under the emission spectra on an energy scale RuCl3(tpy) (250 mg, 0.567 mmol, 1 eq.) and AgBF4 (320 mg, recorded for solutions of the samples and a reference with 1.64 mmol, 2.9 eq.) were dissolved in dry acetone (20 ml) and matching absorbances (ϕ([Ru(bipy)3]Cl2) = 0.094 in deaerated heated to reflux for 2 h in the dark. The mixture was left to MeCN).21 Experimental uncertainty is estimated to be 15%. stand for 1 h and filtered through a syringe filter before remov- Low temperature emission spectra were recorded using an ing the solvent under reduced pressure. The dark residue was Oxford Instruments Optistat DN cryostat with cooling by liquid dissolved in nBuOH (20 ml) and N-acetyl-3,5-dipyrid-2′-ylani- N . ESI+2 and high resolution ESI + mass spectra were recorded line L1 (197 mg, 0.680 mmol, 1.2 eq.) was added. The resulting on a Micromass QTof Ultima API mass spectrometer with mixture was heated to reflux for 16 h to give a dark purple analyte solutions in acetonitrile. Elemental analyses were per- solution. After the removal of the solvent under reduced formed in the microanalytical laboratory of the Chemical Insti- pressure, the residue was dissolved in a minimal amount of tutes of the University of Mainz. NMR spectra were obtained acetonitrile (5 ml). Upon addition of a solution of NH4PF6 with a Bruker Avance II 400 spectrometer at 400.31 (1H) and (220 mg, 1.35 mmol, 2.4 eq.) in water (1 ml) followed by slow 100.66 (13C) at 25 °C. Chemical shifts δ [ppm] are reported addition of more water (80 ml), a black solid precipitated with respect to residual solvent signals as internal standards which was filtered off. Column chromatography on silica gel (1H, 13C): CD3CN δ( 1H) = 1.94 ppm, δ(13C) = 1.32 and (CHCl3 : MeOH = 7 : 1) afforded [Ru(dpb-NHCOCH3)(tpy)](PF6) 118.26 ppm.96 Electrochemical experiments were performed 1(PF6) as a dark purple solid. Yield: 175 mg (0.228 mmol, with a BioLogic SP-50 voltammetric analyzer at a sample con- 40%). Anal. Calc. for C33H25F6N6OPRu (767.6)·H2O: C, 50.45; centration of 10−3 mol l−1 using platinum wire as working and H, 3.46; N, 10.70. Found: C, 50.62; H, 3.31; N, 10.46. MS(ESI+): counter electrodes and a 0.01 mol l−1 Ag/AgNO3 reference elec- m/z (%) = 623.1 (100) [M − PF ]+6 . HR-MS(ESI+, m/z): Calcd for trode. Measurements were carried out at a scan rate of 100 mV C33H25N6ORu [M − PF +6] : 617.1166; Found: 617.1177. 1H NMR s−1 for cyclic voltammetry experiments and at 10 mV s−1 for (CD3CN): δ [ppm] = 8.73 (d, 3JHH = 8 Hz, 2H, H 2A), 8.62 (s, 1H, square-wave voltammetry experiments using 0.1 mol l−1 NH), 8.44–8.35 (m, 4H, H2B, H5A), 8.24 (t, 3JHH = 8 Hz, 1H, [nBu N][PF ] as the supporting electrolyte in acetonitrile. H1A), 8.05 (d, 3J = 8 Hz, 2H, H5B), 7.67 (dt, 34 6 HH JHH = 8 Hz, Potentials are given relative to the ferrocene/ferrocenium 4JHH = 1 Hz, 2H, H 6A), 7.59 (dt, 3J 4HH = 8 Hz, JHH = 1 Hz, 2H, This journal is © The Royal Society of Chemistry 2016 Dalton Trans., 2016, 45, 5640–5658 | 5653 Open Access Article. Published on 29 February 2016. Downloaded on 20/05/2016 14:13:44. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 104 | 3 RESULTS AND DISCUSSION View Article Online Paper Dalton Transactions H6B), 7.11 (d, 3J = 5 Hz, 2H, H8AHH ), 7.01 (d, 3J = 5 Hz, 2H, (100) [M − PF ]+. HR-MS(ESI+HH 6 , m/z): Calcd for C34H26N5O2Ru H8B), 6.94 (m, 2H, H7A), 6.64 (m, 2H, H7B), 2.23 (s, 3H, H11). [M − PF ]+6 : 632.1162; Found: 632.1173. 1H NMR (CD3CN): 13C{1H} NMR (CD3CN): δ [ppm] = 217.2 (C 9B), 169.6 (C10), δ [ppm] = 8.85 (s, 2H, H1B), 8.74 (d, 2H, 3JHH = 8 Hz, H 2A), 8.42 169.4 (C4B), 160.1 (C4A), 155.3 (C8A), 154.0 (C3A), 152.9 (C8B), (d, 2H, 3JHH = 8 Hz, H 5A), 8.30 (t, 1H, 3J 1AHH = 8 Hz, H ), 8.27 (d, 142.5 (C3B), 136.4 (C6B), 135.9 (C6A), 133.6 (C1B), 132.7 (C1A), 2H, 3J = 8 Hz, H5BHH ), 7.74–7.56 (m, 4H, H 6A, H6B), 7.13–7.04 127.2 (C7A), 124.4 (C5A), 123.2 (C2A), 122.5 (C7B), 120.6 (C5B), (m, 4H, H8A, H8B), 6.91 (t, 2H, 3JHH = 7 Hz, H 7A), 6.72 (t, 2H, 117.8 (C2A), 24.3 (C11). IR (KBr disk): ν̃ [cm−1] = 3230 (m, N–H 3JHH = 7 Hz, H 7B), 4.52 (q, 2H, 3JHH = 7 Hz, H 11), 1.52 (t, 3H, amide), 1650 (s, CvO amide), 1600 (m, CvC), 1520 (w, amide 3J = 7 Hz, H12). 13HH C{ 1H} NMR (CD3CN): δ [ppm] = II), 843 (s, P–F). 232.8 (C9B), 168.9 (C4B), 168.6 (C10B), 159.8 (C4A), 155.4 (C8A), Synthesis of [Ru(dpb-NH )(tpy)](PF ) 2(PF ). [Ru(dpb- 153.5 (C3A), 152.8 (C8B), 143.1 (C3A), 136.7 (C6B2 6 6 ), 136.4 (C 6A), NHCOCH3)(tpy)](PF6) 1(PF6) (113 mg, 0.147 mmol) was added 133.8 (C 1A), 127.3 (C7A), 124.6 (C2B), 124.5 (C6A), 123.3 (C2A), to a mixture of water (20 ml), methanol (20 ml), hydrazine 123.0 (C7B), 120.9 (C1B), 61.5 (C11), 14.9 (C12). IR (KBr disk): monohydrate (1 ml) and sodium hydroxide (1 g) and heated to ν̃ [cm−1] = 1695 (s, CvO ester), 1600, 1582 (m, CvC), 843 reflux for 16 h. After removal of the solvent under reduced (s, P–F). pressure, the dark residue was dissolved in a minimal amount Synthesis of [Ru(dpb-COOH)(tpy)](PF6) 4(PF6). [Ru(dpb- of acetonitrile (5 ml) followed by addition of a solution of COOC2H5)(tpy)](PF6) 3(PF6) (154 mg, 0.197 mmol) was added NH4PF6 (153 mg, 0.939 mmol, 6.75 eq.) in water (80 ml). The to a mixture of water (20 ml), methanol (20 ml), hydrazine precipitate was filtered off and washed with water (2 × 5 ml) monohydrate (1 ml) and sodium hydroxide (1 g) and heated to and diethyl ether (2 × 15 ml) giving [Ru(dpb-NH2)(tpy)](PF6) reflux for 16 h. After removal of the solvent under reduced 2(PF6) as a purple solid. Yield: 92 mg (0.127 mmol, 86%). Anal. pressure, the dark residue was dissolved in a minimal amount Calc. for C31H23F6N6PRu (725.6)·0.5H2O: C, 50.69; H, 3.29; N, of acetonitrile (5 ml) followed by slow addition of 1 mol per l 11.58. Found: C, 50.82; H, 3.05; N, 11.34. MS(ESI+): m/z (%) = aqueous H2SO4 to adjust the pH to 1. Upon addition of a solu- 581.1 (100) [M − PF +6] . HR-MS(ESI+, m/z): Calcd for tion of NH4PF6 (145 mg, 0.890 mmol, 4.5 eq.) in water (40 ml) C +31H23N6Ru [M − PF6] : 575.1060; Found: 575.1071. 1H NMR the product precipitated. The complex was filtered off and (CD CN): δ [ppm] = 8.72 (d, 3J = 8 Hz, 2H, H2A3 HH ), 8.40 (d, 3JHH washed with water (2 × 5 ml) and diethyl ether (2 × 15 ml). = 8 Hz, 2H, H5A), 8.20 (t, 3JHH = 8 Hz, 1H, H 1A), 8.00 (d, 3JHH = Column chromatography on silica gel (CHCl3 : MeOH = 5 : 1) 8 Hz, 2H, H5B), 7.74 (s, 2H, H2B), 7.67 (t, 3JHH = 8 Hz, 2H, H 6A), afforded [Ru(dpb-COOH)(tpy)](PF6) 4(PF6) as a dark red 7.56 (t, 3J 6B 3 8AHH = 8 Hz, 2H, H ), 7.18 (d, JHH = 5 Hz, 2H, H ), solid. Yield: 82 mg (0.109 mmol, 55%). Anal. Calc. for 6.97 (dd, 3JHH = 5 Hz, 8 Hz, 2H, H 7A), 6.93 (d, 3JHH = 5 Hz, 2H, C32H22F6N5O2PRu (754.6): C, 50.93; H, 2.94; N, 9.28. Found: C, H8B), 6.58 (dd, 3JHH = 5 Hz, 8 Hz, 2H, H 7B), 4.24 (s, 2H, NH2). 50.64; H, 2.51; N, 9.42. MS(ESI +): m/z (%) = 610.1 (100) 13C{1H} NMR (CD3CN): δ [ppm] = 208.9 (C 9B), 169.9 (C94B), [M − PF +6] . HR-MS(ESI+, m/z): Calcd for C32H22N5O2Ru 160.3 (C4A), 155.2 (C8A), 154.3 (C3A), 152.8 (C8B), 143.6 (C1B), [M − PF6]+: 604.0849; Found: 604.0873. 1H NMR (CD3CN): 142.6 (C3B), 136.2 (C6B), 135.5 (C6A), 132.0 (C1A), 127.2 (C7A), δ [ppm] = 8.85 (s, 2H, H2B), 8.74 (d, 3J = 8 Hz, 2H, H2AHH ), 8.42 124.2 (C5A), 123.2 (C2A), 122.0 (C7B), 120.4 (C5B), 113.0 (C2B). (d, 3J = 8 Hz, 2H, H5A), 8.33–8.24 (m, 3H, H1A, H5BHH ), IR (KBr disk): ν̃ [cm−1] = 3420 (m, N–H amine), 1600 (m, 7.74–7.67 (m, 2H, H6A), 7.67–7.60 (m, 2H, H6B), 7.13–7.05 (m, CvC), 843 (s, P–F). 4H, H8A, H8B), 6.91 (ddd, 3JHH = 7 Hz, 6 Hz, 4JHH = 1 Hz, 2H, Synthesis of [Ru(dpb-COOC H )(tpy)](PF ) 3(PF ). H7A2 5 6 6 ), 6.72 (ddd, 3JHH = 7 Hz, 6 Hz, 4J 7B 13HH = 1 Hz, 2H, H ). C [RuCl3(tpy)] (250 mg, 0.567 mmol, 1 eq.) and AgBF4 (320 mg, { 1H} NMR (CD3CN): δ [ppm] = 233.3 (C 9B), 169.4 (C10), 168.9 1.64 mmol, 2.9 eq.) were dissolved in dry acetone (20 ml) and (C4B), 159.8 (C4A), 155.5 (C8A), 153.5 (C3A), 152.8 (C8B), 143.2 heated to reflux for 2 h in the dark. The mixture was left to (C3B), 136.7 (C6B), 136.4 (C6A), 133.9 (C1A), 127.4 (C7A), 125.0 stand for 1 h and filtered through a syringe filter before remov- (C2B), 124.6 (C5A), 123.4 (C2A), 123.1 (C7B), 122.4 (C1B), 121.0 ing the solvent under reduced pressure. The dark residue was (C5B). IR (KBr disk): ν̃ [cm−1] = 3440 (s, O–H acid), 1665 (s, dissolved in nBuOH (20 ml) and ethyl 3,5-dipyrid-2′-ylbenzoate CvO acid), 1602, 1579 (m, CvC), 843 (s, P–F). L2 (207 mg, 0.680 mmol, 1.2 eq.) was added. The resulting Synthesis of [Ru(dpb-COOBt)(tpy)](PF6) 5(PF6). [Ru(dpb- mixture was heated to reflux for 16 h to give a dark purple COOC2H5)(tpy)](PF6) 4(PF6) (42 mg, 0.056 mmol, 1 eq.), N,N′- solution. After removal of the solvent under reduced pressure, diisopropylcarbodiimide (15 mg, 0.119 mmol, 2.1 eq.) and the residue was dissolved in a minimal amount of acetonitrile N-hydroxybenzotriazole (HOBt, 18 mg, 0.117 mmol, 2.1 eq.) (5 ml). Upon addition of a solution of NH4PF6 (220 mg, were dissolved in dry acetonitrile (20 ml) and stirred for 16 h 1.35 mmol, 2.4 eq.) in water (1 ml) followed by slow addition at room temperature. After removal of the solvent under of more water (80 ml), a dark red solid precipitated that was fil- reduced pressure, the dark residue was dissolved in aceto- tered off and washed with water (2 × 5 ml) and diethyl ether nitrile (5 ml). The product was precipitated by addition of (2 × 15 ml). Column chromatography on silica gel (CHCl3 : NH4PF6 (95 mg, 0.583 mmol, 10.4 eq.) and water (50 ml), fil- MeOH = 7 : 1) afforded [Ru(dpb-COOC2H5)(tpy)](PF6) 3(PF6) as tered off and washed with water (2 × 5 ml) and diethyl ether a dark red solid. Yield: 229 mg (0.293 mmol, 52%). Anal. Calc. (2 × 15 ml). [Ru(dpb-COOBt)(tpy)](PF6) 5(PF6) was obtained as for C34H26F6N5O2PRu (782.6): C, 52.18; H, 3.35; N, 8.95. a dark red solid. Yield: 45 mg (0.052 mmol, 92%). Anal. Calc. Found: C, 52.01; H, 3.34; N, 8.65. MS(ESI+): m/z (%) = 638.1 for C38H25F6N8O2PRu (871.69): C, 52.36; H, 2.89; N, 12.85. 5654 | D alton Trans., 2016, 45, 5640–5658 This journal is © The Royal Society of Chemistry 2016 Open Access Article. Published on 29 February 2016. Downloaded on 20/05/2016 14:13:44. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Section 3.3 | 105 View Article Online Dalton Transactions Paper Found: C, 52.42; H, 2.53; N, 12.54. MS(ESI+): m/z (%) = 699.1 Conclusions (11) [M − PF − N ]+, 727.1 (100) [M − PF ]+6 2 6 , 1599.2 (8) [2M − PF ]+6 . HR-MS(ESI +, m/z): Calcd for C38H25N8O2Ru [M − PF ]+6 : The electrochemical, UV-Vis and excited state properties of a 721.1176; Found: 721.1192. 1H NMR (CD3CN): δ [ppm] = 9.05 series of [Ru(dpb-R)(tpy)] + type of complexes was systematically (s, 2H, H2B), 8.77 (d, 2H, 3JHH = 8 Hz, H 2A), 8.44 (d, 2H, 3JHH = studied. The visible range absorption bands of these com- 8 Hz, H5A), 8.35 (t, 1H, 3J 1A 3HH = 8 Hz, H ), 8.34 (d, 2H, JHH = plexes are dominated by two electronically decoupled 1MLCT 8 Hz, H5B), 8.16 (d, 1H, 3JHH = 9 Hz, H 2C), 7.83 (d, 1H, 3JHH = transitions either involving the dpb ligand (dyz(Ru) → dpb) or 8 Hz, H5C), 7.77–7.66 (m, 5H, H6A, H6B, H4C), 7.58 (t, 1H, H3C), the tpy ligand (dxz(Ru) → tpy). These excitations are followed 7.20 (d, 2H, 3J = 5 Hz, H8BHH ), 7.13 (d, 2H, 3JHH = 5 Hz, H 8A), by intersystem crossing populating an emissive [Ru+(tpy−)] 6.96 (t, 2H, 3JHH = 6 Hz, H 7A), 6.80 (t, 2H, 3J 7B 13 3HH = 6 Hz, H ). C MLCT state in all cases. This state, however, is rapidly {1H} NMR (CD3CN): δ [ppm] = 239.5 (C 9B), 168.2 (C4B), 165.7 depopulated at room temperature via two additional low- (C10), 159.7 (C4A), 155.5 (C8A), 153.3 (C3A), 152.9 (C8B), 144.5 energy triplet excited states yielding very low luminescence (C1C), 144.2 (C3B), 137.0 (C6B), 136.9 (C6A), 134.8 (C1A), 130.1 quantum yields and short excited state lifetimes. VT steady- (C4C), 127.5 (C7A), 126.2 (C3C), 124.9 (C2B), 124.8 (C5A), 123.6 state emission spectroscopy and extended DFT calculations (C7B), 123.5 (C2A), 121.3 (C2C), 121.1 (C6C), 115.4 (C1B), 110.0 revealed their nature as 3LL′CT and 3MC states yielding a bi- (C5C). exponential dependence of the quantum yield on the tempera- Synthesis of [(tpy)Ru(dpb-NHCO-dpb)Ru(tpy)](PF6)2 6(PF6)2. ture. While the 3MC state has been known as a parasitic In separate Schlenk flasks, [Ru(dpb-NH2)(tpy)](PF6) 2(PF6) channel for non-radiative decay in (polypyridine)ruthenium(II) (35 mg, 0.048 mmol, 1 eq.) and [Ru(dpb-COOBt)(tpy)](PF6) complexes for over 30 years, 22 the observation of a 3LL′CT state 5(PF6) (42 mg, 0.048 mmol, 1 eq.) were dissolved in dry aceto- in such ruthenium complexes is unprecedented to the best of nitrile (10 ml). Molecular sieves (3 Å) were added and the mix- our knowledge. We previously referred to the 3LL′CT state as a tures were left to stand overnight to remove crystal water. Both spectroscopically undetectable state (“dark” state).40 However, solutions were then combined in a third Schlenk flask the characteristic temperature dependence of the quantum and tert-butylimino-tris(dimethylamino)phosphorane (P1-tBu) yield clearly is spectroscopic evidence for its presence. Also for (34 mg, 0.144 mmol, 3 eq.) was added. The resulting solution the bis(tridentate)iridium(III) complex [Ir(dpx)(tpy)]2+ (dpxH = was stirred at 50 °C for 16 h. After quenching the reaction by 1,5-di(2-pyridyl)-2,4-xylene), a 3LL′CT state is suggested to be addition of NH4PF6 (180 mg, 1.10 mmol, 23 eq.) dissolved in responsible for its low luminescence quantum yield. 101 Based water (2 ml), the solution was concentrated to 5 ml and the upon the findings of this study, we believe that the excited product was triturated by addition of water (80 ml). The pre- state deactivation in this cyclometalated iridium complex cipitate was filtered off, washed with water (2 × 5 ml) and occurs in an analogous manner via thermal depopulation of diethyl ether (2 × 15 ml) and purified by column chromato- the emissive state via 3LL′CT states. graphy on silica gel (CHCl3 : MeOH = 7 : 1) affording [(tpy) Remarkably, for the acceptor-substituted complexes 3 + and Ru(dpb-NHCO-dpb)Ru(tpy)](PF6)2 6(PF6)2 as a dark red solid. 4 +, the 3LL′CT state resides higher in energy than the 3MLCT Yield: 14 mg (0.0096 mmol, 20%). Anal. Calc. for state, while for the donor-substituted complexes 1+ and 2+, it is C63H43F12N11OP2Ru2 (1462.16)·4H2O: C, 49.32; H, 3.35; N, found to be the lowest triplet state. As a consequence, faster 10.04. Found: C, 49.39; H, 3.76; N, 10.36. MS(ESI+): m/z (%) = deactivation of the emissive 3MLCT states is observed in the 296.6 (3) [M − 2PF ]4+6 , 390.8 (17) [M − 2PF ]3+6 , 586.6 (100) latter complexes associated with substantially lowered emis- [M − 2PF ]2+6 , 1318.3 (5) [M − PF6]+. HR-MS(ESI+, m/z): Calcd sion quantum yields compared to complexes 3+ and 4+. But, for C63H43N11ORu2 [M − 2PF 2+6] : 586.5885; Found: 586.5884. since emission is observed for 1+ and 2+ with increasing 1H NMR (CD3CN): δ [ppm] = 9.63 (s, 1H, NH), 9.09 (s, 2H, quantum yields at lower temperatures, deactivation via the H2A), 8.83 (s, 2H, H2B), 8.80–8.74 (m, 4H, H2,tpy), 8.44 (m, 4H, 3LL′CT state is a thermally activated process and the 3MLCT H5,tpy), 8.39 (d, 3J = 8 Hz, 2H, H5A), 8.32 (t, 3J = 8 Hz, 1H, and 3HH HH LL′CT states are not in thermal equilibrium. H1,tpy), 8.28 (t, 3J 1,tpyHH = 8 Hz, 1H, H ), 8.18 (d, 3JHH = 8 Hz, Upon oxidation of the dinuclear complex 6 2+ to its mixed- 2H, H5B), 7.76–7.65 (m, 6H, H6,tpy, H6A), 7.65–7.59 (m, 2H, valent counterpart 63+, an intense NIR band is detected indi- H6B), 7.19–7.10 (m, 6H, H8,tpy, H8A), 7.07 (d, 3JHH = 6 Hz, 2H, cating a photochemical Ru II → RuIII charge transfer across the H8B), 6.96–6.88 (m, br, 4H, H7,tpy), 6.75 (m, 2H, H7A), 6.68 asymmetric biscyclometalating bridging ligand. Despite the (m, 2H, H7B). 13C{1H} NMR (CD3CN): δ [ppm] = 230.3 (C 9A), substantial redox asymmetry of the two complex subunits 217.9 (C9B), 169.6 (C5B), 169.3 (C5A), 168.0 (C10), 160.2, 160.0 bearing NH- and CO-substituents, a strong electronic com- (C4,tpy), 155.3, 155.2 (C8,tpy), 154.0, 153.6 (C3,tpy), 153.0 (C8A, munication between the donor and acceptor sites of 63+ is C8B), 143.2 (C3A), 142.7 (C3B), 136.8, 136.5, 136.4, 136.0 (C6,tpy, observed. In the excited state of 62+ however, the two complex C6A, C6B), 133.9 (C1B), 133.8, 132.9 (C1,tpy), 127.9 (C1A), 127.3 fragments appear electronically uncoupled with dual emission (C7,tpy), 124.6, 124.4 (C5,tpy), 123.4, 123.3 (C2,tpy), 123.3 (C2A), occurring from 3MLCT states localized at the two remote 123.0 (C7A), 122.6 (C7B), 120.9 (C8A), 120.7 (C8B), 118.9 (C2B). [Ru(tpy)] moieties. This “anti-Kasha” behaviour is explained IR (KBr disk): ν̃ [cm−1] = 3220 (m, N–H amide), 1635 (s, based on the long metal–metal distance and the very rapid CvO amide), 1599, 1582 (m, CvC), 1517 (w, amide II), 843 (s, excited state decay (emissive and non-emissive) that prevents P–F). thermal equilibration in solution via energy transfer entirely. This journal is © The Royal Society of Chemistry 2016 Dalton Trans., 2016, 45, 5640–5658 | 5655 Open Access Article. Published on 29 February 2016. Downloaded on 20/05/2016 14:13:44. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 106 | 3 RESULTS AND DISCUSSION View Article Online Paper Dalton Transactions Just upon freezing of the solution, the excited states become 10 I. Okura and N. Kim-Thuan, J. Mol. Catal., 1979, 5, 311– sufficiently long-lived to allow for thermalization, so that 314. Kasha’s rule is obeyed. 11 J.-M. Lehn and R. Ziessel, Proc. Natl. Acad. Sci. U. S. A., In summary, we were able to show spectroscopically and 1982, 79, 701–704. computationally that the introduction of N^C^N cyclometalat- 12 J. W. Tucker and C. R. J. Stephenson, J. Org. Chem., 2012, ing ligands in bis(tridentate)ruthenium(II) complexes gives rise 77, 1617–1622. to low-lying 3LL′CT states that allow for efficient thermal 13 B. O’Regan and M. Grätzel, Nature, 1991, 353, 737– depopulation of the emissive 3MLCT state. Furthermore, we 740. believe that the existence of 3LL′CT states in polypyridine tran- 14 M. G. Sasso, F. H. Quina and E. J. Bechara, Anal. Biochem., sition metal complexes is more general. As their energy is 1986, 156, 239–243. strongly dependent on the substitution pattern, they become 15 K. K.-W. Lo, T. K.-M. Lee, J. S.-Y. Lau, W.-L. Poon and particularly relevant in strong push–pull systems and can S.-H. Cheng, Inorg. Chem., 2008, 47, 200–208. interfere with luminescence processes and shorten the excited 16 P. D. Beer, Z. Chen, A. J. Goulden, A. Grieve, D. Hesek, state lifetimes substantially. F. Szemes and T. Wear, J. Chem. Soc., Chem. Commun., Additionally, in mixed-valent dinuclear ruthenium com- 1994, 1269–1271. plexes such as 63+, the biscyclometalating bridge was shown to 17 H. J. Bolink, L. Cappelli, E. Coronado and P. Gaviña, be an excellent mediator for photochemical electron transfer Inorg. Chem., 2005, 44, 5966–5968. between the redox sites even in the presence of a substantial 18 J. N. Demas and D. G. Taylor, Inorg. Chem., 1979, 18, redox asymmetry. The triplet excited states of 62+, however, are 3177–3179. electronically uncoupled due to the large spatial separation of 19 S. Yoon, P. Kukura, C. M. Stuart and R. A. Mathies, Mol. the emissive [Ru(tpy)] moieties and their short excited state Phys., 2006, 104, 1275–1282. lifetimes induced by the cyclometalating bridge. 20 J. V. Caspar and T. J. Meyer, J. Am. Chem. Soc., 1983, 105, 5583–5590. 21 K. Suzuki, A. Kobayashi, S. Kaneko, K. Takehira, T. Yoshihara, H. Ishida, Y. Shiina, S. Oishi and S. Tobita, Acknowledgements Phys. Chem. Chem. Phys., 2009, 11, 9850–9860. 22 B. Durham, J. V. Caspar, J. K. Nagle and T. J. Meyer, J. Am. Parts of this research were conducted using the supercomputer Chem. Soc., 1982, 104, 4803–4810. MOGON and advisory services offered by Johannes Gutenberg 23 J. van Houten and R. J. Watts, Inorg. Chem., 1978, 17, Univ. Mainz (http://www.hpc.uni-mainz.de), which is a 3381–3385. member of the AHRP and the Gauss Alliance e.V. This work 24 J. P. Sauvage, J. P. Collin, J. C. Chambron, S. Guillerez, was financially supported by the Deutsche Forschungsge- C. Coudret, V. Balzani, F. Barigelletti, L. de Cola and meinschaft (GSC 266, Materials Science in Mainz, scholarship L. Flamigni, Chem. Rev., 1994, 94, 993–1019. for C. K.). 25 K. Lashgari, M. Kritikos, R. Norrestam and T. Norrby, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 1999, 55, 64–67. Notes and references 26 J. R. Winkler, T. L. Netzel, C. Creutz and N. Sutin, J. Am. Chem. Soc., 1987, 109, 2381–2392. 1 J. P. Paris and W. W. Brandt, J. Am. Chem. Soc., 1959, 81, 27 M. L. Stone and G. A. Crosby, Chem. Phys. Lett., 1981, 79, 5001–5002. 169–173. 2 F. E. Lytle and D. M. Hercules, J. Am. Chem. Soc., 1969, 91, 28 K. Heinze, K. Hempel and M. Beckmann, Eur. J. Inorg. 253–257. Chem., 2006, 2006, 2040–2050. 3 K. Kalyanasundaram, Coord. Chem. Rev., 1982, 46, 159– 29 K. Heinze, K. Hempel and A. Breivogel, Z. Anorg. Allg. 244. Chem., 2009, 635, 2541–2549. 4 A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser 30 S. H. Wadman, M. Lutz, D. M. Tooke, A. L. Spek, F. Hartl, and A. von Zelewsky, Coord. Chem. Rev., 1988, 84, 85–277. R. W. A. Havenith, G. P. M. van Klink and G. van Koten, 5 M. Maestri, N. Armaroli, V. Balzani, E. C. Constable and Inorg. Chem., 2009, 48, 1887–1900. A. M. W. C. Thompson, Inorg. Chem., 1995, 34, 2759–2767. 31 A. Breivogel, C. Kreitner and K. Heinze, Eur. J. Inorg. 6 V. Balzani and A. Juris, Coord. Chem. Rev., 2001, 211, Chem., 2014, 2014, 5468–5490. 97–115. 32 M. Abrahamsson, M. Jäger, T. Österman, L. Eriksson, 7 S. Campagna, F. Puntoriero, F. Nastasi, G. Bergamini and P. Persson, H.-C. Becker, O. Johansson and V. Balzani, Top. Curr. Chem., 2007, 280, 117–214. L. Hammarström, J. Am. Chem. Soc., 2006, 128, 12616– 8 D. M. Hedstrand, W. H. Kruizinga and R. M. Kellogg, 12617. Tetrahedron Lett., 1978, 19, 1255–1258. 33 F. Schramm, V. Meded, H. Fliegl, K. Fink, O. Fuhr, Z. Qu, 9 H. D. Abruna, A. Y. Teng, G. J. Samuels and T. J. Meyer, W. Klopper, S. Finn, T. E. Keyes and M. Ruben, Inorg. J. Am. Chem. Soc., 1979, 101, 6745–6746. Chem., 2009, 48, 5677–5684. 5656 | D alton Trans., 2016, 45, 5640–5658 This journal is © The Royal Society of Chemistry 2016 Open Access Article. Published on 29 February 2016. Downloaded on 20/05/2016 14:13:44. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Section 3.3 | 107 View Article Online Dalton Transactions Paper 34 A. Breivogel, C. Förster and K. Heinze, Inorg. Chem., 2010, 61 K. C. D. Robson, B. D. Koivisto, A. Yella, B. Sporinova, 49, 7052–7056. M. K. Nazeeruddin, T. Baumgartner, M. Grätzel and 35 A. Breivogel, M. Meister, C. Förster, F. Laquai and C. P. Berlinguette, Inorg. Chem., 2011, 50, 5494–5508. K. Heinze, Chem. – Eur. J., 2013, 19, 13745–13760. 62 W. König and R. Geiger, Chem. Ber., 1970, 103, 788– 36 D. G. Brown, N. Sanguantrakun, B. Schulze, U. S. Schubert 798. and C. P. Berlinguette, J. Am. Chem. Soc., 2012, 134, 63 A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652. 12354–12357. 64 A. Schäfer, H. Horn and R. Ahlrichs, J. Chem. Phys., 1992, 37 M. Beley, J. P. Collin and J. P. Sauvage, Inorg. Chem., 1993, 97, 2571–2577. 32, 4539–4543. 65 A. Schäfer, C. Huber and R. Ahlrichs, J. Chem. Phys., 1994, 38 S. H. Wadman, J. M. Kroon, K. Bakker, M. Lutz, 100, 5829–5835. A. L. Spek, G. P. M. van Klink and G. van Koten, Chem. 66 F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys., Commun., 2007, 1907–1909. 2005, 7, 3297–3305. 39 P. G. Bomben, K. C. D. Robson, P. A. Sedach and 67 E. van Lenthe, E. J. Baerends and J. G. Snijders, J. Chem. C. P. Berlinguette, Inorg. Chem., 2009, 48, 9631– Phys., 1993, 99, 4597–4610. 9643. 68 S. Sinnecker, A. Rajendran, A. Klamt, M. Diedenhofen and 40 C. Kreitner, E. Erdmann, W. W. Seidel and K. Heinze, F. Neese, J. Phys. Chem. A, 2006, 110, 2235–2245. Inorg. Chem., 2015, 54, 11088–11104. 69 M. C. Hughes and D. J. Macero, Inorg. Chem., 1976, 15, 41 C. Creutz and H. Taube, J. Am. Chem. Soc., 1969, 91, 3988– 2040–2044. 3989. 70 S. Romain, C. Baffert, C. Duboc, J.-C. Leprêtre, 42 W. Kaim, A. Klein and M. Glöckle, Acc. Chem. Res., 2000, A. Deronzier and M.-N. Collomb, Inorg. Chem., 2009, 48, 33, 755–763. 3125–3131. 43 W. Kaim and G. K. Lahiri, Angew. Chem., 2007, 119, 1808– 71 R. F. Winter, Organometallics, 2014, 33, 4517–4536. 1828, (Angew. Chem., Int. Ed., 2007, 46, 1778–1796). 72 N. G. Connelly and W. E. Geiger, Chem. Rev., 1996, 96, 44 U. Fuerholz, H. B. Buergi, F. E. Wagner, A. Stebler, 877–910. J. H. Ammeter, E. Krausz, R. J. H. Clark, M. J. Stead and 73 C. Creutz, M. D. Newton and N. Sutin, J. Photochem. Photo- A. Ludi, J. Am. Chem. Soc., 1984, 106, 121–123. biol., A, 1994, 82, 47–59. 45 L. T. Zhang, J. Ko and M. J. Ondrechen, J. Am. Chem. Soc., 74 J.-P. Launay, Chem. Soc. Rev., 2001, 30, 386–397. 1987, 109, 1666–1671. 75 L. Wang, W.-W. Yang, R.-H. Zheng, Q. Shi, Y.-W. Zhong 46 S. P. Best, R. J. H. Clark, R. C. S. McQueen and S. Joss, and J. Yao, Inorg. Chem., 2011, 50, 7074–7079. J. Am. Chem. Soc., 1989, 111, 548–550. 76 W.-W. Yang, L. Wang, Y.-W. Zhong and J. Yao, Organo- 47 B. S. Brunschwig, C. Creutz and N. Sutin, Chem. Soc. Rev., metallics, 2011, 30, 2236–2240. 2002, 31, 168–184. 77 D. E. Richardson and H. Taube, J. Am. Chem. Soc., 1983, 48 M. B. Robin and P. Day, Adv. Inorg. Chem., 1968, 10, 247– 105, 40–51. 422. 78 M. Natali, S. Campagna and F. Scandola, Chem. Soc. Rev., 49 N. S. Hush, Prog. Inorg. Chem., 1967, 8, 391–444. 2014, 43, 4005–4018. 50 N. S. Hush, Electrochim. Acta, 1968, 13, 1005–1023. 79 F. Barigelletti, A. Juris, V. Balzani, P. Belser and A. von 51 N. S. Hush, Coord. Chem. Rev., 1985, 64, 135–157. Zelewsky, Inorg. Chem., 1983, 22, 3335–3339. 52 B. S. Brunschwig and N. Sutin, Coord. Chem. Rev., 1999, 80 A. Juris, F. Barigelletti, V. Balzani, P. Belser and A. von 187, 233–254. Zelewsky, Inorg. Chem., 1985, 24, 202–206. 53 G. D. Storrier and S. B. Colbran, Inorg. Chim. Acta, 1999, 81 G. A. Crosby, Acc. Chem. Res., 1975, 8, 231–238. 284, 76–84. 82 F. Barigelletti, A. Juris, V. Balzani, P. Belser and A. von 54 A. Breivogel, K. Hempel and K. Heinze, Inorg. Chim. Acta, Zelewsky, J. Phys. Chem., 1986, 90, 5190–5193. 2011, 374, 152–162. 83 R. Englman and J. Jortner, Mol. Phys., 1970, 18, 145– 55 C. Kreitner, M. Grabolle, U. Resch-Genger and K. Heinze, 164. Inorg. Chem., 2014, 53, 12947–12961. 84 J. V. Caspar, E. M. Kober, B. P. Sullivan and T. J. Meyer, 56 E. C. Constable and M. D. Ward, J. Chem. Soc., Dalton J. Am. Chem. Soc., 1982, 104, 630–632. Trans., 1990, 1405–1409. 85 J. V. Caspar and T. J. Meyer, J. Phys. Chem., 1983, 87, 952– 57 J.-P. Collin, P. Lainé, J.-P. Launay, J.-P. Sauvage and 957. A. Sour, J. Chem. Soc., Chem. Commun., 1993, 434–435. 86 K. A. King and R. J. Watts, J. Am. Chem. Soc., 1987, 109, 58 M. Beley, J. P. Collin, R. Louis, B. Metz and J. P. Sauvage, 1589–1590. J. Am. Chem. Soc., 1991, 113, 8521–8522. 87 T. E. Keyes, Chem. Commun., 1998, 889–890. 59 C. Patoux, J.-P. Launay, M. Beley, S. Chodorowski-Kimmes, 88 E. C. Glazer, D. Magde and Y. Tor, J. Am. Chem. Soc., 2007, J.-P. Collin, S. James and J.-P. Sauvage, J. Am. Chem. Soc., 129, 8544–8551. 1998, 120, 3717–3725. 89 Y. Chen, X. Zhou, X.-H. Wei, B.-L. Yu, H. Chao and 60 J.-Y. Shao and Y.-W. Zhong, Chem. – Eur. J., 2014, 20, L.-N. Ji, Inorg. Chem. Commun., 2010, 13, 1018–1020. 8702–8713. 90 D. L. Dexter, J. Chem. Phys., 1953, 21, 836–850. This journal is © The Royal Society of Chemistry 2016 Dalton Trans., 2016, 45, 5640–5658 | 5657 Open Access Article. Published on 29 February 2016. Downloaded on 20/05/2016 14:13:44. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 108 | 3 RESULTS AND DISCUSSION View Article Online Paper Dalton Transactions 91 F. Barigelletti, L. Flamigni, V. Balzani, J.-P. Collin, 96 G. R. Fulmer, A. J. M. Miller, N. H. Sherden, J.-P. Sauvage, A. Sour, E. C. Constable and H. E. Gottlieb, A. Nudelman, B. M. Stoltz, J. E. Bercaw and A. M. W. C. Thompson, J. Am. Chem. Soc., 1994, 116, K. I. Goldberg, Organometallics, 2010, 29, 2176–2179. 7692–7699. 97 F. Neese, Wiley Interdiscip. Rev.: Comput. Mol. Sci., 2012, 2, 92 V. Grosshenny, A. Harriman and R. Ziessel, Angew. Chem., 73–78. 1995, 107, 1211–1214, (Angew. Chem. Int. Ed., 1995, 34, 98 F. Neese, F. Wennmohs, A. Hansen and U. Becker, Chem. 1100–1102). Phys., 2009, 356, 98–109. 93 B. Schlicke, P. Belser, L. de Cola, E. Sabbioni and 99 R. Izsák and F. Neese, J. Chem. Phys., 2011, 135, 144105. V. Balzani, J. Am. Chem. Soc., 1999, 121, 4207– 100 D. A. Pantazis, X.-Y. Chen, C. R. Landis and F. Neese, 4214. J. Chem. Theory Comput., 2008, 4, 908–919. 94 M. Kasha, Discuss. Faraday Soc., 1950, 9, 14–19. 101 A. J. Wilkinson, H. Puschmann, J. A. K. Howard, 95 B. P. Sullivan, J. M. Calvert and T. J. Meyer, Inorg. Chem., C. E. Foster and J. A. G. Williams, Inorg. Chem., 2006, 45, 1980, 19, 1404–1407. 8685–8699. 5658 | D alton Trans., 2016, 45, 5640–5658 This journal is © The Royal Society of Chemistry 2016 Open Access Article. Published on 29 February 2016. Downloaded on 20/05/2016 14:13:44. