University of Southern California Dissertations and Theses

Molecular modulation to fine-tune optoelectronic properties of OLED materials

(USC Thesis Other)

Molecular modulation to fine-tune optoelectronic properties of OLED materials

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Molecular modulation to fine-tune
optoelectronic properties of OLED materials
Jonas Schaab
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(CHEMISTRY)
May 2024
Copyright 2024 Jonas Schaab

ii
In Memory of Andreas Schaab

iii
This dissertation is dedicated to:
Astrid und Andreas Schaab
Julia Noder-Schaab

iv
Acknowledgements
This work would have not been possible without the help of many. I would like to acknowledge
Karl O. Christe, for making my admission at USC possible and guiding me successfully through my first
1.5 years of graduate school.
Mark E. Thompson, thank you for your guidance and for introducing me to the field of OLEDs. I enjoyed
all the papers and publications we wrote together and all the bigger and smaller projects we accomplished.
I want to thank Aiichiro Nakano for serving on my committee for my screening, quals, and my PhD thesis.
Peter Djurovich, thank you for all the counseling and all the late-night discussions, which helped me and
my projects to move forward. Judy, thank you for all your time for all my special requests and orders and
all the sweets you brought. To my great collaborators and friends Collin Muniz, Jie Ma, and James
Fortwangler: Thank you for going with me through rough and great times and sharing the heavy workloads
as well as the Eureka! moments. The rougher times in the lab were made a lot brighter and easier by our
corner crew. Thanks for all the fun talks, the gossip, the bullshit, and the serious discussions. Thanks,
Marsel, Gemma, Allen, Kelly, and Junru. Thank you, Konstantin, for all the discussions inside and outside
the lab. Thank you to my hood mates Jie and Frances, that you could deal with all my chaos and my many
parallel reactions.
This work was not only made possible by people inside the lab but also by people outside the lab. Thank
you to all who supported me: Without your endless patience, Julia, this PhD would have not been possible.
Your openness to live in a new country and work remotely with such a huge time difference was inevitable.
I want to thank you, that you supported me throughout our many years together and that you also accepted
my long and late working days. Thank you for being such a good partner and friend.
Mama, Papa ich bin euch beiden sehr dankbar, das ihr mich immer auf meinem Weg begleitet habt und ihr
mich unterstützt habt, den richtigen Weg zu finden. Danke auch an meine Geschwister Daniel und Helena
für eure Hilfe und das Übernehmen aller möglichen Dinge in Deutschand, die ich nicht von den USA aus
erledigen kann. Oma und Opa, vielen Dank das ihr mich so unterstützt, auch wenn ich weit weg bin und
euch nur selten sehen kann. Petra, Christiane, Johannes, Sabine und Wolfgang danke euch für die
Unterstützung durch mein Leben, und ganz besonderen Dank für die tolle Unterstützung unserer Familie
nach Papa’s Tod.
Julia Noder, vielen Dank fuer deine Beistand über die Jahre, und deine Unterstütyung von Julia und mir,
wenn wir mal wieder 9000km entfernt voneinander sein mussten. Wanja und Katün, vielen Dank auch für
eure Untestutzung von Julia und mir über all die Jahre weit weg von Zuhause.
Benny, Say, Kathi, Katja, Manu, Mathias und Niklas: Danke für die vielen tollen Telefonate und die gute
Freundschaft, die auch der Entferung und Zeitverschiebung standhält.
For all the fun outside the lab, all the weekend trips, parties, and fun events: Christina, Carlos, Wendy, Ryan,
Jing, Frank, Alex, Aileen, and Amanda, I am thankful to have all of you in my life and am grateful for all
the fun time we had together. Especially since we could all get through the COVID-19 pandemic together
and made the best out of the lockdown time.

v
Table of Contents
Acknowledgements........................................................................................................................ iv
List of Tables ............................................................................................................................... viii
List of Figures................................................................................................................................ ix
Abbreviations............................................................................................................................... xiii
Abstract........................................................................................................................................ xiv
Chapter 1 – 2-coordinate, monovalent copper complexes as chromophores and luminophores.... 1
Introduction................................................................................................................................. 1
Achieving high photoluminescent efficiency for Cu(I) complexes............................................. 2
Two-coordinate complexes......................................................................................................... 4
(carbene)M(amide) (cMa) complexes ........................................................................................ 5
Tuning frontier orbitals on ligands in cMa complexes to achieve a desired excited state
energy.......................................................................................................................................... 8
Solvatochromism and rigidochromism of cMa complexes ........................................................ 9
Controlling emission rate in cMa complexes ........................................................................... 11
OLEDs with cMa complexes as OLED emitters...................................................................... 12
Chapter 2 -Extended Ligands in Two-Coordinate Coinage Metal Complexes.......................... 16
Introduction............................................................................................................................... 16
Synthesis and Characterization................................................................................................. 18
Single Crystal X-Ray diffraction .............................................................................................. 19
Electrochemistry ....................................................................................................................... 20
Computer Modeling.................................................................................................................. 22
Photophysical Properties........................................................................................................... 25
Discussion................................................................................................................................. 30
Conclusion ................................................................................................................................ 33
Synthesis................................................................................................................................... 34
1H-Bim (5H-benzo[4,5]imidazo[1,2]imidazole).................................................................. 34
General Synthesis for 𝑨𝒖𝑪𝒍𝒄𝒂𝒓𝒃𝒆𝒏𝒆.................................................................................. 35
General Synthesis for 𝑨𝒖𝒂𝒎𝒊𝒅𝒆𝒄𝒂𝒓𝒃𝒆𝒏𝒆 ......................................................................... 35
Chapter 3 - Janus carbenes............................................................................................................ 37
Introduction............................................................................................................................... 37
Synthesis................................................................................................................................... 39
Crystallographic Analysis......................................................................................................... 40

vi
Computational Results.............................................................................................................. 42
Electrochemistry ....................................................................................................................... 44
Photophysical properties........................................................................................................... 46
OLEDs...................................................................................................................................... 55
Conclusion ................................................................................................................................ 59
Experimental Procedures.......................................................................................................... 59
Synthesis ClAuBAZAuCl .......................................................................................................... 59
Synthesis CzAuBAZAuCz ......................................................................................................... 59
Synthesis BCzAuBAZAuBCz...................................................................................................... 59
Synthesis BimAuBAZAuBim...................................................................................................... 60
Synthesis BBI ligand............................................................................................................. 60
Synthesis ClAuBBIAuCl ........................................................................................................... 61
Synthesis BCzAuBBIAuBCz ...................................................................................................... 61
Synthesis BZIAuBCz ................................................................................................................ 62
OLED Fabrication................................................................................................................. 62
Chapter 4 – cMa’s with electron deficient amides........................................................................ 63
Introduction............................................................................................................................... 63
Synthesis................................................................................................................................... 64
Synthesis 𝑨𝒖𝑪𝒛(𝑪𝑵)𝟐𝑷𝒁𝑰 .................................................................................................. 64
Synthesis 𝐀𝐮𝐂𝐳(𝐂𝐅𝟑)𝟐𝐏𝐙𝐈.................................................................................................. 64
Synthesis Cz(CF3)(Me)......................................................................................................... 65
Synthesis 𝑨𝒖𝑪𝒛(𝑪𝑭𝟑)(𝑴𝒆)𝑴𝑨𝑪........................................................................................ 65
Synthesis Boc-BimI2 ............................................................................................................. 66
Synthesis Bim(CN)2.............................................................................................................. 66
Synthesis 𝑨𝒖𝑩𝒊𝒎(𝑪𝑵)𝟐𝑷𝒁𝑰............................................................................................... 67
Synthesis 𝑨𝒖𝑩𝒊𝒎(𝑪𝑵)𝟐𝑷𝑨𝑪.............................................................................................. 67
Crystallographic Analysis......................................................................................................... 67
Electrochemistry ....................................................................................................................... 68
Photophysical properties........................................................................................................... 69
Attempts to obtain Bim(CF3)2................................................................................................... 77
Conclusion and Outlook ........................................................................................................... 78
Chapter 5 - Temperature-dependent Photophysics....................................................................... 79
Design and Set-Up of Cryostat ................................................................................................. 79
Fitting of the data...................................................................................................................... 83

vii
Collection of all measured cMa complexes.............................................................................. 85
Conclusion .................................................................................................................................... 94
Experimental Methods.................................................................................................................. 95
Single Crystal Diffraction Analysis...................................................................................... 95
Electrochemistry ................................................................................................................... 95
Modeling methods ................................................................................................................ 95
Photophysical Measurements................................................................................................ 96
OLED Device fabrication ..................................................................................................... 96
References..................................................................................................................................... 98

viii
List of Tables
Table 1-1. Selected performance parameters for OLEDs fabricated by vacuum deposition with cMa
emitters. The devices are sorted from high to low energy emission. *Solution processed device.............15
Table 2-1. Selected bond lengths and angles for the(carbene)M(amide) complexes. ................................19
Table 2-2. Electrochemical data. Electrochemical measurements were carried out in DMF solution
with 0.1 M NBu4PF6 electrolyte, and the potentials are listed relative to a ferrocene internal reference.
The absorption edge is taken as the point where ICT absorbance has dropped to 10% of the peak
absorbance for a toluene solution of the cMa. ............................................................................................21
Table 2-3. Parameters obtained from DFT and TDDFT modeling of the cMa complexes. .......................24
Table 2-4. Photophysical parameters for cMa complexes in toluene (tol) solution and polystyrene (PS)
thin film (1% by weight).............................................................................................................................27
Table 2-5. Energy and rate data from variable temperature photophysical measurements on 1% doped
polystyrene films.........................................................................................................................................29
Table 3-1. Selected X-ray crystallographic data.........................................................................................41
Table 3-2. Electrochemical data. Measurements were performed using 0.1 M TBAPF6 electrolyte in
DMF (except where noted), and the potentials are listed relative to a ferrocene internal reference...........45
Table 3-3. Photophysical data for mono- and bimetallic cMa complexes in solution................................51
Table 3-4. Photophysical data for mono- and bimetallic cMa complexes in different polymer matrices.
....................................................................................................................................................................52
Table 3-5. Energy and rate data from variable temperature photophysical measurements for mono- and
bimetallic cMa complexes in polystyrene films (1 wt%)............................................................................54
Table 3-6. Photoluminescence data of binuclear complexes and MACAuCz in the host material TCTA. .....55
Table 3-7. OLED device performance........................................................................................................58
Table 4-1. Selected X-ray crystallographic data.........................................................................................68
Table 4-2. Electrochemical data. Measurements were performed using 0.1 M TBAPF6 electrolyte in
DMF (except where noted), and the potentials are listed relative to a ferrocene internal reference...........69
Table 4-3. Photophysical data for mono- and bimetallic cMa complexes in solution................................75
Table 4-4. Photophysical data for mono- and bimetallic cMa complexes in different polymer matrices.
....................................................................................................................................................................76
Table 5-1:summary of all analyzed 2-coordinate cMa complexes, which were measured throughout
this PhD thesis. Compounds were measured in a <0.1wt% doped Polystyrene Films. ..............................85

ix
List of Figures
Figure 1-1. (a) shows a simplified kinetic scheme for thermally assisted delayed fluorescence (TADF).
(b) The potential energy diagram at the bottom illustrates the energy of the S1 and T1 states in the
ground state structure (before relaxation, green), the surface after relaxation for an interligand charge
transfer (blue) and for a metal to ligand charge transfer (red). .....................................................................1
Figure 1-2: Energy splitting diagrams for d10 orbitals in tetrahedral, trigonal planar and linear ligand
fields..............................................................................................................................................................3
Figure 1-3. The space filling model is shown (left) for 𝐶𝑢𝐶𝑧𝑀𝐴𝐶 (center). The dipp isopropyl groups
are shown in green. The potential energy was calculated as a function of the dihedral angle between
the MAC-carbene and either a N-carbazolyl or phenyl ligand, at 15° intervals between 0 and 90°. ...........4
Figure 1-4: 𝐴𝑢𝑎𝑟𝑒𝑛𝑒𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes with MLCT (arene = Ph and PhCz) and ICT (arene =
PhNPh2) excited states. The MLCT excited states distort to C-Au-C angles of ~120°. ICT states
maintain the same 180° angle in T1 as is observed the ground state.............................................................5
Figure 1-5. (a) Two series of cMa complexes are shown that illustrate variation of PL with added
steric bulk on a CAAC carbene and of the PL max by changing carbene and amide ligand. The
influence of the metal ion on 𝑘𝑟𝑇𝐴𝐷𝐹 is also illustrated. (b) A representative set of carbene and amide
ligands used to prepare cMa complexes are shown. Acronyms are shown in bold for all of the carbene
and amide ligands of cMa complexes that are discussed in this chapter.......................................................7
Figure 1-6. (a) Electrochemical potentials and ICT absorption energies for 𝐴𝑢𝑏𝑖𝑚𝑐𝑎𝑟𝑏𝑒𝑛𝑒
complexes. The carbene structures are shown above the plot and bim is shown in Figure 1-5. The
reduction potential of the IPr complex (*) is outside of the solvent window and is < -3.0 V. The energy
for the ICT transition also cannot be determined as it overlaps with high energy bands on the carbazolyl
ligand but is >> 3.0 eV. (b) The nature of the interaction between the N lone pairs and carbene -
orbital of the N-heterocycle carbene (NHC) for the cMa complexes is illustrated. In this representation,
the plane of the NHC ligand is perpendicular to the page. ...........................................................................8
Figure 1-7. (a) Emission spectra of cMa complexes at room temperature (RT) and 77 K in 2-MeTHF.
(b) Emission spectra of complex 𝐶𝑢𝐶𝑧𝑀𝐴𝐶 at 78-280 K in 2-MeTHF. (c) Schematic energy diagram
depicting the effect of rigidochromism on the ordering of the ICT/3Cz states. ..........................................10
Figure 1-8. Plots of 𝑘𝑟𝑆1 and EST values as a function of NTO overlap from the study of Li, et al..1

The dashed lines are not fits to data, but guides to the eye. Considering 𝑀𝐶𝑧𝑐𝑎𝑟𝑏𝑒𝑛𝑒 compojunds of
Li and other 𝑀𝐶𝑧𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes optimal NTO overlap values fall between 0.25 and 0.3 (right
plot).............................................................................................................................................................12
Figure 1-9. simplified representation of the layers and components found in an organic light emitting
diode device. ...............................................................................................................................................13
Figure 1-10. OLED Device characteristics with MACAuCz as dopant in the concentrations 10%, 40%
and 100% (neat). (a) current density-voltage-luminance (J-V-L) characteristics, (b) external quantum
efficiency with the electroluminescent spectra (EL) shown inset. Reproduced from reference 2 with
permission from the ACS............................................................................................................................14
Figure 2-1. New compounds considered in this paper (Ar = 2,6-diisopropylphenyl). Previously
reported 𝐴𝑢𝐶𝑧𝐵𝑍𝐼, 𝐴𝑢𝐶𝑧𝐶𝐴𝐴𝐶 and 𝐴𝑢𝐶𝑧𝑀𝐴𝐶 are also discussed...........................................................18
Figure 2-2. Thermal ellipsoid plots 𝐶𝑢𝐵𝐶𝑧𝑃𝐴𝐶 (left), 𝐴𝑢𝑏𝑖𝑚𝑀𝐴𝐶 (center) and 𝐴𝑢𝑀𝑏𝑖𝑚𝐵𝑍𝐼 (right)...20

x
Figure 2-3. Electrochemical redox potentials and transition energies for the 1
ICT state. The energy of
the 1
ICT state (in toluene) was estimated from the onset of the absorption band where the intensity was
0.10 the value at max. ..................................................................................................................................20
Figure 2-4. Molecular orbitals (MOs) of 𝐴𝑢𝐶𝑧𝐵𝑍𝐴𝐶 (left) and 𝐴𝑢𝐵𝑖𝑚𝐵𝑍𝐴𝐶 (right). The HOMO is
displayed with red and blue phases and the LUMO is displayed with turquoise and cream phases
(isovalue = 0.1). A magnified perspective is presented to highlight contribution of the d orbital to the
HOMO and LUMO. The 2,6-isopropyl groups have been removed for clarity..........................................22
Figure 2-5. BZAC/PAC extinction (a) and emission (b) in toluene and polystyrene. Inset shows the
spectra of 𝐴𝑢𝐶𝑧𝐵𝑍𝐴𝐶 in toluene and polystyrene (1 wt %), normalized at the Cz absorbance.
Extinction spectra in toluene (c) and emission spectra in polystyrene (d) of 𝐴𝑢𝑏𝑖𝑚𝑐𝑎𝑟𝑏𝑒𝑛𝑒
complexes. ..................................................................................................................................................25
Figure 2-6. Phosphorescence spectra in MeTHF at 77K of 3Cz and 3Bim.................................................28
Figure 2-7. (a) The ICT transition and the nature of the interaction between the N lone pairs and
carbene p-orbital for the cMa complexes are illustrated. (b) LUMO orbitals are shown for
𝑀𝑏𝑖𝑚𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes (the contribution from the carbene carbon to the LUMO is a pz-orbital,
perpendicular to the C-M bond). The LUMO energies (in eV) from DFT calculations are given below
the acronym of each ligand. ........................................................................................................................30
Figure 2-8. Maps for 𝑀𝐶𝑧𝐵𝑍𝐴𝐶 and 𝑀𝑏𝑖𝑚𝐵𝑍𝐴𝐶 showing the difference in Mulliken charge at each
atom (Q), measured as the difference in atomic charge in T1 relative to S0. The charge distribution is
the same for both the Cu and Au complexes...............................................................................................31
Figure 2-9. (a) Maps for 𝑀𝐶𝑧𝐵𝑍𝐴𝐶 and 𝑀𝑏𝑖𝑚𝐵𝑍𝐴𝐶 showing the difference in Mulliken charge at
each atom (Q), measured as the difference in atomic charge in T1 relative to S0. The charge
distribution is the same for both the Cu and Au complexes, (b) Hole/electron separation distances for
(carbene)Au(amide) complexes. .................................................................................................................32
Figure 2-10. (a) The centers of negative charge (blue spheres) and positive charge (red spheres) are
(b) The rate of TADF emission (𝑘𝑟𝑇𝐴𝐷𝐹) at room temperature for a doped polystyrene film is plotted
as a function of the hole/electron separation distance. The data point for 𝐴𝑢𝐶𝑧𝐵𝑍𝐴𝐶 is actually
𝐴𝑢𝐶𝑧(𝑡𝐵𝑢)2𝐵𝑍𝐴𝐶. Figures reproduced from reference 113 with permission from the ACS. .................33
Figure 2-11. Synthetic chart for all materials prepared in this study. Metal triflates were used to
perform ring closures of PrePAC and PreBZAC where M = Cu(II), Ag(I), and Na+
. (PAC)M’(X) was
achieved for M’ = Cu(I), Ag(I), and Au(I) where X = Cl- or BF4
-
. BZAC M’’(Cl) was achieved for M’’
= Cu(I) and Au(I). (BZI)Au(D) was prepared for D = Bim, MBim, and Obim. (PAC)M’(D’) and
(BZAC)M’’(D’) were isolated where D’ = Cz, BCz, and Bim. Ar = 3,6 – diisopropylphenyl...................36
Figure 3-1. Crystal structures of (a) BCzAuBAZAuBCz , (b) BCzAuBAZAuBCz and (c) bimAuBAZAubim .............41
Figure 3-2. Top: frontier molecular orbitals for CzAuBBIAuCz (left) and BZI’AuCz (right). The isovalue
are set to 0.1. The orbital contributions, energies (oscillator strength) and magnitude of the electronic
dipole moment are given for the ground and Sn states below the structures. The direction of the dipole
moment for each state is indicated with the arrow......................................................................................42
Figure 3-3. Calculated S1, S2 (with oscillator strength) and T1, T2 energies of CzAuBBIAuCz (left) and
BZIAuCz (right) with respect to their dihedral angles ................................................................................43
Figure 3-4. Potential energy surface scan for ligand rotation of the complexes. BZI is the carbene
substituted with a methyl group in place of one dipp moiety......................................................................44
Figure 3-5. CV (black) and DPV (red: oxidation, blue: reduction) traces of complexes and
decamethylferrocene collected in THF or DMF or MeCN with 0.1 M TBAPF6 as an electrolyte. The

xi
asterisk indicates the redox peak of decamethylferrocene. The electrochemical plot at the bottom
shows a comparison of Fc and DMF...........................................................................................................46
Figure 3-6. Absorption and emission spectra of mono- and bimetallic cMa complexes with carbazole
(a, b) and bim (c, d) in toluene....................................................................................................................47
Figure 3-7. Photostability of all complexes and Ir(ppy)3 as reference in degassed toluene. solutions
were in an air-free Schlenk cuvette and were excited with a 375nm LED lamp (1450 mW, 19.2 W
mm-2
)...........................................................................................................................................................48
Figure 3-8. (a) Absorption and emission spectra of BCzAuBAZAuBCz in various solvents. (b) Qualitative
potential surfaces for ground and excited states in solvents polarities. (c) Absorption and emission
maxima vs solvent polarity (ET(30) scale) for BCzAuBAZAuBCz and BZACAuBCz. (d) Slopes of absorption
and emission maxima vs ET(30).................................................................................................................49
Figure 3-9. (left) Emission spectra of BCzAuBAZAuBCz in PS, MeTHF and MeCyHex at room
temperature (solid) and 77K (doted). (right) Absorption and Emission spectra of BCzAuBAZAuBCz in
different polymer matrixes (PMMA, PS and Zeonex)................................................................................50
Figure 3-10. Center of h+(yellow) and ,e- (green) for S1 NTOs for cMa complexes. The d(h+
, e-
)
calculation on the geometry of complexes BCzAuBAZAuBCz, BCzAuBBIAuBCz and bimAuBAZAubim is zero
because the centers of h+
and eare overlapped because of the symmetry. To avoid this situation, these
complexes’ d(h+
, e-
) calculation was done by swapping one amide ligand to a Cl- which are
BCzAuBAZAuCl, BCzAuBBIAuCl and bimAuBAZAuCl...........................................................................................54
Figure 3-11. OLED Devices with 20% BCzAuBBIAuBCz in TCTA as host material with (red) and without
(blue) MoOx. Device structure: ITO/MoOx (5 nm)(with : red, without: blue)/20 % BCzAuBBIAuBCz in
TCTA (30 nm)/TPBi(50 nm)/Liq (1.5 nm)/Al (100 nm) ..............................................................................56
Figure 3-12. Doping concentration-controlled OLED devices with BCzAuBBIAuBCz and TCTA as host
material with the following device structure: ITO/MoOx (5 nm)/X% BCzAuBBIAuBCz in
TCTA (30 nm)/TPBi (50 nm)/Liq (1.5 nm)/Al (100 nm) ...........................................................................57
Figure 3-13. Device characteristics of OLEDs using BCzAuBBIAuBCz, BCzAuBAZAuBCz and MACAuCz
dopants. (a) Molecular structures of host and electron transport materials. (b) Device architecture with
HOMO and LUMO levels in eV. (c) Current-voltage and luminance-voltage curves. (d) Efficiency
(EQE) curves and electroluminescence spectra (inset)...............................................................................58
Figure 4-1. cMa complexes with EWG on the amide ................................................................................64
Figure 4-2. Thermal ellipsoid plots 𝐴𝑢𝐶𝑧(𝐶𝑁)2𝑃𝑍𝐼 (left), 𝐴𝑢𝐶𝑧(𝐶𝐹3)2𝑃𝑍𝐼 (center) and
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2𝑃𝑍𝐼(right).............................................................................................................................68
Figure 4-3. CV and DPV curves in CH2Cl2 for 𝐴𝑢𝐶𝑧(𝐶𝐹3)2𝑃𝑍𝐼 and 𝐴𝑢𝐶𝑧(𝐶𝐹3)(𝑀𝑒)𝑀𝐴𝐶 ................69
Figure 4-4. Absorption and emission spectra of cMa complexes with CzCN2 and Cz(CF3)2 (a, b) and
bimCN2 (c, d) in toluene and polystyrene...................................................................................................70
Figure 4-5. Absorption and emission maxima vs solvent polarity (ET(30) scale) for (a) 𝐴𝑢𝐶𝑧𝐶𝑁2𝑃𝑍𝐼
(green), 𝐴𝑢𝐶𝑧(𝐶𝐹3)2𝑃𝑍𝐼 (black) & 𝐴𝑢𝐵𝐶𝑧𝐵𝑍𝐼 (blue ) and (b) 𝐴𝑢𝐵𝑖𝑚𝐶𝑁2𝑃𝑍𝐼 (red),
𝐴𝑢𝐵𝑖𝑚𝐶𝑁2𝑃𝐴𝐶 (blue) & 𝐴𝑢𝐵𝑖𝑚𝐵𝑍𝐴𝐶 (purple) ....................................................................................71
Figure 4-6. Absorption and emission spectra of 𝐴𝑢𝐶𝑧(𝐶𝑁)2𝑃𝑍𝐼 (a,b), 𝐴𝑢𝐶𝑧(𝐶𝐹3)2𝑃𝑍𝐼 (c,d),
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2𝑃𝑍𝐼(e,f) and 𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2𝑃𝐴𝐶 (g,h) in the solvents MeCy, Toluene, MeTHF, CH2Cl2
and MeCN at room temperature and 77K...................................................................................................72

xii
Figure 4-7. Emission Spectra in MeTHF at 77K of the amides (left) Cz, Cz(CN)2 and Cz(CF3)2 and
(right) Bim and Bim(CN)2.The insets in both graphs show the electron density of the hole of Cz,
Cz(CN)2, Bim and Bim(CN)2 (Iso value: 0.1) ............................................................................................73
Figure 4-8. Emission and excitation spectra of (a), in the rigid polymer matrixes TOPAS, PS and
PMMA at 298K and 77K. ...........................................................................................................................74
Figure 5-1. (a) 45 degree set up and (b) 90 degree modified set up...........................................................79
Figure 5-2. manufacturer sample holder (left) and modified sample holder (core: copper, surface: gold
plating) with a cutout (5x2mm) on the right side, side to detect the emitted light .....................................80
Figure 5-3. (a) 2mm thick sapphire with doped polymer. The edge is polymer free to ensure a better
cold head to Substrate contact, (b)peeling off polymer film upon cooling, (c) sandwiched polymer film
between two 1mm thick Sapphire substrates. .............................................................................................81
Figure 5-4. (a) detected light is collimated and focused on an optical fiber, (b) Cryostat is enclosed by
a black box to reduce background light ......................................................................................................82
Figure 5-5: Temperature dependent lifetime data of 𝐴𝑢𝐵𝑖𝑚𝐵𝑍𝐼 doped PS thin film. In red is the fit
of the data according to the Equation described above...............................................................................84
Figure 5-6:Ahrrenius plot of temperature dependent lifetime of 𝐴𝑢𝐵𝑖𝑚𝐵𝑍𝐼 doped PS thin film. In
red is the linear fit with the equation given in the bottom left ....................................................................84
Figure 5-7: Temperature dependent lifetime data and Arrhenius plot of 𝐴𝑢𝐶𝑧𝐶𝐴𝐴𝐶, 𝐶𝑢𝐶𝑧𝐶𝐴𝐴𝐶,
𝐴𝑢𝐶𝑧𝑀𝐴𝐶 and 𝐶𝑢𝐶𝑧𝑀𝐴𝐶 doped PS thin film. In red is the fits and the fitting parameters are stated in
the insets in each figure...............................................................................................................................86
Figure 5-8: Temperature dependent lifetime data and Arrhenius plot of 𝐴𝑢𝑃ℎ𝐶𝑧𝑀𝐴𝐶, 𝐴𝑢𝐶𝑧𝐵𝑍𝐴𝐶
and 𝐴𝑢𝐵𝐶𝑧𝐵𝑍𝐴𝐶 doped PS thin film. In red is the fits and the fitting parameters are stated in the insets
in each figure...............................................................................................................................................87
Figure 5-9: Temperature dependent lifetime data and Arrhenius plot of 𝐴𝑢𝐵𝐶𝑧𝑃𝐴𝐶, 𝐴𝑔𝐵𝐶𝑧𝑃𝐴𝐶 and
𝐶𝑢𝐵𝐶𝑧𝑃𝐴𝐶 doped PS thin film. In red is the fits and the fitting parameters are stated in the insets in
each figure...................................................................................................................................................88
Figure 5-10: Temperature dependent lifetime data and Arrhenius plot of 𝐴𝑢𝐶𝑧𝑃𝑍𝐼, 𝐴𝑢𝐵𝑖𝑚𝐵𝑍𝐼 and
𝐴𝑢𝐵𝑖𝑚𝐵𝑍𝐴𝐶 doped PS thin film. In red is the fits and the fitting parameters are stated in the insets in
each figure...................................................................................................................................................89
Figure 5-11: Temperature dependent lifetime data and Arrhenius plot of 𝐴𝑢𝐵𝑖𝑚𝐶𝐴𝐴𝐶, 𝐴𝑢𝐵𝑖𝑚𝑀𝐴𝐶
and 𝐴𝑢𝐵𝑖𝑚𝑃𝐴𝐶 doped PS thin film. In red is the fits and the fitting parameters are stated in the insets
in each figure...............................................................................................................................................90
Figure 5-12: Temperature dependent lifetime data and Arrhenius plot of 𝐴𝑢𝐵𝑖𝑚𝑃𝑍𝐼, 𝐴𝑢𝑀𝐵𝑖𝑚𝐵𝑍𝐼
and 𝐴𝑢𝑂𝐵𝑖𝑚𝐵𝑍𝐼 doped PS thin film. In red is the fits and the fitting parameters are stated in the insets
in each figure...............................................................................................................................................91
Figure 5-13: Temperature dependent lifetime data and Arrhenius plot of 𝐴𝑢𝐷𝑀𝐵𝑖𝑚𝐵𝑍𝐴𝐶,
BCzAuBBIAuBCz and BCzAuBAZAuBCz doped PS thin film. In red is the fits and the fitting
parameters are stated in the insets in each figure........................................................................................92
Figure 5-14: Temperature dependent lifetime Arrhenius plot of BimAuBAZAuBim doped PS thin
film. In red is the fits and the fitting parameters are stated in the insets in each figure..............................93

xiii
Abbreviations
Refer to Figure 1-5 for abbreviations used for the carbene or amide ligands
cMa carbene-metal-amide 𝑘𝑟
𝑆1
radiative rate constant for S1
𝑘𝑟
𝑇𝐴𝐷𝐹 radiative rate constant for TADF
CIE Commission Internationale de
l’Éclairage 𝑘𝐼𝑆𝐶
𝑒𝑛𝑑 rate constant for T1→S1, (also called
kRISC)
3Cz carbazolyl triplet state 𝑘𝐼𝑆𝐶
𝑒𝑥𝑜 rate constant for S1→T1
dipp 2,6-diisopropylphenyl LUMO lowest unoccupied molecular orbital
d(h+
,e-
)
distance between the hole and
electron in the ICT state MLCT metal-ligand charge transfer
EST energy difference: S1 - T1 NHC N-heterocyclic carbene
Q
Mulliken charge difference
between S0 and T1
NTO natural transition orbital
EML emissive layer OLED organic light emitting diode
end endergonic PMMA polymethylmethacrylate
EQE external quantum efficiency PS polystyrene
ET/HT electron / hole transport layer
exo exergonic SOC spin orbit coupling
HOMO highest occupied molecular
orbital S1 lowest singlet excited state
PL
photoluminescence quantum
efficiency n decay lifetime for excited state n
ICT interligand charge transfer T1 lowest triplet excited state
ISC intersystem crossing TADF thermally activated delayed
fluorescence
Keq equilibrium constant T1 ⇄ S1 TPA triplet-polaron annihilation
knr non-radiative rate constant TTA triplet-triplet annihilation
kr radiative rate constant VON turn-on voltage

xiv
Abstract
Two-coordinate carbene-MI-amide (cMa, MI = Cu, Ag, Au) complexes have emerged as highly efficient
luminescent materials for use in a variety of photonic applications due to their extremely fast radiative rates
through thermally activated delayed fluorescence (TADF) from an interligand charge transfer (ICT)
process. This thesis presents a series of highly efficient luminescent 2-coordinate carbene-Gold-amide
(cMa) complexes, to achieve very high radiative rates (kr=4x106
s
-1
) and near unity photoluminescence
efficiencies. Temperature dependent photophysics allowed the determination of the singlet and triplet gap
(ΔEST) and the singlet radiative rates. Theoretical calculations on hole and electron separation are used to
explain the high radiative rates and offer a general design approach, to further improve this class of emitter.

