Towards Photoactive Manganese(IV) and Nickel(II) Complexes Dissertation zur Erlangung des Grades „Doktor der Naturwissenschaften“ im Promotionsfach Chemie am Fachbereich Chemie, Pharmazie, Geographie und Geowissenschaften der Johannes Gutenberg-Universität Mainz Nathan R. East geboren in Swansea Mainz, 2023 The present work was carried out in the period from November 2019 to June 2023 in the Department of Chemistry (formerly Institute for Inorganic Chemistry and Analytical Chemistry) at the Johannes Gutenberg University in Mainz under the supervision of Prof. Dr. Katja Heinze. Dean Prof. Dr. Eva Rentschler 1st Supervisor: Prof. Dr. Katja Heinze 2nd Supervisor: Prof. Dr. Eva Rentschler Day of oral examination: ________________ I, Nathan Roy East, hereby certify that I have written this work independently and have not used any sources or aids other than those indicated. I have marked all explanations that were taken from others literally or analogously. _____________________ _____________________ (Date) (Signature) Abstract With increased emphasis on sustainability, photochemistry with earth abundant 3d transition metal complexes has become increasingly important. Primary focus over the last decades has centered on optimising and developing charge transfer phosphorescence. However, in recent years octahedral 3d transition metal complexes containing low energy metal-centered states have received a great deal of attention, particularly with the [Cr(ddpd) 3+2] (ddpd = N,N’- dimethyl-N,N’-dipyridin-2-ylpyridine-2,6-diamine) as prime example. Such successes have guided very recent research with [V(ddpd) ]3+2 as the first octahedral vanadium(III) complex to display room temperature phosphorescence. With the prospect of d2, d3, d4 and d8 electron configurations possessing low energy metal-centered states potentially capable of phosphorescence, research with other 3d metals in relevant oxidation states is of particular interest. To date only one d3 manganese(IV) and no d8 nickel(II) complexes show luminescence from metal-centered states. Part one of this work details the synthesis and characterization of an octahedral MnIV complex [Mn(dgpy) ]4+2 (dgpy = 2,6-diguanidylpyridine) with a d3 electron configuration analogous to chromium(III). This begins with the synthesis of a manganese(II) precursor [Mn(dgpy) ]2+2 , and oxidation to the MnIV complex via a MnIII intermediate. Structural, magnetic, spectroscopic and computational studies for the complete series of oxidation states +II to +IV provides important information to aid in synthesis and design of potentially photoactive manganese(IV) complexes. It is shown that six-membered chelating ligands with mixture of strongly donating and accepting moieties are required to stabilise such a labile electron-rich d5 MnII, Jahn-Teller distorted d4 MnIII and electron-poor d3 MnIV complexes. The photophysics and photochemistry of [Mn(dgpy) )4+2 are subsequently described and documented in part two. The photophysics of such MnIV complexes are poorly understood due to extreme rarity. The present investigation reveals only the second example of MnIV phosphorescence, with luminescence in the low energy NIR-II region (1435 nm). The excited 2LMCT/2MC state (0.86 eV) of this complex was found to be long lived enough (1.6 ns) to participate in bimolecular chemistry. [Mn(dgpy) )4+2 was discovered to be strongly dual state photooxidative following NIR irradiation and is able to oxidise naphthalene dynamically via a 2LMCT/2MC state and also more difficult substrates like benzene statically via a 4LMCT state (1.46 eV). Bimolecular quenching of this photoreactive complex gives valuable insight into further photophysical dynamics and potential design of future photoactive MnIV transition metal complexes capable of bimolecular reactivity following low energy excitation (850 nm). The third part is an investigation into understanding the requirements necessary for a metal- centered emission from octahedral d8 nickel(II) complexes. To date metal-centered emission from octahedral nickel(II) is undocumented; and here an investigation is made to discover why. A strong ligand field is required to get the correct excited state ordering i.e. the intraconfigurational singlet states being lowest in energy. However, there is a limit to the I ligand field strength that can be imposed on d8 systems and still maintain an octahedral coordination, with advantageous excited state ordering. To further examine the effect of ligand field strength on excited state ordering the series of complexes [Ni(dgpy) )2+2 , [Ni(terpy) 2+ 2) (terpy = 2,2';6',2"-terpyridine), [Ni(phen) )2+3 (phen = 1,10-phenanthroline), [Ni(ddpd) )2+ and [Ni(tpe) ]2+2 2 (tpe = 1,1,1-tris(pyrid-2-yl)ethane) are synthesized and structural and electronic properties examined. Additionally, the influence of increased hydrostatic pressure on ligand field states of [Ni(ddpd) )2+2 is also investigated, to evaluate if the inter- and intraconfigurational states can be further separated. This investigation gives valuable information in the pursuit of metal-centered spin-flip emissive octahedral NiII complexes. Part four builds on part three by looking at the impact of increasing ligand field strength on NiII ligand field states. This is done by firstly increasing σ-donation to destabilize the e *g orbitals, with synthesis of the carbene complex [Ni(CNC)(NCN)]2+ (CNC = 1,1’-(pyridin-2,6- diyl)bis(3-methyl-1H-imidazol-3-ylidene)) and NCN = (1,3-bis(2-pyridyl)imidazolylidene). With increased σ-donation it is also conceivable that the complex will adopt square planar geometry, thus structural confirmation of an octahedral environment is the first step, and further characterization will follow. An alternative method of increasing ligand field strength is using π-acceptor ligands to stabilize the t2g orbitals. This method will also be shown with the synthesis of [Ni(dcpp)2]2+ (dcpp = (2,6-bis(2-carboxypyridyl)pyridine)). The dcpp ligand contains two π-accepting carbonyl groups which enhance the π-accepting ability of the ligand. A Lewis acid (Sc[OTf]3] will then be coordinated to the ligand carbonyl groups, and a second coordination sphere will give further insight into the influence of increasing ligand π- acceptance on ligand field states of [Ni(dcpp) 2+2] . This investigation aims to further understand the ligand requirements for NiII in the pursuit of spin-flip emission. II Kurzzusammenfassung Mit zunehmendem Fokus auf den Aspekt der Nachhaltigkeit hat Photochemie mit auf gut verfügbaren 3d-Metallen basierenden Übergangsmetallkomplexen an Bedeutung gewonnen. In den letzten Jahrzehnten lag der Schwerpunkt auf der Optimierung und Entwicklung von charge-transfer-Photosensibilisatoren. In jüngeren Jahren haben jedoch oktaedrische 3d- Übergangsmetallkomplexe, die niederenergetische metallzentrierte Zustände aufweisen, viel Aufmerksamkeit erhalten, insbesondere [Cr(ddpd) ]3+2 (ddpd = N,N'-Dimethyl-N,N'-dipyridin- 2-ylpyridin-2,6-diamin) als Paradebeispiel. Im Rahmen nachfolgender Forschungen konnte [V(ddpd)2]3+ als erster oktaedrischer Vanadium(III)-Komplex, der bei Raumtemperatur Phosphoreszenz zeigt, synthetisiert werden. Im Hinblick auf d2-, d3-, d4- und d8- Elektronenkonfigurationen, die niederenergetische metallzentrierte Zustände besitzen, die potenziell zur Phosphoreszenz fähig sind, ist die Forschung mit anderen 3d-Metallen in relevanten Oxidationszuständen von besonderem Interesse. Bislang sind nur ein d3- Mangan(IV)- und keine d8-Nickel(II)-Komplexe bekannt, die Lumineszenz aus metallzentrierten Zuständen zeigen. Der erste Teil dieser Arbeit beschreibt die Synthese und Charakterisierung eines oktaedrischen MnIV-Komplexes [Mn(dgpy) 4+ 32] (dgpy = 2,6-diguanidylpyridin) mit einer d - Elektronenkonfiguration analog zu Chrom(III). Dies beginnt mit der Synthese eines Mangan(II)- Vorläufers [Mn(dgpy) ]2+ und der Oxidation zum MnIV-Komplex über ein MnIII2 - Zwischenprodukt. Strukturelle, magnetische, spektroskopische und theoretische Studien für die gesamte Reihe der Oxidationsstufen +II bis +IV liefern wichtige Informationen für die Synthese und das Design potenziell photoaktiver Mangan(IV)-Komplexe. Es wird gezeigt, dass sechsgliedrige Chelatliganden mit einer Mischung aus starken Akzeptor- und Donoreigenschaften erforderlich sind, um die unterschiedlichen Oxidationsstufen des Mangans zu stabilisieren. Die Photophysik und Photochemie von [Mn(dgpy) )4+2 wird anschließend im zweiten Teil beschrieben. Über die Photophysik solcher MnIV-Komplexe ist aufgrund der geringen Anzahl von Beispielen nur wenig bekannt. Die vorliegende Untersuchung zeigt erst das zweite Beispiel für MnIV-Phosphoreszenz, mit Lumineszenz im niederenergetischen NIR-II-Bereich (1435 nm). Der angeregte 2LMCT/2MC-Zustand (0,86 eV) dieses Komplexes erwies sich als langlebig genug (1,6 ns), um bimolekulare Reaktionen zu ermöglichen. Es wurde festgestellt, dass [Mn(dgpy) )4+2 nach NIR-Bestrahlung stark photooxidativ ist und Naphthalin dynamisch über den 2LMCT/2MC-Zustand und auch reaktionsträgere Substrate wie Benzol statisch über einen 4LMCT-Zustand (1,46 eV) oxidieren kann. Das bimolekulare Quenchen dieses photoreaktiven Komplexes gibt wertvolle Einblicke in die weitere photophysikalische Dynamik und das potenzielle Design zukünftiger photoaktiver MnIV-Übergangsmetallkomplexe, die nach einer niederenergetischen Anregung (850 nm) bimolekulare Reaktivität zeigen. Im dritten Teil wird untersucht, welche Voraussetzungen für eine metallzentrierte Emission aus oktaedrischen d8-Nickel(II)-Komplexen erforderlich sind. Bislang ist die metallzentrierte Emission von oktaedrisch koordiniertem Nickel(II) nicht dokumentiert. Ein starkes Ligandenfeld ist erforderlich, um die für Spin-Flip Emission notwendige energetische Abfolge der angeregten Zustände zu erhalten, d. h. energiearme intrakonfigurationale Singulett- III Zustände. Es gibt jedoch eine Grenze für die Stärke des Ligandenfeldes, die d8-Systemen auferlegt werden kann, um eine oktaedrische Koordination mit einer vorteilhaften Anordnung der angeregten Zustände aufrechtzuerhalten. Die Reihe der Komplexe [Ni(dgpy) )2+2 , [Ni(terpy) 2+2) (terpy = 2,2'; 6',2"-terpyridin), [Ni(phen) )2+3 (phen = 1,10-phenanthrolin), [Ni(ddpd)2)2+ und [Ni(tpe)2]2+ (tpe = 1,1,1-tris(pyrid-2-yl)ethan) sind Beispiele für NiII im oktaedrischen Ligandenfeld mittlerer Stärke und die daraus resultierenden Auswirkungen auf die inter- und intrakonfigurativen Zustände. Darüber hinaus wird auch der Einfluss von erhöhtem hydrostatischem Druck auf die Ligandenfeldzustände untersucht, um festzustellen, ob die inter- und intrakonfigurationalen Zustände weiter voneinander separiert werden können. Diese Untersuchung liefert wertvolle Informationen für die Suche nach metallzentrierten oktaedrischen NiII-Komplexen mit Spin-Flip-Emissionen. Teil vier baut auf Teil drei auf, indem die Auswirkungen einer Erhöhung der Ligandenfeldstärke auf die NiII-Ligandenfeldzustände untersucht werden. Dazu werden die Carbenliganden CNC = (1,1'-(pyridin-2,6-diyl)bis(3-methyl-1H-imidazol-3-ium) und NCN = (1,3-bis(2- pyridyl)imidazolium) mit ausgeprägterer σ-Donorfähigkeit an NiII koordiniert, um die eg*- Orbitale zu destabilisieren. Durch die stärkere Ligandenfeldaufspaltung ist es denkbar, dass der Komplex eine quadratisch-planare Geometrie annimmt. Somit stellt die strukturelle Bestätigung einer oktaedrischen Komplexgeometrie den ersten Schritt vor weiteren Charakterisierungsmethoden dar. Ein alternativer Ansatz zur Erhöhung der Ligandenfeldstärke ist die Verwendung von π-Akzeptorliganden, um die t2g-Orbitale zu stabilisieren. Diese Methode wird anhand der Synthese von [Ni(dcpp) ]2+2 (dcpp = (2,6-bis(2- carboxypyridyl)pyridin)) näher erläutert. Der dcpp-Ligand enthält zwei elektronenziehende Carbonylgruppen wordurch bei Koordination an das Metall ein stark π-akzeptierendes Ligandenfeld erzeugt wird. Eine Lewis-Säure (Sc[OTf]3]) wird dann an die Carbonylgruppen des Liganden koordiniert. Der Einfluss dieser zweiten Koordinationssphäre soll die π- Akzeptorfähigkeit des Liganden erhöhen. Dies soll tiefere Einblicke in die gezielte Beeinflussung der relativen energetischen Lagen der Ligandenfeldzustände des Komplexes geben. Diese Untersuchung zielt darauf ab, die Ligandenanforderungen für NiII bei der Verwirklichung der Spin-Flip-Emission besser zu verstehen. IV Table of Contents Abstract ........................................................................................................................................ I Abbreviations and Physical Quantities ........................................................................................ VII 1. Introduction ......................................................................................................................... 1 1.1 Excited State Properties of Transition Metal Complexes ........................................................ 4 1.1.1 Fluorescence and Phosphorescence ............................................................................... 5 1.1.2 Non-Radiative Decay Processes ...................................................................................... 6 1.1.3 Excited-State Bimolecular Processes ............................................................................... 9 1.2 Emission from 3d Transition Metal Complexes ..................................................................... 14 1.2.1 Luminescence Enhancement ......................................................................................... 19 1.3 Manganese ............................................................................................................................ 21 1.3.1 Manganese(II) ................................................................................................................ 21 1.3.2 Manganese(III) ............................................................................................................... 24 1.3.3 Manganese(IV) .............................................................................................................. 26 1.4 Nickel(II) ................................................................................................................................. 30 2. Aims of Work ...................................................................................................................... 32 3. Results and Discussion ........................................................................................................ 34 3.1 The Full d3–d5 Redox Series of Mononuclear Manganese Complexes: Geometries and Electronic Structures of [Mn(dgpy) ]n+2 .............................................................................................. 38 3.2 Oxidative Two-State Photoreactivity of a Manganese(IV) Complex using NIR Light ............ 49 3.3 Coupled Potential Energy Surfaces Strongly Impact the Lowest-Energy Spin-Flip Transition in Six-Coordinate Nickel(II) Complexes ............................................................................................. 66 3.4 Influencing Ligand Field States of Nickel(II) Complexes with Strongly σ-Donating and π- Accepting Ligands .............................................................................................................................. 77 3.4.1 Increasing σ-Donation ................................................................................................... 79 3.4.2 Increasing π-Accepting .................................................................................................. 81 3.4.3 Conclusion ..................................................................................................................... 86 4. Summary and Outlook ........................................................................................................ 87 5. References .......................................................................................................................... 89 6. Appendix ............................................................................................................................ 97 4 ........................................................................................................................................... 97 6.1 Supporting Information to Chapter 3.1. (“The Full d3–d5 Redox Series of Mononuclear Manganese Complexes: Geometries and Electronic Structures of [Mn(dgpy) ]n+2 )” ......................... 97 6.2 Supporting Information to Chapter 3.2. (“Oxidative Two-State Photoreactivity of a Manganese(IV) Complex using NIR Light”)...................................................................................... 123 V 6.3 Supporting Information to Chapter 3.3. (“Coupled Potential Energy Surfaces Strongly Impact the Lowest-Energy Spin-Flip Transition in Six-Coordinate Nickel(II) Complexes”) ............. 165 6.4 Supporting Information to Chapter 3.4. (“Influencing Ligand field states of Nickel(II) Complexes with Strongly σ-Donating and π-Accepting Ligands”) ................................................... 192 7. Acknowledgements ............................................................................................................ 197 8. Curriculum Vitae ................................................................................................................ 