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Section 3.4 | 109 3.4 STRONGLY COUPLED CYCLOMETALATED RUTHENIUM TRIARYLAMINE CHROMOPHORES AS SENSITIZERS FOR DSSCS Christoph Kreitner, Andreas K. C. Mengel, Tae Kyung Lee, Woohyung Cho, Kookheon Char, Yong Soo Kang and Katja Heinze Chem. Eur. J. 2016, published online: May, 19th, 2016, DOI: 10.1002/chem.201601001. Anchor-functionalized cyclometalated bis(tridentate) ruthenium(II) triarylamine hybrids featuring mixed-valent states of varying resonance stabilization were employed in dye-sensitized solar cells in combination with different electrolytes. Together with cobalt-based electrolytes, the N-carbazole substituted dye surpasses the N719 dye. Author Contributions The synthesis and characterization of the ruthenium complexes as well as all DFT calculations were performed by Christoph Kreitner. The cobalt electrolytes were synthesized and characterized by Andreas Mengel. After instruction from Prof. Dr. Yong Soo Kang and Prof. Dr. Kookheon Char the DSSCs were built and characterized by Andreas Mengel, Dr. Woohyung Cho and Tea-Kyung Lee. The evaluation of the DSSC results and the experimental arrangements were performed by Andreas Mengel. The manuscript was written by Christoph Kreitner (40 %) and Andreas Mengel (20 %) as well as Katja Heinze (40 %). Supporting Information for this article is found at pp. 240 (excluding Cartesian Coordinates of DFT-optimized geometries). For full Supporting Information, refer to http://onlinelibrary.wiley.com/store/10.1002/ chem.201601001/asset/supinfo/chem201601001-sup-0001-misc_information.pdf?v=1&s= 24bd0e564e6a0c15c98a29abcb13226acfb50978. „Reprinted with permission from Kreitner, C.; Heinze, K. Chem. Eur. J. 2016. Copyright 2016 Jon Wiley and Sons.” 110 | 3 RESULTS AND DISCUSSION DOI: 10.1002/chem.201601001 Full Paper &Cyclometalated Complexes Strongly Coupled Cyclometalated Ruthenium Triarylamine Chromophores as Sensitizers for DSSCs Christoph Kreitner+,[a, d] Andreas K. C. Mengel+,[a] Tae Kyung Lee,[b] Woohyung Cho,[b] Kookheon Char,[c] Yong Soo Kang,[b] and Katja Heinze*[a] Abstract: A series of anchor-functionalized cyclometalated increasing spin delocalization between the metal center and bis(tridentate) ruthenium(II) triarylamine hybrids [Ru(dbp- the triarylamine unit in the order [1a]2+< [1b]2+< [1c]2+ . X)(tctpy)]2 [2a]2–[2c]2 (H3tctpy=2,2’;6’,2’’-terpyridine- Solar cells were prepared with the saponified complexes 4,4’,4’’-tricarboxylic acid; dpbH=1,3-dipyridylbenzene; X= [2a]2–[2c]2 and the reference dye N719 as sensitizers N(4-C6H4OMe) 2 2 ([2a] ), NPh2 ([2b] 2), N-carbazolyl [2c]2) using the I 3 /I  couple and [Co(bpy) ]3+ /2+3 and was synthesized and characterized. All complexes show [Co(ddpd) ]3+ /2+2 couples as [B(C6F ) ]  5 4 salts as electrolytes broad absorption bands in the range 300–700 nm with (bpy=2,2’-bipyridine; ddpd=N,N’-dimethyl-N,N’-dipyridin-2- a maximum at about 545 nm. Methyl esters yl-pyridine-2,6-diamine). Cells with [2c]2 and I /I3 electro- [Ru(Me3tctpy)(dpb-X)] + [1a]+–[1c]+ are oxidized to the lyte perform similarly to cells with N719. In the presence of strongly coupled mixed-valent species [1a]2+–[1c]2+ and cobalt electrolytes, all efficiencies are reduced, yet under the RuIII(aminium) complexes [1a]3+–[1c]3+ at comparably these conditions [2c]2 outperforms N719. low oxidation potentials. Theoretical calculations suggest an Introduction electrode. The major advantages of dye sensitized solar cells over conventional silicon-based or inorganic thin film solar Pioneered by O’Regan and Grtzel in 1991,[1] the dye-sensitized cells are lower costs and their modular architecture allowing solar cell (DSSC) has emerged as a promising light-to-energy for systematic optimization of all components (semiconductor, conversion device.[2,3] Its setup has been optimized and stand- sensitizer, electrolyte) individually.[3–5] ardized over the past 25 years. Typically, its central component Tremendous efforts have been put particularly into the de- is a molecular dye that is absorbed onto a mesoporous wide- velopment of new molecular dyes to optimize cell per- bandgap semiconducting electrode, such as TiO [3,4]2 or ZnO. formance. An ideal sensitizer should be thermally and photo- Upon excitation by visible light, electrons are injected from the chemically stable under working conditions, should rapidly excited state of the dye into the conduction band of the semi- inject electrons into the conduction band of the semiconduc- conductor. The oxidized dye is then regenerated by a redox tor after excitation and, most importantly, should efficiently mediator, which transports the positive charge to the counter absorb light between 400 and 900 nm. Among others, polypyr- idine complexes of iron,[6] copper,[7–9] platinum,[10,11] iridium,[12,13] [a] C. Kreitner,+ A. K. C. Mengel,+ Prof. Dr. K. Heinze and rhenium[14] as well as polyaromatic and conjugated organ- Institute of Inorganic Chemistry and Analytical Chemistry ic compounds,[15,16] porphyrins,[17–19] and quantum dots[20] have Johannes Gutenberg-University of Mainz Duesbergweg 10–14, 55128 Mainz (Germany) proven suitable for sensitization. Particularly, polypyridine com- Fax: (+49)6131-39-27-277 plexes of ruthenium and osmium have emerged as a promising E-mail : katja.heinze@uni-mainz.de class of sensitizers due to their suitable photophysical proper- [b] T. K. Lee, W. Cho, Prof. Y. S. Kang ties.[21–24] The visible range of the electromagnetic spectrum of The Department of Energy Engineering and Center for these complexes is dominated by characteristic metal-to-ligand Next Generation Dye-Sensitized Solar Cells, Hanyang University charge transfer (MLCT) absorptions.[25–27]222 Wangsimni-ro, Seongdong-gu, Seoul 133-791 (Korea) In these transitions, [c] Prof. Dr. K. Char metal orbitals of the t2g set serve as electron donors, while the The National Creative Research Initiative Center for Intelligent polypyridine p* orbitals function as electron acceptors. Hybrids, School of Chemical and Biological Engineering, Seoul The most prominent and well-established sensitizers are the National University, 1 Gwanak-ro, Gwanak-gu, Seoul 151-744 (Korea) complexes [nBu4N]2[Ru(Hdcbpy)2(NCS)2] , N719 (H2dcbpy=2,2’- [d] C. Kreitner+ bipyridine-4,4’-dicarboxylic acid, Scheme 1),[28,29] and the so- Graduate School Materials Science in Mainz Staudingerweg 9, 55128 Mainz (Germany) called “black dye” [nBu4N]3[Ru(Htctpy)(NCS)3] (H3tctpy= [22] [+] These authors contributed equally to the work. 2,2’;6’,2’’-terpyridine-4,4’,4’’-tricarboxylic acid) reaching Supporting information for this article is available on the WWW under power conversion efficiencies (PCE, h) of 10–11% under full air http://dx.doi.org/10.1002/chem.201601001. mass 1.5 (AM 1.5) irradiation. In these complexes, the carboxy Chem. Eur. J. 2016, 22, 1 – 15 1  2016 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim && These are not the final page numbers!  Section 3.4 | 111 Full Paper clometalating ligands as electron acceptors.[37,38,43–48] Addition- ally, cyclometalation substantially increases the energy of the polypyridine-centered lowest unoccupied molecular orbital (LUMO) compared to non-cyclometalated counterparts. This should potentially accelerate charge injection into the TiO2 conduction band.[37, 38,48] The highest occupied molecular orbi- tal (HOMO) on the other side typically extends over the metal center and the anionic part of the cyclometalating ligand. This should facilitate dye regeneration after charge injection.[38,46, 48] Furthermore, the high s-donating strength destabilizes the in- herently photochemically reactive metal-centered (3MC) excit- ed states.[38,48–52] The electron donating or withdrawing charac- ter of the cyclometalating ligands are easily tuned by further substitution (for example [AX,Me]+ , Scheme 1) including hole- transport facilities (X=amines).[38,48–52] Indeed, several ap- proaches have been developed to incorporate electron donors into the dye structure to rapidly detract the positive charge re- maining on the sensitizer after electron injection away from the semiconductor surface.[53–56] Attaching the reversible tri- phenylamine radical cation/triphenylamine redox couple (TPA+ C/TPA0) proved particularly successful in conjunction with several porphyrin dyes, for example, YD2-o-C8, yielding solar cells with h>12%.[19,57] Berlinguette and co-workers demon- strated that the overall cell performance can benefit from a TPA unit linked to a [Ru(pbpy)(tpy)]+ complex via a thiophene spacer. This architecture yields cell efficiencies of up to 8.0% (Scheme 1, [BH]+ , [CH]+).[38] Through clever dye design and ad- justment of relative oxidation potentials of RuIII/II and TPA+ C/0 Scheme 1. N719 reference dye and amine-substituted cyclometalated ruthe- an efficient transfer of the electron hole from the ruthenium nium(II) dyes (top) and cobalt-based electrolytes (bottom). center to the TPA unit is achieved after charge injection. This retards parasitic electron recombination processes with oxi- dized dyes in the DSSC.[38,58–60] The mixed-valent complexes groups serve as anchors to the TiO surface while the [NCS] li- [BR]2+2 are valence-localized and assigned to Robin–Day class II gands are responsible for an efficient charge transfer from the with measurable electronic coupling between the metal center redox mediator onto the dye after charge injection (dye regen- and the TPA unit.[60,61] Recently, Zhong and co-workers present- eration).[29] However, a major drawback of complexes contain- ed a structurally related series of complexes combining bis(tri- ing monodentate ligands is their high lability towards [NCS] dentate) cyclometalated ruthenium complexes with TPA units substitution in photoexcited or oxidized states hampering (Scheme 1, [AX,Me]2+/[AX,Me]+) lacking the thiophene unit.[62,63] long-term application in photovoltaic devices.[30–33] The mixed-valent state [AX,Me]2+ is valence-delocalized (Robin– Recently, bi- and tridentate cyclometalating ligands emerged Day class III) between the metal center and the amine moiety as viable and more robust alternatives for the labile [NCS] li- as evidenced by the shape and bandwidth of the near infrared gands. In 2007, van Koten and co-workers reported the suc- absorption band and by density functional theoretical calcula- cessful sensitization of TiO2 by bis(tridentate) [Ru(pbpy)(tpy)] + tions.[62–65] The parent complex [AH,H]+ lacking the amine sub- complexes (tpy=2,2’;6’,2’’-terpyridine, Hpbpy=6-phenyl-2,2’- stituent (X=H) has been reported recently as well.[66] bipyridine).[34] Shortly thereafter, Grtzel and co-workers pub- In contrast to reported dyes [BR]+ , featuring a valence-iso- lished a dye with record-breaking characteristics meric description of the [BR]2+ state (Robin–Day class II), po- [Ru(H2dcbpy)2(ppy-F2)] + (h>10%) based on a tris(bidentate) tential DSSC sensitizers [AX,H]+ with X=amine that provide cyclometalating motif (Hppy-F2=2-(2,4-difluorophenyl)pyri- a means of detracting the electron hole away from the semi- dine).[35] Since then, much work has been dedicated towards conductor surface in a resonant fashion ([AX,H]2+ ; Robin–Day the development of new cyclometalated ruthenium dyes both class III) have not yet been reported. Saponifying the three in the field of tris(bidentate) and bis(tridentate) complex archi- methyl esters of [AX,Me]+ type complexes should provide suita- tectures.[36–42] These studies indeed reveal several key benefits ble sensitizers [AX,H]+ with a class III mixed-valent state. Herein, of the cyclometalating motif. The introduction of a Ru-C we present a series of three complexes of the general structure s bond in the coordination environment reduces the local [nBu4N]2[Ru(dpb-X)(tctpy)] (Hdpb-X=5-substituted 1,3-di-(2- symmetry around the metal center. This yields a broad absorp- pyridyl)benzene) with different amine substituents X of increas- tion band in the visible range resulting from multiple close- ing electron withdrawing power, namely N,N-bis(4-methoxy- lying MLCT transitions involving both the polypyridine and cy- phenyl)amine (X=N(4-C6H4OMe)2; [nBu4N]2[2a]), N,N-diphenyl- && Chem. Eur. J. 2016, 22, 1 – 15 www.chemeurj.org 2  2016 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim  These are not the final page numbers! 112 | 3 RESULTS AND DISCUSSION Full Paper amine (X=N(C6H5)2 ; [nBu4N]2[2b]) and carbazole (X=N-carba- zolyl ; [nBu4N]2[2c]). We will discuss how the substituents at the dpb ligand affect the degree of valence-delocalization in the mixed-valent state [2] and to what extent such delocalization is beneficial for the application of such sensitizers in DSSCs. As outer-sphere cobalt-based electrolytes[67–80] should deliver higher open-circuit voltages VOC due to their more positive redox potential as compared to the standard triiodide/iodide couple and as they perform extremely well in conjunction with TPA-appended porphyrin dyes (YD2-o-C8)[57] as well as with other potent TPA-appended dyes (Y123, D35),[72,73,74] we study the TPA-appended ruthenium(II) dyes [nBu4N]2[2a]– [nBu4N]2[2c] with cobalt-based electrolytes in addition to the commonly used triiodide/iodide couple. Specifically, we employ the [Co(bpy) ]3+ /2+3 [3] 3+ /2+ and [Co(ddpd) 3+ /2+2] [4]3+ /2+ redox mediators (bpy=2,2’-bipyridine, ddpd=N,N’-di- methyl-N,N’-dipyridin-2-yl-pyridine-2,6-diamine,[80] Scheme 1).[81–83] The DSSCs are studied by incident photon-to- Scheme 2. Synthesis of [1a][PF ]–[1c][PF ] and [nBu N] [2a]–[nBu N] [2c] . current conversion efficiency measurements, by current-volt- 6 6 4 2 4 2 age characteristics under AM 1.5 irradiation and in the dark as well as by electron lifetime measurements. [Ru(dpb)(tpy)]+ complexes, suggesting the presence of a second colored species. Yet, cyclic voltammograms confirm the purity of the synthesized complexes by absence of redox Results and Discussion waves in the range of 3.0 and 1.5 V other than the five ex- pected reversible waves,[62,63a] namely for the [1]3+ /2+ , [1]2+ /+ , Synthesis and characterization of chromophores [1]+ /0, [1]0/ and [1]/2 couples (Figure 1, Supporting Informa- The 5-substituted 1,3-di-(2-pyridyl)benzene ligands La (R=N(4- tion, Figure S5). C6H4OMe)2), L b (R=N(C6H5)2) and L c (R=N-carbazolyl) were syn- thesized starting from the previously reported 1-bromo-3,5-di- (2-pyridyl)benzene under Buchwald–Hartwig cross-coupling re- action conditions similar to a method we,[48] as well Zhong and co-workers employed previously.[63a] In the present study, the dimeric palladium(II) precatalyst bis(m-mesylate)bis[(2-(2’-ami- nophenyl-kN)phenyl-kC1)palladium(II)] [Pd] [84]2 was used along with the phosphane ligand 2-dicyclohexylphosphano-2’,6’-dii- sopropoxybiphenyl[85] to provide a catalytically competent cat- alyst that afforded the ligands in yields of 82–98%. The identi- ty of La was confirmed by comparison of its 1H NMR spectrum with that reported before.[62,63a] The purity and integrity of the new ligands Lb and Lc were ascertained by 1H and 13C NMR spectroscopy, mass spectrometry, and elemental analyses (Ex- perimental Section; Supporting Information, Figures S1–S4). The heteroleptic ester-substituted [Ru(dpb)(tpy)]+ com- plexes [1a]+–[1c]+ were prepared according to a previously employed synthetic method starting from RuCl3(Me3tctpy) (Scheme 2).[22,48] The two-step procedure includes chloride ab- Figure 1. Cyclic voltammograms of a) [1b][PF6] (a) and b) [nBu4N]2[2b] + straction with silver tetrafluoroborate followed by complexa- (c) dyes in [nBu4N][PF6]/CH3CN (E vs. FcH/FcH ). tion with the respective dipyridylbenzene ligand La–Lc under reducing conditions in n-butanol. Similar to observations made by Zhong and co-workers,[62,63a] we were not able to isolate the Subsequent saponification of the three methyl ester groups complexes [1a]+–[1c]+ with high purity. Despite the reducing of [1a]+–[1c]+ in aqueous solution using [nBu4N][OH] as base conditions during their synthesis, substantial amounts of the and hydrazine as reductant yielded the corresponding carboxy- open-shell RuIII complexes [1a]2+–[1c]2+ were obtained, as evi- lates as tetrabutylammonium salts [nBu4N]2[2a]–[nBu4N]2[2c] . denced from ESI mass spectra and the NMR silence of all three This method affords the fully deprotonated complexes [2a]2– compounds (paramagnetic broadening).[62,63a] Additionally, the [2c]2, in contrast to Berlinguette’s procedure,[38] which yields isolated products are black in solution and in the solid state in- the complexes in their neutral zwitterionic form with two pro- stead of the dark purple color typically observed for tonated carboxy groups. Owing to the high solubility of the Chem. Eur. J. 2016, 22, 1 – 15 www.chemeurj.org 3  2016 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim && These are not the final page numbers!  Section 3.4 | 113 Full Paper tetrabutylammonium salts [nBu4N]2[2a]–[nBu4N]2[2c] in organic solvents, the products are isolated straight-forwardly by extrac- tion of the aqueous phase with dichloromethane. Co-extracted [nBu4N][PF6] was removed by subsequent dissolution of the raw products in acetonitrile and addition of a diethylether/hex- anes mixture that precipitates the desired complexes. The in- tegrity of [nBu4N]2[2a]–[nBu4N]2[2c] was confirmed by 1H NMR and 13C NMR spectra as well as by ESI+ and ESI mass spectra (Supporting Information, Figures S6–S19). All NMR spectra lack paramagnetic shifts or broadening, substantiating the absence of RuIII in the pristine samples. The 1H NMR spectra confirm the presence of two equivalents of [nBu N]+4 cations per complex anion in all three cases corroborating the stoichiometry of the salt. IR spectra as KBr disk of the complexes [nBu4N]2[2a]– [nBu4N]2[2c] lack the characteristic vibrations of [PF ]  6 ions at 843 cm1 (asym. stretch) and 588 cm1 (deformation) present Figure 2. Normalized absorption and emission spectra of [nBu4N]2[2a] (c), in the parent complexes [1a][PF6]–[1c][PF6] underlining the [nBu4N]2[2b] (g), and [nBu4N]2[2c] (a) in CH3CN. quantitative [PF6]  removal (Supporting Information, Fig- ure S20). Additionally, 19F NMR spectra of [nBu4N]2[2a]– [nBu4N]2[2c] confirm the absence of [PF ]  6 . Under the acidic accepting sites (dRu!p*tpy and dRu!p*dpb).[38,48] Owing to the and ionizing conditions of the ESI+ mass spectrometry tech- low local symmetry around the metal center, the number of nique, the complexes are observed in their fully protonated absorption bands is larger than for the more symmetric sys- form as monocations [2+3H]+ with RuII metal sites or as dicat- tems containing all-nitrogen donor ligands such as [Ru(tpy) ]2+2 ions [2+3H]2+ with RuIII centers (Supporting Information, Fig- or [Ru(ddpd)(tpy)]2+ , for example.[38,46, 47,86] The lower symmetry ure S18). The ESI mass spectra (Supporting Information, Fig- yields substantially broadened absorption spectra and a more ure S19) show mass peaks at the expected m/z values for the efficient light harvesting throughout the visible range of the dianions [2]2 (RuII) and anions [2] (RuIII) with typical rutheni- electromagnetic spectrum. um isotope patterns. Furthermore, several m/z peaks of decar- The tris(carboxylate) complexes [nBu4N][2a]–[nBu4N][2c] are boxylated complexes are found for all three complexes very weakly emissive at room temperature (Figure 2, Table 1) [nBu4N]2[2a]–[nBu4N]2[2c] , confirming the presence of carbox- with quantum yields below 510 6. The wavelength of the ylate substituents. The carboxylate groups are also evident emission maximum is shifted hypsochromically in the order from the characteristic IR CO stretching vibrations around [nBu4N][2a]> [nBu4N][2b]> [nBu4N][2c] from 817 to 744 nm. 1617 cm1 (Supporting Information, Figure S20). On the one hand, this trend is due to a more pronounced vi- brational progression in [nBu4N][2a] and [nBu4N][2b] than in [nBu4N][2c] similar to that observed for other [Ru(dpb)(tpy)] + Photophysical and electrochemical behavior complexes with a strong push–pull substitution.[48] On the The absorption and emission spectra of the complexes [nBu4N] other hand, the energy of the emissive 3MLCT state increases [2a]–[nBu4N][2c] are depicted in Figure 2 and data are sum- with decreasing donor strength of the amine substituent, as marized in Table 1 (Supporting Information, Figure S21). In the this lowers the energy of the metal orbitals involved in the spectral range between 300 and 800 nm all dyes exhibit very emission process while essentially maintaining the tctpy-cen- similar absorption features. The absorption maximum around tered LUMO energy (Figure 3; Supporting Information, Fig- 536–549 nm is accompanied by three additional bands around ure S22).[48] 500, 425, and 375 nm. These bands characteristic for cyclome- Cyclovoltammetric studies of the ester-substituted com- talated [Ru(dpb)(tpy)]+ complexes arise from metal-to-ligand plexes [1a][PF6]–[1c][PF6] reveal multiple reversible redox pro- charge transfer excitations involving both ligands as electron- cesses (Figure 1, Table 1, Supporting Information, Figure S5). Table 1. Optical and electrochemical data of [1a][PF6]–[1c][PF6] and [nBu4N]2[2a]–[nBu4N]2[2c] . UV/Vis (CH3CN) lmax/nm (e/10 3 m1 cm1) Emission (CH3CN) lem/nm (lexc/nm) Cyclic voltammetry E/V vs. FcH/FcH + [1a][PF ][a]6 323 (29), 339 (26), 420 (16), 507 (13), 583 (10) – [b] 1.87, 1.52, 0.05, +0.31 [1b][PF ] –[b] –[b]6 1.85, 1.49, +0.09, +0.49 [1c][PF ] –[b] –[b]6 1.83, 1.47, +0.34, +0.88 [nBu4N]2[2a] 549 (15.4), 503 (12.1), 424 (9.7), 379 (11.3), 323 (37.5), 289 (62.7) 817 (549) [c] 2.52, 2.09, 0.21, –[d] [nBu N] [2b] 543 (15.4), 501 (12.1), 428 (8.7), 374 (10.4), 325 (36.4), 288 (61.6) 791 (543)[c] 2.54, 2.10, 0.15, –[d]4 2 [nBu N] [2c] 536 (13.9), 499 (12.5), 426 (9.1), 373 (10.0), 328 (29.1), 283 (63.2) 744 (536)[c] 2.51, 2.07, 0.06, –[d]4 2 [a] From Ref. [63a] . [b] No optical data measured due to the presence of RuIII species. [c] Quantum yield <5·106. [d] No second oxidation potential was ob- tained due to precipitation of the neutral dye on the electrode surface. && Chem. Eur. J. 2016, 22, 1 – 15 www.chemeurj.org 4  2016 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim  These are not the final page numbers! 114 | 3 RESULTS AND DISCUSSION Full Paper 600 mV towards more negative values (Figure 1, Table 1; Sup- porting Information, Figure S5). This is consistent with the sub- stantial increase of negative charge density on the tctpy3 ligand in [2]2 and corroborates the tpy centered reduction. The oxidation waves shift to lower potentials as well, but to a lesser extent. This is mainly due to the fact, that the metal or- bital involved with the oxidation process is orthogonal to the tpy ligand. However, a trend of the potential shifts is observed. While the first oxidation wave of [2a]2 occurs 160 mV below that of [1a]+ , the first oxidation potentials of [2c]2 and [1c]+ differ by 400 mV. Consequently, the first oxidation potentials of complexes [2a]2–[2c]2 only differ by 150 mV as opposed to a difference of 390 mV between the esters [1a]+–[1c]+ . This can be understood on the basis of the Mulliken spin popula- tions of the metal center and the amine nitrogen atom in the mixed-valent anions [2a]–[2c] : These amount to 0.49 (Ru)/ 0.17 (N) in [2a] , 0.57 (Ru)/0.13 (N) in [2b] , and 0.74 (Ru)/0.03 (N) in [2c] . Apparently, the charge delocalization over the tri- arylamine fragment is significantly reduced and the spin densi- Figure 3. a) Frontier molecular orbitals of [1a]+ , [1b]+ , and [1c]+ (contour ties of the mixed-valent anions [2]  are more valence-localized value 0.06 a.u.) and b) spin densities of [1a]2+ , [1b]2+ , and [1c]2+ (contour at the electron-rich metal center than their ester counterparts. value 0.01 a.u.). Consequently, the resonance stabilization of the mixed-valent species [2] is not as pronounced as that of [1]2+ resulting in The complexes are oxidized at quite low potentials to the similar oxidation potentials for all three complexes. The stron- mixed-valent counterparts [1]2+ (0.05 to 0.34 V vs. gest impact of deprotection and deprotonation is observed for FcH/FcH+). A second oxidation step occurs at higher poten- [2c]2, since its spin density is basically metal-centered. Ac- tials yielding the RuIII(aminium) complexes [1]3+ (0.31 to 0.88 V cordingly, oxidation occurs in the closest proximity to the neg- vs. FcH/FcH+).[62] The trend of the first and the second oxida- atively charged tctpy3 ligand and is facilitated to the largest tion potentials towards higher values in the order N(4- extent in the dye series [2] . C6H4OCH3)2 650 nm) have emerged as promising for practical applications[1] despite the fascinating photo- candidates for NIR organic light emitting diodes (OLEDs), physical aspects observed in [Cr(ox) ]3¢3 (ox= oxalato) poly- fiber-optic telecommunication applications, night-vision read- meric networks[17a,b] and the use of CrIII complexes as energy able displays, security inks for identification systems, oxygen donors for lanthanide emission in heterometallic complex- sensing, and in vivo imaging.[1–7] Essentially, all currently es.[17c–e] [Cr(bpy)3] 3+ and [Cr(phen)3] 3+ (byp= 2,2’-bipyridine) employed (water-)soluble, NIR emissive dyes are based on complexes have recently found renewed interest as photo- lanthanide complexes,[4–7] complexes of the second- and third- redox catalysts.[18] row metal ions,[8–10] complex organic scaffolds,[11] or a combi- The reasons for the poor quantum yields of CrIII com- nation of them.[12] All of them feature specific advantages, plexes can be understood from ligand field theory.[15] The such as long-lived emissive states and large energy differences desired luminescence of octahedral d3 CrIII complexes with between absorption and emission maxima (lanthanides, 4d/5d a (t 32g) (eg) 0 electron configuration occurs from a transition metal complexes), medium to high quantum yields, and from doublet states (2E and 2T1) to the quartet ground state rational tuning of the emission energy (organic dyes). Typical (4A2), in the red to near-infrared spectral region (for drawbacks are, however, multi-step syntheses and poor water simplicity, we use the Oh point-group classification). The 2E solubility and dye aggregation for the more extended p- and 2T1 spectroscopic terms as well as the 4A2 ground term systems required for NIR emission (organic dyes),[11h] short arise from the (t2g) 3 electron configuration and hence, the geometric reorganization is very minor, yielding sharp [16] [*] S. Otto, Dr. C. Fçrster, C. Kreitner, Prof. K. Heinze emission bands like the ruby emission. At low ligand-field Institute of Inorganic and Analytical Chemistry strength, the doublet states lie above the 4T2 state of electron Johannes Gutenberg-University of Mainz configuration (t )2 12g (eg) yielding weak, broad fluorescence Duesbergweg 10–14, 55128 Mainz (Germany) from 4T2 instead. [19] Even for classical strong-field ligands, E-mail: katja.heinze@uni-mainz.de such as bpy, phen, or 2,2’:6’,2’’-terpyridine (tpy), the energy Dr. M. Grabolle, Dr. U. Resch-Genger difference between 4T2 and the emitting 2E/2T1 states is so Division 1.10 small that back-intersystem crossing occurs, strongly reducing Federal Institute for Materials Research and Testing (BAM) phosphorescence quantum yields and lifetimes.[1, 15] Further- Richard-Willst•tter-Strasse 11, 12489 Berlin (Germany) 4 E-mail: ute.resch@bam.de more, the T2 state is prone to photosubstitution and hence, its C. Kreitner back-population should be avoided. [15, 20] To increase the Graduate School Materials Science in Mainz phosphorescence quantum yield, the energy difference 4 2 Staudingerweg 9, 55128 Mainz (Germany) between the T2 and E states should be large to prevent 4 Supporting information for this article is available on the WWW back-intersystem crossing to the detrimental T2 state. This under http://dx.doi.org/10.1002/anie.201504894. should be achievable by a using a strong ligand-field to shift 11572 Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 11572 –11576 Section 3.6 | 157 Angewandte Chemie the 4T2 state to higher energy in conjunction with a strong nephelauxetic effect lowering the energy of the doublet states 2E and 2T1 and hence should be made possible by proper ligand design. Recently, we introduced the tridentate ddpd ligand (N,N’- dimethyl-N,N’-dipyridin-2-ylpyridine-2,6-diamine) with a large bite angle N-M-N of around 908 in six-coordinate metal complexes to optimize metal–ligand orbital overlap and to induce a stronger ligand field compared to bpy or tpy (Scheme 1).[21] Also, ddpd is a poor p-acceptor ligand, that is, Figure 1. a) Molecular structure of the cation of 1(BF4)3 in the solid state (thermal ellipsoids set at 50% probability); b) space-filling representation of 13+ with the two ligands are shown in yellow and green, respectively (hydrogen atom omitted for clarity). crystals suitable for X-ray diffraction analysis (Figure 1, Supporting Information, Figure S1). The complex cations feature an essentially octahedral CrN6 coordination geometry with Cr¢N distances of 2.028–2.054 è and N-Cr-N angles close to 908 and 1808 as required for a large ligand-field splitting. Similar to structurally comparable [M(ddpd) ]2+2 complexes, the ligands are wrapped around the metal center (Figure 1) and the counter ions fill the pockets between the ligands with Cr···B/P distances between 5.3 and 7.0 è (Supporting Information, Figure S1).[21] Magnetic susceptibility and EPR data are consistent with a quartet ground state (cT= 1.833 cm3Kmol¢1 at 300 K; gav= 1.990 at 77 K, Figure S14, Supporting Information) similar to [Cr(tpy) ]3+.[24] A reversible CrIII/II3 reduction is observed at ¢1.11 V versus ferrocene (Supporting Information, Fig- ure S13). Compared to [Cr(bpy) ]3+3 (E1= =¢0.63 V) and2 3+ Scheme 1. High-yield syntheses of 1(X) and photographs of crystals of [Cr(tpy)2] (E1= =¢0.53 V), this reduction occurs at much3 2 more negative potential.