1
Chapter 1 – 2-coordinate, monovalent copper
complexes as chromophores and luminophores
The following chapter was published in Advances in Inorganic Chemistry, Volume 83.
11 The chapter
was written collaboratively with Prof. Mark E Thompson and Peter Djurovich.
Introduction
Luminescence from Group 11 metal complexes (M(I) = Cu, Ag and Au) has been known for more than
fifty years.12, 13
These emissive materials cover a broad range of molecular complexes and structures,
often giving emission in both the solid state and in solution.
14, 15
The majority of reports of emissive
Cu(I) complexes involve four-coordinate, pseudo tetrahedral complexes. Luminescent three-coordinate
complexes have also been reported but are not as
prevalent as the four-coordinate complexes. Twocoordinate Group 11 complexes have seen a large
increase in interest over the last few years. The first
report of emission from two-coordinate copper
complexes came in 1987 and involved copper
pyrazole complexes.
16
Several other reports have
since cited M(I)L2
+
and LM(I)X complexes (L=
phosphine, carbene, X= halide, acetylide, aryl, amide)
emitting strongly in the solid state and in solution.17-
33 This chapter will focus on a subset of luminescent
Group 11 complexes, i.e. two-coordinate d10 metal
complexes, with an emphasis on copper-based
complexes. More extensive reviews are available for
the reader that wants to dig deeper into the cMa
materials.
14, 34-36 In this chapter we will focus largely
on (carbene)M(amide) (cMa) complexes, as these
have the most promising photophysical properties.
The cMa complexes will be abbreviated 𝑀𝑎𝑚𝑖𝑑𝑒
𝑐𝑎𝑟𝑏𝑒𝑛𝑒
with the identities of the carbene and amide given
through the text. We will cover a range of topics,
including synthesis, photophysics and applications of
these luminescent complexes.
The strong spin orbit coupling (SOC) of the metal
center in Group 11 complexes can induce sufficient
singlet character into the excited triplet state to give
radiative lifetimes in the microsecond regime,
outcompeting nonradiative decay. While some Group
11 complexes emit via fluorescence, the predominant
emission pathway utilizes the lowest triplet excited
state (T1). A substantial number of Group 11

Figure 1-1. (a) shows a simplified
kinetic scheme for thermally assisted
delayed fluorescence (TADF). (b) The
potential energy diagram at the bottom
illustrates the energy of the S1 and T1
states in the ground state structure
(before relaxation, green), the surface
after relaxation for an interligand
charge transfer (blue) and for a metal
to ligand charge transfer (red).
(a)

𝑨 𝑭

𝑘𝐼𝑆𝐶
𝑒𝑛𝑑
𝑘𝐼𝑆𝐶
𝑒𝑥𝑜 
𝑘𝑟
𝑇𝐴𝐷𝐹
(b)

2
compounds display phosphorescence from the triplet state, however, the most common emission
mechanism for Cu(I) complexes is via thermally assisted delayed fluorescence (TADF), originally
referred to as E-type delayed fluorescence (Figure 1-1).37-41
The latter process involves thermal
promotion from the T1 state to the lowest lying singlet excited state (S1), followed by fluorescence
(Figure 1-1). Spin-orbit coupling facilitates this process by enabling fast rates for intersystem crossing
(ISC) between S1 and T1 states (𝑘𝐼𝑆𝐶
𝑒𝑥𝑜) leading to ISC rates in the 1-5 ns-1
regime for copper-based
materials and 100 ns-1
for gold complexes.10, 42, 43
With such high ISC rates a rapid pre-equilibrium
approximation can be applied, leading to an emission rate that is given by the product of the radiative
rate from the S1 state (𝑘𝑟
𝑆1
) and the equilibrium constant for T1⇄S1 (Keq, 𝑘𝐼𝑆𝐶
𝑒𝑥𝑜 𝑘𝐼𝑆𝐶
𝑒𝑛𝑑 ⁄ ), Equation (1-1).
44

Since TADF involves emission from the S1 state the kinetics of the process are intimately tied to the
energy difference between the S1 and T1 states (EST) as this parameter determines Keq.
45

𝑘𝑟
𝑇𝐴𝐷𝐹 = 𝑘𝑟
𝑆1𝐾𝑒𝑞(𝑇1 ⇄ 𝑆1
) (1-1)
Group 11 cMa complexes have great promise for applications in optoelectronics. Their high
photoluminescent quantum efficiencies indicate that the nonradiative decays of the excited states to
heat or other nonproductive processes are minimal. As will be shown below, they can have very fast
TADF radiative rates, making them ideal as phosphorescent emitters in organic light emitting diodes
(OLEDs). In this case the excited state is created electrically within the organic materials, by the
combination of a hole and electron. OLEDs that use cMa complexes as emissive dopants are discussed
and briefly reviewed below.
Achieving high photoluminescent efficiency for Cu(I) complexes
A common approach to increase the photoluminescent quantum yield (PL) of copper complexes has
been to focus on decreasing the nonradiative decay rates of the materials. The SOC from the copper
ions in these complexes is sufficient to induce radiative decay rates (kr) from the T1 state in the 104
-106
s
-1 range. However, the measured luminescence rates (km from T1 = kr + knr) are typically two or more
times faster than this due to high nonradiative rates, leading to low PL values [Φ𝑃𝐿 =
(𝑘𝑟
(𝑘𝑟 + 𝑘𝑛𝑟 ⁄ ))]. The nonradiative decay rates for these complexes show a strong dependence on
the rigidity of the matrix, thus crystalline samples and doped thin films show the highest PL whereas
luminescence is often weak in fluid solutions. The same situation exists for many metal complexes that
emit via a TADF pathway, i.e., PL values are markedly lower in fluid solution than in the solid state.
The principal source of enhanced nonradiative decay of excited Group 11 complexes in fluid solution
is structural distortion of the excited state relative to the ground state, which leads to low energy barriers
for nonradiative crossing to the S0 surface. Minimizing the deleterious effects of this decay pathway
has led to enhanced PL in these complexes.
Many of Group 11 d10 complexes have metal-to-ligand-charge-transfer (MLCT) states46 that can decay
radiatively either through phosphorescence from the T1 state or via a TADF process. In either case the
MLCT excited state involves a reduced ligand and an oxidized metal ion. Four coordinate Cu(I)
complexes typically adopt tetrahedral geometries whereas Cu(II) derivatives are square planar.
Therefore, a tetrahedral Cu(I) complex promoted from a d
10 to a d9
configuration in the MLCT excited
state will undergo to a significant structural distortion to break the degeneracy of the unfilled T2 state

3
(Figure 1-2), a geometric change
attributed to the Jahn-Teller effect.47

Nearly all of the four-coordinate Cu(I)
complexes deviate from the ideal
tetrahedral symmetry required to give the
degenerate set of T2 and E MOs necessary
for a true Jahn-Teller distortion.
Nevertheless, pseudo-tetrahedral Cu(I)
complexes can still distort to a flattened
structure in the MLCT excited state.48-50

This change in molecular geometry
promotes surface crossing from the
lowest excited state to the S0 state in the
distorted structure (see red curve in
Figure 1-1(b)). A crystal lattice restricts the structural change from taking place, keeping better
alignment of the ground and excited state PE surfaces.51
Inhibiting structural distortion by careful
ligand design is also employed to minimize nonradiative decay in the excited states of copper
complexes in fluid media. An approach often used is to introduce bulky steric substituents onto the
ligands to restrict the ability of the complex to distort in the excited state.52-54

Structural distortion can also occur in the MLCT excited state of three-coordinate d10 complexes. The
highest lying d-orbitals in a three-coordinate geometry are a set with E symmetry (Figure 1-2) whose
degeneracy is broken by undergoing a structure change from Y- to T-shape in the MLCT excited state,55
fulfilling the requirements of a Jahn-Teller distortion. However, as with four-coordinate complexes,
most of the three-coordinate complexes do not have ideal trigonal planar structures. Nevertheless, the
PL values for three-coordinate complexes in solution are typically markedly lower than in the solid
state, consistent with an excited state distortion.56

Luminescence from Cu(I) halide cluster compounds was studied from the mid-1970s57 but emission
from mononuclear Cu(I) complexes was given impetus from pioneering work by McMillin, et al. on
cationic homoleptic and heteroleptic derivatives using phenanthroline and phosphine ligands.58-60

Further work established that these and other derivatives of the mononuclear complexes could be used
for photoredox61-63 and electroluminescence applications.64
A significant milestone was achieved when
highly luminescent neutral Cu(I) dinuclear65 and mononuclear66 complexes were developed by Peters,
et al. that used chelating amido-phosphine ligands. This work led to further studies on other
luminescent mononuclear Cu(I) complexes, particularly regarding their temperature dependent behavior
and application to efficient OLEDs fabricated using vacuum deposition methods.39, 67, 68

Prompted by the reports of highly efficient luminescence from four coordinate Cu(I) derivatives, we
embarked on the exploration of three-coordinate Cu(I) complexes using N-heterocyclic carbenes
(NHC) as electron acceptor ligands. We discovered both cationic and neutral luminescent derivatives
and probed the effects of rotation of the metal carbene bond on the photophysical properties.69-71
We
also examined complexes with NHC ligands that had different electron affinities to determine how the
varying the LUMO energies would alter the luminescent colors of the compounds.72, 73
Other work on
three-coordinate Cu(I) complexes using bis- or diphosphine ligands as electron acceptors demonstrated
Figure 1-2: Energy splitting diagrams for d10
orbitals in tetrahedral, trigonal planar and linear
ligand fields.

4
highly emissive species with either halide, thiolate or amide donor ligands.74-76
Efficient OLEDs were
fabricated by vacuum deposition using the diphosphino Cu(I) halide derivatives.74, 77

Two-coordinate complexes
In two-coordinate complexes the highest lying d orbital is neither a doubly or triply degenerate set of
MOs, so an argument based on Jahn-Teller symmetry considerations cannot be used to predict or
explain a geometric distortion in excited state. However, Renner and Teller have analyzed the situation
with linear compounds and determined that bending distortion can occur in the excited state.78, 79
The
effect is manifested in linear two-coordinated complexes that distort from their ground state structure
in the d9 MLCT excited state largely owing to - mixing of the MOs. Adding steric bulk can inhibit
the excited state distortion in the solid state,29 however, these complexes typically give lower PL values
in solution due to structural distortion in the fluid media.26-28, 80 An interesting illustration of this
phenomenon is seen for a series of (carbene)Au(I)(arene) complexes reported by Li, et al., in which the
arenes are phenyl, 4-N-carbazolyl-phenyl (PhCz) and 4-N,N-diphenyl-aniline (PhNPh2), Figure 1-4.
The lowest energy excited states for the phenyl and PhCz complexes are MLCT transitions between
the metal and the carbene -acceptor. In contrast, the excited state of the PhNPh2 derivative is an
interligand charge transfer (ICT) state, involving a triarylamine donor and a carbene acceptor. The
metal center remains a d10 ion when in the ICT excited state as it contributes only a minor portion of its
electron density to the transition (vida infra). Crystal structures and DFT modeling studies give
Ccarbene-Au-CPh bond angles of ~180° all of the complexes in their ground states. For the Ph and PhCz
complexes, a significant distortion of the structure is calculated in their MLCT excited states,
decreasing the bond angle to ~120°. In contrast, the PhNPh2 complex does not show any distortion in
the metal-ligand bond angle in its ICT excited state. The result is that photoluminescence efficiency is
low (PL < 0.01) for the MLCT emitters, whereas values for ICT emitters are much higher (PL ~ 0.4).
This study demonstrates that two-coordinated complexes with ICT excited states do not suffer from

Figure 1-3. The space filling model is shown (left) for 𝐶𝑢𝐶𝑧
𝑀𝐴𝐶 (center). The dipp isopropyl
groups are shown in green. The potential energy was calculated as a function of the dihedral
angle between the MAC-carbene and either a N-carbazolyl or phenyl ligand, at 15° intervals
between 0 and 90°.
0 15 30 45 60 75 90
0
1
2
3
4
5
6 Energy (kcal/mol) Dihedral angle (deg)
Cz phenyl

5
nonradiative decay processes
induced by Jahn-Teller or RennerTeller distortion that severely limits
the luminescence efficiency of
derivatives that emit from MLCT
excited states.
Aside from adding steric constraints
to the ligands of ICT emitters to
prevent excited state distortions,
there are other benefits accrued from
adding steric bulk to the ligand
spheres of two-coordinate d10
complexes. The first is to prevent the
nonradiative decay paths involving
bimolecular processes such as
excimer or exciplex formation.
Sterically encumbering groups that
shield the molecules from each other
and block access to the metal center are important in this regard. Increased steric constraints on the
NHC ligands can also help to slow unimolecular nonradiative decay processes. The addition of 2,6-
diisopropylphenyl (dipp) groups to the nitrogen atoms adjacent to the carbene carbon is common in
cMa complexes, as can be seen in the example in Figure 1-3. The dipp groups also provide a steric
constraint that significantly limits ligand rotation, which is a prominent source of nonradiative decay
in fluid media. The interligand steric interactions of the aryl rings on either side of the carbazole ligand
are key to inhibiting rotation. A space filling model of 𝐶𝑢𝐶𝑧
𝑀𝐴𝐶 in Figure 1-3 shows that the dipp groups
lie perpendicular to the carbene plane. The configuration of isopropyl groups (shown in red) above and
below the plane of the carbene creates a concave pocket which can accommodate the adjacent
carbazolyl ligand. This conformation positions the Cz protons illustrated in the figure directly into the
aryl rings of the dipp groups and leads to a marked up-field shift of these resonances in the 1H NMR
spectra due to the ring current of the adjacent aryl rings.81
Potential energy calculations for the cMa at
a range of dihedral angles between the carbene and amide show a minimum when the two ligands are
coplanar and a maximum when the ligands lie at 75° to each other. When the carbazole is replaced
with a phenyl group, the steric interaction with the isopropyl groups is markedly reduced and the barrier
to rotation of the arene is nearly eliminated.33
Similarly, if the dipp group is replaced with a phenyl or
mesityl group, the energy barrier for ligand rotation drops nearly to zero.
(carbene)M(amide) (cMa) complexes
As mentioned above, luminescent two-coordinate copper complexes have been known for many years.
A great deal of effort has been expended in the investigation of related two-coordinate complexes as
catalysts for organic transformations.82-84 In that regard it was found that carbene ligands were superior
to phosphines owing to the high bond dissociation energy of the (carbene)Cu(I) and (carbene)Au(I)
linkage.85, 86
Recent work from our lab,28, 29 as well as those of Romanov and Bockman,26, 27, 30 and
Stefan32, 87 showed that (carbene)2Cu+
and 𝐶𝑢𝑋
𝑐𝑎𝑟𝑏𝑒𝑛𝑒, where X = halide, alkoxide, thiolate and alkynyl,
Figure 1-4: 𝐴𝑢𝑎𝑟𝑒𝑛𝑒
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes with MLCT (arene = Ph
and PhCz) and ICT (arene = PhNPh2) excited states. The
MLCT excited states distort to C-Au-C angles of ~120°.
ICT states maintain the same 180° angle in T1 as is
observed the ground state.

6
can show emission from MLCT excited states, with emission lifetimes in the tens of s range in the
solid state, but give characteristically low PL in fluid solution. These complexes are relatively
straightforward to prepare. The deprotonated carbene reacts with a metal salt to give the
(carbene)M(halide) complex, and the halide is replaced by an exchange reaction with a metal
nucleophile, Equation 1-2. The same procedure is used to prepare alkoxide, thiolate and amide
complexes.
The first report to show the true potential of cMa complexes as luminescent materials involved the
study of OLEDs with cMa based emitters, by Di, et al..
88
Di’s work suggested that the gold-based cMa
complexes have PL for close to unity and copper-based cMa complexes have PL of ~ 0.50 in the
OLED structures. The authors showed temperature dependent emission that was consistent with TADF
and decay lifetimes in the s regime, however, PL values were not reported. High efficiency OLEDs
were reported, with gold-based cMa complexes (i.e. AuCz
AdCAAC and CuCz
AdCAAC
, Figure 1-5(a), top)
showing efficiencies as high as 25% and copper based cMa OLEDs as high as 10%. Assuming an
isotropic orientation of the dopants within the devices, the internal efficiency is expected to be ~ 4 times
higher than the external quantum efficiency (EQE).
89
Shortly thereafter Hamze, et al., showed that the
measured PL for CuCz
RCAAC complexes could be increased to values near unity in both fluid solution
and doped thin films by increasing the steric bulk the CAAC ligand (Figure 1-5a).90
Moreover, the
emission energy is directly correlated with the energy difference between the carbene ligand reduction
potential (LUMO) and amide oxidation potential (HOMO), consistent with an intramolecular charge
transfer origin for the excited state, not an MLCT state. This paper was followed up by Shi, et al., who
showed high PL and short emission lifetimes could be observed for related cMa complexes involving
MAC and DAC ligands (Figure 1-5a, bottom).2
Two different carbenes and the three different carbazole
ligands were used to prepare copper-based cMa derivatives with emission energies spanning from 430
nm to 700 nm. Luminescence from most of these copper-based cMa complexes is highly efficient in
solution and doped polystyrene thin films (PL > 0.8), however, values for the red and orange emitting
materials are lower (PL < 0.3) owing to vibronic coupling to the ground state (energy gap law).
Biscarbenes, such as BAZ, have also been recently reported in cMa complexes.91
The biscarbenes have
two metal centers and two amides in a single complex. While one could imagine some level of
cooperation or communication between the two carbene centers, the complexes behaved
photophysically very similar mononuclear cMa’s, as the excited state localizes on a single amide and
carbene center. A select number of cMa complexes have been reported recently for Cu, Ag and Au,
several of which are illustrated in Figure 1-5b.

7
Figure 1-5. (a) Two series of cMa complexes are shown that illustrate variation of PL with added
steric bulk on a CAAC carbene and of the PL max by changing carbene and amide ligand. The
influence of the metal ion on 𝑘𝑟
𝑇𝐴𝐷𝐹 is also illustrated. (b) A representative set of carbene and amide
ligands used to prepare cMa complexes are shown. Acronyms are shown in bold for all of the
carbene and amide ligands of cMa complexes that are discussed in this chapter.

8
Tuning frontier orbitals on ligands in cMa complexes to achieve a
desired excited state energy
The excited state energies of cMa complexes that emit from ICT states are directly correlated to the
energy difference between the HOMO (amide) and LUMO (carbene), EHL. One can estimate the
lowest excited state energy of a cMa from the EHL, which can in turn be derived from electrochemical
measurements. This correlation is illustrated in Figure 1-6(a), where the electrochemical potentials and
absorption energies of the ICT state for a series of 𝐴𝑢𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes are shown as a function of the
carbene.92
Note the amide is constant in this series, so the oxidation potential is near invariant, and thus
the EHL value is determined principally by the reduction potential. The decrease in transition energy
is clearly tied to an increase in electron affinity on going from BZI to PZI. Substitution of electron
donating and accepting groups onto carbazole and other amides can be also used to tune the HOMO
energy (and ICT energy) in a relatively straightforward manner.2, 4, 81, 90, 91, 93-96
Tuning the LUMO (carbene) energies in these complexes is also readily accomplished by ligand design,
but in a less obvious manner than with the amide donor. To first order, the LUMO in a cMa is a vacant
p-orbital of the N-heterocyclic carbene. The lone pair electrons on the adjacent nitrogen(s) atoms mix
with the vacant p-orbital and act to destabilize the LUMO [Figure 1-6 (b)]. A greater participation of
the lone pair electrons on nitrogen(s) in the LUMO will lead to greater has two N-atoms in the NHC
whereas CAAC ligand has only a single N-atom, which leads to a significant destabilization of the
LUMO on IPr relative to CAAC. The benzannulated ring of BZI withdraws electron density from the
imidazole ring, leading to less mixing of the N lone pairs in the MO, stabilizing the LUMO relative to
IPr. The BZAC ligand would be expected to have a destabilization and thus increase the energy of the
Figure 1-6. (a) Electrochemical potentials and ICT absorption energies for 𝐴𝑢𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒
complexes. The carbene structures are shown above the plot and bim is shown in Figure 1-5.
The reduction potential of the IPr complex (*) is outside of the solvent window and is < -3.0
V. The energy for the ICT transition also cannot be determined as it overlaps with high energy
bands on the carbazolyl ligand but is >> 3.0 eV. (b) The nature of the interaction between the
N lone pairs and carbene -orbital of the N-heterocycle carbene (NHC) for the cMa complexes
is illustrated. In this representation, the plane of the NHC ligand is perpendicular to the page.
Ipr BZI BZAC CAAC MAC PAC PZI -3.0
-2.5
-2.0
0.0
0.5 Oxidation Reduction
Absorption energy
(carbene)Au(bim) Electrochemical Potential (vs. Fc/Fc
+) *
(a)
2.0
2.5
3.0
3.5 Absorption energy (eV)

9
ICT state in the cMa complexes. The IPr ligand LUMO energy between that of IPr and BZI, since
only one nitrogen is attenuated by the benzannulation, however, the ring expansion from a five- to a
six-membered ring stabilizes the carbene p-orbital,97 leading to similar LUMO energies for both BZI
and BZAC. The PZI ligand is an aza-analog of BZI formed by substitution of nitrogen into the arene
ring. The pyrazine ring of PZI has a greater electron affinity than the arene of BZI, leading to greater
stabilization of the carbene LUMO for PZI. The carbonyl groups of MAC and PAC compete
effectively for the nitrogen lone pair, therefore the LUMO energies for these two carbenes are lower
than for BZAC. It is evident upon comparison of reduction potentials for BZAC and MAC that the
carbonyl group is more effective than benzannulation in stabilizing the LUMO.
The combination of tuning carbene and amide energies to influence the emission energy of cMa
complexes is illustrated in Figure 1-6(b). Here two carbenes (MAC and DAC) and three amides (Cz,
Cz(CN) and Cz(CN)2) were used to prepare a series of six complexes, whose emission spans the visible
spectrum.2
Solvatochromism and rigidochromism of cMa complexes
In nearly all of the cMa complexes illustrated in Figure 1-5, the lowest energy excited state at room
temperature is an ICT state involving the carbene and amide ligands. The electronic structure of these
complexes leads to large magnitudes for ground state dipole moments with calculated values typically
ranging from 10-15 D.2, 10, 98
The excited states in cMa complexes also have a reasonably large dipole
moment; however, since the ICT state leads to an oxidized amide and reduced carbene, the dipole
moment is oriented in the opposite direction to that of the ground state dipole. Thus, cMa complexes
display negative solvatochromism, i. e., hypsochromic shifts of their absorption spectra and
bathochromic shifts of their emission spectra with increasing solvent polarity, with the smallest Stokes
shifts found in nonpolar solvents.10, 91
Solvatochromism is also observed for rigid media, such as doped polymer films. Commonly used
aliphatic polymers like ZEONEX®, a copolymer of ethylene and norbornene, polystyrene (PS) and
polymethylmethacrylate (PMMA) have a polarities similar to hexane, toluene and ethyl acetate,
respectively.99 The cMa complexes undergo a hypsochromic shift of their CT absorption band with
increased polarity of the polymer matrix. In contrast to behavior in solution, the emission profile of
cMa complexes also displays a hypsochromic shift with increasing polarity of the polymer matrix. The
Stokes shifts increase along with increasing polymer polarity, however, the difference is smaller in
magnitude than seen for the same complexes in fluid solution.
91 The solvatochromic behavior upon
drying comes about as the polymer chains organize around the cMa in response to the permanent dipole
moment, which leads to a stabilization of the ground state. The rigid matrix cannot change the
“solvation” environment after excitation of the cMa complex, leading to destabilization of the excited
state, which has a dipole moment that is the opposite of the permanent ground state dipole. This
immobile solvent environment leads to small Stokes shifts and therefore, to a net blue shift in emission
in polar polymeric media.
91
The reversal of dipole moments between the ground and excited state in cMa complexes also gives rise
to interesting rigidochromic phenomena in frozen solutions. The emission spectra of the complexes
undergo marked hypsochromic shifts in frozen glass-like solvents, as shown by comparing the emission
spectra for several cMa complexes in 2-MeTHF at RT and 77 K in Figure 1-7(a). The spectral shifts

10
upon freezing the fluid solutions are consistent
with a change in the emission character from
an ICT to a ligand localized phosphorescence,
typically on the carbazole ligand (3Cz). In fluid
solution the solvent molecules reorient in
response to the excited state dipole, however,
in a frozen solvent this is not possible. At 77 K
the dipoles of the solvent molecules are frozen
in orientations that stabilize the ground state
dipole moment of the cMa. A solvation shell
organized around the ground state will
destabilize the excited state, thereby shifting
the ICT state to higher energy, leading to 3Cz
based emission at 77 K. Differing ratios of
3Cz/ICT emission are observed for different
cMa complexes. The spectrum of CuCzCN
MAC
displays ICT emission at room temperature but
only 3CzCN based emission is observed at
77 K. The ICT state in this derivative is
destabilized to the point that emission comes
from the 3Cz state. Both 3Cz and ICT emission
are observed for CuCz
MAC at 77 K as the two
states are similar in energy in frozen 2Me-THF.
In contrast, CuCz(CN)2
DAC only shows a broad ICT
emission band upon cooling to 77 K despite
undergoing a 125 nm (4300 cm-1
)
rigidochromic blue shift. Apparently, the
destabilization of the excited state for this
derivative in frozen solvent is insufficient to
raise the energy of the ICT state above that for
the 3Cz state.
To further probe the nature of the
rigidochromic shift on cooling, emission
spectra of CuCz
MAC in 2-MeTHF were obtained
at various temperatures between 77 K and 140
K (Figure 1-7(b)). The luminescence spectrum
at 77 K is dominated by 3Cz phosphorescence.
Minor changes in the emission profile were
observed between 77-93 K, which is close to
the glass transition temperature of 2-MeTHF
(Tg = 90-91 K).100-102 At temperatures above Tg,
the dielectric relaxation rate of the solvent
increases and the solvent molecules reorient in
Figure 1-7. (a) Emission spectra of cMa
complexes at room temperature (RT) and
77 K in 2-MeTHF. (b) Emission spectra of
complex 𝐶𝑢𝐶𝑧
𝑀𝐴𝐶 at 78-280 K in 2-MeTHF.
(c) Schematic energy diagram depicting the
effect of rigidochromism on the ordering of
the ICT/3Cz states.
400 500 600 700
0.0
0.5
1.0
0.0
0.5
1.0
400 500 600 700
0.0
0.5
1.0 Normalized emission
RT
77 K CuMAC
CNCz CuMAC
Cz
CuDAC
(CN)2Cz
(a)
Wavelength (nm)
400 500 600 700
0.0
0.5
1.0 Normalized Emission Wavelength (nm)
78 K
93 K
95 K
98 K
100 K
105 K
120 K
140 K
280 K
(b)

11
response to the excited state dipole. The softening of the matrix causes the ICT band to red-shift and
increase in intensity between 93 K and 105 K. Emission from the 3Cz state is not observed at
temperatures above 105 K and luminescence is due exclusively to the ICT state. The ICT state is fully
stabilized even before the melting point of 2-MeTHF (Tm = 137 K) as the emission profile is nearly
unchanged from 105 K to room temperature.
Controlling emission rate in cMa complexes
Thermal population of the S1 state must markedly out-compete the rate for both radiative and
nonradiative decay from the T1 state for efficient TADF. Two factors that determine the rate for thermal
promotion are the energy difference between the S1 and T1 states (EST) and the strength of spin orbit
coupling which promotes ISC. The ISC rate for T1→S1 (𝑘𝐼𝑆𝐶
𝑒𝑛𝑑) depends on both EST and the strength
of spin orbit coupling available for the transition. TADF emitters need EST to be < 0.2 eV to achieve
a sufficient Boltzmann population of the S1 state at room temperature. This metric is typically achieved
in organic ICT emitters by using steric constraints to force the donor and acceptor moieties into an
orthogonal conformation. The orthogonality minimizes orbital overlap and decreases interaction
between the two spins in the T1 state, thus diminishing the exchange energy, which in turn decreases
EST. However, the weak SOC of purely organic TADF emitters, even those with low EST values,
ultimately limits 𝑘𝐼𝑆𝐶
𝑒𝑛𝑑 to rates slower than 107
s
-1
. Compounding this problem is that fact that twisted
geometries weaken the donor-acceptor electronic interaction for the S1 state, which decreases the
oscillator strength and slows down the rate for 𝑘𝑟
𝑇𝐴𝐷𝐹. The cMa complexes in contrast have strong
SOC which gives rise to fast 𝑘𝐼𝑆𝐶
𝑒𝑛𝑑 and 𝑘𝐼𝑆𝐶
𝑒𝑥𝑜. These rates are fast enough that the S1 and T1 states are
in rapid equilibrium during the ICT lifetime. Moreover, a small EST can also be beneficial to achieve
a fast rate for 𝑘𝑟
𝑇𝐴𝐷𝐹. Thus, the TADF rate in cMa complexes can be increased using two parameters,
EST (which determines Keq) and 𝑘𝑟
𝑆1
(Equation 1-1).
As described above, the cMa complexes have been designed to have a near zero dihedral angle between
the acceptor (carbene) and donor (amide) ligand planes. Electronic interactions between the ligands are
sufficiently insulated by the bridging d10 metal ion to maintain a small exchange energy in the ICT
excited states. Twisting the two ligands to a large dihedral angle will further decrease EST but also
markedly lower the oscillator strength for the S1 state, ultimately slowing the rate for 𝑘𝑟
𝑇𝐴𝐷𝐹
.
90, 103, 104

For this reason the impetus has been to prepare cMa complexes with a coplanar orientation carbene and
amide ligands.
The apparent trade-off between EST and 𝑘𝑟
𝑆1
requires further analysis. A decrease in the interaction
between the hole and electron of the ICT state leads to a decrease in EST and a concomitant increase
in Keq, but it also slows 𝑘𝑟
𝑆1
. The fact that these two parameters inversely effect 𝑘𝑟
𝑇𝐴𝐷𝐹 suggest that
there should be an optimal level of hole/electron interaction to give a maximal value for 𝑘𝑟
𝑇𝐴𝐷𝐹. A
recent paper explored the hole/electron interaction in the TADF properties of a set of 12 cMa
complexes.1
In this study, a series of Cu(I), Ag(I) and Au(I) cMa complexes were prepared (Figure 1-8)
and investigated theoretically and experimentally. The degree of hole/electron interaction was
evaluated for each of the complexes by examining the overlap of the natural transition orbitals (NTOs)
for the hole and electron wavefunctions, with values ranging from 1.0 (complete overlap) to zero (no
overlap). The series of cMa compounds gave NTO overlaps between 0.21 and 0.41. The Ag(I)

12
complexes had the largest donor-acceptor separation10 and thus gave smallest NTO overlap, with
intermediate overlap values for Cu(I) complexes and highest NTO overlaps for the Au(I) complexes. The
expected reciprocal relationship was found between NTO overlap and values for EST and 𝑘𝑟
𝑆1
(Figure
1-8). Note that the value for Keq will decrease as EST increases, thus limiting 𝑘𝑟
𝑇𝐴𝐷𝐹. It was then
determined that the optimal level of NTO overlap to give the fastest 𝑘𝑟
𝑇𝐴𝐷𝐹 was 0.25-0.30, based on
data from the series of 12 compounds reported in this work as well as other cMa complexes in the
literature.1
OLEDs with cMa complexes as OLED emitters
Organic light emitting diodes (OLEDs) convert electricity into light, with emission colors covering the
visible spectrum as well as broadband white. They have been intensively studied for several decades
and appeared in commercial mobile displays in the early 2000’s.34, 36, 105 OLED displays have gained
popularity due to their high color purity and high contrast (due to the true black of an OLED when off).
The commercial OLED displays rely heavily on iridium and platinum-based dopants to collect both
singlet and triplet excitations formed in the electroluminescent process, so utilizing earth abundant
Figure 1-8. Plots of 𝑘𝑟
𝑆1
and EST values as a function of NTO overlap from the study of Li, et
al..1
The dashed lines are not fits to data, but guides to the eye. Considering 𝑀𝐶𝑧
𝑐𝑎𝑟𝑏𝑒𝑛𝑒
compojunds of Li and other 𝑀𝐶𝑧
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes optimal NTO overlap values fall between
0.25 and 0.3 (right plot).