198 VI Abbreviations and Physical Quantities (Ph)OLED (Phosphorescent) organic light-emitting diode (phenN,N’^C)2 2-(3-(tert-butyl)phenyl)-1,10-phenanthroline 𝑯 Hamiltonian 𝑯𝟐𝒂𝒃 Electronic coupling matrix element µS Microsecond 4, 4`-bpy(NO2)2 4, 4`-dinitro-2,2‘-bipyridine Abs Absorption als 3-((2-hydroxybenzylidene)amino)propanoic acid B, C Racah parameters big Biguanide bISC Back-intersystem crossing Bn-TPEN N-benzyl-N,N’,N’-tris(2-pyridylmethyl)-1,2-diamino-ethane) bpmp 2,6-bis(2-pyridyl-methyl)pyridine bpy 2,2′-bipyridine btz 3,3′dimethyl-1,1′-bis(p-tolyl)-4,4′-bis(1,2,3-triazol-5-ylidene) Cbz Carbazole CNAr (th)5 NC 1,3-bis(N-formyl-4-methyl-6-phenylanilin-2-yl)benzene CNAr5NC 1,3-bis(N-formyl-4-methyl-6-phenylanilin-2-yl)thiophene CNC 1,1’-(pyridin-2,6-diyl)bis(3-methyl-1H-imidazol-3-ylidene) CO Carbon monoxide CT Charge transfer D Axial zero-field splitting parameter dcpp 2,6-bis(2-carboxypyridyl)pyridine ddpd N,N’-dimethyl-N,N’-dipyridin-2-ylpyridine-2,6-diamine dgpy 2,6-diguanidylpyridine dmp 2,9-dimethyl-1,10-phenanthroline dpb 1,3-bis(N-alkylbenzimidazol-2'-yl)benzene dpc 3,6-di-tert-butyl-1,8-di(pyridine-2-yl)-carbazolato DSSCs Dye-sensitized solar cell E00 One electron potential at zero vibrational transition E1/2 Half-wave potential EnT Energy transfer Eox Ground state oxidation potential eq Equivalents Ered Ground state reduction potential ES Excited state eV Electron volt VII FCWD Franck-Condon weighed density Fl Fluorescence fs Femtosecond GS Ground state HB(3, 5-Mepz) 3, 5-dimethyl pyrazolyl borate L-6H Hydrazine clathrochelate Hbig H-biguanide HOMO Highest occupied molecular orbital HS High-spin ILCT Intraligand charge transfer Imp 1,1′-(1,3-phenylene)bis(3-methyl-1-imidazol-2-ylidene) ISC Intersystem crossing kB Boltzman constant kEnT Rate constant for energy transfer kfl Fluorescence rate constant kISC Rate constant for intersystem crossing knr Non-radiative decay rate constant kph Phosphorescence rate constant kET Rate constant for electron transfer kr Radiative decay rate constant 𝝀𝑺 Solvent reorganization energy Lbi 2,5-bis(3,5-di-tert-butyl-2-isocyanophenyl)thiophene LC Ligand-centered LCNC Tert-butyl-carbazole dicyclohexylmesoionic carbene LF Ligand field LLCT Ligand-to-ligand transfer LMCT Ligand-to-metal charge transfer Lme 1,3-dimethyl-2-phenyl-1,3-diazaphospholidine-2-oxide Lpy bispidine pyridine LS Low-spin LUMO Lowest occupied molecular orbital MC Metal-centered M-L Metal-to-ligand MLCT Metal-to-ligand charge transfer ms Millisecond NCN 1,3-bis(2-pyridyl)imidazolylidene NIR Near infrared (760 -1000 nm) NIR-II Near infrared II (1000-1700 nm) nm Nanometers nr Non-radiative decay VIII ns Nanosecond OLED Organic light-emitting diode PSEN Photosensitizer PES Potential energy surface PET Photoelectron transfer Ph4P Tetraphenylphosphonium cation phen 1,10-phenanthroline Ph Phosphorescence Phtmeimb Phenyl[tris(3-methylimidazol-1-ylidene)] ppy Deprotonated 2-phenylpyridine ps Picosecond RT Room temperature S0 Singlet ground state S 1st1 excited singlet state S2 2nd excited singlet state sal Salicylic acid SCO Spin crossover SF Spin-flip SMM Single molecule magnet SOC Spin-orbit coupling SOMO Singly occupied molecular orbital Sub Substrate T1 Lowest energy triplet excited state tBu-terpy Tert-butyl 2,2';6',2"-terpyridine terpy 2,2';6',2"-terpyridine ThiaSO2 P-tert-butylsulphonylcalix[4]arene TM Transition metal TMCs Transition metal complexes tpe 1,1,1-tris(pyrid-2-yl)ethane tppb Hydro-tris(3-phenylpyrazol-1-yl)borate Tripp 2,4,6-triisopropylphenyl V Volt VR Vibrational relaxation Zeff Effective nuclear charge ZFS Zero field splitting ΔO Octahedral ligand field splitting λem Emission wavelength πL Ligand π-orbitals π *L Ligand π-antibonding orbitals σML Metal-ligand sigma bonding orbitals IX Ea Activation energy  Quantum Yield  Lifetime 𝜻 Spin-orbit coupling constant X XI 1. Introduction Photochemistry using transition metals (TMs) has become increasingly important in recent decades as a means to achieve more sustainable energy usage. Current world energy consumption is heavily reliant on fossil fuels, which contribute largely to greenhouse gas emissions and have finite availability.[1] A significant challenge is transitioning from these types of energy sources to more sustainable, less polluting alternatives. Photochemistry is the concept of chemically harnessing sunlight and converting into energy that can be exploited for use in various ways including  but not limited to  catalytic chemical transformations,[2] use as light harvesters in dye sensitized solar cells (DSSCs)[3,4] or for use in other energy efficient applications such as organic light emitting diodes (OLEDs) or phosphorescent organic light emitting diodes ((Ph)OLEDs).[5] TM photochemistry typically uses transition metal complexes (TMCs) to perform such light-harvesting, as they possess favorable excited state (ES) dynamics. TMs such as Ru, Ir or Pt are primarily used in such applications, as their ES dynamics are particularly advantageous. A typical example is [Ru(bpy) ]2+3 (bpy = 2,2′- bipyridine) (12+, scheme 1.1) as used in DSSCs[6] or in photoredox catalysis,[7] due to high stability and ES lifetime, excellent redox properties and broad absorption in the visible region of the spectrum. Although heavier 4d and 5d TMs are typically better suited for photochemistry, there are some inherent drawbacks associated with their use that are largely related to their abundance. For instance, first row 3d TMs are several orders of magnitude more abundant in the earth’s crust than their heavier analogues e.g. Ru is ≈ 106 % abundant, while Fe is ≈ 5 % abundant as mass percentage of the earth’s crust.[8,9] In addition extraction of these metals is challenging which inflates cost further.[10] A cheaper and more sustainable alternative is the use of 3d metals instead.[8,9,11] 1 Scheme 1.1: Molecular structures of RuII and FeIII,II complexes 12+, 22+, 33+ and 4. While more sustainable there are some inherent challenges associated with using 3d TMs over 4d and 5d. This is evident when comparing [Fe(bpy) 2+ 3] (lifetime ()= 150 fs) (22+, scheme 1.1) [12] and [Ru(bpy)3]2+ (λem = 620 nm, = 0.80 µs, quantum yield ()= 9.5 % in deoxygenated CH3CN at RT). [Fe(bpy)3]2+ is not phosphorescent and cannot be used in the same applications as its Ru analogue.[13] This is largely a result of the intrinsically strong ligand field (LF) splitting of 4d and 5d metals. 3d metals suffer from much weaker intrinsic LF splitting as a result of the primogenic effect,[14] and from much smaller spin-orbit coupling (SOC), which scales with Zeff (‘heavy-atom effect’), and might slow down intersystem crossing (ISC).[15] Thus, in recent decades there has been increased research into 3d luminescent metal complexes and there has been significant progress made.[16] Including such examples as the [Fe(btz) ]3+ 3 (btz = 3,3′dimethyl-1,1′-bis(p-tolyl)-4,4′-bis(1,2,3-triazol-5-ylidene)) (33+, scheme 1.1) complex,[17] which shows room temperature (RT) 2LMCT based fluorescence at 600 nm (= 0.1 ns,  = 0.03 %); and the more recent [Fe(phenN,N’^C)2] (phenN,N’^C = 2-(3-(tert- butyl)phenyl)-1,10-phenanthroline) (4, scheme 1.1) complex,[18] which shows MLCT phosphorescence at 1200 nm (= 14 ns at 77K in frozen toluene/THF solution). Such progress is a consequence of sophisticated ligand design and has helped lead to further efforts using other 3d TMs. 2 To date considerable research has focused on manipulation of 3d metal charge transfer (CT) ESs. However, metal-centered (MC) ESs can also be photoactive, and according to work by Y. Tanabe and S. Sugano achieving such depends on LF strength,[19,20] illustrating that certain electronic configurations can be emissive in strong octahedral LFs.[21] Their work has led to substantial developments in the area of MC based emission, particularly with chromium(III) with a d3 electronic configuration.[22–24] Such exemplary progress can be seen with the more recent example of [Cr(ddpd) 3+2] (ddpd = N,N’-dimethyl-N,N’-dipyridin-2-ylpyridine-2,6- diamine) (53+, scheme 1.2) showing MC spin-flip (SF) luminescence at 738/775 nm with exceptional quantum yield and lifetime in CH3CN under deoxygenated conditions (= 1122 µs,  = 13.7 %).[25] The photophysical characteristics of 53+ are largely attributed to ligand design, and this work has led to other outstanding contributions in this area including the NIR-II SF emissive mer-[V(ddpd)2]3+ complex (λem = 1109/1123 nm, = 1.35 µs,  = 1.8 x 104 % in deoxygenated CH CN at RT) (63+, scheme 1.2),[26]3 with these aforementioned developments in 3d luminophores it heralds work with other 3d metals. Scheme 1.2: [Cr(ddpd) ]3+2 (53+) and [V(ddpd) 3+ 3+2] (6 ) exhibiting NIR and NIR-II luminescence respectively. Manganese(IV) is also capable of SF emission as it posesses a d3 electronic configuration and so is isoelectronic with chromium(III). Moreover, in a strong field ligand environment it can also possess low energy MC states capable of phosphorescence. Molecular octahedral MnIV complexes are rare[27–34] particularly with only nitrogen donors ligands  and there is only one previously known emissive octahedral MnIV complex. Thus, their photophysical and chemical properties are poorly understood. Nickel(II) octahedral TMCs have other associated challenges. There are many more documented examples of molecular octahedral nickel(II) d8 compounds even in strong ligand fields (e.g. [Ni(ddpd) ]3+2 ). However, there is no report to date of SF emission from these complexes.[35] There is a need for further fundamental understanding to help overcome these challenges. Thus, work with these alternative 3d TMs is necessary on a fundamental level to increase understanding and open the way for eventual use in applications. 3 1.1 Excited State Properties of Transition Metal Complexes Photoexcitation of a TMC creates an ES species, which has many possible routes to return to the ground state (GS) (figure 1.1). With the correct ES ordering a TMC can be emissive, and it is also possible that such a low energy ES can be quenched in photochemical applications e.g. bimolecular redox (ET) or energy transfer (EnT) processes. It is important to understand the various routes of decay in order to design photoactive TMCs for use in application.[15] Figure 1.1: Simplified Jablonski energy diagram showing processes that follow absorption of a photon by a singlet GS (S0) molecule. Absorption (Abs), 1st singlet ES (S1), 2nd singlet ES (S2), lowest energy triplet ES (T1), vibrationally excited oscillators (XH) (X = C, N, O). Rate constants (k) for: vibrational relaxation (VR), internal conversion (IC), intersystem crossing (ISC), back-intersystem crossing (bISC), fluorescence (Fl), non-radiative decay (nr), phosphorescence (Ph). It is possible for an ES complex to relax to the GS radiatively (fluorescence/ phosphorescence), non-radiatively or via bimolecular quenching. Each of these processes will be described and discussed in subsequent sections. 4 1.1.1 Fluorescence and Phosphorescence Following photoexcitation of a TMC to the first ES (S1) (also be described as a Franck-Condon state), the ES can radiatively relax back to the GS. Decay can occur from S1 to S0 (i.e. fluorescence) or from a T1 to S0 (i.e. phosphorescence). Each process leads to emission of radiative energy (figure 1.1). Fluorescence is a spin-allowed process where the spin multiplicity is conserved. The resulting fluorescence is relatively fast, occurring on a timescale of approximately 109 to 107 s. Phosphorescence however is spin-forbidden involving a change in multiplicity, more specifically a ‘flip’ in the excited electrons spin. This spin forbidden process of triplet state relaxation to a singlet GS typically occurs on a much longer scale 106 to 103 s. The lowest energy T1 is lower in energy than the lowest energy S1, due to Hunds rule of maximum multiplicity and so decay from these states is energetically preferred, occurring at longer wavelengths. The process that governs the change of an electrons spin (transition from a singlet ES to a triplet ES) is the relative rate of intersystem crossing (kISC). Phosphorescence is preferential for many applications as the state is longer lived and can be further exploited.[15] Lifetimes and Quantum Yields To further describe an ES molecule, the parameters of lifetime () and quantum yield () are used. The lifetime of an ES is expressed as the reciprocal sum of radiative (kr) and non-radiative (knr) rate constants (figure 1.1, equation. 1.1).[36] 1 𝜏 = (equation. 1.1) ∑𝑘 + ∑𝑘 The term kr includes the terms kfl and kph, and knr consists of the various non-radiative rate terms. The quantum yield () of a process describes the efficiency of a radiative process in terms of the ratio of photons absorbed (kr + knr) to photons emitted (kr)(equation. 1.2).[36–38] 𝑘 𝛷 = (equation. 1.2) ∑𝑘 + ∑𝑘 The smaller the non-radiative decay term the closer to 1 the quantum yield will be. It is evident from these two equations that the lifetime and quantum yield of an emissive process is heavily dependent on competitive non-radiative processes. 5 1.1.2 Non-Radiative Decay Processes An ES can also depopulate without emission, where the energy of the absorbed photon is dissipated via another process. Deactivation of an ES via vibrational relaxation (kVR) and internal conversion (kIC) are dominant non-radiative mechanisms that are ultrafast (1014tosand so very likely to occur immediately following excitation. The non- radiative mechanistic description is further enhanced with two general mechanisms, known as strong and weak coupling limits. Strong coupling describes non-radiative decay that typically accompanies distorted ESs; weak coupling describes non-radiative decay processes associated with a nested ES. Strong Coupling Limit Strong coupling limit (figure 1.2) describes non-radiative decay (knr) that occurs due to strong ES distortion (equation. 1.3).[39–42] 𝑘 𝑇 2𝜋 Δ𝐸 𝑘 = 𝐻 × exp − (equation. 1.3) ℏ 𝐸 (𝑘 𝑇) 𝑘 𝑇 The electronic coupling matrix element (𝐻 ) describes the transition from GS to ES. 𝐸 describes the half stokes shift of emission from excitation, and Δ𝐸 represents the energy gap between the lowest excited vibronic state and the crossing point between the ES and GS PESs. In this case, the knr displays temperature dependent Arrhenius behavior; Δ𝐸 is the energy barrier governing non-radiative decay. In situations where ES distortion is large (figure 1.2), the activation energy required for non-radiative decay approaches zero; the resultant non- radiative decay in this case is almost barrier-free at RT. Such strong coupling effects can be reduced by carrying out emission measurements at cryogenic temperatures or with ligand rigidification, both reducing ES distortion.[43,44] 6 Figure 1.2: Ground (GS) and excited state (ES) PESs at the strong coupling limit, the red arrow indicates the non-radiative decay route. Half Stokes shift (Em), GS vibrational energy level (nogap (E) between GS (noandES (no and activation energy (Ea). Weak Coupling Limit Weak coupling limit (figure 1.3) can occur when the ES PES is nested with the GS PES. First tunneling from an ES into a vibrationally excited GS occurs, followed by vibrational relaxation (equation. 1.4 and figure 1.3).[40,41,45] 2𝜋 𝐻 S (Δ𝐸 − n ℏ𝜔 − 𝜆 ) 𝑘 = ⋅ ⋅ ⋅ exp(−𝑆 ) ⋅ exp − (equation. 1.4) ℏ 4𝜋𝜆 𝑘 𝑇 𝑛 ! 4𝜆 𝑘 𝑇 Here knr dependence includes the electronic coupling matrix (𝐻 ); solvent reorganizational energy (𝜆 ); the Huang-Rhys factor (𝑆 ) which describes geometric distortion between GS and ES; the energy difference between the GS and ES vibrational levels (Δ𝐸) and the quantum number for intraligand vibrational modes of the GS (𝑛 ). This relationship is also known as the ‘energy gap law’ i.e. the exponential dependence of knr on Δ𝐸, knr is inversely proportional to Δ𝐸 therefore the larger the energy gap between ES and GS the smaller k .[38,41]nr Figure 1.3: Ground (GS) and excited state (ES) PESs at the weak coupling limit, the red arrow horizontal arrow indicates the energy transfer via tunneling (EnT) and the vertical red arrow indicates vibrational relaxation (VR). Vibrational energy level (n and the gap (E) between GS (noandES (no Deactivation occurs via a horizontal EnT transition followed by vibrational relaxation to the GS. Equation 1.4 shows that knr increases with geometric ES displacement Sm, and with vibrational energy levelswith the lowest quantum numbers nm. Vibrational modes (with low quantum number overtones) that contribute most include high energy CH, OH and NH 7 (𝜔 (XH) = 30003600 cm) stretching vibrations, and will affect ES energies of 800014500 cmTransfer of energy can occur to high energy oscillators present on the ligands of a TMC, or alternatively to surrounding solvent molecules. Intersystem Crossing The rate at which intersystem crossing (kISC) occurs relative to other competing processes will primarily determine whether phosphorescence will occur. This process has many competing processes which prevent population of a triplet state (mainly kVR and kIC) however, kf will also compete with kISC. For phosphorescence to occur a change of spin multiplicity happens during a transition. This process is spin selection rule forbidden (S = 0), and so has a low probability of occurring. Facilitation of this change in multiplicity can occur because of SOC which is a relativistic effect where the spin and orbital angular momentum magnetically mix and interact. The total angular momentum of a system (J) should remain unchanged and can remain unchanged with change in spin angular momentum (S), but only if a simultaneous change of orbital angular momentum (L) occurs (equation. 1.5, El Sayed’s rules). This mixing of orbital and spin angular momentum allows electronic states of different multiplicity to couple and mix, assisting with overcoming the selection rule restriction and thus increasing the probability of an electron ‘flip’.[15,37,39] J=L+S (equation. 1.5) The quantified SOC constant (𝜁), is shown by the simplified Hamiltonian (𝐻 ) (equation. 1.6), and it includes several variables. The operator is the sum of the angular momentum operator (𝑙 ), spin angular momentum operator (?̂? ) and SOC constant (𝜁). The SOC constant importantly depends on the specific electronic configuration of an atom, and approximately scales with Z 4eff . ‘The heavy atom effect’ describes this increase in SOC in 4d and 5d TMs (e.g. 𝜁 Ru2+) 1exp = 1159 cm ) compared with 3d TMs (e.g. 𝜁Fe2+)exp = 436 cm1) which can significantly increase kISC and the probability of phosphorescence.[24] 𝐻 = 𝑙 ?̂? 𝜁 (equation. 1.6) The effect of SOC on kISC is apparent when applying Fermi’s golden rule approximation (equation. 1.7);[47] this approximation also includes the relation of the T1 Franck-Condon weighted density of accepting states (FCWD) available for coupling with the S1 state. 2𝜋 𝑘 = S 𝐻 T ⋅ FCWD (equation. 1.7) ℏ 8 Spin-vibronic coupling is another mechanism that can facilitate high ISC rates. This type of coupling assumes a breakdown of the Born-Oppenheimer approximation instead assuming nuclear and electron motion are on a similar timescale. Coupling or mixing of electronic states with vibronic progressions, resulting from molecular geometric distortions can then occur.[39] Smaller SOC can present an intrinsic problem when attempting to achieve phosphorescence in 3d TMCs. ISC for TMCs typically occurs on a short timescales (fs) (e.g. ruthenium(II) (12+) and chromium(III) (53+)),[48,49] though slower (ns-ps) timescales have been observed in some other complexes.[26,39,50–52] Such examples include [FeO ]2-4 (72) with slower ISC being attributed to nested high energetic barriers for surface crossing,[53] and [Cu(dmp) +2] (dmp = 2,9-dimethyl- 1,10-phenanthroline) (8+) with slower ISC described as a result of weakened SOC due to geometric distortion and ‘flattening’ of the complex.[54,55] 1.1.3 Excited-State Bimolecular Processes Bimolecular quenching is an alternative non-radiative process that can follow excitation of a TMC, and can occur as EnT (Förster and Dexter) and ET (oxidative and reductive). It is essential that the energy of an ES photosensitized donor (D*) can excite an acceptor molecule (A) for EnT to occur.[56] Whereas for ET to occur the excited state redox potential of a photosensitizer (PSEN) and GS redox potential of a substrate (sub) must be compatible.[2,57] Bimolecular quenching can be further classified by timescale. Dynamic (collisional) bimolecular quenching requires a sufficiently long ES lifetime to be quenched on a diffusion controlled timescale, as described by the Stern-Volmer relation i.e. > 1 ns (equation. 1.8).[38] 𝛷 𝜏 = = 1 + K 𝜏 [𝑄] = 1 + K [𝑄] (equation. 1.8) 𝛷 𝜏 Here,𝛷 and 𝛷 represent luminescence intensities without and with quencher; 𝜏 and 𝜏 the lifetime of emitter without and with quencher present; [𝑄] the concentration of quenching species and K the dynamic quenching constant. The Stern-Volmer constant K for a specific system can be extracted from concertation dependent experiments i.e. plotting 𝛷 ⁄𝛷 or 𝜏 ⁄𝜏 vs [𝑄], where the slope is equal to K . Additional factors such as medium viscosity, temperature and sterics (sensitizer and quencher) can affect diffusion and thus collision kinetics.