[23]1(X) grown from CH CN solutions. DFT calculations (B3LYP, RIJ-3 3 COSX, Def2-SVP/J, Def2-SVP, ZORA) confirm the metal- centered reduction to CrII (Supporting Information, Fig- rather electron rich and difficult to reduce, but a quite strong ure S25,S26). The next reduction at Ep=¢1.94 V is irrever- s-donor ligand.[21] With these properties of ddpd in mind, we sible as coordinated ddpd cannot be reduced to its radical envisaged that ddpd could increase the energy of the 4T2 state anion. Interestingly, 1(BF4)3 is highly soluble in water in a [Cr(ddpd) ]3+2 complex 1 3+ (Scheme 1) and simultane- (0.0479 molL¢1) while 1(PF6)3 is more soluble in CH3CN ously decrease the energy of the 2E state, resulting in an (0.208 molL¢1), enabling different applications of the two enlarged 4T 22/ E energy gap, which impedes back-intersystem salts. The absorption spectra of 1 3+ in H2O or CH3CN show crossing. maxima at 220(sh), 302, 315(sh), 350(sh), and 435 nm [Cr(bpy) 3+3] and [Cr(tpy) 3+ 2] are substitutionally labile (Figure 2, Supporting Information, Figure S5) which can be under alkaline conditions giving the hydroxido complexes assigned to pp*, ligand-to-metal charge transfer (LMCT) and [Cr(bpy) (OH) ]+ and [Cr(tpy)(OH) ] (3¢x)n.[22]2 2 x n Possibly, the mixed metal-centered (MC)/LMCT excitations according to p-accepting ligands bpy and tpy reduce the electron density time-dependent DFT calculations (Supporting Information, between the ligand axes by back-donation from t2g orbitals, Figure S20). No metal-to-ligand charge transfer (MLCT) facilitating a nucleophilic attack of hydroxide. The p-accept- transitions were identified in this energy region because of ing nature of bpy/tpy also accounts for the special redox the weak electron-accepting properties of ddpd and the properties, as reduction of [CrIII(bpy) ]3+ or [CrIII3 (tpy) 3+ 2] inaccessible Cr III/IV oxidation. The low-energy absorption does not yield CrII, CrI, Cr0, Cr¢I oxidation states but is maximum is ascribed to the 4A2!4T2 transition (TD-DFT: ligand centered.[23] The envisaged ddpd complex 13+ should 427.7, 436.9, and 439.0 nm) and an LMCT (Supporting resist ligand-centered reductions and nucleophilic attack at Information, Figure S20). Three Laporte- and spin-forbidden the metal center due to the strong electron donating power of transitions are found at 697, 736, and 776 nm in the single- ddpd. crystal absorption spectrum of 1(BF4)3. These are assigned to The synthesis of 13+ is straightforward from CrCl and 42 A2!2T2(tentative), 2T1, and 2E excitations (Supporting ddpd[21a] in water. Ion exchange with (BF )¢ or (PF )¢4 6 gives Information, Figure S10). Excitation of a solution of 1(BF4)3 the bright orange salts 1(BF4)3 and 1(PF6)3 (Scheme 1, in water or CH3CN (Supporting Information, Figure S8) at Supporting Information). Both were obtained as single 435 nm leads to emission spectra that can be superimposed, as Angew. Chem. Int. Ed. 2015, 54, 11572 –11576 Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 11573 158 | 3 RESULTS AND DISCUSSION Angew.andte Communications 8.0%, t= 3.0 ns; Supporting Information, Figure S12), the weak 500 nm emission cannot be assigned to ddpd fluores- cence but is ascribed to the spontaneous 4T !42 A2 fluores- cence of 13+. Delayed 4T !4A fluorescence[15, 19]2 2 fed by back- intersystem crossing from 2E/2T1 states is ruled out on the basis of the short lifetime. Hence, back-intersystem crossing is efficiently prevented in 13+ which accounts for its exception- ally high quantum yield and lifetime. The minimal energy difference between the relaxed 2E and 4T2 states is estimated at around 7100 cm¢1 (0.88 eV; 85 kJmol¢1) from the emission spectra. Although the geometry of the 2E state is close to that of the 4A2 ground state, a large reorganization energy barrier is expected as the relaxed 4T2 state features a Jahn–Teller distorted octahedron with Cr¢Nax bonds elongated by Figure 2. Absorption factor (blue), excitation (lobs=775 nm, green) approximately 0.3 è according to DFT calculations (Support- and emission spectrum (lexc=435 nm, red) of 1(BF4)3 in deaerated ing Information, Figure S24–S26).[26] For back-intersystem H2O at room temperature; the inset shows the emission decay curves crossing (2E!4T2), the large energy gap and the reorganiza-of 1(BF4)3 in H2O with and without O2. tional barrier must be overcome which is clearly impossible at room temperature (Figure 3).[15] Direct intersystem crossing depicted in Figure 2 for CH3CN. The strong, sharp emission from 4T 2 2 22 to the vibrationally excited T1/ E states or to the T2 band at 775 nm (full width at half maximum height (FWHM)= 420 cm¢1) is ascribed to the 2E emission and the weaker band at 738 nm to the 2T emission.[15, 25]1 A single crystal of 1(BF4)3 emits at 740 and 778 nm (Supporting Information, Figure S11). Clearly, these two intraconfigura- tional doublet states equilibrate at room temperature both in solution and in the solid state. At 100 K in a frozen butyronitrile glass, only the 2E emission at 779 nm is observed (Supporting Information, Figure S9). The emission of 13+ is considerably red shifted relative to [Cr(bpy) ]3+3 (727 nm) and [Cr(phen)3] 3+ (730 nm), but similar to that of [Cr(tpy)2] 3+ (770 nm).[1, 15] The solid material ruby emits at 694 nm.[16] The luminescence quantum yields (F) of 13+ in deaerated CH3CN and H2O were determined absolutely with an Figure 3. Jablonski diagram of 13+ constructed from experimental integrating-sphere setup to F= 12.1% and 11.0%, respec- solution data (2T2 state tentatively from single-crystal absorption). tively. In D2O,F increases to 14.2%. To our knowledge, these ISC= intersystem crossing, IC= internal conversion. Fvalues are by far the highest values reported for CrIII complexes in solution at room temperature to date.[1, 15] For instance, [Cr(bpy) 3+3] , [Cr(phen) ] 3+ 3 , and [Cr(tpy) ] 3+ 2 have state and subsequent internal conversion is conceivable F= 0.089%, 0.15%, and < 0.00089% in water.[1] The life- (Figure 3). For Cr(acac)3 (acac= acetylacetonato), McCusker times (t) of the emitting doublet states of 13+were determined et al. have shown that intersystem crossing to 2E is faster than to t= 899, 898, and 1164 ms in deaerated CH3CN, in H2O, and vibrational cooling within the 4T2 state along the Jahn–Teller in D2O, respectively. Again, these are the highest values modes. [27] Intersystem crossing might also occur from vibra- reported to date for a molecular CrIII complex in solution at tionally hot states in 13+ before the Jahn–Teller distortion. room temperature. The lifetimes of [Cr(bpy)3] 3+, [Cr- Independent of the details of the intersystem crossing (phen) ]3+3 , and [Cr(tpy) 3+ 2] are t= 63 ms, 270 ms, and processes, the use of the strong-field ddpd ligand is very ! 30 ms, respectively.[1] The solid laser material ruby has t= efficient in inducing high phosphorescence quantum yields 4270 ms[16] while a single crystal of 1(BF4)3 reveals t= 443 ms. and lifetimes as a result of the large barrier for back- Excitation spectra recorded at 775 nm in CH3CN andH2O intersystem crossing. [15] perfectly match with the absorption spectrum around the As expected, the phosphorescence quantum yield is 435 nm maximum (Figure 2, Supporting Information, Fig- sensitive to the presence of O [28, 29]2. In air, F is reduced by ure S8) suggesting efficient population of the 2E/2T1 states factors of 5.2 (H2O) and 17 (CH3CN) and the lifetimes are from these 4T2 ligand-field and LMCT states. At higher correspondingly shortened from 898 ms to 177 ms (H2O) and energies, the excitation spectra deviate from the absorption 51 ms (CH3CN)(Figure 2). The bimolecular O2 quenching factor suggesting that not all high-energy states of 13+ constant has been estimated from a Stern–Volmer plot of populate the 2E/2T1 states. Excitation at 435 nm also yields 1(BF4)3 in H2O as kq= 1.77 × 10 7m¢1s¢1 and the Stern–Volmer a very weak broad emission band around 500 nm with t of constant as KSV= kd× t= 1.59 × 10 4m¢1 (Supporting Informa- 3 ns, independent of the presence of O2 (Supporting Infor- tion, Figure S15). These quenching efficiencies [29] suggest mation, Figure S7). As ddpd emits at 398 nm in CH3CN (F= possible applications of 1 3+ in optical oxygen sensors,[2, 30] 11574 www.angewandte.org Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 11572 –11576 Section 3.6 | 159 Angewandte Chemie with the large difference between excitation and emission (www.hpc.uni-mainz.de), which is a member of the AHRP easing the combination with a spectrally distinguishable O2- and the Gauss Alliance e.V. This work was financially inert reference dye. The quenching efficiency is explained on supported by the Deutsche Forschungsgemeinschaft (GSC the basis of the very long 2E lifetime and on the basis of spin 266, Materials Science in Mainz, scholarship for C.K.). statistics, although kq is not particularly large. [29c] The kq value might be associated with an effective shielding of CrIII by the Keywords: chromium complexes · intersystem crossing · ligands and the counterions (Figure 1, Figure S1). Commonly ligand-field splitting · NIR luminescence · photophysics employed optical oxygen sensors are based on the quenching of their dyeÏs excited triplet states, for example, 3MLCT or How to cite: Angew. Chem. Int. Ed. 2015, 54, 11572–11576 3pp*, by 3O yielding the dyeÏs singlet ground state and Angew. Chem. 2015, 127, 11735–117392 1O .[2, 30]2 For these triplet states, spin statistics predict that 1/9 (11%) of the possible encounters (quintet, triplet, singlet: [1] H. Xiang, J. Cheng, X. Ma, X. Zhou, J. Chruma, Chem. Soc. Rev. 9 possibilities), namely the singlets, are productive. For the 2E 2013, 42, 6128 – 6185. state of 13+ and 3O2, a quartet and a doublet encounter [2] M. Quaranta, S. M. Borisov, I. Klimant, Bioanal. Rev. 2012, 4, complex is conceivable giving six microstate possibilities. The 115 – 157. 4 3+ [3] a) “Luminescent lanthanide complex, and articles and inksquartet encounter is productive giving the A2 state of 1 and 1 4 1 containing the luminescent complex”: F. Thomas, C. Laporte,O2. Hence, /6 (67%) of the encounters should yield O2 PCT Int. Appl. WO 2014048702A1, 2014 ; b) “Secure document which explains the O2 sensitivity of Cr III complexes in general. comprising luminescent chelates”: V. Aboutanos, T. Tiller, C. The substitutional stability of 1(BF4)3 was probed in Reinhard, S. RascagnÀres, PCT Int. Appl. WO 2010130681A1, aqueous solution (pH 7) as well as in the presence of HCl 2010. (pH 2.1) and NaOH (pH 11.9). The cation 13+ is stable for at [4] A. J. Amoroso, S. J. A. Pope, Chem. Soc. Rev. 2015, 44, 4723 – least 2.5 months according to UV/Vis spectroscopy (Fig- 4742. [5] E. Pershagen, K. E. Borbas, Coord. Chem. Rev. 2014, 273 – 274, ure S16,S17). This stability is in stark contrast to the lability of 30 – 46. [Cr(bpy) ]3+ and [Cr(tpy) ]3+.[22]3 2 Also, 1 3+ is perfectly stable in [6] E. J. New, D. Parker, D. G. Smith, J. W. Walton, Curr. Opin. 0.1m [nBu4N]Cl and in [nBu4N](OH) (pH 11.4) H2O/CH3CN Chem. Biol. 2010, 14, 238 – 246. (1:1) solution under illumination with LEDs at 430 nm in air [7] S. V. Eliseeva, J.-C. G. Bînzli, Chem. Soc. Rev. 2010, 39, 189 – according to absorption and emission spectra while an 227. isoabsorptive solution of [Cr(bpy) ]3+ undergoes complete [8] a) Q. Zhao, C. Huanga, F. Li, Chem. Soc. Rev. 2011, 40, 2508 –3 photosubstitution within a few hours (Figure S18).[15] These 2524; b) O. S. Wenger, Chem. Rev. 2013, 113, 3686 – 3733. 3+ [9] V. W.-W. Yam, K. M.-C. Wong, Chem. Commun. 2011, 47,experiments demonstrate the superior stability of 1 com- 11579 – 11592. pared to [Cr(bpy)3] 3+ in aqueous solution. [10] P.-T. Chou, Y. Chi, Chem. Eur. J. 2007, 13, 380 – 395. Thanks to the difficult CrIII/CrII reduction and the low 2E [11] Selection of examples: a) S. Wiktorowski, C. Rosazza, M. J. energy, the oxidative power of the 2E state of 13+ is rather Winterhalder, E. Daltrozzo, A. Zumbusch, Chem. Commun. small [E(CrIII/II)*=E(CrIII/II)+E (2E)=¢1.11 V+ 1.60 V= 2014, 50, 4755 – 4758; b) T. Marks, E. Daltrozzo, A. Zumbusch,00 0.49 V versus ferrocene (+ 1.12 V vs. normal hydrogen Chem. Eur. J. 2014, 20, 6494 – 6504; c) D. Frath, J. Massue, G. Ulrich, R. Ziessel, Angew. Chem. Int. Ed. 2014, 53, 2290 – 2310; electrode (NHE))]. Hence, no photooxidative damage to Angew. Chem. 2014, 126, 2322 – 2342; d) J. C. Er, C. Leong, C. L. organic material is expected. In contrast [Cr(bpy) ]3+3 or Teoh, Q. Yuan, P. Merchant, M. Dunn, D. Sulzer, D. Sames, A. [Cr(ttpy)2] 3+ photooxidize dGMP and hence, cleave DNA in Bhinge, D. Kim, S.-M. Kim, M.-H. Yoon, L. W. Stanton, S. H. Je, their excited states (ttpy=p-tolylterpyridine, dGMP= deoxy- S.-W. Yun, Y.-T. Chang, Angew. Chem. Int. Ed. 2015, 54, 2442 – guanosine monophosphate).[31] Indeed, dGMP (E= 1.29 V vs. 2446; Angew. Chem. 2015, 127, 2472 – 2476; e) S. Wiktorowski, NHE) quenches the emission of [Cr(bpy) ]3+ under our E. Daltrozzo, A. Zumbusch, RSC Adv. 2015, 5, 29420 – 29423;3 conditions but not that of 13+ (Figure S19). f) D. J•nsch, C. Li, L. Chen, M. Wagner, K. Mîllen, Angew. Chem. Int. Ed. 2015, 54, 2285 – 2289; Angew. Chem. 2015, 127, Based upon rational ligand design, we could obtain the 2314 – 2319; g) E. Heyer, P. Lory, J. Leprince, M. Moreau, A. first molecular, water- or CH3CN-soluble ruby analogues Romieu, M. Guardigli, A. Roda, R. Ziessel, Angew. Chem. Int. 1(BF4)3 and 1(PF6)3 with excellent phosphorescence quantum Ed. 2015, 54, 2995 – 2999; Angew. Chem. 2015, 127, 3038 – 3042; yields. Together with their high stability, their simple high- h) U. Resch-Genger, M. Grabolle, S. Cavaliere-Jaricot, R. yield synthesis, their large excitation/emission energy gap and Nitschke, T. Nann, Nat. Methods 2008, 5, 763 – 775. long lifetime these properties will allow a plethora of [12] M. Schulze, A. Steffen, F. Wîrthner, Angew. Chem. Int. Ed. applications, such as time-gated imaging, the design of optical 2015, 54, 1570 – 1573; Angew. Chem. 2015, 127, 1590 – 1593. [13] a) J. V. Caspar, E. M. Kober, B. P. Sullivan, T. J. Meyer, J. Am. probes, and integration into multiplexed sensing schemes. We Chem. Soc. 1982, 104, 630 – 632; b) J. V. Caspar, T. J. Meyer, J. aim to explore their full potential in the near future. Am. Chem. Soc. 1983, 105, 5583 – 5590; c) J. V. Caspar, T. J. Meyer, J. Phys. Chem. 1983, 87, 952 – 957; d) T. J. Meyer, Pure Appl. Chem. 1986, 58, 1193 – 1206. Acknowledgements [14] a) N. Armaroli, G. Accorsi, F. Cardinali, A. Listorti, Top. Curr. Chem. 2007, 280, 69 – 115; b) C. E. McCusker, F. N. Castellano, Inorg. Chem. 2013, 52, 8114 – 8120; c) M. S. Lazorski, F. N. We thank Luca Carrella and Eva Rentschler for collecting the Castellano, Polyhedron 2014, 82, 57 – 70. SQUID data of 1(PF6)3. Parts of this research were conducted [15] a) A. D. Kirk, Chem. Rev. 1999, 99, 1607 – 1640; b) L. S. Forster, using the supercomputer Mogon and advisory services Coord. Chem. Rev. 2002, 227, 59 – 92; c) N. A. P. Kane-Maguire, offered by Johannes Gutenberg University Mainz Top. Curr. Chem. 2007, 280, 37 – 67; d) V. Balzani, P. Ceroni, A. Angew. Chem. Int. Ed. 2015, 54, 11572 –11576 Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 11575 160 | 3 RESULTS AND DISCUSSION Angew.andte Communications Juris, Photochemistry and Photophysics, Wiley-VCH,Weinheim, Housecroft, M. Neuburger, J. Schçnle, J. A. Zampese, Dalton 2014, pp. 216 – 219; e) M. A. Jamieson, N. Serpone, M. Z. Hoff- Trans. 2014, 43, 7227 – 7235. man, Coord. Chem. Rev. 1981, 39, 121 – 179; f) C. K. Ryu, J. F. [23] a) C. C. Scarborough, K. M. Lancaster, S. DeBeer, T. Weyher- Endicott, Inorg. Chem. 1988, 27, 2203 – 2214. mîller, S. Sproules, K. Wieghardt, Inorg. Chem. 2012, 51, 3718 – [16] a) T. H. Maiman, Nature 1960, 187, 493 – 494; b) C. Degli Es- 3732; b) C. C. Scarborough, S. Sproules, T. Weyhermîller, S. posti, L. Bizzocchi, J. Chem. Educ. 2007, 84, 1316 – 1318. Serena DeBeer, K. Wieghardt, Inorg. Chem. 2011, 50, 12446 – [17] a) A. Hauser, M. E. Von Arx, R. Pellaux, S. Decurtins, Mol. 12462. Cryst. Liq. Cryst. 1996, 286, 225 – 230; b) M. Milos, S. Kairouani, [24] U. Casellato, R. Graziani, R. P. Bonomo, A. J. Di Bilio, J. Chem. S. Rabaste, A. Hauser,Coord. Chem. Rev. 2008, 252, 2540 – 2551; Soc. Dalton Trans. 1991, 23 – 31. c) L. Aboshyan-Sorgho, H. Nozary, A. Aebischer, J.-C. G. [25] N. A. P. Kane-Maguire, J. Conway, C. H. Langford, J. Chem. Soc. Bînzli, P.-Y. Morgantini, K. R. Kittilstved, A. Hauser, S. V. Chem. Commun. 1974, 801 – 802. Eliseeva, S. Petoud, C. Piguet, J. Am. Chem. Soc. 2012, 134, [26] a) F. Gilardoni, J. Weber, K. Bellafrouh, C. Daul, H. U. Gîdel, J. 12675 – 12684; d) M. Cantuel, F. Gumy, J.-C. G. Bînzli, C. Chem. Phys. 1996, 104, 7624 – 7632; b) O. S. Wenger, H. U. Piguet, Dalton Trans. 2006, 2647 – 2660; e) S. Torelli, D. Gîdel, J. Chem. Phys. 2001, 114, 5832 – 5841. Imbert, M. Cantuel, G. Bernardinelli, S. Delahaye, A. Hauser, [27] a) E. A. Juban, J. K. McCusker, J. Am. Chem. Soc. 2005, 127, J.-C. G. Bînzli, C. Piguet, Chem. Eur. J. 2005, 11, 3228 – 3242. 6857 – 6865; b) E. A. Juban, A. L. Smeigh, J. E. Monat, J. K. [18] S. M. Stevenson, M. P. Shores, E. M. Ferreira,Angew. Chem. Int. McCusker, Coord. Chem. Rev. 2006, 250, 1783 – 1791. Ed. 2015, 54, 6506 – 6510; Angew. Chem. 2015, 127, 6606 – 6610. [28] a) N. Serpone, M. A. Jamieson, M. S. Henry, M. Z. Hoffman, F. [19] a) H. Yersin, P. Huber, G. Gietl, D. Trîmbach, Chem. Phys. Lett. Bolletta, M. Maestri, J. Am. Chem. Soc. 1979, 101, 2907 – 2916; 1992, 199, 1 – 9; b) F. Castelli, L.-S. Forster, J. Am. Chem. Soc. b) B. Brunschwig, N. Sutin, J. Am. Chem. Soc. 1978, 100, 7568 – 1975, 97, 6306 – 6309. 7577. [20] a) N. Sabbatini, V. Balzani, J. Am. Chem. Soc. 1972, 94, 7587 – [29] a) M. Isaacs, A. G. Sykes, S. Ronco, Inorg. Chim. Acta 2006, 359, 7589; b) H. F. Wasgestian, J. Phys. Chem. 1972, 76, 1947 – 1951. 3847 – 3854; b) K. D. Barker, K. A. Barnett, S. M. Connell, J. W. [21] a) A. Breivogel, C. Fçrster, K. Heinze, Inorg. Chem. 2010, 49, Glaeser, A. J. Wallace, J. Wildsmith, B. J. Herbert, J. F. Wheeler, 7052 – 7056; b) K. Mack, A. Wînsche von Leupoldt, C. Fçrster, N. A. P. Kane-Maguire, Inorg. Chim. Acta 2001, 316, 41 – 49; M. Ezhevskaya, D. Hinderberger, K. W. Klinkhammer, K. c) M. Z. Hoffman, F. Bolletta, L. Moggi, G. L. Hug, J. Phys. Heinze, Inorg. Chem. 2012, 51, 7851 – 7858; c) A. Breivogel, M. Chem. Ref. Data 1989, 18, 219 – 543. Meister, C. Fçrster, F. Laquai, K. Heinze, Chem. Eur. J. 2013, 19, [30] J. Napp, T. Behnke, L. Fischer, C. Wîrth, M. Wottawa, D. M. 13745 – 13760; d) C. Fçrster, K. Mack, L. M. Carrella, V. Katschinski, F. Alves, U. Resch-Genger, M. Sch•ferling, Anal. Ksenofontov, E. Rentschler, K. Heinze, Polyhedron 2013, 52, Chem. 2011, 83, 9039 – 9046. 576 – 581; e) A. Breivogel, M. Park, D. Lee, S. Klassen, A. [31] a) R. T. Watson, N. Desai, J. Wildsmith, J. F. Wheeler, N. A. P. Kîhnle, C. Lee, K. Char, K. Heinze, Eur. J. Inorg. Chem. 2014, Kane-Maguire, Inorg. Chem. 1999, 38, 2683 – 2687; b) V. G. 288 – 295; f) A. Breivogel, C. Kreitner, K. Heinze, Eur. J. Inorg. Vaidyanathan, B. U. Nair, Eur. J. Inorg. Chem. 2004, 1840 – 1846. Chem. 2014, 5468 – 5490; g) A. K. C. Mengel, C. Fçrster, A. Breivogel, K. Mack, J. R. Ochsmann, F. Laquai, V. Ksenofontov, K. Heinze, Chem. Eur. J. 2015, 21, 704 – 714. [22] a) M. Maestri, F. Bolletta, N. Serpone, L. Moggi, V. Balzani, Received: May 29, 2015 Inorg. Chem. 1976, 15, 2048 – 2051; b) E. C. Constable, C. E. Published online: August 12, 2015 11576 www.angewandte.org Ó 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2015, 54, 11572 –11576 | 161 4 SUMMARY AND OUTLOOK In this work, the successful synthesis and characterization of novel mono- and dinuclear push-pull substituted polypyridine ruthenium complexes is presented. The dinuclear bis(terpyridine)ruthenium complex [(EtOOC-tpy)Ru(tpy-NHCO-tpy)Ru(tpy-NHAc)]4+ [I b]4+ (section 3.1) does not exhibit any measurable electronic coupling between the metal centers in the mixed valent state [I b]5+. This was attributed to high tunneling barriers for electron transfer as the bridge’s frontier orbitals differ drastically from the metal dπ orbitals in their energies. In contrast, in the photo-excited state of [I b]4+, the bridging ligand is reduced by one electron, while one of the metal centers is oxidized. The resulting mixed-valent system is partially delocalized between the two valence isomers [(EtOOC-tpy)RuII(tpy-NHCO-tpy−)RuIII(tpy-NHAc)]4+ and [(EtOOC- tpy)RuIII(tpy-NHCO-tpy−)RuII(tpy-NHAc)]4+. Both valence isomers are in thermal equilibrium mediated by electron transfer between the two metal sites. This is possible as the formal reduction of the bridge in the photo-excited state opens up additional electron transfer channels that are not accessible in [I b]5+, leading to the detection of dual emission at room temperature. Furthermore, several mononuclear cyclometalated polypyridine ruthenium complexes with a [Ru(N^N^N)(N^C^N)]+ coordination cage were synthesized and studied (sections 3.2 and 3.3). The UV-Vis absorption properties of these [Ru(dpb-R’)(tpy-R’’)]+ systems were examined on a theoretical level and the obtained results verified using resonance Raman spectroscopy: The low- energy absorption band in the range of 470-600 nm consists of Ru→tpy and Ru→dpb transitions to a similar extent in all cases, while 350-450 nm band is predominantly composed of Ru→tpy excitations involving higher unoccupied orbitals at the tpy ligand. All of the studied [Ru(dpb-R’)(tpy-R’’)]+ complexes are weakly emissive at room temperature with quantum yields in the range of 10−6 – 10−5 excluding [Ru(dpb-NHAc)(tpy-COOEt)]+ (sections 3.2 and 3.3). The latter is non-emissive both at room temperature and at 77 K. The lack of emission was attributed to the strong push-pull substitution induced by the cyclometalation which is further amplified by the carboxy and amide substituents. This reduces the emission energy to an extent that non-radiative decay via tunneling into vibrationally excited states of the singlet ground state becomes very efficient leading to complete quenching of the excited state via this channel (section 3.2). Using temperature-dependent quantum yield measurements complemented with quantum chemical results, it was shown for the luminescent complexes, that the emissive 3MLCT state is flanked by two parasitic triplet states that promote radiationless excited state decay, namely a 3MC state and a 3LL’CT state. Both states are thermally accessible with activation barriers of around 25 and 5 kJ mol−1, respectively, leading to a biexponential dependence of the emission quantum yield on the temperature. The experimentally determined activation barriers were correlated with the electronic 3MLCT→3MC and 3MLCT→3LL’CT transitions based on the excellent agreement with the energies of the DFT-calculated transition states with deviations as small as ± 2 kJ mol−1 (section 3.3). The dinuclear cyclometalated polypyridine ruthenium complex [(tpy)Ru(dpb-NHCO- dpb)Ru(tpy)]2+ [1]2+ with an amide-containing bridging ligand was synthesized and characterized as well (section 3.3). As suggested by preliminary DFT results, the odd electron is predominantly 162 | 4 SUMMARY AND OUTLOOK located at the carboxy-substituted complex fragment in the mixed-valent state: [[(tpy)Ru(dpb- NHCO-dpb)RuIII(tpy)]3+ [1]3+. However, the UV-Vis-NIR spectrum of [1]3+ shows an NIR absorption band at 1100 nm characteristic for an IVCT transition. This indicates that [1]3+ belongs to Robin- Day class II in contrast to the similar non-cyclometalated class I complex [I b]5+. Also the emissive properties of [1]2+ differ from those of [I b]4+. At room temperature, dual emission is observed arising from two RuIII(tpy−) 3MLCT states localized at the two capping fragments, respectively. However, these states are spatially so far apart that thermal equilibration to the lower-energy [(tpy−)RuIII(dpb-NHCO-dpb)RuII(tpy)]2+ state via Dexter energy transfer is very slow. Hence, on the timescale of the excited state lifetime, which is estimated to be well below 1 ns, energy transfer does not occur. Only in frozen solution at 77 K, single emission from the N-substituted complex fragment is observed, as the transition to the solid phase slows down all non-emissive decay channels thus allowing energy transfer to occur prior to emission. Finally, [Ru(dpb-NR2)(tpy(-COO) 2−3)] complexes bearing diarylamine substituents at the cyclometalating dipyridylbenzene ligand were synthesized (section 3.4). These are known to exhibit substantial charge delocalization between the metal center and the amine functionality in the mixed-valent state [Ru(dpb-NR2)(tpy-COO)]2− (Robin-Day class II/III regime). The suitability of such mesomeric charge delocalization away from the semiconductor surface was tested with respect to an application in DSSCs. Unfortunately, the charge delocalization in the mixed-valent state yields a resonance-stabilization that hampers the dye regeneration by the employed electrolytes. Nonetheless, the carbazole-substituted dye reached an external efficiency of 𝜂 = 3.3 % using the iodide/triiodide electrolyte while benchmark dye N719 reached 𝜂 = 5.8 % using the same cell setup. In the presence of cobalt electrolytes, on the other hand, the performances of the carbazole-substituted cyclometalated dye and N719 are very similar with efficiencies just above 1 %. In summary, the photophysical properties of cyclometalated polypyridine ruthenium complexes with a [Ru(N^N^N)(N^C^N)]+ coordination cage have been illustrated using a variety of experimental and theoretical techniques. Especially the discovery that the 3MLCT and 1GS potential energy surfaces intersect in the proximity of the minimum 3MLCT geometry, sheds new light on the research of highly emissive ruthenium-based NIR emitters (section 3.5). This surface crossing seems to be a common feature of cyclometalated bis(tridentate) complexes, as it was also found for [Ru(tpy)(pbpy)]+ with a [Ru(N^N^N)(N^N^C)]+ environment. However, an evaluation of the importance of this phenomenon for the excited state deactivation of cyclometalated ruthenium complexes definitely requires future work for example by studying the temperature dependence of the quantum yield of [Ru(tpy)(pbpy)]+. The 3MC state of this complex is thermally inaccessible at room temperature and its responsibility in thermally activated emission quenching be excluded. If this research suggests, that a direct 3MLCT→1GS surface crossing is indeed a viable excited state decay channel in cyclometalated bis(tridentate) ruthenium complexes this will mean that not only the energies of 3MC and potential 3LL’CT states need to be adjusted in order to improve the emission quantum yield of such systems. Additionally, the energy of this surface crossing point will have to be taken into consideration. As it is inherently linked to the distortion of the excited state with respect to the ground state, more strained systems might provide a possible solution. For example, anullated ligands such as 2-pyrid-2’-yl-benzo[h]chinoline | 163 as N^N^C ligand or 9-pyrid-2’-yl-benzo[h]chinoline as N^C^N ligand might render excited state distortions more energetically demanding. This could be combined with very weakly π-accepting ligands to force Ru→(cyclometalating ligand) 3MLCT states as lowest triplet excited states, which might be considerably less distorted than Ru→(polypyridine ligand) excited states in cyclometalated complexes. Two such exemplary complexes are given in Figure 4.1. Additionally, bis(cyclometalation) could be a potential path to luminescent cyclometalated ruthenium complexes as suggested previously.114 Figure 4.1 Potentially luminescent ruthenium complexes with more restrained cyclometalating ligands. However, before synthetic efforts in any of the directions are undertaken, quantum chemical calculations can already provide valuable information about the energies and geometries of all excited states. The work presented herein has highlighted, that the predictive power of DFT calculation is sufficient to acquire a profound understanding of the photophysical properties of a cyclometalated polypyridine ruthenium complex without actually synthesizing it. This allows to scan for electronically suited structures or motives that then can be tackled synthetically. In conclusion, this work has provided insight into the photophysical properties of [Ru(N^N^N)(N^C^N)]+ complexes and given answers to the question why these complexes are so weakly emissive. This should lay the foundation for more in-depth studies in this field that might yield luminescent complexes | 165 5 REFERENCES 1 J. P. Paris and W. W. Brandt, J. Am. Chem. Soc., 1959, 81, 5001–5002. 2 G. D. Hager and G. A. Crosby, J. Am. Chem. Soc., 1975, 97, 7031–7037. 3 G. D. Hager, R. J. Watts and G. A. Crosby, J. Am. Chem. Soc., 1975, 97, 7037–7042. 4 J. van Houten and R. J. Watts, Inorg. Chem., 1978, 17, 3381–3385. 5 M. Maestri, F. Bolletta, L. Moggi, V. Balzani, M. S. Henry and M. Z. Hoffman, J. Am. Chem. Soc., 1978, 100, 2694–2701. 6 B. Durham, J. V. Caspar, J. K. Nagle and T. J. Meyer, J. Am. Chem. Soc., 1982, 104, 4803– 4810. 7 J. V. Caspar, E. M. Kober, B. P. Sullivan and T. J. Meyer, J. Am. Chem. Soc., 1982, 104, 630– 632. 8 F. Barigelletti, A. Juris, V. Balzani, P. Belser and A. von Zelewsky, Inorg. Chem., 1983, 22, 3335–3339. 9 C. Chiorboli, M. T. Indelli and F. Scandola, Top. Curr. Chem., 2005, 257, 63–102. 10 S. Campagna, G. Denti, S. Serroni, M. Ciano and V. Balzani, Inorg. Chem., 1991, 30, 3728– 3732. 11 K. Heinze, K. Hempel and M. Beckmann, Eur. J. Inorg. Chem., 2006, 2006, 2040–2050. 