13
metals would be economical and environmentally beneficial. The cMa complexes are a promising
family of emissive dopants in OLEDs, due to their high PL, high radiative rates, and tunable emission.
An OLED consists of an emissive layer sandwiched between transporting layers and electrodes. A
general device architecture is shown in Figure 1-9, consisting of hole and electron transporting layers
(HT and ET, respectively) and an emissive layer (EML). The highest efficiency devices involve more
than the three layers shown in Figure 1-9,
105, 106 but this device scheme is sufficient to explain the
electroluminescent process in OLEDs. When bias is applied to the organic stack of materials, holes
and electrons are injected into the emission layer from the HT and ET layers, respectively. Holes and
electrons are subsequently injected into the EML by the HT and ET materials, respectively. The EML
consists of a host matrix and an emitter dopant. The emitter is present at low concentration (typically
1-20%), to prevent self-quenching of the dopant emission. The photons generated in an OLED device
are a result of the recombination of holes and electrons on either the matrix or dopant molecules in the
EML forming an excited molecule, in both the S1 and T1 excited states. If the recombination takes place
at the host matrix the energy is transferred to and trapped at the dopant. Relaxation of that excited state
(also referred to as an exciton) leads to OLED light emission. This excited state is the same one formed
in optical excitation, so the PL of the dopant in the chosen host matrix is an important parameter and
limits the overall OLED efficiency, i.e. the total electroluminescent efficiency PL.
OLEDs with cMa emitters typically give broad and featureless emission, matching the
photoluminescence in doped thin films. The first demonstration of a cMa based OLED was in 2017 and
involved solution processed OLEDs utilizing the emitter CuCz
AdCAAC and analogous gold-based cMa
emitters.
95 The first efficient OLEDs prepared by thermal evaporation utilizing cMa complexes
involved CuCz
MAC as the emissive dopant (see Figure 1-10).2
Increasing the doping concentration from
10% to 40% to 100% resulted in a redshift in EL from 537 nm to 555 nm, due to aggregation of the
dopant. The efficiency of the devices increased from 17.0% (10% doping) to 19.4% (40% doping) and
dropped again to 16.3% (100% dopant). The OLED with an EML comprised of a neat film of the cMa
gave a high efficiency due to relatively low self-quenching in this material (PL of the neat film is >
0.8). The bulky sterically demanding diisopropylphenyl (dipp) groups play an important role by
Figure 1-9. simplified representation of the layers and components found in an organic light
emitting diode device.

14
preventing close stacked packing, which can lead to self-quenching. The turn-on voltage (VON) is a
good indicator for the efficiency of charge carrier injection and transport. The VON decreased from
3.0 V to only 2.5 V with increasing doping concentration. Another important variable is the roll-off,
which is defined as the efficiency decrease when driving the device to higher current densities. The
small roll-off visible in the devices is attributed to the very high radiative rates of the dopant (𝑘𝑟
𝑇𝐴𝐷𝐹
= 6 x105
s
-1
in PS and neat films).10
A large number of papers reporting OLEDs with cMa emitters have appeared since the seminal work
of Li, et al., in 2017.88
Selected device performance data taken from literature reports is shown in the
Table 1-1. The reader is directed to recent reviews for a more thorough listing of the cMa OLED studies
reported to date.34
Here we have focused on devices prepared by vacuum deposition since this is the
route to the most efficient OLEDs. The CIE (Commission Internationale de l’Éclairage) coordinate
given in the last column of Table 1-1, describes the EL emission as color coordinates on the CIE 1931
diagram, allowing precise characterization of the luminescence in terms of how color is perceived by
the human eye.107
The data in is focused on devices using Cu(I) cMa complexes as this text focuses on first row transition
metal elements. More device work has been reported using gold cMa complexes, as these derivatives
show higher overall thermal stability, radiative rates, and device efficiencies. Only one device utilizing
a silver cMa has been reported to date, owing to the low thermal stability of Ag cMa complexes.
9
Comparing the max values it is clear that the majority of cMa based OLEDs emit green light, fewer
OLEDs emitting blue, and a much smaller number of yellow-to-red emitting devices. The small number
of yellow-to-red OLEDs is related to the energy gap law, which describes the increase of nonradiative
rates for lower energy emitters. Since the PL of the emitter limits the device efficiency, the low
efficiencies for the low energy emitters (typically < 0.3) severely limits OLED performance (remember
EQE  0.25  total efficiency). The majority of the OLED data are on green emitters, as these emitters
do not suffer markedly from the effects of nonradiative decay caused by the energy gap law.
Figure 1-10. OLED Device characteristics with MACAuCz as dopant in the concentrations 10%,
40% and 100% (neat). (a) current density-voltage-luminance (J-V-L) characteristics, (b)
external quantum efficiency with the electroluminescent spectra (EL) shown inset. Reproduced
from reference 2 with permission from the ACS.

15
An added benefit of the green emitters is that they are likely less prone to suffer from the degradation
processes associated with triplet-triplet annihilation (TTA) and triplet-polaron annihilation (TPA). TTA
and TPA processes are considered to be dominant degradation pathways for blue emitting OLEDs that
utilize phosphorescent emitters.
108, 109
The issue here is the long excited state lifetime of the phosphors,
which enables TTA and TPA and the high energy of the blue emitter that leads to high energy products
from TTA and TPA that degrade. The cMa emitters offer a path to minimize TTA and TPA based
degradation pathways owing to high radiative rates that can be ten times faster than conventional heavy
metal-based emitters. The fast radiative rates (short excited state lifetime) will minimize the probability
that the bimolecular TTA or TPA take place prior to relaxation to the ground state. The cMa emitters
are available for these studies, but their thermal stability is insufficient to make high quality OLEDs for
device lifetime testing. The study of these materials is ongoing.
The cMa’s also can be used to sensitize fluorescent dopants in OLEDs.110 Sensitizers improve the
efficiency and narrow the spectral linewidth, by transferring energy via FRET or Dexter mechanisms
from an emitter that harvests both singlets and triplets to a singlet excited state of a fluorescence
sensitizer,111 utilizing the high radiative rates and the narrow emission profile of the fluorescent singlet
state. The basic operation of the OLED is similar to a device utilizing a cMa dopant, but in this case
the cMa is used to equilibrate the singlet and triplet states so they could be efficiently transferred to a
highly efficient singlet emitting (fluorescent) dye.
Table 1-1. Selected performance parameters for OLEDs fabricated by vacuum deposition with
cMa emitters. The devices are sorted from high to low energy emission. *Solution processed
cMa Dopant level
in EML (wt%)
VON
(V)
max
(nm)
EQEmax
(%)
CIE
(x,y) Ref.
AuCz
BZI 5% 4.2 430 11.8 0.16,0.06 4
CuIndole
MAC 100% 3.4 459 20.6 0.22,0.31 6
CuCz
PYI 4% - 474 23.6 0.14,0.22 7
CuCz(CN)
PZI 2% - 492 21.1 0.19,0.42 7
CuCz
AdCAAC 20% * 2.6 510 26.3 0.27, 0.48 8
AgCz(tBu)2
AdCAAC 20% 4.3 509 13.7 0.28,0.46 9
AuCz(tBu)2
AdCAAC 20% 4.7 518 18.7 0.29,0.49 9
AuCz
MAC
10% 3.4 508 13.8 0.25,0.50 10
40% 2.6 516 18.0 0.28,0.54 10
100% 2.5 528 13.5 0.34,0.56 10
CuCz
MAC
10 % 3.0 537 17.0 0.37,0.56 2
40 % 2.5 543 19.4 0.40,0.56 2
100% 2.5 555 16.3 0.44,0.53 2
CuCz(Ph)2
PZI 2% - 580 18.0 0.50,0.49 7
CuCz
PZI 6% - 582 18.7 0.51,0.48 7
CuCz(tBu)2
PZI 2% - 619 14.4 0.58,0.42 7
AuCz
BPZI 2% - 632 17.4 0.61,0.39 3
AuCz(tBu)2
BPZI 10% - 705 9.9 0.69,0.30 3

16
Chapter 2 -Extended Ligands in Two-Coordinate
Coinage Metal Complexes
The following chapter was published in J. Am. Chem. Soc. 2022, 144, 39, 17916–17928.93 Dr. Collin
N. Muniz was responsible for synthesizing the carbenes, as well as performing the theoretical
publication, electrochemistry and the photophysics measurement of the carbazole complexes. Anton
Razgoniaev synthesized the amide bim. I, on the other hand, contributed to the project by scaling the
synthesis of the amide bim, conducting photophysics experiments as well as all temperature dependent
analysis and analyzing all complexes via single crystal X-Ray diffraction. Additionally, I was also
involved in the general characterization and writing the publication together with Collin N Muniz.
Introduction
The luminescent properties of coinage metal complexes (M(I) = Cu, Ag and Au) were reported over fifty
years ago, 12, 13 with the first report of emission from a two-coordinate d10 coinage metal complex in
1987.16
Several papers have highlighted emission in the solid state and in solution for M(I)L2
+
and
LM(I)X complexes (L = phosphine, carbene, X = halide, acetylide, aryl, amide).17-19, 21-26, 28-33, 2, 9, 10, 90,
112-118
Of particular interest here is the promise of (carbene)M(I)(amide) (cMa) complexes as efficient
luminescent materials.2, 9, 10, 90, 112-118
The cMa complexes can have high photoluminescent quantum
yields (PL), short luminescence decay lifetimes () in the s regime and shorter, and emission color
tunable over the entire visible spectrum in solid, solution and doped films.2, 9, 10, 32, 90, 113-119
These
luminophores have properties similar to transition metal phosphors that contain Ru, Os, Ir and Pt used
in a range of applications including organic electronics and LEDs2, 9, 10, 18, 90, 112-118, 120-123
,
photocatalysis82, 124, 125, chemo- and bio-sensing126-128 and solar energy conversion.
Unlike the noble metal phosphors which luminesce from triplet states, the majority of the reported cMa
complexes emit via thermally assisted delayed fluorescence (TADF), Figure 1-1.
44
The carbene ligand
serves as an electron acceptor (A) and amide ligand serves as an electron donor (D). The lowest energy
(emitting) excited state is an interligand charge transfer (ICT) transition between the two ligands. The
energy of the ICT state depends on the choice of ligands but is relatively insensitive to the identity of
the metal atom.10
The linear geometry of the cMa complexes leads to a large spatial separation between
the donor and acceptor groups/ligands of ~4 Å. This spacing restricts the overlap between the -orbitals
of the two ligands, consequently limiting the exchange energy and thus the energy gap between lowest
singlet (S1) and triplet (T1) states (EST). A small EST enhances thermal population of the singlet state,
which improves the luminescence efficiency for TADF by increasing the radiative rate for emission.44

Organic TADF molecules have distinct lifetimes for prompt ( = 1-100 ns) and delayed ( = 1-1000 s)
emission that are controlled by EST and the rate of intersystem crossing S1→T1 (typically kISC < 107
s
-
1
).129-134
In contrast, the intersystem crossing (ISC) rates in metal containing TADF complexes are fast
enough (kISC ≥ 1010 s
-1
) to outcompete the radiative rates for the S1 state, making delayed emission
(TADF) independent of kISC.
44
The result is extremely fast prompt emission ( < 200 ps) and
comparatively short TADF lifetimes, TADF = 0.5-3 s, leading to high luminescence efficiency. 9, 10, 33,
87, 113

17
For compounds where the ISC rate exceeds 𝑘𝑟
S1
, a pre-equilibrium approximation can be made such
that the equilibrium constant (𝐾𝑒𝑞, 𝑇1 ⇄ 𝑆1) becomes a principal factor in determining 𝑘𝑟
TADF as shown
in Eqn. 2-1:44
𝑘𝑟
TADF = 𝑘𝑟
S1
∙ 𝐾𝑒𝑞 (2-1)
In this equation, 𝑘𝑟
TADF is only dependent on 𝑘𝑟
S1
and 𝐾𝑒𝑞, with Keq tied to EST, Eqn. 2-2.
45

𝐾𝑒𝑞(𝑇1 ⇄ 𝑆1) =
1
3
𝑒
−
∆𝐸𝑆𝑇
𝑘𝑏𝑇
(2-2)
Therefore, predictions can be made regarding the TADF properties for TADF emitters with fast ISC
(high spin orbit coupling) without prior knowledge of the ISC rates since only 𝑘𝑟
S1
and EST are needed
to estimate 𝑘𝑟
TADF. Boltzmann fits of temperature dependent luminescence data can be used to
accurately derive EST and 𝑘𝑟
S1 values, and thus Keq (Eqn. 2-2).
1, 2, 10, 44
The values of 𝑘𝑟
S1
can also be
estimated experimentally from absorption spectra using the Strickler-Berg analysis.135
Being able to control the rate of TADF with EST and 𝑘𝑟
S1
is useful when designing chromophores for
different applications. A small EST and large 𝑘𝑟
S1
leads to a fast 𝑘𝑟
TADF
,
44 which is important for
applications where a short excited state lifetime is important, such as organic LEDs. To that end, this
paper describes cMa complexes with some of the shortest TADF lifetimes reported to date ( ~ 250 ns).
Conversely, one can design molecules where the EST value is large, which will make 𝐾𝑒𝑞(𝑇1 ⇄ 𝑆1
)
very small and push (TADF) in the 10-100 s regime. These long-lived materials could be used as
sensitizers for photoelectrochemical reactions where the diffusion of the excited species to an
electrocatalysis in solution is a key step in the process. A EST value of 1000 cm-1
is sufficient to
increase the TADF lifetime to the s regime, while only sacrificing ~100 meV in electrochemical
potential for the excited state.
Both 𝑘𝑟
S1
and EST parameters are related to the degree of overlap between the hole and electron
wavefunctions that describe the excited state. This overlap can be evaluated using the natural transition
orbitals (NTOs) and is referred to as 𝛬𝑁𝑇𝑂.
1, 136, 137
While 𝛬𝑁𝑇𝑂 appears to be useful in predicting both
𝑘𝑟
S1
and Keq, it affects the two parameters in opposite ways. A large 𝛬𝑁𝑇𝑂 leads to a high value of 𝑘𝑟
S1
,
but it also increases the value for EST and thus lowers Keq. In a previous study we found that an 𝛬𝑁𝑇𝑂
= 0.25-0.3 is an optimal range to give the fastest 𝑘𝑟
TADF for cMa complexes of Cu, Ag and Au with a
variety of carbene and amide ligands.
1
The lowest 𝛬𝑁𝑇𝑂 values were obtained for silver based cMa
complexes.
In this paper we report the synthesis and characterization of a family of new cMa materials and study
their photophysical properties, with an eye to further explore how ligands can tune 𝛬𝑁𝑇𝑂 and how these
changes affect the excited state properties. The compounds discussed here are illustrated in Figure 2-1.
Neither the BZAC and PAC carbene ligands, nor the use of bim as the amide, in a cMa complex has
been reported previously. We find that the choice of carbene ligand affects the excited state energy but
has a weaker influence on the photophysical properties and EST of the cMa complexes. Interestingly,

18
shifting from a carbazolide to a bim ligand markedly increases 𝑘𝑟
TADF even though the energy of the
HOMO is effectively the same for both complexes.
Synthesis and Characterization
A range of cMa complexes have been prepared (Figure 2-1). Detailed syntheses and full
characterization data for all of the new compounds are given in the Synthesis part of this chapter and
in ref93. The basic synthetic route is illustrated in Scheme 2-1. All of the compounds studied here are
air stable in the solid state indefinitely and in solution for prolonged periods, with the exception of
𝐴𝑔𝐵𝐶𝑧
𝑃𝐴𝐶 which decomposes in air and solution over prolonged periods.
Scheme 2-1
In these materials we have varied the carbene fairly extensively and looked at both N-carbazolide (Cz)
and N-benzo[d]benzo[4,5]imidazo[1,2-a]imidazolyl (bim) amides. Part of this study is to investigate
two new carbene ligands for cMa complexes, i.e. BZAC and PAC, so the carbazolide complexes were
made for these two carbenes with Cu, Ag and Au to compare to other (carbene)M(Cz) compounds.
Although we were able to prepare pure samples of the PAC-based materials for all three metals, we
were only able to synthesize the Cu and Au complexes with BZAC. It appears that thermal instability
of 𝐴𝑔𝐶𝑧
𝐵𝑍𝐴𝐶 and 𝐴𝑔𝐵𝐶𝑧
𝐵𝑍𝐴𝐶 preclude their isolation. Although much of previous work involved
unsubstituted carbazole groups, we found that the addition of tBu groups to the 3,6-positions of the
𝑀𝑋
𝑃𝐴𝐶: M = Cu, Ag, Au; X = Cz, BCz and 𝑀𝑋
𝐵𝑍𝐴𝐶: M = Cu, Au; X = Cz, BCz
𝐴𝑢𝑎𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒: carbene = BZI, CAAC, MAC, PAC, BZAC; am = bim, Mbim, Obim
Figure 2-1. New compounds considered in this paper (Ar = 2,6-diisopropylphenyl). Previously
reported 𝐴𝑢𝐶𝑧
𝐵𝑍𝐼
, 𝐴𝑢𝐶𝑧
𝐶𝐴𝐴𝐶 and 𝐴𝑢𝐶𝑧
𝑀𝐴𝐶 are also discussed.

19
carbazole (BCz) led to greater solubility which facilitated our solution and thin film photophysical
studies. The solubility of the bim based cMa complexes are similar to their BCz analogs.
Single Crystal X-Ray diffraction
Single crystal X-ray diffraction was used to determine the molecular structures of 𝐶𝑢𝐵𝐶𝑧
𝑃𝐴𝐶
, 𝐴𝑢𝐶𝑧
𝐵𝑍𝐴𝐶
,
𝐴𝑢𝑏𝑖𝑚
𝐵𝑍𝐴𝐶
, 𝐴𝑢𝑏𝑖𝑚
𝐶𝐴𝐴𝐶
, 𝐴𝑢𝑏𝑖𝑚
𝑀𝐴𝐶
, 𝐴𝑢𝑏𝑖𝑚
𝑃𝐴𝐶 and 𝐴𝑢𝑀𝑏𝑖𝑚
𝐵𝑍𝐼
. Crystallographic data for the seven compounds are
given in Table 2-1 and representative structures of the compounds of 𝐶𝑢𝐵𝐶𝑧
𝑃𝐴𝐶
, 𝐴𝑢𝑏𝑖𝑚
𝑀𝐴𝐶 and 𝐴𝑢𝑏𝑖𝑚
𝐵𝑍𝐼 are
shown in Figure 2-2. Each structure was deposited in the Cambridge Crystallographic Data Centre
with the following accension codes: 𝐴𝑢𝐶𝑧
𝐵𝑍𝐴𝐶 2168084, 𝐶𝑢𝐵𝐶𝑧
𝑃𝐴𝐶 2170320, 𝐴𝑢𝑏𝑖𝑚
𝐵𝑍𝐴𝐶 2182717, 𝐴𝑢𝑏𝑖𝑚
𝐶𝐴𝐴𝐶
2155241, 𝐴𝑢𝑏𝑖𝑚
𝑀𝐴𝐶 2167514, 𝐴𝑢𝑏𝑖𝑚
𝑃𝐴𝐶 2170000 and 𝐴𝑢𝑀𝑏𝑖𝑚
𝐵𝑍𝐼 2168086. All the complexes have a coplanar
conformation of carbene and amide ligands (dihedral angles = 0.7-7.3°) and close to 180˚ angle around
the metal center (C-Au-N = 174 - 179˚). The C-Au bond lengths are in the range 1.97-2.01 Å and the
Au-N bond is 2.01 – 2.03 Å. Values for the C-Au and Au-N bond lengths are similar to analogous
(carbene)M(Cz) complexes published previously.
10, 90, 112, 115, 138 The only Cu compound structurally
characterized in this report, 𝐶𝑢𝐵𝐶𝑧
𝑃𝐴𝐶, has C-Cu and Cu-N bond lengths of 1.89 Å and 1.85 Å, respectively,
that are shorter than the Au based compounds and consistent with the difference in ionic radii of the
metals.10
Two different conformers are possible for 𝐴𝑢𝐵𝑖𝑚
𝑀𝐴𝐶, one with the carbonyl anti to the phenylene
(pictured in Figure 2-2) and the other where the carbonyl is syn to the phenylene. Both conformers
were present in the crystal and the diffraction data was best fit by treating the crystal as disordered with
a 10% “impurity” of the syn-conformer in a crystal of the anti-conformer. Only a single conformer for
𝐴𝑢𝐵𝑖𝑚
𝐵𝑍𝐴𝐶
, 𝐴𝑢𝐵𝑖𝑚
𝑃𝐴𝐶 and 𝐴𝑢𝐵𝑖𝑚
𝐶𝐴𝐴𝐶 was observed, even though a similar stereocenter to that of 𝐴𝑢𝐵𝑖𝑚
𝑀𝐴𝐶 was
present in these molecules. The crystals of 𝐴𝑢𝑀𝑏𝑖𝑚
𝐵𝑍𝐼 contain two tautomers, with the major component
(84%) having the methyl group closest to the Au-N bond (pictured in Figure 2-2). The ratio of
tautomers in the single crystals is close to the 75:25 ratio observed using NMR spectroscopy.
Table 2-1. Selected bond lengths and angles for the(carbene)M(amide) complexes.
compound C-M (A) M-N (A) C-M-N (˚) Torsion (˚)
NC-M-NC
Conformer
ratio CCDC #
𝐴𝑢𝐶𝑧
𝐵𝑍𝐴𝐶 2.004(7)
1.984(8)
2.020(6)
1.999(7)
179.1(3)
1.79.5(3) 3.7 100/0 2168084
𝐶𝑢𝐵𝐶𝑧
𝑃𝐴𝐶 1.879(2) 1.854(2) 173.9(1) 6.1 100/0 2170320
𝐴𝑢𝑏𝑖𝑚
𝐵𝑍𝐴𝐶 2.005(1) 2.025(7) 176.1(4) 2.7 100/0 2182717
𝐴𝑢𝑏𝑖𝑚
𝐶𝐴𝐴𝐶 1.974(4) 2.014(3) 178.1(2) 7.2 100/0 2155241
𝐴𝑢𝑏𝑖𝑚
𝑀𝐴𝐶 1.992(7) 2.023(6) 178.8(3) 4.0 90/10 2167514
𝐴𝑢𝑏𝑖𝑚
𝑃𝐴𝐶 1.995(3) 2.018(3) 178.6(2) 7.3 100/0 2170000
𝐴𝑢𝑀𝑏𝑖𝑚
𝐵𝑍𝐼 2.005(7) 2.031(6) 174.1(3) 0.7 84/16 2168086

20
Electrochemistry
The redox properties of the cMa complexes were examined by cyclic and differential pulse voltammetry
in dimethylformamide (DMF), and the potentials relative to an internal ferrocene reference are listed
in
Table 2-2c and graphically presented in Figure 2-3. Oxidations are irreversible for all cMa complexes
with Cz and bim amides. A common decomposition pathway for carbazoles upon oxidation is
dimerization at the 3,6-positions. Alkyl substitution at these sites inhibits this reaction pathway,
139
leading to reversible oxidation for the 𝐴𝑢𝐵𝐶𝑧
𝐵𝑍𝐴𝐶 and 𝐴𝑢𝐵𝐶𝑧
𝑃𝐴𝐶 complexes. The oxidation potentials for
silver and gold derivatives of 𝑀𝐵𝐶𝑧
𝑃𝐴𝐶 are similar; however, the copper analog is easier to oxidize. This
is contrary to what was observed for the 𝑀𝐶𝑧
𝑃𝐴𝐶 and 𝑀𝐶𝑧
𝑃𝐴𝐶 complexes, where the oxidation and reduction
potentials were unaffected by the choice of metal ion.10
A similar trend is observed for 𝑀𝐵𝐶𝑧
𝐵𝑍𝐴𝐶 where
the oxidation potential is lower for the copper derivative than the gold analog. Addition of the tBu
groups to Cz destabilizes the oxidation potential for a given complex by ~100 mV, e.g. 𝐴𝑢𝐶𝑧
𝐵𝑍𝐴𝐶/𝐴𝑢𝐵𝐶𝑧
𝐵𝑍𝐴𝐶
and 𝐴𝑢𝐶𝑧
𝑃𝐴𝐶/𝐴𝑢𝐵𝐶𝑧
𝑃𝐴𝐶
. This effect parallels the shift observed for the energy for the longest wavelength
absorption band in the two complexes, which also decreases by roughly 100 mV. The oxidation

Figure 2-2. Thermal ellipsoid plots 𝐶𝑢𝐵𝐶𝑧
𝑃𝐴𝐶 (left), 𝐴𝑢𝑏𝑖𝑚
𝑀𝐴𝐶 (center) and 𝐴𝑢𝑀𝑏𝑖𝑚
𝐵𝑍𝐼 (right)
Figure 2-3. Electrochemical redox potentials and transition energies for the 1
ICT state. The energy
of the 1
ICT state (in toluene) was estimated from the onset of the absorption band where the
intensity was 0.10 the value at max.
Cu-BCz Au-BCz Au-Cz Cu-BCz Ag-BCz Au-BCz Au-CzPACMACCAACBZACBZI AuBZI Mbim AuBZI Obim
0.4
0.2
0.0
-2.0
-2.2
-2.4
-2.6
-2.8
-3.0 Potential vs. Fc/Fc+
1.5
2.0
2.5
3.0
3.5
4.0 Absorption Energy (eV)
Abs. Energy Reduction Oxidation Aucarbene M bim PAC M Cz, BCz BZAC Cz, BCz

21
potentials for complexes with bim ligands are ca. 60 mV greater than those of the analogous Cz based
complexes, leading to a similar shift in the absorption energies. The dependence of the absorption
energy on the oxidation and reduction potentials of the complex is consistent with an interligand charge
transfer (ICT) transition for these complexes, as seen for other cMa complexes. Reduction waves were
observed for all complexes except 𝐴𝑢𝑏𝑖𝑚
𝐼𝑝𝑟 , which falls outside the solvent window for DMF. All
reductions are reversible, except for derivatives with BZAC ligands. The identity of the metal ion for
the complex does not affect the reduction potential (see 𝑀𝐵𝐶𝑧
𝑃𝐴𝐶). A gradual shift to more negative
potentials, along with a concomitant increase in energy for the lowest absorption band, is found upon
going from complexes with the PAC, MAC, CAAC, BZAC and BZI carbenes, which is again consistent
with an ICT assignment for the transition.
Table 2-2. Electrochemical data. Electrochemical measurements were carried out in DMF solution
with 0.1 M NBu4PF6 electrolyte, and the potentials are referenced to a ferrocene. The absorption edge
is the point where ICT absorbance has dropped to 10% of the peak absorbance for a toluene solution..
Complex Eox (V) Ered (V) Eredox (V) Abs. edge (eV)
𝑪𝒖𝑩𝑪𝒛
𝑩𝒁𝑨𝑪 0.06 -2.90 2.96 2.74
𝑨𝒖𝑩𝑪𝒛
𝑩𝒁𝑨𝑪 0.15 -2.83 2.98 2.77
𝑨𝒖𝑪𝒛
𝑩𝒁𝑨𝑪 0.26 -2.81 3.07 2.89
𝑪𝒖𝑩𝑪𝒛
𝑷𝑨𝑪 0.12 -2.28 2.40 2.25
𝑨𝒈𝑩𝑪𝒛
𝑷𝑨𝑪 0.02 -2.24 2.26 2.27
𝑨𝒖𝑩𝑪𝒛
𝑷𝑨𝑪 0.21 -2.24 2.45 2.33
𝑨𝒖𝑪𝒛
𝑷𝑨𝑪 0.29 -2.24 2.53 2.49
𝑨𝒖𝑪𝒛
𝑷𝒁𝑰 0.29 -1.99 2.30 2.36
𝑨𝒖𝒃𝒊𝒎
𝑷𝒁𝑰 0.36 -1.92 2.28 2.38
𝑨𝒖𝒃𝒊𝒎
𝑷𝑨𝑪 0.32 -2.16 2.48 2.52
𝑨𝒖𝒃𝒊𝒎
𝑴𝑨𝑪 0.33 -2.37 2.70 2.65
𝑨𝒖𝒃𝒊𝒎
𝑪𝑨𝑨𝑪 0.32 -2.62 2.94 2.86
𝑨𝒖𝒃𝒊𝒎
𝑩𝒁𝑨𝑪 0.31 -2.69 3.00 2.99
𝑨𝒖𝒃𝒊𝒎
𝑩𝒁𝑰 0.33 -2.79 3.12 3.08
𝑨𝒖𝑴𝒃𝒊𝒎
𝑩𝒁𝑰 0.26 -2.75 3.01 3.00
𝑨𝒖𝑶𝒃𝒊𝒎
𝑩𝒁𝑰 0.18 -2.81 2.99 2.93
𝑨𝒖𝒃𝒊𝒎
𝑰𝒑𝒓 0.33 1 1 3.59
1 The reduction potential of 𝑨𝒖𝒃𝒊𝒎
𝑰𝒑𝒓 was outside of the solvent window. This is consistent
with it having the largest optical LUMO of all compounds in this study.

22
Computer Modeling
The electronic properties of the complexes were modeled using Density Functional Theory (DFT) and
Time Dependent DFT (TDDFT), details are given in the Experimental section. The HOMO and LUMO
energies from these calculations are listed in Table 2-3 and representative orbital isosurfaces shown in
Figure 2-4. TDDFT methods were used to estimate the oscillator strengths for the singlet transitions,
as well as vertical energies and dipole moments for the singlet and triplet excited states. In all cases
but 𝐴𝑢𝑏𝑖𝑚
𝐼𝑝𝑟 , the transitions from S0 to the S1 are ICT in nature and have > 99.5% HOMO (amide) →
LUMO (carbene) character. For the majority of the cMa complexes the T1 state is comprised of the
same orbitals as the S1 state. The 𝐴𝑢𝑏𝑖𝑚
𝐼𝑝𝑟 and 𝐴𝑢𝐶𝑧
𝐼𝑝𝑟140 complexes are a special case where the energy of
the LUMO is sufficiently destabilized that the lowest triplet state is 3bim and 3Cz, respectively, rather
than 3
ICT. The dipole moments for the ground, 1
ICT and 3
ICT states are large in magnitude ( = 10-
20 D); however, the excited state dipoles are antiparallel to those of the ground state. This feature is
common in cMa complexes and is a consequence of the high degree of charge transfer in the ICT
excited state (Figure 2-4). The antiparallel alignment of dipoles gives rise to hypsochromic shifts in
absorption and bathochromic shifts for emission in solvents with increasing polarity. The natural
transition orbitals (NTOs) for the S1 state were calculated using the S0 optimized geometry and the
overlap between the hole and electron wavefunctions (NTO) determined as described in the
experimental section (Table 2-3). The transition energies in the series of Group 11 metals in 𝑀𝐶𝑧
𝑃𝐴𝐶
follow the same trends as reported previously for 𝑀𝐶𝑧
𝐶𝐴𝐴𝐶 and 𝑀𝐶𝑧
𝑀𝐴𝐶
.
10
The energies for the S1 and T1
states are independent of the metal center, whereas values for EST fall in the order Au > Cu > Ag,
which mirror NTO values of 0.39, 0.36 and 0.26, respectively. The interligand carbeneCꞏꞏꞏN distances for
the three complexes fall in the order Ag > Au > Cu. A long interligand distance might be expected to
give rise to a small NTO, but the higher value for the Au complex suggests a greater participation of the
metal ion in the excited state than for the Cu and Ag complexes.
Figure 2-4. Molecular orbitals (MOs) of 𝐴𝑢𝐶𝑧
𝐵𝑍𝐴𝐶 (left) and 𝐴𝑢𝐵𝑖𝑚
𝐵𝑍𝐴𝐶 (right). The HOMO is
displayed with red and blue phases and the LUMO is displayed with turquoise and cream
phases (isovalue = 0.1). A magnified perspective is presented to highlight contribution of the d
orbital to the HOMO and LUMO. The 2,6-isopropyl groups have been removed for clarity.

23
Examining the data for the 𝐴𝑢𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes it is apparent that the carbene ligand markedly affects
the S1 and T1 energies, whereas the EST, NTO and oscillator strengths for the S1 state are only
moderately altered by the nature of this ligand. Small differences in EST and NTO among the
complexes with various carbene ligands suggests that the stabilization imparted on the carbene
−system by addition of electron withdrawing carbonyl groups (MAC and PAC) or benzannulated
arene rings (BZAC, BZI, and PAC) does not markedly alter overlap between the electron and hole
wavefunctions. In contrast, both EST and NTO decrease upon shifting from a Cz to a bim donor,
despite there being only a minor increase in energy for the S1 and T1 states.
The lowest energy excited state in these cMa complexes is best characterized as interligand, not metalto-ligand, charge-transfer in character. In the cMa complexes the metal ion contributes equally to both
the HOMO and LUMO in the ICT state, so there is no net charge transfer between the metal and the
ligands. However, the metal ion is still an important participant in these transitions. This can be seen
in comparing the modeling data for the complexes in
Table 2-2 to data for 𝐻𝐶𝑧
𝑀𝐴𝐶 and 𝐿𝑖𝐶𝑧
𝑀𝐴𝐶. The separation between the central ion (neither of which have
accessible d-orbitals available for bonding) and the ligands was kept at the same distance as for the
copper ion in the geometry optimized structure of 𝐶𝑢𝐶𝑧
𝑀𝐴𝐶. Predictably, the values for EST and NTO
decrease precipitously in these two complexes, illustrating the contribution of the metal ion to the
valence orbitals of the cMa complexes.