[38] Conversely, static bimolecular quenching is not diffusion controlled and has a pseudo Stern-Volmer relationship. It can therefore occur on shorter timescales (< 1 ns), as there is already close pre-association between ES sensitizer and quencher (equation. 1.9).[58– 60] 𝛷 = 1 + K [𝑄] (equation. 1.9) 𝛷 9 Where, K represents the association constant of the complex formation, and 𝛷 ⁄𝛷 vs [𝑄], will yield a straight line with slope equal to the static quenching constant K . Transient absorption spectroscopy (TA) can also help to assign static quenching, as this type of quenching typically does not have an effect on the lifetime of a sensitizer. Such time resolved spectroscopy can also be used in the case of non-emissive bimolecular processes to extract information bimolecular quenching kinetics.[2,61,62] Although Stern-Volmer analysis and ultrafast spectroscopy can distinguish between static or dynamic quenching, they cannot reveal whether ET and EnT has occurred. Further analysis of product formation will confirm if ET or EnT has occurred (i.e. looking for presence of oxidized or reduced photosensitizer/quencher confirms that ET has occurred).[24,60] Energy transfer Photophysical EnT occurs from a D* molecule to an A molecule in a lower energy state. This EnT leads to dissipation of D* and excitation of A. The thermodynamic potential is connected to the E *00 of both D and A i.e. Δ𝐺 = 𝐸 − 𝐸 (with entopic and 𝜆 effects assumed negligible). This enables the possibility of a Marcus-type treatment (equation. 1.10) for such processes.[24,36] 4𝜋 𝑘 = (𝐻 ) 𝐹𝐶𝑊𝐷 (equation. 1.10) ℎ Here kEnT depends on the electronic coupling between D* and A (𝐻 ) and is composed of electronic Coulombic and exchange elements that regulate EnT. The Franck-Condon weighted density of states (𝐹𝐶𝑊𝐷) corresponds with the spectral overlap integral between the absorption and emission spectra. The two mechanisms for EnT i.e. electronic Coulombic (Förster) and exchange (Dexter), are described in the following section.[38] Förster Resonance Mechanism The mechanism of Förster EnT is dipole-dipole induced, distance dependent, and can follow when sufficient spectral overlap between the emission of D* and the absorption of A occurs. (equation. 1.11, figure 1.4). 𝐾 𝛷 𝑘 = 8.8 x 10 𝐽 (equation. 1.11) 𝑛 𝑟 𝜏 Here 𝑛 is the solvent refractive index and 𝑟 is distance between D* and A; 𝜏 and 𝛷 are the lifetime and quantum yield of D* respectively; 𝐽 signifies the normalized spectral overlap between the emission of D* and absorption of A, while 𝐾 describes directional dipole-dipole 10 interactions. Importantly this mechanism can occur over a relatively long range (𝑟 ). Förster EnT does not involve a change in multiplicity i.e. singlet to singlet EnT.[38] Figure 1.4: Schematic representation of Förster (Coloumbic) EnT from excited singlet donor molecule (D*) to a singlet acceptor molecule (A). Dexter Mechanism Dexter, unlike Förster EnT, is a double electron transfer and can occur between D* and A states of different multiplicities. The multiplicity of D* and A can change during the exchange and the overall multiplicity of the process can remain conserved. This mechanism describes how an electron transfers from D* to the LUMO of A, while simultaneous movement of a second electron from the HOMO of A to the SOMO of D* (figure 1.5).[24,36] Figure 1.5: Schematic representation of Dexter (exchange) EnT from an excited triplet donor molecule (D*) to the HOMO of an acceptor molecule (A) and concurrent transfer of a second electron from HOMO on A to SOMO on D*. Dexter EnT is strongly distance dependent as it requires orbital/wavefunction overlap of D* and A (r < 10 Å); its rate decreases exponentially with increased distance (equation. 1.12). 4𝜋 𝛽 𝑘 = 𝐻 (0) ⋅ exp − (𝑟 − 𝑟 ) 𝐽 (equation. 1.12) ℎ 2 Here, the electronic coupling factor 𝐻 includes the interaction contact distance 𝐻 (0), along with attenuation factor for the exchange energy 𝛽 . 𝐽 describes normalized overlap of the emission of D* and absorption of A. One common example of Dexter EnT is quenching of a triplet emission by triplet oxygen, producing singlet oxygen. Although detrimental for 11 emission, this can be harnessed for use in optical oxygen sensing or production of 1O2 for use in further reactions.[25,63,64] Electron Transfer Photoexcited species can act as stronger oxidants or reductants compared to their GS counterparts. When in contact with molecules that have suitable potentials, photo-induced electron transfer (PET) can happen to quench the ES. This can occur when an electron is transferred from substrate to the now strongly oxidizing ES photosensitizer (figure 1.6a) or where ET occurs from the strongly reducing ES photosensitizer to a substrate molecule that is lower in energy (figure 1.6b). Figure 1.6: Schematic representation of reductive (a) and oxidative (b) photo-induced electron transfer (PET). The ES photoredox potentials of a species can be estimated using equations 1.13 and 1.14.[2,62] 𝐸∗ = 𝐸 − 𝐸 (equation. 1.13) 𝐸∗ = 𝐸 + 𝐸 (equation. 1.14) 𝐸 represents the energy gap between the ES and GS at zero vibrational levels (00 transition); 𝐸 the one electron GS oxidation potential and 𝐸 the one electron GS reduction potential. Kinetically ET is described by expanded Marcus theory (equation. 1.15).[65–67] ∆ k = H (0) ⋅ exp − (r − r ) ⋅ exp − (equation. 1.15) 12 The rate constant 𝑘 has a PSEN/Sub distance-dependent electronic coupling factor 𝐻 , which includes the interaction contact distance 𝐻 (0), along with an attenuation factor 𝛽 . The Franck-Condon factor contains the reorganizational energy 𝜆, which reflects the energetics associated with structural or solvent redistribution tied to ET, the Boltzmann constant k and the thermodynamic driving force ∆𝐺 . Scheme 1.3: Schematic representation of the stages of ET (oxidative quenching) between photosensitizer (PSEN) and substrate (sub). For an outer sphere ET (i.e. ET between two moieties that do not undergo ligand substitution) (scheme 1.3), PSEN and Sub diffuse together to form an encounter complex. This encounter complex is surrounded by a cage of solvent molecules wherein ET can occur if there is sufficient driving force and electronic coupling. Following ET, products can then escape the solvent cage and diffuse apart. The cage escape yield of an electron transfer process, i.e. the yield with which products charge separate and escape a solvent cage, and the rate of back ET will have an effect on the overall process rate and outcome.[62] Examples of factors that can affect such cage escape and back electron transfer include solvent polarity/viscosity, temperature, spin-parity and SOC.[62,68–71] If a photosensitizer can be reversibly recovered, it is possible to use them in a photocatalytic cycle.[57] Regeneration of the photoactive species can be carried out using a sacrificial oxidant/reductant or electrolysis[72–74] to give a closed catalytic cycle, and subsequent turnovers following absorption of light.[7,71] An example of photocatalysis can be seen with the trisaminocyclopropenium radical dication.[74] This photocatalyst was found to be strongly photooxidative (E *ox = 2.95 vs ferrocene/ferrocenium) and capable of catalyzing Nicewicz-type coupling of benzene to a deactivated pyrazole (1H- pyrazole-4-carboxylic acid ethyl ester) with 65% yield following irradiation (compact fluorescent light (CFL), 23 W) for 60 hrs. Re-oxidation of the catalyst was carried out using electrolysis,[72–74] due to its high GS potential (E1/2 = +0.88 V ferrocene/ferrocenium) and significant absorption in the visible region (up to 600 nm).[74] 13 1.2 Emission from 3d Transition Metal Complexes Phosphorescence from 3d TMCs can have different character depending on the electronic nature and structures of the TMCS. The main types are charge-transfer (CT), metal-centered (MC/SF) and ligand-based (Intraligand-charge transfer (ILCT), ligand-to-ligand charge transfer (LLCT)).[16] Examples of some CT and MC emissive TMCs include [Fe(phtmeimb) + 2] (phtmeimb = phenyl[tris(3-methylimidazol-1-ylidene]) (10+, scheme 1.4) which shows 2LMCT emission (em = 655 nm, = 2 ns,  = 2.1 % at RT in CH3CN),[75] [Mn(Lbi) + bi3] (L = 2,5-bis(3,5-di-tert-butyl- 2-isocyanophenyl)thiophene) (11+, scheme 1.4) which shows 3MLCT emission (em = 485 nm, = 0.74 ns,  = 0.05 % at RT in deaerated CH [76]3CN) and [Cr(tpe) 3+2] (tpe = 1,1,1-tris(pyrid-2- yl)ethane) (123+, scheme 1.4) which shows 2MC emission (em = 748 nm, = 4500 µs,  = 8.2 % at RT in deaerated D O/DClO ).[77]2 4 Scheme 1.4: Example TMC complexes showing emission originating from various types of phosphorescence. Shown here a LMCT transition (left) which involves transfer of charge from occupied πL orbitals or σML bonding orbitals to the metal; an MLCT transition (middle) which involves transfer of charge from the π-metal t2g orbitals to unoccupied π *L orbitals; and a MC/SF transition (right) involving only metal d orbitals. 14 Spin-Flip Spin-flip emission is described as phosphorescence from a TMC intraconfigurational MC state, and to understand the requirements for MC/SF emission (scheme 1.4, right), a further the electronic description using potential energy surfaces (PESs) and Tanabe-Sugano diagrams can be used.[19,20] Both can be used to understand which d electron configurations are suitable for SF emission. Configurations d2, d3, d4 and d8 are candidates for such emission as they contain nested intraconfigurational states i.e. SF states share the same overall electronic configuration as the GS. This means that the ES is not geometrically distorted (e.g. M-L bond elongation) due to no bonding orbitals being depopulated or anti-bonding orbitals being populated following excitation and SF transition (figure 1.7, bottom). This is not the case for other configurations such d6 low-spin (LS) that contains distorted interconfigurational states (figure 1.7, top).[21] This nesting of SF states gives a sharp emission band profile that is typically low in energy (NIR). The benefit of having nested states means that knr is less competitive with kph, leading to phosphorescence in spite of the transition being Laporte and spin selection rule forbidden.[78,79] Figure 1.7: Example GS (blue) and lowest SF/MC (red) potential energy curves for d2 electronic configuration (bottom) and d6 LS (top), illustrating a geometrically distorted (5T 62 d LS) compared with a nested ES (1T2 d2). Tanabe-Sugano diagrams are important descriptors of intra- and interconfigurational states of TMCs. In an octahedral ligand environment LF terms are plotted as a function of LF strength, and both terms are a scaled by the Racah parameters. The Racah parameters (B and C) describe the d-d interelectronic repulsion in octahedral TMCs where the relative state energies are proportional to the C/B ratio and so will vary with varied TMCs.[21] The Tanabe- Sugano diagram for the d2 electronic configuration shows this relation (figure 1.8); As LF 15 strength increases, so does the energy of the distorted 1st excited state (3T2) beyond the point where it crosses (black circle) with the LF independent SF states (1E/1T2). This leaves SF states as the lowest energy states and allows for potential SF emission. Larger LF strength is advantageous because a larger energy gap between intra- and interconfigurational states (O/B > 40) helps suppresses thermally activated bISC.[78] Figure 1.8: Simplified Tanabe-Sugano diagram for octahedral d2 electronic configuration in octahedral symmetry (C/B = 4, scaled by Racah B parameter) with microstates of the triplet, spin allowed states (blue) and singlet MC states (red). Configurations d3, d8 and d4 will be further discussed in sections 1.3 and 1.4.[80] Tanabe-Sugano diagrams apply for strict octahedrons; deviations from this symmetry in the GS or ES mean deviations from these descriptions. These diagrams do not factor in SOC, which also can lead to state mixing and deviations (not as significant for 3d metals compared with 4d and 5d). They also neglect CT states which can have an effect on SF emission.[21] When it comes to adjusting emission energies for SF emitters, it is challenging; the energies of SF states are difficult to predict because of d-d interelectronic repulsion. Furthermore, a clear relationship between various ligands, metal identity and charge does not exist beyond the nephelauxetic relationship[81,82] i.e. the relation of d-d interelectronic repulsion as described by the Racah parameters B and C and the effect on M-L bond covalency. An increase in M-L bond strength and larger degree of covalency correlates with a decrease in Racah parameter B, and a lowering in energy of SF states, and a red shift in emission energies. This effect is seen when comparing complex 53+ to other octahedral chromium(III) complexes such as fac- [Cr(ppy)3] (ppy = deprotonated 2-phenylpyridine) (13, scheme 1.5) (λem = 910 nm at 77K in frozen 2-MeTHF)[83] and [Cr(dpc)2]+ (dpc = 3,6-di-tert-butyl-1,8-di(pyridine-2-yl)-carbazolato) (14+, scheme 1.5) (λ [84]em = 1067 nm at 77K in frozen CH3CN). Increasing the Racah B parameter has the opposite effect, with the increase of interelectronic repulsion and blue shift of emission compared to 53+, demonstrated in [Cr(bpmp) 3+ 2] (bpmp = 2,6-bis(2-pyridyl- methyl)pyridine) (153+, scheme 1.5) (λem = 709 nm at RT in H [63]2O). This effect can also be linked to the charge of the metal center where the Racah B and C parameters increase with charge.[85] 16 Scheme 1.5: Molecular structures of emissive chromium(III) complexes 13153+. SF emission (energy, lifetime and quantum yield) are affected by various external factors, such as temperature and pressure. While increased temperature can have the effect of leading to thermal deactivation processes, it can also affect the Boltzmann population of various energetically close SF states. In turn, this can affect the emission band shape and number of bands, which is the case of complex 53+ (scheme 1.2). 53+ has dual band luminescence at RT, however the relative intensity of these band changes with temperature allowing for further use as an optical ratiometric thermometer.[86] External pressure has been seen to also have an effect on SF emission with complex 53+ which shows a shift in emission of 14.8 cm1 kbar1 in H2O. Increase in hydrostatic pressure leads to small structural changes in the complex and consequently the M-L bonding changes, changing the nephelauxetic effect and leading to a red shift in emission. This effect can be harnessed for use as an optical pressure sensor.[87] SF emission can also be switched off by altering the pH as seen with 153+ (scheme 1.5). Increasing the pH leads to deprotonation of the ligand methylene bridge and the complex becomes non- luminescent, as this leads to de-aromatization of one of the coordinating pyridine moieties and thus alteration of ligand donor properties. Such reversible deprotonation allows for ratiometric optical pH sensing.[63] 17 Charge Transfer Charge transfer emission involves movement of charge within a molecule. LMCT requires an electron deficient metal centre (e.g. Iron(III), scheme 1.4, left); MLCT requires an electron rich metal centre (manganese(I), scheme 1.4, middle). This electronic transition involves bonding orbital depopulation or anti-bonding orbital population, leading to ES distortion (similar to d6 LS PESs (figure 1.7, top)). CT transitions are not Laporte-forbidden, unlike SF transitions and so generally give intense, broad emission profiles.[44] The emission energy for CT bands is conceptually more straightforward to adjust than with SF emitters, by adjusting the energy gap between the metal d-orbitals and the π *L orbitals (MLCT) or πL/σML orbitals (LMCT). The narrowing of this energy gap would cause a red shift emission; an MLCT example would be to use more σ-donating ligands to de-stabilise the t2g orbitals relative to the πL*, or to introduce electron withdrawing substituents onto a ligand lowering the π * orbital energy relative to the t orbitals.[88]L 2g This can be seen with the example of 12+ (λem = 620 nm) vs. [Ru(4,4’-bpy(NO 2+2)2)3] (4,4’-bpy(NO2)2 = 4, 4’-dinitro-2,2’-bipyridine) (162+)(λem = 700 nm).[2,36,89–94] The MLCT emission energy can also be lowered by changing the metal centre e.g. Ru to Os (4d vs. 5d orbitals) shown by complex 12+ (λem = 620 nm) vs. [Os(bpy) ]2+ (172+3 )(λem = 735 nm).[44,95,96] CT emitters are also subject to local environment changes; they have a larger transition dipole moment than SF emitters and so are more subject to changes in solvent polarity (solvatochromism) or changes in counter cation/anion.[97] Cryogenic temperatures and high pressure have the effect of suppressing deactivation via solvent vibrational modes of ligand vibrational modes.[98] Temperature changes can also lead to state occupation differences like with MC emitters, affecting emission profiles, with each effect previously reported to occur with complex 12+ (scheme 1.1).[44,99] Both SF and CT emissive states are sometimes not composed of ‘pure’ MC or CT character. Typically such states can be close in energy and so can mix, having properties of both types including LC admixtures.[100] Emissive state mixing can additional offer additional benefits. Increased MC character could offer an increase in lifetime (transition becomes more parity forbidden) and additional CT character can offer an increase in absorption of a complex. Alternatively, mixed character of states can offer challenges such as ES distortion and vibrational deactivation as seen with 14+, noted as an admixture of 2LMCT and 2MC states.[84] 18 1.2.1 Luminescence Enhancement There are a number of ways to enhance emission for 3d TMCs, beginning with ligand design. The initial challenge is a result of the primogenic effect which describes how 3d TMs have the inherent problem of smaller LF splitting due to the absence of radial nodes in the 1st d shell electronic functions. This results in radial contraction of the 3d orbitals in contrast to the 4d and 5d orbitals. This contraction leads to poor overlap of 3d orbitals with ligand orbitals leading to smaller LF splitting (figure 1.9).[101,102] Figure 1.9: Schematic representation of the primogenic effect, illustrating the contraction of 3d orbitals, relative to the 4d and 5d homologues. There are a few noteworthy ways to overcome this challenge, including the use of strong field ligands and/or maximizing metal-to-ligand (M-L) bond overlap. The use of strong field ligands is a clear way to overcome such issues i.e. increasing the σ-donating character of the ligand will destabilize the σ-bonding metal e *g orbitals (e.g. use of carbenes);[103] increasing the π- accepting ability of a ligands will stabilize the π-metal non-bonding t2g orbitals (e.g. use of CO).[104] Another way is by maximising M-L bite angle (L-M-L) and bond overlap; this is the idea of increasing the bonding angle of the ligand to overlap more efficiently with the orthogonal 3d bonding d 2 2z and dx  2y orbitals. This can be clearly seen when comparing the 3d 2 2x y orbital overlap with the ddpd ligand and the terpy ligand (figure 1.10).