12 J. M. Calvert, J. V. Caspar, R. A. Binstead, T. D. Westmoreland and T. J. Meyer, J. Am. Chem. Soc., 1982, 104, 6620–6627. 13 C. Creutz and H. Taube, J. Am. Chem. Soc., 1969, 91, 3988–3989. 14 W. Kaim, A. Klein and M. Glöckle, Acc. Chem. Res., 2000, 33, 755–763. 15 W. Kaim and G. K. Lahiri, Angew. Chem., 2007, 119, 1808–1828. Angew. Chem. Int. Ed., 2007, 46, 1778–1796. 16 S. Campagna, F. Puntoriero, F. Nastasi, G. Bergamini and V. Balzani, Top. Curr. Chem., 2007, 280, 117–214. 17 H. D. Abruna, A. Y. Teng, G. J. Samuels and T. J. Meyer, J. Am. Chem. Soc., 1979, 101, 6745– 6746. 18 J.-M. Lehn and R. Ziessel, Proc. Natl. Acad. Sci., 1982, 79, 701–704. 19 J. Hawecker, J.-M. Lehn and R. Ziessel, J. Chem. Soc., Chem. Commun., 1983, 536. 20 H. Ozawa, M. Haga and K. Sakai, J. Am. Chem. Soc., 2006, 128, 4926–4927. 21 S. Ji, W. Wu, W. Wu, P. Song, K. Han, Z. Wang, S. Liu, H. Guo and J. Zhao, J. Mater. Chem., 2010, 20, 1953–1963. 22 B. O'Regan and M. Grätzel, Nature, 1991, 353, 737–740. 23 M. Grätzel, J. Photochem. Photobiol. C, 2003, 4, 145–153. 24 M. A. Green, Solar Energy, 2004, 76, 3–8. 166 | 5 REFERENCES 25 M. K. Nazeeruddin, P. Péchy, T. Renouard, S. M. Zakeeruddin, R. Humphry-Baker, P. Comte, P. Liska, Le Cevey, E. Costa, V. Shklover, L. Spiccia, G. B. Deacon, C. A. Bignozzi and M. Grätzel, J. Am. Chem. Soc., 2001, 123, 1613–1624. 26 M. K. Nazeeruddin, A. Kay, I. Rodicio, R. Humphry-Baker, E. Mueller, P. Liska, N. Vlachopoulos and M. Graetzel, J. Am. Chem. Soc., 1993, 115, 6382–6390. 27 M. K. Nazeeruddin, S. M. Zakeeruddin, R. Humphry-Baker, M. Jirousek, P. Liska, N. Vlachopoulos, V. Shklover, C.-H. Fischer and M. Grätzel, Inorg. Chem., 1999, 38, 6298– 6305. 28 G. Sauvé, M. E. Cass, S. J. Doig, I. Lauermann, K. Pomykal and N. S. Lewis, J. Phys. Chem. B, 2000, 104, 3488–3491. 29 T. Bessho, E. C. Constable, M. Graetzel, A. Hernandez Redondo, C. E. Housecroft, W. Kylberg, M. K. Nazeeruddin, M. Neuburger and S. Schaffner, Chem. Commun., 2008, 3717– 3719. 30 T. Kinoshita, J.-i. Fujisawa, J. Nakazaki, S. Uchida, T. Kubo and H. Segawa, J. Phys. Chem. Lett., 2012, 394–398. 31 C. E. Housecroft and E. C. Constable, Chem. Soc. Rev., 2015, 44, 8386–8398. 32 H. J. Bolink, L. Cappelli, E. Coronado and P. Gaviña, Inorg. Chem., 2005, 44, 5966–5968. 33 H. J. Bolink, L. Cappelli, E. Coronado, M. Grätzel and M. K. Nazeeruddin, J. Am. Chem. Soc., 2006, 128, 46–47. 34 M. Maestri, N. Armaroli, V. Balzani, E. C. Constable and A. M. W. C. Thompson, Inorg. Chem., 1995, 34, 2759–2767. 35 M. Abrahamsson, M. Jäger, T. Österman, L. Eriksson, P. Persson, H.-C. Becker, O. Johansson and L. Hammarström, J. Am. Chem. Soc., 2006, 128, 12616–12617. 36 F. Schramm, V. Meded, H. Fliegl, K. Fink, O. Fuhr, Z. Qu, W. Klopper, S. Finn, T. E. Keyes and M. Ruben, Inorg. Chem., 2009, 48, 5677–5684. 37 A. Breivogel, C. Förster and K. Heinze, Inorg. Chem., 2010, 49, 7052–7056. 38 P. Reveco, J. H. Medley, A. R. Garber, N. S. Bhacca and J. Selbin, Inorg. Chem., 1985, 24, 1096–1099. 39 E. C. Constable and J. M. Holmes, J. Organomet. Chem., 1986, 301, 203–208. 40 M. Beley, J. P. Collin, R. Louis, B. Metz and J. P. Sauvage, J. Am. Chem. Soc., 1991, 113, 8521–8522. 41 S. H. Wadman, J. M. Kroon, K. Bakker, M. Lutz, A. L. Spek, G. P. M. van Klink and G. van Koten, Chem. Commun., 2007, 1907–1909. 42 T. Bessho, E. Yoneda, J.-H. Yum, M. Guglielmi, I. Tavernelli, H. Imai, U. Rothlisberger, M. K. Nazeeruddin and M. Grätzel, J. Am. Chem. Soc., 2009, 131, 5930–5934. 43 I. M. Dixon, F. Alary and J.-L. Heully, Dalton Trans., 2010, 39, 10959–10966. 44 I. M. Dixon, F. Alary, M. Boggio-Pasqua and J.-L. Heully, Dalton Trans., 2015, 44, 13498– 13503. | 167 45 K. C. D. Robson, B. D. Koivisto, A. Yella, B. Sporinova, M. K. Nazeeruddin, T. Baumgartner, M. Grätzel and C. P. Berlinguette, Inorg. Chem., 2011, 50, 5494–5508. 46 P. G. Bomben, K. C. Robson, B. D. Koivisto and C. P. Berlinguette, Coord. Chem. Rev., 2012, 256, 1438–1450. 47 K. Kalyanasundaram, Coord. Chem. Rev., 1982, 46, 159–244. 48 J. N. Demas and D. G. Taylor, Inorg. Chem., 1979, 18, 3177–3179. 49 S. Yoon, P. Kukura, C. M. Stuart and R. A. Mathies, Mol. Phys., 2006, 104, 1275–1282. 50 S. J. Strickler and R. A. Berg, J. Chem. Phys., 1962, 37, 814–822. 51 G. S. Ming Tong, K. T. Chan, X. Chang and C.-M. Che, Chem. Sci., 2015, 6, 3026–3037. 52 D. G. Brown, N. Sanguantrakun, B. Schulze, U. S. Schubert and C. P. Berlinguette, J. Am. Chem. Soc., 2012, 134, 12354–12357. 53 R. Englman and J. Jortner, Mol. Phys., 1970, 18, 145–164. 54 B. S. Brunschwig and N. Sutin, Comments Inorg. Chem., 1987, 6, 209–235. 55 J. V. Caspar and T. J. Meyer, J. Phys. Chem., 1983, 87, 952–957. 56 T. E. Keyes, F. Weldon, E. Müller, P. Pechy, M. Grätzel and J. G. Vos, J. Chem. Soc., Dalton Trans., 1995, 2705–2706. 57 M. Schulz, J. Hirschmann, A. Draksharapu, G. Singh Bindra, S. Soman, A. Paul, R. Groarke, M. T. Pryce, S. Rau, W. R. Browne and J. G. Vos, Dalton Trans., 2011, 40, 10545–10552. 58 F. Barigelletti, P. Belser, A. von Zelewsky, A. Juris and V. Balzani, J. Phys. Chem., 1985, 89, 3680–3684. 59 J. R. Lakowicz, ed., Principles of Fluorescence Spectroscopy, Springer US, Boston, MA, 1983. 60 V. Balzani, G. Bergamini, S. Campagna and F. Puntoriero, Top. Curr. Chem., 2007, 280, 1– 36. 61 T. Förster, Ann. Phys., 1948, 437, 55–75. 62 D. L. Dexter, J. Chem. Phys., 1953, 21, 836–850. 63 R. A. Marcus, J. Chem. Phys., 1956, 24, 979–989. 64 R. A. Marcus, J. Chem. Phys., 1965, 43, 679–701. 65 R. Marcus and N. Sutin, Biochim. Biophys. Acta - Bioenergetics, 1985, 811, 265–322. 66 N. S. Hush, Prog. Inorg. Chem., 1967, 8, 391–444. 67 N. S. Hush, Coord. Chem. Rev., 1985, 64, 135–157. 68 J. T. Hupp and M. J. Weaver, J. Electroanal. Chem. Interfacial Electrochem., 1983, 152, 1– 14. 69 D. Rehm and A. Weller, Ber. Bunsenges. Phys. Chem., 1969, 73, 834–839. 70 O. Stern and M. Volmer, Physik. Z., 1919, 20, 183–188. 71 G. C. Allen and N. S. Hush, Prog. Inorg. Chem., 1967, 8, 357–389. 72 M. B. Robin and P. Day, Adv. Inorg. Chem., 1968, 10, 247–422. 73 B. S. Brunschwig, C. Creutz and N. Sutin, Chem. Soc. Rev., 2002, 31, 168–184. 74 B. S. Brunschwig and N. Sutin, Coord. Chem. Rev., 1999, 187, 233–254. 168 | 5 REFERENCES 75 W. Kaim and B. Sarkar, Coord. Chem. Rev., 2007, 251, 584–594. 76 R. F. Winter, Organometallics, 2014, 33, 4517–4536. 77 S. Trofimenko, Inorg. Chem., 1973, 12, 1215–1221. 78 M. I. Bruce, Angew. Chem., 1977, 89, 75–89. Angew. Chem. Int. Ed., 1977, 16, 73–86. 79 Y. Chi and P.-T. Chou, Chem. Soc. Rev., 2010, 39, 638–655. 80 V. W.-W. Yam and K. M.-C. Wong, Chem. Commun., 2011, 47, 11579–11592. 81 V. Balzani and S. Campagna, Top. Curr. Chem., 2007, 280, 1–273. 82 V. Balzani and S. Campagna, Top. Curr. Chem., 2007, 281, 1–327. 83 E. C. Constable and M. J. Hannon, Inorg. Chim. Acta, 1993, 211, 101–110. 84 E. C. Constable, A. M. W. C. Thompson and S. Greulich, J. Chem. Soc., Chem. Commun., 1993, 1444–1446. 85 S. Sprouse, K. A. King, P. J. Spellane and R. J. Watts, J. Am. Chem. Soc., 1984, 106, 6647– 6653. 86 Y. Ohsawa, S. Sprouse, K. A. King, M. K. DeArmond, K. W. Hanck and R. J. Watts, J. Phys. Chem., 1987, 91, 1047–1054. 87 P. G. Bomben, K. C. D. Robson, P. A. Sedach and C. P. Berlinguette, Inorg. Chem., 2009, 48, 9631–9643. 88 B. D. Koivisto, K. C. D. Robson and C. P. Berlinguette, Inorg. Chem., 2009, 48, 9644–9652. 89 M. K. Nazeeruddin, F. de Angelis, S. Fantacci, A. Selloni, G. Viscardi, P. Liska, S. Ito, B. Takeru and M. Grätzel, J. Am. Chem. Soc., 2005, 127, 16835–16847. 90 W. Zeng, Y. Cao, Y. Bai, Y. Wang, Y. Shi, M. Zhang, F. Wang, C. Pan and P. Wang, Chem. Mater., 2010, 22, 1915–1925. 91 S. Mathew, A. Yella, P. Gao, R. Humphry-Baker, B. F. E. Curchod, N. Ashari-Astani, I. Tavernelli, U. Rothlisberger, M. K. Nazeeruddin and M. Grätzel, Nat. Chem., 2014, 6, 242– 247. 92 A. Yella, H.-W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran, M. K. Nazeeruddin, E. W.-G. Diau, C.- Y. Yeh, S. M. Zakeeruddin and M. Grätzel, Science, 2011, 334, 629–634. 93 D. Siebler, M. Linseis, T. Gasi, L. M. Carrella, R. F. Winter, C. Förster and K. Heinze, Chem. Eur. J., 2011, 17, 4540–4551. 94 D. Siebler, C. Forster and K. Heinze, Dalton Trans., 2011, 40, 3558–3575. 95 A. Breivogel, K. Hempel and K. Heinze, Inorg. Chim. Acta, 2011, 374, 152–162. 96 C. Kreitner, Diploma thesis, 2012. 97 S. H. Wadman, M. Lutz, D. M. Tooke, A. L. Spek, F. Hartl, R. W. A. Havenith, G. P. M. van Klink and G. van Koten, Inorg. Chem., 2009, 48, 1887–1900. 98 M. L. Muro-Small, J. E. Yarnell, C. E. McCusker and F. N. Castellano, Eur. J. Inorg. Chem., 2012, 2012, 4004–4011. 99 B. Schulze, D. Escudero, C. Friebe, R. Siebert, H. Görls, S. Sinn, M. Thomas, S. Mai, J. Popp, B. Dietzek, L. González and U. S. Schubert, Chem. Eur. J., 2012, 18, 4010–4025. | 169 100 C. D. Ertl, D. P. Ris, S. C. Meier, E. C. Constable, C. E. Housecroft, M. Neuburger and J. A. Zampese, Dalton Trans., 2015, 44, 1557–1570. 101 C.-J. Yao, H.-J. Nie, W.-W. Yang, J.-Y. Shao, J. Yao and Y.-W. Zhong, Chem. Eur. J., 2014, 20, 17466–17477. 102 A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652. 103 A. Vlček and S. Záliš, Coord. Chem. Rev., 2007, 251, 258–287. 104 T. Yanai, D. P. Tew and N. C. Handy, Chem. Phys. Lett., 2004, 393, 51–57. 105 J. Preiß, M. Jäger, S. Rau, B. Dietzek, J. Popp, T. Martínez and M. Presselt, ChemPhysChem, 2015, 16, 1395–1404. 106 C. Adamo and V. Barone, J. Chem. Phys., 1999, 110, 6158–6170. 107 A. Schäfer, H. Horn and R. Ahlrichs, J. Chem. Phys., 1992, 97, 2571–2577. 108 A. Schäfer, C. Huber and R. Ahlrichs, J. Chem. Phys., 1994, 100, 5829–5835. 109 F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys., 2005, 7, 3297–3305. 110 S. Sinnecker, A. Rajendran, A. Klamt, M. Diedenhofen and F. Neese, J. Phys. Chem. A, 2006, 110, 2235–2245. 111 D. Andrae, U. Häußermann, M. Dolg, H. Stoll and H. Preuß, Theoret. Chim. Acta, 1990, 77, 123–141. 112 E. van Lenthe, E. J. Baerends and J. G. Snijders, J. Chem. Phys., 1993, 99, 4597–4610. 113 S. Grimme, S. Ehrlich and L. Goerigk, J. Comput. Chem., 2011, 32, 1456–1465. 114 E. Y. Li, Y.-M. Cheng, C.-C. Hsu, P.-T. Chou, G.-H. Lee, I.-H. Lin, Y. Chi and C.-S. Liu, Inorg. Chem., 2006, 45, 8041–8051. | 171 6 APPENDIX 6.1 SUPPORTING INFORMATION TO 1.1: REDOX AND PHOTOCHEMISTRY OF BIS(TERPYRIDINE) RUTHENIUM(II) AMINO ACIDS AND THEIR AMIDE CONJUGATES – FROM UNDERSTANDING TO APPLICATIONS General procedures: Et2O was distilled from sodium, CH3CN and CH2Cl2 from CaH2 and THF from potassium under an argon atmosphere. TentaGel-Wang resin and Fmoc-Gly-OH were purchased from IRIS Biotech. DIC, PyBOP and HOBT were purchased from Fluka (DIC = N,N'- diisopropylcarbodiimide, PyBOP = benzotriazole-1-yloxy)tripyrrolidinophosphonium hexafluorophosphate, HOBT = 1-hydroxybenzotriazole). All other reagents were used without further treatment from commercial suppliers (Acros and Sigma-Aldrich). Microwave heating was performed in a Discover Benchmate Plus (CEM Synthesis) single-mode microwave cavity, producing continuous irradiation at 2.455 GHz with 100 W (maximum power). The temperature and irradiation power were monitored during the course of the reaction. NMR spectra were recorded on a Bruker Avance DRX 400 spectrometer at 400.31 MHz (1H) and 100.66 MHz (13C{1H}). All resonances are reported in ppm versus the solvent signal as an internal standard (CH 13CN ( H, δ = 1.94; 13C, δ = 1.24 ppm). Figure S1 shows the SPPS of [36](PF6)2 and [37](PF6)2 as well as the atom numbering for NMR signal assignment. IR spectra were recorded on a BioRad Excalibur FTS 3100 spectrometer as CsI disks. Electrochemical experiments were carried out with a Bio Logic SP- 50 voltammetric analyzer using platinum wires as counter- and working electrodes and 0.01 M Ag/AgNO3 as reference electrode. The measurements were performed with a scan rate of 50 – 333 mV s–1 for cyclic voltammetry experiments and 100 – 200 mV s–1 for square-wave voltammetry experiments using 0.1 M [n-Bu4N](PF6) as the supporting electrolyte and a 10–3 M solution of the sample in dry and degassed CH3CN. Potentials are referenced to the ferrocene/ferrocenium couple (E1/2 = 85 ± 5 mV under our experimental conditions). UV/Vis/near-IR spectra were recorded on a Varian Cary 5000 spectrometer using 1.0 cm cells (Hellma, Suprasil). Emission spectra were recorded on a Varian Cary Eclipse spectrometer. Quantum yields were determined by comparing the areas under the emission spectra on an energy scale (cm–1) recorded for optically matched solutions of the sample and the reference [Ru(bpy)3]Cl2 (Φ = 9.4% in deaerated CH3CN).1 ESI mass spectra were recorded on a Micromass Q-TOF-Ultima spectrometer. Elemental analyses were performed by the microanalytical laboratory of the chemical institutes of the University of Mainz. 1 K. Suzuki, A. Kobayashi, S. Kaneko, K. Takehira, T. Yoshihara, H. Ishida, Y. Shiina, S. Oishi, S. Tobita, Phys. Chem. Chem. Phys. 2009, 11, 9850–9860. 172 | 6 APPENDIX Figure S1 SPPS of complexes [36](PF6)2 and [37](PF6)2. Tentagel-Gly--Gly-Fmoc was synthesized according to a literature procedure.2 The progress of the reaction was monitored by treating a small portion of Tentagel-Gly--Gly-Fmoc (2 – 3 mg) with TFA (4 ml) and stirring for 30 min. The solvent was removed under reduced pressure, the product was re-dissolved in CH3CN (2 ml) and filtered. ESI mass spectra of the resulting solution showed signals for the Fmoc-protected tripeptide [H-Gly--Gly-Fmoc]2+ = [M]2+ at m/z = 481.6 (100%) [M]2+, 962.2 (29%) [M–H]+. Tentagel-Gly--Gly-Gly-Fmoc: Tentagel-Gly--Gly-Fmoc (0.167 mmol active centers, 1.0 equiv) was shaken in a piperidine/CH2Cl2 (1:5, 15 ml) mixture for 30 min. After filtration and washing with CH2Cl2 (3 × 15 ml) the residue was dried under reduced pressure. A solution of Fmoc- Gly-OH (209.5 mg, 0.705 mmol, 4.2 equiv), HOBT (118.5 mg, 0.877 mmol, 5.2 equiv) and DIC (0.15 ml, 0.97 mmol, 5.8 equiv) in DMF (20 ml) was prepared, stirred for 30 min and added. The mixture was shaken for 16 h, filtered and washed with DMF (3 × 15 ml) and with CH2Cl2 (3 × 15 ml). The 2 K. Heinze, K. Hempel, Chem. Eur. J. 2009, 15, 1346–1358. Section 6.1 | 173 dark red product was dried under reduced pressure. The progress of the reaction was monitored by treating a small portion of Tentagel-Gly--Gly-Gly-Fmoc (2 – 3 mg) with TFA (4 ml) and stirring for 30 min. The solvent was removed under reduced pressure, the product was re- dissolved in CH3CN (2 ml) and filtered. ESI mass spectra of the resulting solution showed signals for the Fmoc-protected tetrapeptide [H-Gly--Gly-Gly-Fmoc]2+ = [M]2+ at m/z = 510.1 (100%) [M]2+, 1019.3 (49%) [M–H]+. [36](PF6)2: Tentagel-Gly--Gly-Gly-Fmoc (0.167 mmol active centers, 1.0 equiv) was shaken in a piperidine/CH2Cl2 (1:5, 15 ml) mixture for 30 min. After filtration and washing with CH2Cl2 (3 × 15 ml) the residue was dried under reduced pressure. A solution of 7-diethylaminocumarin-3- carboxylic acid (104.1 mg, 0.398 mmol, 2.4 equiv), PyBOP (235.0 mg, 0.452 mmol, 2.7 equiv) and pyridine (0.20 ml, 2.5 mmol, 15 equiv) in CH2Cl2 (20 ml) was prepared, stirred for 30 min and added. The mixture was shaken for 40 h, filtered and washed with CH2Cl2 (6 × 15 ml). The red product was dried under reduced pressure. Trifluoroacetic acid (10 ml) was added, the mixture was shaken for 45 min, filtered and washed with CH3CN (2 × 10 ml). The solvent of the combined filtrates was removed under reduced pressure. The dark red powder was dissolved in CH3CN (2 ml). Water was added (10 ml) and the product precipitated upon adding an aqueous solution of [NH4](PF6) (194.4 mg, 1.19 mmol, 7.1 equiv, 3 ml H2O). After filtration the dark red product was washed with cold water and dried under reduced pressure. Yield: 140.0 mg (0.105 mmol, 63%). 1H NMR (CD3CN, 300 K):  = 9.93 (s, 1 H, NHb), 9.43 (t, 3JHH = 5.2 Hz, 1 H, NHd), 9.25 (s, 2 H, H2'), 9.08 (s, 2 H, H2), 8.69 (s, 1 H, H17), 8.59 (d, 3JHH = 8.0 Hz, 2 H, H5), 8.41 (d, 3JHH = 8.0 Hz, 2 H, H5'), 8.10 (t, 3JHH = 5.6 Hz, 1 H, NHa), 7.94 (m, 2 H, H6), 7.87 (m, 2 H, H6'), 7.57 (t, 3JHH = 5.2 Hz, 1 H, NHc), 7.47 (d, 3J = 5.2 Hz, 2 H, H8HH ), 7.32 (d, 3JHH = 5.2 Hz, 2 H, H8'), 7.18 (m, 2 H, H7), 7.12 (m, 2 H, H7'), 7.08 (m, 3J 15 12,14 3HH = 9.6 Hz, 1 H, H ), 6.55 (m, 2 H, H ), 4.32 (d, JHH = 5.6 Hz, 2 H, CH a2 ), 4.16 (d, 3JHH = 5.2 Hz, 2 H, CH b2 ), 4.13 (d, 3JHH = 5.2 Hz, 2 H, CH c 32 ), 3.44 (q, JHH = 7.0 Hz, 4 H, CH ethyl2 ), 1.15 ppm (t, 3JHH = 7.0 Hz, 6 H, CH3), no resonance for OH was observed. 13C{1H} NMR (CD3CN, 300 K):  = 171.4 (COb), 170.9 (COOH), 168.4 (COd), 166.2 (COc), 165.2 (COa), 163.3 (C10), 158.9 (C9), 158.8 (C4'), 158.7 (C4), 157.2 (C3), 156.0 (C3'), 154.1 (C16), 153.8 (C8'), 153.4 (C8), 148.9 (C17), 147.5 (C1'), 140.4 (C1), 139.2 (C6'), 139.0 (C6), 132.2 (C15), 128.6 (C7), 128.4 (C7'), 125.5 (C5), 125.3 (C5'), 122.2 (C2), 114.6 (C2'), 111.2 (C14), 109.9 (C11), 108.7 (C13), 97.1 (C12), 45.6 (CH Ethyl2 ), 45.1, 44.7 (CH b2 , CH c2 ), 42.3 (CH a2 ), 12.6 ppm (CH +3). MS (ESI ): m/z 520.6 (65%) [M – 2 PF 2+6] , 1186.3 (100%) [M – PF6]+. HR-MS (ESI+): m/z calcd for C51H45N 96 2+11O8 Ru : 517.6264; found: 517.6273; calcd for C 96 +51H45F6N11O8P Ru : 1180.2170; found: 1180.2195. IR (CsI): ν̃ = 3425 (br, m, OH, NH), 3101 (w, CH), 2963 (w, CH), 2930 (w, CH), 2874 (w, CH), 2858 (w, CH), 1699 (m, C=O), 1616 (s, C=N, amide I), 1580 (m, C=N, amide II), 1512 (s, C=N), 1475 (m), 1427 (m), 1354 (s), 1288 (m), 1232 (s), 1190 (m), 1165 (w), 1134 (m), 1096 (m), 1036 (w), 845 (vs, PF), 791 (m), 756 (w), 613 (m), 559 (m) cm–1. UV/Vis (CH3CN): abs(ε) = 490 (17100), 422 (33700), 309 (42200), 275 (56300), 236 (33500), 207 nm (62300 M–1cm–1). Emission (CH3CN, 295 K, exc = 490 nm): emiss = 668 nm. (CH3CN, λexc = 456 nm, 295 K): 0.12%. CV (CH3CN): E1/2 = +0.89 (1e, rev.), +0.75 (1e, rev.), –1.15 (1e, irrev.), –1.55 (2e, irrev.) V vs. FcH/FcH+. Elemental analysis calcd (%) for C51H45F12N11O8P2Ru·× 10 H2O: C 41.03, H 4.25, N 10.32; found: C 41.12, H 3.97, N 9.21. 174 | 6 APPENDIX [37](PF6)2: Tentagel-Gly--Gly-Gly-Fmoc (0.183 mmol reaction centers, 1.0 equiv) was shaken in a piperidine/CH2Cl2 (1:5, 15 ml) mixture for 30 min. After filtration and washing with CH2Cl2 (3 × 15 ml) the residue was dried under reduced pressure. A solution of acetyl chloride (0.10 ml, 1.4 mmol, 7.7 equiv) and pyridine (0.40 ml, 5.0 mmol, 27 equiv) in CH2Cl2 (15 ml) was added. The mixture was shaken for 12 h, filtered and washed with CH2Cl2 (5 × 15 ml). Trifluoroacetic acid (10 ml) was added, the mixture was shaken for 45 min, filtered and washed with CH3CN (2× 10 ml). The solvent of the filtrate was removed under reduced pressure. The dark red powder was dissolved in CH3CN (2 ml). Water was added (10 ml) and the product precipitated upon adding an aqueous solution of [NH4](PF6) (194.4 mg, 1.09 mmol, 5.9 equiv, 3 ml H2O). After filtration the dark red product was washed with cold water and dried under reduced pressure. The product was dissolved in CH3CN (2 ml), precipitated by the addition of Et2O (30 ml), filtered and dried at 90°C for 2 h. Yield: 125.5 mg (0.111 mmol, 61%). 1H NMR (CD CN, 300 K):  = 9.93 (s, 1 H, NHb3 ), 9.34 (s, 2 H, H2'), 9.09 (s, 2 H, H2), 8.59 (d, 3JHH = 8.0 Hz, 2 H, H5), 8.36 (d, 3JHH = 8.2 Hz, 2 H, H5'), 8.16 (t, 3JHH = 5.4 Hz, 1 H, NHa), 7.94 (m, 2 H, H6), 7.90 (m, 2 H, H6'), 7.51 (t, 3JHH = 6.1 Hz, 1 H, NHc), 7.47 (d, 3JHH = 5.5 Hz, 2 H, H8), 7.35 (t, 3JHH = 5.0 Hz, 1 H, NHd), 7.31 (d, 3JHH = 4.8 Hz, 2 H, H8'), 7.21 (m, 2 H, H7), 7.11 (m, 2 H, H7'), 4.32 (d, 3JHH = 5.4 Hz, 2 H, CH a2 ), 4.15 (d, 3JHH = 6.1 Hz, 2 H, CH b2 ), 3.84 (d, 3JHH = 5.0 Hz, 2 H, CH c), 2.18 ppm (s, 3 H, CH ). 13C{1H} NMR (CD CN, 300 K):  = 174.3 (COd2 3 3 ), 171.2; 171.1 (3 C) (COb, COc, COOH), 165.3 (COa), 158.8 (C4'), 158.7 (C4), 157.2 (C3), 156.0 (C3'), 153.8 (C8'), 153.4 (C8), 147.6 (C1'), 140.4 (C1), 139.2 (C6'), 139.0 (C6), 128.7 (C7), 128.4 (C7'), 125.5 (C5), 125.2 (C5'), 122.2 (C2), 114.3 (C2'), 45.1 (CH c2 ), 44.5 (CH b), 42.4 (CH a2 2 ), 23.2 ppm (CH3). MS (ESI+): m/z 420.1 (30%) [M – 2 PF 2+6] , 985.2 (100%) [M – PF6]+. HR-MS (ESI+): m/z calcd for C39H34F6N O 9610 6 Ru2+: 979.1381; found: 979.1395. IR (CsI): ν̃ = 3433 (br, m, OH, NH), 3341 (w, NH), 3102 (w, CH), 2931 (w, CH), 1661 (s, C=O), 1604 (m, C=N, amide I), 1529 (m, C=N, amide II), 1477 (m, C=N), 1358 (s), 1292 (m), 1232 (s), 1165 (w), 1097 (m), 1036 (w), 1028 (w), 841 (vs, PF), 791 (m), 764 (w), 756 (w), 613 (m), 559 (s), 405 (w) cm–1. UV/Vis (CH3CN): abs(ε) = 490 (16500), 308 (38900), 275 (49700), 235 (27200), 204 nm (41500 M–1cm–1). Emission (CH3CN, 295 K, exc = 490 nm): emiss = 667 nm. (CH3CN, exc = 447 nm, 295 K): 0.080%. CV (CH3CN): E1/2 = +0.88 (1e, rev.), –1.14 (1e, irrev.), – 1.49 (2e, irrev.) V vs. FcH/FcH+. Elemental analysis calcd (%) for C39H34F12N10O6P2Ru·2.5 HPF6: C 31.34, H 2.46, N 9.37; found: C 31.56, H 2.46, N 9.19. Section 6.2 | 175 6.2 SUPPORTING INFORMATION TO 3.1: DUAL EMISSION AND EXCITED-STATE MIXED-VALENCE IN A QUASI-SYMMETRIC DINUCLEAR RU−RU COMPLEX Synthesis of 8(PF6)2: [(HOOC-tpy)Ru(tpy-NH2)](PF6)2 1(PF6)2 (202 mg, 0.221 mmol) was dissolved in abs. acetonitrile (15 ml) and pentafluorophenol (53.4 mg, 0.290 mmol) and N,N’- diisopropylcarbodiimide (39.0 mg, 0.310 mmol) were added. After stirring at room temperature for 60 min the reaction mixture was concentrated to 5 ml under reduced pressure and the product triturated by addition of a solution of NH4PF6 (263 mg) in water (70 ml). The product was collected via filtration, washed with a small amount of water and diethyl ether and dried under reduced pressure to give [(C6F5OOC-tpy)Ru(tpy-NH2)](PF6)2 8(PF6)2 as red powder. Yield: 203.5 mg (0.188 mmol, 85%). Anal. Calc. C37H22F17N7O2P2Ru (1086.6)·1.5H2O: C, 40.05; H, 2.27; N, 8.84. Found: C, 40.06; H, 2.22; N, 8.84%. MS (ESI+): m/z (%) = 396.6 (83) [M-2PF6]2+, 938.1 (100) [M-PF6]+, 1479.1 (3) [3M-2PF 2+6] . HR-MS (ESI+, m/z): calcd. for C37H22F11N7O2PRu [M-PF ]+6 : 932.0473; found: 932.0459. 1H NMR (CD3CN): δ = 9.33 (s, 2H, H2), 8.67 (d, 3JHH = 8 Hz, 2H, H5), 8.28 (d, 3JHH = 8 Hz, 2H, H5’), 8.01 - 7.95 (m, 4H, H2’, H6), 7.86 (td, 3JHH = 8 Hz, 4JHH = 1.5 Hz, 2H, H6’), 7.60 (d, 3JHH = 5 Hz, 2H, H8), 7.35 - 7.28 (m, 2H, H7), 7.19 (d, 3JHH = 5 Hz, 2H, H8’), 7.07 - 7.02 (m, 2H, H7’), 6.04 (s, 2H, NH ). 13C{12 H} NMR (CD3CN): δ = 162.1 (s, COOC6F5), 159.2 (s, C4’), 158.6, 158.7 (2s, C4, C3), 154.7, 154.7 (2s, C1’, C3’), 153.9 (s, C8), 153.1 (s, C8’), 139.3 (s, C6’), 138.9 (s, C6), 131.3 (s, C1), 129.1 (s, C7), 128.0 (s, C7’), 125.8 (s, C5), 124.9 (s, C5’), 124.2 (s, C2), 109.7 (s, C2’), (C6F5-C-atoms not observed). 19F NMR (CD CN): δ = −73.3 (d, 13 JFP = 705 Hz, 12F, PF6), −154.8 (d, 3JFF = 17 Hz, 2F, o-F), −159.0 (t, 3JFF = 21 Hz, 1F, p-F), −163.8 (dd, 3JFF = 21, 17 Hz, 2F, m-F). Synthesis of 9(PF6)2: [(C6F5OOC-tpy)Ru(tpy-NH2)](PF6)2 8(PF6)2 (71 mg, 0.066 mmol) was dissolved in abs. acetonitrile (15 ml) and tert-butylamine (0.2 ml, 1.9 mmol, excess) was added. After stirring the mixture for 60 min at room temperature water (80 ml) was added. Addition of NH4PF6 (113 mg) in water (2 ml) precipitated the product which was filtered off, washed with small amounts of water and diethyl ether and dried under reduced pressure to give [(tBuHNOC- tpy)Ru(tpy-NH2)](PF6)2 9(PF6)2 as red powder. Yield: 31.6 mg (0.033 mmol, 50%). Anal. Calc. C35H32F12N8OP2Ru (971.7)·2.5H2O: C, 41.35; H, 3.67; N, 11.02. Found: C, 41.26; H, 3.34; N, 10.82%. MS (ESI+): m/z (%) = 341.1 (100) [M-2PF ]2+6 , 827.2 (57) [M-PF ]+6 . HR-MS (ESI+, m/z): calcd. for C35H32F6N8OPRu [M-PF6]+: 792.0788; found: 792.0782. 1H NMR (CD3CN): δ = 8.98 (s, 2H, H2), 8.60 (d, 3J = 8 Hz, 2H, H5), 8.26 (d, 3J = 8 Hz, 2H, H5’), 8.00 - 7.88 (m, 4H, H2’, H6), 7.84 (t, 3HH HH JHH = 7 Hz, 2H, H6’), 7.54 (d, 3J = 5 Hz, 2H, H8HH ), 7.28 (s, 1H, CONH), 7.27 - 7.20 (m, 2H, H7), 7.18 (d, 3JHH = 5 Hz, 2H, H8’) 7.07 - 6.99 (m, 2H, H7’), 5.97 (s, 2H, NH2), 1.61 (3, 9H, CH3). 13C{1H} NMR (CD3CN): δ = 164.5 (s, CONHtBu), 159.4 (s, C4), 159.1 (s, C4’), 157.5 (s, C3), 156.9 (s, C1’), 155.1 (s, C3’), 153.6 (s, C8), 153.1 (s, C8’), 141.9 (c, C1), 139.0 (s, C6), 138.7 (s, C6’), 128.7 (s, C7), 127.9 (s, C7’), 125.4 (s, C5), 124.8 (s, C5’), 122.2 (s, C2), 109.6 (s, C2’), 53.4 (s, C(CH3)3), 28.9 (s, C(CH3)3). 176 | 6 APPENDIX Figure S1. 1H NMR spectrum of 9(PF6)2 in CD3CN. Figure S2. 13C NMR spectrum of 9(PF6)2 in CD3CN. Section 6.2 | 177 Figure S3. 1H NMR spectrum of 7(PF6)2 in CD3CN. Figure S4. 13C NMR spectrum of 7(PF6)2 in CD3CN. 178 | 6 APPENDIX Figure S5. 19F NMR spectrum of 7(PF6)2 in CD3CN. Figure S6. 1H NMR spectrum of 8(PF6)2 in CD3CN. Section 6.2 | 179 Figure S7. 13C NMR spectrum of 8(PF6)2 in CD3CN. Figure S8. 19F NMR spectrum of 8(PF6)2 in CD3CN. 180 | 6 APPENDIX Figure S9. CH-HSQC spectrum of the aromatic region of 3(PF6)4 in CD3CN. Figure S10. CH-HMBC spectrum of the aromatic region of 3(PF6)4 in CD3CN. Section 6.2 | 181 Figure S11. Emission spectra of 34+ at room temperature in deaerated CH3CN with varying excitation wavelength. Figure S12. DFT (B3LYP, LANL2DZ) optimized geometric structures of 33+ (doublet, IEFPCM: acetonitrile) and 34+ (triplet, gasphase) including tpy-NHCO-tpy dihedral angles (°) and calculated spin densities. Contour value: 0.01, CH hydrogen atoms are omitted. Figure S13. Square wave voltammogram of 34+ in the presence of 0.1 M [nBu4N][PF6] as supporting electrolyte in acetonitrile. Black line shows the SWV in the absence of ferrocene, blue line in the presence of 2 eq. of ferrocene as internal standard for referencing. 182 | 6 APPENDIX Figure S14. UV/Vis absorption spectra of 34+ and 42+ in dry CH3CN upon titration with a solution of CoCp*2 in CH3CN (0 eq.  4 eq. for 34+, 0 eq.  2 eq. for 42+). Arrows indicate spectral changes. Dashed lines indicate spectra of 34+ and 42+, bold lines show deprotonated species 3-H3+ and 4-H+. Figure S15. 1H NMR spectra of 34+ in CD3CN upon deprotonation to 3-H3+ with substoichiometric amounts of CoCp*2. Blue frames highlight most significant spectral changes. Section 6.2 | 183 Figure S16. Stern-Volmer plots of the mono- and dinuclear complexes 42+ (blue) and 34+ (red) employing ferrocene derivatives as quenchers. Plots were obtained using complex concentrations of c = 210−5 mol l−1. Figure S17. Linearized Stern-Volmer plot I0/(I0−I) of the dinuclear complex 34+ employing N,N- dimethylaniline as quencher. Plot was obtained using a complex concentration of c = 210−5 mol l−1. 184 | 6 APPENDIX 6.3 SUPPORTING INFORMATION TO 3.2: UNDERSTANDING THE EXCITED STATE BEHAVIOR OF CYCLOMETALATED BIS(TRIDENTATE)RUTHENIUM(II) COMPLEXES: A COMBINED EXPERIMENTAL AND THEORETICAL STUDY Synthesis of ethyl 3,5-di-(pyridin-2-yl)benzoate L3: Ethyl 3,5-dibromobenzoate LD (1.21 g, 3.93 mmol, 1 eq.) was dissolved under argon in abs. toluene (50 ml) and tri-n-butyl-(pyridin-2- yl)stannane1 (4.94 g, 13.4 mmol, 3.4 eq), Pd(PPh3)4 (0.17 g, 0.15 mmol, 4 mol%) and lithium chloride (1.39 g, 8.3 eq) were added. The reaction mixture was heated to reflux for 23 h. After cooling to room temperature the reaction was quenched with aqueous sodium hydroxide solution (1 M, 50 ml). The mixture was extracted with ethyl acetate (3100 ml) and the combined organic phases were dried over magnesium sulfate before evaporation of the solvent under reduced pressure. The crude product was purified by column chromatography on silica gel (diethyl ether/hexanes 1:2, after elution of organotin impurities: diethyl ether/hexanes 2:1). Ethyl 3,5-di- (pyridin-2-yl)benzoate was obtained as colorless solid. Yield: 767 mg (2.52 mmol, 64 %). Anal. Calc. C19H16N2O2 (304.34): C, 74.98; H, 5.30; N, 9.20. Found: C, 74.66; H, 5.64; N, 9.12. MS(FD+): m/z (%) = 304.2 (100) [M]+. 1H NMR (CD2Cl2): δ [ppm] = 8.96 (s, 1H, H9), 8.74 (d, 3JHH = 4 Hz, 2H, H8), 8.70 (s, 2H, H2), 7.93 (d, 3JHH = 8 Hz, 2H, H5), 7.83 (dd, 3JHH = 4 Hz, 3JHH = 11 Hz, 2H, H6), 7.31 (m, 2H, H7), 4.44 (q, 3J 11 3HH = 7 Hz, 2H, H ), 1.45 (t, JHH = 7 Hz, 3H, H12). 13C{1H} NMR (CD3CN): δ [ppm] = 166.8 (s, C10), 156.5 (s, C4), 150.4 (s, C8), 140.8 (s, C3), 137.4 (s, C6), 132.3 (s, C1), 130.0 (s, C9), 128.6 (s, C2), 123.3 (s, C7), 121.1 (s, C5), 61.8 (s, C11), 14.8 (s, C12). Characterization of [Ru(dpb-COOEt)(tpy-NH2)](PF6) 3(PF6) Anal. Calc. for C34H27F6N6O2PRu (797.7): C, 51.20; H, 3.41; N, 10.54. Found: C, 51.49; H, 3.48; N, 10.22. MS(ESI+): m/z (%) = 653.1 (100) [M-PF6]+. HR-MS(ESI+, m/z): Calcd. for C34H27N6O2Ru [M- PF +6] : 647.1271; Found: 647.1273. 1H NMR (CD3CN): δ [ppm] = 8.81 (s, 2H, H2B), 8.25 (d, 3JHH = 8 Hz, 2H, H5B), 8.19 (d, 3JHH = 8 Hz, 2H, H5A), 8.06 (s, 2H, H2A), 7.69 – 7.55 (m, 4H, H6A, H6B), 7.34 (d, 3JHH = 5 Hz, 2H, H8B), 6.90 (d, 3JHH = 5 Hz, 2H, H8A), 6.83 – 6.72 (m, 4H, H7A, H7B), 5.92 (s, 2H, NH2), 4.50 (q, 3JHH = 7 Hz, 2H, H11), 1.51 (t, 3J 12 13HH = 7 Hz, 3H, H ). C{1H} NMR (CD3CN): δ [ppm] = 235.6 (s, C9B), 169.2 (s, C4B), 168.8 (s, C10), 160.3 (s, C4A), 155.7 (s, C8A), 155.3 (s, C1A), 153.4 (s, C3A), 152.6 (s, C8B), 143.9 (s, C3B), 136.2 (s, C6B), 136.1 (s, C6A), 126.9 (s, C7A), 124.3 (s, C2B), 124.0 (s, C5A), 122.9 (s, C7B), 121.8 (s, C1B), 120.6 (s, C5B), 108.3 (s, C2A), 61.4 (s, C11), 15.0 (s, C12). 