24
Table 2-3. Parameters obtained from DFT and TDDFT modeling of the cMa complexes.
HO
(eV)
LU
(eV)
ELU-HO
(eV)
S1→S0
(eV/f)
T1→S0
(eV)
EST
a
(eV) NTO (S1)
𝐶𝑢𝐶𝑧
𝑃𝐴𝐶
-4.19 -2.20 1.99 2.43/0.13 2.21 0.22 0.36
𝐴𝑔𝐶𝑧
𝑃𝐴𝐶
-4.07 -2.29 1.78 2.39/0.09 2.27 0.11 0.26
𝐴𝑢𝐶𝑧
𝑃𝐴𝐶
-4.35 -2.20 2.15 2.61/0.19 2.36 0.25 0.39
𝐶𝑢𝐶𝑧
𝑀𝐴𝐶
-4.17 -1.99 2.17 2.48/0.11 2.27 0.21 0.36
𝐴𝑢𝐶𝑧
𝑀𝐴𝐶
-4.32 -1.99 2.33 2.66/0.16 2.41 0.25 0.40
𝐴𝑢𝐶𝑧
𝐶𝐴𝐴𝐶
-4.29 -1.60 2.68 2.86/0.16 2.56 0.30 0.43
𝐶𝑢𝐶𝑧
𝐵𝑍𝐴𝐶
-4.04 -1.39 2.65 2.95/0.12 2.76 0.19 0.36
𝐴𝑢𝐶𝑧
𝐵𝑍𝐴𝐶
-4.18 -1.41 2.76 3.10/0.19 2.87 0.23 0.39
𝐴𝑢𝐶𝑧
𝐵𝑍𝐼
-4.22 -1.46 2.76 3.09/0.20 2.87 0.22 0.40
𝐴𝑢𝐶𝑧
𝐼𝑝𝑟
-4.16 -0.81 3.35 3.44 3.15b 0.29 0.42
𝐴𝑢𝐶𝑧
𝑃𝑍𝐼
-4.38 -2.54 1.84 2.36/0.20 2.13 0.23 0.37
𝐴𝑢𝑏𝑖𝑚
𝑃𝑍𝐼
-4.44 -2.58 1.85 2.45/0.16 2.29 0.16 0.32
𝐴𝑢𝑏𝑖𝑚
𝑃𝐴𝐶
-4.43 -2.21 2.22 2.75/0.17 2.56 0.19 0.35
𝐴𝑢𝑏𝑖𝑚
𝑀𝐴𝐶
-4.41 -2.0 2.41 2.81/0.14 2.63 0.18 0.35
𝐴𝑢𝑏𝑖𝑚
𝐶𝐴𝐴𝐶
-4.39 -1.61 2.78 3.02/0.15 2.80 0.22 0.38
𝐴𝑢𝑏𝑖𝑚
𝐵𝑍𝐴𝐶
-4.29 -1.42 2.87 3.25/0.17 3.08 0.17 0.35
𝐴𝑢𝑏𝑖𝑚
𝐵𝑍𝐼
-4.30 -1.50 2.80 3.22/0.21 3.04 0.18 0.35
𝐴𝑢𝑏𝑖𝑚
𝐼𝑝𝑟
-4.27 -0.82 3.46 3.62/0.16 3.32b 0.30 0.36
𝐻𝐶𝑧
𝑀𝐴𝐶
-3.99 -2.15 1.84 2.35/0.001 2.34 0.01 0.06
𝐿𝑖𝐶𝑧
𝑀𝐴𝐶
-3.83 -2.08 1.75 2.17/0.03 2.11 0.06 0.18
a EST is based on energies calculated for the S1 and T1 states.
b The T1 state for this compound has
substantial 3
amide character as opposed to the 3
ICT state observed for the other complexes.

25
Photophysical Properties
The UV-visible absorption spectra of the complexes were recorded in toluene solution. Spectra for the
carbazole-based cMa complexes with BZAC and PAC are shown in Figure 2-5(a). The absorption
spectra of the 𝑀𝐵𝐶𝑧
𝐵𝑍𝐴𝐶 and 𝑀𝐵𝐶𝑧
𝑃𝐴𝐶 complexes display bands at high energy ( < 325 nm) that are assigned
to transitions localized on the carbene and carbazolide ligands. Structured bands at 375 nm are assigned
transitions on the carbazolide ligand, whereas bands assigned to absorption from the 1
ICT state are
centered at 410 and 480 nm for the BZAC and PAC complexes, respectively. As observed for other
cMa complexes,10 values for the extinction coefficients of the ICT bands fall in the order Au > Cu >
Ag. Changing the methylene moiety in BZAC to carbonyl in PAC stabilizes the LUMO and leads to a
marked red shift for complexes with the PAC ligand.
The absorption energies of the ICT state vary depending on the carbene ligands used here, which can
best be seen by comparing spectra of the Au complexes using the bim donor (𝐴𝑢𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒), Figure 2-5(c).

Figure 2-5. BZAC/PAC extinction (a) and emission (b) in toluene and polystyrene. Inset shows
the spectra of 𝐴𝑢𝐶𝑧
𝐵𝑍𝐴𝐶 in toluene and polystyrene (1 wt %), normalized at the Cz absorbance.
Extinction spectra in toluene (c) and emission spectra in polystyrene (d) of 𝐴𝑢𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes.

26
In these complexes the absorption transitions localized on the amide (bim) ligand that appear at 310 nm
ensures minimal overlap with the ICT band of the complexes. An exception is in 𝐴𝑢𝑏𝑖𝑚
𝐼𝑝𝑟 where the
LUMO is destabilized sufficiently to raise the energy of the ICT transition to be comparable to that of
the bim ligand. For the other complexes the ICT bands are distinct and have energies that fall in the
order BZI > BZAC > CAAC > MAC > PAC.
The absorption spectra of 𝐴𝑢𝐶𝑧
𝐵𝑍𝐴𝐶 in toluene, overlaid with that in polystyrene at 1 wt% loading and
normalized at 375 nm, is shown in the inset to Figure 2-5(a). The close match in absorption profiles
confirm that the two media have similar solvation properties. However, the polystyrene matrix is
expected to hinder geometric rearrangement, particularly rotation around the metal-ligand bond axis.
Luminescence spectra for all cMa complexes were thus recorded in toluene and polystyrene (Figure
2-5(b, d)). The spectra are broad and featureless at room temperature, indicative of emission from an
ICT state. Spectra for complexes measured in toluene are slightly red shifted relative to the same
compounds in polystyrene (Table 2-4), suggesting that only minor structural changes take place in the
excited state; however, the shift is greater for the complexes with the bim donor. This difference in the
rigidochromic effects in polystyrene suggests that the bim-based cMa complexes undergo a greater
structural distortion in their ICT state than do the carbazole-based materials. The emission energies
from the cMa complexes parallel values observed for the absorption energies of the ICT state (Table
2-4). Compounds ligated with the electrophilic PAC emit orange-red whereas those with BZI and
BZAC luminesce in the blue spectral region. For Au derivatives sharing a common carbene, the spectra
with BCz are red-shifted relative to their Cz analogs, whereas the bim congeners emit at higher energies.
Addition of methyl or methoxy groups to the bim ligand destabilizes the HOMO and leads to a
concomitant red shift in emission relative to the parent 𝐴𝑢𝑏𝑖𝑚
𝐵𝑍𝐼 complex. The 𝑀𝐵𝐶𝑧
𝑃𝐴𝐶 complexes display
only minor variation in emission energy with respect to the identity of the metal center. The
luminescence properties 𝐴𝑢𝑏𝑖𝑚
𝐼𝑝𝑟 are consistent with assignment to a combined ICT/ligand-triplet
transition, as observed for 𝐴𝑢𝐶𝑧
𝐵𝑍𝐼
,
118 leading to a slow radiative rate relative to the ICT emitters.
Cooling solutions of the 𝑀𝐵𝐶𝑧
𝐵𝑍𝐴𝐶 complexes to 77 K leads to a marked change in the luminescence
spectrum with a structured band appearing to the blue of the room temperature spectrum (Figure
2-5(b)). This rigidochromic transformation upon freezing the solvent has previously been shown to
come about from destabilization of the ICT state to an energy higher than that of the triplet state
localized on the carbazolide ligand (3Cz).10, 90, 118
The ICT state is destabilized to a lesser extent in a
rigid polystyrene matrix such that it remains the lowest energy state upon cooling to 77 K. Cooling
both toluene and polystyrene samples of 𝑀𝐵𝐶𝑧
𝑃𝐴𝐶 leads to a blue shift of the ICT band, but this change in
energy is not large enough to access the 3Cz state. In contrast, the energy of the triplet state for bim (ET
= 365 nm) is much higher than that of carbazole (ET = 415 nm) (Figure 2-6). The ligand localized triplet
states of the carbene ligands are also in the UV region so the 𝐴𝑢𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes emit from largely
featureless ICT transitions with Gaussian line shapes at all temperatures (Figure 2-5(d)).

27
Table 2-4. Photophysical parameters for cMa complexes in toluene (tol) solution and polystyrene (PS)
thin film (1% by weight).
max (PL) (nm) PL (%)  (μs) kr (106
s
-1
) knr (106
s
-1
)
tol PS tol PS tol PS tol PS tol PS
𝐶𝑢𝐵𝐶𝑧
𝐵𝑍𝐴𝐶 502 459 >95 93 0.71 1.24a 1.4 -- 0.04 --
𝐴𝑢𝐵𝐶𝑧
𝐵𝑍𝐴𝐶 500 484 >95 >95 0.56 0.72 1.7 1.4 0.07 <0.01
𝐶𝑢𝐵𝐶𝑧
𝑃𝐴𝐶 595 594 47 75 0.56 0.95 0.84 0.79 0.95 0.26
𝐴𝑔𝐵𝐶𝑧
𝑃𝐴𝐶 610 588 10 51 0.58 0.26 0.17 1.0 1.5 0.33
𝐴𝑢𝐵𝐶𝑧
𝑃𝐴𝐶 588 586 42 76 0.44 0.74 0.95 1.0 1.3 0.33
𝐴𝑢𝐶𝑧
𝑃𝑍𝐼 600 570 62 92 0.41 0.45 1.5 2.0 0.94 0.22
𝐴𝑢𝐶𝑧
𝑃𝐴𝐶 546 546 73 >95 0.74 0.81 1.0 1.2 0.36 0.04
𝐴𝑢𝐶𝑧
𝑀𝐴𝐶 ref 10
-- 508 -- 85 -- 0.8 -- 1.0 -- 0.18
𝐴𝑢𝐶𝑧
𝐶𝐴𝐴𝐶 b , ref 10
-- 472 -- >95 -- 1.14 -- 0.88 -- < 0.01
𝐴𝑢𝐶𝑧
𝐵𝑍𝐴𝐶 480 479 >95 >95 0.69 1.98a 1.5 -- <0.01 --
𝐴𝑢𝐶𝑧
𝐵𝑍𝐼 ref 118 448 432 94 >95 1.11 2.28a 0.85 -- 0.05 --
𝐴𝑢𝑏𝑖𝑚
𝑃𝑍𝐼 600 552 31 91 0.21 0.24 1.5 3.8 3.2 0.37
𝐴𝑢𝑏𝑖𝑚
𝑃𝐴𝐶 562 532 30 81 0.17 0.27 1.8 3.0 4.2 0.7
𝐴𝑢𝑏𝑖𝑚
𝑀𝐴𝐶 548 506 19 88 0.17 0.40 1.1 2.2 4.8 0.3
𝐴𝑢𝑏𝑖𝑚
𝐶𝐴𝐴𝐶 514 476 87 >95 0.63 0.55 1.4 1.8 0.02 <0.01
𝐴𝑢𝑏𝑖𝑚
𝐵𝑍𝐴𝐶 484 452 >95 >95 0.43 0.28 2.3 3.7 <0.01 <0.01
𝐴𝑢𝑏𝑖𝑚
𝐵𝑍𝐼 454 429 >95 >95 0.42 0.25 2.3 4.0 <0.01 <0.01
𝐴𝑢𝑏𝑖𝑚
𝐼𝑝𝑟 340 400 <1 <5 c 13 c 0.004 c c
𝐴𝑢𝑀𝑏𝑖𝑚
𝐵𝑍𝐼 460 436 >95 >95 0.42 0.29 2.4 3.4 <0.01 <0.01
𝐴𝑢𝑂𝑏𝑖𝑚
𝐵𝑍𝐼 480 450 92 >95 0.37 0.38a 2.5 -- 0.2 --
a The lifetime given is the weighted average of a biexponential fit; 𝐶𝑢𝐵𝐶𝑧
𝐵𝑍𝐴𝐶: 0.79 (73%), 2.47 (27%);
𝐴𝑢𝐶𝑧
𝐵𝑍𝐴𝐶: 0.84 (57%), 3.40 (43%); 𝐴𝑢𝐶𝑧
𝐵𝑍𝐼: 0.74 (46%); 3.6 (54%); 𝐴𝑢𝑂𝑏𝑖𝑚
𝐵𝑍𝐼 : 0.25 (80%), 0.90 (20%).
b
The CAAC ligand on this complex has a menthyl instead of the adamantyl group. c Values cannot be
accurately determined.

28
The photoluminescence efficiency for several of the cMa complexes are high (PL > 0.95) in both fluid
toluene and rigid polystyrene, and all have microsecond to sub-microsecond emission lifetimes. The
PAC-based cMa compounds show markedly lower PL, an effect attributed to the energy gap law where
vibrational deactivation increases the rate of nonradiative decay for the excited state. In particular, the
Ag complex has a low PL in solution which may be related to photodecomposition of this derivative.
Lower values for PL in toluene solution owe largely to increased nonradiative rates in fluid solutions,
likely ligand rotation and/or excimer/exciplex formation. The radiative rates for the bim based cMa
complexes are 1.8-4 x 106
s
-1
, which are some of the highest values reported for TADF emitters and
lead to radiative lifetimes as fast as 250 ns. The fastest radiative rates are observed for the complexes
with the highest excited state energy (i.e., 𝐴𝑢𝑏𝑖𝑚
𝐵𝑍𝐴𝐶 and 𝐴𝑢𝑏𝑖𝑚
𝐵𝑍𝐼 ) as expected from Einstein’s
relationship.141
In the cases where an analogous carbazole based complex is available for comparison,
the radiative rate for the bim based cMa complexes are two-fold faster than the Cz analogs.
Unfortunately, the near degeneracy in energy between the ICT and 3Cz states in 𝐴𝑢𝐶𝑧
𝐵𝑍𝐴𝐶 and 𝐴𝑢𝐶𝑧
𝐵𝑍𝐼
leads to non-first order behavior in the luminescence decay traces for these derivatives precluding direct
comparisons with the bim analogs.
As discussed in the introduction, the radiative rate of TADF from luminophores with effective spin orbit
coupling (SOC) is controlled by the radiative rate from the S1 state and the equilibrium constant
between the T1 and S1 states (Eqn. 2-1). For 𝑀𝐶𝑧
𝑀𝐴𝐶 and 𝑀𝐶𝑧
𝐶𝐴𝐴𝐶 complexes, the principal factor that
leads to faster TADF rates for the silver complexes over the copper and gold analogs is that the silver
complex has larger equilibrium constant, owing to its smaller EST (Eqn. 2-2). The modeling presented
above suggests that the EST values for bim based cMa complexes should be lower than their
carbazolide counterparts, which may account for their faster TADF rates. To validate this conjecture,
temperature dependent photophysical measurements were performed between 4-300 K to determine
the two components that control 𝑘𝑟
TADF (i.e., 𝑘𝑟
S1
and EST). These measurements were conducted
using polystyrene thin films doped at ~1 wt% with the cMa complex. Fitting the temperature dependent
lifetimes to a three-level model gives values for EST, the zero-field splitting (ZFS) and the radiative
rate from the S1 and T1 states.1, 2, 10, 44
The high PL in these cMa complexes allows us to neglect
nonradiative processes in our model. In this analysis, the ZFS is the energy difference between the two
Figure 2-6. Phosphorescence spectra in MeTHF at 77K of 3Cz and 3Bim 350 400 450 500 550
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
3bim is blue shifted into the UV normalized Intensity (arb. Units) Wavelength (nm)
bim
Cz
solid state at 77K

29
closely spaced triplet sublevels and the highest energy triplet sublevel. The energy spacing between
the two lowest sublevels could not be determined from data obtained at temperatures down to 4 K,
which is the limit of our cryogenic system. The fits to the data are shown for each complex in the
supporting information, and the energy and rate data derived from the fits is given in Table 2-5.
The 𝑀𝐵𝐶𝑧
𝑃𝐴𝐶 complexes show the same trends in TADF parameters observed previously for the MAC and
CAAC analogs.10
The copper and gold complexes give similar values for EST and S1 radiative rates,
but the silver complex gives a smaller EST and slower 𝑘𝑟
S1
, leading to comparable rates of TADF for
Table 2-5. Energy and rate data from variable temperature photophysical measurements on 1% doped
polystyrene films.
ZFS
(meV / cm-1
)
 10%
EST
(meV / cm-1
)
 3%
𝜏𝑆1
(ns)
 9%
𝜏𝑇1
(s)
 5%
𝐴𝑢𝐶𝑧
𝐵𝑍𝐴𝐶 a 65 / 520 18 a
𝐶𝑢𝐵𝐶𝑧
𝑃𝐴𝐶 b 70 / 570 18 b
𝐴𝑔𝐵𝐶𝑧
𝑃𝐴𝐶 b 26 / 210 35 b
𝐴𝑢𝐵𝐶𝑧
𝑃𝐴𝐶 0.9 / 7 72 / 580 14 36
𝐴𝑢𝐶𝑧
𝑀𝐴𝐶 1.2 / 10 87 / 700 13 28
𝐴𝑢𝐶𝑧
𝑃𝑍𝐼 0.9 / 7 54 / 440 21 82
𝐴𝑢𝑏𝑖𝑚
𝑃𝑍𝐼 1.1 / 9 31 / 250 26 76
𝐴𝑢𝑏𝑖𝑚
𝑃𝐴𝐶 1.0 / 8 45 / 360 17 38
𝐴𝑢𝑏𝑖𝑚
𝑀𝐴𝐶 2.0 / 16 51 / 410 14 38
𝐴𝑢𝑏𝑖𝑚
𝐶𝐴𝐴𝐶 1.1 / 9 53 / 430 18 16
𝐴𝑢𝑏𝑖𝑚
𝐵𝑍𝐴𝐶 1.2 / 10 41 / 330 19 19
𝐴𝑢𝑏𝑖𝑚
𝐵𝑍𝐼 1.5 / 12 41 / 330 15 19
𝐴𝑢𝑀𝑏𝑖𝑚
𝐵𝑍𝐼 1.5 / 12 44 / 350 16 11
𝐴𝑢𝑂𝑏𝑖𝑚
𝐵𝑍𝐼 1.2 / 10 41 / 330 19 17
a Luminescence from the 3Cz state at temperatures below 250 K prevents accurate Boltzmann fits for
the 3
ICT parameters.
b
Insufficient data was available at low temperature. Data below 4 K would be
required to determine accurate values for these parameters.

30
complexes with the three different metals. The 𝑀𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes show similar values for 𝑘𝑟
S1
to
their carbazole analogs (compare the MAC and PAC derivatives in Table 2-5), but the EST values for
the bim based cMa complexes are uniformly lower than those of the analogous carbazole based
materials by roughly 30%. The fast 𝑘𝑟
TADF for the bim based cMa complexes is thus due largely to a
significant increase in Keq caused by the small EST. We will discuss the origin of this decrease in EST
for the bim complexes in the following discussion section, but there is a clear connection to the smaller
NTO for the bim based complexes.
Discussion
The ICT transition in the cMa complexes is essentially an electron transfer from the amide group to the
carbene. While the transition utilizes the full spatial extent of the HOMO and LUMO (Figure 2-4), one
can consider the process as simply being a charge transfer from the nitrogen lone pair of the amide to
the vacant p-orbital of the N-heterocyclic carbene. With this line of reasoning, it is apparent that if the
amide is kept constant, the energy of the vacant p-orbital on the carbene will determine the energy of
the ICT state. The energy of the LUMO in the carbene ligands chosen for the present study span a
range of values (
Table 2-2 and Figure 2-3). The LUMO energy in N-heterocyclic carbene ligands is destabilized by
electron donation from the nitrogen(s) adjacent to the p-orbital of the carbene [Figure 2-7(a)]. Orbital
contours from DFT calculations illustrate the antibonding nature of the N-C
carbene bond in the LUMO
[Figure 2-7(b)], consistent with the energy diagram presented in Figure 2-7(a). The carbene p-orbital
will be destabilized by greater the participation of the nitrogen(s) in this MO, thus raising the transition
energy of the ICT state. The CAAC ligand has only a single N-atom, whereas the others have two Natoms destabilizing the carbene p-orbital, which explains the significant stabilization of the CAAC
LUMO relative to that of the closest analog (Ipr). The benzannulated phenyl ring of BZI accepts
Figure 2-7. (a) The ICT transition and the nature of the interaction between the N lone pairs
and carbene p-orbital for the cMa complexes are illustrated. (b) LUMO orbitals are shown for
𝑀𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes (the contribution from the carbene carbon to the LUMO is a pz-orbital,
perpendicular to the C-M bond). The LUMO energies (in eV) from DFT calculations are given
below the acronym of each ligand.
Ipr
-0.82
BZI
-1.50
BZAC
-1.42
CAAC
-1.61
MAC
-2.00
PAC
-2.21
ICT
(a) (b)
PZI
-1.85

31
electron density from the nitrogen atoms and leads to less mixing of the N lone pair with the carbene
p-orbital, stabilizing the LUMO relative to Ipr. The BZAC ligand would be expected to have a LUMO
between that of Ipr and BZI, since only one nitrogen is attenuated by the benzannulation, however, the
ring expansion from a five- to a six-membered ring stabilizes the carbene p-orbital,97 leading to similar
LUMO energies for BZI and BZAC. The carbonyl groups of MAC and CAAC compete effectively for
the nitrogen lone pair, therefore the LUMO energies for these two carbenes are lower that for BZAC.
It is evident upon comparison of BZAC and MAC that the carbonyl group leads to a much greater
stabilization of the LUMO than does benzannulation. The lowest LUMO energy is for PAC since the
-system is stabilized from both benzannulation and the carbonyl group.
An important set of observations that deserves further discussion are the short lifetimes for the cMa
complexes coordinated to the bim ligand. In all cases, the radiative rate for the 𝐴𝑢𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complex is
a factor of two faster that the analogous 𝐴𝑢𝐶𝑧
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complex. This difference occurs despite both
derivatives having similar excited state energies and extinction coefficients. Interestingly, it was found
that the values for NTO overlap are essentially the same for 𝐴𝑢𝑎𝑚𝑖𝑑𝑒
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes (excluding the Ipr
complex, which emits form a 3bim state) having either amide and a common carbene and give rise to
nearly the same 𝑘𝑟
𝑆1
for both congeners (Table 2-3 and Table 2-5). The increase in 𝑘𝑟
𝑇𝐴𝐷𝐹 rates for bim
based complexes is due to EST values that are ~ 40 % lower than for the Cz complexes. Remarkably,
this is the first case where a reciprocal relationship is NOT observed between EST and 𝑘𝑟
𝑆1
in a TADF
emitter. Shifting to carbenes with shallower LUMO energies leads to a blue shift in absorption and
emission and increasing 𝑘𝑟
𝑇𝐴𝐷𝐹, due to a progressive decrease in EST and a near constant 𝑘𝑟
𝑆1
.
1
Why do correlations of NTO overlap with EST and 𝑘𝑟
𝑆1 differ for the carbazole and bim based cMa
complexes? The reason for this can be seen in comparing the differences in Mulliken charges on each
atom between the ground and ICT excited states. The images in Figure 2-8 show the Mulliken change
differences at each atom for two different cMa complexes. It is expected that the amide will have a
positive charge and the carbene a negative charge, but it is striking how much of the charge is localized
on just the amide nitrogen and the carbene carbon. While the NTO overlap concerns the wavefunction
overlap of the hole and electron for all of the atoms of the complex, the charge in the ICT state is largely
on the three coordinated C-M-N atoms.
Figure 2-8. Maps for 𝑀𝐶𝑧
𝐵𝑍𝐴𝐶 and 𝑀𝑏𝑖𝑚
𝐵𝑍𝐴𝐶 showing the difference in Mulliken charge at each
atom (Q), measured as the difference in atomic charge in T1 relative to S0. The charge
distribution is the same for both the Cu and Au complexes.
-0.334
-0.19
-0.0461
0.0978
0.242
0.324
-0.33
-0.22
-0.11
-0.0032
0.11
0.22
0.25

32
A key difference between Cz and bim in the cMa complexes is the location of the center of positive
charge (Figure 2-9(a)) in the ICT excited states of the two amides. The positions calculated for the
electron and hole for the 𝑀𝑎𝑚𝑖𝑑𝑒
𝑐𝑎𝑟𝑏𝑒𝑛𝑒complexes discussed here are illustrated in Figure 2-10(a). Both
charges lie within the plane of the carbene and amide ligand, respectively. For Cz the center of positive
charge lies near the N atom, whereas in 𝑀𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes the three nitrogen atoms of the core
guanidinium unit disperse the positive charge and shift the center away from the nitrogen atom bound
to the metal. The position of the electron in each carbene for the ICT excited state is independent on
the choice of Cz or bim, and similarly the position of the hole for each amide is unaffected by the choice
of carbene. The center of electron charge resides within the C-M bond for Ipr and CAAC, whereas the
charge shifts into the carbene ligand for BZI, BZAC, MAC and PAC. This is consistent with the
enhanced delocalization in the LUMO by -extending the carbene backbone. The distance between
the centers of positive and negative charge, d(h+
, e-
), for each cMa is given in Figure 2-10(b). Values
for d(h+
, e-
) are spaced further apart for the bim complexes than the Cz-based ones. Figure 2-10(b)
shows the relationship of d(h+
, e-
) for 𝑀𝑎𝑚𝑖𝑑𝑒
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes to 𝑘𝑟
𝑇𝐴𝐷𝐹
. Three other carbenes have been
added to the compounds used to generate Figure 2-10. These carbenes are similar to BZI and MAC
and have reported 𝑘𝑟
𝑇𝐴𝐷𝐹 values.1, 138
Radiative rates for both bim and Cz complexes show a
dependence on d(h+
, e-
), with larger separations leading to faster 𝑘𝑟
𝑇𝐴𝐷𝐹 (Figure 2-10(b)); however, the
degree to which 𝑘𝑟
𝑇𝐴𝐷𝐹 increases with d(h+
, e-
) differ for the two derivatives. In all cases where data
are available for a given carbene, the 𝑘𝑟
𝑇𝐴𝐷𝐹 values are markedly faster for the bim-based cMa
complexes than for the Cz-based analogs (see dashed lines in Figure 2-10(b)). Part of the reason for
this difference in rates stems from the nature of the excited states for 𝑀𝐶𝑧
𝐵𝑍𝐼 and 𝑀𝐶𝑧
𝐵𝑍𝐴𝐶. The energies
of the 3
ICT and 3Cz states are nearly degenerate in both of these complexes, which promotes mixing of
the states and a concomitant increase in the measured lifetime. Although the energy separation between
the 3
ICT and ligand localized tripletstates for the 𝑀𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 and 𝑀𝐶𝑧
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes with PAC, MAC,
CAAC and PZI carbenes is large enough to diminish mixing between these excited states, the bim
complexes still display a marked increase in the radiative rate relative to their carbazole analogs. The
large d(h+
, e-
) separations for the bim complexes are consistent with lower NTO values, and thus the
𝐴𝑢𝑎𝑚𝑖𝑑𝑒
𝐶𝑎𝑟𝑏𝑒𝑛𝑒 d(h+
,e-
) (Å)
Cz Bim
Ipr 3.94 4.31
BZI 5.15 5.51
BZAC 5.08 5.36
CAAC 4.43 4.74
MAC 5.16 5.47
PAC 5.55 5.86
PZI 6.54 6.96
Figure 2-9. (a) Maps for 𝑀𝐶𝑧
𝐵𝑍𝐴𝐶 and 𝑀𝑏𝑖𝑚
𝐵𝑍𝐴𝐶 showing the difference in Mulliken charge at each
atom (Q), measured as the difference in atomic charge in T1 relative to S0. The charge
distribution is the same for both the Cu and Au complexes, (b) Hole/electron separation
distances for (carbene)Au(amide) complexes.