[79,105] 19 Figure 1.10: Comparison of 3d orbital overlap with a terpy (left) and ddpd (right) ligand; N-M-N bond angles for the 5-membered terpy ligand are ca. 78°, whereas angles with the 6-membered ddpd ligand are close to 90° enabling much more efficient M-L orbital overlap and leading to a stronger LF. These strategies of course can be  and usually are  combined, as many ligand types are both accepting and donating. Effective ligand design has also been shown to help supress quenching with confined SF states (Dexter EnT) via triplet oxygen. For example making a ligand more sterically bulky, as seen with substituted 53+ (scheme 1.2). Bulky Tripp groups (Tripp = 2, 4, 6-triisopropylphenyl) were added at the 5-position of the terminal pyridines of ddpd ligand. This enhanced the lifetime and quantum yield in oxygenated CH3CN from 52 µs and 0.8 % to 518 µs and 5.1 %.[25] In addition to effective ligand design other methods can be employed to increase the lifetime and quantum yield of 3d TMCs. For instance, as previously mentioned cryogenic temperatures along with the rigidification of the complex can reduce geometric distortion and intramolecular vibrations.[106] Increasing the energy separation between the emissive states and deactivating states is another method to reduce deactivation.[79] Another example for SF emitters is the introduction of an inversion center to a TMC, increasing the Laporte-forbidden nature of a MC/SF transition and prolonging the lifetime as seen with 123+ (scheme 1.4) showing a lifetime of 4500 µs[77] compared with 1122 µs for 53+ (scheme 1.2).[78] A particularly successful approach to suppress the multiphonon relaxation is the deuteration of ligand and/or solvent. This approach aims to significantly reduce the energy of the XH oscillators, increasing their quantum number overtone and thus decreasing non-radiative decay via multiphonon relaxation (figure 1.1). This method is especially useful for NIR emitters whose energy is most at risk from deactivation via weak coupling. This was seen with 53+, where the lifetime and quantum yields were extended by changing H2O to D2O, which increases the quantum yield and lifetime from 11 % and 898 µs to 14.2 % and 1164 µs (in deoxygenated conditions). Further enhancement was achieved by the statistical deuteration of the ddpd ligand; the combined effect of D2O and d9-ddpd enhances the quantum yield and lifetime to 30 % and 2300 µs (deoxygenated conditions).[78,79,107,108] This can also be seen for complex 153+ (scheme 1.5), with quantum yield and lifetime increasing from 15.8 % and 1550 µs to 24.6 % and 2500 µs (in acidic deoxygenated media) upon introduction of deuterated solvent and 20 ligand.[63] A similar yet smaller effect is seen in the CT emitter 12+ (scheme 1.1), where the lifetime is prolonged by 75 ns upon deuteration.[44,109] 1.3 Manganese Manganese is one of the most abundant TMs on earth at ≈ 0.1 % of the earth’s crust.[9,110] It is commonly found in the Earth’s crust as the minerals pyrolusite (MnO2) and rhodochrosite (MnCO3), and when isolated it is used in various industrial processes, particularly in making alloys such as steel to improve strength and hardness.[111] It is used in catalysis,[112] found in biological systems[113] and is vital in photosynthesis, as part of the oxygen evolving cluster (Mn4O5Ca) that splits water as part of the photosystem II process.[114,115] Manganese possesses rich redox chemistry and is relatively stable in eight oxidation states, with 0, I, II, IV and VII compounds being most prevalent. Common examples include MnCl2, MnBr(CO)5, MnO2 and the well-known aqueous oxidant KMnO .[116]4 The oxidation states III, V, VI are less stable and compounds containing manganese in these oxidation states more rare; some examples include such as Mn2O3 or K2MnO [110]4. Notably, manganese has rich oxide chemistry, as the hard dianionic oxygen ligand can better stabilize the higher oxidation states of Mn.[117] With greater general emphasis on sustainability, there has been increased interest in the photophysical properties of Mn based sensitizers. There are various reports of emissive doped MnII,IV,V solids and their potential in applications such as in OLEDs.[118–120] Most examples of emissive molecular manganese TMCs are with the MnII cation, which is largely related to the ease of synthesis from readily available manganese(II) starting materials.[121] Manganese(IV) is an attractive prospect as an analogue to chromium(III), with the rich well documented CrIII SF photochemistry.[16] However, TMCs containing MnIV are comparatively few and synthesis begins with MnII salts, requiring challenging oxidations to achieve MnIV;[122] thus, the photophysical properties are not well understood.[8] The unique properties and (photo)physical characteristics of molecular MnII compounds advancing through MnIII to MnIV will be discussed, with focus on MnIV, to aid in the understanding, design and synthesis of photoactive MnIV containing TMCs. 1.3.1 Manganese(II) MnII is the most common oxidation state of Mn and has a d5 electronic configuration. When in an octahedral LF it can have a high-spin (HS) (S = 5/2) or LS (S = 1/2) configuration. Typically, MnII exists in a highly paramagnetic HS configuration, which is a consequence of the inherent stability of a half-filled d electron shell and Hund’s rule of maximum multiplicity.[81] Octahedral HS MnII is also characteristically substitutionally labile, resulting from the larger ionic radii and 21 lack of LF stabilization energy. This lability also means the barrier for isomerization is low, making isolation of single isomers a challenge.[123] Suppressing such ligand substitution and isomerization proves challenging. It is possible with careful ligand design, for example using rigid chelate ligands or steric bulk e.g. [Mn(tppb) ]2+2 (tppb = hydro-tris(3-phenylpyrazol-1- yl)borate) (182+, scheme 1.6)[124] or [Mn(Lpy) ]+ (19+2 , scheme 1.6) (Lpy = bispidine pyridine).[125,126] The Tanabe-Sugano diagram for octahedral MnII TMCs (figure 1.11) shows the photophysical MC landscape of these complexes. Figure 1.11: Simplified Tanabe-Sugano diagram for d5 electronic configuration in octahedral symmetry (C/B = 4, scaled by Racah B parameter) with microstates of the doublet (blue), sextet (green) and quartet, MC states (red).[80] In HS configuration, LF transitions are Laporte-forbidden and spin-forbidden leading to weakly colored complexes.[127,128] However, molecular (and solid state) luminescent MnII complexes are most common amongst Mn TMCs.[121,129,130] An important difference between the d5 Tanabe-Sugano diagram with those of d2-d4 and d8, is that the quartet states (4T 42 and T1), are not nested (antibonding orbitals are populated) and structurally distorted. As the LF strength increases, a spin-crossover point is reached and HS converts to LS; such conversion with MnII requires strong field ligands, e.g. CN-.[131] LS MnII compounds are rare with only a few reported.[132–136] 22 Scheme 1.6: Molecular structures of HS d5- MnII complexes 182+–22; complexes 20–22 show MC luminescence. Due to the forbidden nature of the MC transitions in HS MnII complexes, strategies for luminescence aim at reducing the overall symmetry of the system. This is done by making transitions ‘more’ allowed, for example with [MnX me2(L )2] (X = Cl, Br, I, Lme = 1,3-dimethyl-2- phenyl-1,3-diazaphospholidine-2-oxide) (20, scheme 1.6) (emBr = 509 nm, Br683 sand  Br= 23 % in solid state at RT), utilizing a tetrahedral environment that does not possess a center of inversion,[137–139] or with the structurally restrained [Mn4(ThiaSO2)2F]+ (ThiaSO2 = p-tert-butylsulphonylcalix[4]arene) (21, scheme 1.6) (em = 666 nm, 1.08 msand  = 15 % in deoxygenated DMF at RT under inert conditions).[140] Another method involves introducing heavy halides to increase ISC rates via the heavy atom effect as with [Ph4P]2[MnBr4] (Ph4P = tetraphenylphosphonium cation) (22, scheme 1.6) (em = 516 nm, 355 sand  = 0.98 % in solid state at RT).[141,142] Most strategies typically combine the aforementioned approaches.[139,143] Luminescence from these complexes is MC based and the strictly spin-forbidden nature of the relaxation leads to relatively long lifetimes (µs-ms). HS MnII complexes present an interesting prospect for OLEDs[141] and also as triboluminescent sensors that respond to mechanical stress.[129,139,144] In strong LFs, MnII is LS and has an electron hole in the t2g orbitals analogous to Fe(III) complexes, which promotes a parity-allowed strongly colored LMCT transition. It is 23 conceivable to achieve emission from LS d5 compounds if the LF splitting is strong enough and low energy MC states are destabilized enough. Such 2LMCT emission has been seen in very few Fe(III) complexes such as 33+, 10+ (schemes 1.1 and 1.4) and [Fe(ImP) ]+2 (ImP = 1,1′-(1,3- phenylene)bis(3-methyl-1-imidazol-2-ylidene)) (23+) (em = 736 nm, 0.2 ns  = < 1 % in CH3CN at RT).[103,145] There are no known emissive LS MnII complexes. MnII may require stronger field ligands to achieve 2LMCT emission, originating from the lower intrinsic LF of the doubly-charged Mn ion vs. the triply-charged Fe ion.[103] To increase the oxidation state of MnII compounds a suitable chemical (or electrochemical) oxidant must be selected.[72,122] The oxidation of MnII TMCs to MnIV is dependent on the II/III and III/IV redox couples;[128,146] complexes with weakly donating ligands will have the intrinsic problem of having the III/IV oxidation couple being too high in potential to achieve and, if achieved, being too unstable and decomposing. This is seen with [Mn(terpy) 2+2] (terpy = 2,2';6',2"-terpyridine) (322+) that has a very high III/IV oxidation couple of 1.39 V (vs ferrocene/ferrocenium).[32] Introduction of more strongly-donating ligands such as phtmeimb makes the oxidation couples more negative (II/IV = 2.09 V and III/IV = 0.77 V vs ferrocene/ferrocenium)[30] and higher oxidation states more stable. Thus, higher oxidation states can be reached with various strong one electron chemical oxidations such as thianthrene radical cation [C H S +•12 8 2] (0.87 V vs ferrocene/ferrocenium in CH3CN), nitrosonium salts [NO]+ (0.87 V vs ferrocene/ferrocenium in CH3CN), or silver salts (Ag+) (0.65 V vs ferrocene/ferrocenium in DCM).[32] 1.3.2 Manganese(III) Octahedral MnIII d4 TMCs can exist in both HS (S = 2) or LS (S = 1) configurations. HS MnIII TMCs are more common than LS[127,147] but can have a tendency to disproportionate with weak field ligands to MnII and MnIV products,[148] driven largely by the high stability and insolubility of MnO .[123]2 Another factor that can increase the instability of MnIII is the Jahn-Teller distortion present in HS MnIII TMCs.[127,149,150] HS MnIII TMCs have unsymmetrically filled e *g orbitals (t 32g e 1g ) which leads to geometric distortion and tetragonal elongation along the z-axis as seen with the [Mn(acac)3] (acac = acetylacetone) (24) complex.[151,152] This Jahn-Teller axial distortion increases zero-field splitting (ZFS) of 5E GS and, coupled with SOC mixing of states, leads to axial magnetic anisotropy (D). A larger D means the barrier for magnetization can be large; and magnetization can be retained by the TMC following removal of a magnetic field. This ability can enable use of a TMC as a single-molecule magnet (SMM), and has potential in various applications such as information storage.[153] Relaxation to the magnetic GS is slow with large barriers, but there is always a possibility for fast decay via a different mechanism (e.g. quantum tunneling).[154] There are a number of MnIII TMCs that are used as SMMs as a result of Jahn-Teller induced axial anisotropy.[153,155] LS MnIII TMCs are more scarce, the LF strength required for MnIII spin-crossover (SCO) is smaller than LF strength for MnII SCO due to its larger charge, thus (in the right ligand environment) the possibility for switchable SCO 24 i.e. using temperature or pressure.[156,157] There are few reports of LS MnIII TMCs with stronger field ligands including [Mn(CN)6]3(253, scheme 1.7),[131] [Mn(phtmeimb) ]+2 (26+, scheme 1.7),[158] [Mn(HB(3, 5-Mepz) ) ]+ 3 2 (HB(3, 5-Mepz) = 3, 5-dimethyl pyrazolyl borate) (27+, scheme 1.7)[159] and [Mn(LCNC) +2] (LCNC = tert-butyl-carbazole dicyclohexylmesoionic carbene) (29+, scheme 1.7).[29] Scheme 1.7: Molecular structures of HS d4- MnIII complex 24 and LS d4- MnIII complexes 253–27+ and 29+. Mn d4 compounds are more strongly colored than their d5 counterparts, as d-d transitions are no longer spin-forbidden, although still Laporte-forbidden. Photophysically the Tanabe- Sugano diagram shows that MC SF emission from MnIII is conceivable in very strong LFs, where the 5E state is sufficiently destabilised and the SF states (1E 12 and T2) become the lowest energy ESs (black circle, figure 1.12).[21] 25 Figure 1.12: Simplified Tanabe-Sugano diagram for d4 electronic configuration in octahedral symmetry (C/B = 4, scaled by Racah B parameter) with microstates of the pentet (orange), triplet (blue) and singlet SF states (red).[80] However SF MC luminescence from molecular Mn d4 TMCs, has yet to be reported for HS or LS, with only few reports of emission e.g. MnIII clusters[160] or LC emission.[161] 1.3.3 Manganese(IV) When oxidized mononuclear MnIII and MnII TMCs can yield MnIV TMCs. These compounds are particularly uncommon with only a few octahedral complexes reported.[127] MnIV is highly charged and has a small ionic radius, and so exists largely as its stable oxide (MnO2) or with other strongly donating or anionic ligands capable of stabilizing the MnIV cation. Most reports of octahedral MnIV TMCs contain both nitrogen and oxygen donor ligands, e.g. [Mn(sal)2(bipy)] (sal = salicylic acid) (31, scheme 1.8)[162,163] or [Mn(als) 2] (als = 3-((2- hydroxybenzylidene)amino)propanoic acid)) (32, scheme 1.8)[164] with a Schiff base N, O type ligand.[165,166] There are even fewer MnIV TMCs complexed solely with nitrogen donors due to the fact they are not as strongly donating and don’t stabilise the higher Mn oxidation states as well. However, there are examples where ligands have been designed in such a way to increase stabilization e.g. [Mn(tBu-terpy) 4+ t2] ( Bu-terpy = tert-butyl 2,2';6',2"-terpyridine) (33, scheme 1.8)[32] and [Mn(bigH) ]4+2 (bigH = H-biguanide) (34, scheme 1.8)[31] where stabilization is enhanced by the donating capability of the ligands. Other examples are seen with [Mn(HB(3, 5-Mepz) ) ]2+ (283 2 , scheme 1.7),[167] [Mn(big) +3] (big = biguanide) (35, scheme 1.8)[27] and [Mn(L-6H)]2 (L-6H = hydrazine clathrochelate) (36, scheme 1.8)[34,168] where stability is enhanced by using anionic ligands and/or steric bulk. As interest in synthesizing and isolating MnIV TMCs has increased to harness their unique properties and photophysics, more emphasis has been placed on ligand modification with strongly donating and/anionic ligands. This is apparent with recent examples using carbene-based anionic ligands e.g. [Mn(LCNC) +2] 26 (LCNC = tert-butylcarbazole dicyclohexylmesoionic carbene) (302, scheme 1.7)[29] that stabilise the MnIV centre.[123,127] Scheme 1.8: Molecular structures of d3- MnIV complexes 31–362 As an analogue to CrIII d3 configuration, simplified octahedral MnIV photophysics (without ES distortion or CT state considerations) can be better understood with the Tanabe-Sugano diagram (figure 1.13). Figure 1.13: Simplified Tanabe-Sugano diagram for d3 electronic configuration in octahedral symmetry (C/B = 4, scaled by Racah B parameter) with microstates of the quartet (blue) and doublet SF states (red).[80] Population of the lowest energy doublet states (2T1 and 2E) requires a LF strength beyond the crossing point (black circle, figure 1.13). Back-intersystem crossing (i.e. 2T /21 E to 4T2) is still feasible when close to the crossing point. To avoid this, stronger LFs are used to increase the 27 energy separation between the quartet and doublets states. Such emission has been reported with MnIV in doped solid state materials, reporting red emission,[119,169–172] whereas CrIII SF emitters have been primarily reported as NIR emitters. Such a blue shift in emission is thought to be a result of the increased metal ion charge. The smaller MnIV cation (MnIV = 0.53 Å vs CrIII = 0.62 Å)[81] leads to increased interelectronic repulsion, seen by the comparative Racah B and C parameters (CrIII B ≈ 918 cm1 and C ≈ 3850 cm1 vs MnIV B ≈ 1160 cm1 and C ≈ 4303 cm1) for free ions.[85] The leads to the increased energy of the 2T 21 and E states (relative to Cr(III)) and a blue shift in emission to the visible region).[85,173] While important, the reason for emission blue shift is more multifaceted. For example in MnIV doped fluorides and oxides the effect of Mn-L bond length can be seen. In doped fluorides with shorter fixed bond lengths they have generally higher Racah B parameters and higher energy emission, compared with doped oxides.[169] This larger degree of bond covalency within the doped oxides leads to an increased nephelauxetic effect (leading to lower interelectronic repulsion and lower emission energies). This effect translates to the only emissive molecular MnIV compound [Mn(phtmeimb) ]2+ (372+2 , scheme 1.9), which is subject to increased M-L covalency, with significantly red shifted emission to NIR region (em= 828 nm,= 1.5 µs in solid state at 85 K).[174] Scheme 1.9: Molecular structure of emissive [Mn(phtmeimb) 2+ 2] 372 While weakly emissive, the red shift in emission is interesting considering the increase in charge compared with Cr(III). However, with only one molecular emissive octahedral MnIV complex reported, the photophysics and the nephelauxetic effect for these types of TMCs are not well understood.[82] Nevertheless, it is clear from 372+ that ligand design is critical. Strongly donating ligands such as carbenes or even anionic ligands are required to ensure SF states are lowest in energy and to stabilise the high oxidation state. An additional challenge that is coupled with the +IV oxidation state is the presence of low energy CT (4LMCT and 2LMCT) states, which can deactivate emission from SF states.[174] The 4LMCT absorption band for 372+ peaks at 500 nm, with a tail up to 625 nm, and emission is reported at 828 nm. In this case, the 4LMCT and associated 2LMCT states are not causing deactivation of the SF state. However, with complex 302+ (scheme 1.7) the 4LMCT band is 28 reported as panchromatic with the peak at 730 nm and tail up to 930 nm. Although no emission has been reported, SF emission within that range will probably be deactivated.[29] If SF emission were to occur from complexes with such panchromatic LMCT absorption it would have to be significantly red shifted to avoid deactivation. It is possible for MnIV complexes to be photoactive without being emissive. This is seen with the [(Bn-TPEN)Mn(O)]2+-[Sc(OTf)3]2 (Bn-TPEN = N-benzyl-N,N′,N′-tris(2-pyridylmethyl)-1,2- diamino-ethane) (382+, scheme 1.10); although not emissive, it is still able to photooxidize and hydroxylate challenging compounds including benzene with irreversible decomposition of the complex. Scheme 1.10: Molecular structure of non-emissive strong photooxidant, together with Lewis acid activation from 2 eq of Sc[OTf] 382+.[61]3 This bimolecular reactivity is attributed to a non-emissive long lived mixed 2MC/2LMCT state (= 6.4 µs in CH3CN/TFE mixture) that, following irradiation at 440 nm, oxidises benzene and transfers an oxygen atom to the benzene radical cation forming phenol.