1 Bolm, C.; Ewald, M.; Felder, M.; Schlingloff, G. Chem. Ber. 1992, 125, 1169–1190. Section 6.3 | 185 Ligand syntheses The synthesis of ethyl dipyridylbenzoate L3 was carried out adopting the literature-known synthesis of the analogous methyl ester.2,3 For the synthesis of the amino acid derivatives 1+ and 2+ synthetic procedures to the amino- substituted ligands L1 and L2 needed to be developed. In order to minimize the amount of undesired side reactions during the complexation we decided to protect the rather reactive amino functions by acetyl groups. The introduction of acetamide via palladium-catalyzed cross-coupling reactions has only been shown to occur with rather elaborate catalyst/ligand systems4,5,6 and is highly dependent on the type of leaving group and the electronics of the aromatic system.7 Screening experiments with one of Buchwald’s amidation precatalysts ([Pd] 82) showed reasonable reactivity in the reaction of 1-bromo-3,5-dipyrid-2-ylbenzene and acetamide in refluxing toluene. With XantPhos as chelating ligand and sodium tert-butoxide as supporting base N-acetyl-3,5-di- (pyridin-2-yl)aniline L1 was isolated in 78 % yield. Remarkably, under nearly the same conditions 4’-chloroterpyridine could be successfully reacted with acetamide in xylenes. Slightly higher reaction temperatures (130 °C instead of 110 °C) and longer reaction times (20 h instead of 8 h) gave N-acetyl-4’-amino-2,2’:6’,2’’-terpyridine L2 in essentially identical yields (76 %). Repeatedly, 4’-amino-2,2’:6’,2’’-terpyridine was isolated as major byproduct in this reaction because the terpyridine fragment substantially increases the nucleophilicity of the carbonyl carbon and facilitates hydrolysis under the basic reaction conditions. The byproduct could eventually be transformed to the desired product by treatment with acetylchloride in refluxing dichloromethane (see Procedure b). Both N-acetylated ligands L1 and L2 were fully characterized by FD mass spectra (Figure S1) and NMR spectroscopy (Figures S4-S7) and their purity was confirmed by elemental analysis (see Experimental Section). 2 Chen, L. S.; Chen, G. J.; Tamborski, C. J. Organomet. Chem. 1981, 215, 281–291. 3 Williams, J. A. G.; Beeby, A.; Davies, E. S.; Weinstein, J. A.; Wilson, C. Inorg. Chem. 2003, 42, 8609– 8611. 4 Yin, J.; Buchwald, S. L. J. Am. Chem. Soc. 2002, 124, 6043–6048. 5 Fors, B. P.; Dooleweerdt, K.; Zeng, Q.; Buchwald, S. L. Tetrahedron 2009, 65, 6576–6583 6 Crawford, S. M.; Lavery, C. B.; Stradiotto, M. Chem. Eur. J. 2013, 19, 16760–16771. 7 Surry, D. S.; Buchwald, S. L. Chem. Sci. 2010, 2, 27–50. 8 Bruno, N. C.; Tudge, M. T.; Buchwald, S. L. Chem. Sci. 2013, 4, 916–920. 186 | 6 APPENDIX Figure S1 FD mass spectra of L1, L2 and L3 in CH2Cl2. Figure S2 1H NMR spectrum (400 MHz, CD2Cl2) of L3. Section 6.3 | 187 Figure S3 13C NMR spectrum (100 MHz, CD2Cl2) of L3. Figure S4 1H NMR spectrum (400 MHz, CD2Cl2) of L1. 188 | 6 APPENDIX Figure S5 13C NMR spectrum (100 MHz, CD2Cl2) of L1. Figure S6 1H NMR spectrum (400 MHz, 0.5 ml CD2Cl2 + 0.1 ml d6-DMSO) of L2. Section 6.3 | 189 Figure S7 13C NMR spectrum (100 MHz, 0.5 ml CD2Cl2 + 0.1 ml d6-DMSO) of L2. Figure S8 ESI mass spectra of 1(PF6) (top, blue) and 2(PF6) (bottom, red). Inset shows isotope pattern of the most intense peak [M-PF +6] including calculated mass distribution for C36H29N6O3Ru. 190 | 6 APPENDIX Figure S9 1H NMR spectrum (400 MHz, CD3CN) of 1(PF6). Figure S10 13C NMR spectrum (100 MHz, CD3CN) of 1(PF6). Section 6.3 | 191 Figure S11 1H COSY spectrum (400 MHz, CD3CN) of 1(PF6). Figure S12 1H-13C HSQC spectrum (CD3CN) of 1(PF6). 192 | 6 APPENDIX Figure S13 1H-13C HMBC spectrum (CD3CN) of 1(PF6). Figure S14 1H NMR spectrum (400 MHz, CD3CN) of 2(PF6). Section 6.3 | 193 Figure S15 13C NMR spectrum (100 MHz, CD3CN) of 2(PF6). Figure S16 1H COSY spectrum (400 MHz, CD3CN) of 2(PF6). 194 | 6 APPENDIX Figure S17 1H-13C HSQC spectrum (CD3CN) of 2(PF6). Figure S18 1H-13C HMBC spectrum (CD3CN) of 2(PF6). Section 6.3 | 195 Figure S19 1H NMR spectrum (400 MHz, CD3CN) of 3(PF6). Figure S20 13C NMR spectrum (100 MHz, CD3CN) of 3(PF6). 196 | 6 APPENDIX Figure S21 IR spectra of 1(PF6) and 2(PF6) in the solid state (KBr disk). Section 6.3 | 197 Figure S22 MO diagram of 1+ and 2+ generated at the B3LYP, def2-TZVP, DKH, COSMO(acetonitrile) level of theory. Orbitals are plotted at a contour value of 0.07. Hydrogen atoms are omitted for clarity. 198 | 6 APPENDIX Figure S23 Comparison of experimental (RT, acetonitrile solution) and calculated absorption spectra of 1+ using different functionals. Figure S24 Comparison of experimental (RT, acetonitrile solution) and calculated absorption spectra of 2+ using different functionals. Section 6.3 | 199 Table S1 First ten TD-DFT excitations of 1+ and all further with oscillator strengths f > 0.01 (B3LYP, def2-TZVP, DKH, COSMO(acetonitrile)) and corresponding electronic difference densities ES − GS at a contour value of 0.005 (purple lobes indicate loss, orange lobes show increase of electron density upon excitation). Hydrogen atoms are omitted for clarity. State 1: E = 12104 cm−1, λ = 826 nm, f = State 2: E = 15095 cm−1, λ = 663 nm, f = 1.110−5 5.510−3 State 3: E = 15151 cm−1, λ = 660 nm, f = State 4: E = 18194 cm−1, λ = 550 nm, f = 0.030 6.810−4 State 5: E = 19150 cm−1, λ = 522 nm, f = 0.040 State 6: E = 20914 cm−1, λ = 478 nm, f = 0.091 State 7: E = 21962 cm−1, λ = 455 nm, f = 0.213 State 8: E = 21843 cm−1, λ = 457 nm, f = 2.610−3 State 9: E = 22623 cm−1, λ = 442 nm, f = 0.022 State 10: E = 21505 cm−1, λ = 465 nm, f = 1.310−4 State 13: E = 24838 cm−1, λ = 403 nm, f = 0.112 State 14: E = 25216 cm−1, λ = 397 nm, f = 0.010 200 | 6 APPENDIX State 17: E = 25476 cm−1, λ = 393 nm, f = 0.016 State 18: E = 25689 cm−1, λ = 389 nm, f = 0.034 State 19: E = 26158 cm−1, λ = 382 nm, f = 0.089 State 21: E = 27068 cm−1, λ = 369 nm, f = 0.034 State 22: E = 28129 cm−1, λ = 356 nm, f = 0.063 State 23: E = 28520 cm−1, λ = 351 nm, f = 0.034 State 24: E = 27202 cm−1, λ = 368 nm, f = 0.011 State 26: E = 29117 cm−1, λ = 343 nm, f = 0.030 State 28: E = 28269 cm−1, λ = 354 nm, f = 0.010 State 29: E = 28688 cm−1, λ = 349 nm, f = 0.037 State 33: E = 31513 cm−1, λ = 317 nm, f = 0.313 State 34: E = 31238 cm−1, λ = 320 nm, f = 0.077 State 36: E = 32539 cm−1, λ = 307 nm, f = 0.142 State 40: E = 29117 cm−1, λ = 291 nm, f = 0.027 Section 6.3 | 201 State 42: E = 34484 cm−1, λ = 290 nm, f = 0.215 State 43: E = 32391 cm−1, λ = 309 nm, f = 0.011 State 46: E = 35609 cm−1, λ = 281 nm, f = 0.054 State 48: E = 35420 cm−1, λ = 282 nm, f = 0.058 State 49: E = 36342 cm−1, λ = 275 nm, f = 0.326 202 | 6 APPENDIX Table S2 First ten TD-DFT excitations of 2+ and all further with oscillator strengths f > 0.01 (B3LYP, def2-TZVP, DKH, COSMO(acetonitrile)) and corresponding electronic difference densities ex − GS at a contour value of 0.005 (purple lobes indicate loss, orange lobes show increase of electron density upon excitation). Hydrogen atoms are omitted for clarity. State 1: E = 15955 cm−1, λ = 627 nm, f = State 2: E = 17378 cm−1, λ = 575 nm, f = 7.110−5 5.010−3 State 3: E = 16776 cm−1, λ = 596 nm, f = State 4: E = 18332 cm−1, λ = 546 nm, f = 0.014 8.410−4 State 5: E = 18593 cm−1, λ = 538 nm, f = 0.060 State 6: E = 20657 cm−1, λ = 484 nm, f = 5.110−3 State 7: E = 21675 cm−1, λ = 461 nm, f = 0.170 State 8: E = 21681 cm−1, λ = 461 nm, f = 0.096 State 9: E = 22130 cm−1, λ = 452 nm, f = State 10: E = 22631 cm−1, λ = 442 nm, f = 5.110−5 6.410−3 State 11: E = 24120 cm−1, λ = 415 nm, f = 0.111 State 16: E = 26366 cm−1, λ = 379 nm, f = 0.026 Section 6.3 | 203 State 17: E = 27656 cm−1, λ = 362 nm, f = 0.100 State 18: E = 27600 cm−1, λ = 362 nm, f = 0.049 State 24: E = 29114 cm−1, λ = 344 nm, f = 0.191 State 25: E = 28016 cm−1, λ = 357 nm, f = 0.039 State 27: E = 29771 cm−1, λ = 336 nm, f = 0.014 State 28: E = 29764 cm−1, λ = 336 nm, f = 0.012 State 32: E = 30746 cm−1, λ = 325 nm, f = 0.138 State 36: E = 30849 cm−1, λ = 324 nm, f = 0.018 State 38: E = 32940 cm−1, λ = 304 nm, f = 0.318 State 43: E = 35975 cm−1, λ = 278 nm, f = 0.082 State 44: E = 36195 cm−1, λ = 276 nm, f = 0.420 State 45: E = 34743 cm−1, λ = 288 nm, f = 0.030 204 | 6 APPENDIX State 46: E = 36077 cm−1, λ = 277 nm, f = 0.020 State 47: E = 36739 cm−1, λ = 272 nm, f = 0.445 State 48: E = 36125 cm−1, λ = 277 nm, f = 0.079 State 50: E = 36720 cm−1, λ = 272 nm, f = 0.158 Section 6.3 | 205 Figure S25 Spectral decomposition of the absorption spectra of 1(PF6) (top, black) and 2(PF6) (bottom, black) in acetonitrile solution between 10000 and 30000 cm-1 into separate absorption bands (light blue for 1(PF6) and light red for 2(PF6)). Sum of the individual contributions is shown as blue and red curve, respectively. Fit parameters are given in Tables 1 and 2, respectively. Figure S26 Emission spectra of 1(PF6) (top, blue) and 2(PF6) (bottom, red) in deaerated acetonitrile at room temperature. 206 | 6 APPENDIX Figure S27 Emission spectrum of 2(PF6) at 77 K in a frozen butyronitrile glass. Figure S28 Cyclic and square-wave voltammograms of 1(PF6) (c = 1 mM) in acetonitrile with 0.1 M [NBu4][PF6] as supporting electrolyte. All potentials are referenced against the FcH/FcH+ couple. Blue insets show the individual redox waves. Asterisks indicate reoxidation waves of follow-up products after three subsequent reduction steps. Figure S29 Cyclic and square-wave voltammograms of 2(PF6) (c = 1 mM) in acetonitrile with 0.1 M [NBu4][PF6] as supporting electrolyte. All potentials are referenced against the FcH/FcH+ couple. Red insets show the individual redox waves. Section 6.3 | 207 Figure S30 DFT calculated spin density (B3LYP/def2-TZVP/DKH/COSMO(acetonitrile), contour value: 0.01) of 12+ and experimental X-band EPR spectra (ν ≈ 9.4 GHz) obtained from frozen acetonitrile solutions of 12+ (c = 5 mM) in situ generated with [N(C6H4-4-Br)3][SbCl6]. CH hydrogen atoms are omitted for clarity. Figure S31 DFT calculated spin density (B3LYP/def2-TZVP/DKH/COSMO(acetonitrile), contour value: 0.01) of 22+. CH hydrogen atoms are omitted for clarity. 208 | 6 APPENDIX Figure S32 DFT calculated spin densities (B3LYP/def2-SVP/DKH/COSMO(acetonitrile), contour value: 0.01) of 13+ (top) and 23+ (bottom). Significant Mulliken spin density populations ( > 0.04) are given at the respective atoms. CH hydrogen atoms are omitted for clarity. Section 6.4 | 209 6.4 SUPPORTING INFORMATION TO 3.3: THE PHOTOCHEMISTRY OF MONO- AND DINUCLEAR CYCLOMETALATED BIS(TRIDENTATE)RUTHENIUM(II) COMPLEXES: DUAL EXCITED STATE DEACTIVATION AND DUAL EMISSION Figure S1 1H NMR spectrum (400 MHz) of 1(PF6) in CD3CN. 210 | 6 APPENDIX Figure S2 13C NMR spectrum (100 MHz) of 1(PF6) in CD3CN. Figure S3 1H NMR spectrum (400 MHz) of 2(PF6) in CD3CN. Section 6.4 | 211 Figure S4 13C NMR spectrum (100 MHz) of 2(PF6) in CD3CN. Figure S5 1H NMR spectrum (400 MHz) of 3(PF6) in CD3CN. 212 | 6 APPENDIX Figure S6 13C NMR spectrum (100 MHz) of 3(PF6) in CD3CN. Figure S7 1H NMR spectrum (400 MHz) of 4(PF6) in CD3CN. Section 6.4 | 213 Figure S8 13C NMR spectrum (100 MHz) of 4(PF6) in CD3CN. Figure S9 1H NMR spectrum (400 MHz) of 5(PF6) in CD3CN. 214 | 6 APPENDIX Figure S10 13C NMR spectrum (100 MHz) of 5(PF6) in CD3CN. Figure S11 1H NMR spectrum (400 MHz) of 6(PF6) in CD3CN. Section 6.4 | 215 Figure S12 13C NMR spectrum (100 MHz) of 6(PF6) in CD3CN. Figure S13 Aromatic region of the 1H-13C HSQC NMR spectrum of 6(PF6) in CD3CN. 216 | 6 APPENDIX Figure S14 Aromatic region of the 1H-13C HMBC NMR spectrum of 6(PF6) in CD3CN. Section 6.4 | 217 Figure S15 ESI mass spectra of a) 1(PF6), b) 2(PF6), c) 3(PF6), d) 4(PF6), e) 5(PF6), and f) 6(PF6) in CH3CN. Insets show experimental and calculated isotope pattern of the most intense peak. 218 | 6 APPENDIX Figure S16 IR spectra of complexes a) 1(PF6), b) 2(PF6), c) 3(PF6), d) 4(PF6) and e) 6(PF6)2 in the solid state (KBr disk). Section 6.4 | 219 Table S1 Selected molecular orbitals of 1+, 2+, 3+, and 4+ (B3LYP, def2-SV(P), ZORA, COSMO(acetonitrile)) including orbital number (energy E in eV) (contour value 0.07). Hydrogen atoms are omitted for clarity. 1+ 2+ 3+ 4+ MO162 (−1.50) MO151 (−1.45) MO162 (−1.57) MO158 (−1.58) MO161 (−1.82) MO150 (−1.74) MO161 (−1.86) MO157 (−1.87) MO160 (−2.23) MO149 (−2.20) MO160 (−2.27) MO156 (−2.28) MO159 (−2.27) MO148 (−2.23) MO159 (−2.33) MO155 (−2.34) MO158 (−4.99) MO147 (−4.71) MO158 (−5.31) MO154 (−5.33) MO157 (−5.35) MO146 (−5.27) MO157 (−5.47) MO153 (−5.48) HOMO-1 HOMO LUMO LUMO+1 LUMO+2 LUMO+3 220 | 6 APPENDIX MO156 (−5.43) MO145 (−5.36) MO156 (−5.52) MO152 (−5.53) Table S2 DFT calculated spin densities of complexes 1+ – 4+ after single and double oxidation as well as single and double reduction (B3LYP, def2-SV(P), ZORA, COSMO(acetonitrile); contour value: 0.01). charg 2+ 3+ 0 1− e 1 2 3 4 HOMO-2 Section 6.4 | 221 63+ 64+ 61+ 60 Figure S17 DFT calculated spin densities of the singly and doubly oxidized complexes 63+ and 64+ as well as the singly and doubly reduced complexes 6+ and 60 (B3LYP, def2-SV(P), ZORA, COSMO(acetonitrile); contour value: 0.01). Table S3 Selected molecular orbitals of 62+ (B3LYP, def2-SV(P), ZORA, COSMO(acetonitrile)) including orbital number (energy E in eV) (contour value 0.07). Hydrogen atoms are omitted for clarity. MO304 (−1.54) MO297 (−2.36) MO303 (−1.61) MO296 (−5.00) MO302 (−1.86) MO295 (−5.36) LUMO+5 LUMO+6 LUMO+7 HOMO-1 HOMO LUMO 222 | 6 APPENDIX MO301 (−1.92) MO294 (−5.39) MO300 (−2.26) MO293 (−5.46) MO299 (−2.31) MO292 (−5.49) MO291 (−5.55) MO298 (−2.31) Table S4 Selected vertical TD-DFT transitions of 1+ (B3LYP, def2-SV(P), ZORA, COSMO(acetonitrile)) sorted by their energy including difference density plots |ΨES|2 − |Ψ 2GS| (contour value 0.005, purple: depletion, orange: gain in electron density). Hydrogen atoms are omitted for clarity. ?̃? / λ / 2 2 ?̃? / λ / −1 fosc |ΨES| − |ΨGS| −1 fosc |Ψ | 2 ES − |ΨGS|2 cm nm cm nm 1 13967 716 4.30∙10−8 10 22891 437 0.00001 2 14972 668 0.00066 13 24853 402 0.09267 LUMO+1 LUMO+2 LUMO+3 LUMO+4 HOMO-5 HOMO-4 HOMO-3 HOMO-2 Section 6.4 | 223 3 17170 582 0.00723 14 25051 399 0.03568 4 18458 542 0.00512 19 28073 356 0.03405 5 18728 534 0.05845 20 28080 356 0.03646 6 20594 486 0.09011 21 28236 354 0.03297 7 21029 476 0.00695 23 28339 353 0.04613 8 21890 457 0.14519 33 32361 309 0.01697 9 22432 446 0.01191 34 32565 307 0.02433 Table S5 Selected vertical TD-DFT transitions of 2+ (B3LYP, def2-SV(P), ZORA, COSMO(acetonitrile)) sorted by their energy including difference density plots |Ψ |2 − |Ψ |2ES GS (contour value 0.005, purple: depletion, orange: gain in electron density). Hydrogen atoms are omitted for clarity. ?̃? / λ / ?̃? / λ / −1 fosc |Ψ 2 ES| − |Ψ |2GS −1 f |Ψ | 2 osc ES − |Ψ 2GS| cm nm cm nm 1 12179 821 4.57∙10−7 15 24593 407 0.02282 2 12929 773 0.00062 14 24780 404 0.10567 224 | 6 APPENDIX 3 16901 592 0.00626 18 26283 381 0.01072 4 18183 550 0.00412 23 27516 363 0.07089 5 18431 543 0.06130 24 28019 357 0.06192 6 19735 507 0.11450 29 30082 332 0.01847 7 20703 483 0.05125 32 31177 321 0.01062 8 20931 478 0.00062 36 32993 303 0.03505 9 21795 459 0.03127 46 33168 302 0.02112 10 22232 450 0.06565 Section 6.4 | 225 Table S6 Selected vertical TD-DFT transitions of 3+ (B3LYP, def2-SV(P), ZORA, COSMO(acetonitrile)) sorted by their energy including difference density plots |ΨES|2 − |Ψ 2GS| (contour value 0.005, purple: depletion, orange: gain in electron density). Hydrogen atoms are omitted for clarity. ?̃? / λ / 2 2 ?̃? / λ / f |Ψ 2 2−1 osc ES| − |ΨGS| −1 fosc |ΨES| − |ΨGS| cm nm cm nm 1 15881 630 1.75∙10−6 10 23117 433 2.95∙10−7 3 17096 585 0.00062 9 23166 432 0.00512 2 17563 569 0.00902 11 24728 404 0.10876 4 18873 530 0.00626 18 28165 355 0.05247 5 19215 520 0.05380 19 28534 351 0.08369 8 21496 465 0.00058 24 29891 335 0.11957 6 21914 456 0.06001 26 30040 333 0.03531 7 22283 449 0.17235 28 30815 325 0.05395 226 | 6 APPENDIX Table S7 Selected vertical TD-DFT transitions of 4+ (B3LYP, def2-SV(P), ZORA, COSMO(acetonitrile)) sorted by their energy including difference density plots |ΨES|2 − |Ψ 2GS| (contour value 0.005, purple: depletion, orange: gain in electron density). Hydrogen atoms are omitted for clarity. ?̃? / λ / ?̃? / λ / −1 fosc |Ψ | 2 − |Ψ |2ES GS −1 f 2 2 osc |ΨES| − |ΨGS| cm nm cm nm 1 15985 626 3.06∙10−6 10 23117 433 1.66∙10−6 3 17210 581 0.00063 9 23179 431 0.00662 2 17603 568 0.00919 11 24694 405 0.10794 4 18916 529 0.00616 18 28199 355 0.05170 5 19264 519 0.05350 19 28564 350 0.08182 8 21518 465 0.00191 26 29633 338 0.04010 6 21948 456 0.05880 24 29888 335 0.12162 7 22325 448 0.16638 28 30892 324 0.03754 Section 6.4 | 227 Table S8 Selected vertical TD-DFT transitions of 62+ (B3LYP, def2-SV(P), ZORA, COSMO(acetonitrile)) sorted by their energy including difference density plots |ΨES|2 − |ΨGS|2 (contour value 0.005, purple: depletion, orange: gain in electron density). Hydrogen atoms are omitted for clarity. ?̃? / λ / 2 2 −1 fosc |ΨES| − |ΨGS| cm nm 1 13941 717 2.65∙10−6 3 14926 670 0.00063 2 15509 645 5.04∙10−6 12 15932 628 0.00149 21 16101 621 4.49∙10−6 25 16358 611 1.44∙10−6 6 16487 607 0.00070 19 16676 600 2.95∙10−6 228 | 6 APPENDIX 26 16800 595 3.25∙10−7 23 16842 594 0.00013 11 17953 557 0.02374 10 18348 545 0.02801 7 18491 541 0.00986 9 18780 533 0.05506 8 18854 530 0.01090 13 20498 488 0.07937 14 20813 481 0.05640 Section 6.4 | 229 22 21003 476 0.00912 15 21806 459 0.34875 18 22284 449 0.13827 41 22291 449 0.01702 20 22393 447 0.00896 36 24755 404 0.04056 38 24977 400 0.16626 44 25891 386 0.03412 50 26642 375 0.01756 230 | 6 APPENDIX Figure S18 Absorption spectra of 62+ after addition of 1.50, 1.75 and 2.00 equivalents of (NH ) [Ce(NO ) ] as oxidant (63+64+4 2 3 6 ). Table S9 Selected vertical TD-DFT transitions of 63+ (B3LYP, def2-SV(P), ZORA, COSMO(acetonitrile)) sorted by their energy including difference density plots |Ψ |2ES − |Ψ |2GS (contour value 0.005, purple: depletion, orange: gain in electron density). Hydrogen atoms are omitted for clarity. ?̃? / λ / −1 f 2 osc |ΨES| − |ΨGS|2 cm nm 1 4255 2350 4.63∙10−7 2 4495 2225 0.00026 3 7172 1394 0.06523 4 8319 1202 2.50∙10−6 5 8674 1153 0.00050 Section 6.4 | 231 11 13059 766 0.00423 6 14873 672 0.00036 9 15530 644 8.00∙10−8 7 15758 635 0.19415 8 15988 626 0.04104 10 16361 611 5.85∙10−6 19 17520 571 0.00070 13 17850 560 0.00943 44 17998 556 0.01475 232 | 6 APPENDIX 37 18303 546 0.01029 21 19180 521 0.01038 24 19526 512 0.05026 25 19889 503 0.00272 30 20522 487 0.00723 35 20717 483 0.00345 31 20725 483 0.00114 42 21775 459 0.05831 43 22161 451 0.20749 Section 6.4 | 233 49 22512 444 0.01892 48 22618 442 0.07026 Figure S19 UV-Vis absorption spectra of 1+ (top, blue) and 3+ (bottom, red) in dry acetonitrile upon addition of 01 equivalents of (NH4)2[Ce(NO3)6] as oxidant. 234 | 6 APPENDIX Figure S20 Fit of the IVCT band of 63+ (generated in situ by oxidation of 62+ with one equivalent of (NH4)2[Ce(NO3)6] in acetonitrile). The figure shows the experimental spectrum (black), the fit of the spectral range between 3500 and 16000 cm−1 (red, dashed), the band fits of the LMCT bands (grey) and the fit of the IVCT band (red). Fit parameters: IVCT: LMCT 1: LMCT 2: 𝜈max = 8585 cm −1 𝜈max = 13780 cm −1 𝜈max = 15225 cm −1 ε = 2600 M−1max cm−1 εmax = 4010 M−1 cm−1 ε −1 −1 max = 3640 M cm 𝜈1/2 = 6020 cm−1 𝜈 −1 −1 1/2 = 1480 cm 𝜈1/2 = 2900 cm Section 6.4 | 235 Figure S21 Spectral decomposition of the emission spectra of a) 1+ b) 2+ c) 3+ and d) 4+ (recorded at 155 K in butyronitrile solution) into individual gaussians. Vibrational progression energies are 740 cm−1 (1+), 710 cm−1 (2+) and 670 cm−1 (3+ and 4+).1,2 Figure S22 Emission spectra of [Ru(bpy)3](PF6)2 in butyronitrile in the temperature range between 300 K and 200 K. 1 Z. Murtaza, D. K. Graff, A. P. Zipp, L. A. Worl, Jones, Wayne E. Jr., W. D. Bates and T. J. Meyer, J. Phys. Chem., 1994, 98, 10504–10513. 2 K. Heinze, K. Hempel and M. Beckmann, Eur. J. Inorg. Chem., 2006, 2006, 2040–2050. 236 | 6 APPENDIX Figure S23 Variable-temperature emission plot ln() vs. T−1 of [Ru(bpy)3](PF6)2 in air-equilibrated butyronitrile in the temperature range between 300 K and 200 K. Activation barrier ΔE of 36.0 kJ mol−1 has been determined from the fit using Meyer’s equation (literature value of ΔE = 42.6 kJ mol−1 for degassed butyronitrile).3 (0 kJ mol−1) (6.3 kJ mol−1) Figure S24 DFT calculated spin densities of the 3MLCT states of 62+ (B3LYP, def2-SV(P), ZORA, COSMO(acetonitrile); contour value: 0.01). The relative electronic energies are given in parentheses. 3 B. Durham, J. V. Caspar, J. K. Nagle and T. J. Meyer, J. Am. Chem. Soc., 1982, 104, 4803–4810. Section 6.4 | 237 Figure S25 Difference of the emission spectra of 62+ at 155 K in butyronitrile solution upon excitation at 560 nm and 480 nm. 238 | 6 APPENDIX Derivation of the equation used to fit the 𝐥𝐧(𝝓) vs. 𝑻−𝟏 plots Excited state reaction pathways: 1GS → 1MLCT Excitation 1 ϕ~1MLCT → 3MLCT Intersystem crossing 3 𝑘nrMLCT → 1GS Direct non-radiative deactivation 3 𝑘rMLCT→ 1GS Phosphorescence 𝑘𝑛𝑟 + 𝑘𝑟 = 𝑘1 𝑘2 3MLCT ⇌ 3MC 3MLCT−3MC surface crossing 𝑘−2 3 𝑘3MC→ 1GS 3MC deactivation pathways 3 𝑘4MC→ photo products 𝑘5 3MLCT ⇌ 3LL′CT 3MLCT−3LL’CT surface crossing 𝑘−5 𝑘5 𝐾𝐿𝐿′𝐶𝑇/𝑀𝐿𝐶𝑇 = 𝑘−5 3 𝑘6LL′CT → 1GS 3LL’CT deactivation We assume that the 3LL’CT state is chemically stable. Based on the above reactions, the lifetime 𝜏0 of the 3MLCT state can be expressed as follows: 1 𝑘3+𝑘 𝑘= 𝑘 + 𝑘 ( 4 ) + 𝑘 ( 61 2 5 ) (1) 𝜏0 𝑘−2+𝑘3+𝑘4 𝑘−5+𝑘6 In principle, all rate constants have to be considered as temperature-dependent. However, Meyer[4] argued, that the rate constants 𝑘1 and 𝑘3 for intersystem crossing describe processes at the respective Franck-Condon point and therefore are independent from the temperature. For the same reason, 𝑘6 can be considered temperature-independent. Following Meyer’s argumentation, the back reaction from the 3MC state to the 3MLCT state is slow compared to the 3MC state deactivation (𝑘−2 ≪ 𝑘3 + 𝑘4). Therefore, the first fraction 𝑘3+𝑘( 4 ) of equation (1) equals 1: 𝑘−2+𝑘3+𝑘4 1 𝑘 = 𝑘1 + 𝑘2(𝑇) + 𝑘 6 5 ( ) (2) 𝜏0(𝑇) 𝑘−5+𝑘6 4 B. Durham, J. V. Caspar, J. K. Nagle and T. J. Meyer, J. Am. Chem. Soc., 1982, 104, 4803–4810. Section 6.4 | 239 According to Meyer, 𝑘2(𝑇) is composed of a rate constant at infinite temperature (𝑘 0 2) and an Arrhenius-like activation barrier term, taking the 3MLCT−3MC activation barrier Δ𝐺‡1 into account. Δ𝐺‡ 𝑘2(𝑇) = 𝑘 0 2 exp (− 1) (3) 𝑅𝑇 𝑘 For the second fraction 𝑘 65 ( ), a differentiation into two limiting cases is necessary: 𝑘−5+𝑘6 a) When 𝑘−5 is small compared to 𝑘6, the surface crossing to the 3LL’CT state is irreversible and for the lifetime of the 3MLCT state follows: 1 = 𝑘 𝜏 (𝑇) 1 + 𝑘2(𝑇) + 𝑘5(𝑇) (4) 0 In this case, just as above for 𝑘2(𝑇), 𝑘5(𝑇) is composed of a rate constant at infinite temperature and an Arrhenius term, associated with the 3MLCT-3LL’CT activation barrier Δ𝐺‡2 . 0 Δ𝐺 ‡ 𝑘 25(𝑇) = 𝑘5 exp (− ) (5) 𝑅𝑇 b) When the back reaction from the 3LL’CT to the 3MLCT state is faster than the depopulation of the 3LL’CT state into the ground state (𝑘−5 < 𝑘6), the 3LL’CT and 3MLCT states are in thermal equilibrium: 1 = 𝑘 + 𝑘 (𝑇) + 𝐾 ′ ⋅ 𝑘 (6) 𝜏0(𝑇) 1 2 𝐿𝐿 𝐶𝑇/𝑀𝐿𝐶𝑇 6 An exponential term is required to describe the temperature dependence of the last component of the equation in this case as well, but it contains the difference between the Gibbs free enthalpies of the 3LL’CT and 3MLCT states. Δ𝐺0 𝐾 0 0 0𝐿𝐿′𝐶𝑇/𝑀𝐿𝐶𝑇 ⋅ 𝑘6 = 𝑘6 exp (− ) with Δ𝐺 = 𝐺 ′ − 𝐺 (7) 𝑅𝑇 𝐿𝐿 𝐶𝑇 𝑀𝐿𝐶𝑇 Thus, the mathematical description with a sum over two exponential terms, which was used to fit the ln(𝜙) vs. 𝑇−1 plots, is identical for both limiting cases, although the physical implications differ substantially: 1 Δ𝐺 ‡ Δ𝐺‡ a) = 𝑘 + 𝑘0 exp (− 1) + 𝑘0 exp (− 21 2 5 ) (8a) 𝜏0(𝑇) 𝑅𝑇 𝑅𝑇 1 Δ𝐺 ‡ Δ𝐺0 b) = 𝑘1 + 𝑘 0 2 exp (− 1) + 𝑘 exp (− ) (8b) 𝜏 60(𝑇) 𝑅𝑇 𝑅𝑇 A decision between the two cases a) or b) can only be made based on experimental data. When Δ𝐺0, the difference between the Gibbs free enthalpies of the 3LL’CT and 3MLCT states, is negative, a decrease of the 3MLCT lifetime 𝜏0(𝑇) with decreasing temperature is expected for case b), while for case a) the lifetime should increase. In this study, the quantum yield 𝜙(𝑇) instead of the 3MLCT lifetime 𝜏0(𝑇) is measured. This, however, does not affect the obtained data neither qualitatively nor quantitatively, because 𝜙 and 𝜏0(𝑇) are linearly related when kr is independent from the temperature: 𝜙(𝑇) = 𝑘𝑟 ⋅ 𝜏0(𝑇). This temperature-independence has been observed in all studies on luminescent polypyridine ruthenium(II) complexes. 240 | 6 APPENDIX 6.5 SUPPORTING INFORMATION TO 3.4: STRONGLY COUPLED CYCLOMETALATED RUTHENIUM TRIARYLAMINE CHROMOPHORES AS SENSITIZERS FOR DSSCS Figure S1 1H NMR spectrum (400 MHz) of ligand Lb in CD2Cl2. Figure S2 13C NMR spectrum (100 MHz) of ligand Lb in CD2Cl2. Section 6.5 | 241 Figure S3 1H NMR spectrum (400 MHz) of ligand Lc in CD2Cl2. Figure S4 13C NMR spectrum (100 MHz) of ligand Lc in CD2Cl2. 242 | 6 APPENDIX Figure S5 1H NMR spectrum (400 MHz) of dye [nBu4N]2[2a] in CD3CN. Figure S6 13C NMR spectrum (100 MHz) of dye [nBu4N]2[2a] in CD3CN. Section 6.5 | 243 Figure S7 1H-13C HSQC NMR spectrum of the aromatic region of dye [nBu4N]2[2a] in CD3CN. Figure S8 1H-13C HMBC NMR spectrum of the aromatic region of dye [nBu4N]2[2a] in CD3CN. 244 | 6 APPENDIX Figure S9 1H NMR spectrum (400 MHz) of dye [nBu4N]2[2b] in CD3CN. Figure S10 13C NMR spectrum (100 MHz) of dye [nBu4N]2[2b] in CD3CN. Section 6.5 | 245 Figure S11 1H-13C HSQC NMR spectrum of the aromatic region of dye [nBu4N]2[2b] in CD3CN. Figure S12 1H-13C HMBC NMR spectrum of the aromatic region of dye [nBu4N]2[2b] in CD3CN. 246 | 6 APPENDIX Figure S13 1H NMR spectrum (400 MHz) of dye [nBu4N]2[2c] in CD3CN. Figure S14 13C NMR spectrum (100 MHz) of dye [nBu4N]2[2c] in CD3CN. Section 6.5 | 247 Figure S15 1H-13C HSQC NMR spectrum of the aromatic region of dye [nBu4N]2[2c] in CD3CN. Figure S16 1H-13C HMBC NMR spectrum of the aromatic region of dye [nBu4N]2[2c] in CD3CN. 248 | 6 APPENDIX Figure S17 ESI+ mass spectra of dyes a) [1a][PF6], b) [1b][PF6], c) [1c][PF n6], d) [ Bu4N]2[2a], e) [nBu4N]2[2b] and f) [nBu4N]2[2c] from CH3CN solution. Section 6.5 | 249 Figure S18 ESI+ mass spectra of dyes a) [nBu n n4N]2[2a], b) [ Bu4N]2[2b] and c) [ Bu4N]2[2c] from CH3CN solution. 250 | 6 APPENDIX Figure S19 Solid state IR spectra (KBr disk) of ligands Lb and Lc and dyes [nBu4N]2[2a], [1b][PF6], [nBu4N]2[2b], [1c][PF n6] and [ Bu4N]2[2c]. Section 6.5 | 251 Figure S20 Cyclic voltammograms of dyes [nBu4N]2[2a], [1b][PF6], [nBu4N]2[2b], [1c][PF6] and [nBu4N]2[2c] in CH3CN (c = 0.1 mol l−1; supporting electrolyte: [nBu4N][PF6], c = 10−3 mol l−1). 252 | 6 APPENDIX Figure S21 UV-Vis spectra (200-800 nm) of dyes [nBu4N]2[2a], [nBu4N]2[2b] and [nBu4N]2[2c] in CH3CN solution. Section 6.5 | 253 Figure S22 DFT calculated MO diagram of ester complexes [1a]+, [1b]+ and [1c]+ and frontier orbitals of [1b]+ (contour value: 0.06). 