33
lower EST values, for the bim complexes relative to their carbazole counterparts. It is interesting that
even though 𝐴𝑢𝑏𝑖𝑚
𝑃𝑍𝐼 has a substantially larger d(h+
, e-
) than 𝐴𝑢𝑏𝑖𝑚
𝐵𝑍𝐼 and 𝐴𝑢𝑏𝑖𝑚
𝐵𝑍𝐴𝐶 the 𝑘𝑟
𝑇𝐴𝐷𝐹 values for the
three complexes are comparable. This may indicate that the d(h+
, e-
) value observed for 𝐴𝑢𝑏𝑖𝑚
𝑃𝑍𝐼 (6.8 Å)
is approaching a limit for enhancing 𝑘𝑟
𝑇𝐴𝐷𝐹. Moreover, at first glance an increase in d(h+
, e-
) might also
be expected to decrease 𝑘𝑟
𝑆1
, as observed upon substituting Ag+
(with a large ionic radius) in place of
Cu+
and Au+
for the 𝑀𝐵𝐶𝑧
𝑃𝐴𝐶 complexes, as well as for the 𝑀𝐶𝑧
𝑀𝐴𝐶 and 𝑀𝐶𝑧
𝐶𝐴𝐴𝐶 derivatives.
10
However, this
is not the case for 𝐴𝑢𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes since 𝑘𝑟
𝑆1
remains nearly constant (in the 15-20 ns range) for
complexes with either carbazole or bim ligands (Table 2-5). Thus, the decrease in EST brought about
by the enhanced d(h+
, e-
) is the principal factor leading to the fast 𝑘𝑟
TADF values for the 𝐴𝑢𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒
complexes.
Conclusion
Two-coordinate, coinage metal (carbene)M(I)(amide) complexes have attracted a great deal of attention
recently, due in part to their excellent photophysical and electroluminescent properties.2, 9, 10, 90, 112-118, 138

These cMa complexes give high photoluminescent and electroluminescent efficiencies and have very
short phosphorescence (TADF) lifetimes. Most of the work published to date for these complexes has
focused on CAAC, MAC and BZI type ligands, with the most common amide being a carbazolide. In
this paper we have explored the impact of -extending the carbene and amide ligands on the physical
and photophysical properties of cMa complexes. The carbene ligand was -extended via
benzannulation (BZI, BZAC and PAC). The amide ligand was -extended via replacement of the
central “pyrrole” moiety of carbazole with a guanidinium group (i.e., the bim ligand). We have found
that -extending the carbene and amide together leads to some of the highest radiative rates observed
for a triplet-controlled emission process, with rates as high as 4 x 106
s
-1
.
Figure 2-10. (a) The centers of negative charge (blue spheres) and positive charge (red spheres)
are (b) The rate of TADF emission (𝑘𝑟
𝑇𝐴𝐷𝐹) at room temperature for a doped polystyrene film
is plotted as a function of the hole/electron separation distance. The data point for 𝐴𝑢𝐶𝑧
𝐵𝑍𝐴𝐶 is
actually 𝐴𝑢𝐶𝑧(𝑡𝐵𝑢)2
𝐵𝑍𝐴𝐶 . Figures reproduced from reference 113 with permission from the ACS.
1x106 2x106 3x106 4x106
3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
RED amide = bim
BLACK amide = Cz
PAC
MAC
CAAC
BZAC
BZI
PZI d(h
+
, e-) (Å)
kTADF
r (s-1
)
Aucarbene
amide
(b)

34
The lowest energy absorption and emission bands in the 𝑀𝐶𝑧
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes are due to transitions
involving the amide (HOMO) and carbene (LUMO). The energy of the LUMO shifts significantly
across the series of carbenes explored here, leading to the ordering of energies for the ICT bands of
𝑀𝐶𝑧
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 being BZI > BZAC > CAAC > MAC > PAC. Other than this shift in energy, the choice of
carbene in the 𝑀𝐶𝑧
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complex has a modest impact on the photophysical properties since the
complexes have similar values for extinction coefficients, as well as for radiative and nonradiative
decay rates. In contrast, the effect of −extending the amide ligand (𝑀𝐶𝑧
𝑐𝑎𝑟𝑏𝑒𝑛𝑒→ 𝑀𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒) has almost
no effect on the energy of the ICT transition, but other photophysical properties are markedly altered.
The radiative decay rate (𝑘𝑟
𝑇𝐴𝐷𝐹) for a given 𝑀𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 is between two-to-four times greater than the
rate for the 𝑀𝐶𝑧
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complex of the same carbene. The principal source of the rate enhancement for
bim-based cMa complexes comes from a decrease in the EST for these complexes compared to the Cz
counterparts. This decrease in EST is related to the smaller NTO for 𝑀𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes, brought
about by a shift of the hole density (HOMO) away from the electron localized on the carbene. While
this picture explains the difference between Cz- and bim-based complexes with a common carbene
ligand, it does not explain the differences in radiative rates seen for the 𝑀𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes. The
calculated values of EST and NTO for the 𝑀𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes are similar, but their radiative rates
differ by more than a factor of two in comparing MAC to BZI complexes. Consistent with the
experimental 𝑘𝑟
𝑇𝐴𝐷𝐹 data, values for EST are lowest for 𝑀𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes with -extended carbene
ligands, i.e. BZI, BZAC and PAC. Interestingly, 𝑘𝑟
𝑆1
for the 𝑀𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes only marginally
decreases with a decrease in NTO and increase in hole/electron separation. Therefore, the increase
𝑘𝑟
𝑇𝐴𝐷𝐹 for 𝑀𝑏𝑖𝑚
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 complexes versus their 𝑀𝐶𝑧
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 analogs owes primarily to a large increase in Keq,
rather than to a change in 𝑘𝑟
𝑆1
(Eqn. 2-1), and is most prominent for the complexes with benzannulated
carbene ligands. Thus, −extension of both ligands (carbene and amide) imparts the most significant
enhancements in radiative rate.
While this report has focused on achieving fast radiative lifetimes, it is important to stress that
knowledge of 𝛬𝑁𝑇𝑂 and d(h+
, e-
) can be used to tailor the ICT excited states in related compounds for
use as sensitizers in photoelectrochemical processes. A small 𝛬𝑁𝑇𝑂 will lead to a large EST value and
small Keq, as described above, but the large 𝛬𝑁𝑇𝑂 will conversely improve hole/electron overlap, and
thus increase extinction coefficients for light absorption. Therefore, it should be possible to achieve
both long lifetimes and high molar absorptivity from other cMa complexes.
Synthesis
1H-Bim (5H-benzo[4,5]imidazo[1,2]imidazole)
1
st step: This compound was synthesized based on modified literature procedure.142
The following
synthesis was carried out under Schlenk conditions in a 500 mL 3-neck flask with additional funnel. 2-
Chloro-benzimidazole (50.0 g, 328 mmol, 1.0 eq) was charged in the flask and the system was pump-

35
purged; N-methylpyrrolidine (180 mL) was added via cannula transfer and bubble degassed for 20 min.
Under vigorously stirring, methanesulfonic acid (31.5 g, 21.3 mL, 328 mmol, 1.0 eq) was added
dropwise using an additional funnel. After 1 h stirring at ambient temperature, 2-bromoaniline (56.4 g,
328 mmol, 1.0 eq) was added and heated to react at 100 ˚C overnight. (Tip: preheat everything which
gets in contact with bromoaniline, like the beaker and funnel, which facilitates the transfer). After
cooling down, the reaction was quenched with 100 mL water, and afterwards neutralized using 30 wt%
aqueous KOH solution. Precipitate was filtered and dried at 90 ˚C in vacuum overnight. N-(2-
bromophenyl)-1H-benzo[d]imidazol-2-amine was obtained NMR pure in 80% yield (76.5 g, 265
mmol). [6.93 (td, 1H); 7.01 (dd, 2H); 7.33 (dd, 2H); 7.39 (td, 1H); 7.62 (dd, 1H); 8.62 (d, 1H); 11.25
(s, 2H)]
2
nd Step: N-(2-bromophenyl)-1H-benzo[d]imidazol-2-amine (76.5 g, 265 mmol, 1.0 eq), Cs2CO3
(130 g, 398 mmol, 1.5 eq) and CuBr2 (1.19 g, 5.3 mmol, 0.02 eq) was added to a Schlenk flask and
pump purged. 275 mL dry, bubble degassed DMF was added via a cannula. After reaction at 130 ˚C
overnight, rection was cooled down and quenched with water. Vacuum filtration yielded the crude
product, which was recrystallized in hot acetic acid. The acetic acid was removed by distillation and
the remaining solid was NMR pure bim and was used directly in further reactions. Alternatively, the
crude bim can be sublimed in a high-vacuum sublimator starting at 230 ˚C with steady increase to
290 ˚C over 4-5 h to yield a high purity snow-white product in 90% yield (49.5 g, 265.5 mmol).
1H NMR (400 MHz, DMSO) δ 11.99 (s, 1H), 8.06 (dd, J = 7.8, 0.8 Hz, 2H), 7.46 (d, J = 7.9 Hz, 2H),
7.23 (dtd, J = 24.3, 7.5, 1.2 Hz, 4H).
General Synthesis for 𝑨𝒖𝑪𝒍
𝒄𝒂𝒓𝒃𝒆𝒏𝒆

A Schlenk flask with a stir bar was charged with the carbene precursor (1.0 eq, 0.2-1 mmol) and pump
purged and dry bubble-degassed THF was added. 0.5 M potassium hexamethylsilylamide (KHMDS)
in THF (1.0 eq) was added and the reaction was stirred for 1 hour. AuS(CH3)2Cl (1.1 mmol) was added,
the flask covered in Aluminum foil and the reaction was stirred overnight.. The solution was filtered
through celite, washed with dichloromethane and the solvent was evaporated to almost dryness.
Hexanes were added and most of the solvent was removed, causing a colorless precipitate, which was
filtered and washed with solvents as needed. Yielding the 𝐴𝑢𝐶𝑙
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 in ~50% yield. For more detailed
synthesis and compound specific workup see ref93
.
General Synthesis for 𝑨𝒖𝒂𝒎𝒊𝒅𝒆
𝒄𝒂𝒓𝒃𝒆𝒏𝒆

A 25 mL Schlenk flask with a stir bar with the amide (1.1 eq, 0.3-1 mmol) was pump purged and bubble
degassed dry THF (10 mL) was added via cannula transfer. 2 M sodium tert-butoxide (NaOtBu) (1.1
eq) solution was added dropwise. After 1h stirring, 𝐴𝑢𝐶𝑙
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 (1.0 eq) was added and reaction was
stirred overnight. Filtration through Celite, washed with dichloromethane and solvent was removed.
Product was precipitated from dichloromethane by adding hexanes. Solid was filtered and washed if
necessary with a variety of solvents, yielding to the correcponding 𝐴𝑢𝑎𝑚𝑖𝑑𝑒
𝑐𝑎𝑟𝑏𝑒𝑛𝑒 in >50% yield. If used in
OLED devices, materials were sublimed in a high-vac sublimator around 230-260C at 10-7 torr. For
more detailed synthesis and compound-specific workup see ref93
.

36
Figure 2-11. Synthetic chart for all materials prepared in this study. Metal triflates were used
to perform ring closures of PrePAC and PreBZAC where M = Cu(II), Ag(I), and Na+
. (PAC)M’(X)
was achieved for M’ = Cu(I), Ag(I), and Au(I) where X = Cl- or BF4
-
. BZAC M’’(Cl) was achieved
for M’’ = Cu(I) and Au(I). (BZI)Au(D) was prepared for D = Bim, MBim, and Obim. (PAC)M’(D’)
and (BZAC)M’’(D’) were isolated where D’ = Cz, BCz, and Bim. Ar = 3,6 – diisopropylphenyl.

37
Chapter 3 - Janus carbenes
The following chapter was published in J. Am. Chem. Soc. 2023, 145, 36, 20097–20108.91 Dr. Jie Ma
and myself were responsible equally for Synthesis of compounds and photophysical characterization as
well as the OLED device fabrication. Sritoma Paul repeated our device structure on better tools, to
achieve higher performances. Dr Jie Ma solely was responsible for theoretical calculations. I, on the
other hand, conducted temperature dependent photophysics experiments as well as analyzed all
complexes via single crystal X-Ray diffraction. The publication was written in equal parts by Jie Ma
and me.
Introduction
The first observation of photoluminescent coinage metal (Cu, Ag, Au) complexes was in 1970,143
although they were rarely considered for practical applications because of the poor luminescence
efficiency of compounds known at that time. A resurgence in interest in the photophysics of coinage
metal complexes came with the report of highly luminescent two-coordinate carbene-metal-amide
(cMa) complexes of copper, silver and gold.2, 4, 7, 9, 10, 30, 33, 93, 113, 115, 116, 119, 138, 144-148 The cMa complexes
are composed of a carbene ligand serving as an electron acceptor (A) and an amide ligand as an electron
donor (D) bridged by the monovalent metal ion. By judicious choice of the two ligands, a metal
perturbed amide-to-carbene interligand charge transfer (ICT) is the lowest energy excited state, which
undergoes thermally assisted delayed fluorescence (TADF) with high photoluminescence efficiency
(PL). The high luminescence efficiencies of cMa complexes has enabled application in organic
electronics,9, 113
whereas the high redox potentials of the ICT state for the cMa complexes also make
them promising candidates as sensitizers for photocatalysis.124
Emission from TADF is observed when the lowest excited singlet (S1) and triplet (T1) state are close
enough in energy (ΔEST < 2000 cm-1
, 250 meV) that the two states are in dynamic equilibrium at room
temperature. The cMa complexes have a large separation (4 Å) between donors and acceptor ligands
which leads to a small energy difference between the S1 and T1 states, on the order of 100 meV.2, 144, 148

Rates for intersystem crossing (kISC) of over 1010 s
-1 have been observed for cMa complexes leading to
a rapid equilibrium between the S1 and T1 states.10
Considering the high PL values for the cMa
complexes, as well as the fast ISC rate, the excited state lifetime is approximated by the TADF rate. A
steady-state pre-equilibrium approximation allows the radiative lifetime of TADF (and thus the decay
rate of the excited state) to be equated to the product of the radiative rate from the S1 state (𝑘𝑟
𝑆1
) and the
equilibrium constant between these states Keq(T1 ⇄ S1), equation 3-1.44
𝑘𝑟
𝑇𝐴𝐷𝐹 = 𝑘𝑟
𝑆1
∙ 𝐾𝑒𝑞 (3-1)
Here, Keq depends on the value of ΔEST (eq. S1), which in turn is determined by the Coulombic repulsion
between the two unpaired spins in the S1 and T1 states. Spatial separation between the hole (h+
) and
electron (e-
) of the natural transition orbital (NTO) in the lowest excited states decreases ΔEST, thereby
increasing Keq. However, there is a trade-off when optimizing for both EST (Keq) and 𝑘𝑟
𝑆1
. The rate of
𝑘𝑟
𝑆1 depends on the transition dipole moment 𝜇𝑆0
𝑆1
, which is tied to the product of h+
and e- overlap in

38
the NTO (eq. S2). An increase in the h+
and eseparation (and concomitant decrease in orbital overlap)
will decrease EST (and consequently Keq) but also decrease 𝑘𝑟
𝑆1
.
Several research groups have performed experimental and theoretical investigations on monometallic
cMa complexes having various donors and acceptors with the goal to adjust 𝑘𝑟
𝑆1
and EST.
1, 93, 115, 138

For example, by selectively extending the -system of the carbene or/and amide ligands we prepared
cMa complexes with lifetimes ranging from 250 ns to 1 s.93, 138
In contrast, photophysical studies of
bimetallic cMa complexes applying this approach are scarce. A previous report from our group
described a bimetallic cMa complex prepared by extending a mono-nuclear carbene by addition of a
second metal-carbene, i.e. C:→Au−C:→Au-Cz where C: is a carbene acceptor, −C: is a bridging ditopic
donor-carbene, and Cz in an N-carbazolyl donor.145
This bimetallic complex has a small EST =
50 meV. However, the overlap between the donor-carbene and the amide is poor, leading to a slow 𝑘𝑟
𝑆1
and radiative lifetime for TADF of 0.5 s.
Scheme 3-1. Mono- and bimetallic cMa complexes.

39
To further investigate the trade-off between 𝑘𝑟
𝑆1
and EST, an alternative design strategy for bimetallic
cMa complexes is presented here. The complexes utilize Janus carbenes,149, 150 facially-opposed ditopic
ligands, as an acceptor, where each end is coordinated by a metal-amide donor moiety, providing a
general structure of amide-metal:carbene~carbene:→metal-amide (aMccMa), e.g. BAZ and BBI in
Scheme 1. The benefit of this symmetrical D-A-D structural motif is that 𝑘𝑟
𝑇𝐴𝐷𝐹 can be enhanced by an
increase in value of 𝑘𝑟
𝑆1 or a decrease in the value of ΔEST by increasing the h+
and eseparation.
151-153

Extending the excited states over two amides and a Janus carbene moiety is expected to increase the
oscillator strength of S1 state of the chromophore. A similar approach to increase 𝑘𝑟
𝑇𝐴𝐷𝐹 was reported
by Yersin, et al, invoking exciton coupling.
152, 154
Alternatively, if the molecular symmetry is broken in
the excited state, the bimetallic complex can still benefit from the extended electronic -system of the
Janus carbene, leading to a decrease in ΔEST relative to the monometallic analogs.93
Therefore, based
on the equation 3-1, the overall value for 𝑘𝑟
𝑇𝐴𝐷𝐹 will be enhanced by either mechanism.
Herein we report a series of bimetallic cMa complexes using Janus carbene ligands that have high PL
efficiencies (PL > 0.95) and lifetimes for TADF ( < 300 ns) that are about one third of their
monometallic analogs (Scheme 1). The fast TADF rates for the bimetallic complexes come about from
short radiative lifetimes from S1 to S0 states (𝜏𝑆1
12 ns) and a decrease in the exchange energies (EST
= 40 to 50 meV) relative to monometallic analogs. The absorption and emission spectra of these
bimetallic complexes are strongly solvent dependent, indicating that two-fold symmetry is lost in the
excited states of these materials. Lastly, two of the bimetallic complexes were used as emissive dopants
in solution-processed organic light emitting devices (OLEDs) that display high quantum efficiency with
low roll-off at high brightness.
Synthesis
The chemical structures of two Janus carbenes used here, 1,3,6,8-tetrakis(2,6-diisopropylphenyl)-
3,4,8,9-tetrahydropyrimido[4,5-g]quinazoline-1,6-diium (BAZ) and 3,4-dimethyl-1,7-bis(2,6-
diisopropylphenyl)benzobis(imidazolium) iodide (BBI) are shown in Scheme 1. The synthesis of the
BAZ ligand was adapted from literature procedures,149 whereas BBI ligand was prepared by modifying
a protocol of Bielawski, et al..
149, 150 The bulky BBI imidazolate was prepared by treating
dichlorodinitrobenzene with an excess of 2,6-diisopropylaniline (DIPA) as both reactant and solvent
and heated to 150 C overnight. The metal chloride precursor complexes were prepared by treatment
of BAZ ligand with potassium bis(trimethylsilyl)amide followed by reaction with (MeS)2AuCl,
whereas the BBI gold chloride complex was prepared using a weaker base (Ag2O) to prevent
polymerization of the BBI ligand (Scheme 2). The final cMa complexes were made using three various
deprotonated amides: N-carbazole (Cz), 3,6-di-tert-butylcarbazole (BCz) and Nbenzo[d]benzo[4,5]imidazo[1,2-a]-imidazolyl (bim). Their corresponding monometallic complexes
were synthesized in an analogous procedure using either 1,3-bis(2,6-diisopropylphenyl)-3,4-
dihydroquinazolin-1-ium (BZAC) or 1,3-bis(2,6-diisopropylphenyl)benzo-imidazolium (BZI) acceptor
carbenes. The details of the synthesis and characterization of the complexes are given at the end of
Chapter 3.

40
The bimetallic complexes are significantly less soluble than their monometallic analogs. For example,
complexes with unsubstituted carbazole ligands are soluble in a moderately polar solvent like THF but
insoluble in both polar solvents, such as DMSO and MeCN, as well as nonpolar solvents, such as
cyclohexane and toluene. Substitution of tert-butyl groups onto the 3,6-positions of carbazolyl (BCz)
markedly improves the solubility of the complexes in all solvents, although BCzAuBAZAuBCz remains less
soluble than BCzAuBBIAuBCz. Thus, discussion of electrochemical and photophysical properties on these
complexes will be focused on the complexes with BCz and bim ligands (BCzAuBAZAuBCz, BCzAuBBIAuBCz
and bimAuBAZAubim) and their corresponding monometallic analogues (BZACAuBCz,
BZIAuBCz and
BZACAubim).
Crystallographic Analysis
Single-crystal X-ray structures were determined for bimetallic complexes with BAZ coordinated to Cz,
BCz and bim amides, as well as for the monometallic complexes BZACAuBCz and BZIAuBCz. Unfortunately,
X-ray quality crystals could not be obtained for bimetallic complexes using the BBI carbene.
Crystallographic data are given in
Table 3-1, and have been deposited in the Cambridge Crystallographic Data Centre (CCDC numbers:
2206141, 2233511, 2206146, 2216913 and 2216906). Representative structures of the compounds of
BCzAuBAZAuBCz and bimAuBAZAubim are shown in Figure 3-1. The bimetallic complexes crystallize into
centrosymmetric space groups with the inversion center located in the middle of the bridging aromatic
ring of the carbene. There are no significant differences in metrical parameters between the bimetallic
complexes and the related bonds in the monometallic cMa analogs.4, 10, 93 All complexes display near
linear two-coordinate geometries (Ccarbene-Au-Namide = 175177°) along with near equal Au-Ccarbene and
Au-Namide bond lengths (1.989-2.020 Å) (see Table 3-1). The sum of angles around Ccarbene and Namide
are 360°, indicating a trigonal planar geometry for the ligated atoms. The bimetallic complexes
crystallize with the carbene and both amides in a near coplanar conformation as dihedral angles for
these ligands vary between 0° to 5°. The amide ligands in bimAuBAZAubim are oriented antiparallel to
each other, with the longer side of the amide opposite to the methylene group of the BAZ carbene
(Figure 3-1c). A notable S-shaped curvature is also apparent in the molecular plane of the complex.
This distortion is attributed to crystal packing forces as a planar arrangement of carbene and amides is
found in geometry optimized calculations for the same complex.
Scheme 3-2. Synthesis of bimetallic cMa complexes

41
Table 3-1. Selected X-ray crystallographic data.
Compound
Bond length Bond angle
Ccarbene-AuNamide (˚)
Dihedral angle
Ccarbene-AuNamide (˚)
Angles
Ccarbene-Au
(Å)
Au-Namide
(Å)
Ccarbene / Namide
(˚)
BCzAuBAZAuBCz 1.989(4) 2.007(4) 177.0(2) 5.5 360 / 360
BZACAuBCz 2.009(4) 2.013(4) 176.8(2) 0.6 360 / 358
CzAuBAZAuCz 1.995(9) 2.020(7) 177.9(3) 0.6 360 / 358
BZIAuBCz 1.995(5) 2.027(4) 175.6(2) 10.4 360 / 359
bimAuBAZAubim 1.993(4) 2.017(3) 175.4(2) 5.4 360 / 360

Figure 3-1. Crystal structures of (a) BCzAuBAZAuBCz , (b) BCzAuBAZAuBCz and (c) bimAuBAZAubim
(a) (b)
(c)

42
Computational Results
The electronic structure of the mono- and bimetallic complexes were examined using density functional
theory (DFT) and time-dependent DFT (TD-DFT) calculations, with B3LYP/LACVP and
cam-B3LYP/LACVP methods, respectively. Simplified carbenes substituted with methyl groups (BBI
and BZI), in place of bulky dipp moieties, were used to streamline the calculations (Figure 3-2) The
CzAuBBIAuCz complex was optimized using D2h symmetry, and C2V symmetry was applied for the
BZI’AuCz complex. The results for the related derivatives, CzAuBAZAuCz and bimAuBAZAubim, optimized
under Ci symmetry, along with their corresponding monometallic complexes, follow the same trends.
More detailed information can be found in the published paper.
91
The CzAuBBIAuCz complex has a HOMO and HOMO-1 comprised of in phase and out of phase
combinations of the carbazole -orbitals that are similar in energy (-4.57 and -4.62 eV, respectively,
Figure 3-2). In contrast, the energy of HOMO-1 for the monometallic analog is roughly 0.6 eV deeper
than the HOMO and has no contribution from the nitrogen of the carbazole (Figure 3-2 right). The
lowest unoccupied molecular orbital (LUMO) of CzAuBBIAuCz is delocalized throughout the bridging
carbene ligand and stabilized by more than 1 eV relative to the LUMO for BBIAuCz. The electronic
dipole moments for ground and excited states for CzAuBBIAuCz in D2h symmetry are zero. Therefore,
Figure 3-2. Top: frontier molecular orbitals for CzAuBBIAuCz (left) and BZI’AuCz (right). The isovalue
are set to 0.1. The orbital contributions, energies (oscillator strength) and magnitude of the electronic
dipole moment are given for the ground and Sn states below the structures. The direction of the
dipole moment for each state is indicated with the arrow.

43
the hole density of the S1 and the lowest triplet excited state (T1) states distribute evenly over both
amides. The high symmetry of the bimetallic complex has a significant influence on the oscillator
strength for the ICT state. There are two ICT transitions in CzAuBBIAuCz, an S1 state with large oscillator
strength (ƒ = 0.62) and a symmetry forbidden S2 state (ƒ = 0). Notably, the oscillator strength for the S1
state of the bimetallic complex is three times larger than that for the S1 state of BBIAuCz.
Rotation about the metal-ligand bond can break the two-fold molecular symmetry in bimetallic
complexes that markedly alters their electronic properties. Thus, the geometries of the bimetallic
complexes in the ground state were also optimized without symmetry. Asymmetry was imparted by
constraining one dihedral angle between the carbene and the amides (Cz or bim) at 0 while twisting
the other dihedral angle () away from 0º. The energies and compositions for the S1 state are not
significantly affected with twist angles up to 20. However, when  increases to 30 the hole of the S1
state becomes localized on the twisted Cz, the oscillator strength decreases to 0.45, and the net dipole
moment for the S1 state increases to 16 D. The magnitude for the dipole moment is comparable to the
value for the monometallic analog and indicative of long-range charge transfer in the excited state
(Figure 3-2).
2, 148 The oscillator strength of the S1 state decreases as the dihedral angle increases to 90º
at which point the value drops to zero (Figure 3-3) In brief, a large angle for  (> 20 for the BAZ
based complexes and 30 for the BBI based complex) leads to a low oscillator strength and a large
increase in the net dipole moment for the S1 state of the bimetallic complexes.
To evaluate the energy barrier for rotation about the metal-ligand bond, potential energy surface (PES)
calculations were performed using the B3LYP/LACVP method that included a DFT-D3(BJ) dispersion
correction. Both mono- and bimetallic complexes with dipp groups were examined by varying the
dihedral angles between a carbene and amide from 0° to 180°. The results for these calculations are
shown in Figure 3-4. Energy barriers for rotation from 0° to 180° are similar for corresponding
monometallic and bimetallic complexes. The energy barriers for all the derivatives remain low
(<1.6 kcal/mol) until dihedral angles reach 30°, which allows facile libration about the metal-ligand
bond. Such geometric twisting of the metal-ligand bonds breaks the two-fold molecular symmetry in
the bimetallic complexes, particularly in a fluid medium. It is noteworthy that the energy barriers for
BZACAubim and bimAuBAZAubim are the lowest among cMa complexes despite the presence of bulky dipp
Figure 3-3. Calculated S1, S2 (with oscillator strength) and T1, T2 energies of CzAuBBIAuCz (left)
and BZIAuCz (right) with respect to their dihedral angles 
0 30 60 90 120 150 180
2.0
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8 Energy (eV) Dihedral Angle (°)
S1
S2
T1
T2 0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
f for S1
f for S2 Osillator Strength
CzAuBBIAuCz
0 30 60 90 120 150 180
2.4
2.6
2.8
3.0
3.2
3.4
3.6
3.8
4.0 Energy (eV) Dihedral Angle (°)
S1
S2
T1
T2
BZIAuCz
0.0
0.2
0.4
0.6
0.8
1.0
f for S1
f for S2 Osillator Strength

44
groups on the BAZ ligand. The energy barrier is 5.4 kcal/mol for CzAuBAZAuCz but only 1.5 kcal/mol
for bimAuBAZAubim. The significant decrease in the rotation barriers for cMa complexes with the bim
ligand is due to a combination of lower steric hinderance and loss of an attractive edge-to- interactions
present between the C-H bonds of the carbazole and dipp moieties. An intermediate energy barrier for
the CzAuBBIAuCz complex of 2.8 kcal/mol is attributed to the absence of one dipp group on each side of
carbene and reduced steric interactions imparted by the narrow N-C-N angle of the five-membered ring.
Electrochemistry
The electrochemical properties of the cMa complexes were examined using cyclic and differential pulse
voltammetry (Figure 3-5), and their redox potentials relative to an internal ferrocene reference are listed
in
Table 3-2. The measurements were performed in N,N-dimethylformamide (DMF) solution except for
the BAZ based complexes which were carried out in tetrahydrofuran (THF) due to poor solubility in
DMF. Unfortunately, the reduction wave of BAZ based complexes overlaps with the THF solvent
window and obscures the reduction peak. Thus, the reduction potential for ClAuBAZAuCl measured in
DMF was used to estimate the value for BCzAuBAZAuBCz and bimAuBAZAubim. Previous studies have
shown that energies for the LUMO in related two-coordinated complexes are determined by the identity
of the carbene and metal ion, but independent of anionic ligand.
Both mono- and bimetallic complexes exhibit reversible anodic waves at similar potentials, ranging
from 0.1 V to 0.2 V, that are assigned to oxidation of BCz or bim ligands. The absence of a second
oxidation wave for the bimetallic complexes within the potential window of the solvent indicates that
the amide ligands are electronically uncoupled. In contrast, reduction waves are irreversible, except
for BZIAuBCz. The potentials for the bimetallic complexes are shifted to less negative values by 0.3 V to
0.4 V than their monometallic counterparts (Table 3-2). This difference is due to stabilization from the
Figure 3-4. Potential energy surface scan for ligand rotation of the complexes. BZI is the
carbene substituted with a methyl group in place of one dipp moiety.
0 30 60 90 120 150 180
0
1
2
3
4
5
6
BZACAuCz BZIAuCz BZI"AuCz
BZACAubim Energy barrier (kcal/mol)
Dihedral angle (°)
CzAuBAZAuCz CzAuBBIAuCz bimAuBAZAubim

45
extended conjugation in the -system of the Janus carbene ligands and replicated in the LUMO energies
calculated for both types of species.
Table 3-2. Electrochemical data. Measurements were performed using 0.1 M TBAPF6 electrolyte in
DMF (except where noted), and the potentials are listed relative to a ferrocene internal reference.
compound oxidation
(V)
reduction
(V)
redox gap
(V)
HOMOd
(eV)
LUMOd
(eV)
BCzAuBAZAuBCz 0.07 a b
2.65
-4.87 -
ClAuBAZAuCl - -2.58 - -1.79
BZACAuBCz
c 0.15 c
-2.83 c 2.95 -4.96 -1.49
BCzAuBBIAuBCz
d 0.20 -2.44 2.64 -5.02 -1.95
BZIAuBCz
d 0.21 -2.85 3.06 -5.03 -1.47
bimAuBAZAubim 0.22 a b 2.80 -5.04 -1.79
BZACAubim 0.31 c
-2.69 c 3.00 -5.15 -1.66
a
in THF; b
insoluble in DMF; c
from reference93; d
calculated using the equations: HOMO
= -1.15(Eox) − 4.79; LUMO = -1.18(Ered) – 4.83 according to reference5
.

46
Photophysical properties
The UV-visible absorption and emission spectra of the complexes were recorded in a toluene solution
(Figure 3-6). The absorption spectra of all complexes display structured bands at high energy (BCz: λ
< 350 nm, bim: λ < 380 nm) that are assigned to -

transitions localized on the carbene and amide
ligands. Absorption maxima at 336 and 352 nm in BCzAuBAZAuBCz are assigned to -* transitions on
the carbene as absorption bands at the same wavelength are found in the precursor complex,
ClAuBAZAuCl. Broad featureless absorption bands at lower energy (λ  350 nm) are assigned to
transitions from the ICT state. The ICT transitions in the bimetallic complexes are at lower energy than
the monometallic analogs consistent with the stabilization of the LUMO observed in the reduction
potentials of complexes. Luminescence at room temperature from all complexes is broad and
Figure 3-5. CV (black) and DPV (red: oxidation, blue: reduction) traces of complexes and
decamethylferrocene collected in THF or DMF or MeCN with 0.1 M TBAPF6 as an electrolyte. The
asterisk indicates the redox peak of decamethylferrocene. The electrochemical plot at the bottom shows
a comparison of Fc and DMF.
-1.0 -0.5 0.0 0.5
-60
-40
-20
0
20
Current (µA)
Potential (V vs. Fc+/0)
BCzAuBAZAuBCz
*
-2.5 -2 -1.5 -1 -0.5
-100
-80
-60
-40
-20
0
20
40
Current (µA)
Potential (V vs. Fc+/0)
ClAuBAZAuCl
*
- 1.0 - 0.5 0.0 0.5
- 40
- 30
- 20
- 10
0
10
Current (µA)
Potential (V vs. Fc+/0)
bimAuBAZAubim
*
-2.5 -2 -1.5 -1 -0.5 0 0.5
-60
-40
-20
0
20
Current (µA)
Potential (V vs. Fc+/0)
BCzAuBBIAuBCz
*

47
featureless in all solvents except MeCy, where emission is vibronically structured (Figure 3-6 and
Figure 3-8) as found in previous literature.10, 30, 33
The degree of electronic coupling between the amides and Janus carbene in the bimetallic complexes
can be assessed from the intensity of the absorption band for the ICT transition.155 For example, the
molar absorptivities for the ICT transitions in the bimetallic complexes with BCz donors are larger than
those in the monometallic analogs. Integration of the low energy half of the ICT bands (from ICT max
till low energy baseline) show an increase by a factor 1.9 for BCzAuBAZAuBCz and 1.3 for BCzAuBBIAuBCz.
In contrast, the ICT bands for BZACAubim and bimAuBAZAubim have similar molar absorptivities (Figure
3-6). The intensities follow the same order as the energy barrier to rotation calculated in the PES scan
(BCzAuBAZAuBCz > BCzAuBBIAuBCz > bimAuBAZAubim). This correlation suggests that a high percentage of
BCzAuBAZAuBCz molecules have both amides and carbene ligands in a near coplanar geometry since the
oscillator strength for the ICT transition is markedly lower for highly twisted geometries. Conversely,
the low energy barrier to rotation for bimAuBAZAubim leads to a high percentage of molecules with the
two amides in a random orientation, thus mitigating any enhancement in molar absorptivity from
electronic coupling.
The photostability of each of the bimetallic complexes and their monometallic analogs were tested in
degassed toluene using 375 nm excitation (see Figure 3-7). The complexes with BZAC and BAZ
carbenes displayed similar photostability, undergoing less than 10% decrease in absorbance at the max
Figure 3-6. Absorption and emission spectra of mono- and bimetallic cMa complexes with
carbazole (a, b) and bim (c, d) in toluene.
35000 30000 25000 20000
0
1×104 2×104 3×104 4×104 5×104 Molar Absorptivity (M-1cm-1
)
Energy (cm-1)
BZIAuBCz BCzAuBBIAuBCz BZACAuBCz BCzAuBAZAuBCz
(a) 300 350 400 450 500 550 Wavelength (nm)
25000 20000 15000
0.0
0.5
1.0 Photoluminisence (AU)
Energy (cm-1)
BZIAuBCz BCzAuBBIAuBCz BZACAuBCz BCzAuBAZAuBCz
(b) 400 500 600 700 800 Wavelength (nm)
35000 30000 25000 20000
0.0
5.0×103 1.0×104 1.5×104 2.0×104 Molar Absorptivity (M-1 cm-1
)
Energy (cm-1)
BZACAubim
BimAuBAZAuBim
(c) 300 400 500 Wavelegth (nm)
25000 20000 15000
0.0
0.5
1.0 Photoluminisence (AU)
Energy (cm-1)
BZACAubim
bimAuBAZAubim
(d) 400 500 600 700 800 Wavelength (nm)