[33,61,175] Such reactivity is enabled by the second coordination sphere, i.e. the coordination of 2 eq of Sc[OTf]3. This coordination significantly enhances the GS oxidation potentials of the [(Bn-TPEN)Mn(O)]2+ complex (0.38 V with no Sc[OTf]3 to 0.96 V (vs ferrocene/ferrocenium) with 2 eq). As a result, the ES potential reaches 1.7 V (vs ferrocene/ferrocenium). Similar shifts also occurred with coordination of 2 eq of Sc(NO3)3 (ESox = 1.72 V, = 7.1 µs in CH3CN/TFE mixture).[176] This difference enables the oxidation of aryl substrates and significantly increases the ET transfer rates due to a large increase in driving force and a decrease in reorganizational energy, i.e. the Mn=O bond length change following ET is much smaller due to prior elongation at Sc[OTf]3 coordination.[177,178] This complex represents one of the very few examples of the unique photoredox capabilities of MnIV TMCs, particularly with challenging substrates.[62,175,178–182] 29 1.4 Nickel(II) The area of emissive molecular nickel TMCs is relatively underdeveloped, and there are only very few reports of emission. For example a 3MLCT emission with Ni0 d10 tetrahedral isocyanide complexes [Ni(CNAr5NC)2] (CNAr5NC = 1,3-bis(N-formyl-4-methyl-6-phenylanilin-2- yl)benzene) (39) (em = 510 nm, = 1.1 µs at 77 k in toluene) and [Ni(CNAr (th)NC) ] (CNAr (th)5 2 5 NC = 1,3-bis(N-formyl-4-methyl-6-phenylanilin-2-yl)thiophene) (40, scheme 1.11) (em = 560 nm, = 1.2 µs at 77 K in toluene).[183,184] Emission from NiII has also been reported in the solid state[35] and one recent report of an emissive square planar NiII d8 complex, [Ni(dpb)(Cbz)] (dpb = 1,3-bis(N-alkylbenzimidazol-2'-yl)benzene) and Cbz = carbazole) (41, scheme 1.11) which shows metal perturbed 3ILCT emission (em = 468 nm, = 0.11 µs at 77 K in solid state).[9,16,185,186] Scheme 1.11: Molecular structures of emissive Ni0 TMCs 39 and 40, emissive NiII complex 41 and non- emissive [Ni(ddpd)2]2+ 422+, [Ni(phen) ]2+ 432+ and hexammine NiII complexes 442+3 . Based on the Tanabe-Sugano diagram it is theoretically possible to achieve MC SF emission from molecular NiII in an octahedral environment, but this has yet to be achieved. The Tanabe- Sugano diagram for d8 configurations demonstrates like for d2, d3 and d4 configurations  that as LF strength increases, the intraconfigurational SF states become the lowest ES (black circle, figure 1.14), while the interconfigurational ES are linearly destabilized. The main difference here is that the SF transition involved the e *g orbitals, not the t2g orbitals as with the other configurations.[21] The impact of this difference on state splitting is not well understood due to the lack of emissive molecular NiII octahedral complexes.[187] 30 Figure 1.14: Simplified Tanabe-Sugano diagram for d8 electronic configuration (C/B = 4, scaled by Racah B parameter) with microstates of the triplet (blue) and singlet SF states (red).[80] There are many octahedral NiII complexes known, with their synthesis being relatively uncomplicated from readily available precursors (e.g. Ni[BF4]2۰6H2O), yet none show SF luminescence. This phenomenon has been explained by detrimental singlet/triplet state mixing and this effect is visible in the absorbance spectra due to intensity borrowing, i.e. when close to the crossing point of the 3T and 12 E states (black circle) the proximity of the states leads to mixing and singlet intensity borrowing from the spin allowed transition via SOC.[187– 189] With stronger LFs [Ni(ddpd) ]2+ and [Ni(phen) ]2+ (phen = 1,10-phenanthroline) (422+2 3 and 432+, scheme 1.11), the 1E state can be detected as a low energy shoulder on the 3T2 band, weaker field ligands such as NH3 in the complex [Ni(NH3)6]2+ (442+, scheme 1.11) have a spectrum where the 1E state manifests as a high energy shoulder or band.[187,190,191] As states mix, the SF nested transition 1E gains more 3T2 character and becomes more structurally distorted, leading to efficient non-radiative decay.[192] It is possible to increase the energy separation between these states by hydrostatically increasing pressure (53+, section 1.1), leading to less mixing and a more ‘pure’ SF state.[193–195] However, it is clear that stronger LFs are required to split the triplet and singlet states and avoid state mixing/bISC. This is a challenge with d8 configurations, as LF strength increases a square planar configuration is preferred over octahedral coordination.[196] This can be seen with the heavier NiII analogues PdII and PtII that almost exclusively prefer square planar configurations. Thus, a balance is required with the strength of LF imposed on NiII, enough to separate states but not too much as to promote square planar geometry. Another method could be to impose an octahedral geometry with a rigid strong field (-donation, π-accepting or a combination of both) or a cage ligand, thus ensuring octahedral geometry is maintained along with a strong LF.[197–201] 31 2. Aims of Work There are very few MnIV transition metal complexes and only one emissive complex,[174] and so the photophysical properties of such complexes are poorly understood. Thus, the primary objective of this work is to design and synthesize photoactive octahedral MnIV complexes as analogues to MC emissive CrIII complexes, and to further investigate their photophysical properties and applications. This work initially centers on the idea of using strongly donating tridentate ligands capable of 6-membered coordination, to stabilise the highly charged MnIV ion and ensure that the intraconfigurational states (2E/2T1) are lowest in energy. The literature known ligand 2,6- diguanidylpyridine (dgpy) (scheme 1.12a) offers such characteristics, having already yielded 3LMCT luminescence with cobalt(III).[202] Synthesis will begin with a readily available MnII precursor (Mn[OTf]2), following isolation and characterization of [Mn(dgpy) ]2+2 the primary aim is to find a viable synthetic route to [Mn(dgpy) 4+2] (e.g. chemical oxidation via [Mn(dgpy)2]3+) (scheme 1.12b). Full characterization (structurally, electronically, theoretically and magnetically) will follow isolation of [Mn(dgpy) 4+2] . Photophysical investigation will then be a significant focus; absorption, variable temperature emission and time resolved spectroscopy, along with detailed theoretical studies (DFT) will be utilized to better understand the ES dynamics. Application work will follow photophysical characterization, pending the discovery of advantageous ES ordering and lifetimes. Scheme 1.12: Structures of a) 2, 6-diguanidylpyridine (dgpy) ligand and b) [Mn(dgpy)2]n+ (n = 24) complex. The other objective of this work is to more fully understand and investigate the nature of MC SF transitions in octahedral NiII complexes, and how they can be influenced to ultimately yield SF emission from molecular octahedral NiII complexes. To achieve this, a series of three literature known complexes [Ni(terpy) ]2+, 422+2 , 432+ and two novel octahedral homoleptic nickel(II) complexes (scheme 1.13) will be synthesised and fully characterized. 32 Scheme 1.13: Structures of a) [Ni(dgpy)2]2+ and b) [Ni(tpe) 2+2] complexes. The photophysical properties will be described using absorption and emission spectroscopies, and theoretically using detailed LF theory analysis along with DFT, CASSCF and coupled potential energy surface analysis. These compounds will also be tested to see if increased hydrostatic pressure can separate the intra- and interconfigurational LF states.[87] To further this study the influence of increased σ-donating and π-accepting ligands will be looked at. The synthesis of [Ni(CNC)(NCN)]2+ (CNC = 1,1’-(pyridin-2,6-diyl)bis(3-methyl-1H- imidazol-3-ylidene) and NCN = 1,3-bis(2-pyridyl)imidazolylidene) will explore the influence of increased σ-donation, and synthesis of [Ni(dcpp) ]2+2 (dcpp = 2,6-bis(2-carboxypyridyl)pyridine) will look at increased ligand π-acceptance (scheme 1.14). Scheme 1.14: Structures of a) [Ni(CNC)(NCN)]2+ and b) [Ni(dcpp) ]2+2 complexes. As mentioned in a strong enough ligand field it is possible for NiII to adopt square planar coordination. These complexes will first be characterized structurally to confirm an octahedral environment and then characterized optically to investigate the position of ligand field states and determine if SF emission is possible. 33 3. Results and Discussion Most of the results and findings detailed and documented in this thesis have been submitted to and/or published as scientific articles in peer-reviewed chemistry journals. Each article discussed is reprinted with permission from respective publishers. Synthesis, characterization and quantum chemical calculations of the novel [Mn(dgpy) ]n+2 complex in three different oxidation states (II, III, IV) is presented in section 3.1 “The Full d3– d5 Redox Series of Mononuclear Manganese Complexes: Geometries and Electronic Structures of [Mn(dgpy) n+2] ”. The investigation and study of [Mn(dgpy) ]n+2 in three different oxidation states represents the first of its type for Mn, showing the direct geometric and electronic effects related to the challenging sequential removal of two electrons from e *g orbitals on the way to the MnIV complex. Synthesis of this series begins with [Mn(dgpy)2][OTf]2 from Mn[OTf]2 and dgpy ligand, the complex precipitated from dry THF. Both almost colorless cis-fac and mer isomers were isolated due to a result of the lack of ligand field stabilization and the inherent substitution lability of HS MnII complexes. Following isolation of [Mn(dgpy)2][OTf]2, electrochemical characterization was carried out to reveal the oxidative potentials required to achieve +IV oxidation state from +II. Cyclic voltammograms revealed the +II/+III couple at 0.26 V (vs ferrocene/ferrocenium) and the +III/+IV couple at 0.58 V (vs ferrocene/ferrocenium). Thus, 1 eq of a suitable oxidant (AgPF +/06, E1/2(Ag ) = +0.04 V in CH3CN) was selected and used to make orange [Mn(dgpy) 3+2] , which following removal of solid Ag was isolated as mainly the mer isomer, [Mn(dgpy)2][PF6]3. [Mn(dgpy) 4+2] was synthesised in a similar fashion, but with 2 eq of a stronger chemical oxidant (C +٠ +/012H8S2 , E1/2([C12H8S2] ) = +0.86 V in CH3CN), yielding mer-[Mn(dgpy)2][PF6]4 as deep purple crystals. The geometric studies and analysis using computational methods (DFT) and X-ray diffraction (XRD) showed a clear relationship of decreased bond length with decreased antibonding orbital population. Superconducting quantum interference magnetic measurements (SQUID) confirmed the spin- only magnetic moments and thus the spin state of the complexes. Furthermore, the axial magnetic anisotropy (D) of [Mn(dgpy)2]3+ was confirmed to be 3.84 cm-1 which agrees well with the CASSCF/NEVPT2 calculated axial anisotropy value of 4.54 cm-1. Most importantly, absorbance spectroscopy of [Mn(dgpy) 4+2] revealed an intense (Ɛ > 2700 M1 cm1) panchromatic LMCT band up to 950 nm, which masks any (calculated with CASSCF/NEVPT2) d-d transitions. The [Mn(dgpy) 3+2] and [Mn(dgpy) 4+2] complexes represent rare examples of MnIII and MnIV supported solely by nitrogen donors. 34 In Section 3.2, “Oxidative Two-State Photoreactivity of a Manganese(IV) Complex using NIR Light”, the photophysics and resulting applications of the [Mn(dgpy) 4+2] complex are detailed and listed. [Mn(dgpy) 4+2] displays weak NIR-II phosphorescence at 1435 nm in solid state at 77K following laser excitation at 730 nm. Using DFT it was found that the emission originates from a mixed 2MC/2LMCT state. Following transient absorption spectroscopy this state was found to be long lived enough (1.6 ns) to participate in dynamic bimolecular chemistry. Using steady state absorption spectroscopy [Mn(dgpy) ]4+2 was found to be strongly photooxidative following NIR irradiation (850 nm). The complex is able to participate in dynamic quenching via the lower energy 2MC/2LMCT state and oxidise naphthalene (Eox ≈ 0.93 – 1.16 V vs ferrocene/ferrocenium) forming a naphthalene radical cation and [Mn(dgpy) 3+2] . The naphthalene radical cation was trapped using an electron-poor pyrazole (ethyl 1-phenyl-1H- pyrazole-4-carboxylate) in a Nicewicz-type oxidative coupling and characterized with ESI mass spectroscopy (C16H15N2O2 m/z = 267). Mechanistic aspects were examined using DFT and confirmed with time resolved spectroscopy. They showed following low energy excitation a 4LMCT state forms and ISC occurs in 780 fs to a lower energy mixed 2LMCT/2MC, this state then oxidises naphthalene forming a radical cation and LS [Mn(dgpy) ]3+2 . LS [Mn(dgpy) 3+2] , then undergoes SCO to form HS [Mn(dgpy) 3+2] . [Mn(dgpy) 4+2] also displays surprising reactivity and is able to oxidise substrates with significantly higher potentials up to benzene (Eox ≈ 1.98 V vs ferrocene/ferrocenium) and even solvents such as CH3CN and CH3NO2. This reactivity is attributed to the more reactive short lived 4LMCT (1.46 eV) state and enabled by slow ISC. Photooxidation occurs more slowly and statically via a close pre-organized CH3CN molecule, which was confirmed with molecular dynamics simulations and with ESI mass spectroscopy (H/D isotopic) showing a benzene nitrilium ion (m/z = 118 for C8H8N and m/z = 123 for C8H3D5N). Steady state absorption measurements were carried out with mesitylene, toluene and benzene all showing similar reactivity. The radical cations of mesitylene (C15H19N2O2 m/z = 259) and benzene (C12H13N2O2 m/z = 217 and C12H8D5N2O2 m/z = 222) were also trapped in the same Nicewicz-type oxidative coupling as the naphthalene radical cation. The formation of the benzene cross-coupled product 1H-pyrazole-4-carboxylic acid ethyl ester was also confirmed with HPLC (comparing peak area to a known standard) and the reaction yield was found to be 11%. Molecular dynamics simulations also showed close interaction of PF6 anions and complex in solution suggested ion pairing. To further investigate if ion pairing has an effect on reactivity steady state absorption measurements were carried out with increased ionic strength (0, 50, 100 and 200 mM [nBu4N][PF6]). Results showed a slowing of rate with increased in ionic strength which suggests slower photooxidation or slower cage escape. This coupled with no changes observed in transient spectroscopy at high ionic strength (100 mM) indicates that ion pairing likely affects cage escape rather than the initial static quenching step. 35 In Section 3.3, “Coupled Potential Energy Surfaces Strongly Impact the Lowest-Energy Spin-Flip Transition in Six-Coordinate Nickel(II) Complexes”, describes the synthesis and characterization of a series of five NiII octahedral complexes. Three literature known complexes [Ni(terpy) )2+, 2 [Ni(phen) 2+ 2) and [Ni(ddpd) )2+2 and two novel complexes [Ni(dgpy) )2+2 and [Ni(tpe) )2+2 to help understand the specific dynamics governing SF states in octahedral NiII d8 complexes. Unlike d2, d3 and d4, the SF transition occurs in the antibonding e *g orbitals and there are no emissive molecular octahedral complexes NiII that exist to date.[187] Each of the five complexes show increasing LF strength beginning with [Ni(dgpy) )2+2 which was synthesized from Ni[BF4]2۰6H2O and 2 eq of dgpy ligand. LF states can be seen with absorption spectroscopy as a result of 1E state intensity borrowing, and this complex shows weakest LF strength of the five, with the 1E state manifest as a high energy shoulder (12 600 cm1) on the 3T2 state (11 990 cm1). [Ni(terpy)2)2+ shows increased LF strength with 3T2 state being higher in energy (12 420 cm1), however the 1E state (12 630 cm1) remains as a higher energy shoulder. As ligand field strength increases with the [Ni(phen)2)2+ and [Ni(ddpd)2)2+ complexes the 1E state moves to lower energy (phen = 11 650 cm1, ddpd = 11 300 cm1), compared with the 3T2 state (phen = 12 680 cm1, ddpd = 12 700 cm1). [Ni(tpe) )2+2 was synthesised in a similar way to [Ni(dgpy)2)2+, with 2 eq of tpe ligand stirred with Ni[BF4]2۰6H2O. Separation of LF states reaches its largest with this complex (3T2 = 13 380 cm1, 1E = 11 650 cm1). None of these complexes are luminescent, even at high hydrostatic pressure as seen with [Ni(ddpd)2)2+ at 62 kbar. With detailed DFT, LF studies and coupled potential energy surface studies, it is shown that singlet/triplet state mixing is the primary reason for this occurrence. This mixing via SOC leads to the 1E state taking on more anharmonic distorted character, which leads to increased non- radiative decay. Even though LF strength increases with this series (also seen with CASSCF/NEVPT2 calculations) the singlet/triplet state proximity remains an issue which inhibits the possibility of SF emission. 36 In Section 3.4, “Influencing Ligand field states of Nickel(II) Complexes with Strongly σ-Donating and π-Accepting Ligands”, further investigates increasing ligand field strength in NiII complexes by using strongly σ-donating and π-accepting ligands. This sheds further light on required ligand characteristics for SF emission from octahedral NiII. This chapter describes synthesis of a NiII carbene complex [Ni(CNC)(NCN)]2+, beginning with a [NiBr(CNC)]Br precursor the NCN ligand is deprotonated with a weak base (NaOAc) and heated for complexation to occur. Here the LF strength is higher resulting from the three strong σ-donor carbene ligands. Following XRD structural characterization it is seen that the LF is too strong and has lead NiII to adopt a square planar and 4+1 geometry, meaning an octahedral d8 photophysical description no longer applies. The second complex [Ni(dcpp) 2+2] synthesised looks at the effect of increasing LF strength with π-accepting ligands. This complex was synthesized by stirring Ni[BF4]2۰6H2O and 2 eq of dcpp ligand. Following isolation and XRD it can be seen that the complex exists in an octahedral geometry. Absorption spectroscopy reveals that even at this increased π-accepting LF the 3T2 and 1E LF states remain mixed and are not separate. The 1E band is seen as a low energy should (11 480 cm1) on the 3T2 band (12 750 cm1). Thus SF emission does not occur from this complex. Furthermore, the dcpp ligand contains two carbonyl groups, this provides the possibility to further increase the π-accepting ability of dcpp upon Lewis acid coordination. Sc[OTf]3 (12 eq) were coordinated to [Ni(dcpp) ]2+2 as a second coordination sphere. Absorption spectroscopy showed unexpected results with the LF states 3T2 and 3T1 moving to lower energies with increased equivalence of Sc[OTf]3 and movement of the 1E state to higher energies. The movement of the 3T2 and 3T1 to lower energies could be a result of decreased the σ-donation of the dcpp ligand lowering Δo. Movement of the 1E state to higher energies could be a result of changes in coordinated geometry upon addition of Sc[OTf]3, which would change the nephelauxetic effect. 