254 | 6 APPENDIX Figure S23 Paramagnetic 1H NMR spectrum (400 MHz) of electrolyte [3][B(C6F5)4]2 in CD3CN. Figure S24 19F NMR spectrum (377 MHz) of electrolyte [3][B(C6F5)4]2 in CD3CN. Section 6.5 | 255 Figure S25 1H NMR spectrum (400 MHz) of electrolyte [3][B(C6F5)4]3 in CD3CN. Figure S26 19F NMR spectrum (377 MHz) of electrolyte [3][B(C6F5)4]3 in CD3CN. 256 | 6 APPENDIX Figure S27 Paramagnetic 1H NMR spectrum (400 MHz) of electrolyte [4][B(C6F5)4]2 in CD3CN. Figure S28 19F NMR spectrum (377 MHz) of electrolyte [4][B(C6F5)4]2 in CD3CN. Section 6.5 | 257 Figure S29 1H NMR spectrum (400 MHz) of electrolyte [4][B(C6F5)4]3 in CD3CN. Figure S30 19F NMR spectrum (377 MHz) of electrolyte [4][B(C6F5)4]3 in CD3CN. 258 | 6 APPENDIX Figure S31 ESI+ mass spectra of electrolytes a) [3][B(C6F5)4]2, b) [3][B(C6F5)4]3, c) [4][B(C6F5)4]2 and d) [4][B(C6F5)4]3 from CH3CN solution. Section 6.6 | 259 6.6 SUPPORTING INFORMATION TO 3.5: EXCITED STATE DECAY OF CYCLOMETALATED POLYPYRIDINE RUTHENIUM COMPLEXES: INSIGHT FROM THEORY AND EXPERIMENT DFT optimized Cartesian coordinates of the 1GS of [Ru(bpy)2(ppy)]+ Ru -0.02190206217445 0.02482014297240 -0.04958866603494 C -0.10260797878980 -0.00097278653297 1.99617280883337 C 1.16113920144658 -0.15226189203276 2.64364539250481 C -1.23041515264235 0.10784665775614 2.84095041203804 C 1.27259945119644 -0.18355793104483 4.04916281068880 C -1.12183902913617 0.07575156227903 4.23709126001905 H -2.22434743689963 0.22483267747609 2.40366863700977 C 0.13417172808340 -0.07036146115855 4.84896123325055 H 2.24568448406269 -0.29750595230729 4.52865888631826 H -2.01875255553463 0.16425528639324 4.85424426028089 H 0.22205322599048 -0.09508671920532 5.93597064932104 N 2.01910435642795 -0.22665535415604 0.41429033688720 C 2.32131315340800 -0.26412076151074 1.74806006044785 C 3.01105158519582 -0.32394593262768 -0.49207481221912 C 3.65892494204589 -0.40389829502693 2.16241189620959 C 4.35092919660698 -0.46576416277989 -0.14107475611643 H 2.71329727409568 -0.29011114550564 -1.53800669127460 C 4.67965174907348 -0.50534098675499 1.21991675412589 H 3.89644088676804 -0.43171739488212 3.22359742028317 H 5.11022536027170 -0.54180848473995 -0.91770957257283 H 5.71559133881184 -0.61385899735353 1.54090244194931 N -0.50646607849372 -1.98679396862291 -0.07776645898051 C 0.36535169941325 -3.00560862564891 0.06962691777329 C -1.82899977927372 -2.27659271927598 -0.25216783835754 C -0.02300009282522 -4.34264364373979 0.04547074362070 H 1.40734258119760 -2.73028388573022 0.21097473759375 C -2.28413132000890 -3.60307140194975 -0.28340426748911 C -1.37660916263980 -4.65047965163888 -0.13399979014487 H 0.72696814097076 -5.12228634823042 0.16861578477495 H -3.34100639826499 -3.81697224981407 -0.42244085238233 H -1.71921651967936 -5.68418714698018 -0.15583041851668 N 0.26208469983924 0.28051551340496 -2.19439158229186 C 0.18654858723719 -0.68804051299028 -3.12345932500422 C 0.48998677452576 1.55958971979255 -2.59018132089517 C 0.33308767128236 -0.44195521446586 -4.48798847565623 H 0.00440218654620 -1.69727934601571 -2.75681213172899 C 0.64664903859725 1.88085684567562 -3.94779854013224 C 0.56781162281716 0.87185555522741 -4.90768584875744 H 0.26264435688411 -1.26389884441838 -5.19861635487084 H 0.82502411810682 2.90715811467083 -4.25829037209472 H 0.68603334823092 1.10921285021622 -5.96434051020693 N -2.07440801556631 0.10186799643092 -0.32709943756473 C -2.70679038943560 -1.10463032801178 -0.40335257903076 C -2.80936334048413 1.22591940052618 -0.45066239485484 C -4.09181904879649 -1.18766358252074 -0.61035902462464 C -4.18623125249344 1.21138634273544 -0.65725888254241 H -2.26927637511847 2.16656200445894 -0.37790266183219 C -4.84412392369910 -0.02139018200199 -0.74031584776151 260 | 6 APPENDIX H -4.57993865216700 -2.15722392223119 -0.66995625703049 H -4.72552229989484 2.15263182820108 -0.74963072016661 H -5.92014288206236 -0.07387172192168 -0.90128568954490 N 0.33168656868028 2.08466991947955 -0.24145053574220 C 0.37413861672408 2.95110447022406 0.79166969400839 C 0.55524565175926 2.55887515786798 -1.50183661224162 C 0.63300265165631 4.31077425329070 0.63244725966143 H 0.18920928846883 2.52763405866212 1.77551524774962 C 0.82257664707099 3.91755870983914 -1.72753858969592 C 0.86332325785208 4.80667296374059 -0.65450100258574 H 0.65205096422032 4.95991709508930 1.50635896195243 H 0.99959201607371 4.28168675054615 -2.73626040900053 H 1.07045132444103 5.86291267687124 -0.82167137735545 DFT optimized Cartesian coordinates of the 3MLCT state of [Ru(bpy)2(ppy)]+ Ru -0.00582391831081 0.02594705213624 0.00487677542443 C -0.10202330324492 -0.03969146739102 2.03124169889945 C 1.15659134281037 -0.17837325186830 2.68619803888731 C -1.25770683013009 0.03806604541994 2.83378338676050 C 1.23568772265027 -0.23137468601398 4.09075133824811 C -1.17318613225537 -0.00145643624829 4.23076308969006 H -2.23778168038295 0.13762756517188 2.36706334451717 C 0.07409961339392 -0.13932749668553 4.86077834399596 H 2.19716069286433 -0.34304902849711 4.59169553423964 H -2.08091036314520 0.06878879798131 4.83227586007964 H 0.13891063428178 -0.17666115510330 5.94864840481599 N 2.04528648956580 -0.18737990648968 0.47330776932281 C 2.33060113314030 -0.26143965512106 1.80646654424385 C 3.04236587646479 -0.22610491024292 -0.42993897189471 C 3.66279727703305 -0.39587976687114 2.23383418993475 C 4.37950369757505 -0.35698689615406 -0.06596321029442 H 2.75332426617555 -0.15984971686612 -1.47703026209407 C 4.69225580202768 -0.44566658923048 1.29641875269381 H 3.88966694614924 -0.45885135996021 3.29560110649707 H 5.14989423447918 -0.38992017125593 -0.83423343136147 H 5.72633778877892 -0.55095697273243 1.62348164248203 N -0.45622813175899 -1.98043710781621 -0.06530525349620 C 0.40762618711169 -3.00460186339516 0.12954378072723 C -1.81226893225132 -2.25066239979692 -0.28716973364106 C 0.03283691843167 -4.33364041399523 0.08630684495762 H 1.44280141254435 -2.72581410724968 0.31787347778438 C -2.24110112209771 -3.61203557393101 -0.33444836715334 C -1.34078354025446 -4.63827506806540 -0.15652007859108 H 0.77468219591658 -5.11540308981447 0.23659475659152 H -3.29192753624521 -3.83488081687903 -0.50868714413035 H -1.67689189797397 -5.67461966814571 -0.19390605275837 N 0.20105967964042 0.27435735113347 -2.19764928687935 C 0.06181212514665 -0.70126462470621 -3.10683245478162 C 0.41007184875120 1.54969444029494 -2.60351711718399 C 0.12888405868561 -0.46043772001204 -4.47964711378695 H -0.10589453393870 -1.70520896756133 -2.71956519204026 C 0.48339948723078 1.86673483906580 -3.96795490708361 C 0.34199707524101 0.85073293581749 -4.91469858490092 H 0.01220858687737 -1.28327375781662 -5.18262493707074 Section 6.6 | 261 H 0.64615411532060 2.88994193067142 -4.29558426191443 H 0.39605589427788 1.08368885774694 -5.97748842360238 N -2.04843567650735 0.12636240198787 -0.32775817085132 C -2.66238266648466 -1.12003218707435 -0.43815891787566 C -2.78413187394331 1.24780958671937 -0.47249498190253 C -4.06707525257734 -1.17759414330527 -0.69128360950276 C -4.14729058356568 1.23986101295940 -0.71960775286486 H -2.24702810060867 2.19028089052149 -0.37572718474738 C -4.80112193252053 -0.01984812468570 -0.83026801293197 H -4.55656029901912 -2.14569712265762 -0.77587983356487 H -4.68852256117734 2.17833883144470 -0.82195870690058 H -5.87292935572475 -0.07002437187492 -1.02320201057537 N 0.40853307570580 2.09826614943241 -0.24844966293506 C 0.54695967873187 2.95695848478663 0.77934570880299 C 0.55306241821879 2.55425529798644 -1.52367241072608 C 0.82337537291786 4.31054429912261 0.59667107241334 H 0.42385912716453 2.54204170855464 1.77678412718717 C 0.83170348126075 3.90534593210521 -1.77251664641185 C 0.96678912530747 4.79432859216638 -0.70605119213873 H 0.92136374040117 4.96153672850165 1.46332967579251 H 0.94626953303180 4.26481336927161 -2.79148640281630 H 1.18249756881237 5.84542449451461 -0.89268098358470 DFT optimized Cartesian coordinates of the 3MC state of [Ru(bpy)2(ppy)]+ Ru -0.35189428368518 -0.11080288179678 -0.05974649397658 C -0.34901844072056 -0.22258880017789 1.99369582165052 C 0.89558155394524 -0.26743906776530 2.69007269353272 C -1.52556485870115 -0.25923407240878 2.77366297101283 C 0.92209728827479 -0.35205513418613 4.09892713468511 C -1.49169902122391 -0.32714410492679 4.17211493068486 H -2.50073637639817 -0.23183896256260 2.28106509053827 C -0.26003793749430 -0.37965945131319 4.84126225889500 H 1.87254922396188 -0.41287800302189 4.62955784078633 H -2.42433864986063 -0.34858925730853 4.73967478739579 H -0.22172660258688 -0.44766266934687 5.92932836042573 N 2.02625502479739 -0.39943294359546 0.56370376400592 C 2.15109374194929 -0.22022539164127 1.90122110697632 C 3.10680457439803 -0.35748076028195 -0.22544652865885 C 3.42405481668825 0.01524491410816 2.46134641538907 C 4.39624249863895 -0.13747039177008 0.25958399534136 H 2.93458116740599 -0.51448904167909 -1.29238526576665 C 4.54934908537201 0.05504959968866 1.63853452757487 H 3.53422430752092 0.18072199396642 3.53078646633095 H 5.24688116259279 -0.11686791362497 -0.42038569773421 H 5.53413507027030 0.24065668248636 2.06801115019476 N -0.79665043593874 -2.22408161948253 -0.06229150271323 C 0.17155843926770 -3.13412859131363 0.15685603000235 C -2.05863071668810 -2.65880791637264 -0.32342497706353 C -0.05346373671636 -4.50854300043366 0.11593242836546 H 1.15915732251452 -2.73303277477334 0.37412269903248 C -2.35183406775500 -4.03063470990228 -0.39023293166980 C -1.34369332111656 -4.96761327617780 -0.16903411300847 H 0.76902735847453 -5.19647480012259 0.30453373329281 H -3.35928956034204 -4.36371004125669 -0.62733660020379 H -1.56136689084997 -6.03388789394416 -0.22113978798046 262 | 6 APPENDIX N 0.13188411024017 0.16535467431698 -2.19200353389406 C 0.16471858763154 -0.79880819002213 -3.12678499298309 C 0.46839187232406 1.43686755756237 -2.52882449938327 C 0.53693295671145 -0.55569623059909 -4.44759372609184 H -0.12031350197792 -1.79861587869172 -2.80189007330876 C 0.85437226054378 1.75397570244116 -3.84195053984326 C 0.88969562418421 0.74983008675972 -4.80877896048399 H 0.54790971411256 -1.37037161548553 -5.16970636994875 H 1.12672549789420 2.77110290948964 -4.11153660303338 H 1.18840662684080 0.98424387272893 -5.82986410794510 N -2.62162250760602 -0.36788623937064 -0.75025382734202 C -3.08978503863820 -1.61472784671164 -0.53123195583218 C -3.48225002066925 0.63872997602241 -0.95085193685515 C -4.47020431808744 -1.87647864628565 -0.49740051447223 C -4.86618683851640 0.45984884612459 -0.94190639886141 H -3.04622172241065 1.62275708733736 -1.12996411285820 C -5.36647018735682 -0.82695865619953 -0.70628772333563 H -4.84486764272740 -2.87741007430948 -0.29590907913036 H -5.52966493044018 1.30539296830052 -1.11728487290905 H -6.44027237816649 -1.01055386004129 -0.67965040701961 N 0.03243893364515 1.95055143465520 -0.21074547417542 C -0.08028677839855 2.81890450310220 0.81688566618201 C 0.38659927309255 2.43173227845597 -1.43987562136782 C 0.14822658434747 4.18599311058372 0.68336297429577 H -0.36138605967395 2.38966768302938 1.77545910717867 C 0.63657766168796 3.79859177971044 -1.63557457063955 C 0.51846732599325 4.68792001377219 -0.56872215373812 H 0.03949610985895 4.83663906638382 1.54955195187367 H 0.91702401911271 4.17019749136812 -2.61770487027826 H 0.70856416945341 5.75055490850930 -0.71434653013758 DFT optimized Cartesian coordinates of the 3MLCT−3MC transition state of [Ru(bpy) (ppy)]+2 Ru 0.10059725744503 0.03829869157653 -0.09700174905223 C -0.15352728395743 -1.82459573913717 -0.92130778095472 C -1.21808973912787 -2.65625211459010 -0.46392696206675 C 0.67965137748452 -2.33933453814504 -1.93802670376463 C -1.40839828605785 -3.93950702936710 -1.02099484663494 C 0.47692233212446 -3.60842147318772 -2.49450688877076 H 1.51072217738310 -1.73617115519426 -2.31255764516431 C -0.57123359104014 -4.41733582736774 -2.03078342862540 H -2.20866795065657 -4.58514670112045 -0.65872228549305 H 1.14021402617894 -3.97020667859999 -3.28291922743888 H -0.73022577903239 -5.41233289216810 -2.44790383277715 N -1.70582458997020 -1.00398423129891 1.22226071899370 C -2.11243222793549 -2.13722418352761 0.59924456744569 C -2.45790552598107 -0.45768007291293 2.18524321433237 C -3.33385810824475 -2.73905964464111 0.96647635059209 C -3.67333399417409 -1.00020687927066 2.60279544835145 H -2.06706287998340 0.45002628791350 2.64903046688954 C -4.11542227151162 -2.16978162244131 1.97160819967498 H -3.67937104043125 -3.63824739540112 0.46159589150274 H -4.24939094148002 -0.52209742907689 3.39387802499092 H -5.06198264097320 -2.63048199723850 2.25484541639260 N 1.69142858346631 -0.81271934139596 1.07696945430955 C 1.43817925461553 -1.59190042044933 2.14604941832058 Section 6.6 | 263 C 2.98559951320335 -0.54156376081456 0.75476522538997 C 2.44145398035724 -2.13013855242927 2.94884220487235 H 0.38804490390033 -1.78587911587348 2.35356517664839 C 4.04465355695325 -1.04450115074892 1.52799380433923 C 3.77538564011639 -1.84685034603079 2.63546977849461 H 2.17507597715224 -2.75687819313285 3.79834408011491 H 5.07248476627246 -0.79736126261098 1.27331219243783 H 4.58991584817286 -2.23915123230388 3.24313902973132 N -0.19539341657397 1.95067027707156 0.97702028304368 C 0.43364226083295 2.33665758859259 2.09870425206889 C -1.15168651123689 2.74683980076651 0.43562903425056 C 0.14413735291649 3.53436495810224 2.75021425909436 H 1.19650562361062 1.66092513171241 2.48368952467310 C -1.49341194944431 3.96960872206092 1.03741245628436 C -0.84165782353244 4.36526282787629 2.20496723245500 H 0.67961460323228 3.80454760193168 3.65862182595187 H -2.26098791747188 4.60713185177708 0.60657992263524 H -1.10032000483803 5.30984472406176 2.68189697179019 N 2.11523638429280 0.97142626600519 -0.88521946278472 C 3.20311757979836 0.30336948723742 -0.44050694117762 C 2.21972880588658 1.76526972751847 -1.96117346492508 C 4.44181998730280 0.41540796572223 -1.09458240102588 C 3.41575518226549 1.93813000261030 -2.65676332814452 H 1.31241870988765 2.28611117820700 -2.27047669634397 C 4.54903172357435 1.24328308374834 -2.21289576011697 H 5.30780917032027 -0.14467754093894 -0.74917970698148 H 3.45674323430642 2.60153104672019 -3.51952249089448 H 5.50084753596859 1.33924813366251 -2.73434295787464 N -1.37655326329130 1.01064323437773 -1.22794462823011 C -1.90523557409249 0.51230193682800 -2.36588985637480 C -1.77943035675092 2.24601249149476 -0.80475388424883 C -2.84180466056843 1.20050163772230 -3.13254651682109 H -1.55872631450012 -0.47505837429523 -2.66098291072190 C -2.72327369124507 2.98648866537061 -1.53268693749894 C -3.26282831020308 2.46461517381446 -2.70726262544314 H -3.23002948778144 0.74549608963249 -4.04220498222241 H -3.03229400862902 3.96998144651246 -1.18887554438983 H -3.99380820830492 3.03553286508338 -3.27852897910883 DFT optimized Cartesian coordinates of the 3MC−1GS MECP of [Ru(bpy)2(ppy)]+ Ru -0.59266234665190 -0.11908882786205 0.12845431123761 C -0.16082339088075 -0.19633714902559 2.12895221891764 C 1.16262849239242 -0.24693419190606 2.65104922928732 C -1.22726677951433 -0.19957793271765 3.05663239383905 C 1.36985873432140 -0.30773490170160 4.04672227856993 C -1.01191308956290 -0.25288588902361 4.43921649900127 H -2.25949415523287 -0.16038268571262 2.69687150710392 C 0.29698389922326 -0.31110048101609 4.93964182089557 H 2.38159433414344 -0.37412928058268 4.44745813645221 H -1.86260297868274 -0.25617154496532 5.12378963452572 H 0.47875086449473 -0.36557681411539 6.01378391485999 N 2.03424302586296 -0.52441359686505 0.41997417634422 C 2.31178464335867 -0.23989042349256 1.71190487451598 C 3.01935313760749 -0.54853900404469 -0.48232229094358 C 3.63528700565228 0.05040931741136 2.10622482044833 264 | 6 APPENDIX C 4.35490826085719 -0.28574930910277 -0.16886071378847 H 2.73214911472861 -0.79377922737322 -1.50698198842960 C 4.66071486218842 0.02594444409950 1.16112103832845 H 3.86090862012508 0.30936856746712 3.13836631473969 H 5.12264270392067 -0.32112509193795 -0.94091837091665 H 5.68499339199535 0.25270679507225 1.45876256776160 N -1.02425503032493 -2.20210820187459 0.13889815440617 C -0.11952357863853 -3.09876943348954 0.58098315451189 C -2.22782076219924 -2.65489811441466 -0.31146191352042 C -0.35607488960351 -4.47130891053913 0.60081532287165 H 0.82608036079971 -2.68565261678103 0.92146480768856 C -2.52562647883391 -4.02668602802610 -0.32425585731023 C -1.58640850506816 -4.94752660325657 0.13757877524202 H 0.41422573041153 -5.14542630869619 0.97150730397946 H -3.48291376085525 -4.37615944686626 -0.70184046959022 H -1.81028829757984 -6.01363963997818 0.13037232922920 N 0.09090040118975 0.18354333338031 -1.99923440180989 C 0.21219452056460 -0.76493062013385 -2.93854963407635 C 0.42818193560241 1.46326894564688 -2.28283220985838 C 0.68134588324105 -0.49427022184952 -4.22362105325207 H -0.07774359436890 -1.77483060805172 -2.64932556890763 C 0.90887238118108 1.81122754261691 -3.55737246387819 C 1.03685817151589 0.82369218357879 -4.53385484423740 H 0.76433838277969 -1.29545713113300 -4.95602906508599 H 1.18149152756798 2.83714575853037 -3.79097610040257 H 1.40982030609515 1.08106440077055 -5.52456563306996 N -2.70118669302528 -0.37695630629877 -0.90820156439993 C -3.18893751390749 -1.63009881833555 -0.78719054305308 C -3.49832218896968 0.61068497024689 -1.33406618953817 C -4.52972581626980 -1.91510656059439 -1.09584885094059 C -4.83851967378742 0.40675948766016 -1.66595937067190 H -3.04613551150879 1.59993685756225 -1.41630340881023 C -5.36088286128265 -0.88542891033660 -1.54026187482425 H -4.92912862978528 -2.91992840972347 -0.98377591655978 H -5.45039416564827 1.23795048746778 -2.01250968934636 H -6.40344653676139 -1.09029555711416 -1.78099929241512 N -0.21887835707593 1.93486261701095 0.00024114918388 C -0.44110964418517 2.79369264542459 1.01930154728025 C 0.24147299520419 2.43739778576858 -1.18539201365929 C -0.21696368796179 4.16418464847209 0.92297565771131 H -0.80522574554315 2.35249931440115 1.94376025498898 C 0.49528805193757 3.80895075644994 -1.34028169995831 C 0.26840476566444 4.68564975310141 -0.28084934652842 H -0.41747053803874 4.80255827286308 1.78188167855160 H 0.86409602105339 4.19424559665492 -2.28722187976773 H 0.46191767606818 5.75139931728034 -0.39565765292269 DFT optimized Cartesian coordinates of the 1GS of [Ru(dpb)(tpy)]+ C 12.70986134899085 1.98220383100623 3.71162115152764 C 12.38739122164992 0.73236027679925 3.16650298798295 C 13.21034724882957 0.20338594067972 2.16426246650927 C 14.33194571368258 0.91593461096831 1.72139318351112 C 14.60988425973646 2.16114775521095 2.30051821563134 N 13.79985449005086 2.64921569505625 3.26758731284669 H 11.51672264423244 0.17657163685905 3.50756238536650 Section 6.6 | 265 H 14.96902557531010 0.50235430663386 0.94264602309717 C 11.86970819091090 4.64415544049135 6.08305472728813 C 10.71473505847850 4.20060876325122 6.72594460304821 C 10.17656552171168 2.96041785332860 6.36741902760661 C 10.81608875965492 2.21389532091951 5.37748635325243 C 11.97441360344888 2.71130146773052 4.76661263447539 N 12.49494219348044 3.93084550491562 5.12718804289036 H 9.27610109914716 2.58004474374099 6.84788515291804 H 12.31895487713878 5.60256086189951 6.33425691988814 H 10.25323217881131 4.82185660073879 7.49181775936554 H 10.41799928087446 1.24640533383714 5.07981435113701 C 15.73687564630091 3.06276226496691 1.97976443884069 C 16.70839126852085 2.76405686636038 1.01559095158435 C 17.75001784228006 3.66095381591704 0.77649870535925 C 17.79564337773860 4.84855838813408 1.51417316547163 C 16.79819370526775 5.09391431842504 2.45678085539192 N 15.78832953179915 4.23507152484043 2.69426787745731 H 18.51009891166827 3.43502306991568 0.02961217146751 H 16.65033738504896 1.83245151093874 0.45695395114841 H 18.58792604455079 5.58075239114786 1.36680215760469 H 16.79748443231684 6.00725063649152 3.04776582515725 C 14.10936767167928 7.40821571552064 4.26929500563293 C 14.41863914896342 8.66583848027731 4.81742481760086 C 15.25893614232609 8.74571330506218 5.94135137075609 C 15.79712901666393 7.58845236436631 6.53001485326376 C 15.49145860831185 6.32728034660117 5.98784397415586 H 14.02049727018489 9.58646716214816 4.38688095565026 H 16.44682670427978 7.68975444161002 7.40135557302214 C 12.39824561343533 5.40391605561244 1.74947915998820 C 11.70676252003390 6.31050550056326 0.94911968327575 C 11.80404811533112 7.67562152226252 1.24877631644268 C 12.58572381055274 8.07351771148857 2.33181311764658 C 13.26176454352422 7.11592189924528 3.10628164373781 N 13.15494451035803 5.77703405297390 2.79808402061305 H 11.27763870099253 8.41656752343506 0.64741881935551 H 12.34620272290756 4.33507635520533 1.54762401431819 H 11.10891443078957 5.94903201712825 0.11404674232119 H 12.67632111549822 9.12810200244582 2.58640984809851 C 15.95288782145127 5.01195548841332 6.45014507648960 C 16.79533721184075 4.78723544776261 7.55177552952038 C 17.17243315816100 3.49013720528868 7.89413921563184 C 16.69675361110377 2.42092447790364 7.12384924083098 C 15.86183601182768 2.69989098029918 6.04400208672058 N 15.49161574475104 3.94866735433826 5.70452936505871 H 17.82676630612323 3.31383491218595 8.74774124974577 H 17.15074501549084 5.63518557396702 8.13452943144196 H 16.96300359763747 1.38970795060190 7.34983349047346 H 15.47206570056079 1.89503915462450 5.42265807992537 Ru 14.23265442144270 4.47814384567735 4.08266421595330 C 14.64818125651387 6.23987032925084 4.85600174946847 H 12.97739854986061 -0.76631175809354 1.72737255314975 H 15.49755253577058 9.72218785062950 6.36351140188492 DFT optimized Cartesian coordinates of the 3MLCT state of [Ru(dpb)(tpy)]+ C 15.36665615195286 -0.89853322002545 4.68994605991724 266 | 6 APPENDIX C 14.88457240480534 -1.81309109686364 5.64374845494015 C 15.79167157945354 -2.54416010106605 6.42476756375261 C 17.17611091056390 -2.37554990144005 6.26854974720763 C 17.66347084431477 -1.46168910853326 5.31717862139610 C 16.75848870202374 -0.72662409681138 4.53157288272409 H 13.81351352133364 -1.96356481920765 5.78568863719652 H 17.85717732825504 -2.95812966656071 6.89002505116377 C 14.73031116148055 1.55992268368591 2.08335062691179 C 13.34530662691568 1.64489480017839 1.96952516531245 C 12.55400248247823 0.84360635482172 2.80261107501053 C 13.17399550712772 -0.00897606321773 3.71466785764950 C 14.57397545636277 -0.05642425772373 3.78880221356516 N 15.33005932305306 0.73670806259900 2.96367134825601 H 11.46675095087376 0.88365605110639 2.74131102743718 H 15.38225859787418 2.16494747006657 1.45513307276697 H 12.90503428020984 2.32517377705668 1.24291722001465 H 12.57990227441996 -0.64074939090569 4.37210125678823 C 19.06675157842557 -1.15401702773983 5.01682428295496 C 20.18027688928500 -1.73274440144371 5.64484778266370 C 21.46660737309161 -1.35058290291154 5.26726273719149 C 21.62124621694255 -0.38965285104760 4.26127739786702 C 20.47854862884783 0.14939415223024 3.67356146084525 N 19.23685307248518 -0.21640123795757 4.03417669985906 H 22.33599958177971 -1.79553836558249 5.75061524356128 H 20.03204647423579 -2.47755198655414 6.42455840777223 H 22.60471908319295 -0.05896393448418 3.93217738966370 H 20.55179017186756 0.89810429230936 2.88772111747233 C 18.67143779523406 1.32231300372970 0.60741890092396 C 19.19067922955207 2.20066017498465 -0.34196994437363 C 19.46230076140825 3.53317142777739 0.03092105946836 C 19.16969185764628 3.98945439605053 1.33277559329724 C 18.65136890106161 3.09308980601764 2.26574406314528 H 19.36709873236083 1.87791281484083 -1.36631396576656 H 19.32985407528335 5.03497337239241 1.58908308053532 C 17.20037142756006 -1.94767952498993 1.34629176520592 C 17.38207897452566 -2.69557639155800 0.18916371374320 C 18.03488290285438 -2.09402810796930 -0.90468123091905 C 18.47060061291586 -0.78017463369439 -0.79127189688418 C 18.26008988010880 -0.06398148656781 0.40430878716368 N 17.62152060446778 -0.67370189457712 1.47176985074281 H 16.70114064270322 -2.37303021368228 2.21508680541945 H 17.02333093890095 -3.72216398664305 0.14661813784394 H 18.98102423630040 -0.29871638845301 -1.62318077912754 C 18.22230723617704 3.38098771419146 3.63157181960776 C 18.41510534790169 4.62206643023500 4.27109955773080 C 17.96684257921330 4.81499809205838 5.57137365431691 C 17.31888302467993 3.75543414376194 6.23616467044822 C 17.15368028629182 2.55125762507436 5.56229271403035 N 17.58718202783463 2.34860677508943 4.30251893467043 H 18.92244489242124 5.42610762583708 3.74139454942572 H 16.95122305353820 3.86007495325570 7.25509160255380 H 16.65881341752837 1.70660702500387 6.03841518970751 Ru 17.40942670195578 0.58443327014827 3.15436814542743 N 18.47449829597344 1.77734826799580 1.89397143613881 H 18.19959080025723 -2.64861153873635 -1.82798367364019 H 19.86573211719298 4.22674798177651 -0.70496969040556 H 18.11816987924663 5.77346716251991 6.06690325553282 Section 6.6 | 267 H 15.41632983825181 -3.25255585584749 7.16306075017540 DFT optimized Cartesian coordinates of the 3MC state of [Ru(dpb)(tpy)]+ C 4.43137034484600 4.22899941565427 -1.13779472312929 C 4.18731407035750 2.92929435415081 -0.66983521566719 C 5.18077692790083 2.27350091205141 0.05941969588672 C 6.39301476017192 2.92304164610276 0.29767846825431 C 6.57735864305469 4.22347407802599 -0.19524244667518 N 5.60288128006359 4.85604423026593 -0.88692586214195 H 3.24723308363562 2.42665883365681 -0.88313660014390 H 7.18563861745090 2.41520427819620 0.84163892597410 C 3.15383310338480 6.85112718324117 -3.32136596654016 C 1.82526802210954 6.51243137258190 -3.57993851244884 C 1.30058018817933 5.36570377270782 -2.97120378992425 C 2.12624929133062 4.59778770445542 -2.14899745620421 C 3.45932374854661 4.99567028958229 -1.95234695992346 N 3.94566575816885 6.11971835779045 -2.52485729921983 H 0.26386781354066 5.07269659258513 -3.13410279776255 H 3.60276850221639 7.74043887154761 -3.76492425608636 H 1.22110977683629 7.13386352029228 -4.23946005469714 H 1.73305959518829 3.70621331708748 -1.66586867114274 C 7.83817359424459 4.98387977237753 -0.02819330118053 C 8.86717301843402 4.59639706028682 0.84637742360268 C 10.03673739848677 5.35589716123102 0.90039483572207 C 10.15239057639080 6.48337048710147 0.07767224228022 C 9.07831230441005 6.81433447017137 -0.74922715110665 N 7.95020878441423 6.09173643099441 -0.79529999078295 H 10.84465911978192 5.07189397394056 1.57402917808771 H 8.75803574138186 3.72056472234539 1.48223961393751 H 11.05237153997474 7.09658515793941 0.07613816000042 H 9.11795652917271 7.69094809382159 -1.39679111144298 C 6.69364156572647 9.23207719273884 -3.32018036999425 C 6.86707946004530 10.59293105625108 -3.64024016352025 C 6.51662050743235 11.57376772336614 -2.70274313636216 C 5.99389236846280 11.22514677273860 -1.44307380662827 C 5.81587552773686 9.87247401189237 -1.11608510751040 C 6.16716965707022 8.87684159785335 -2.05833879492193 H 7.27154383177196 10.90072590974169 -4.60554246815011 H 5.73470663021082 12.01833883470414 -0.74058011016842 C 7.01064732712790 5.73276771078638 -4.25797564543846 C 7.53145225325543 5.72951684025948 -5.55036099565535 C 7.79695438733869 6.95998962794887 -6.16733506551159 C 7.53403116256976 8.13593422750823 -5.46746957564886 C 7.00979395135365 8.07374140038114 -4.16473377674761 N 6.75094186262210 6.86134995635885 -3.57605725817554 H 8.20279317611410 6.99988571843603 -7.17783559772158 H 6.78933769718370 4.79926309251595 -3.74224115773359 H 7.72141164981534 4.78477809054288 -6.05732246670769 H 7.73148726534015 9.10407228180995 -5.92440763376452 C 5.28142846004700 9.33903031154309 0.15529697162401 C 4.85960197821285 10.12086916114516 1.23919180946441 C 4.36821179160060 9.50659774534675 2.39370319062240 C 4.30543508916379 8.11092907448740 2.44854025948141 C 4.74022514568119 7.38625817914954 1.33794198778719 N 5.21354965322560 7.96955737584114 0.22590502629043 268 | 6 APPENDIX H 4.03931055164529 10.11194775425862 3.23816421239678 H 4.91504094353513 11.20604561022724 1.17850945047070 H 3.93049281616743 7.58782506514367 3.32649043269590 H 4.70992625784011 6.29783820163401 1.33714991465397 Ru 5.91638982221294 6.98334803518108 -1.57405642124779 H 5.01824994487403 1.26050873883475 0.42454050053655 H 6.65206276994232 12.62655976618829 -2.95161058394094 DFT optimized Cartesian coordinates of the 3LL'CT state of [Ru(dpb)(tpy)]+ C 15.31816003118714 -0.86461821038032 4.66172210129033 C 14.80153839632097 -1.77189047912937 5.60281957774737 C 15.69367644262167 -2.51932818627397 6.39029033386642 C 17.09138285810335 -2.38650662284775 6.26845878628807 C 17.61400511506414 -1.48179268788625 5.33281658306258 C 16.71604889621880 -0.73418719569880 4.54371868535377 H 13.72826939865165 -1.91033380119154 5.73752223646208 H 17.73698008283872 -2.99030426522132 6.90689256353865 C 14.79092005361711 1.64021742518002 2.05938780902454 C 13.40644881795057 1.75747291727453 1.92691339203636 C 12.58679710371503 0.96659928334539 2.73795248907554 C 13.17858669067851 0.09043561813476 3.64987983278192 C 14.57451643581129 0.01538637222474 3.73995964177250 N 15.36569732470048 0.79992519587603 2.93335748434879 H 11.50148066302608 1.03010905286565 2.66356871660050 H 15.46549515985138 2.23708966907102 1.44811869709894 H 12.99059147313711 2.45548573261133 1.20242681455281 H 12.56236253600744 -0.53508314038299 4.29299914405880 C 19.03991130593370 -1.18801580481354 5.05529406835875 C 20.12076022159477 -1.78895600226085 5.70916897457336 C 21.42815148901933 -1.42681270645868 5.36999193842585 C 21.63178350369356 -0.46493592176749 4.37792698298560 C 20.51372076150788 0.09907030489894 3.75887532696968 N 19.25942915373436 -0.24576542833526 4.07960307210279 H 22.27336561026839 -1.89152539968246 5.87683780460535 H 19.93996745258168 -2.53517185310203 6.48063247215481 H 22.63095431165447 -0.15094857194969 4.08175674256303 H 20.62367976253702 0.85211241872546 2.98005401648782 C 18.58636691435896 1.36511788926079 0.57433519601395 C 19.19251134689961 2.18938435529465 -0.36816054211499 C 19.49948847112861 3.52272751979220 0.00908143140102 C 19.19896082319207 3.98177690381957 1.28689194636359 C 18.58480216133518 3.11119199251819 2.21711341292324 H 19.42673004309111 1.83285490110262 -1.36848391349806 H 19.43270621654200 5.00826317288793 1.56462741630169 C 17.16587798917905 -1.91265276806066 1.37169753909916 C 17.32772052184989 -2.66873816251429 0.21662212957660 C 17.94746532305503 -2.07880738460766 -0.89579153392850 C 18.37302030130879 -0.75614630997651 -0.80441505016311 C 18.18561665675324 -0.03795001497247 0.38570516064769 N 17.58730647283327 -0.63299391873071 1.47097537747489 H 16.68950042342210 -2.33538728245677 2.25319228418658 H 16.97239476710286 -3.69716954227278 0.19425967108860 H 18.85092696360376 -0.27314332521836 -1.65387145016767 C 18.19895633589941 3.39239069660753 3.56993821120698 C 18.39841017448824 4.63261760044709 4.23221664207184 Section 6.6 | 269 C 17.96752247703751 4.81627656617437 5.53100170416272 C 17.31410965449109 3.74530888653804 6.20139489494394 C 17.14150793249257 2.55030853671808 5.52528345305090 N 17.56941611717202 2.34627341191526 4.25686577905629 H 18.89784444320352 5.44108470381455 3.70039701534052 H 16.94929792482478 3.84642260203043 7.22162021646012 H 16.64722962919427 1.70688595673842 6.00422661609176 Ru 17.44584940623572 0.53101299866710 3.23337622644892 N 18.30090921626586 1.82582209010985 1.81507642064533 H 18.09218918782569 -2.64091196202861 -1.81752995677433 H 19.97034566747865 4.19114089293389 -0.71064478453923 H 18.12450573740016 5.77078155590843 6.03287087699009 H 15.29276463830833 -3.22349666126575 7.11940183545242 DFT optimized Cartesian coordinates of the 3MLCT−3LL'CT transition state of [Ru(dpb)(tpy)]+ C -2.881571 -0.680049 -0.217668 C -4.258789 -0.405259 -0.318463 C -4.697161 0.926754 -0.281285 C -3.791652 1.995890 -0.145681 C -2.416707 1.728736 -0.054864 C -1.975879 0.391153 -0.094804 H -4.992938 -1.205699 -0.418957 H -4.176199 3.015950 -0.112694 C -0.137204 -3.070809 -0.008616 C -0.727868 -4.330473 -0.078505 C -2.118222 -4.411258 -0.223561 C -2.864129 -3.234916 -0.285182 C -2.219148 -1.991607 -0.205350 N -0.851391 -1.933381 -0.077050 H -2.614297 -5.379355 -0.285542 H 0.940946 -2.961120 0.094886 H -0.106171 -5.222204 -0.025433 H -3.946769 -3.272680 -0.390722 C -1.316627 2.703243 0.078061 C -1.466506 4.095453 0.107883 C -0.341192 4.914674 0.221237 C 0.925277 4.327403 0.303704 C 1.014627 2.935776 0.275605 N -0.065120 2.145779 0.166098 H -0.453389 5.998401 0.245894 H -2.460721 4.532533 0.040020 H 1.829376 4.926265 0.395815 H 1.978148 2.434106 0.346008 C 2.612035 -0.496663 1.281468 C 3.984230 -0.709253 1.355745 C 4.720582 -0.744661 0.142201 C 4.082630 -0.576508 -1.085721 C 2.688847 -0.359686 -1.123969 H 4.489328 -0.849275 2.308618 H 4.662055 -0.619008 -2.006554 C -0.590158 -0.205688 3.039696 C -0.279170 -0.367180 4.385601 C 1.060990 -0.569416 4.748623 C 2.030103 -0.607659 3.749770 C 1.661595 -0.442861 2.405630 270 | 6 APPENDIX N 0.343369 -0.234371 2.066524 H -1.616571 -0.041569 2.718191 H -1.072864 -0.331082 5.129785 H 3.075178 -0.768237 4.005536 C 1.847381 -0.180782 -2.280897 C 2.319239 -0.153452 -3.617926 C 1.441014 0.014413 -4.672657 C 0.053784 0.157105 -4.403814 C -0.367509 0.123445 -3.084939 N 0.474783 -0.031420 -2.039067 H 3.385690 -0.265733 -3.807100 H -0.673744 0.284105 -5.203276 H -1.422419 0.223344 -2.834211 Ru -0.032480 0.005427 0.002862 N 2.012682 -0.320370 0.076435 H 1.342910 -0.696356 5.793146 H 5.796490 -0.912102 0.172376 H 1.811629 0.034513 -5.697329 H -5.763581 1.139508 -0.354488 DFT optimized Cartesian coordinates of the 3MLCT−3MC transition state of [Ru(dpb)(tpy)]+ C -2.36864489353110 0.87237850700313 0.64134773038059 C -3.77175419604035 0.81029682913068 0.53419705634385 C -4.35849977449903 -0.15210047293970 -0.29942069263167 C -3.57351806188466 -1.06091338750200 -1.03370266436684 C -2.17534465403174 -1.00777628763679 -0.93014264514859 C -1.57986506629813 -0.03850095827932 -0.09206776953627 H -4.41373486504454 1.49698434783136 1.08766474950324 H -4.06977450301763 -1.79423415168201 -1.67077251572694 C 0.59475931865583 2.42340156815806 2.07716428397209 C 0.12751620339968 3.43118070287481 2.91982656463684 C -1.25608417037032 3.62423410406554 3.02537194253272 C -2.11158675780258 2.80575256911000 2.28940081590516 C -1.57891638236871 1.80463872426504 1.45994032853066 N -0.22242000313658 1.63096373253122 1.36636043087018 H -1.66135007006418 4.40211061130326 3.67239287982139 H 1.66247592642519 2.23897737473531 1.96395299175217 H 0.83307832078664 4.04633435891516 3.47595181943287 H -3.19039814222348 2.93622154773989 2.35511938058548 C -1.20399424189514 -1.88538698058313 -1.61916005023477 C -1.54253668297091 -2.92322472063148 -2.49655833818450 C -0.53747431068967 -3.69469052559331 -3.08588101656628 C 0.79922605070807 -3.41517331044270 -2.78761937523357 C 1.07590309285950 -2.36728389597940 -1.90843548799212 N 0.11517182859326 -1.62259958732707 -1.33935738117067 H -0.79763630197486 -4.50358250159505 -3.76840406654809 H -2.58887054549005 -3.12609665716230 -2.71602024519577 H 1.61532126756384 -3.98970170388861 -3.22188141704935 H 2.10209563182703 -2.11306362971649 -1.64874844267586 C 3.26933157194439 0.77762578644604 -0.80618016507176 C 4.67061491188745 0.80261127170093 -0.79126984168853 C 5.35158662547154 0.02946933358082 0.15480068566400 C 4.62673877879601 -0.74886419320453 1.06309857831940 C 3.22648109875654 -0.73478833363951 1.00226175128286 H 5.22431330236420 1.42565055699280 -1.49027561489357 Section 6.6 | 271 H 5.14582228729734 -1.34004790394485 1.81447389549954 C 0.26235618234000 2.35675239746061 -2.12088628784997 C 0.67148122174460 3.04129753453992 -3.26074082526546 C 2.01763877014468 2.95243622984876 -3.65403759060309 C 2.89677889340120 2.20248757729197 -2.87623775683675 C 2.42144634653038 1.55548876102032 -1.72261346093852 N 1.10461623079144 1.62070385229292 -1.36977549421033 H -0.77447618039482 2.39204410965432 -1.78730152098667 H -0.04705829997013 3.63398589634836 -3.82495161890087 H 3.94297227148681 2.11500085105806 -3.16301656014388 C 2.33518374300430 -1.47833423978259 1.90563163533736 C 2.76179734298391 -2.51067750272028 2.75869493871587 C 1.84314635126242 -3.12062896262539 3.60998568210153 C 0.50771423725767 -2.68327813811820 3.59188181670661 C 0.14757880726321 -1.67298797548783 2.70594433755566 N 1.02826452249471 -1.08600246327056 1.87223544529511 H 3.80061719902034 -2.83478403014716 2.75003276005315 H -0.24003425207711 -3.11514659074657 4.25527797243395 H -0.87911716023599 -1.31149664246892 2.65350666037516 Ru 0.39000721066647 0.01704629050968 0.02672952446421 N 2.58127612313474 0.01294059482083 0.07510899805288 H 2.37145740671836 3.45918757627996 -4.55114402273257 H 6.44005183114239 0.04574152680685 0.19405738163836 H 2.15813489904788 -3.92241261638068 4.27702064954704 H -5.44464134176029 -0.19863699682019 -0.38112021492668 DFT optimized Cartesian coordinates of the 3MC−1GS MECP of [Ru(dpb)(tpy)]+ C 4.38470679815292 4.17076500507583 -1.09998123054573 C 4.15570297011802 2.90751097353839 -0.53319727102153 C 5.14497926167141 2.33861579656771 0.26920459203631 C 6.33710689545686 3.03414197675186 0.47278222361019 C 6.50048171372369 4.29712191586051 -0.11806731599752 N 5.52686574441872 4.85206415282219 -0.86922470337152 H 3.23397424687978 2.36544654497053 -0.72864116218649 H 7.13364553691555 2.58862180359476 1.06350599441438 C 3.12626394549021 6.53774904566726 -3.56135332103208 C 1.82197992445969 6.13191633868455 -3.84866597497328 C 1.30520618820921 5.02611096980045 -3.16357198860359 C 2.11356621836272 4.36390753479239 -2.23776928574841 C 3.42174657761863 4.82653467976509 -2.01875768419102 N 3.89980062408706 5.90824987301357 -2.66734432975930 H 0.28753933828004 4.68278249409209 -3.34687385421043 H 3.56946875174746 7.39931426910931 -4.06267744404180 H 1.23111504308422 6.67073681294023 -4.58793848536903 H 1.72500407885772 3.50587284217991 -1.69407196063948 C 7.75944660801955 5.07732046201249 0.01007418509262 C 8.74380324865493 4.79578200276583 0.97359023776954 C 9.91576886126860 5.55480526570807 0.98473892910646 C 10.07381360966569 6.57245618644621 0.03690774736822 C 9.03510268811746 6.79872036368934 -0.86994098425172 N 7.91060437965419 6.07575558541433 -0.87877637269440 H 10.69150253977390 5.35538206433896 1.72378818364238 H 8.59901850845686 4.00815194953261 1.70997964385010 H 10.97583842607923 7.18192437386980 0.00220717394617 H 9.10546129200951 7.59466686979074 -1.61327721523492 272 | 6 APPENDIX C 6.75201604159672 9.31255158123966 -3.28856096686528 C 7.06883413972527 10.66477443819301 -3.52526259254639 C 6.84637653798825 11.61280268479290 -2.51814374229603 C 6.31485525986489 11.24027413147174 -1.26900383011694 C 5.99298126199607 9.89668986944868 -1.02629529714516 C 6.20673542131676 8.93711290351122 -2.04171149802541 H 7.49307302280703 10.98847833927682 -4.47665183206832 H 6.16188711605762 12.00777693596642 -0.50927408951358 C 6.75232936288312 5.85678030832646 -4.41229174562837 C 7.28803539199305 5.87354489217600 -5.69853715304554 C 7.66382327902605 7.10533024423442 -6.25189292458435 C 7.49576982103393 8.26361723576016 -5.49482009533608 C 6.95368505896774 8.18100517668822 -4.20135791963512 N 6.57875541821678 6.96906614595016 -3.67718806782415 H 8.08232540228768 7.16008432473178 -7.25647951089809 H 6.44150876835871 4.92272838582588 -3.94706009753480 H 7.40280744761232 4.94198415018393 -6.25056949629427 H 7.78363014508894 9.23230554846735 -5.89967117006893 C 5.42888547781248 9.34057673174020 0.22372915605402 C 5.08932452921728 10.09427976103457 1.35409050163174 C 4.54354284984739 9.46245393329997 2.47502655080767 C 4.34425829381511 8.07976417025305 2.44670188212162 C 4.70924805036117 7.38265468258685 1.29334322665000 N 5.23731711025868 7.98214331912406 0.21571440707853 H 4.27747094146672 10.04532228088540 3.35660178470147 H 5.24655061625848 11.17104103293965 1.35528090676358 H 3.92177638998087 7.54370838375779 3.29463225306805 H 4.57907177529512 6.30380172597418 1.23180405420115 Ru 5.74444966740833 7.04973776124694 -1.68290178493078 H 4.99772659829462 1.35513108381861 0.71325230335696 H 7.09407678392897 12.65844168429891 -2.70183253904138 DFT optimized Cartesian coordinates of the 3MLCT−1GS MECP of [Ru(dpb)(tpy)]+ C 15.40484049240013 -0.88345726144979 4.67290211562383 C 14.94754647926214 -1.80500092938010 5.63322591973987 C 15.86735006528119 -2.52420595782918 6.40183923356242 C 17.24219374658701 -2.32987650540788 6.21944508723134 C 17.70861701688645 -1.41080383444737 5.26240030674557 C 16.79373759322178 -0.68206275263216 4.48267281298562 H 13.87957419464140 -1.96371072826052 5.78354054786139 H 17.94413827218692 -2.89929253547154 6.82856280321915 C 14.60367056497956 1.54813304150502 2.09915391747670 C 13.21533549491118 1.59380415067267 2.02896082321080 C 12.47404042959764 0.76923877985877 2.88625244894212 C 13.14349893839878 -0.06687880094169 3.77812484941638 C 14.54592823577888 -0.07435454747098 3.80805400998993 N 15.24461566620492 0.73814594136014 2.96277924583522 H 11.38492041166255 0.77935584760720 2.85834707333634 H 15.22176861306172 2.17122828682848 1.45253168244898 H 12.73345435534352 2.26157603897678 1.31775017373802 H 12.58589350980225 -0.71425315985898 4.45149640219587 C 19.12255532924218 -1.12334564026840 4.98527776079038 C 20.21646046876207 -1.71950408366729 5.63357009473006 C 21.51138957338004 -1.34856564534365 5.27032967626743 C 21.68726225760729 -0.39005175073395 4.26642450935622 Section 6.6 | 273 C 20.55153654339953 0.15758389963834 3.66679411291781 N 19.31165738967484 -0.19796095276491 4.01536784084155 H 22.37055393716787 -1.80185399187783 5.76456223918181 H 20.05542613005032 -2.46313551633017 6.41163486392774 H 22.67854439649378 -0.06967383824452 3.95000439047635 H 20.62749612572942 0.90600903222417 2.87963933841034 C 18.78681451586253 1.26982533793107 0.65456728844805 C 19.30023402574685 2.15533155127804 -0.28580277170369 C 19.60407800472820 3.48256335794422 0.08505209491452 C 19.27933826967863 3.93569682979084 1.38140003530421 C 18.76683212733781 3.04862864723617 2.32084386103413 H 19.43031275158510 1.83918038296745 -1.31975835617745 H 19.39281529875844 4.98976726410595 1.63032688117575 C 17.04347640177903 -1.87351775344298 1.27713732176576 C 17.24585344640456 -2.63990436921467 0.13644740323133 C 18.00912505998313 -2.08985363822064 -0.91279528642626 C 18.53345259576657 -0.81243872856842 -0.77586230897963 C 18.29965720486892 -0.07689343090251 0.40679395401633 N 17.55162638513612 -0.63491207103361 1.42352948642596 H 16.45793954068919 -2.25347125347459 2.11268935761466 H 16.81973037811792 -3.63901219432983 0.07087418785775 H 19.13440769998231 -0.37588775237086 -1.57123027221005 C 18.26232220220065 3.37777925903336 3.64345443489264 C 18.47834312714679 4.60714132667052 4.30399350526601 C 17.94031590026054 4.82019397592822 5.56510524337559 C 17.18087217520412 3.80095892786556 6.17392556452204 C 16.99552597831762 2.61206017740364 5.48000733587735 N 17.51749650353924 2.39167157264199 4.25820588204822 H 19.07685287020168 5.37858363859300 3.82326163752581 H 16.74471758565209 3.92608761311111 7.16296506093197 H 16.41364801678941 1.79690845636811 5.90717965534584 Ru 17.30659631396221 0.68815508234196 3.04135337450720 N 18.66157963728374 1.68718557357099 1.99011971029106 H 18.19241723012228 -2.66153646890871 -1.82227445152951 H 20.00138664867259 4.17752601566264 -0.65224632647679 H 18.10998561793944 5.76595474792969 6.07918242163703 H 15.51403849956687 -3.23790341019833 7.14553705303492 DFT optimized Cartesian coordinates of the 1GS of [Ru(tpy)(pbpy)]+ C 12.74521562833022 1.95490289053854 3.76119933536019 C 12.45868542841642 0.68376668504466 3.25081750186654 C 13.29339122522544 0.13994414217331 2.26765659789958 C 14.39715473410017 0.86827698210373 1.80886616481376 C 14.64886090445029 2.13580298441614 2.34586889086865 N 13.82479841068104 2.64662475926470 3.30209044814486 H 11.59982651966680 0.12140975235201 3.61062787084164 H 15.04921511690488 0.44966441067639 1.04532772292855 C 11.85352524592396 4.65285729437298 6.05037937087397 C 10.70477003003662 4.21148083138506 6.70401207012244 C 10.18425839678606 2.95314113797450 6.37927176195252 C 10.83472037145227 2.18741111632069 5.41223822543540 C 11.98749182484208 2.68575303653428 4.78997578740968 N 12.48777453843325 3.92095627952269 5.11522778685950 H 9.28863119704088 2.57448561500970 6.87001167563632 H 12.28834314160832 5.62389344877784 6.27899039493245 274 | 6 APPENDIX H 10.23374011271928 4.84632861709586 7.45273622364213 H 10.45103107895906 1.20623850585168 5.14136834329680 C 15.75465771801265 3.04169856923670 1.99483358098373 C 16.73487580837206 2.74470264025971 1.03788514372457 C 17.75117497588445 3.66409569375002 0.77948647860197 C 17.76346481987185 4.87103600338486 1.48892647799367 C 16.75970875444248 5.11065949652407 2.42497479205648 N 15.77405987775666 4.22962419572461 2.67987668318202 H 18.51883963438869 3.44202503682480 0.03924374971418 H 16.70256833154536 1.79893329572430 0.50143867003726 H 18.53637756739475 5.62011527289818 1.32478283217553 H 16.73787318232410 6.03680896782337 2.99571163664174 C 14.11743304774658 7.38479875076768 4.27983650473617 C 14.41296225511687 8.64451392436043 4.81493800864275 C 15.24726638329367 8.71810780996924 5.93627159675053 C 15.76693048574328 7.54994739902167 6.49849461855912 C 15.44528407470801 6.30586540975312 5.93091654758392 H 14.00656130146252 9.55072146335351 4.37253126936735 H 16.41668451231850 7.60442269517592 7.36945604716985 C 12.34936712491264 5.51146141753226 1.69027799751470 C 11.68818883275178 6.47240555103074 0.92511452151309 C 11.82910147270612 7.81901263655948 1.27640272375149 C 12.62255881004128 8.15017190313750 2.37567615840372 C 13.25872758986386 7.13258069330615 3.10118581737121 N 13.11380968528862 5.82434638108386 2.74883377668281 H 11.32902547171545 8.60020016745984 0.70512839544242 H 12.26420403771407 4.45254139464700 1.44866697714437 H 11.07940374360197 6.16524031929787 0.07631896764249 H 12.74438084598587 9.19120522689144 2.66573082761323 C 15.89368883481598 4.97596717591248 6.37204534323800 C 16.73895848178298 4.78314217741729 7.48280524728292 C 17.13686003277949 3.49611203750343 7.85035864994381 C 16.68544906910564 2.39639372840521 7.10265968677629 C 15.84287698813811 2.58552363472172 5.99862099222482 C 15.41923828873027 3.86838661996780 5.59690272007006 H 17.79135866317775 3.34890152246904 8.71020385546479 H 17.09015731990586 5.63598712590451 8.06543195870517 H 16.99320683393514 1.38701396366127 7.38407031905902 H 15.51121660108683 1.70673522432742 5.44108596676036 Ru 14.20645585487862 4.43348288455966 4.04334006984424 N 14.63122689144188 6.26242596309963 4.83997060347489 H 13.08510716604031 -0.84811850056666 1.86097570306393 H 15.49099772363989 9.68741363770414 6.36967891020631 DFT optimized Cartesian coordinates of the 3MLCT state of [Ru(tpy)(pbpy)]+ C 12.67660189415259 1.91482973878531 3.61413229680271 C 12.42637189112245 0.64744300646951 3.09555754029943 C 13.27462145110415 0.13055773303360 2.09521881065385 C 14.38890168372808 0.87015762765675 1.64820691251308 C 14.62579286357009 2.13588470103827 2.17605698974852 N 13.73650635197756 2.64703730552963 3.10685895981100 H 11.59487473715519 0.04946227021980 3.46346361464945 H 15.06229403181304 0.44150201147662 0.90845132899273 C 11.87219329906830 4.50100468392905 6.05946342518713 C 10.71819753927229 4.05669370543875 6.69537016950609 Section 6.6 | 275 C 10.16443594246182 2.82385700152531 6.30063539443034 C 10.78425383264120 2.09815574206977 5.29107237841135 C 11.95075817443141 2.59725471817551 4.67848994391582 N 12.48097117706997 3.80879172296293 5.07838209954298 H 9.26196628065677 2.44155472993422 6.77629567441647 H 12.33488331372072 5.44599992118967 6.33931169180159 H 10.26611686174535 4.66036339021994 7.48023228908911 H 10.37116739593959 1.14473753978012 4.96784819679103 C 15.74696926531942 3.02203965463301 1.88345158052032 C 16.73131507839703 2.76281495152257 0.90919396137380 C 17.77035350488123 3.66441323229853 0.71279444667641 C 17.82115412035600 4.83203708172638 1.49852669895284 C 16.82254943632511 5.03969843795263 2.44327504790970 N 15.80812181671036 4.17678287970843 2.63987900305274 H 18.53283357999221 3.46659327273021 -0.03997469307286 H 16.67307100957535 1.85506734343309 0.31174991380135 H 18.61655228475724 5.56575545052230 1.38062563328048 H 16.82639549270988 5.92865480085416 3.07190681639383 C 14.16612539173117 7.32947066673166 4.29075394268052 C 14.47950630887131 8.58633485950216 4.81933373820394 C 15.31834176506314 8.65674163231982 5.93727428565609 C 15.82208096710873 7.48595195366940 6.50296659233320 C 15.48013773447466 6.24655957133265 5.93759582149975 H 14.08128715164628 9.49448395247068 4.37428992644514 H 16.47304111619703 7.53075017292230 7.37326325610846 C 12.34420071666518 5.48617096477300 1.72743839417577 C 11.69104373725377 6.45466342670496 0.96332585992551 C 11.85895495742060 7.79955533541796 1.30672344968737 C 12.66811288175587 8.12674024292332 2.39674032873860 C 13.29277784761941 7.10018921303927 3.11818907049088 N 13.12172678249681 5.80164129604649 2.77172053423177 H 11.36631781565324 8.58558814813179 0.73573419192373 H 12.24373674170179 4.42574158724832 1.49798882193539 H 11.06842039903441 6.15426246536548 0.12236399999221 H 12.80844568619657 9.16745608126187 2.67869934076175 C 15.90961559284557 4.91865398887904 6.38856842644080 C 16.74524188641425 4.69473666473318 7.49936277062373 C 17.10126522853694 3.39290386702486 7.85632166050833 C 16.61747636831403 2.31100415305526 7.10389766481483 C 15.78088415945359 2.52540542156106 5.99932195427156 C 15.41540719733644 3.82896183847970 5.61900226280487 H 17.74912275350825 3.21932543740001 8.71572347071845 H 17.11914010295562 5.53344670549776 8.08729888351095 H 16.89192682855390 1.29135662774279 7.37997178834502 H 15.41929536503553 1.66366756012848 5.43657105445727 Ru 14.23827460543424 4.31595126514310 4.03833961686097 N 14.66641456668263 6.20239196544968 4.84945558098664 H 13.08632682310878 -0.86068022308641 1.68676454204212 H 15.57540119727573 9.62508161331428 6.36466886937299 DFT optimized Cartesian coordinates of the 3MC state of [Ru(tpy)(pbpy)]+ C -2.76102499548434 1.12702377386616 0.28765099569250 C -4.15808118418278 1.19140498309187 0.18398184481036 C -4.85149381031882 0.12042195258441 -0.37959852187425 C -4.13373016560589 -0.99157279154958 -0.81967530715916 276 | 6 APPENDIX C -2.73777832829448 -1.00377864329344 -0.68865200962060 N -2.06918849527796 0.04477611529524 -0.14735916616594 H -4.69825530960821 2.07247598437968 0.52092502451219 H -4.65419952136940 -1.83014520714206 -1.27531054565348 C 0.18068837184451 3.12032886664225 1.09101259665743 C -0.29200213845034 4.22140238133657 1.80694181730009 C -1.66605547086170 4.30472773838347 2.06444304784830 C -2.50873605695000 3.29965752883088 1.58766520899194 C -1.95478553574724 2.23064065706248 0.86161336279000 N -0.62759759511806 2.15445482609416 0.63911350722326 H -2.07652535483121 5.14040811122249 2.63059575656627 H 1.24275866932303 3.00840305222785 0.86764245156253 H 0.39913375988177 4.98939813714506 2.15083226730454 H -3.57636983585324 3.34615410121787 1.78991465967890 C -1.90836169800412 -2.14589833550791 -1.14091553393319 C -2.43632136123881 -3.40959925535929 -1.45855694331224 C -1.57406026609137 -4.41242266914456 -1.90408454507284 C -0.20688923535313 -4.13274574744873 -2.02253319778427 C 0.23993169449312 -2.85834691433661 -1.67022270042102 N -0.58672399899018 -1.89926747974854 -1.23726672907252 H -1.96420385612774 -5.39897093792591 -2.15263679464414 H -3.49857419969389 -3.61670233663012 -1.35060755652596 H 0.49862792149579 -4.88218147181379 -2.37801205233901 H 1.29616450813154 -2.59390034293569 -1.74015061969802 C 2.90822206585381 0.38642394473329 -0.88014516708982 C 4.31236793371106 0.32113140653227 -0.78157984483075 C 4.89494988374734 -0.20868574618442 0.36604170278995 C 4.08631129087374 -0.67135900063847 1.41821770396057 C 2.70036226188503 -0.58970919400853 1.28629324301975 N 2.14845663675068 -0.07292859483304 0.15594216357365 H 4.93618344070649 0.68187417104632 -1.59661131181770 H 4.54341537040503 -1.08286198725524 2.31449337348782 C -0.10223802625385 1.35262430621099 -2.86425641920941 C 0.46531947634252 1.88762773719419 -4.02468593169533 C 1.86404758520133 1.93448806562996 -4.18073062312080 C 2.68417889990032 1.44571456858316 -3.16580843966906 C 2.11378916432691 0.90668244288069 -1.99176229419185 C 0.69082091068909 0.84315959126758 -1.81339358183238 H 2.30357036623667 2.35066755957031 -5.08804939318695 H -1.19065373054366 1.33075834993555 -2.77423428818642 H -0.17884792969117 2.27232888581874 -4.81838592538433 H 3.76776068170319 1.48596697598358 -3.28878305590183 C 1.72264714916337 -1.03311032435366 2.31726764746447 C 2.10831500768407 -1.57173347737862 3.55171996040048 C 1.12853949578574 -1.95989711677798 4.46935611624238 C -0.21879584747297 -1.80261653321321 4.13481002901728 C -0.53055190260299 -1.25907600514790 2.88676196574170 N 0.40587608282102 -0.88501983409767 2.00500715084026 H 1.41860140225541 -2.37859166583536 5.43234057951693 H 3.16018202089079 -1.68864924115065 3.80095380184794 H -1.01514625974496 -2.09082017626086 4.81848043494780 H -1.56714238823843 -1.11779467899539 2.58237948129498 Ru 0.13252028802661 0.01376489879761 -0.03092398559529 H -5.93499637389381 0.15602092876591 -0.48371310386762 H 5.97945525376477 -0.26503269236338 0.45283417577188 Section 6.6 | 277 DFT optimized Cartesian coordinates of the 3MLCT−3MC transition state of [Ru(tpy)(pbpy)]+ C -2.75354927205834 1.12758518422665 0.29199139926459 C -4.14890455291455 1.19887895866046 0.17857089251439 C -4.84222399792926 0.13008922160635 -0.39254658695705 C -4.12780261280195 -0.98651773676722 -0.83037403549527 C -2.73339165236175 -1.00567658550670 -0.68971892013974 N -2.06727658242835 0.03948000838534 -0.13822289981865 H -4.68659695438285 2.08283879879914 0.51319146575650 H -4.64871248835143 -1.82255944733834 -1.29103803029254 C 0.21620401085835 3.09302898611763 1.04502309934211 C -0.23169044543041 4.18054520188965 1.79410514570107 C -1.59750138300368 4.26196856436891 2.10052526920488 C -2.45715338891589 3.26814744021846 1.63278799870300 C -1.93158681591202 2.21322046512345 0.86633330664698 N -0.60808558671484 2.13427239918587 0.59987928613843 H -1.98586524714157 5.08694321392209 2.69705111406689 H 1.27004003804592 2.98496993202617 0.78493731916276 H 0.47170326356084 4.94298436353355 2.12525413522650 H -3.51795878917875 3.30859872224918 1.87082527559947 C -1.89187056302891 -2.13712794341868 -1.13168073555929 C -2.39629823933873 -3.41616180387722 -1.42515727531284 C -1.52003994962333 -4.40673067382607 -1.86815015925306 C -0.15908019311940 -4.09933600704339 -2.00732140589539 C 0.26760378853993 -2.81401623166349 -1.67594759058606 N -0.57240186644720 -1.86265499726940 -1.24437444364842 H -1.89189261968521 -5.40456719848801 -2.09877042768828 H -3.45340952577055 -3.64047497396670 -1.29951487826535 H 0.55670778512377 -4.83748990544315 -2.36583644342333 H 1.31728945459858 -2.53036694305712 -1.76341697449292 C 2.89354379083846 0.39073350681475 -0.88051500500028 C 4.29746809569368 0.32965899233466 -0.77717647168420 C 4.87670325043192 -0.19669175099252 0.37385298802477 C 4.06599272318631 -0.66007316502573 1.42395078386271 C 2.68020834401003 -0.58251718914363 1.28667337355504 N 2.13274320039798 -0.06898031381324 0.15356628212668 H 4.92351926550284 0.69072979551232 -1.59027026376289 H 4.52125309946026 -1.06875969556847 2.32241811678225 C -0.11339010198234 1.34580411484654 -2.87814871326597 C 0.45754358771864 1.88036611250692 -4.03744634726869 C 1.85646634380249 1.93061508579101 -4.18788640753580 C 2.67418844068793 1.44572130350688 -3.16906870074413 C 2.10088166021199 0.90735474706025 -1.99623467909088 C 0.67786185302632 0.84142897850853 -1.82475009592372 H 2.29841249466701 2.34640151709699 -5.09415523720539 H -1.20192700511116 1.32090924817627 -2.79164161067231 H -0.18456439123069 2.26190016731916 -4.83424151324845 H 3.75805081037573 1.48852279868006 -3.28797121490458 C 1.69887842753254 -1.02692775706340 2.31423192546475 C 2.08134616031624 -1.56303835174878 3.55061370812389 C 1.09902732410391 -1.95277525462822 4.46496780193879 C -0.24734320141478 -1.79953893212514 4.12497021249209 C -0.55588458858931 -1.25823069944859 2.87510865059128 N 0.38295323711400 -0.88259254339838 1.99672867462915 H 1.38647652982874 -2.36959035627777 5.42953064897016 H 3.13255506535007 -1.67690542466568 3.80397807650171 H -1.04551132358399 -2.08924309503007 4.80587701864418 278 | 6 APPENDIX H -1.59162798444991 -1.12009513353675 2.56634005418172 Ru 0.11493087974475 0.01555961731688 -0.04224186810891 H -5.92457256379568 0.17179841707972 -0.50605620666362 H 5.96102796196758 -0.24973775273201 0.46482011869129 DFT optimized Cartesian coordinates of the 3MC-1GS MECP of [Ru(tpy)(pbpy)]+ C -2.61493034534291 1.04366822521681 0.45781862424740 C -4.00902873226880 1.19595118017195 0.52515865914716 C -4.83393164864842 0.21435938553557 -0.02325654033724 C -4.25040685884401 -0.89614634818210 -0.62866264218616 C -2.85093570865961 -0.99199204634059 -0.66474760147989 N -2.05812228724584 -0.03709095823884 -0.12915905286159 H -4.45177383241122 2.07346038723464 0.98769594380962 H -4.87546087534866 -1.67042481532353 -1.06517465633768 C 0.50603512357210 2.78932667479091 1.16352933181865 C 0.14953700495120 3.92633173595996 1.89435029646300 C -1.20147231400113 4.11421350603726 2.19955537523526 C -2.13246004894730 3.17587443195657 1.75121010070758 C -1.68634188403353 2.06829295613102 1.00710577412434 N -0.38294440461319 1.88825911560673 0.74047308504594 H -1.52941910887059 4.97798299738384 2.77733058497422 H 1.54950315691442 2.60206065964940 0.91068858375602 H 0.91162382262634 4.63504536233777 2.21469312666611 H -3.18494724162375 3.31044826427166 1.98701097017676 C -2.14382101157311 -2.12974047644463 -1.28901517350702 C -2.79317253786459 -3.21346341489474 -1.90324703205642 C -2.03579390080484 -4.23558159918576 -2.47277904346079 C -0.63884739526004 -4.14975811852020 -2.42145369651687 C -0.06248672245600 -3.04394281052425 -1.80190261424575 N -0.79369230571411 -2.06247518858956 -1.24171172291568 H -2.52684979585084 -5.08203783430813 -2.95133217053059 H -3.87865302539915 -3.26143820188229 -1.93940072985290 H -0.00350394714571 -4.91867010009843 -2.85763235335441 H 1.02049148759010 -2.93142181080828 -1.75361681096138 C 2.86591558506496 0.70196830458019 -0.75554964132898 C 4.21023442231151 1.01551690840609 -0.49386959203120 C 4.75249608580065 0.72115180774536 0.75900584245974 C 3.96694218150152 0.10855376100458 1.74616294790564 C 2.63226816567514 -0.18468430622669 1.45327789326376 N 2.12036130939026 0.12811290797377 0.23438903635863 H 4.82642327620221 1.48028992922595 -1.26038318353723 H 4.39749251129751 -0.13473604065568 2.71414353852674 C 0.03531607067485 0.61877558429846 -3.19875712209368 C 0.59639540060129 1.24046270276736 -4.32250689460856 C 1.92444676030077 1.69890540540096 -4.29656539951659 C 2.68366989203777 1.53094087573898 -3.13659321853824 C 2.12000489392573 0.90856434629251 -2.00490256268650 C 0.76572136886746 0.42855283172671 -2.00746846169419 H 2.36008608657419 2.18139386981678 -5.17214003114659 H -0.99936307288434 0.27312848631279 -3.25461844294640 H -0.00204190244596 1.36932036789953 -5.22712611469691 H 3.71444603617835 1.88797604975387 -3.11832943178858 C 1.67699061892084 -0.84392449306159 2.37833901993527 C 2.00000770526967 -1.16928647929053 3.70261679221722 C 1.05000726398870 -1.80204504046085 4.50757864650347 Section 6.6 | 279 C -0.20423089295620 -2.09864709199312 3.96690260185440 C -0.45782150429933 -1.74382754443286 2.64023542895543 N 0.44769723708164 -1.13172264547892 1.86340312121954 H 1.28811538695899 -2.05653272943761 5.53962698402379 H 2.98057500054063 -0.92880999351841 4.10637827803992 H -0.97663582813041 -2.59224791800700 4.55434370624187 H -1.42353420611855 -1.95594143524479 2.18287486764461 Ru 0.22663176006425 -0.43154395689027 -0.23322094483818 H -5.91744273587323 0.31760227088872 0.01658582794517 H 5.79449746075236 0.95913710592292 0.96922289278892 DFT optimized Cartesian coordinates of the 3MLCT−1GS MECP of [Ru(tpy)(pbpy)]+ C 12.61358178681654 1.97126149828555 3.54868678378587 C 12.49352295068037 0.63615942223376 3.19029890132533 C 13.36289985175564 0.07715983749838 2.22824603939377 C 14.43823169408147 0.85277829001862 1.74170062947695 C 14.55672556297348 2.18701346027740 2.10267348828781 N 13.53842772336366 2.80237279794331 2.87094838689399 H 11.75839075249189 0.00408808150430 3.68672701857455 H 15.20420044501372 0.38788087804342 1.12236777609929 C 11.89263451810404 4.48616207616055 6.08343881124608 C 10.71612640178375 4.06962753157569 6.69738486490413 C 10.11517427454121 2.87835579931180 6.25132898702288 C 10.70419534542836 2.16331863346184 5.21578879840548 C 11.89652394957860 2.63561706358213 4.62893698080609 N 12.46963740038362 3.80063630819390 5.08166774579514 H 9.19158274603602 2.52089045740653 6.70589660516705 H 12.39933645539078 5.39614936962469 6.40063811188129 H 10.28439594476710 4.66094706026866 7.50258939228465 H 10.24462200516214 1.24946372389694 4.84473578528681 C 15.69243792413750 3.05283820273245 1.81404438252031 C 16.64997400736016 2.81984287895913 0.80483818327567 C 17.70665182243004 3.70698626667666 0.63864048191869 C 17.80407839667354 4.83066122155712 1.47974632907618 C 16.82949691060908 5.01143494363075 2.45550763841105 N 15.80353354042125 4.15927671566227 2.62241832743179 H 18.44750084973798 3.53428848626346 -0.14182454981401 H 16.54839089784376 1.95359610365765 0.15397561353576 H 18.61508750660947 5.54984019056508 1.38149060614923 H 16.86563497596011 5.86281386933985 3.13334496982937 C 14.15122130153326 7.30481056925188 4.30353217386350 C 14.44548827594954 8.55941398553222 4.84764308049253 C 15.27132215016172 8.63191827515839 5.97411703449319 C 15.78290519756611 7.46189370295850 6.53014360288512 C 15.45919429924853 6.22533524500600 5.94815137731063 H 14.04177872851538 9.46609834726248 4.40515462163460 H 16.42739114880366 7.50168575346438 7.40525653485800 C 12.38888801329946 5.49990531497811 1.68613515080719 C 11.74295203247803 6.47532326880566 0.92230320689088 C 11.89386406966817 7.81572109761726 1.28890606261785 C 12.67948185189074 8.13359812978422 2.39951976276460 C 13.29514614668341 7.09670514850593 3.11430205564634 N 13.13937707920070 5.80892043735605 2.74730510767862 H 11.40696551967098 8.60677187888171 0.71974692963491 H 12.31098246757531 4.43817688468965 1.45153785663076 280 | 6 APPENDIX H 11.13954070472020 6.18438614816110 0.06407603180235 H 12.80705945685031 9.17124977028875 2.69802798680928 C 15.91259150385345 4.90863548354901 6.39870136000802 C 16.74709340567725 4.68154030553692 7.50963879748664 C 17.12208630186260 3.37992656540976 7.84695933992996 C 16.66279099529401 2.30030625116827 7.07566319474583 C 15.82966380982320 2.51585640334509 5.96890557858786 C 15.44704900700391 3.82059218619813 5.61993574539542 H 17.76851834454947 3.20373756319255 8.70662956149355 H 17.10486457584656 5.51684895408664 8.11177428089967 H 16.95414442271228 1.28193711804760 7.33693250718034 H 15.48634509614720 1.65754283876209 5.38797602393103 Ru 14.25514072257642 4.22105532389564 4.04783517732174 N 14.65382424365441 6.17538343391092 4.85271395766795 H 13.27443726330260 -0.97183650483572 1.95253468184238 H 15.51255122374571 9.59928295169886 6.41264112771804 Section 6.7 | 281 6.7 SUPPORTING INFORMATION TO 3.6: [CR(DDPD)2]3+: A MOLECULAR, WATER-SOLUBLE, HIGHLY NIR-EMISSIVE RUBY ANALOGUE General procedures CrCl2 (95%, ABCR), deoxyguanosine monophosphate (dGMP) and (HOCH2)3CNH3Cl (Tris-HCl) (Sigma-Aldrich) and 2,2’-bipyridine (bpy) (Alpha Aesar) were purchased from commercial suppliers. The ligand ddpd was synthesized according to a literature procedure.1 [Cr(bpy)3](PF6)2 was prepared similar to a literature procedure.2 Air- or moisture-sensitive reactions were performed in dried glassware under inert gas atmosphere (argon, quality 4.6). Acetonitrile was refluxed over CaH2 and distilled under argon prior to use in these reactions. UV/Vis spectra were recorded on a Varian Cary 5000 spectrometer in 1 cm cuvettes. Emission spectra were recorded on a Varian Cary Eclipse spectrometer (single crystal 1(BF4)33CH3CN and 1(PF6)3 in solution) or on an Edinburgh Instruments spectrometer (FSP 920). Luminescence decay curves in the µs-range were measured with an Edinburgh Instruments spectrometer (FSP 920) using a µs Xe-flashlamp and multi-channel scaling mode. Fluorescence decays in the ns-range were recorded using an Edinburgh Instruments lifetime spectrometer (FLS 920) equipped with a MCP-PMT (R3809U-50, Hamamatsu), and a TCSPC module (TCC 900). A supercontinuum laser (SC400-PP, Fianium) was used for excitation wavelengths > 400 nm, for excitation at 330 nm a ps-laserdiode (EPLED, Edinburgh Instruments) was used. All luminescence measurements were performed using magic angle condition (polarization 0° in the excitation and 54.7° in the emission channel). Luminescence quantum yields were determined using an Ulbricht integrating sphere (Quantaurus-QY C11347- 11, Hamamatsu).3 Relative uncertainty is estimated to be +/- 5 %. Oxygen was removed from the solvents by purging with argon and the oxygen concentration in the sample solutions was measured using an Neofox-GT (sensor Phosphor-R) optical detection system (OceanOptics). For the single crystal absorption and emission spectra a single crystal of 1(BF4)33CH3CN of approximate 0.5 x 0.3 x 0.1 cm3 dimension was placed a microcuvette, covered with heptane and analyzed by absorption and emission spectroscopy. ESI+ mass spectra were recorded on a Micromass QTof Ultima API mass spectrometer with analyte solutions in acetonitrile. Elemental analyses were performed by the microanalytical laboratory of the chemical institutes of the University of Mainz. Electrochemical experiments were performed with a BioLogic SP-50 voltammetric analyser using platinum wire working and counter electrodes and a 0.01 M Ag/AgNO reference electrode. Measurements were carried out at a scan rate of 100 mV s−13 for 1 A. Breivogel, C. Förster, K. Heinze, Inorg. Chem. 2010, 49, 7052-7056. 