48
of the ICT band after irradiation
for 24 h. For comparison, facIr(ppy)3 also showed a 10%
decrease in the max for the
absorption band (380 nm) after
24 h under the same conditions.
The complexes with BZI and BBI
carbenes decomposed much faster
as the absorbance of
BCzAuBBIAuBCz complex at the ICT
max decreased 20%, and a 40%
decrease was observed for
BZIAuBCz after 24 hours of
irradiation.
Solvatochromism is a
characteristic feature for the
optical properties of monometallic
cMa complexes owing to large differences in the magnitude and direction of the dipole moments for
the ground and excited states.2, 10 This effect gives rise to large hypsochromic shifts in absorption and
smaller bathochromic shifts for emission with increasing polarity of solvents. This response to solvent
polarity is also displayed in bimetallic complexes (Figure 3-8(a), exemplarily for BCzAuBAZAuBCz). The
energy of the absorption band undergoes a larger solvatochromic shift than does the emission band.
This effect can be explained by the solvent reorganization energy being greater in the excited state upon
absorption than in the ground state upon emission (see Figure 3-8(b)). To evaluate the solvatochromic
shift in the ground and excited states, absorption and emission maxima for BZACAuBCz and
BCzAuBAZAuBCz are plotted against the solvent polarity indexed using the ET(30) scale in Figure
3-8(c).
156
The absorption and emission values for each compound were fit using a linear least squares
routine and the slopes are listed in Figure 3-8(d). All other complexes follow the same trend and their
data can be found in the supplementary information of the publication.91 The absorption and emission
maxima of all compounds show a linear dependence with respect to the solvent polarity, indicating that
emission originates from a common state in all solvents. The absolute slope for absorption is larger than
the one for emission, which is true for all cMa complexes. The slopes of absorption and emission of the
bimetallic complexes in the ET(30) plots parallel the slopes of their monometallic analogs, indicating
the strength of the transition dipoles are similar for both absorption and emission in mono- and
bimetallic complexes. Note that if the bimetallic complex remains symmetric in the ground and excited
state, the modeling studies described above predict solvatochromism to be negligible. However, the
strong solvatochromic response is indicative of an asymmetric geometry, since rotation of the ligand(s)
will cause the excited state to localize largely on only half of the complex. This rotation leads to a large
Figure 3-7. Photostability of all complexes and Ir(ppy)3
as reference in degassed toluene. solutions were in an airfree Schlenk cuvette and were excited with a 375nm LED
lamp (1450 mW, 19.2 W mm-2
)
0 4 8 12 16 20 24
40
50
60
70
80
90
100
CT Band decay (%)
Time (h)
BBI-Au-BCz
BZI-Au-BCz
BAZ-Au-BCz
BZAC-Au-BCz
BAZ-Au-Bim
BZAC-Au-Bim
Ir(ppy)3

49
dipole moment for the ICT state, which is similar in magnitude to the value found for the monometallic
analog.
Emission spectra from cMa complexes are distinguished by large rigidochromic shifts upon going from
fluid solutions to frozen media at low temperatures. Likewise, luminescence spectra for the bimetallic
complexes recorded in fluid 2-MeTHF and methylcyclohexane (MeCy) change markedly upon cooling
solutions to frozen glass at 77 K (Figure 3-9(left)). The featureless ICT emission bands from
BCzAuBAZAuBCz and BCzAuBBIAuBCz blueshift at 77 K and are replaced by narrow, highly structured
bands, which are assigned to phosphorescence from a triplet excited state localized on the BCz ligand
(
3LE). This change occurs because the immobile solvent molecules freeze around the ground state
dipole, destabilizing the ICT state so much that it rises above the 3LE state.2, 10
Destabilization of the
ICT state occurs to a lesser extent in a polystyrene (PS) matrix at 77 K, leading only to a small blue
shift in the ICT emission band. Similarly, the bimAuBAZAubim complex displays a narrow, vibronically
structured band at 430 nm in 2-MeTHF, MeCy and a PS matrix at 77 K (Figure 3-9(left)). However,
this emission band in this derivative is assigned to a 3LE state localized on the BAZ ligand. The energy
Figure 3-8. (a) Absorption and emission spectra of BCzAuBAZAuBCz in various solvents. (b)
Qualitative potential surfaces for ground and excited states in solvents polarities. (c)
Absorption and emission maxima vs solvent polarity (ET(30) scale) for BCzAuBAZAuBCz and
BZACAuBCz. (d) Slopes of absorption and emission maxima vs ET(30).
30000 25000 20000 15000
0.0
0.2
0.4
0.6
0.8
1.0
1.2 Absorption (arb.U.)
Energy (cm-1)
MeCyHex Toluene MeTHF
CH2Cl2 MeCN
(a)
BCzAuBAZAuBCz 400 500 600 700 800 Wavelength (nm)
(d)
(b)
30 32 34 36 38 40 42 44 46
1.6
1.8
2.0
2.2
2.4
2.6
2.8
3.0
Abs Em
Abs Em Energy (x10
4 cm-1
)
Solvent Polarity Index
BCzAuBAZAuBCz
(c)
BZACAuBCz

50
of the 3LE state for the bim ligand is higher (E00 = 365 nm) than BAZ and therefore the amide does not
contribute to emission.
The solvatochromic effects in rigid media at room temperature were examined in polymer matrices.
The series ZEONEX® (fully aliphatic copolymer of ethylene and norbornene), PS and
polymethylmethacrylate (PMMA) go from nonpolar to moderately polar. The absorption and emission
maxima of the CT band for all mono- and bimetallic complexes blue shift upon going from nonpolar
(ZEONEX®) to weakly polar PS to moderately polar (PMMA). The Stokes shift between absorption
and emission maxima for all complexes is smallest in ZEONEX and largest in PMMA (Figure
3-9(right)). The blue shift in absorption in PMMA is consistent with the polymer organizing in response
to the ground state dipole moment on deposition, leading to a destabilization the excited state. Unlike
a fluid solvent, the rigid polymer matrix cannot rearrange at room temperature so the “solvent”
environment around each emitter remains fixed, leading to a small Stokes shift and a net blue shift in
emission relative to the nonpolar ZEONEX® and PS matrices.
An attractive feature of the cMa complexes is their high photoluminescence quantum yields and short
emission lifetimes for TADF ( < 1 s). The bimetallic complexes, like monometallic complexes, can
have near unity quantum yields in nonpolar solvents. Quantum yields and emission lifetimes decrease
with increasing solvent polarity (Table 3-3). Similarly, radiative rates decrease in more polar solvents.
The rate for non-radiative decay in all bimetallic complexes is greater than that of the monometallic
complexes. The enhanced nonradiative decay rates for bimetallic complexes are likely due to the
additional rotational degrees of freedom introduced by the second metal-amide ligand rotor and lower
energy for emission (energy gap law).154, 157, 158 Importantly, the bimetallic complexes with BCz donor
ligands have radiative rates for TADF that are significantly faster than their monometallic analogs
(Table 3-3). Hence, BCzAuBAZAuBCz and BCzAuBBIAuBCz complexes have radiative rates around 34
 

s
-1
, whereas the highest radiative rates of monometallic analogs are near 1.5  

s
-1
. Both
bimetallic complexes also have near unit quantum yields in PS films (PL ≥ 0.9) (Table 3-4). In contrast,
the radiative rate for bimAuBAZAubim in toluene is only marginally faster with respect to its monometallic
Figure 3-9. (left) Emission spectra of BCzAuBAZAuBCz in PS, MeTHF and MeCyHex at room
temperature (solid) and 77K (doted). (right) Absorption and Emission spectra of BCzAuBAZAuBCz
in different polymer matrixes (PMMA, PS and Zeonex).
400 450 500 550 600 650 700 750
0
0.2
0.4
0.6
0.8
1 Photoluminisence BAZ-Au-BCz (AU)
Wavelength (nm)
PS
PS 77K
MeTHF
MeTHF 77K
MeCyHex
MeCyHex 77K
BCzAuBAZAuBCz
30000 25000 20000 15000
0.0
0.5
1.0
Photoluminisence (AU)
Energy (cm-1)
BCzAuBAZAuBCz
PMMA
PS
ZEONEX
350 400 450 500 550 600 650
Wavelength (nm)

51
analog. The photoluminescence quantum yields in PS films for this derivative are also lower than the
other bimetallic derivatives (PL =0.76).
Table 3-3. Photophysical data for mono- and bimetallic cMa complexes in solution
Complexes
Abs
λmax
(nm)
PL
λmax
(nm)
ΦPL
(%)
τ
(s)
kr
(106
s
-1
)
knr
(106
s
-1
)
λmax
77K
(nm)
τ77 K
(s) (%)
MeCy
BZIAuBCz 478 514 93 0.370 2.5 0.19 480 60
BCzAuBBIAuBCz 430 495 >95 0.637 1.5 <0.01 430 92 (47)
390 (53)
BZACAuBCz 495 530 >95 0.254 3.8 <0.01 475 12.84
BCzAuBAZAuBCz 423 452 90 0.723 1.2 0.14 438 265
bimAuBZAC 435 468 >95 0.265 3.7 0.11 451 790 (62)
273 (38)
bimAuBAZAubim 385 454 >95 0.290 3.4 0.07 400 259
Toluene
BZIAuBCz 450 525 90 0.33 2.7 0.3 - -
BCzAuBBIAuBCz 410 500 >95 0.56 1.7 <0.01 - -
BZACAuBCz 455 535 >95 0.24 3.9 <0.01 - -
BCzAuBAZAuBCz 400 475 94 0.64 1.5 0.09 - -
bimAuBZAC 410 505 78 0.29 2.7 0.8 - -
bimAuBAZAubim 380 480 >95 0.43 2.3 <0.01 - -
2-MeTHF
BZIAuBCz 426 544 83 0.420 2.0 0.4 442 250 (44)
530 (56)
BCzAuBBIAuBCz 395 515 88 0.664 1.3 0.18 430 533(0.35)
875(0.65)
BZACAuBCz 426 551 89 0.278 3.2 0.4 438 80 (25)
307 (75)
BCzAuBAZAuBCz 375 485 97 0.868 1.1 0.03 438 358
bimAuBZAC 388 522 63 0.327 1.9 1.1 442 563
bimAuBAZAubim 363 494 78 0.310 2.5 0.71 400 467
MeCN
BZIAuBCz 360 550 75 0.371 2.0 0.67 - -
BCzAuBBIAuBCz 375 525 69 0.801 0.86 0.39 - -
BZACAuBCz 375 600 34 0.139 2.4 0.47 - -
BCzAuBAZAuBCz 350 510 65 2.82 0.23 0.12 - -
bimAuBZAC 367 540 30 2.468 0.12 0.28 - -

52
Table 3-4. Photophysical data for mono- and bimetallic cMa complexes in different polymer matrices.
bimAuBAZAubim 311 522 42 - - - - -
CH2Cl2
BZIAuBCz 410 545 89 0.374 2.3 0.29 - -
BCzAuBBIAuBCz 375 515 81 0.664 1.2 0.29 - -
BZACAuBCz 415 580 77 0.274 2.8 0.84 - -
BCzAuBAZAuBCz 370 500 66 1.80 0.36 0.19 - -
bimAuBZAC 375 520 80 0.581 1.4 0.34 - -
bimAuBAZAubim 350 494 83 0.325 2.6 0.52 - -
Complexes
Abs
λmax
(nm)
PL
λmax
(nm)
ΦPL
(%)
τ
(s)
kr
(106
s
-1
)
knr
(106
s
-1
)
λmax
77K
(nm)
τ 77 K
(s) (%)
Zeonex
BZIAuBCz 415 454 92 0.777 1.2 0.1 - -
BCzAuBBIAuBCz 438 518 74 0.24 (0.58)
1.04 (0.42) - - - -
BZACAuBCz 426 489 95 0.532 1.8 0.1 - -
BCzAuBAZAuBCz 470 510 69 0.325 2.1 1.0 - -
bimAuBZAC 386 452 99 0.250 4.0 0.04 - -
bimAuBAZAubim 419 472 80 0.325 2.5 0.6 - -
PS
BZIAuBCz 402 460 94 1.58 (0.5)
8.98 (0.5) - - 445 65 (0.3)
219 (0.7)
BCzAuBBIAuBCz 455 515 93 0.21 4.3 0.33 505 25
BZACAuBCz 409 484 >95 0.72 1.4 <0.01
BCzAuBAZAuBCz 450 506 90 0.30 3 0.33 506 50
bimAuBZAC 374 452 >95 0.28 3.7 <0.01 - -
bimAuBAZAubim 407 475 76 0.33 (0.8)
2.85 (0.2) - - 455 27 (0.2)
346 (0.8)
PMMA
BZIAuBCz 368 447 87 34.7 (27%)
261 (73%) - - - -
BCzAuBBIAuBCz 420 471 48 0.29 (54%)
1.41 (46%) - - - -

53
To probe the origin of the fast radiative rates for TADF in the bimetallic complexes, temperature
dependent photophysical measurements were carried out from 4 K to 300 K in doped PS films (for
details on this measurement see Chapter 5). An Arrhenius model for emission decay was used to fit data
in the temperature region of 200-300 K, where TADF is the dominant mechanism emission, to extract
values for 𝑘𝑆1
and EST (Table 3-5). At temperature below 200 K, the emission decay at each
temperature was also fit to a three-level Boltzmann model (0), which gives the zero-field splitting (ZFS)
and the radiative rate for the T1 state. Data for BZIAuBCz is not included in Table 3-5 because the 3
ICT
state and 3LE state on the BCz ligand have energies close enough to only give biexponential emission
decays traces that change in contributions upon cooling to cryogenic temperatures (Figure 3-9). Data
for bimAuBAZAubim is included, despite also having transient decay traces in PS films that required
biexponential fits. However, as opposed to BZIAuBCz, a second longer-lived component contributed
steadily by 20% to the overall decay trace from 200-300 K. This allowed us to assume that the decay
mechanism is invariant over this temperature range. Thus, only the first (faster) decay trace was used
for data evaluation of this complex as TADF is the sole mechanism for this process.
Comparison of the 𝑘𝑆1
and EST parameters allows us to determine the origin of the fast rates for TADF
found for the bimetallic complexes (Table 3-3 & Table 3-4). Both mono- and bimetallic complexes
have similar values for 𝑘𝑆1
, whereas values of EST for the latter derivatives are found to be much
smaller (EST = 70, 54 and 43 meV for BZACAuBCz, BCzAuBAZAuBCz and BCzAuBBiAuBCz, respectively).
Thus, the high radiative rate for the bimetallic complexes is mainly attributed to the significant increase
in Keq induced by the small EST imparted by the -extended Janus carbene (see eq 3-1). This
conclusion is consistent with the calculated distance between the centers of positive and negative charge
(d(h+
, e-
)) in the S1 state. We recently determined the relationship between d(h+
, e
-
) and 𝑘𝑟
𝑇𝐴𝐷𝐹 for related
monometallic cMa complexes.93 In general, a larger separation of h+
and eleads to a smaller EST, and
therefore a faster rate for 𝑘𝑟
𝑇𝐴𝐷𝐹
. The d(h+
, e-
) values calculated for BCzAuBAZAuBCz and BCzAuBBIAuBCz
(6.17 and 6.87 Å, respectively), are larger than for BZACAuBCz (5.19 Å) (Figure 3-10), which is consistent
with a decrease in EST for the bimetallic complexes. The details for calculating d(h+
, e-
) are included
in the Supporting Information.
The value of 𝑘𝑟
𝑇𝐴𝐷𝐹 for bimAuBAZAubim complex is not faster than the mono-nuclear analog despite
having a value for EST of 20 meV, one of the smallest EST values reported to date for 2-coordinate
TADF complexes.7, 9, 10, 93, 115 The difference in EST values is consistent with the larger value of d(h+
,
e
-
) for bimAuBAZAubim (6.29 Å) than for BZACAubim (5.39 Å) (Figure 3-10); however, the S1 state for
bimAuBAZAubim has a slower radiative lifetime (𝜏𝑆1
= 50 ns) than does BZACAubim (𝜏𝑆1
= 19 ns). These
parameters likely originate from the low energy barrier to rotation for the bim ligand (Figure 3-4). The
BZACAuBCz 387 458 100
3.24 (22%)
20.2 (33%)
11.8 (45%)
- - - -
BCzAuBAZAuBCz 424 487 69 0.36 (65%)
1.40 (35%) - - - -
bimAuBZAC 359 440 95 0.280 - - - -
bimAuBAZAubim 388 470 79
1.27 (15%)
10.3 (36%)
68.2 (50%)
- - - -

54
presence of two rotation centers in bimAuBAZAubim increases the probability that at least one amide ligand
is rotated out of plane in the molecule. The large fraction of molecules with amide ligands in highly
twisted conformations will lead to an overall increase in 𝜏𝑆1
. Thus, the opposing effects of a decrease
in EST and increase in 𝜏𝑆1
results in similar radiative rates of TADF for BZACAubim and bimAuBAZAubim.
Table 3-5. Energy and rate data from variable temperature photophysical measurements for mono- and
bimetallic cMa complexes in polystyrene films (1 wt%).
𝜏𝑆1
(ns)
 9%
EST
(meV / cm-1
)
 3%
𝜏𝑇1
(s)
 5%
ZFS
(meV / cm-1
)
 10%
h
+
/e- pair distance in
S1 state (Å)
BZIAuBCz
a a a a a
BCzAuBBIAuBCz 12.6 43 / 346 33.1 1.1 / 9 6.87
BZACAuBCz 14.0 70 / 564 46.7 1.0 / 8 5.19
BCzAuBAZAuBCz 12.7 54 / 432 32.9 0.9 / 7 6.17
BZACAubim 19b 41 / 330b 19b 1.2 / 10b 5.39
bimAuBAZAubim 50c 20 / 161c d d 6.26
a Luminescence from the 3LE state below room temperature contributed to the lifetime and was changing contribution
upon cooling, preventing a determination of all values.
b Data from reference
c Biexponential decay is observed due to aggregation, even at concentrations <0.1 wt-%. Only the fast component was
used to determine EST and 𝜏𝑆1
as the relative amplitude stayed constant at 0.65 in the measured temperature range.
d Luminescence from the 3LE state on BAZ at temperatures below 150 K prevented accurate fits of the 3
ICT parameters
to determine 𝜏𝑇1
and ZFS.
6.17 Å 6.87 Å 6.29 Å

5.19 Å 5.30 Å 5.39 Å
Figure 3-10. Center of h+(yellow) and ,e- (green) for S1 NTOs for cMa complexes. The d(h+
,
e
-
) calculation on the geometry of complexes BCzAuBAZAuBCz, BCzAuBBIAuBCz and bimAuBAZAubim
is zero because the centers of h+
and eare overlapped because of the symmetry. To avoid this
situation, these complexes’ d(h+
, e-
) calculation was done by swapping one amide ligand to a
Cl- which are BCzAuBAZAuCl, BCzAuBBIAuCl and bimAuBAZAuCl.

55
OLEDs
The high photoluminescence efficiencies and radiative rates of the bimetallic complexes make them
promising candidates as luminescent dopants for OLEDs. Thermogravimetric analysis under flowing
nitrogen showed that BCzAuBBIAuBCz and BCzAuBAZAuBCz both undergo a 10% weight loss at 315 ºC.
Attempts to sublime the materials under vacuum at temperatures approaching 315 ºC led to
decomposition of the material and no observable sublimation. Therefore, emissive layers for the
devices were prepared using solution processing methods. The fabrication procedure is described in
chapter 7. The monometallic MACAuCz complex was chosen as an emissive dopant for comparative
purposes as this derivative has HOMO, LUMO and emission energies that are similar to the bimetallic
complexes.
The host material tris(4-carbazoyl-9-ylphenyl)amine (TCTA) was selected due to its low barrier for
hole injection, high hole mobility and ability to disperse the dopants in the matrix. Photoluminescence
data for the cMa complexes doped into TCTA films are given in Table 3-6. Photoluminescence (PL)
spectra of the films were exclusively from the dopant. The BCzAuBBIAuBCz complex has a higher PL
than BCzAuBAZAuBCz in a TCTA thin film. The lower PL could be due to a higher degree of aggregation
for BCzAuBAZAuBCz, leading to greater self- quenching for BCzAuBAZAuBCz than the BBI analog. This
observation is consistent with lower solubility of BCzAuBAZAuBCz in liquid solution compared to
BCzAuBBIAuBCz. All three complexes show a decrease in PL and radiative rate with increasing
concentration that is consistent with static quenching caused by aggregation of the dopant (see Table
3-6).
Table 3-6. Photoluminescence data of binuclear complexes and MACAuCz in the host material TCTA.
OLEDs were fabricated with the general structure ITO/MoOx (5 nm)/10% dopant in TCTA
(30 nm)/TPBi(50 nm)/Liq (1.5 nm) / Al (100 nm). The layer of MoOx deposited onto the ITO was used
to passivate the ITO surface, facilitate hole injection and improve device stability (Figure 3-11) relative
to the commonly used PEDOT:PSS.159
PEDOT:PSS based solution OLED devices have a large hole
injection barrier, indicated by the high turn-on voltage (Von, defined at brightness of 1 cd/m2) generally
around 4 to 6 V.159
MoOx has not been widely used in OLEDs as a HIL despite the suitable frontier
energies (-9.7 eV/-5.5 eV). A device with MoOx as the HIL with structure ITO/ MoOx (5 nm)/ 20%
BCzAuBBIAuBCz in TCTA (30 nm)/TPBi (50 nm)/Liq (1.5 nm)/Al (100 nm), was fabricated. Pure PL
spectrum (0.28, 0.61) has been observed for the device. The MoOx device exhibits much better current
Emitter Conc. in
TCTA PL (%) t
(ns)
kr
(106
s
-1
)
knr
(106 s
-1
)
BCzAuBBIAuBCz
5 86 203 4.2 0.7
10 74 209 3.5 1.2
20 63 232 2.7 1.6
30 46 176 2.6 3.0
BCzAuBAZAuBCz
5 64 277 2.3 1.3
10 50 250 2.0 2.0
MACAuCz
5 100 752 1.3 <0.01
10 79 752 1.1 0.28

56
conduction and luminance than the PEDOT device in literature at a certain voltage as shown in the
current-luminance-voltage (J-L-V). The turn-on voltage (Von, defined at brightness of 1 cd/m2) is 2.8 V
for the MoOx devices. At 4.8 V, the MoOx device gives a current density of 10 mA/cm2
and a
luminance 3423 cd/m2
.The absence of MoOx as an HIL leads to an increased dark current and therefore
poor device performance (see Figure 3-11). Furthermore, in the absence of PEDOT:PSS in the devices,
annealing is not required. The devices with the same structure were fabricated, with and without
annealing of the EML layer after spin coating. Their J-L-V characteristics are found to be similar. The
optimal thickness for the MoOx layer is 5 nm. Devices with a 10 nm of MoOx layer lead to less uniform
organic film coverage and give high dark currents and turn-on voltages. Devices were fabricated at
doping levels of 5%, 10%, 15% and 20%, with the 10% doping giving the best device performance
(Figure 3-12).
The characteristics of optimized devices are shown in Figure 3-13 and Table 3-7. The energy levels for
frontier orbitals for the materials used in the device are shown in Figure 3-13(b). The energy levels of
the dopants are nested within those of hosts, so it is expected that charges will be both carried and
trapped by the dopants. The electroluminescence spectra are identical to the photoluminescence spectra
of a given dopant, indicating no contribution from either host or ETL material. The current-voltage
characteristics show higher current at a given bias for devices using BCzAuBAZAuBCz compared to those
using BCzAuBBIAuBCz and MACAuCz (Figure 3-13(c)). Further, the BCzAuBAZAuBCz device turns on at a
lower voltage (2.5 V) than either the BCzAuBBIAuBCz (2.7 V) or MACAuCz device (2.8 V), and shows a
higher current density between turn on and 10 V, likely due to differences in charge mobility. The
MACAuCz device has an external quantum efficiency (EQE) of 14% at 6.2 V with a luminance of 300
cd/m2
, whereas the BCzAuBBIAuBCz device reaches a higher EQE (17%) at 6.1 V at a comparable
luminance (Figure 3-13(c,d)). In contrast, devices using BCzAuBAZAuBCz reach an EQE = 11% at 3.5 V
with a luminance of 6 cd/m2
and achieves a stable performance of EQE = 10% at 5.5 V with a luminance
of 280 cd/m2
.
Figure 3-11. OLED Devices with 20% BCzAuBBIAuBCz in TCTA as host material with (red) and without
(blue) MoOx. Device structure: ITO/MoOx (5 nm)(with : red, without: blue)/20 % BCzAuBBIAuBCz in TCTA
(30 nm)/TPBi(50 nm)/Liq (1.5 nm)/Al (100 nm)
0.01 0.1 1 10 100
0
2
4
6
8
10
12
14
16
18
20
EQE (%)
Current Density (mA/cm2
)
BBI
BBI_no MoOx
0 2 4 6 8 10 12
1E-6
1E-5
1E-4
0.001
0.01
0.1
1
10
100
1000
Current density (mA/cm2
)
Voltage (V)
BBI
BBI_no MoOx
0 2 4 6 8 10 12
0.1
1
10
100
1000
10000
100000
Luminance (cd/m2
)
Voltage (V)
BBI
BBI_no MoOx

57
The PL limits the highest possible EQE that can be achieved for a given dopant in the EML. If the
EQEmax is divided by the  PL of that dopant in the same matrix, one can obtain an EQE that would be
achieved if the dopant had PL = 1.0. The EQE values adjusted to unity PL efficiency are 18, 22 and
23 % for MACAuCz, BCzAuBAZAuBCz and BCzAuBBIAuBCz, respectively. Therefore, the solution-processed
OLEDs give the maximum theoretical efficiency for OLEDs of 20% without extrinsic or intrinsic
(dopant alignment) outcoupling enhancement.89 The brightness and EQE values for all the devices are
comparable to other solution processed OLEDs using cMa dopants.114, 116, 147, 160, 161
The roll-off in EQE with increasing current density for the MACAuCz device is 50% from 1 to 100
mA/cm2
. In contrast, the roll-off for the device with the BCzAuBBIAuBCz emitter is only 40% in the same
range, which suggests that a fast radiative rate helps suppress luminescence quenching by triplet-triplet
(TTA) or triplet-polaron annihilation (TPA).
Figure 3-12. Doping concentration-controlled OLED devices with BCzAuBBIAuBCz and TCTA as
host material with the following device structure: ITO/MoOx (5 nm)/X% BCzAuBBIAuBCz in
TCTA (30 nm)/TPBi (50 nm)/Liq (1.5 nm)/Al (100 nm)
0 2 4 6 8 10 12
1E-6
1E-5
1E-4
0.001
0.01
0.1
1
10
100
1000
10000
Current density (mA/cm2
)
Voltage (V)
20%
15%
10%
5%
0 2 4 6 8 10 12
0.1
1
10
100
1000
10000
100000
1000000
Luminance (cd/m2
)
Voltage (V)
20%
15%
10%
5%
0.01 0.1 1 10 100
0
2
4
6
8
10
12
14
16
18
20
EQE (%)
Current Density (mA/cm2
)
20%
15%
10%
5%
300 400 500 600 700 800
0.0
0.2
0.4
0.6
0.8
1.0
Normalized spectrum
Wavelength (nm)
20%
15%
10%
5%

58
Table 3-7. OLED device performance.
dopant Von
(V)
EQEmax
(%)
1 mA cm-2 100 mA cm-2
λmax (nm),
EQE CIE
(%)
luminance
(cd/m2
)
EQE
(%)
luminance
(cd/m2
)
MACAuCz 2.8 14 14 440 7 22,000 531,
(0.25, 0.45)
BCzAuBBIAuBCz 2.7 17 17 610 10 34,000 526,
(0.21, 0.48)
BCzAuBAZAuBCz 2.5 11 10 320 5 18,000 534,
(0.27, 0.50)
Figure 3-13. Device characteristics of OLEDs using BCzAuBBIAuBCz, BCzAuBAZAuBCz and
MACAuCz dopants. (a) Molecular structures of host and electron transport materials. (b) Device
architecture with HOMO and LUMO levels in eV. (c) Current-voltage and luminance-voltage
curves. (d) Efficiency (EQE) curves and electroluminescence spectra (inset).
0.01 0.1 1 10 100
0
5
10
15
20 400 500 600 700
0.0
0.5
1.0 EL Intensity
Wavelength (nm) EQE (%)
Current Density (mA/cm2
)
BCzAuBBIAuBCz BCzAuBAZAuBCz MACAuCz
(b) (d)
0 2 4 6 8 10 12
10−6 10−5 10−4 10−3 10−2 10−1 100 101 102 103 Current density (mA/cm2
)
Voltage (V)
BCzAuBBIAuBCz BCzAuBAZAuBCz MACAuCz
10−1 100 101 102 103 104 105 106 107 Luminance (cd m-2
)
(a) (c)

59
Conclusion
In summary, we have designed and synthesized a series of luminescent bimetallic cMa complexes by
using Janus carbenes to bridge Au-amide moieties. The presence of one donor on each side of Janus
carbene maintains the high oscillator strength between the donors and central carbene acceptor.
Luminescence from the bimetallic complexes occurs via TADF and spans from blue (em =  nm) to
green (em =  nm) with PL efficiencies close to unity. The bimetallic complexes can have radiative
rates that are 23 times faster than their monometallic analogs. Theoretical and photophysical analyses
show that the fast radiative rate is due to a decrease in the singlet-triplet energy gap caused by spatial
extension of the ICT exciton over the entire Janus carbene ligand. OLEDs fabricated using solutionbased deposition methods utilizing the bimetallic complexes as dopants give high luminance efficiency.
Moreover, MoOx as the hole injection material is a viable alternative to PEDOT:PSS. The small roll-off
demonstrates that fast radiative rate of the emitter can indeed suppress the TTA and TPA in the devices.
Experimental Procedures
Synthesis ClAuBAZAuCl
The BAZ Ligand was synthesized according to Literature.149 To a 100 mL Schlenk flask with bubble
degassed dry THF (45 mL) 500 mg (0.444 mmol, 1.0 eq) BAZ-Ligand was added and cooled to -78C.
1.86 mL (0.931 mmol, 2.1 eq) 0.5M KHMDS THF solution was added dropwise over 20 min. After 2 h
stirring at -78C Au(Me2S)Cl (287 mg, 0.976 mmol, 2.2 eq) was added and reaction was slowly warmed
up to room temperature overnight. Reaction was worked up by filtration through Celite, washing with
dichloromethane and solvent removal. Product was precipitated from dichloromethane by adding
hexanes, yielding to the off-white ClAuBZACAuCl in 86% yield (495 mg, 0.383 mmol). 1H NMR (400
MHz, acetone-d6) δ 7.54 – 7.49 (m, 2H), 7.44 – 7.36 (m, 6H), 7.33 – 7.27 (m, 4H), 6.46 (s, 2H), 4.89
(s, 4H), 3.24 (p, J = 6.7 Hz, 4H), 3.10 (p, J = 6.9 Hz, 4H), 1.39 (dd, J = 10.9, 6.8 Hz, 24H), 1.25 (d, J
= 6.8 Hz, 12H), 1.16 (d, J = 6.8 Hz, 12H).
Synthesis CzAuBAZAuCz
To a 50 mL Schlenkflask with bubble degassed dry THF (25 mL) 54 mg (0.325 mmol, 2.1 eq) 1Hcarbazole ligand was added. 0.159 mL (0.317 mmol, 2.05 eq) 2M sodium tert-butoxide (NaOtBu) THF
solution was added dropwise. After 1 h stirring at ambient temperature, ClAuBZACAuCl (200 mg, 0.155
mmol, 1.0 eq) was added and reaction was stirred overnight. Reaction was worked up by filtration
through Celite, washing with dichloromethane and solvent removal. Product was precipitated from
dichloromethane by adding hexanes. Solid was filtered and washed with diethyl ether, yielding to the
colorless CzAuBZACAuCz in 80% yield (192 mg, 0.124 mmol). Under a UV light the solid is blue
emissive.1H NMR (400 MHz, acetone-d6) δ 7.86 – 7.66 (m, 8H), 7.60 (d, J = 7.8 Hz, 4H), 7.51 (d, J =
7.8 Hz, 4H), 6.89 (ddd, J = 8.3, 7.0, 1.3 Hz, 4H), 6.74 (ddd, J = 7.8, 7.0, 1.1 Hz, 4H), 6.61 (s, 2H), 6.08
(dt, J = 8.2, 0.9 Hz, 4H), 5.03 (s, 4H), 3.41 (hept, J = 7.7 Hz, 4H), 3.33 – 3.20 (m, 4H), 1.37 (d, J = 6.8
Hz, 12H), 1.33 (t, J = 6.8 Hz, 24H), 1.24 (d, J = 6.9 Hz, 12H). CHN: C: 62.74%, H: 5.98%, N: 5.28%;
calculated: C: 63.39%; H: 5.84%; N: 5.41%
Synthesis BCzAuBAZAuBCz
To a 50 mL Schlenk flask with bubble degassed dry THF (25 mL) 90 mg (0.325 mmol, 2.1 eq) 3,6-Ditert-butylcarbazole ligand was added. 0.158 mL (0.317 mmol, 2.05 eq) 2M sodium tert-butoxide