37 3.1 The Full d3–d5 Redox Series of Mononuclear Manganese Complexes: Geometries and Electronic Structures of [Mn(dgpy) ]n+2 Nathan R. East, Christoph Förster, Luca M. Carrella, Eva Rentschler and Katja Heinze. Inorg. Chem. 2022, 61, 37, 14616–14625 This article reports the novel [Mn(dgpy) n+2] complexes isolated in three oxidation states (II, III, IV). Aided by DFT and CASSCF studies, this complete series gives unique insight on how structural, electronic, optical and magnetic properties change with sequential removal of an electron from the e *g orbitals. In particular, [Mn(dgpy) ]4+ 2 is a rare example of an octahedral MnIV complex stabilized solely by nitrogen donor ligands, showing strong (Ɛ > 2700 M1 cm1) panchromatic LMCT absorbance. Author contributions Synthesis and characterization of the title compounds, along with DFT studies, CASSCF calculations and optical studies were carried out by N. R. East. All crystal structures were solved and refined by Dr. C. Förster. SQUID magnetic measurements were performed by Dr. L. M. Carrella (group of Prof. Dr. E. Rentschler). The manuscript and supplementary information were written by Prof. Dr. Katja Heinze and N. R. East. Supporting Information Found on page 97 “N. R East, C. Förster, L. M. Carrella, E. Rentschler, K. Heinze, Inorg. Chem. 2022, 61, 37, 14616– 14625. Copyright 2022 American Chemical Society. Reproduced with permission” 38 39 40 41 42 43 44 45 46 47 48 3.2 Oxidative Two-State Photoreactivity of a Manganese(IV) Complex using NIR Light Nathan R. East, Robert Naumann, Christoph Förster, Charusheela Ramanan, Gregor Diezemann and Katja Heinze. Preprint [DOI: 10.26434/chemrxiv-2023-bhl82] (submitted) This article reports the photophysical properties of the [Mn(dgpy) 4+2] complex, possessing unique dual state photoreactivity using NIR excitation, acting as a luminescent strong photooxidant capable of oxidising challenging aryl substrates such as benzene via two mechanisms. At 730 nm excitation, a low energy NIR-II phosphorescent (1435 nm at 77 K) state is dynamically quenched by naphthalene, with transient absorption spectroscopy confirming a lifetime of 1.6 ns, and DFT studies confirming the states identity as a mixed 2MC/2LMCT. Unconventionally, a high energy 4LMCT state (1.42 V vs ferrocene/ferrocenium) is statically quenched by benzene via a CH3CN solvent molecule intermediate. Radical cations of these substrates were also trapped via Nicewicz-type oxidative coupling and products characterized with ESI mass spectrometry and HPLC, confirming reactivity. This is the first example of an emissive molecular MnIV complex capable of dual state photooxidative reactivity upon NIR excitation, and is a significant step in the pursuit of earth-abundant photoactive transition metal complexes. Author contributions N. R. East performed syntheses, reactivity studies, irradiation experiments and computational studies. Dr. R. Naumann performed and analyzed the luminescence and ultrafast time- resolved experiments and provided data interpretation. Dr. C. Förster performed and assisted with the computational studies. Dr. C. Ramanan assisted with the time-resolved experiments. Prof. Dr. G. Diezemann performed and analyzed the molecular dynamics simulations. The manuscript and supplementary information were written by Prof. Dr. Katja Heinze, Dr. R. Naumann and N. R. East. Supporting Information Found on page 123 “N. R East, R. Naumann, C. Förster, C. Ramanan, G. Diezemann, K. Heinze, Preprint [DOI: 10.26434/chemrxiv-2023-bhl82] Copyright 2023, Reproduced with permission” 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 3.3 Coupled Potential Energy Surfaces Strongly Impact the Lowest-Energy Spin-Flip Transition in Six-Coordinate Nickel(II) Complexes Nathan R. East, Chahinez dab, Christoph Förster, Katja Heinze and Christian Reber. Inorg. Chem. 2023, [DOI: 10.1021/acs.inorgchem.3c00779] (accepted) This article illustrates that spin- flip luminescence from six- coordinate d8 nickel(II) complexes is not easily observed as for molecular d3 chromium(III) compounds. A series of five nickel(II) complexes is explored and their lowest-energy singlet and triplet excited states characterized by DFT, CASSCF, experimental spectroscopy and theoretical models, revealing a large degree of detrimental mixing of states via spin-orbit coupling, leading to non-radiative decay. Author contributions N. R. East performed synthesis and characterization of the title compounds, along with DFT studies, CASSCF calculations and optical studies. Dr. Chahinez Dab (group of Prof. Dr. Christian Reber) measured all Raman and luminescence spectra on crystalline samples at variable temperature and pressure and analyzed the spectroscopic results. All crystal structures were solved and refined by Dr. Christoph Förster. The manuscript and supplementary information were written by Prof. Dr. Christian Reber, Prof. Dr. Katja Heinze and N. R. East. Supporting Information Found on page 165 “N. R East, C. Dab, C. Förster, K. Heinze, C. Reber, Inorg. Chem. 2022, [DOI: 10.1021/acs.inorgchem.3c00779] Copyright 2022 American Chemical Society. Reproduced with permission” 66 67 68 69 70 71 72 73 74 75 76 3.4 Influencing Ligand Field States of Nickel(II) Complexes with Strongly σ-Donating and π-Accepting Ligands The prospect of SF emissive octahedral NiII complexes is attractive, however there are none reported to date. To achieve emission from NiII octahedral complexes the intra- and interconfigurational states must be separated, and intraconfigurational states must be lowest in energy.[188] To do this a strong field σ-donating ligand is required, this will destabilize e *g orbitals. However, if the LF is too strong NiII will adopt a square planar or five-coordinate configuration, thus a balance is required.[81] An alternative route is to use a more π-accepting ligand which can stabilize the t2g orbitals. Both methods aim to increase Δo and the energy of the 3T2 state.[187] In principle this can better separate the LF states, however the effect of σ- donation and π-acceptance on the intraconfigurational 1E state is less clear. Increasing the LF strength using σ-donors is not a new concept and has been used on NiII with strong field ligands such as phen or ddpd.[190,191] These ligands however do not provide a strong enough LF for NiII, thus stronger σ-donating ligands such as carbenes are required. Detailed in this chapter is the synthesis along with structural characterization of a NiII heteroleptic carbene complex (scheme 3.1a) containing the literature known CNC ligand (1,1’- (pyridin-2,6-diyl)bis(3-methyl-1H-imidazol-3-ylidene)) (scheme 3.1b) and NCN ligand (1,3- bis(2-pyridyl)imidazolylidene) (scheme 3.1c).[203,204] Scheme 3.1: Molecular structures of a) [Ni(CNC)(NCN)]2+ and ligands b) CNC and c) NCN. To investigate the effect of increased ligand π-accepting ability, the literature known dcpp (2,6-bis(2-carboxypyridyl)pyridine) (scheme 3.2b) ligand[205] containing two electron- withdrawing carbonyl groups is used to make the homoleptic [Ni(dcpp) ]2+2 complex (scheme 3.2a). Detailed in this chapter is the synthesis along with structural and optical characterization of [Ni(dcpp) ]2+2 which contains a strong π-accepting ligands. 77 Scheme 3.2: Molecular structures of a) [Ni(dcpp) ]2+2 and ligand b) dcpp. What will also further be discussed is the influence of the second coordination sphere on the LF parameters of [Ni(dcpp)2]2+. This is carried out by coordinating the Lewis acid Sc[OTf]3 to the ligand carbonyl groups and measuring changes to the LF states using absorption spectroscopy. Supporting Information Found on page 192 78 3.4.1 Increasing σ-Donation It is well known that nickel(II) unlike its heavier analogues PtII and PdII can form octahedral complexes, this is a result of NiII containing a weaker intrinsic LF splitting.[101] However, there is a boundary where this is not the case and it becomes more energetically favourable too adopt square planar coordination over octahedral. A strong ligand field is required for favourable separate ES ordering and so a balance is required. Few octahedral NiII complexes containing carbenes have been reported,[198–200] but combination of CNC and NCN ligands can potentially provide a strong ligand field with three carbenes that is not too strong. Synthesis 2Br Br PF6 N NNaOAc NN N NN N N + DMSO N NiII N II N N HOAc NNi N N N NaPF6 N Br Scheme 3.3: Synthesis route of [Ni(CNC)(NCN)]2+ from the known bromide CNC NiII complex [NiBr(CNC)]Br,[203] NaOAc acts as a weak base to deprotonate the H-NCN+ pro-ligand to the carbene (see section 6.4, pg 192 for details) Structural Characterization To further evaluate photophysical properties it was first required to verify if the complex was coordinated as octahedral or square planar via single-crystal XRD. The complex [Ni(CNC)(NCN)][PF6]2 crystallized in the monoclinic space group P21/c with two independent cations in the asymmetric unit. 79 Figure 3.1: Molecular structures of cation [Ni(CNC)(NCN)]2+ with photo of crystals. Structures show 4+1 coordination (left) and square planar coordination (right). Thermal ellipsoids are displayed at 50% probability. Hydrogen atoms, counterions, and solvent molecules are omitted for clarity. Table 3.1: Selected bond lengths/Å and angles/deg of the two cations of [Ni(CNC)(NCN)]2+, for comparison also obtained from DFT calculations. exp / Å DFT / Å exp / deg DFT / deg Ni(1) C(14) 1.856(4) 2.003 C(14)  Ni(1)  N(3) 177.70(15) 173.00 Ni(1) N(3) 1.862(3) 2.076 C(14)  Ni(1)  C(1) 98.01(17) 106.92 Ni(1) C(1) 1.920(4) 2.084 N(3)  Ni(1)  C(1) 81.26(16) 76.40 Ni(1) C(10) 1.924(4) 2.086 C(14)  Ni(1)  C(10) 99.61(19) 100.53 Ni(2) C(36) 1.920(4) 2.092 N(3)  Ni(1)  C(10) 81.24(17) 76.52 Ni(2) N(12) 1.883(3) 2.099 C(1)  Ni(1)  C(10) 162.17(18) 152.47 Ni(2) C(27) 1.903(4) 2.093 C(40)  Ni(2)  N(12) 172.86(15) 171.44 Ni(2) C(40) 1.860(4) 2.002 C(40)  Ni(2)  C(36) 99.95(17) 101.34 Ni(2) C(15) 2.436(3) 2.066 N(12)  Ni(2)  C(36) 80.29(16) 75.93 C(40)  Ni(2)  C(27) 98.31(18) 105.78 N(12)  Ni(2)  C(27) 80.71(17) 75.74 C(36)  Ni(2)  C(27) 160.39(18) 151.02 80 What is clear from the structural analysis of this complex is that the LF strength imposed on NiII with the CNC and NCN ligands is too strong and the complex adopts a square planar and 4+1 coordination. The bond angle CNiC is a restrained 162.17° due to five-membered nature of the CNC ligand, bond lengths of the centrally coordinated atoms vary from 1.856 – 1.924 Å. The analysis also shows one cation with a 4+1 coordination, the bond length Ni(2) N(15) is very long 2.436(3) Å, and can most likely be viewed as a weak interaction rather than a genuine single bond. In both cases (square planar (figure 3.2) and 4+1) the orbitals with z-components are lowered in energy (in particular d 2Z ). This will lead to spin pairing i.e. a singlet ground state (figure 3.2). Figure 3.2: Ligand field splitting diagram for square planar coordinated complexes. Orbitals with z- components are lowered in energy. The effect is similar with 4+1 coordination but to a lesser extent. The Tanabe-Sugano diagram for d8 octahedrons no longer applies to this complex. It is clear that if strong σ-donors are used to increase the LF splitting for NiII, they would have to be used in conjunction with a more rigid or cage ligand to maintain an octahedral coordination. 3.4.2 Increasing π-Accepting Introduction of π-accepting ligands to a TM is an alternative way to increase LF splitting. The dcpp ligand contains two carbonyl groups as strong π-acceptors, this ligand also offers additional potential to enhance its accepting capability via Lewis acid coordination. This can help further explore the effect of π-accepting ligand behavior on LF states. The method of Lewis acid coordination has been already successfully used with MnIV complex (372+), where Lewis acid (Sc[OTf]3) coordination significantly enhanced the GS oxidation potential of the complex.[177] 81 Synthesis [BF4]2 O O N O NN CH3CN 2 N + Ni[BF4]2 6H2O N Ni II N -6H2O N O N N O O Scheme 3.4: Synthesis route of [Ni(dcpp) 2+2] . (see section 6.4, pg 192 for details) Structural Characterization Figure 3.3: Molecular structure of the cation [Ni(dcpp) 2+2] with photo of crystals. Thermal ellipsoids are displayed at 50% probability. Hydrogen atoms, counterions, and solvent molecules are omitted for clarity. 82 Table 3.2: Selected bond lengths/Å and angles/deg of the cation of [Ni(dcpp) 2+2] , for comparison also obtained from DFT calculations. exp / Å DFT / Å exp / deg DFT / deg Ni(1) N(2) 2.0819(11) 2.087 N(2) Ni(1) N(4) 177.48(6) 179.96 Ni(1) N(1) 2.0939(11) 2.115 N(2) Ni(1) N(1) 92.07(4) 88.22 Ni(1) N(3) 2.0967(11) 2.114 N(2) Ni(1) N(6) 94.83(4) 88.22 Ni(1) N(4) 2.0819(11) 2.086 N(2) Ni(1) N(5) 86.16(4) 91.74 Ni(1) N(5) 2.0939(11) 2.115 N(2) Ni(1) N(3) 86.94(4) 91.78 Ni(1) N(6) 2.0967(11) 2.114 N(4) Ni(1) N(5) 92.07(4) 88.21 N(4) Ni(1) N(1) 86.16(4) 91.77 N(4) Ni(1) N(3) 94.83(4) 88.25 N(4) Ni(1) N(6) 86.94(4) 91.76 N(5) Ni(1) N(3) 173.10(4) 176.46 As seen from XRD analysis the homoleptic [Ni(dcpp) ]2+2 complex is octahedral. With NNiN bond angles close to 180°, the average NiN bond length ranges from 2.0819  2.0967 Å. With and octahedral coordination the position of LF states can be determined. 83 Absorption Spectroscopy Figure 3.4: UV/Vis/NIR spectrum of [Ni(dcpp)2][PF6]2 in acetonitrile with absorption band assignments. The absorption spectrum of [Ni(dcpp)2][PF6]2 shows LF states 3T2 and 1E are mixed. The 1E state is seen as a lower energy shoulder (11 480 cm1) on the 3T2 state (12 750 cm1), this lack of separation (ΔE = 1270 cm1) hinders any potential SF emission. The separation as seen in the absorption spectrum is comparable to [Ni(ddpd) ]2+2 (432+) (ΔE = 1400 cm1) with the 1E state also seen as a lower energy shoulder (11 300 cm1) on the 3T2 state (12 700 cm1). The 3T1 band is masked by CT bands and cannot be detected in the absorption spectrum preventing Racah B calculation. Increasing the π-accepting nature of the ligand to this extent does not have the desired effect of splitting the LF states sufficiently. The dcpp ligand enables the possibility of further increase in the π-accepting effect, by coordinating a Lewis acid to the carbonyl groups on the ligand. Coordination of 2 eq of Sc[OTf]3 gave an unexpected outcome (figure 3.5). 84 Figure 3.5: UV/Vis/NIR spectra of [Ni(dcpp)2][PF6]2 in acetonitrile following addition of 12 eq of Sc[OTf]3. Photos of [Ni(dcpp)2][PF6]2 in acetonitrile (right) following addition of 1 eq of Sc[OTf]3 (middle) and 2 eq of Sc[OTf]3 (left). It could be expected that with an increase in the π-accepting effect the energy of the t2g orbitals would be more stabilized and Δo would increase, moving 3T2 state to higher energy. Although the effects of increased π-acceptance on the 1E state are not clear, it could also be surmised that there would be an increase in the nephelauxetic effect resulting from increased electron delocalization and enhanced M-L covalency moving the 1E state to lower energy (however, the opposite occurred experimentally).[82] As the equivalence of Sc[OTf]3 increases (up to 2 eq) Δo decreases, with a shift of the 3T2 state to lower energy (1 eq Sc[OTf]3 = 11 730 cm1 and 2 eq Sc[OTf]3 = 10 430 cm1). The energy of the 1E state moves to higher energies with increased Sc[OTf]3 (1 eq Sc[OTf]3 = 13 300 cm1 and 2 eq Sc[OTf] = 13 910 cm13 ). The increase in Racah B parameter for the Lewis acid coordinated complexes of 954 cm1 (1 eq Sc[OTf]3) and 1205 cm1 (2 eq Sc[OTf]3) also reflects this change. This unexpected shift of the 3T2 state to lower energies could be the result of the Lewis acid coordination decreasing the σ-donation of the dcpp ligand, this would lead to a lowering of Δo in-line with experiment. The increase in energy of 1E state and increase in energy of the Racah B parameter implies a decrease in covalency of the M-L σ-bonds, this could be a result of 85 changes in coordination geometry upon addition of Sc[OTf]3, which would alter the nephelauxetic effect. 3.4.3 Conclusion SF luminescence from octahedral NiII complexes remains a challenge due to the close proximity of 3T2 and 1E LF states. This leads to mixing of states and deactivation of the 1E state preventing SF emission. Increasing the LF splitting is a clear way to separate these states and this can be done by increasing the σ-donation or π-accepting natures of coordinating ligands. A significant challenge with NiII is that it adopts a square planer geometry in strong LFs, and so a balance of LF strength is required. This study shows that as σ-donation increases with the combination of carbene ligands CNC and NCN, NiII adopts a square planar or 4+1 configuration i.e. dissociates one to two axial pyridine ligands. The LF strength accompanying ligands with three carbene moieties is too strong. If the strategy to separate LF states in NiII is by increasing LF splitting with σ-donors, it should be accompanied with a very rigid or cage strong field ligands, to ensure octahedral geometry is maintained.[197–201] Increasing the π-accepting natures of the ligands is another way to increase LF spitting with stabilization of the t2g orbitals. [Ni(dcpp)2]2+ even with strongly π-accepting ligands still shows mixing of LF states demonstrating LF splitting is not high enough. To further enhance the π- accepting character, a second sphere coordination is introduced with the Lewis acid Sc[OTf]3. What is seen is that increased ligand π-accepting has a more multifaceted effect on LF splitting and LF states. This is demonstrated by unanticipated shift of the 3T2 state to lower energies and the 1E state to higher energies. Further investigation is required to determine how the π- accepting nature of ligands affect NiII LF states, this will help understand ligand requirements needed in the pursuit of SF luminescence from octahedral NiII complexes. 86 4. Summary and Outlook In recent years, there have been great advances made in the field of photoactive 3d transition metal complexes. However, little research has been devoted to octahedral MnIV and NiII compounds. It is evident from this thesis that the photophysical properties of octahedral MnIV complexes are of great interest fundamentally and for eventual use in applications.[16,21] Herein is the report of the successful preparation of the emissive strongly dual state photooxidative [Mn(dgpy)2]4+ (dgpy = 2,6-diguanidylpyridine) complex. Synthesis, characterization and isolation of [Mn(dgpy) ]n+2 (n = 24) in three oxidative states aids with key understanding as to the importance of ligand design; this represents one of the very few octahedral MnIV complexes supported solely by nitrogen donors.[127] The tridentate dgpy ligand combines accepting (pyridines) and strongly donating (guanidine) moieties capable of offering a strong LF that can stabilise the electron rich high-spin d5 [Mn(dgpy) ]2+2 , a Jahn-Teller distorted high-spin d4 [Mn(dgpy)2]3+, and an electron poor d3 [Mn(dgpy) ]4+2 complex. What is also apparent through CASSCF is that the dgpy ligand gives the desired excited state ligand field state ordering. There is only one other known example of octahedral MnIV emission[174] and so the photophysics of these compounds are not well understood. [Mn(dgpy) 4+2] provides a second example of long lived phosphorescence ( = 1.6 ns) from a mixed 2MC/2LMCT state that is significantly red shifted to 1435 nm,[21,174] highlighting the range at which emission can be achieved with octahedral MnIV complexes. An important finding (aided by DFT studies) is related to the bimolecular reactivity of this compound.[62] [Mn(dgpy) ]4+2 photooxidises very challenging substrates ranging up to benzene and CH3CN; such reactivity demonstrates the strongest practical photooxidative power possessed by a transition metal complex.