2 B. R. Baker, B. D. Metha, Inorg. Chem. 1962, 4, 848-854. 3 a) C. Würth, D. Geißler, T. Behnke, M. Kaiser, U. Resch-Genger, Anal. Bioanal. Chem. 2015, 407, 59–78; b) C. Würth, M. G. González, R. Niessner, U. Panne, C. Haisch, U. Resch-Genger, Talanta 2012, 90, 30–37; c) C. Würth, J. Pauli, C. Lochmann, M. Spieles, U. Resch-Genger, Anal. Chem. 2012, 84, 1345–1352. 282 | 6 APPENDIX cyclic voltammetry experiments using 0.1 M [nBu4N][PF6] as supporting electrolyte in acetonitrile. Potentials are given relative to the ferrocene/ferrocenium couple (0.40 V vs. SCE4, E1/2 = 0.90±5 mV under the given conditions). EPR spectra were recorded on a Miniscope MS 300 X-band CW spectrometer (Magnettech GmbH, Germany). Values of g are referenced to Mn2+ in ZnS as external standard (g = 2.118, 2.066, 2.027, 1.906, 1.986, 1.946). Simulations were performed with the EasySpin program package.5 Magnetic susceptibility measurements were carried out with a Quantum Design MPMS-XL7 SQUID magnetometer under an applied magnetic field of 1 T. Experimental susceptibility data were corrected by the underlying diamagnetism using Pascal’s constants. The magnetic contribution of the holder was experimentally determined and substracted from the measured susceptibility data. Density functional theoretical calculations were carried out using the ORCA program package (version 3.0.2).6 Tight convergence criteria were chosen for all calculations (Keywords TightSCF and TightOpt, convergence criteria for the SCF part: energy change 1.0·10–8 Eh, 1-El. energy change 1.0·10–5 E , orbital gradient 1.0·10–5h , orbital rotation angle 1.0·10–5, DIIS Error 5.0·10–7; for geometry optimizations: energy change: 1.0·10–6 Eh, max. gradient 1.0·10–4 Eh bohr–1, RMS gradient 3.0·10–5 E –1h bohr , max. displacement 1.0·10–3 bohr, RMS displacement 6.0·10–4 bohr). All calculations make use of the resolution of identity (Split-RI-J) approach for the coulomb term in combination with the chain-of-spheres approximation for the exchange term (COSX).7 Geometry optimizations were performed using the B3LYP functional8 in combination with Ahlrichs’ split- valence double-ξ basis set def2-SV(P) for all atoms which comprises polarization functions for all non-hydrogen atoms.9 The optimized geometries were confirmed to be local minima on the respective potential energy surface by subsequent numerical frequency analysis (Nimag = 0). TD- DFT calculations were performed based on the B3LYP/def2-SV(P) optimized geometry. The ZORA relativistic approximation10 was used to describe relativistic effects in all calculations. Fifty vertical transitions were calculated in TD-DFT calculations. Explicit counterions and/or solvent molecules were neglected. 4 N. G. Connelly, W. E. Geiger, Chem. Rev. 1996, 96, 877–910. 5 S. Stoll, A. Schweiger, J. Magn. Reson. 2006, 178, 42–55. 6 F. Neese, WIREs Comput Mol Sci 2012, 2, 73–78. 7 a) F. Neese, F. Wennmohs, A. Hansen, U. Becker, Chem. Phys. 2009, 356, 98–109; b) R. Izsák, F. Neese, J. Chem. Phys. 2011, 135, 144105. 8 A. D. Becke, J. Chem. Phys. 1993, 98, 5648-5642. 9 a) A. Schäfer, H. Horn, R. Ahlrichs, J. Chem. Phys. 1992, 97, 2571; b) A. Schäfer, C. Huber, R. Ahlrichs, J. Chem. Phys. 1994, 100, 5829. 10 a) E. van Lenthe, E. J. Baerends, J. G. Snijders, J. Chem. Phys. 1993, 99, 4597; b) C. van Wüllen, J. Chem. Phys. 1998, 109, 392; c) D. A. Pantazis, X.-Y. Chen, C. R. Landis, F. Neese, J. Chem. Theory Comput. 2008, 4, 908–919. Section 6.7 | 283 Crystal Structure Determinations. Intensity data were collected with a Bruker AXS Smart 1000 CCD diffractometer with an APEX II detector and an Oxford cooling system and corrected for absorption and other effects using Mo K radiation ( = 0.71073 Å) at 173(2) K. The diffraction frames were integrated using the SAINT package, and most were corrected for absorption with MULABS.11,12 The structures were solved by direct methods and refined by the full-matrix method based on F2 using the SHELXTL software package.13,14 All non-hydrogen atoms were refined anisotropically, while the positions of all hydrogen atoms were generated with appropriate geometric constraints and allowed to ride on their respective parent carbon atoms with fixed isotropic thermal parameters. See Table S2 for crystal and structure refinement data. 1(PF6)32CH3CN crystallized as small plates resulting in weakly diffracting crystals and a low observed/unique data ratio. Furthermore, 1(PF6)32CH3CN features two independent cations in the unit cell together with the corresponding (partially disordered) counter anions and solvent molecules. The disordered anions have been refined with split-models with the following occupancies for the disordered atoms: 1(PF6)32CH3CN (anion P2, P4, P6: 0.5:0.5), 1(BF4)33CH3CN (anion B2, B3: 0.896(4):0.104(4)). SAME and SADI geometric restraints have been used and the SIMU and DELU instructions in some cases to enable anisotropic refinement of the disordered anions. Crystallographic data (excluding structure factors) for the structure reported in this paper have been deposited with the Cambridge Crystallographic Data Centre as supplementary publication no CCDC-1059802 [1(BF4)33CH3CN] and CCDC-1059801 [1(PF6)32CH3CN]. Copies of the data can be obtained free of charge upon application to CCDC, 12 Union Road, Cambridge CB2 1EZ, U.K. [fax (0.44) 1223-336-033; e-mail deposit@ccdc.cam.ac.uk]. Synthesis of [Cr(ddpd)2](BF4)3 [1(BF4)3]: Anhydrous chromium(II) chloride (220 mg, 1.95 mmol) and ddpd1 (810 mg, 2.78 mmol) were dissolved in deaerated CH3CN/H2O (1:1, 45 ml). The deep green solution was stirred under argon for 12 h. Addition of a solution of ammonium tetrafluoroborate (5 ml, 1.06 M in H2O) yielded a green precipitate. The solid was removed by filtration and extracted once with diethyl ether. The solvents of the orange solution were removed under reduced pressure. The orange residue was dissolved in CH3CN. Addition of diethyl ether yielded orange crystals, which were dried under reduced pressure. Diffusion of diethyl ether into a concentrated CH3CN solution yielded large diffraction quality crystals. Yield: 970 mg (1.08 mmol, 78 %). Synthesis of [Cr(ddpd)2](PF6)3 [1(PF6)3] (route I): Anhydrous chromium(II) chloride (90 mg, 0.732 mmol) and ddpd1 (304 mg, 1.04 mmol) were dissolved in deaerated water (50 ml). The deep green 11 SMART Data Collection and SAINT-Plus Data Processing Software for the SMART System, various versions; Bruker Analytical X-ray Instruments, Inc.: Madison, WI, 2000. 12 R. H. Blessing, Acta Crystallogr. 1995, A51, 33–38. 13 G. M. Sheldrick, SHELXTL, version 5.1; Bruker AXS: Madison, WI, 1998. 14 G. M. Sheldrick, SHELXL-97; University of Göttingen: Göttingen, Germany, 1997. 284 | 6 APPENDIX solution was stirred under argon for 15 h. Addition of a solution of potassium hexafluorophosphate (10 ml, 0.17 M in H2O) yielded a green and orange colored precipitate. The solids were collected by filtration and washed once by diethyl ether. The orange residue was dissolved in CH3CN. Addition of diethyl ether yielded orange crystals, which were dried under reduced pressure. Diffusion of diethyl ether into a concentrated CH3CN solution yielded diffraction quality crystals. Yield: 200 mg (0.187 mmol, 36 %). Synthesis of [Cr(ddpd)2](PF6)3 [1(PF6)3] (route II): Potassium hexafluorophosphate (97 mg, 0.527 mmol) was added to a concentrated aqueous solution of [Cr(ddpd)2](BF4)3 (98.6 mg, 0.110 mmol). The resulting orange precipitate was collected by filtration and dissolved in acetonitrile. Slowly adding diethyl ether resulted in precipitation of yellow crystals. Yield: 90.7 mg (0.085 mmol, 77 %). Synthesis of [Cr(bpy)3](ClO4) 23 : 2,2’-Bipyridine (750 mg , 4.80 mmol) was added to a solution of anhydrous chromium(II) chloride (99 mg, 0.806 mmol) in deaerated 0.1 M perchloric acid (40 ml). After stirring the resulting black-purple suspension for 15 minutes at room temperature air was bubbled through the reaction mixture for 5 hours. Yellow crystals precipitated from the yellow solution overnight. The crystals were collected by filtration, washed with ethanol and dried under reduced pressure. Yield: 560 mg (0.684 mmol, 84 %). Synthesis of [Cr(bpy)3](PF6)3: Potassium hexafluorophosphate (480 mg, 2.61 mmol) was added to a concentrated aqueous solution of [Cr(bpy)3](ClO4)3 (410 mg, 0.500 mmol). The resulting yellow precipitate was collected by filtration and dissolved in acetonitrile. Slowly adding diethyl ether resulted in precipitation of yellow crystals. Yield: 348 mg (0.364 mmol, 73 %). Stability tests: Isoabsorptive solutions (at 430 nm) of 1(PF6)3 and [Cr(bpy)3](PF6)3 in 0.1 M [nBu4N]Cl H2O/MeCN (1:1) solution were irradiated with an LED torch at 430 nm under aerobic conditions. Isoabsorptive solutions (at 430 nm) of 1(PF6)3 and [Cr(bpy)3](PF6)3 in 0.1 M [nBu4N]Cl H2O/MeCN (1:1) solution with pH = 11.4 adjusted with [nBu4N](OH) were irradiated under the same conditions. The reaction progress was monitored at the respective emission maximum (777 and 727 nm) over a time of 5 hours. Quenching with dGMP: 3.3 x 10-5 M solutions of 1(PF6)3 and [Cr(bpy)3](PF6)3 in 50 mM aqueous Tris-HCl buffer were titrated with a 6.45 mM solution of deoxyguanosine monophosphate (dGMP) dissolved in the same buffer. Emission quenching was monitored at the respective emission maximum (777 and 727 nm) up to a dGMP concentration of 3.1410-4 M. Section 6.7 | 285 Table S1. Analytical data of 1(BF4)3 and 1(PF6)3. 1(BF4)3 1(PF6)3 molecular formula C34H34B3CrF12N10 C34H34CrF18N10P3 molecular mass 895.11 g mol-1 1069.59 g mol-1 solubility in CH3CN 122 g l–1 (0.136 mol l–1) 222 g l–1 (0.208 mol l–1) solubility in H2O 42.9 g l–1 (0.048 mol l–1) 1.73 g l–1 (0.0016 mol l–1) MS (ESI): m/z = 171.5 (13) [M−3BF ]3+, 171.5 (7) [M−3PF ]3+4 6 , 292.1 (47) [ddpd+H]+, 291.1 (12) [ddpd]+, 362.1 (23) [M−2BF 2+4] , 389.6 (4) [M−2PF6]2+, 808.2 (100) [M−BF +4] , 924.1 (100) [M−PF6]+, 1703.4 (15) [2M−BF4]+ 1994.2 (22) [2M−PF +6] IR (KBr): 𝜈 = 1606 (vs), 1585 (s), 1568 1609 (vs), 1585 (s), 1570 (m), 1497 (vs), 1455 (s), (m), 1499 (vs), 1455 (s), 1435 (vs), 1365 (w), 1343 1437 (vs), 1369 (w), 1346 (s), 1237 (m), 1141 (s), (s), 1240 (m), 1178 (w), 1095-1035 (vs br, BF) cm–1 1141 (s), 838 (vs, PF) cm–1 Magnetism (300 K): χT = - 1.833 cm3 K mol–1 EPR (77 K) in CH3CN gav = - 1.990 (broad) UV/Vis (CH –1 –13CN): λmax (ε/M cm ) 436 (3770, LMCT+MC), 436 (4095, LMCT+MC), = 315 (sh, 25500, LMCT+MC), 315 (sh, 25700, 302 (28100, LMCT), LMCT+MC), 220 (sh, 53600, ππ*) nm 301 (28700, LMCT), 218 (56000, ππ *) nm UV/Vis (H2O): λmax (ε/M–1 cm–1) = 435 (3980, LMCT+MC), 436 (3530, LMCT+MC), 315 (sh, 24700, LMCT+MC), 315 (sh, 21840,LMCT+MC), 301 (27600, LMCT), 301 (24500, LMCT), 217 (sh, 53400, ππ *) nm 219 (48000, ππ *) nm UV/Vis (single crystal): 776 (<1), 736 (<1), 697(<1) - λmax (ε/M–1 cm–1) = nm Emission (CH3CN, λexc = 435): 776 (1.0), 739 (0.20), 500 776 (1.0), 738 (0.2), 500 λmax (rel. intensity) = (very weak) nm (very weak) nm Emission (H2O, λexc = 435): 777 (1.0), 739 (0.20), 500 - λmax (rel. intensity) = (very weak) nm Emission (single crystal, λexc = 778 (1.0), 740 (0.2), 594 - 435): λmax (rel. intensity) = (0.2) nm Emission (single crystal, λexc = 443 µs - 435): = CV ([nBu4N][PF6]/CH3CN, vs. Fc): E½ - –1.11 V (CrIII/CrII) = Elemental analysis found / calcd. C 44.98 (45.22) C 37.90 (38.18) H 4.17 (3.83) H 2.95 (3.20) N 15.04 (15.65) N 12.98 (13.10) Photographs of crystals 286 | 6 APPENDIX Table S2. Summary of X-ray data of 1(BF4)3x3CH3CN and 1(PF4)3x2CH3CN. 1(BF4)2x3CH3CN 1(PF4)3x2CH3CN Empirical formula C40H43B3CrF12N13 C38H40CrF18N11P3 Formula weight 1018.30 1151.73 Crystal color, habit red block orange plate Crystal dimensions / mm 0.74 x 0.50 x 0.34 0.32 x 0.09 x 0.07 Crystal system monoclinic monoclinic Space group Pn P21/c a / Å 11.5125(8) 28.2212(12) b / Å 16.5554(11) 11.7062(5) c / Å 12.9721(9) 34.0629(13) α / ° 90 90 β / ° 111.890(2) 124.479(3) γ / ° 90 90 V / Å3 2294.2(3) 9312.7(7) Z 2 8 F(000) 1042 4664 Density (calcd) / g cm–3 1.474 1.643 Absorption coefficient µ / mm–1 0.345 (MULABS) 0.467 (MULABS) Theta range / ° 2.02 – 27.90 1.20 – 28.04 Index ranges –14 ≤ h ≤ 15 –37 ≤ h ≤ 36 –21 ≤ k ≤ 21 –13 ≤ k ≤ 15 –17 ≤ l ≤ 17 –44 ≤ l ≤ 44 Reflections collected 27213 90048 Independent reflections 9828 (Rint = 0.0504) 22407 (Rint = 0.1510) Observed reflections 9828 22407 Parameters, restraints 666, 22 1439, 688 Max. / min. transmission 0.784 / 0.892 0.9681 / 0.8650 Goodness-of-fit on F2 1.041 1.042 Largest difference peak and hole / e Å–3 0.315 / –0.347 1.647 / –1.018 R1 (I>2σ(I)) 0.0409 0.1087 R1 (all data) 0.0439 0.2304 wR2 (I>2σ(I)) 0.1105 0.3165 wR2 (all data) 0.1127 0.3529 absolute structure parameter –0.005(12) - Section 6.7 | 287 Table S3. Selected distances [Å] and angles [deg] of 1(BF4)2x3CH3CN and 1(PF4)3x2CH3CN. 1(BF4)2x3CH3CN 1(PF4)3x2CH3CN 1(PF4)3x2CH3CN (molecule A) (molecule B) Cr1-N1 2.0485 (0.0018) 2.0410 (0.0058) 2.0395 (0.0069) Cr1-N3 2.0393 (0.0018) 2.0535 (0.0065) 2.0296 (0.0068) Cr1-N5 2.0394 (0.0019) 2.0280 (0.0062) 2.0327 (0.0068) Cr1-N6 2.0446 (0.0017) 2.0398 (0.0067) 2.0400 (0.0061) Cr1-N8 2.0444 (0.0018) 2.0538 (0.0071) 2.0465 (0.0066) Cr1-N10 2.0485 (0.0018) 2.0476 (0.0068) 2.0302 (0.0059) N1-Cr1-N3 85.13 (0.08) 85.74 (0.25) 86.78 (0.28) N1-Cr1-N5 170.86 (0.08) 172.27 (0.25) 172.57 (0.27) N1-Cr1-N6 91.06 (0.07) 89.98 (0.25) 90.45 (0.27) N1-Cr1-N8 95.23 (0.08) 95.47 (0.25) 93.71 (0.27) N1-Cr1-N10 89.49 (0.07) 90.37 (0.26) 89.23 (0.27) N3-Cr1-N5 85.74 (0.07) 86.71 (0.26) 86.03 (0.27) N3-Cr1-N6 95.10 (0.07) 93.89 (0.27) 95.33 (0.26) N3-Cr1-N8 178.94 (0.08) 178.61 (0.28) 179.01 (0.29) N3-Cr1-N10 94.02 (0.07) 95.13 (0.27) 91.60 (0.26) N5-Cr1-N6 89.68 (0.07) 92.20 (0.27) 88.38 (0.27) N5-Cr1-N8 89.68 (0.07) 92.10 (0.26) 93.51 (0.27) N5-Cr1-N10 91.22 (0.07) 88.64 (0.27) 92.82 (0.27) N6-Cr1-N8 85.89 (0.07) 85.42 (0.27) 85.54 (0.25) N6-Cr1-N10 170.88 (0.07) 170.97 (0.27) 173.03 (0.27) N8-Cr1-N10 84.99 (0.07) 85.56 (0.27) 87.54 (0.26) 288 | 6 APPENDIX Figure S1 Short anion…cation contacts (Å) in crystals of a) 1(BF4)2x3CH3CN and b) 1(PF4)3x2CH3CN (F and H atoms omitted) and structure of c) 1(BF4)2x3CH3CN and d) 1(PF4)3x2CH3CN with thermal ellipsoids at 30 % probability. Section 6.7 | 289 Figure S2 a) ESI mass spectrum of 1(BF4)3 including experimental and calculated isotopic pattern of [M-BF ]+4 and b) ESI mass spectrum of 1(PF6)3 including experimental and calculated isotopic pattern of [M-PF +6] . a) b) 290 | 6 APPENDIX Figure S3 IR spectrum of 1(BF4)3 as KBr disk. Figure S4 IR spectrum of 1(PF6)3 as KBr disk. Section 6.7 | 291 Figure S5 UV/Vis spectra of 1(BF4)3 a) in CH3CN and b) in H2O. Figure S6 UV/Vis spectrum of 1(PF6)3 in CH3CN. 292 | 6 APPENDIX Figure S7 a) Emission spectrum of 1(BF4)3 in air-saturated CH3CN in the region 400 – 600 nm (exc = 430 nm) and b) decay curves of the broad band luminescence at 500 nm in air-saturated and oxygen-free acetonitrile (exc = 450 nm, obs = 500 nm). The fast initial decay resembling the pulse profile of the excitation light pulse (IRF) is caused by Raman and Rayleigh scattered excitation light. Section 6.7 | 293 Figure S8 Absorption factor, excitation (obs = 775 nm) and emission spectrum (exc = 435 nm) of 1(BF4)3 in CH3CN (inset shows decay curve in the presence and absence of O2, exc = 435 nm, obs = 775 nm). 294 | 6 APPENDIX Figure S9 Temperature dependent emission spectra of 1(BF4)3 in butyronitrile (100 K – 300 K). Section 6.7 | 295 Figure S10 Absorption spectrum of a single crystal of 1(BF4)33CH3CN and photographs of the measured single crystal of 1(BF4)33CH3CN. Figure S11 Emission spectrum of a single crystal of 1(BF4)33CH3CN. 296 | 6 APPENDIX Figure S12 a) Absorption and emission spectra (exc = 330 nm) of ddpd in CH3CN and b) fluorescence decay curve (exc = 330 nm, obs = 398 nm). Section 6.7 | 297 Figure S13 Cyclic voltammogram of 1(PF6)3 in 0.1 M [nBu4N][PF6]/CH3CN, Pt electrodes, referenced against ferrocene. 298 | 6 APPENDIX Figure S14 EPR spectrum of 1(PF6)3 at 77 K in CH3CN, frequency 9.410 GHz. Figure S15 Stern-Volmer plot of 1(BF4)3 in H2O by quenching with O2 (exc = 435 nm, obs = 775 nm). Dotted line is a linear regression of the data. Section 6.7 | 299 Figure S16 UV/Vis absorption spectra of 1(BF4)3 under air in H2O at different pH over time. Figure S17 Traces of the intensity of the absorption band at 435 nm of 1(BF4)3 under air in H2O at different pH over time. 300 | 6 APPENDIX Figure S18 a) Emission spectra of isoabsorptive solutions of 1(BF4)3 and [Cr(bpy)3](PF6)2 in 0.1 mM [nBu4N]Cl H2O/MeCN (1:1) solution under aerobic conditions with 430 nm irradiation over time, b) emission spectra of isoabsorptive solutions of 1(BF4)3 and [Cr(bpy)3](PF6)2 in H2O/MeCN (1:1) solution with pH = 11.4 adjusted with [nBu4N](OH) under aerobic conditions with 430 nm irradiation over time and c) traces of the emission intensity over time. Section 6.7 | 301 Figure S19 Stern-Volmer plots of 1(PF4)3 (obs = 777 nm, red) and [Cr(bpy)3](PF6)3 (obs = 727 nm,blue) by quenching with dGMP in H2O (exc = 435 nm). Dotted line is a linear regression of the data. 302 | 6 APPENDIX Figure S20 TD-DFT calculated transitions of 13+ (B3LYP, RIJCOSX, Def2-SVP/J, Def2-SVP, ZORA), assignments and corresponding difference electron densities ES − GS at a contour value of 0.005 (purple lobes indicate loss, orange lobes show increase of electron density upon excitation, hydrogen atoms omitted for clarity). # λ / nm character from to difference electron density 1 441.7 LMCT p (amine-N) t2g (dxy) 2 439.0 MC t2g (dxy) eg (dx2-y2) 3 436.9 MC t2g (dxz) eg (dz2) 4 431.8 LMCT p (amine-N) t2g (dxz) 5 428.3 LMCT p (amine-N) t2g (dxy) z 6 427.7 MC t2g (dxz) eg (dx2-y2) y x 7 422.2 LMCT p (amine-N) t2g (dxz) 8 409.8 LMCT p (amine-N) t2g (dyz) Section 6.7 | 303 9 399.5 LMCT p (amine-N) t2g (dyz) 10 375.7 LMCT p (amine-N) t2g (dxy) 11 373.4 LMCT p (amine-N) t2g (dxy) 12 369.5 LMCT p (amine-N) eg (dz2) 13 367.6 LMCT p (amine-N) eg (dx2-y2) 14 363.6 LMCT p (amine-N) eg (dx2-y2 + dz2) 15 358.4 MC t2g (dyz) eg (dz2) 16 357.3 LMCT p (amine-N) t2g (dxz) 17 356.0 LMCT p (amine-N) t2g (dxz) 304 | 6 APPENDIX 18 352.3 LMCT p (amine-N) t2g (dxz) 19 350.8 LMCT p (amine-N) t2g (dyz) 20 348.9 LMCT + ππ* p (amine-N) eg (dx2-y2) 21 348.0 LMCT p (amine-N) t2g (dyz) 22 347.7 MC t2g (dxy) eg (dx2-y2) 23 339.2 LMCT p (amine-N) eg (dz2) 24 338.8 LMCT p (amine-N) t2g (dyz) 25 336.4 LMCT p (amine-N) t2g (dyz) 26 333.0 LMCT p (amine-N) t2g (dxy) Section 6.7 | 305 27 329.2 LMCT p (amine-N) t2g (dxy) 28 327.1 MC t2g (dxy) eg (dz2) 29 323.5 LMCT p (amine-N) t2g (dxy + dxz + dyz) 30 322.7 LMCT p (amine-N) t2g (dxz) 306 | 6 APPENDIX Figure S21 DFT calculated spin density of 13+ (4A2 ground state) (B3LYP, RIJCOSX, Def2-SVP/J, Def2- SVP, ZORA); isosurface value 0.01 a.u.; hydrogen atoms omitted for clarity; distances in Å. Figure S22 DFT calculated spin density of 13+ (2E state) (B3LYP, RIJCOSX, Def2-SVP/J, Def2-SVP, ZORA); isosurface value 0.01 a.u.; hydrogen atoms omitted for clarity; distances in Å. Section 6.7 | 307 Figure S23 DFT calculated spin density of 13+ (2T2 state) (B3LYP, RIJCOSX, Def2-SVP/J, Def2-SVP, ZORA); isosurface value 0.01 a.u. (hydrogen atoms omitted for clarity). Figure S24 DFT calculated spin density of 13+ (4T2 state) (B3LYP, RIJCOSX, Def2-SVP/J, Def2-SVP, ZORA); isosurface value 0.01 a.u. (hydrogen atoms omitted for clarity). 308 | 6 APPENDIX Figure S25 DFT calculated spin density of 12+ (5E state) (B3LYP, RIJCOSX, Def2-SVP/J, Def2-SVP, ZORA); isosurface value 0.01 a.u. (hydrogen atoms omitted for clarity). Figure S26 DFT calculated spin density of 12+ (3T1 state) (B3LYP, RIJCOSX, Def2-SVP/J, Def2-SVP, ZORA); isosurface value 0.01 a.u. (hydrogen atoms omitted for clarity). Section 6.7 | 309 Table S4 Selected distances [Å] and angles [deg] of DFT optimized geometries (B3LYP, RIJCOSX, Def2-SVP/J, Def2-SVP, ZORA). 13+ 12+ 4A 2E 2T 42 2 T2 5E 3T 1 Cr1-N1 / Å 2.094 2.087 2.084 2.421 2.282 2.095 Cr1-N3 / Å 2.072 2.065 2.040 2.240 2.116 2.094 Cr1-N5 / Å 2.094 2.087 2.084 2.421 2.282 2.095 Cr1-N6 / Å 2.094 2.087 2.084 2.142 2.282 2.095 Cr1-N8 / Å 2.072 2.065 2.040 2.144 2.116 2.094 Cr1-N10 / Å 2.094 2.087 2.084 2.142 2.282 2.095 N1-Cr1-N3 / ° 86.61 86.71 85.89 78.43 83.32 86.41 N1-Cr1-N5 / ° 173.23 173.41 171.77 156.86 166.64 172.81 N1-Cr1-N6 / ° 92.29 92.60 91.43 92.48 95.88 92.21 N1-Cr1-N8 / ° 93.39 93.30 94.11 101.58 96.69 93.59 N1-Cr1-N10 / ° 88.11 87.78 89.16 90.00 85.68 88.24 N3-Cr1-N5 / ° 86.61 86.70 85.89 78.43 83.32 86.41 N3-Cr1-N6 / ° 93.39 93.30 94.11 96.23 96.68 93.59 N3-Cr1-N8 / ° 180.00 180.00 180.00 179.99 179.99 180.00 N3-Cr1-N10 / ° 93.39 93.30 94.12 96.22 96.67 93.60 N5-Cr1-N6 / ° 88.11 87.78 89.16 90.00 85.67 88.24 N5-Cr1-N8 / ° 93.39 93.30 94.12 101.56 96.68 93.60 N5-Cr1-N10 / ° 92.29 92.60 91.43 95.51 95.89 92.21 N6-Cr1-N8 / ° 86.61 86.71 85.89 83.78 83.32 86.41 N6-Cr1-N10 / ° 173.23 173.41 171.77 167.55 166.65 172.81 N8-Cr1-N10 / ° 86.61 86.70 85.89 83.78 83.32 86.41 | 311 7 ACKNOWLEDGMENTS | 313 8 CURRICULUM VITAE Christoph Kreitner Date of Birth: 21.09.1988 Place of Birth: Wiesbaden, Germany Nationality: German EDUCATION Johannes Gutenberg-University Mainz, Germany Doctorate (Chemistry) 01/2013 – 06/2016 „Synthesis and Characterization of new Ruthenium complexes: Functional chromophores and electron transfer relays” Advisor: Johannes Gutenberg-University Mainz, Germany Diploma (Chemistry, very good, 1.0) 10/2007 – 09/2012 Main: Inorganic chemistry; Elective: Theoretical chemistry Diploma thesis (very good, 1.0) 01/2012 – 09/2012 „Synthesis, experimental and theoretical characterization of new mixed-valent ruthenium complexes” Advisor: Teaching Assistant, inorganic and physical chemistry 10/2008 – 08/2011 University of Toronto, Toronto, Canada Research internship 09/2010 – 03/2011 “Reactivity of Frustrated Lewis Pairs with Lactones and Lactide” Advisor: Gymnasium Theresianum Mainz, Germany Abitur (very good, 1.1) 08/1999 – 03/2007 Research paper (chemistry, very good, 1.0) 02/2006 “The basics of coordination chemistry on the basis of the synthesis of tetraamminecopper(II)-sulfate” BEGYS (school intern talent training) 09/2001 – 07/2004 International Language School, Cannes, France French language course, 2 weeks 08/2005 Pinkerton Academy, Derry (NH), USA Student exchange, 4 weeks 09/2004 314 | 8 CURRICULUM VITAE AWARDS AND SCHOLARSHIPS Gutenberg Academy, Johannes-Gutenberg University Mainz since 03/2014 Junior Membership Awarded to the 25 best PhD students of the University Materials Science in Mainz (MAINZ), Graduate School of Excellence since 05/2013 Graduate Student Scholarship Poster Prize of the International Union of Pure and Applied Chemistry 09/2013 (IUPAC) at the GDCh Wissenschaftsforum 2013, Darmstadt, Germany Adolf Todt Award 2013 of the Johannes Gutenberg-University Mainz 04/2013 for an excellent diploma thesis Travel stipend of the Graduate School Materials Science in Mainz 09/2010 – 02/2011 Award of the GdCh for "the best Abitur in Chemistry“ 03/2007 Award of the “Förderverein Theresianum e.V.” “for exemplary behavior 03/2007 within the school community, for outstanding scholastic achievements and for the longtime engagement in the instrumental ensembles of the school“ LANGUAGE SKILLS German native language English business fluent French very good (CEFR level C1) Latin Latin proficiency certificate SUMMER SCHOOLS AND WORKSHOPS Max-Planck-Institute for Chemical Energy Conversion, Mülheim, Germany Summer School 09/2014 “Methods in Molecular Energy Research: Theory and Spectroscopy” University of Manchester, UK Summer School 05/2013 “Introductory workshop on the Theory and Practice of EPR spectroscopy” Johannes Gutenberg-University Mainz, Germany Workshops Poster Design and Communication 10/2013 Presenting in English 02/2014 Intercultural Communication 09/2014 Introductory Workshop on High Performance Computing 11/2014 Mainz, 1st of June, 2016 Section 8.1 | 315 8.1 LIST OF PUBLICATIONS C. Kreitner, S. J. Geier, L. J. E. Stanlake, C. B. Caputo, D. W. Stephan, Ring openings of lactone and ring contractions of lactide by frustrated Lewis pairs. Dalton Trans. 2011, 40, 6771–6777. C. B. Caputo, S. J. Geier, E. Y. Ouyang, C. Kreitner, D. W. Stephan, Chloro- and phenoxy-phosphines in frustrated Lewis pair additions to alkynes. Dalton Trans. 2012, 41, 237–242. A. Breivogel, C. Kreitner, K. Heinze, Redox and Photochemistry of Bis(terpyridine)ruthenium(II) Amino Acids and Their Amide Conjugates - from Understanding to Applications. Eur. J. Inorg. Chem. 2014, 2014, 5468–5490. – Highlighted Cover Article C. Kreitner, M. Grabolle, U. Resch-Genger, K. Heinze, Dual Emission and Excited-State Mixed- Valence in a Quasi-Symmetric Dinuclear Ru–Ru Complex. Inorg. Chem. 2014, 53, 12947–12961. C. Kreitner, E. Erdmann, W. W. Seidel, K. Heinze, Understanding the Excited State Behavior of Cyclometalated Bis(tridentate)ruthenium(II) Complexes: A Combined Experimental and Theoretical Study. Inorg. Chem. 2015, 54, 11088–11104. – Highlighted Cover Article S. Otto, M. Grabolle, C. Förster, C. Kreitner, U. Resch-Genger, K. Heinze, [Cr(ddpd) 3+2] : ein molekulares, wasserlösliches, hoch NIR-lumineszentes Rubin-Analogon. Angew. Chem. 2015, 127, 11735–11739. [Cr(ddpd)2]3+: A Molecular, Water-Soluble, Highly NIR-Emissive Ruby Analogue. Angew. Chem. Int. Ed. 2015, 54, 11572–11576. Highlighted as Hot Paper C. Kreitner, K. Heinze, The photochemistry of mono- and dinuclear cyclometalated bis(tridentate)ruthenium(II) complexes: Dual excited state deactivation and dual emission. Dalton Trans. 2016, 45, 5640–5658. C. Kreitner, A. K. Mengel, T. K. Lee, W. Cho, K. Char, Y. S. Kang, K. Heinze, Strongly Coupled Cyclometalated Ruthenium Triarylamine Chromophores as Sensitizers for DSSCs. Chem. Eur. J. 2016, published online 19 May 2016, DOI: 10.1002/chem.201601001. C. Kreitner, K. Heinze, Excited State Decay of Cyclometalated Polypyridine Ruthenium Complexes: Insight from Theory and Experiment, 2016, submitted. 316 | 8 CURRICULUM VITAE 8.2 CONFERENCE CONTRIBUTIONS 09/2013 GDCh Wissenschaftsforum, Darmstadt Germany Poster presentation 08/2014 41st International Conference on Coordination Chemistry, Singapore Poster presentation 09/2014 Summer School “Methods in Molecular Energy Research: Theory and Spectroscopy”, Gelsenkirchen, Germany Poster presentation 09/2014 Vortragstagung der Wöhlervereinigung, Saarbrücken, Germany Poster presentation 03/2015 Koordinationschemie-Tagung, Paderborn, Germany Oral presentation