60
(NaOtBu) solution was added dropwise. After 1 h stirring at ambient temperature, ClAuBAZAuCl (200
mg, 0.155 mmol, 1.0 eq) was added and reaction was stirred overnight. Reaction was worked up by
filtration through Celite, washing with dichloromethane and solvent removal. Product was washed with
-40C hexanes and collected the precipitations. Then repeat this step three times, yielding to the yellow
powder BCzAuBAZAuBCz in 72% yield (198 mg, 0.111 mmol). Under a UV light the solid is skyblue
emissive.
1H NMR (400 MHz, acetone-d6) δ 7.88 – 7.80 (m, 6H), 7.72 (t, J = 7.8 Hz, 2H), 7.62 (d, J
= 7.8 Hz, 4H), 7.53 (d, J = 7.8 Hz, 4H), 7.00 (dd, J = 8.6, 2.0 Hz, 4H), 6.61 (s, 2H), 6.04 (dd, J = 8.6,
0.6 Hz, 4H), 5.03 (s, 4H), 3.41 (sept, J = 7.0 Hz, 4H), 3.27 (sept, J = 6.5 Hz, 4H), 1.40 (d, J = 6.9 Hz,
36H), 1.38 – 1.31 (m, 60H), 1.26 (d, J = 6.8 Hz, 12H).
13C NMR (100 MHz, acetone-d6) δ 229.2,
228.3, 227.8, 223.4, 223.1, 219.3, 219.2, 214.2, 211.5, 210.0, 208.8, 205.2, 202.6, 201.1, 195.5, 194.4,
192.4, 191.0, 190.1, 189.7, 185.9, 182.5, 180.6, 177.9, 177.3, 173.7, 173.6, 166.6, 164.9, 162.8, 153.1,
148.3, 148.1, 146.8, 145.7, 145.3, 137.6, 125.5, 125.3, 123.4, 120.4, 118.7, 114.5, 113.4, 38.4, 38.1,
37.8, 37.4, 34.0, 31.7, 24.0, 24.0, 23.6. CHN: C: 65.32%, H: 6.55%, N: 4.93%; calculated: C: 66.20%;
H: 6.92%; N: 4.73%
Synthesis BimAuBAZAuBim
The bim ligand was synthesized according to Literature.93 To a 50 mL Schlenk flask with bubble
degassed dry THF (25 mL) 61 mg (0.293 mmol, 2.1 eq) 1H-Bim ligand was added. 0.143 mL (0.286
mmol, 2.05 eq) 2M sodium tert-butoxide (NaOtBu) THF solution was added dropwise. After 1 h
stirring at ambient temperature, ClAuBAZAuCl (180 mg, 0.139 mmol, 1.0 eq) was added and reaction was
stirred overnight. Reaction was worked up by filtration through Celite, washing with dichloromethane
and solvent removal. Product was precipitated from dichloromethane by adding hexanes. Solid was
filtered and washed with diethyl ether, yielding to the off-white BimAuBAZAuBim in 76% yield (174 mg,
0.107 mmol). Under a UV light the solid is blue emissive. 1H NMR (400 MHz, acetone-d6) δ 7.70 –
7.57 (m, 6H), 7.57 – 7.48 (m, 4H), 7.46 – 7.39 (m, 4H), 7.30 (ddd, J = 8.0, 1.2, 0.6 Hz, 2H), 7.06 (ddd,
J = 8.0, 7.3, 1.2 Hz, 2H), 6.96 – 6.84 (m, 8H), 6.61 (s, 2H), 6.27 – 6.18 (m, 2H), 5.05 (s, 4H), 3.41
(sept, J = 6.8 Hz, 4H), 3.27 (sept, J = 6.5 Hz, 4H), 1.48 (dd, J = 14.2, 6.8 Hz, 24H), 1.32 (d, J = 6.9 Hz,
12H), 1.25 (d, J = 6.8 Hz, 12H).
13C NMR (100 MHz, acetone-d6) δ 146.5, 145.1, 145.0, 136.7, 130.6,
129.8, 127.2, 127.1, 126.7, 126.5, 126.1, 126.0, 125.5, 125.2, 121.0, 120.9, 118.8, 118.5, 117.4, 116.7,
115.1, 113.2, 109.0, 108.8, 24.1, 23.7. CHN: C: 42.83%, H: 2.06%, N: 14.31%; calculated: C: 43.74%;
H: 2.24%; N: 14.17%
Synthesis BBI ligand
Step 1: In a 500-mL two-neck flask containing a stir bar and 5.0 g (21.1 mmol) of the
dichlorodinitrobenzene was equipped with a water-condenser. The system was pumped down and
backfilled with nitrogen gas for three cycles. Nitrogen-degassed 2,6-diisopropoylaniline (23.9 mL, 126.6
mmol) was cannula-transferred into the reaction flask. The mixture was heated at 150°C for 48 hours after
which it was allowed to cool down to room temperature. Methanol (~20 mL) was added, and the flask was
placed in a freezer (-40°C) to encourage precipitation. The precipitate was collected by vacuum filtration
and washed with cold methanol and dried under vacuum to yield 8.5 g (78%) of a yellow powder. 1H NMR
(400 MHz, Chloroform-d) δ = 9.43 (s, 2H), 9.39 (s, 1H), 7.23 (t, J = 7.8 Hz, 2H), 7.04 (d, J = 7.8 Hz, 4H), 5.09

61
(s, 1H), 2.75 (hept, J = 6.8 Hz, 4H), 1.07 (d, J = 6.8 Hz, 12H), 0.85 (d, J = 6.9 Hz, 12H). 13C NMR (101 MHz,
Chloroform-d) δ = 149.35, 146.30, 132.03, 129.66, 129.03, 125.15,124.08, 94.82, 28.73, 24.93, 22.37.
Step 2:A 500-mL round-bottom flask was charged with a magnetic stir-bar and formic acid (98%)/water
mixture (90:10 mL). NaHCO3 (9.72 g, 115.7 mmol) was added portion-wise with vigorous stirring. To
the mixture was added Pd/C (0.616 g, 10 wt %, 0.578 mmol Pd) and dinitroarene (3.0 g, 5.78 mmol).
The flask was fitted with a water condenser and heated in an oil bath with vigorous stirring at 120°C
for 48 h. The mixture was then allowed to cool to room temperature and filtered through a plug of celite
with the aid of 70 mL of Ethyl acetate. On a rotary evaporator, the filtrate volume was reduced to a
slurry mixture. 50 mL of de-ionized water was added into the mixture, swirled, and slowly poured into
a vigorously stirring solution of saturated aqueous K2CO3. A beige precipitate developed and was
collected via vacuum filtration, rinsed with plenty of water H2O and dried under vacuum to yield 2.73
g (98%). 1H NMR (400 MHz, Chloroform-d): δ = 8.39 (s, 1H), 7.87 (s, 2H), 7.47 (t, J = 7.8 Hz, 2H),
7.27 (d, J = 7.9 Hz, 4H), 6.52 (s, 1H), 2.26 (hept, J = 6.7 Hz, 4H), 1.06 (dd, J = 6.9, 0.7 Hz, 12H), 0.90
(dd, J = 6.8, 0.7 Hz, 12H). 13C NMR (101 MHz, CDCl3) δ 147.95, 144.81, 140.84, 134.61, 130.99,
130.49, 124.38, 110.63, 89.89, 28.41, 24.88, 23.95.
Step 3:In a 100-mL pressure flask, benzobis(imidazole) (500 mg, 1.04 mmol), 2-iodomethane (0.52
mL, 8.36 mmol), and acetonitrile (5 mL) were added. The pressure flask was sealed and heated at 90°C
for 24 hours. After the completion of the reaction, the mixture was cooled down to room temperature
and solvents were removed under reduced pressure. Diethyl ether was added to the resulting crude
solid, which was sonicated for 10 min, and collected by vacuum filtration to yield a beige solid (680
mg, 85%). 1H NMR (400 MHz, Acetone-d6) δ = 10.73 (s, 2H), 9.91 (s, 1H), 8.24 (s, 1H), 7.63 (t, 2H),
7.45(d, J = 7.9 Hz, 4H), 4.73 (s, 6H), 2.45 (hept, J = 6.8 Hz, 4H), 1.14 (d, J = 6.8 Hz, 12H), 1.00 (d, J
= 6.8 Hz,12H). 13C NMR (101 MHz, acetone) δ 146.89, 146.78, 133.35, 132.32, 127.59, 125.10,
102.47, 98.68, 36.23, 24.25, 23.15, -2.00
Synthesis ClAuBBIAuCl
To a 100 mL Schlenk flask with bubble degassed dry THF (35 mL) 400 mg (0.496 mmol, 1.0 eq) BBILigand was added. 1.49 mL (1.04 mmol, 2.1 eq) 0.7M KHMDS THF solution was added dropwise.
After 1h stirring at room temperature Au(Me2S)Cl (321 mg, 1.09 mmol, 2.2 eq) was added and reaction
was stirred overnight. Reaction was worked up by filtration through Celite, washing with
dichloromethane and solvent removal. Product was precipitated from dichloromethane by adding
hexanes. Solid was filtered and washed with methanol, yielding to the off-white ClAuBBIAuCl in 72%
yield (348 mg, 0.358 mmol).
1H NMR (400 MHz, acetone-d6) δ 8.52 (s, 1H), 7.51 (t, 2H), 7.36 (d, J
= 7.8 Hz, 4H), 6.87 (s, 1H), 4.37 (s, 6H), 2.27 (sept, J = 6.5 Hz, 4H), 1.22 (d, J = 6.8 Hz, 12H), 0.94
(d, J = 6.9 Hz, 12H).
Synthesis BCzAuBBIAuBCz
To a 25 mL Schlenk flask with bubble degassed dry THF (10 mL) 120 mg (0.432 mmol, 2.1 eq) 3,6-
Di-tert-butylcarbazole ligand was added. 0.211 mL (0.422 mmol, 2.05 eq) 2M sodium tert-butoxide
(NaOtBu) solution was added dropwise. After 1 h stirring at ambient temperature, ClAuBBIAuCl (200
mg, 0.206 mmol, 1.0 eq) was added and reaction was stirred overnight. Reaction was worked up by
filtration through Celite, washing with dichloromethane and solvent removal. Product was purified by
addition of Hexanes, sonicating 30 min, cooling the flask slowly in a Dewar to -40C in a deep freezer.
Solid was collected at low temperature. The purification step was repeated if unreacted BCz was

62
detected by NMR. The bright yellow solid BCzAuBBIAuBCz was obtained in 73% yield (219 mg, 0.150
mmol). Under a UV light the solid is green emissive.
1H NMR (600 MHz, acetone-d6) δ 8.65 (d, J =
0.8 Hz, 1H), 8.00 (dd, J = 2.0, 0.7 Hz, 4H), 7.71 (t, J = 8.0 Hz, 2H), 7.52 (d, J = 8.0 Hz, 4H), 7.20 (dd,
J = 8.5, 2.0 Hz, 4H), 7.13 – 7.10 (m, 5H), 4.62 (s, 6H), 2.76 (s, 2H), 2.48 (sept, J = 6.6 Hz, 4H), 1.42
(d, J = 6.6 Hz, 4H), 1.38 (s, 35H), 1.30 (d, J = 6.8 Hz, 12H), 1.04 (d, J = 6.9 Hz, 12H).
13C NMR (151
MHz, acetone-d6) δ 148.1, 147.1, 138.1, 134.0, 132.6, 131.8, 131.1, 124.8, 123.9, 121.1, 114.9, 113.0,
110.0, 109.6, 36.3, 35.6, 34.1, 31.7, 31.5, 24.1, 23.6. CHN: C: 62.08%, H: 6.17, N: 6.44%, calculated:
C: 61.17%; H: 6.58%; N: 5.49% (includes 1 THF)
Synthesis BZIAuBCz
BZIAuCl Ligand was synthesized according to Literature.4 To a 25 mL Schlenk flask with bubble
degassed dry THF (10 mL) 83 mg (0.297 mmol, 1.05 eq) 3,6-Di-tert-butylcarbazole ligand was added.
0.149 mL (0.297 mmol, 1.05 eq) 2M sodium tert-butoxide (NaOtBu) solution was added dropwise.
After 1 h stirring at ambient temperature, BZIAuCl (190 mg, 0.283 mmol, 1.0 eq) was added and reaction
was stirred overnight. Reaction was worked up by filtration through Celite, washing with
dichloromethane and solvent removal. Product was precipitated from dichloromethane by adding
hexanes. Solid was filtered and washed with diethyl ether, yielding to the colorless BZIAuBCz in 82%
yield (212 mg, 0.232 mmol). Under a UV light the solid is blue emissive.
1H NMR (400 MHz, acetoned6) δ 7.92 (dd, J = 2.1, 0.7 Hz, 2H), 7.87 – 7.78 (m, 2H), 7.68 – 7.58 (m, 6H), 7.43 (dd, J = 6.1, 3.2 Hz,
2H), 7.07 (dd, J = 8.5, 2.0 Hz, 2H), 6.69 (dd, J = 8.5, 0.6 Hz, 2H), 2.63 (sept, J = 6.8 Hz, 4H), 1.39 (d,
J = 6.9 Hz, 12H), 1.34 (s, 18H), 1.19 (d, J = 6.9 Hz, 12H).
13C NMR (100 MHz, acetone-d6) δ 148.0,
146.9, 137.9, 134.9, 131.7, 131.1, 125.8, 124.7, 123.7, 120.9, 114.8, 112.8, 112.1, 34.0, 31.7, 31.4, 24.0,
23.3. CHN: C: 67.20%, H: 6.95%, N: 5.10%, calculated: C: 67.02%; H: 6.84%; N: 4.60%
OLED Fabrication
OLED devices were fabricated on pre-patterned ITO-coated glass substrates (20 ± 5 Ω cm2, Thin Film
Devices, Inc.). Prior to deposition, the substrates were cleaned with soap, rinsed with deionized water
and sonicated for 10 minutes. Afterwards, two subsequent rinses and 15-minute sonication baths were
performed in acetone and isopropyl alcohol sequentially, followed by 15 min UV ozone exposure. After
the MoOx deposition (5nm) with an EvoVac (Angstrom Engineering) the substrates were transferred
without contact to air into a particle free Nitrogen Glovebox. The EML was spin coated from a toluene
solution, which was stirred overnight at 65 C and filtered before usage. A consistent thickness of 30 nm
was achieved with 65 L of solution with concentration of 8mg per mL toluene at a spin rate of 3000
rpm for 90s. If applicable substrates were annealed under N2 atmosphere for 10min @110C. Transfer
into the deposition system occurred without contact to air. TPBi (ETL) was deposited using a Vacuum
Thermal Evaporation (Angstrom Engineering) and the Cathode (Liq and Al) was deposited using a
Vacuum Thermal Evaporation (Kurt J. Lesker).
Current-voltage-luminescence (J-V-L) curves were measured in an by using a Keithley power source
meter model 2400 and a Newport multifunction optical model 1835-C, PIN-220DP/SB blue enhanced
silicon photodiodes (OSI optoelectronics Ltd.). The sensor was set to measure power at an energy of
520 nm, followed by correcting to the average electroluminescence wavelength for each individual
device during data process. Electroluminescence (EL) spectra of OLEDs were measured using the
fluorimeter (model C-60 Photon Technology International QuantaMaster) at several different voltages.
Thicknesses were determined on Silicon wafers using a Filmsense FS-1 Ellipsometer.

63
Chapter 4 – cMa’s with electron deficient amides
The following chapter was written by Jonas Schaab solely, but part of the data was achieved in
collaboration with James Fortwengler and is not yet published. James Fortwengler was responsible
for part of the synthesis and for calculations I, on the other hand, contributed to the project by synthesis,
conducting photophysics experiments as well as electrochemical analysis and analyzing all complexes
via single crystal X-Ray diffraction.
Introduction
The previous chapters showed that cMa complexes can achieve extraordinary high radiative rate and
can be easily color tuned by modifying carbenes. cMa’s with Bim as amide can reach radiative rates of
4.0 x 106
s
-1
and emitting deep blue (CIE 0.16 0.17). As mentioned in Chapter 1, cMa’s can be used as
dopants to achieve high efficient OLED devices with up to 26.3% EQE and up to 20 000h lifespan.2, 3,
7, 11, 34 As seen in Table 1-1 most of the published OLEDs are green in color and only a few blue OLEDs
are published. On the other hand industrially mainly blue is from interest, as blue OLEDs have the
shortest device lifespan, due to TTA and TPA. But why are there so little blue cMa OLEDs published?
Comparing the HOMO and LUMO energies of commonly used blue Iridium based dopants like
Ir(cbp)3 or Ir(xxx)3 and cMa complexes it is notable that the HOMO energies are around 0.5-1 eV
shallower. This shallower HOMO energy requires also shallower LUMO energies to obtain the same
emission color. For blue complexes like 𝐴𝑢𝐶𝑧
𝐶𝐴𝐴𝐶or 𝐴𝑢𝐵𝑖𝑚
𝐵𝑍𝐼 the LUMO energies are as shallow as -1.8
and -1.5 eV. These high LUMO energies make it difficult to pair these complexes with electron
transporting host materials, as most Host materials have LUMO energies below -2 eV or above 1.0
eV. Therefore the blue cMa’s can only be paired with wide band gap host materials like mCBP,
which requires the dopant to carry electrons and holes, resulting in lower efficiencies and short
device lifespans. Too achieve blue cMa emitters, it would be necessary to lower both HOMO and
LUMO by around 0.5 eV to be able to be paired with hole and electron transporting host materials,
which in turn, would allow a longer device lifetime. As discussed in chapter 1, LUMO energies can
be varied by changing the carbene, which can lower the LUMO to below -2.0 eV by using MAC, PAC
or PZI as carbenes. The HOMO on the other hand can be tuned the easiest by changing the
substituents on the amide. Electron donating groups like tBu are rising the HOMO and electron
withdrawing groups (EWG) like are lowering the HOMO. Adding a cyanide (CN) or trifluoromethane (CF3) group in the 3 position on carbazole results in a HOMO lowering of ~0.3 eV and
adding two EWG in the 3 and 6 position in a reduction of 0.6 eV (see Figure 4-1). Previously known
complexes with EWG on Cz like 𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑀𝐴𝐶
, 𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝐶𝐴𝐴𝐶 have very long excited state lifetimes in the
order of s. In these cases the CT band is close to an ligand localized excited state, which causes
the excited state lifetime to elongate significantly. By using carbenes with lower LUMO, the
emission can be slightly redshifted, and therefore lower the CT band below the LE excited state,
resulting in shorter lifetimes. As bim has a higher ligand localized excited state (see Figure 2-6),
adding EWGs should further improve the radiative rate.
In this chapter we will investigate cMa complexes with EWG on the amide, to achieve blue emitters
with lower HOMO and LUMO energies, compared to above discussed complexes with Cz and Bim.
The complexes can be seen in the following Figure:

64
Synthesis
Synthesis 𝑨𝒖𝑪𝒛(𝑪𝑵)𝟐
𝑷𝒁𝑰

𝐴𝑢𝐶𝑙
𝑃𝑍𝐼was synthesized as described in Chapter 2 and in Ref.
93 3,6-Di-cyano-carbazole (CzCN2) was
synthesized after Ref.162 To a 25 mL Schlenk flask with degassed dry THF (10 mL) CzCN2 ligand (589
mg, 2.71 mmol, 1.1 eq) was added. 1.29 mL (2.59 mmol, 1.05 eq) 2M sodium tert-butoxide (NaOtBu)
solution was added dropwise. After 1 h stirring at 0C, 𝐴𝑢𝐶𝑙
𝑃𝑍𝐼 (400 mg, 2.46 mmol, 1.0 eq) was added
and reaction was stirred overnight. Reaction was worked up by filtration through Celite, washing with
dichloromethane and solvent removal. Product was precipitated from dichloromethane by adding
hexanes. Solid was filtered and washed with diethyl ether, yielding to the colorless 𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼 in 88%
yield. Under a UV light the solid is blue emissive.
1H NMR (400 MHz, Acetone-d6) δ = 8.78 (s, 2H), 8.49 (d, 2H), 7.90 (td, J=7.8, 1.9, 2H), 7.68 (dd,
J=7.8, 1.9, 4H), 7.42 (dt, J=8.6, 1.7, 2H), 6.70 (dd, J=8.5, 1.9, 2H), 2.66 (hept, J=6.8, 4H), 1.33 (dd,
J=7.0, 1.9, 12H), 1.15 (dd, J=6.9, 1.9, 12H).
13C NMR (100 MHz, Acetone-d6) δ 188.40, 151.76, 147.17, 142.23, 140.40, 131.67, 130.85, 127.84,
125.34, 124.80, 123.47, 120.27, 114.50, 100.13, 29.19, 23.75, 23.35.
Synthesis 𝐀𝐮𝐂𝐳(𝐂𝐅𝟑)𝟐
𝐏𝐙𝐈

𝐴𝑢𝐶𝑙
𝑃𝑍𝐼was synthesized as described in Chapter 2 and in Ref.93 3,6-Di-(trifluoromethane)-carbazole
(Cz(CF3)2) was synthesized after Ref.163 To a 25 mL Schlenk flask with degassed dry THF (10 mL)
Cz(CF3)2 (1.20g, 3.94 mmol, 1.60 eq) ligand was added. 1.85mL (3.70 mmol, 1.50 eq) 2M sodium tertbutoxide (NaOtBu) solution was added dropwise. After 1 h stirring at 0C, 𝐴𝑢𝐶𝑙
𝑃𝑍𝐼 (400 mg, 2.46 mmol,
Figure 4-1. cMa complexes with EWG on the amide

65
1.0 eq) was added and reaction was stirred overnight and reached ambient temperature. Reaction was
worked up by filtration through Celite, washing with dichloromethane and solvent removal. Product
dissolved in diethyl ether, filtered and solvent removed. Product was washed with cold (-40C) hexanes.
If remaining Cz(CF3)2 was visible in NMR, the remaining free ligand could be removed by sublimation
at 120C in high vacuum (10-7
torr), which afforded a highly pure 𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 in the boat.
1H NMR (400 MHz, acetone-d6) δ = 8.76 (s, 2H), 8.41 (s, 2H), 7.89 (t, J=7.8, 2Hz), 7.68 (d, J=7.9,
4H), 7.35 (dd, J=8.6, 1.9, 2H), 6.74 (d, J=8.4, 2H), 2.65 (h, J=6.7, 4H), 1.35 (d, J=6.9, 12H), 1.17 (d,
J=6.8, 12H).
19F NMR (376 MHz, acetone-d6) δ = -58.9 (referenced to CFCl3).
13C NMR (100 MHz, acetone-d6) δ 151.64, 147.11, 142.05, 140.41, 131.56, 130.88, 124.73, 123.22,
121.04, 118.89, 115.57, 113.77, 29.24, 23.76, 23.39.
Synthesis Cz(CF3)(Me)
Cz(CF3)(Me) was synthesized based on a modified from preparation from reference.163 p-Toluidine
(1.2g, 11.2 mmol, 1.0 eq), K3PO4 (4.75g, 22.4 mmol, 2.0 eq) and XPhos Pd-G4 (622 mg, 0.784 mmol,
0.07 eq) were pump-purged and 1-iodo-4-(trifluoromethyl)benzene (1.65mL 3.05 g, 11.2 mmol, 1.0 eq)
was added via syringe as well as bubble degassed toluene. The reaction mixture was heated to 100C
overnight. The reaction mixture was extracted from ethylacetate and brine. Column purification with
20% EA in Hexanes yielded the product in 85% yield.
The intermediate was dissolved in Acetic Acid and Pd(OAc)2 (10wt%) was added. Under an oxygen
atmosphere (Ballon with O2), the reaction was refluxed overnight. The reaction mixture was quenched
with aqueous NaHCO3 and extracted with ethyl acetate. Column chromatography (Hex/EA 3:1) yielded
Cz(CF3)(Me) in a 80% yield.
Synthesis 𝑨𝒖𝑪𝒛(𝑪𝑭𝟑)(𝑴𝒆)
𝑴𝑨𝑪

To a 25 mL Schlenk flask with degassed dry THF (10 mL) Cz(CF3)(Me) (115mg, 0.463 mmol, 1.05 eq)
ligand was added. 0.23mL (0.463 mmol, 1.05 eq) 2M sodium tert-butoxide (NaOtBu) solution was
added dropwise. After 1 h stirring at 0C, 𝐴𝑢𝐶𝑙
𝑀𝐴𝐶 (300 mg, 0.441 mmol, 1.0 eq) was added and reaction
was stirred overnight and reached ambient temperature. Reaction was worked up by filtration through
Celite, washing with dichloromethane and solvent removal. Product was washed with diethylether,
which afforded a highly pure 𝐴𝑢Cz(𝐶𝐹3)(𝑀𝑒)
𝑀𝐴𝐶
.
1H NMR (400 MHz, acetone-d6) δ 8.11 (dt, J=1.6, 0.8, 1H), 7.83 – 7.69 (m, 3H), 7.58 (d, J=7.9, 2H),
7.53 (d, J=7.8, 2H), 7.15 (dd, J=8.6, 1.8, 1H), 6.85 (ddd, J=8.3, 1.7, 0.6, 1H), 6.07 (dt, J=8.6, 0.7, 1H),
6.01 (dd, J=8.3, 0.6, 1H), 4.30 (s, 2H), 3.50 (hept, J=6.8, 2H), 3.25 (hept, J=6.8, 2H), 2.37 (s, 3H), 1.69
(s, 6H), 1.41 (d, J=6.9, 6H), 1.36 (d, J=6.8, 7H), 1.33 (d, J=6.9, 6H), 1.23 (d, J=6.8, 6H).
19F NMR (376 MHz, acetone-d6) δ = -60.3 (referenced to CFCl3).

66
13C NMR (100 MHz, acetone-d6) δ 170.9, 150.2, 147.5, 145.5, 144.4, 139.7, 135.7, 129.3, 129.0, 126.8,
124.8, 124.8, 124.5, 123.5, 122.6, 122.0, 118.2, 118.2, 118.0, 115.7, 115.3, 115.3, 113.2, 112.8, 60.1,
37.1, 23.0, 23.0, 22.8, 22.7, 22.3, 19.6.
Synthesis Boc-BimI2
BimH (30g, 145 mmol, 1.0eq) Potassiumiodide (31.24g, 188 mmol, 1.3eq) and Potassiumiodate (31.0g,
145 mmol, 1.0eq) was dissolved in Acetic Acid and refluxed overnight. The reaction mixture was
cooled and added to DI-water. The precipitate was filtered and washed with aqueous sodiumbicarbonate
solution till no gas formation was visible. The Filtercake was dried at 110C overnight, yielding a brown
solid crude mixture of BimI2. BimI2 crude was used as is for the 2nd step.
BimI2 crude (15.8g, 34.38 mmol, 1.0 eq) was added to a Schlenk flask and pump purged. Dry MeCN
was added, and bubble degassed with nitrogen. Ditertbutyl dicarbonate (11.85 mL, 11.25g, 51.57 mmol,
1.5 eq) was added via a syringe and mixture. 4-Dimethylaminopyridine (4.20 g, 34.38 mmol, 1.0 eq)
was added as one portion and stirred overnight. The reaction mixture was purified by column
chromatography (Hexane/Ethyl acetate with a 0-100% EA ramp), yielding pure BOC-BimI2 in a 10%
overall yield as a off-white solid.
1H NMR (400 MHz, acetone) δ = 8.60 (d, J=1.7, 1H), 8.57 (d, J=1.7, 1H), 8.03 (d, J=8.6, 1H), 7.81
(dd, J=8.6, 1.7, 1H), 7.70 (dd, J=8.5, 1.7, 1H), 7.54 (d, J=8.5, 1H), 1.74 (s, 9H).
Synthesis Bim(CN)2
This synthesis was done as a variation of the Synthesis of CzCN2 in ref162
. BOC-BimI2 (500mg) was
pump purged in a Schlenk flask and 30 mL DMF with 10% H2O added and bubble degassed for 30-
40min. Subsequenlty Zn dust (2.34 mg), Zn(CN)2 (126 mg), Zn(OAc)2 (6.56 mg, ) and Pd2(dba)3
(8.19mg) were added and heated to 100C overnight. The cooled reaction mixture was poured into aq
NH4Cl/aq NH3/DI water (4:1:5, 100 mL) and filtered and washed again with the same solution mixture,
followed by three washes with DI-Water, Toluene and MeOH. The dried filter cake yielded Bim(CN)2
as an off white solid in 90% yield.
1H NMR (400 MHz, dmso) δ = 8.90 (s, 2H), 7.71 (d, J=25.2, 4H).

67
Synthesis 𝑨𝒖𝑩𝒊𝒎(𝑪𝑵)𝟐
𝑷𝒁𝑰

To a 25 mL Schlenk flask with degassed dry THF (10 mL) Bim(CN)2 (100mg, 0.389 mmol, 1.05 eq)
ligand was added. 0.19mL (0.389 mmol, 1.05 eq) 2M sodium tert-butoxide (NaOtBu) solution was
added dropwise. After 1 h stirring at 0C, 𝐴𝑢𝐶𝑙
𝑃𝑍𝐼 (250 mg, 0.370 mmol, 1.0 eq) was added and reaction
was stirred overnight and reached ambient temperature. Reaction was worked up by filtration through
Celite, washing with dichloromethane and solvent removal. Product was washed with diethyl ether and
methanol, which afforded a highly pure 𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼
.
1H NMR (500 MHz, acetone) δ = 8.80 (d, J=6.6, 2H), 8.45 (dd, J=4.9, 1.6, 1H), 7.88 – 7.77 (m, 2H),
7.65 (dd, J=7.8, 3.8, 4H), 7.58 – 7.42 (m, 4H), 6.63 (d, J=8.4, 1H), 2.75 – 2.62 (m, 4H), 1.45 (dd, J=6.9,
3.2, 12H), 1.17 (dd, J=6.8, 2.0, 12H).
Synthesis 𝑨𝒖𝑩𝒊𝒎(𝑪𝑵)𝟐
𝑷𝑨𝑪

To a 25 mL Schlenk flask with degassed dry THF (10 mL) Bim(CN)2 (10mg, 0.389 mmol, 1.05 eq)
ligand was added. 0.19mL (0.389 mmol, 1.05 eq) 2M sodium tert-butoxide (NaOtBu) solution was
added dropwise. After 1 h stirring at 0C, 𝐴𝑢𝐶𝑙
𝑃𝐴𝐶 (259 mg, 0.370 mmol, 1.0 eq) was added and reaction
was stirred overnight and reached ambient temperature. Reaction was worked up by filtration through
Celite, washing with dichloromethane and solvent removal. Product was washed with diethyl ether and
methanol, which afforded a highly pure 𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝐴𝐶 in 49% yiekld.
Crystallographic Analysis
Single crystal X-ray diffraction was used to determine the molecular structures of
𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼
, 𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 and 𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼
. Crystallographic data for the seven compounds including
CCDC deposition numbers are given in Table 4-1 and representative structures are shown in Figure
4-2. All the complexes have a coplanar conformation of carbene and amide ligands (dihedral angles =
3.5-8.1°) and close to 180˚ angle around the metal center (C-Au-N = 170 - 176˚). The C-Au bond
lengths are in the range 1.97-1.98 Å and the Au-N bond is 1.999 – 2.026 Å. Values for the C-Au and
Au-N bond lengths are similar to analogous (carbene)M(Cz) complexes mentioned in previous chapters
and published previously.
10, 90, 91, 112, 115, 138, 164, 165 The sum of angles around the Ccarbene and Namide are
358˚-360˚, showing a planar geometry.