[60] Even more uniquely [Mn(dgpy) 4+2] displays unconventional dual state reactivity upon NIR irradiation,[2,62] oxidising a lower potential substrate (naphthalene) dynamically from a mixed 2LMCT/2MC state and high potential substrates (benzene) statically from a 4LMCT state via a pre-organised CH3CN (solvent) intermediate. Unusually slow ISC (780 fs) enables this reactivity from the 4LMCT state. These results add significant understanding to manganese transition metal (photo)chemistry, helping further understand the requirements for isolation of octahedral MnIV complexes. What is shown is the range at which emission can be found, the photooxidative potential and the possibility of accessing the full oxidising power of [Mn(dgpy) 4+2] due to static quenching and slow ISC. It also shows how complex design can lead to panchromatic absorption and the possibility of using low energy NIR excitation for such photooxidative applications.[206] With such understanding, it can be concluded that strongly donating ligands are required to stabilise MnIV centres and low energy LMCT absorption bands should be pushed to higher energies to get higher energy emission. Thus, ligand design is paramount and using tridentate ligands containing carbene donors can offer these characteristics. It is clear that the range at 87 which MnIV emission can occur is large, thus more MnIV examples are required to further understand the nephelauxetic effect and CT state admixtures associated with MnIV complexes. The photooxidative capabilities of [Mn(dgpy) 4+2] show how the utilization of both static and dynamic quenching can aid with photoredox reactions. Not much is understood regarding ISC rates of MnIV but it is apparent from this case that ‘slower’ ISC rates are not necessarily detrimental for photoreactivity.[50] Lastly, future studies with such a compound class could focus on photoredox catalysis. To do this it would be important to make complexes such as [Mn(dgpy) ]4+2 water stable, and to reduce the GS oxidation potential. Lowering the GS oxidation potential is important to enhance GS stability and ensure the possibility of catalyst recovery with a stable sacrificial oxidant. This could accomplished with bulky strongly donating ligands, which would both lower the GS potential and reduce complex substitutional lability.[29] The investigation of a series of five NiII octahedral complexes highlights the challenges associated with LF strength in d8 systems. In order to achieve MC SF emission from such compounds, strong LFs are required for MC states to be lowest in energy.[187,189] However, a limit is reached when the d8 NiII system adopts a square planar geometry.[186] The complexes described i.e. [Ni(dgpy) )2+2 , [Ni(terpy) 2+2) (terpy = 2,2';6',2"-terpyridine), [Ni(phen) )2+ 2 (phen = 1,10-phenanthroline), [Ni(ddpd) )2+2 (ddpd = N,N’-dimethyl-N,N’-dipyridin-2-ylpyridine-2,6- diamine) and [Ni(tpe) )2+2 (tpe = 1,1,1-tris(pyrid-2-yl)ethane), each contain medium to strong field ligands that have yielded emission with other metals, e.g. CrIII, CoIII, MnIV.[77,79,202] However, these NiII complexes do not show emission even at high hydrostatic pressures. It is clear from DFT and coupled surface studies that detrimental mixing of intra- and interconfigurational states leads to the non-radiative decay.[188] This mixing can be seen as ‘intensity borrowing’ using absorption spectroscopy, with the spin forbidden states shown as shoulders on the spin allowed absorption bands. Evidently, it is important to separate these states with strong LF, so that mixing cannot occur while also maintaining an octahedral ligand environment. Attempts to further increase ligand field strength via σ-donation by N- heterocyclic carbenes resulted with the square planar [Ni(CNC)(NCN)][PF6]2 (CNC = (1,1’- (pyridin-2,6-diyl)bis(3-methyl-1H-imidazol-3-ylidene)) and (NCN = (1,3-bis(2- pyridyl)imidazolylidene). Attempts to do the same with a more π-accepting ligand gave octahedrally coordinated [Ni(dcpp) 2+ 2] (dcpp = 2,6-bis(2-carboxypyridyl)pyridine), however the ligand field states were again mixed. Coordination of the Lewis acid Sc[OTf]3 shows that a second sphere coordination can influence ligand field states. Coordination showed and unanticipated shift of the 3T2 state to lower energies and 1E state to higher energies. Further investigation is required to determine how the π-accepting nature of ligands affect NiII LF states, and more careful ligand design is required in order to increase ligand field strength and maintain an octahedral ligand field. This could be overcome by using more sterically demanding strong field ligands such as cage ligands to impart a strong field and impose an octahedral coordination.[201] 88 5. References [1] L. Li, J. Lin, N. Wu, S. Xie, C. Meng, Y. Zheng, X. Wang, Y. Zhao, Energy Built Environ. 2022, 3, 139–157. [2] D. M. Arias-Rotondo, J. K. McCusker, Chem. Soc. Rev. 2016, 45, 5803–5820. [3] A. Carella, F. Borbone, R. Centore, Front. Chem. 2018, 6, 481. [4] A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo, H. Pettersson, Chem. Rev. 2010, 110, 6595– 6663. [5] H. Xu, R. Chen, Q. Sun, W. Lai, Q. Su, W. Huang, X. Liu, Chem. Soc. Rev. 2014, 43, 3259– 3302. [6] S. Aghazada, M. Nazeeruddin, Inorganics 2018, 6, 52. [7] J. M. R. Narayanam, C. R. J. Stephenson, Chem. Soc. Rev. 2011, 40, 102–113. [8] C. Förster, K. Heinze, Chem. Soc. Rev. 2020, 49, 1057–1070. [9] O. S. Wenger, J. Am. Chem. Soc. 2018, 140, 13522–13533. [10] N. Haque, A. Hughes, S. Lim, C. Vernon, Resources 2014, 3, 614–635. [11] A.W. Adamson, W.L. Waltz, E. Zinato, D.W. Watts, P.D. Fleischauer, R.D. Lindholm, Chem. Rev. 1968, 68, 541–585. [12] W. Zhang, R. Alonso-Mori, U. Bergmann, C. Bressler, M. Chollet, A. Galler, W. Gawelda, R. G. Hadt, R. W. Hartsock, T. Kroll, K. S. Kjær, K. Kubiček, H. T. Lemke, H. W. Liang, D. A. Meyer, M. M. Nielsen, C. Purser, J. S. Robinson, E. I. Solomon, Z. Sun, D. Sokaras, T. B. van Driel, G. Vankó, T.-C. Weng, D. Zhu, K. J. Gaffney, Nature 2014, 509, 345–348. [13] O. S. Wenger, Chem. Eur. J. 2019, 25, 6043–6052. [14] M. Kaupp, J. Comput. Chem. 2007, 28, 320–325. [15] A. Steffen, B. Hupp, Comprehensive inorganic chemistry III. Vol. 2, (Ed.: J. Reedijk), Elesvier, 2021. [16] C. Wegeberg, O. S. Wenger, JACS Au 2021, 1, 1860–1876. [17] P. Chábera, Y. Liu, O. Prakash, E. Thyrhaug, A. E. Nahhas, A. Honarfar, S. Essén, L. A. Fredin, T. C. B. Harlang, K. S. Kjær, K. Handrup, F. Ericson, H. Tatsuno, K. Morgan, J. Schnadt, L. Häggström, T. Ericsson, A. Sobkowiak, S. Lidin, P. Huang, S. Styring, J. Uhlig, J. Bendix, R. Lomoth, V. Sundström, P. Persson, K. Wärnmark, Nature 2017, 543, 695– 699. [18] W. Leis, M. A. Argüello Cordero, S. Lochbrunner, H. Schubert, A. Berkefeld, J. Am. Chem. Soc. 2022, 144, 1169–1173. [19] Y. Tanabe. S. Sugano, J. Phys. Soc. Jpn. 1954, 9, 753–766. [20] Y. Tanabe. S. Sugano, J. Phys. Soc. Jpn. 1954, 9, 766–799. [21] W. R. Kitzmann, J. Moll, K. Heinze, Photochem. Photobiol. Sci. 2022, 21, 1309-1331. [22] A. D. Kirk, Chem. Rev. 1999, 99, 1607–1640. [23] M. Maestri, F. Bolletta, L. Moggi, V. Balzani, M. S. Henry, and M. Z. Hoffman, J. Am. Chem. Soc. 1978, 100, 2694–2701. [24] Dorn, M. East, N. R. Förster, C. Kitzmann, W. R. Moll, J. Reichenauer, F. Reuter, T. Stein, L. Heinze, K. Comprehensive inorganic chemistry III (Third Edition). Vol. 8, (Ed.: J. Reedijk), Elsevier, 2023. [25] L. Stein, C. Wang, C. Förster, U. Resch-Genger, K. Heinze, Dalton Trans. 2022, 51, 17664–17670. 89 [26] M. Dorn, J. Kalmbach, P. Boden, A. Päpcke, S. Gómez, C. Förster, F. Kuczelinis, L. M. Carrella, L. A. Büldt, N. H. Bings, E. Rentschler, S. Lochbrunner, L. González, M. Gerhards, M. Seitz, K. Heinze, J. Am. Chem. Soc. 2020, 142, 7947–7955. [27] Robert O. C. Hart, Simon G. Bott, Jerry L. Atwood, Stephen R. Cooper, J. Chem. Soc., Chem. Commun. 1992, 21, 894–895. [28] M. K. Chan, W. H. Armstrong, Inorg. Chem. 1989, 28, 3777–3779. [29] B. Wittwer, N. Dickmann, S. Berg, D. Leitner, L. Tesi, D. Hunger, R. Gratzl, J. van Slageren, N. I. Neuman, D. Munz, S. Hohloch, Chem. Commun. 2022, 58, 6096–6099. [30] A. P. Forshaw, R. P. Bontchev, J. M. Smith, Inorg. Chem. 2007, 46, 3792–3794. [31] G. Das, P. K. Bharadwaj, D. Ghosh, B. Chaudhurib, R. Banerjee, Chem. Commun. 2001, 37, 323–324. [32] S. Romain, C. Baffert, C. Duboc, J.-C. Leprêtre, A. Deronzier, M.-N. Collomb, Inorg. Chem. 2009, 48, 3125–3131. [33] H. Yoon, Y. Morimoto, Y.-M. Lee, W. Nam, S. Fukuzumi, Chem. Commun. 2012, 48, 11187–11189. [34] S. I. Shylin, J. L. Pogrebetsky, A. O. Husak, D. Bykov, A. Mokhir, F. Hampel, S. Shova, A. Ozarowski, E. Gumienna-Kontecka, I. O. Fritsky, Chem. Commun. 2021, 57, 11060– 11063. [35] O. S. Wenger, S. Bénard, H. U. Güdel, Inorg. Chem. 2002, 41, 5968–5977. [36] S. Campagna, F. Puntoriero, F. Nastasi, G. Bergamini, V. Balzani, Top. Curr. Chem. 2007, 280, 1–36. [37] P. W. Atkins, J. de Paula, Physical Chemistry. 9th ed., W.H Freeman and Company, New York, 2010. [38] J. R. Lakowicz, Principles of fluorescence spectroscopy. 3rd ed., Springer, New York, 2006. [39] T. J. Penfold, E. Gindensperger, C. Daniel, C. M. Marian, Chem. Rev. 2018, 118, 6975– 7025. [40] R. Englman, J. Jortner, Mol. Phys. 1970, 18, 145–164. [41] W. R. Browne, J. G. Vos, Coord. Chem. Rev. 2001, 219-221, 761–787. [42] J. V. Caspar, E. M. Kober, B. P. Sullivan, T. J. Meyer, J. Am. Chem. Soc 1982, 104, 630– 632. [43] L. A. Büldt, X. Guo, A. Prescimone, O. S. Wenger, Angew. Chem. Int. Ed. 2016, 55, 11247–11250. [44] W. R. Kitzmann, K. Heinze, Angew. Chem. Int. Ed. 2022, e202213207. [45] R. Englman, J. Lumin. 1970, 1, 2, 134–142. [46] A. Beeby, I. M. Clarkson, R. S. Dickins, S. Faulkner, D. Parker, J. Chem. Soc., Perkin Trans. 2 1999, 493–503. [47] C. M. Marian, WIREs Comput. Mol. Sci. 2012, 2, 187–203. [48] J. K. McCusker, Acc. Chem. Res. 2003, 36, 876–887. [49] W. R. Kitzmann, C. Ramanan, R. Naumann, K. Heinze, Dalton Trans. 2022, 51, 6519– 6525. [50] L. S. Forster, Coord. Chem. Rev. 2006, 250, 2023–2033. [51] E. V. Dose, M. A. Hoselton, N. Sutin, M. F. Tweedle, L. J. Wilson, J. Am. Chem. Soc. 1978, 100, 1141–1147. 90 [52] P. J. Gonçalves, L. de Boni, I. E. Borissevitch, S. C. Zílio, J. Phys. Chem. A. 2008, 112, 6522–6526. [53] C. Antolini, C. D. Spellman, C. J. Otolski, G. Doumy, A. M. March, D. A. Walko, C. Liu, X. Zhang, B. T. Young, J. E. Goodwill, D. Hayes, J. Am. Chem. Soc. 2022, 144, 22514–22527. [54] G. B. Shaw, C. D. Grant, H. Shirota, E. W. Castner, G. J. Meyer, L. X. Chen, J. Am. Chem. Soc. 2007, 129, 2147–2160. [55] L. Bergmann, G. J. Hedley, T. Baumann, S. Bräse, I. D. W. Samuel, Sci. Adv. 2016, 2, e1500889. [56] F. Strieth-Kalthoff, M. J. James, M. Teders, L. Pitzer, F. Glorius, Chem. Soc. Rev. 2018, 47, 7190–7202. [57] C. B. Larsen, O. S. Wenger, Chem. Eur. J. 2018, 24, 2039–2058. [58] D. Genovese, M. Cingolani, E. Rampazzo, L. Prodi, N. Zaccheroni, Chem. Soc. Rev. 2021, 50, 8414–8427. [59] V. Balzani, L. Moggi, M. F. Manfrin, F. Bolletta, Coord. Chem. Rev. 1975, 15, 321–433. [60] D. Kim, T. S. Teets, Chem. Phys. Rev. 2022, 3, 21302. [61] N. Sharma, J. Jung, K. Ohkubo, Y.-M. Lee, M. E. El-Khouly, W. Nam, S. Fukuzumi, J. Am. Chem. Soc. 2018, 140, 8405–8409. [62] C. Förster, K. Heinze, Chem. Phys. Rev. 2022, 3, 41302. [63] F. Reichenauer, C. Wang, C. Förster, P. Boden, N. Ugur, R. Báez-Cruz, J. Kalmbach, L. M. Carrella, E. Rentschler, C. Ramanan, G. Niedner-Schatteburg, M. Gerhards, M. Seitz, U. Resch-Genger, K. Heinze, J. Am. Chem. Soc 2021, 143, 11843–11855. [64] S. Otto, A. M. Nauth, E. Ermilov, N. Scholz, A. Friedrich, U. Resch-Genger, S. Lochbrunner, T. Opatz, K. Heinze, ChemPhotoChem 2017, 1, 344–349. [65] R. A. Marcus, Angew. Chem. Int. Ed. Engl. 1993, 32, 1111–1121. [66] E. J. Piechota, G. J. Meyer, J. Chem. Educ. 2019, 96, 2450–2466. [67] R. A. Marcus, N. Sutin, Biochim. Biophys. Acta, Bioenerg. 1985, 811, 265–322. [68] A. Aydogan, R. E. Bangle, A. Cadranel, M. D. Turlington, D. T. Conroy, E. Cauët, M. L. Singleton, G. J. Meyer, R. N. Sampaio, B. Elias, L. Troian-Gautier, J. Am. Chem. Soc 2021, 143, 15661–15673. [69] J. Olmsted, T. J. Meyer, J. Phys. Chem. 1987, 91, 1649–1655. [70] S. V. Rosokha, J. K. Kochi, Acc. Chem. Res. 2008, 41, 641–653. [71] S. Sittel, R. Naumann, K. Heinze, Front. Chem. 2022, 10, 887439. [72] J. P. Barham, B. König, Angew. Chem. Int. Ed. 2020, 59, 11732–11747. [73] S. Wu, J. Kaur, T. A. Karl, X. Tian, J. P. Barham, Angew. Chem. Int. Ed. 2022, 61, e202107811. [74] H. Huang, Z. M. Strater, M. Rauch, J. Shee, T. J. Sisto, C. Nuckolls, T. H. Lambert, Angew. Chem. Int. Ed. 2019, 58, 13318–13322. [75] K. S. Kjær, N. Kaul, O. Prakash, P. Chábera, N. W. Rosemann, A. Honarfar, O. Gordivska, L. A. Fredin, K.-E. Bergquist, L. Häggström, T. Ericsson, L. Lindh, A. Yartsev, S. Styring, P. Huang, J. Uhlig, J. Bendix, D. Strand, V. Sundström, P. Persson, R. Lomoth, K. Wärnmark, Science 2019, 363, 249–253. [76] P. Herr, C. Kerzig, C. B. Larsen, D. Häussinger, O. S. Wenger, Nat. Chem. 2021, 13, 956– 962. 91 [77] S. Treiling, C. Wang, C. Förster, F. Reichenauer, J. Kalmbach, P. Boden, J. P. Harris, L. M. Carrella, E. Rentschler, U. Resch-Genger, C. Reber, M. Seitz, M. Gerhards, K. Heinze, Angew. Chem. Int. Ed. 2019, 58, 18075–18085. [78] S. Otto, M. Grabolle, C. Förster, C. Kreitner, U. Resch-Genger, K. Heinze, Angew. Chem. Int. Ed. 2015, 54, 11572–11576. [79] S. Otto, M. Dorn, C. Förster, M. Bauer, M. Seitz, K. Heinze, Coord. Chem. Rev. 2018, 359, 102–111. [80] Oppenheim, J. Miller, J., Tanabe-Sugano for Mathematica. https://github.com/JulesOpp/Tanabe-Sugano, Accessed 08 May 2023. [81] P. W. Atkins, T. L. Overton, J. P. Rourke, M. T. Weller, F. A. Armstrong, Inorganic Chemistry. 5th ed., W.H Freeman and Company, New York, 2010. [82] N. Sinha, P. Yaltseva, O. S. Wenger, Angew. Chem. Int. Ed. 2023, e202303864. [83] L. Stein, P. Boden, R. Naumann, C. Förster, G. Niedner-Schatteburg, K. Heinze, Chem. Commun. 2022, 58, 3701–3704. [84] N. Sinha, J.-R. Jiménez, B. Pfund, A. Prescimone, C. Piguet, O. S. Wenger, Angew. Chem. Int. Ed. 2021, 60, 23722–23728. [85] M. G. Brik, A. M. Srivastava, ECS J. Solid State Sci. Technol. 2018, 7, R3079-R3085. [86] S. Otto, N. Scholz, T. Behnke, U. Resch-Genger, K. Heinze, Chem. Eur. J. 2017, 23, 12131–12135. [87] S. Otto, J. P. Harris, K. Heinze, C. Reber, Angew. Chem. Int. Ed. 2018, 57, 11069–11073. [88] H. Xiang, J. Cheng, X. Ma, X. Zhou, J. J. Chruma, Chem. Soc. Rev. 2013, 42, 6128–6185. [89] T. M. Stonelake, K. A. Phillips, H. Y. Otaif, Z. C. Edwardson, P. N. Horton, S. J. Coles, J. M. Beames, S. J. A. Pope, Inorg. Chem. 2020, 59, 2266–2277. [90] J. A. Treadway, G. F. Strouse, R. R. Ruminski, T. J. Meyer, Inorg. Chem. 2001, 40, 4508– 4509. [91] A. K. Pal, S. Nag, J. G. Ferreira, V. Brochery, G. La Ganga, A. Santoro, S. Serroni, S. Campagna, G. S. Hanan, Inorg. Chem. 2014, 53, 1679–1689. [92] M. J. Cook, A. P. Lewis, G. S. G. McAuliffe, V. Skarda, J. Chem. Soc., Perkin Trans. 2 1984. [93] D. B. Nemez, I. B. Lozada, J. D. Braun, J. A. G. Williams, D. E. Herbert, Inorg. Chem. 2022, 61, 13386–13398. [94] J. T. Yarranton, J. K. McCusker, J. Am. Chem. Soc. 2022, 144, 12488–12500. [95] D. W. Thompson, A. Ito, T. J. Meyer, Pure Appl. Chem. 2013, 85, 1257–1305. [96] T. J. Meyer, Pure Appl. Chem. 1986, 58, 1193–1206. [97] P. Chen, T. J. Meyer, Chem. Rev. 1998, 98, 1439–1478. [98] J. M. Lang, Z. A. Dreger, H. G. Drickamer, J. Phys. Chem. 1993, 97, 2289–2294. [99] E. Krausz. J. Ferguson, Progress in Inorganic Chemistry. Vol. 37, Wiley, Chichester, 1989. [100] P. S. Wagenknecht, P. C. Ford, Coord. Chem. Rev. 2011, 255, 591–616. [101] P. Pyykko, Chem. Rev. 1988, 88, 563–594. [102] J. K. McCusker, Science 2019, 363, 484–488. [103] P. Chábera, L. Lindh, N. W. Rosemann, O. Prakash, J. Uhlig, A. Yartsev, K. Wärnmark, V. Sundström, P. Persson, Coord. Chem. Rev. 2021, 426, 213517. [104] P. S. Wagenknecht, P. C. Ford, Coord. Chem. Rev. 2011, 255, 591–616. 92 [105] T. Reuter, A. Kruse, R. Schoch, S. Lochbrunner, M. Bauer, K. Heinze, Chem. Commun. 2021, 57, 7541–7544. [106] J. A. Treadway, B. Loeb, R. Lopez, P. A. Anderson, F. R. Keene, T. J. Meyer, Inorg. Chem. 1996, 35, 2242–2246. [107] C. Wang, S. Otto, M. Dorn, E. Kreidt, J. Lebon, L. Sršan, P. Di Martino-Fumo, M. Gerhards, U. Resch-Genger, M. Seitz, K. Heinze, Angew. Chem. Int. Ed. 2018, 57, 1112– 1116. [108] S. Otto, C. Förster, C. Wang, U. Resch-Genger, K. Heinze, Inorg. Chem. 2018, 24, 12555–12563. [109] S. F. McClanahan, J. R. Kincaid, J. Am. Chem. Soc 1986, 108, 3840–3841. [110] R. G. Lucchini, M, Aschner, Y, Kim, M. Šarić, Handbook on the Toxicology of Metals. 4th ed., Elsevier, London, 2015. [111] R. L. M. Baric, Handbook on the toxicology of metals. 3rd ed., Elsevier, London, 2007. [112] J-B. Sortais, Manganese Catalysis in Organic Synthesis, Wiley-VCH Weinheim, Germany, 2022. [113] D. S. Avila, R. L. Puntel, M. Aschner, Met. Ions life Sci. 2013, 13, 199–227. [114] N. Cox, M. Retegan, F. Nesse, D. A. Pantazis, A. Boussac, W. Lubitz, Science 2014, 345, 804–808. [115] C. Herrero, A. Quaranta, W. Leibl, A. W. Rutherford, A. Aukauloo, Energy Environ. Sci. 2011, 4, 2353. [116] J. C. Vites, M. M. Lynam, Coord. Chem. Rev. 1995, 146, 167–189. [117] W. M Haynes, D. R. Lide, T. J Bruno, CRC Handbook of Chemistry and Physics 95th Edition, CRC Press, Boca Raton, Florida, 2014. [118] L. Hung, C. Chen, Mater. Sci. Eng., R 2002, 39, 143–222. [119] M. H. Du, J. Mater. Chem. C 2014, 2, 2475–2481. [120] M. D. Dramićanin, Ł. Marciniak, S. Kuzman, W. Piotrowski, Z. Ristić, J. Periša, I. Evans, J. Mitrić, V. Đorđević, N. Romčević, M. G. Brik, C.-G. Ma, Light Sci. Appl. 2022, 11, 279. [121] Y. Qin, P. She, X. Huang, W. Huang, Q. Zhao, Coord. Chem. Rev. 2020, 416, 213331. [122] N. G. Connelly, W. E. Geiger, Chem. Rev. 1996, 96, 877–910. [123] G. W. Brudvig. R. H. Crabtree, Progress in Inorganic Chemistry. Vol. 37, Wiley, Chichester, 1989. [124] D. M. Eichhorn, W. H. Armstrong, Inorg. Chem. 1990, 29, 3607–3612. [125] P. Cieslik, P. Comba, B. Dittmar, D. Ndiaye, É. Tóth, G. Velmurugan, H. Wadepohl, Angew. Chem. Int. Ed. 2022, 61, e202115580. [126] S. Wang, T. D. Westmoreland, Inorg. Chem. 2009, 48, 719–727. [127] A. D. M.-N. Collomb, Encyclopedia of Inorganic Chemistry. 1st Ed., Wiley, Chichester, 2006. [128] F.A. Cotton, G. Wilkinson, P.L. Gaus, Advanced Inorganic Chemistry. 2nd Ed., John Wiley & Sons, Toronto, 1987. [129] P. Tao, S.-J. Liu, W.-Y. Wong, Adv. Optical Mater. 2020, 8, 2000985. [130] S. Zhang, Y. Zhao, J. Zhou, H. Ming, C.-H. Wang, X. Jing, S. Ye, Q. Zhang, Chemical Engineering Journal 2021, 421, 129886. [131] J. J. Alexander, H. B. Gray, J. Am. Chem. Soc. 1968, 90, 4260–4271. [132] I. Syiemlieh, A. Kumar, S. D. Kurbah, A. K. De, R. A. Lal, J. Mol. Struct. 2018, 1151, 343– 352. 93 [133] A. Saha, P. Majumdar, S. Goswami, J. Chem. Soc., Dalton Trans. 2000, 1703–1708. [134] C. P. Basu, Inorg. Chem. 1992, 31, 4980–4986. [135] R. A. Lal, D. Basumatary, A. K. De, A. Kumar, Transit. Met. Chem. 2007, 32, 481–493. [136] D. Basumatary, R. A. Lal, A. Kumar, J. Mol. Struct. 2015, 1092, 122–129. [137] M. Bortoluzzi, J. Castro, A. Gobbo, V. Ferraro, L. Pietrobon, S. Antoniutti, New J. Chem. 2020, 44, 571–579. [138] V. Morad, I. Cherniukh, L. Pöttschacher, Y. Shynkarenko, S. Yakunin, M. V. Kovalenko, Chem. Mater. 2019, 31, 10161–10169. [139] J. Chen, Q. Zhang, F.-K. Zheng, Z.-F. Liu, S.-H. Wang, A.-Q. Wu, G.-C. Guo, Dalton Trans. 2015, 44, 3289–3294. [140] Y. Suffren, N. O'Toole, A. Hauser, E. Jeanneau, A. Brioude, C. Desroches, Dalton Trans. 2015, 44, 7991–8000. [141] L.-J. Xu, C.-Z. Sun, H. Xiao, Y. Wu, Z.-N. Chen, Adv. Mater. 2017, 29. [142] D. G. M. Wrighton, Chem. Phys. 1974, 295–299. [143] A. S. Berezin, K. A. Vinogradova, V. A. Nadolinny, T. S. Sukhikh, V. P. Krivopalov, E. B. Nikolaenkova, M. B. Bushuev, Dalton Trans. 2018, 47, 1657–1665. [144] G. E. Hardy, J. I. Zink, Inorg. Chem. 1976, 15, 3061–3065. [145] J. Steube, A. Kruse, O. S. Bokareva, T. Reuter, S. Demeshko, R. Schoch, M. A. Argüello Cordero, A. Krishna, S. Hohloch, F. Meyer, K. Heinze, O. Kühn, S. Lochbrunner, M. Bauer, Nature chemistry 2023, 15, 468–474. [146] N. Elgrishi, K. J. Rountree, B. D. McCarthy, E. S. Rountree, T. T. Eisenhart, J. L. Dempsey, J. Chem. Educ. 2018, 95, 197–206. [147] A. Saju, J. R. Griffiths, S. N. MacMillan, D. C. Lacy, J. Am. Chem. Soc. 2022, 144, 16761– 16766. [148] A. W. Adamson, W. L. Waltz, E. Zinato, D. W. Watts, P. D. Fleischauer, R. D. Lindholm, Chem. Rev. 1968, 68, 541–585. [149] M. Drosou, G. Zahariou, D. A. Pantazis, Angew. Chem. Int. Ed. 2021, 60, 13493–13499. [150] H. Aghabozorg, G. J. Palenik, R. C. Stoufer, J. Summers, Inorg. Chem. 1982, 21, 3903– 3907. [151] R. J. F. Berger, G. V. Girichev, N. I. Giricheva, A. A. Petrova, N. V. Tverdova, Angew. Chem. Int. Ed. 2017, 56, 15751–15754. [152] C. Mantel, A. K Hassan, J. Pecaut, A. Deronzier, M-N. Collomb, C. Duboc-Toia, J. Am. Chem. Soc. 2003, 125, 12337–12344. [153] G. A. Craig, M. Murrie, Chem. Soc. Rev. 2015, 44, 2135–2147. [154] J. Vallejo, A. Pascual-Álvarez, J. Cano, I. Castro, M. Julve, F. Lloret, J. Krzystek, G. de Munno, D. Armentano, W. Wernsdorfer, R. Ruiz-García, E. Pardo, Angew. Chem. Int. Ed. 2013, 52, 14075–14079. [155] J. M. Frost, K. L. M. Harriman, M. Murugesu, Chem. Sci. 2016, 7, 2470–2491. [156] Y. Garcia, P. Gütlich, Top. Curr. Chem. 2004, 234, 49–62. [157] P. Greig Sim and Ekk Sinn. [158] A. P. Forshaw, J. M. Smith, A. Ozarowski, J. Krzystek, D. Smirnov, S. A. Zvyagin, T. D. Harris, H. I. Karunadasa, J. M. Zadrozny, A. Schnegg, K. Holldack, T. A. Jackson, A. Alamiri, D. M. Barnes, J. Telser, Inorg. Chem. 2013, 52, 144–159. [159] F. Hossain, M. A. Rigsby, C. T. Duncan, P. L. Milligan, R. L. Lord, M.-H. Baik, F. A. Schultz, Inorg. Chem. 2007, 46, 2596–2603. 94 [160] A. A. Alaimo, D. Takahashi, L. Cunha-Silva, G. Christou, T. C. Stamatatos, Inorg. Chem. 2015, 54, 2137–2151. [161] A. Majumder, S. Goswami, S. R. Batten, M. Salah El Fallah, J. Ribas, S. Mitra, Inorg. Chim. Acta 2006, 359, 2375–2382. [162] P. Pavacik, J. C. Huffman, G. Christou, J. Chem. Soc., Chem. Commun. 1986, 1, 43–44. [163] T. Weyhermüller, T. K Paine, E. Bothe, E. Bill, P. Chaudhuri, Inorg. Chim. Acta 2002, 337, 344–356. [164] S. K. Chandra, P. Basu, D. Ray, S. Pal, A. Chakravorty, Inorg. Chem. 1990, 29, 2423– 2428. [165] D. F. Back, G. Manzoni de Oliveira, C. Mensch Canabarro, B. Almeida Iglesias, Z. Anorg. Allg. Chem. 2015, 641, 941–947. [166] N. Sarkar, K. Harms, A. Bauzá, A. Frontera, S. Chattopadhyay, ChemistrySelect 2017, 2, 2975–2984. [167] Michael K. Chan, William H. Armstrong, Inorg. Chem. 1989, 3777–3779. [168] Y. Xu, C. Li, X. Wu, M.-X. Li, Y. Ma, H. Yang, Q. Zeng, J. L. Sessler, Z.-X. Wang, J. Am. Chem. Soc. 2022, 144, 18834–18843. [169] M. G. Brik, A. M. Srivastava, J. Lumin. 2013, 133, 69–72. [170] Y. Li, S. Qi, P. Li, Z. Wang, RSC Adv. 2017, 7, 38318–38334. [171] Q. Zhou, L. Dolgov, A. M. Srivastava, L. Zhou, Z. Wang, J. Shi, M. D. Dramićanin, M. G. Brik, M. Wu, J. Mater. Chem. C 2018, 6, 2652–2671. [172] S. Adachi, Journal of Luminescence 2018, 202, 263–281. [173] M. G. Brik, S. J. Camardello, A. M. Srivastava, N. M. Avram, A. Suchocki, ECS J. Solid State Sci. Technol. 2016, 5, R3067-R3077. [174] V. Baslon, J. P. Harris, C. Reber, H. E. Colmer, T. A. Jackson, A. P. Forshaw, J. M. Smith, R. A. Kinney, J. Telser, Can. J. Chem. 2017, 95, 547–552. [175] J. R. Mayfield, E. N. Grotemeyer, T. A. Jackson, Chem. Commun. 2020, 56, 9238–9255. [176] N. Sharma, Y.-M. Lee, W. Nam, S. Fukuzumi, Isr. J. Chem. 2020, 60, 1049–1056. [177] H. Yoon, Y.-M. Lee, X. Wu, K.-B. Cho, R. Sarangi, W. Nam, S. Fukuzumi, J. Am. Chem. Soc 2013, 135, 9186–9194. [178] C. Saracini, S. Fukuzumi, Y.-M. Lee, W. Nam, Dalton Trans. 2018, 47, 16019–16026. [179] N. Sharma, J. Jung, Y.-M. Lee, M. S. Seo, W. Nam, S. Fukuzumi, Chem. Eur. J. 2017, 23, 7125–7131. [180] N. Sharma, Y.-M. Lee, X.-X. Li, W. Nam, S. Fukuzumi, Inorg. Chem. 2019, 58, 14299– 14303. [181] D. G. Karmalkar, M. S. Seo, Y.-M. Lee, Y. Kim, E. Lee, R. Sarangi, S. Fukuzumi, W. Nam, Inorg. Chem. 2021, 60, 16996–17007. [182] J. Chen, H. Yoon, Y.-M. Lee, M. S. Seo, R. Sarangi, S. Fukuzumi, W. Nam, Chem. Sci. 2015, 6, 3624–3632. [183] L. A. Büldt, C. B. Larsen, O. S. Wenger, Chem. Eur. J. 2017, 23, 8577–8580. [184] L. A. Büldt, O. S. Wenger, Dalton Trans. 2017, 46, 15175–15177. [185] Y.-S. Wong, M.-C. Tang, M. Ng, V. W.-W. Yam, J. Am. Chem. Soc 2020, 142, 7638–7646. [186] T. Ogawa, N. Sinha, B. Pfund, A. Prescimone, O. S. Wenger, J. Am. Chem. Soc. 2022, 144, 21948–21960. [187] E. González, A. Rodrigue-Witchel, C. Reber, Coord. Chem. Rev. 2007, 251, 351–363. 95 [188] M-C. Nolet, A. Michaud, C. Bain, D. Zargarian, C. Reber, Photochem. Photobiol. 2006, 82, 57–63. [189] C. R. G. Bussiere, J. Am. Chem. Soc. 1998, 120, 6306–6315. [190] M. Dorn, K. Mack, L. M. Carrella, E. Rentschler, C. Förster, K. Heinze, Z. Anorg. Allg. Chem. 2018, 644, 706–712. [191] D. A. V. Griend, D. K. Bediako, M. J. DeVries, N. A. DeJong, L. P. Heeringa, Inorg. Chem. 2008, 47, 656–662. [192] M.-C. Nolet, R. Beaulac, A.-M. Boulanger, C. Reber, Struct. Bond. 2004, 107, 145–158. [193] R. W. Parsons, H. G. Drickamer, J. Chem. Phys. 1958, 29, 930–937. [194] J. C. Zahner, H. G. Drickamer, J. Chem. Phys. 1961, 35, 1483–1490. [195] S. Poirier, H. Lynn, C. Reber, E. Tailleur, M. Marchivie, P. Guionneau, M. R. Probert, Inorg. Chem. 2018, 57, 7713–7723. [196] A. Mrutu, K. I. Goldberg, R. A. Kemp, Inorg. Chim. Acta 2010, 364, 115–119. [197] X. Zhang, B. Liu, A. Liu, W. Xie, W. Chen, Organometallics 2009, 28, 1336–1349. [198] S. Gu, W. Chen, Organometallics 2009, 28, 909–914. [199] S. Gu, J. Du, J. Huang, Y. Guo, L. Yang, W. Xu, W. Chen, Dalton Trans. 2017, 46, 586– 594. [200] S. Gu, D. Xu, W. Chen, Dalton Trans. 2011, 40, 1576–1583. [201] N. Takeda, D. Shimizu, N. Tokitoh, Inorg. Chem. 2005, 44, 8561–8568. [202] A. K. Pal, C. Li, G. S. Hanan, E. Zysman-Colman, Angew. Chem. Int. Ed. 2018, 57, 8027– 8031. [203] K. Inamoto, J-I. Kuroda, K. Hiroya, Y. Noda, T. S. M. Watanabe, Organometallics 2006, 25, 3095–3098. [204] J. C. C. Chen, I. J. B. Lin, Organometallics 2000, 19, 5113–5121. [205] C. R. Goldsmith, T. D. P. Stack, Inorg. Chem. 2006, 45, 6048–6055. [206] T. Jiang, N. F. Polizzi, J. Rawson, M. J. Therien, J. Am. Chem. Soc 2017, 139, 8412–8415. [207] X-Area;, STOE & Cie GmbH, Darmstadt, Germany. [208] R. H. Blessing, Acta Crystallogr., Sect. A: Found. Crystallogr. 1995, 51, 33–38. [209] A. L. Spek, Acta Crystallogr., Sect. D: Biol. Crystallogr. 2009, 65, 148–155. [210] G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr. 2015, 71, 3–8. [211] G. M. Sheldrick, Acta Crystallogr., Sect. C: Struct. Chem 2015, 71, 3–8. [212] G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr. 2008, 64, 112–122. [213] C. B. Hübschle, G. M. Sheldrick, B. Dittrich, J. Appl. Crystallogr. 2011, 44, 1281–1284. [214] F. Neese, WIREs Comput. Mol. Sci. 2012, 2, 73–78. [215] A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652. [216] B. Miehlich, A. Savin, H. Stoll, H. Preuss, Chem. Phys. Lett. 1989, 157, 200–206. [217] C. Lee, W. Yang, and R. G. Parr, Phys. Rev. B. 1988, 37, 785–789. [218] R. Izsák, F. Neese, J. Chem. Phys. 2011, 135, 144105. [219] F. Neese, F. Wennmohs, A. Hansen, U. Becker, Chem. Phys. 2009, 356, 98–109. [220] D. A. Pantazis, X.-Y. Chen, C. R. Landis, F. Neese, J. Chem. Theory Comput. 2008, 4, 908– 919. [221] S. Miertus, E. Scrocco, J. Tomasi, Chem. Phys. 1981, 55, 117–129. [222] F. Weigend, Phys. Chem. Chem. Phys. 2006, 8, 1057–1065. [223] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297–3305. 96 6. Appendix 6.1 Supporting Information to Chapter 3.1. (“The Full d3– d5 Redox Series of Mononuclear Manganese Complexes: Geometries and Electronic Structures of [Mn(dgpy) ]n+2 )” 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 6.2 Supporting Information to Chapter 3.2. (“Oxidative Two- State Photoreactivity of a Manganese(IV) Complex using NIR Light”) 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 6.3 Supporting Information to Chapter 3.3. (“Coupled Potential Energy Surfaces Strongly Impact the Lowest- Energy Spin-Flip Transition in Six-Coordinate Nickel(II) Complexes”) 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 6.4 Supporting Information to Chapter 3.4. (“Influencing Ligand field states of Nickel(II) Complexes with Strongly σ- Donating and π-Accepting Ligands”) Experimental section General procedures. All reactions were performed under an argon atmosphere. The literature known [NiBr(CNC)]Br precursor[203] was received from TCI chemicals. NaOAc, dry DMSO (>97%), Ni[BF4]2٠6H n2O and [ BuN4][PF6] were received from Sigma-Aldrich. The H-NCN[204] pro-ligand and dcpp[205] ligand were prepared according to literature known procedures. Solvents CH3CN and Et2O were received from Alfa Aesar and distilled from calcium hydride and sodium respectively. The reagents were used as received from commercial suppliers. Gloveboxes (UniLab/Mbraun Ar 4.8, O2 <0.1 ppm) were used to store and weigh sensitive compounds for synthesis as well as to prepare any measurement sample that required the absence of oxygen and water. Intensity data for crystal structure determinations were collected with a STOE IPDS-2T diffractometer from STOE & CIE GmbH and an Oxford cooling system and corrected for absorption and other effects using Mo Kα radiation (λ = 0.71073 Å). The diffraction frames were integrated using the STOE X-Area[207] package, and most were corrected for absorption with MULABS[208] of the PLATON software package.[209] The structures were solved with SHELXT[210] refined by the full-matrix method based on F2 using SHELXL[211] of the SHELX[212] software package and the ShelXle[213] graphical interface. 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. UV/Vis/NIR spectra were recorded on a Jasco V-770 spectrometer using 1.0 cm cells (Hellma, Suprasil). Electrospray ionization mass spectra were recorded on an Agilent 6545 QTOF-MS spectrometer. Elemental analyses were performed by the microanalytical laboratory of the Department of Chemistry of the University of Mainz using an Elementar vario EL Cube Density functional theory calculations on the nickel(II) complex cations ([Ni(CNC)(NCN)]2+ and [Ni(dcpp) ]2+2 ) were carried out using the ORCA program package (version 4.2.1).[214] Tight convergence criteria were chosen for all calculations (keywords tightscf and tightopt). All calculations were performed using the B3LYP functional[215–217] employing the RIJCOSX approximation.[218,219] Relativistic effects were calculated at the zeroth order regular approximation (ZORA) level.[220] The ZORA keyword automatically invokes relativistically adjusted basis sets. To account for solvent effects, a conductor-like screening model (CPCM) modeling acetonitrile was used in all calculations.[221] Geometry optimizations were 192 performed using Ahlrichs’ polarized valence triple-ζ basis set (def2-TZVPP).[222,223] Atom- pairwise dispersion correction was performed with the Becke-Johnson damping scheme (D3BJ). The energy of the electronic states and presence of energy minima were checked by numerical frequency calculations. Explicit counterions and/or solvent molecules were not taken into account. Synthesis of [Ni(CNC)(NCN)][PF6]. A solution of 52 mg (0.627 mmol, 3.0 eq) of NaOAc in dry DMSO (2 mL) was added to a solution of 96 mg (0.209 mmol, 1.0 eq) [NiBr(CNC)]Br[203] and 100 mg (0.272 mmol, 1.3 eq) [H-NCN][PF6] in dry DMSO (4 mL). The solution turned light orange and was stirred and heated to 100°C for 24 h. The DMSO was removed at 80°C under reduced pressure, leaving an orange powder, which was purified by crystallization via slow diffusion of dry diethyl ether into a concentrated dry acetonitrile solution to yield 155 mg (0.191 mmol, 91%) of [Ni(CNC)(NCN)][PF6]2 as orange crystals. Mass spectrometry confirmed composition (ESI+, CH3CN): m/z (%) = 259.57 (37, [Ni(CNC)(NCN)]2+), 664.10 (91, [Ni(CNC)(NCN)]2+) + PF ]+6 ). For single-crystal X-ray diffraction (XRD) analysis, a solution of 334 mg of [nBu4N][PF6] (0.586 mmol, 5 eq) in dry CH3CN (4 mL) was added to 140 mg of [Ni(CNC)(NCN)][PF6]2. Crystallization via slow diffusion of dry diethyl ether into this solution gave orange crystals of [Ni(CNC)(NCN)][PF6]2, suitable for single-crystal XRD analysis. The PF6 salt crystallizes in the P21/c space group. Synthesis of [Ni(dcpp)2][PF6]2. A solution of 100 mg (0.345 mmol, 2.0 eq) of dcpp in CH3CN (1.5 mL) was added to a solution of 59 mg (0.173 mmol, 1.0 eq) Ni[BF4]2۰6H2O in CH3CN (1.5 mL). The solution turned dark orange and was stirred at room temperature for 30 mins. The complex was precipitated from solution with Et2O (7 ml) and the solution was filtered, leaving an orange powder, which was purified by crystallization via slow diffusion of diethyl ether into a concentrated dry acetonitrile solution to yield 129 mg (0.159 mmol, 92%) of [Ni(dcpp)2][BF4]2 as orange crystals. Elemental analysis and mass spectrometry confirmed the composition and purity. Elem. anal. calcd. (%) for C34H22NiF12NO4P2 (927.21): C 44.04 H 2.39, N 9.06; found C 44.35, H 2.08 N 9.35. MS (ESI+, CH3CN): m/z (%) = 318.05 (14, [Ni(dcpp)2]2+), 655.10 (76, [Ni(dcpp)2 + F]+). For single-crystal X-ray diffraction (XRD) analysis and further characterization, a solution of 239 mg of [nBu4N][PF6] (0.616 mmol, 5 eq) in CH3CN (3 mL) was added to 100 mg of [Ni(dcpp)2][BF4]2. Crystallization via slow diffusion of diethyl ether into this solution gave orange crystals of [Ni(dcpp)2][PF6]2, suitable for single-crystal XRD analysis. The PF6 salt crystallizes in the Pbcn space group. 193 Structural Characterization Table 6.1: Crystallographic data structure refinement for [Ni(CNC)(NCN)][PF6]2. Empirical formula C26 H23 F12 N9 Ni P2 Formula weight 810.18 Temperature 120(2) K Wavelength 0.71073 Å Crystal system Monoclinic Space group P21/c Unit cell dimensions a = 17.1436(9) Å a= 90° b = 17.7956(6) Å b= 99.907(4)° c = 20.2978(11) Å g = 90° Volume 6100.1(5) Å3 Z 8 Density (calculated) 1.764 Mg/m3 Absorption coefficient 0.852 mm-1 F(000) 3264 Crystal size 0.200 x 0.143 x 0.050 mm3 Theta range for data collection 2.464 to 30.964°. Index ranges -24<=h<=23, -24<=k<=25, -27<=l<=29 Reflections collected 43127 Independent reflections 17138 [R(int) = 0.0758] Completeness to theta = 25.242° 99.4 % Absorption correction Semi-empirical from equivalents Max. and min. transmission 0.8519 and 0.3829 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 17138 / 414 / 1033 Goodness-of-fit on F2 0.899 Final R indices [ > 2sigma()] R1 = 0.0728, wR2 = 0.1764 R indices (all data) R1 = 0.1178, wR2 = 0.1928 Extinction coefficient n/a Largest diff. peak and hole 1.446 and -1.334 e.Å-3 194 Table 6.2: Crystallographic data structure refinement for [Ni(dcpp)2][PF6]2. Empirical formula C34 H22 F12 N6 Ni O4 P2 Formula weight 927.22 Temperature 120(2) K Wavelength 0.71073 Å Crystal system Orthorhombic Space group Pbcn Unit cell dimensions a = 14.5984(6) Å a= 90° b = 15.6099(5) Å b= 90° c = 14.9803(4) Å g = 90° Volume 3413.7(2) Å3 Z 4 Density (calculated) 1.804 Mg/m3 Absorption coefficient 0.780 mm-1 F(000) 1864 Crystal size 0.750 x 0.683 x 0.640 mm3 Theta range for data collection 2.610 to 31.044°. Index ranges -19<=h<=20, -21<=k<=21, -21<=l<=16 Reflections collected 18065 Independent reflections 4901 [R(int) = 0.0247] Completeness to theta = 25.242° 99.8 % Absorption correction Semi-empirical from equivalents Max. and min. transmission 0.2168 and 0.1860 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 4901 / 0 / 267 Goodness-of-fit on F2 1.022 Final R indices [ > 2sigma()] R1 = 0.0319, wR2 = 0.0883 R indices (all data) R1 = 0.0377, wR2 = 0.0901 Extinction coefficient n/a Largest diff. peak and hole 0.462 and -0.622 e.Å-3 195 Mass spectra Figure 6.1: ESI+ mass spectrum of [Ni(CNC)(NCN)][PF6]2 in acetonitrile. Figure 6.2: ESI+ mass spectrum of [Ni(dcpp)2][PF6]2 in acetonitrile. 196 7. Acknowledgements I would like to thank Prof. Dr. Katja Heinze for giving me the opportunity to work in her research group. I am grateful for the freedom I had to look at many interesting ideas. In particular I am grateful for her support in all matters scientific and for her patience. Our many discussions helped me work through countless challenges. I am thankful to Christoph Förster, for many helpful discussions and resolving of the many complicated crystal structures, and for his brilliant taste in music. I am sincerely grateful to Robert Naumann for helping me a lot with manganese luminescence and photochemistry. Robert helped me come to terms with endless artifact finding. His enthusiasm for chemistry is infectious. A special thank you to my collaboration partners Chahinez Dab, Prof. Dr Gregor Diezemann and Charusheela Ramanan their help and work made this thesis possible. I want to especially thank Prof. Dr. Christian Reber for his work on our NiII paper and his general helpfulness. I’m thankful for the hard work that my undergraduate student Nico Blum did, he was genuinely interested in photochemistry which made working together great. I would like to thank the Graduate School of Excellence (MAINZ) and Max Planck Graduate School (MPGC) for their financial and technical support. I’m am particularly grateful for the help given when I moved over here from the UK with my family. I am grateful to have had such help during my PhD through such generous support to attend conferences and purchase materials to further my research. I would like to give a special thanks to the Heinze group members past and present. It is difficult to put into words the support each of you gave. There were some tough times, I’m grateful for taking the time to listen and help with challenges related to work or otherwise (a special thanks is due to Winald, Thomas, Lukas, Steven and Matthias for helping without thought of return). There were many great times, and times I laughed a lot, the meme making was truly first rate, the dad jokes not so much. Thanks for the help with the glovebox (Matthias, Steven and Alex). Best wishes to Sandra with manganese. A big thank you to those who took the time to proof read my thesis (Steven, Winald, Laura, Lukas, Phillip and Holly). I am grateful to my parents and my siblings for their love and support though this PhD time, for being there on the other end of the phone when I needed a chat. I am very thankful for all of the help and support from my parent’s in-law during this time. I am grateful to my dearest wife Janine, she has been a pillar of support, I am thankful for her love and kindness. My son Isaac has been such a blessing to have during this time. I am thankful for him. I am also deeply thankful for the divine support I often received. Thank You! 197 8. Curriculum Vitae 198 199 200