68
Table 4-1. Selected X-ray crystallographic data.
Electrochemistry
Cyclic and differential pulse voltammetry in CH2Cl2 was used to determine the redox properties of the
cMa complexes. Potentials relative to an internal ferrocene reference are listed in Table 4-2 and the
CVs are shown in Figure 4-3. The reduction is reversible for all complexes. The reductions potentials
of all complexes are similar to other cMs with the same carbene, indicating that the carbene is reduced.
For all cMas in this chapter oxidation is irreversible in CH2Cl2, even though the 3,6 positions of
Carbazole are blocked with a CN or a CF3 group.139 reversibility for cMa complexes with BCz as amide
are irreversible in CH2Cl2 but reversible in DMF.92, 93 Cz(CF3)2 HOMO is 70 meV shallower than the
one of CzCN2, which is due to the stronger electron withdrawing character of the CN group compared
to CF3. 𝐴𝑢𝐶𝑧(𝐶𝐹3)(𝑀𝑒)
𝑀𝐴𝐶 HOMO is at -5.26 eV, which is 30 meV deeper than 𝐴𝑢𝐶𝑧𝐶𝑁
𝑀𝐴𝐶 and 200 meV
shallower than 𝐴𝑢𝐶𝑧(𝐶𝐹3)(𝑡𝐵𝑢)
𝐴𝑑𝐶𝐴𝐴𝐶
.
2, 119
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼 HOMO energy is 60 meV shallower compared to 𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼
, similar to all cMa complexes
with bim compared to Cz.93 The dependence of the absorption energy on the oxidation and reduction
potentials of the complex is consistent with an interligand charge transfer (ICT) transition for these
complexes, as seen for other cMa complexes.
Compound
Bond length Bond angle
Ccarbene-AuNamide (˚)
Dihedral
angle
Ccarbene-AuNamide (˚)
Angles
CCDC Ccarbene-Au
(Å)
Au-Namide
(Å)
Ccarbene /
Namide (˚)
𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼 1.984(4) 1.999(3) 174.9(2) 8.1 360/359 2320329
𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 1.979(3) 2.016(2) 176.4(9) 3.9 360/358 2320330
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼 1.972(4) 2.026(3) 169.8(1) 3.5 360/360 2325219
Figure 4-2. Thermal ellipsoid plots 𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼 (left), 𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 (center) and 𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼 (right)

69
Photophysical properties
The UV-visible absorption and emission spectra of the complexes were recorded in a toluene solution
(Figure 4-4). The absorption spectra of all complexes display structured bands at high energy (BCz: λ
< 350 nm, bim: λ < 380 nm) that are assigned to -

transitions localized on the carbene and amide
ligands. Broad featureless absorption bands at lower energy (λ  370 nm) are assigned to transitions
from the ICT state. The ICT state of 𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 is redshifted 10nm compared to 𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼 , due to the
slightly shallower HOMO of Cz(CF3)2. 𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼 has a stabilized LUMO compared to 𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝐴𝐶
leading to a 10 nm redshift of the ICT band. Luminescence is broad and featureless for all complexes
and all solvents except MeCy, where emission is vibronically structured (Figure 4-4 and Figure 4-6),
which is similar to previous literature.10, 30, 33, 91 𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 has a 18 nm redshifted emission compared
to 𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼 in toluene.
Figure 4-3. CV and DPV curves in CH2Cl2 for 𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 and 𝐴𝑢𝐶𝑧(𝐶𝐹3)(𝑀𝑒)
𝑀𝐴𝐶
Table 4-2. Electrochemical data. Measurements were performed using 0.1 M TBAPF6 electrolyte in
DMF (except where noted), and the potentials are listed relative to a ferrocene internal reference.
compound oxidation
(V)
reduction
(V)
redox gap
(V)
HOMOd
(eV)
LUMOd
(eV)
𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼

ref 3
0.86 -1.87 2.74 -5.78 -2.62
𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 0.80 -2.10 2.90 -5.71 -2.35
𝐴𝑢𝐶𝑧(𝐶𝐹3)(𝑀𝑒)
𝑀𝐴𝐶 0.41 -2.55 2.96 -5.26 -1.82
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼 0.80 -2.05 2.85 -5.71 -2.41
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝐴𝐶 0.84 -2.36 3.20 -5.76 -2.05
d
calculated using the equations: HOMO = -1.15(Eox) − 4.79; LUMO = -1.18(Ered) – 4.83
according to reference5
.

70
2-coordinate cMa complexes are known to have a strong solvatochromism, due to large differences in
the magnitude and direction of the dipole moments for the ground and excited states.
2, 10, 91 As described
in Chapter 1 and 3 with increasing polarity of solvents large hypsochromic shifts are measured in
absorption and smaller bathochromic shifts in emission. The complexes with EWG at the amide show
the same solvatochromic behavior, as the complexes without EWGs. The complexes with Bim(CN)2
ligand show a to be expected shifts in Absorption, but in Emission, there was no shift in emission going
from MeTHF to CH2Cl2. In MeCN the emission redshifted to 400 nm, which is likely arising from a
degradation of the Bim(CN)2 complexes. To quantify the solvatochromic shifts, absorption and
emission maxima were plotted against the ET(30) scale and fitted linearly (see Chapter 3) (Figure 4-5).
To compare the complexes with EWGs on the amide, 𝐴𝑢𝐵𝐶𝑧
𝐵𝑍𝐼 to compare the carbazole complexes and
𝐴𝑢𝐵𝑖𝑚
𝐵𝑍𝐴𝐶 to compare the bim complexes, were chosen, due to their similar absorption and emission
energies. The carbazole complexes have all similar slopes in absorption. The emission slopes are
slightly getting steeper from 𝐴𝑢𝐵𝐶𝑧
𝐵𝑍𝐼 to 𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼 and to 𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼
. The bim complexes have very
similar absorption slopes compared to each other and compared to carbazole. The emission slopes on
the other hand are also steeper for the BimCN2 complexes than 𝐴𝑢𝐵𝑖𝑚
𝐵𝑍𝐴𝐶 with 𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼 having a 2.3
times steeper slope than 𝐴𝑢𝐵𝑖𝑚
𝐵𝑍𝐴𝐶
. Adding EWGs to the amide, does not change the solvatochromic
behavior of the Cz and bim complexes. On the other hand, EWGs are increasing the bathochromic shift
in emission with the CF3-group having a greater influence than the CN-group. This arises from a greater
stabilization of the excited state in more polar solvents due to the EWGs. This effect is also greater for
Figure 4-4. Absorption and emission spectra of cMa complexes with CzCN2 and Cz(CF3)2 (a,
b) and bimCN2 (c, d) in toluene and polystyrene.

71
the bim complexes than the Cz complexes, likely due to the lower rotational barrier of bim, which
allows a greater stabilization of the excited state.
Emission spectra from cMa complexes are distinguished by large rigidochromic shifts upon going from
fluid solutions to frozen media at low temperatures. Likewise, luminescence spectra for the complexes
with modified EWGs recorded in fluid 2-MeTHF and methylcyclohexane (MeCy) change markedly
upon cooling solutions to frozen glass at 77 K (Figure 4-6). At room temperature all complexes have
featureless emission in MeTHF, which shifts to a highly structured blue shifted emission at 77K. The
frozen solvents reorganizes around the ground state dipole moment, and cannot rearrange upon
excitation. This leads to a destabilization of the excited state, due to the opposing oriented excited state
dipole moment. Therefore the excited state CT band is destabilized further than the amide ligand
localized (LE) excited state. Emission is showing a clear Cz(CN)2 and Bim(CN)2 emission. Both
Bim(CN)2 complexes are showing LE emission of the 3Bim(CN)2 state, something which was not being
observed in the parent bim complex 𝐴𝑢𝐵𝑖𝑚
𝐵𝑍𝐴𝐶. LE spectra of Bim and Bim(CN)2 ligand in MeTHF at
77K show a redshift of the E00 from 365 nm to 408 nm (Figure 4-7). This stabilization of the LE arises
from an elongation of the -system onto the CN-groups. On the other hand, the 3Cz triplet has the same
energy than 3Cz(CN)2 and (E00= 408 nm) and 3Cz(CF3)2 is destabilized by 6 nm (E00= 402nm). This
isoenergetic character arises from a node in the HOMO orbitals in the 3,6 position of Cz, causing no
participation of the CN-group in the HOMO. In Bim(CN)2 the HOMO electron density spreads onto
the CN-groups (see insets in Figure 4-7). In MeCyHex at 77K the emission spectrum of 𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼 and
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼 is featureless and therefore still emitting from via CT. The emission spectra of
𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 and 𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝐴𝐶 is broadening with a clear shoulder on the blue side of the CT band, which
correlates with an emission from an amide-LE. Due to the lower polarity of MeCyHex (compared to
MeTHF), the CT excited state is less destabilized, leading to an emission with CT and LE character.

Figure 4-5. Absorption and emission maxima vs solvent polarity (ET(30) scale) for (a)
𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼 (green), 𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 (black) & 𝐴𝑢𝐵𝐶𝑧
𝐵𝑍𝐼 (blue ) and (b) 𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼 (red), 𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝐴𝐶
(blue) & 𝐴𝑢𝐵𝑖𝑚
𝐵𝑍𝐴𝐶 (purple)

72
Figure 4-6. Absorption and emission spectra of 𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼 (a,b), 𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼
(c,d),
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼 (e,f) and 𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝐴𝐶 (g,h) in the solvents MeCy, Toluene, MeTHF, CH2Cl2 and
MeCN at room temperature and 77K.

73
Emission of these complexes in the ridged media PS show broad and featureless emission spectra for
all complexes. Emission in PS is blue shifted to the one in Toluene for all complexes. 𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 and
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼 are shifting 40 nm, compared to only 25nm for 𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼 and 𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝐴𝐶 . The larger shift
indicates a larger change of the dipole moment in the excited state, leading to a larger destabilization
of the excited state ICT and therefore a blue shift. Due to the restriction in rotation around the CarbeneM-Amide bond, due to the ridged matrix, the emission spectra have a narrower FWHM then the same
complex in Toluene solution. Upon cooling to 77K both Cz complexes are blue shifted by 10 nm and
𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼 has a shoulder on the blue side of the emission spectra, indicating a close lying 3Cz LE-state.
Both Bim(CN)2 complexes show have an emission max at 420 nm indicating a 3bim contribution to the
emission. This contrasts with the parent Bim complex 𝐴𝑢𝐵𝑖𝑚
𝐵𝑍𝐴𝐶 and caused by the stabilization of the
3Bim-LE state. In a more polar matrix (PMMA) the absorption is blue shifting by around 15 nm for all
complexes compared to PS. The emission is unchanged in shift and spectral width. This is true at room
temperature and 77K.
Like solutions, the polarity of the matrix can be changed. Unlike a fluid solvent, the rigid polymer
matrix cannot rearrange at room temperature so the “solvent” environment around each emitter remains
fixed, leading to a small Stokes shift and a net blue shift in absorption and emission by increasing the
polarity of the polymer matrix. In the less polar polymer matrix TOPAS (fully aliphatic copolymer of
ethylene and norbornene) the CT band in the absorption is redshifted by 5 nm to the PS spectra. Both
Cz compounds have a 5nm redshifted emission in TOPAS compared to PS at 298K. On the other hand,
their 77K spectra is peaking at the same wavelength than the PS 77K emission, but the shoulder at 420
nm in PS is not visible in the TOPAS spectra. The emission spectra of the Bim(CN)2 compounds are
redshifted by 10 nm at room temperature. Upon cooling to 77K the 𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼 spectrum blueshifts by
20 nm and keeps a typical featureless broad character, which indicates an emission from a CT state.
The spectrum of 𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝐴𝐶 is blue shifting to 420 nm and showing characteristic highly structured
triplet LE emission.
Figure 4-7. Emission Spectra in MeTHF at 77K of the amides (left) Cz, Cz(CN)2 and Cz(CF3)2
and (right) Bim and Bim(CN)2.The insets in both graphs show the electron density of the hole
of Cz, Cz(CN)2, Bim and Bim(CN)2 (Iso value: 0.1)

74
cMa complexes are known to have high photoluminescence quantum yields and radiative rates in the
order of up to 4x106
s
-1
. All complexes, like their parent complexes, can have near unity quantum yields
in nonpolar solvents. Quantum yields and emission lifetimes decrease with increasing solvent polarity
(Table 4-3). All complexes have very similar quantum yields to their parent complexes 𝐴𝑢𝐵𝐶𝑧
𝐵𝑍𝐼 and
𝐴𝑢𝐵𝑖𝑚
𝐵𝑍𝐴𝐶
. The Bim(CN)2 complexes have in most solvents lower quantum yields then the Cz complexes.
All complexes show monoexponential excited state lifetime decay curves, indicating emission from
only the ICT state. Solvent molecules are rearranging upon excitation around the excited state dipole
moment, and therefore stabilizing the ICT below all LE states. Cz complexes have higher radiative
rates in all solvents than it parent 𝐴𝑢𝐵𝐶𝑧
𝐵𝑍𝐼, except in the coordinating solvent MeCN. Bim complexes
have, except in Toluene, lower quantum yields and shorter lifetimes. But their radiative rate in MeCy
and toluene is greater than the one of the Cz complexes. Both Bim complexes have similar lifetimes
than their parent but has lower radiative and higher nonradiative rates in all solvents, due to the lower
quantum yields.
In ridged media all complexes have similar quantum yields between 70-90% with non-reaching >95%
(Table 4-4). Cz complexes have monoexponential lifetimes in TOPAS and 𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 in toluene. In
PMMA both are biexponential with shorter component in the 107
s and a slower in the 106
s magnitude.
Figure 4-8. Emission and excitation spectra of (a), in the rigid polymer matrixes TOPAS, PS
and PMMA at 298K and 77K.

75
This character arises from a participation of a ligand centered 3Cz state in the emission process. The
monoexponential lifetime of 𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 in PS arises either from a greater difference between CT and
3Cz band, a lower aggregation of the complex due to the CF3-group or a combination thereof. Both
Bim(CN)2 complexes show in all polymers multiexponential emission decays with the fast component
being in the 300 ns range and the slower component being around 1s. The character arises from a
participation of a ligand centered 3Bim state in the emission process or from a aggregation in the
polymer matrices due to the lower solubility of the Bim(CN)2 complexes compared to the parent and
the Cz complexes.
Unlike cMa complexes discussed in Chapter 2 and 3, EST and 𝜏𝑆1
, to probe the reason behind the fast
radiative rate and their changes to their parent complexes, could not be determined due to the
biexponential excited state decay curves in PS films. This was true for all four complexes.
Table 4-3. Photophysical data for mono- and bimetallic cMa complexes in solution
Complexes
Abs
λmax
(nm)
PL λmax
(nm)
ΦPL
(%)
τ
(ns)
kr
(106
s
-
1
)
knr
(106
s
-
1
)
λmax
77K
(nm)
τ 77 K
(s) (%)
MeCy
𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼 422 440 71 476 1.5 0.6 446 59 (0.7)
355 (0.3)
𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 440 457 82 439 1.9 0.4 450 63
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼 420 440 53 223 2.4 2.1 444
29 (58)
722 (42)
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝐴𝐶 404 450 100 436 2.3 0 444 21 (57)
427 (43)
Toluene
𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼 402 488 99 338 2.9 0.03 - -
𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 414 506 93 414 2.2 0.2 - -
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼 402 495 100 290 3.4 <0.01 - -
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝐴𝐶 392 470 100 228 4.4 <0.01 - -
2-MeTHF
𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼 384 510 95 368 2.6 0.1 419 -
𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 394 522 93 461 2.0 0.2 418 376 (68)
82 (32)
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼 385 535 68 304 2.2 1.1 418 881(58)
2235 (42)
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝐴𝐶 375 500 59 278 2.1 1.5 418 748 (36)
1830 (64)
CH2Cl2
𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼 374 528 90 362 2.5 0.3 - -
𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 388 551 70 332 2.1 0.9 - -
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼 380 540 53 254 2.1 1.8 - -
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝐴𝐶 372 490 67 356 1.9 0.9 - -
MeCN

76
Table 4-4. Photophysical data for mono- and bimetallic cMa complexes in different polymer matrices.
𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼 360 572 27 127 2.1 5.8 - -
𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 360 595 9 58 1.6 15.7 - -
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼 375 400 <1 - - - - -
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝐴𝐶 368 (400) 42 - - - - -
Complexes
Abs
λmax
(nm)
PL
λmax
(nm)
ΦPL
(%)
τ
(s)
kr
(106
s
-
1
)
knr
(106
s
-
1
)
λmax
77K
(nm)
τ 77 K
(s) (%)
Topas/Zeonex
𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼 414 472 85 486 1.7 0.3 450 67
𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 422 478 90 431 2.1 0.2 458 61
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼 410 466 73 268 (83)
823 (17) - - 448 57 (61)
16 (39)
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝐴𝐶 400 466 79 466 (49)
1957 (51) - - 423 85 (20)
438 (80)
PS
𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼 404 463 79 550 (68)
1750 (32) 1.5 0.4 450 59 (71)
123 (29)
𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 412 465 82 470 1.7 0.4 455 53
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼 402 456 78
310 (61)
1419 (26)
7554 (13)
0.5
(1013
ns)
0.14 438 414 (57)
38 (43)
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝐴𝐶 396 446 55 315 (43)
1540 (57) - - 422 431 (80)
78 (20)
PMMA
𝐴𝑢𝐶𝑧(𝐶𝑁)2
𝑃𝑍𝐼 387 440 -
944 (50)
6854 (50) - - - -
𝐴𝑢𝐶𝑧(𝐶𝐹3)2
𝑃𝑍𝐼 395 462 90 623 (76)
2175 (24) - - - -
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝑍𝐼 385 454 78 434 (51)
4222 (49) - - 422 590 (72)
43 (28)
𝐴𝑢𝐵𝑖𝑚(𝐶𝑁)2
𝑃𝐴𝐶 382 449 74
813 (37)
10453
(63)
- - 420 870 (80)
198 (20)

77
Attempts to obtain Bim(CF3)2
Bim has a significant higher energetic triplet than Cz, and therefore can form deep blue emitting cMa
complexes with very fast radiative lifetimes. Attaching CN to Bim, lowers the HOMO of Bim, but are
extending the -system and therefore lowering the triplet energy significantly, causing the complexes
to have a mixed CT and LE emission in ridged matrixes. CF3 groups have a similar electron
withdrawing character like CN groups, but are not able to extend the p-system and should therefore be
the better choice to achieve blue emitting cMa’s. During this work there were many attempts to
synthesize Bim(CF3)2, but no one was successful. In the following schemes the unsuccessful attempts
are summarized:
Different trifluoromethylation agents were tested according to the following publications.163, 166-168
None of them yielded in Bim(CF3)2. Reaction conditions and ratios were first used as in the publication
and varied after that. Solvents, temperature, addition of CuI was varied, but non yielded in Bim(CF3)2
This Synthesis route was designed to obtain Bim(CF3)2 by synthesizing bim with CF3 containing
starting materials. The 1st step worked in 50% yield and the 2nd step in 85% yield. Unfortunately the
triple ring closure utilizing triphosgene did not work
This synthesis is the only synthesis, which gives the Bim(CF3)2, but it forms as two different isomers,
which have identical NMR spectra. The separation of both isomers was not successful. Column
chromatography did not separate the two isomers. Taking the isomeric mixture into a metalation step
resulted in the following reaction:

78
This reaction yielded a blue emissive compound as a solid. By NMR both products can be identified,
but separation by washing, recrystallization and sublimation was not successful.
Conclusion and Outlook
In summary, a new group of cMa complexes, utilizing electron withdrawing groups on the amide is
being presented. The synthesis of BimI2 and BimBr2 to obtain Bim(CN)2, opens pathways to a range
of new complexes utilizing the Bim-amide in a modified way. The EWG leads to a stabilization of the
HOMO, allowing to obtain blue emitting complexes with the carbenes PZI and PAC, leading to blue
emitters with lower HOMO and LUMO energies than similarly emitting Cz or Bim complexes.
Luminescence is blue and occurs via TADF with high PL efficiencies. The EWG groups does not
change their solvatochromic and rigidochromic character. Radiative rates are similar or higher than
their parent. In ridged media all complexes have a 3LE state close by the CT state, leading to longer
and biexponential decays. Obtaining Bim(CF3)2 as amide, would open a new group of blue emitters,
which should have a higher 3Bim triplet state, due to having bim and a non--extending EWG group.

79
Chapter 5 - Temperature-dependent Photophysics
Design and Set-Up of Cryostat
To be able to understand the cMa complexes better and increase their radiative rate further, it is
necessary to determine the energy differences between the excited states (EST and ZFS) (see Figure
1-1) and the radiative lifetimes of the singlet and triplet excited states. This can be done using
temperature dependent lifetime measurements from 3-300K. To obtain these measurements a new
Cryogenic Probe for variable temperature dependent Photophysics was designed and installed.
Cryogenic photophysical measurements were carried out on closed-cycle Helium Cryostat with solid
copper coldhead (Janis SHI-4-2) with a Lakeshore 335 Temperature controller and evacuated by a
Drytel 31 Turbomolecular pump to 1.2*10-4 mTorr.
The manufacturer designed the cryostat, that excitation and emission should be in a 45degree angle to
the sample (see Figure 5-1(a)). This set up caused that a significant amount of scattered excitation light
could be detected in the emission spectra, which does not allow to measure an emission spectrum
starting within 15 nm of the excitation light. The sample holder was modified in a way, that it is able to
measure the light, which is sidewise outcoupled through the sapphire substrate (see Figure 5-2), on the
edge of the substrate. This allows the sample to be mounted perpendicular to the excitation source (see
Figure 5-1(b)). This set up allows to measure emission spectra starting 5 nm to the red of the excitation
maximum (LED Thorlabs: M365L3).
Figure 5-1. (a) 45 degree set up and (b) 90 degree modified set up

80
The sample holder is designed to fit 1 inch diameter optical windows. To ensure good thermal
conductivity and accurate temperature control at the sample, instead of quartz (2 Wm-1K-1
) Saphire
windows (46 Wm-1K-1
) are used. To analyze a compound a <0.5 wt% doped polymer film is used.
Polymer (PS, TOPAS or PMMA) is dissolved in Toluene (400mg in 4mL), drop casted onto the
substrate, dried for 2h and afterwards dried in vacuum overnight. If the doped polymer is drop casted
onto one substrate (Figure 5-3(a)) and mounted in the cryostat, it is cracking and falling of the substrate
upon cooling (Figure 5-3(b)). This results in a temperature offset, and therefore the sample is at a higher
temperature than the cryostat is set to. This problem could be solved by using two 1mm thick sapphires,
sandwiching the polymer film (Figure 5-3(c)).
To ensure that temperature at the sample is accurate, a PMMA film with Ir(ppy)3 was used as standard.
The temperature dependance of the lifetime of Ir(ppy) was thoroughly studied by Yersin et. al. and is
used as a reference.169 As an additional check the sample was measured in a cooling and a warming
cycle, and both curves lined up without offset, indicating an accurate temperature control. Sample
temperature equilibrates in less than one minute.
The cryostat is connected to a TCSPC and a Fluorimeter, in order to allow to collect both spectra at
each temperature. The light collection set-up is shown in Figure 5-4(a). The emitted light was
collimated by an Edmund Optics 45-716 Lens 1 (Focus: 75mm, Ø 50 mm, VIS-NIR coating for 400-
Figure 5-3. (a) 2mm thick sapphire with doped polymer. The edge is polymer free to ensure a
better cold head to Substrate contact, (b)peeling off polymer film upon cooling, (c) sandwiched
polymer film between two 1mm thick Sapphire substrates.
Figure 5-2. manufacturer sample holder (left) and modified sample holder (core: copper,
surface: gold plating) with a cutout (5x2mm) on the right side, side to detect the emitted light

81
1000nm) and focused with an Edmund Optics 47-393 Lens 2 (Focus: 125mm, Ø 50 mm, VIS-NIR
coating for 400-1000nm) Lens onto a Thorlabs BF13LSMA02 (400-2200 nm, Ø 1.3 mm) optical fiber
connected to a Thorlabs bifurcated fiber BF19Y2LS02 (250-1200 nm, 19 Fiber, Ø 200 m) with 10
fibers ending in the TCSPC and 9 Fibers in the fluorimeter. The entire collection unit is mounted on a
XYZ-Stage to align the lenses. The optical fiber is mounted onto an independent linear stage, to fine
adjust the fiber position and is partly inside copper pipes to prevent bending and breakage of the fibers.
Emission spectra were collected with a Photon Technology International QuantaMaster
spectrofluorimeter. As Excitation source a 365nm LED (Thorlabs M365L3 365nm LED, 1000 mW)
equipped with the Thorlabs SM1U25-A Adjustable Collimation Adapter and driven with a Thorlabs
LEDD1B driver.
For luminescence lifetime studies a Horibia Fluorohub+ with a Horiba Jobin Yvon detector with
monochromator was used. As excitation source a NanoLED 407N (405 nm) or IBH SpectraLED S-03
(372 nm) was used.
The Cryostat is incased by a black-out curtain and walls lined with black fabric to reduce the amount
of background light and making it possible to measure even very dim samples (see Figure 5-4(b)).
Figure 5-4. (a) detected light is collimated and focused on an optical fiber, (b) Cryostat is
enclosed by a black box to reduce background light

82
Fitting of the data
Values for the measured lifetime (s) were plotted against temperature (Figure 5-5). The Values were
fitted using a 3-level Boltzman fit using the following equation:
With this equation the Energy difference between T3 and S1 (EST) and the energy differences between
the triplet substates T1, T2 and T3 (ZFS) as well as the excited state lifetimes of the individual excited
states. The used cryostat cannot cool below 3K, which is for most complexes not enough to resolve T1
and T2 and therefore there were treated degenerate. Statistical weighing was used in fits to prevent a
higher contribution of long lifetimes. Not using statistical weighing undervalues the high temperature
lifetimes and the fit would poorly align with the high temperature data.
In case only high temperature data above 150K could be obtained (e.g. Au𝐵𝐶𝑧
BZAC), an Arrhenius plot was
used to determine EST and 𝜏𝑆1
. This is based on the following equation:
𝜏𝑇𝐴𝐷𝐹(𝑇) = 3𝜏𝑆1
𝑒
∆𝐸𝑆𝑇
𝑘𝐵𝑇
Which can be transformed into a linear equation:
(
1
𝜏𝑇𝐴𝐷𝐹
) = −
∆𝐸𝑆𝑇
𝑘𝐵
1
𝑇
+ ln(
1
3𝜏𝑆1
)
The lifetime data from 150 K – 300 K were plotted as an Arrhenius plot as ln(kTADF) vs. 1/T (Figure
5-6) and fitted linear. The linear fit allows to obtain EST from the slope and S1 from the y-intercept:
∆𝐸𝑆𝑇 = −𝑘𝐵(𝑠𝑙𝑜𝑝𝑒)
𝜏𝑆1 =
1
3 𝑒
𝑦−𝑖𝑛𝑡𝑒𝑟𝑠𝑒𝑝𝑡

83
Figure 5-5: Temperature dependent lifetime data of 𝐴𝑢𝐵𝑖𝑚
𝐵𝑍𝐼 doped PS thin film. In red is the fit of
the data according to the Equation described above.
Figure 5-6:Ahrrenius plot of temperature dependent lifetime of 𝐴𝑢𝐵𝑖𝑚
𝐵𝑍𝐼 doped PS thin film. In
red is the linear fit with the equation given in the bottom left

84
Collection of all measured cMa complexes
Table 5-1:summary of all analyzed 2-coordinate cMa complexes, which were measured throughout this
PhD thesis. Compounds were measured in a <0.1wt% doped Polystyrene Films.
cpd EST (meV) 𝜏𝑆1
(ns) ZFS (meV) 𝜏𝑇1
𝐼𝐼𝐼 (s) 𝜏
𝑇1
𝐼,𝐼𝐼 (s) Literature
Au𝐶𝑧
𝐴𝑑CAAC 91 13 1.73 21 140 170, 171
Cu𝐶𝑧
𝐴𝑑CAAC 114 18 0.9 26 75 170, 171
Au𝐶𝑧
MAC 87 13 1.2 28 -
170, 171
Cu𝐶𝑧
MAC 80 22 1.3 246 383 171
Cu𝑃ℎ𝐶𝑧
𝑀𝐴𝐶 67 16 0.5 277 830 -
Au𝐶𝑧
𝐵𝑍𝐴𝐶 65 18 - - -
93
Au𝐵𝐶𝑧
𝐵𝑍𝐴𝐶 70 14 1.0 47 91
Au𝐵𝐶𝑧
𝑃𝐴𝐶 72 14 0.9 36 -
93
Ag𝐵𝐶𝑧
𝑃𝐴𝐶 26 35 - - -
93
Cu𝐵𝐶𝑧
𝑃𝐴𝐶 70 18 - - -
93
Au𝐶𝑧
𝑃𝑍𝐼 54 21 0.9 82 -
93
AuBim
BZI 41 15 1.5 15 -
93
Au𝐵𝑖𝑚
𝐵𝑍𝐴𝐶 41 19 1.2 19 -
93
Au𝐵𝑖𝑚
𝐴𝑑CAAC 53 18 1.1 16 -
93
Au𝐵𝑖𝑚
𝑀𝐴𝐶 51 14 2.0 38 -
93
Au𝐵𝑖𝑚
𝑃𝐴𝐶 45 17 1.0 38 -
93
Au𝐵𝑖𝑚
𝑃𝑍𝐼 31 26 1.1 76 -
93
AuMBim
BZI 44 16 1.5 11 -
93
AuOBim
BZI 41 19 1.2 17 -
93
AuDMBim
BZAC 36 16 3.6 38 - -
BCzAuBBIAuBCz 43 13 1.1 33 -
91
BCzAuBAZAuBCz 54 13 0.9 33 -
91
BimAuBAZAuBim 20 50 - - -
91

85
0 50 100 150 200 250 300
0
20
40
60
80
100
120
140 CAAC-Au-Cz Lifetime (us)
Temperature (K)
Model TADFBoltzmann (User) Equation
(2+exp(ZFS/(-0.0000861733*x)
)+exp(dEst/(-0.0000861733*x)))
/((2/T12)+((exp(ZFS/(-0.000086
1733*x)))/T3)+((exp(dEst/(-0.00
00861733*x)))/S1)) Plot Lifetime ZFS 0.00173 ± 6.09333E-5
dEst 0.09094 ± 0.00192 T12 137.49178 ± 1.87768 T3 21.10015 ± 0.34664 S1 0.01275 ± 0.00157 Reduced Chi-Sqr 0.04858 R-Square (COD) 0.9989 Adj. R-Square 0.99878
50 100 150 200 250 300 350
0
10
20
30
40
50
60
70 CAAC-Cu-Cz Lifetime (us)
Temperature (K)
Model TADFBoltzmann (User) Equation
(2+exp(ZFS/(-0.0000861733*x)
)+exp(dEst/(-0.0000861733*x)))
/((2/T12)+((exp(ZFS/(-0.000086
1733*x)))/T3)+((exp(dEst/(-0.00
00861733*x)))/S1)) Plot Lifetime ZFS 8.92532E-4 ± 1.0832E-4
dEst 0.11364 ± 0.00194 T12 75.3144 ± 2.8461 T3 25.80218 ± 0.59223 S1 0.01787 ± 0.00176 Reduced Chi-Sqr 0.01968 R-Square (COD) 0.99913 Adj. R-Square 0.99903
0 50 100 150 200 250 300
0
20
40
60
80
100
120
140
160
180
200
220
240
260
280
300 MAC-Au-Cz Lifetime (us)
Temperature (K)
Model FourLevelFit Equation
(1+(exp(-dE2/(kB*x)))+(exp(-dE
3/(kB*x)))+(exp(-dE4/(kB*x))))/(( 1/tau1)+((1/tau2)*exp((-dE2)/(k B*x)))+((1/tau3)*exp((-dE3)/(kB* x)))+((1/tau4)*exp((-dE4)/(kB*x))
)) Plot Lifetime
dE2 0 ± 1.00196
dE3 10.07374 ± 0.91846
dE4 616.30589 ± 0
tau1 500 ± 0
tau2 207.01161 ± 29.14239
tau3 28.35514 ± 0.49419
tau4 0.02161 ± 0 Reduced Chi-Sqr 0.0328 R-Square (COD) 0.99954 Adj. R-Square 0.99951 0.0030 0.0035 0.0040 0.0045 0.0050
12.0
12.5
13.0
13.5
14.0 MAC-Au-Cz ln k (s-1
)
1/T (K-1)
Equation y = a + b*x Plot ln k Weight No Weighting
Intercept 17.09229 ± 0.06507 Slope -1011.17517 ± 16.43144 Residual Sum of Squares 0.01263 Pearson's r -0.99855 R-Square (COD) 0.9971 Adj. R-Square 0.99684
0 50 100 150 200 250 300
0
50
100
150
200
250
300
350 MAC-Cu-Cz Lifetime (s)
Temperature (K)
Model TADFBoltzmann (User) Equation
(2+exp(ZFS/(-0.0000861733*x)
)+exp(dEst/(-0.0000861733*x)))
/((2/T12)+((exp(ZFS/(-0.000086
1733*x)))/T3)+((exp(dEst/(-0.00
00861733*x)))/S1)) Plot Lifetime ZFS 0.00293 ± 8.45239E-4
dEst 0.08989 ± 0.00261 T12 340.76324 ± 6.54527 T3 175.68475 ± 12.42758 S1 0.01843 ± 0.00362 Reduced Chi-Sqr 0.48344 R-Square (COD) 0.99711 Adj. R-Square 0.99678
0.003 0.004 0.005 0.006 0.007 0.008
9.5
10.0
10.5
11.0
11.5
12.0
12.5
13.0
13.5
14.0 MAC-Cu-Cz ln k (s-1
)
1/T (K-1)
Equation y = a + b*x Plot ln k Weight No Weighting
Intercept 16.526 ± 0.0581 Slope -924.60674 ± 11.64542 Residual Sum of Squares 0.07477 Pearson's r -0.9985 R-Square (COD) 0.997 Adj. R-Square 0.99684
Figure 5-7: Temperature dependent lifetime data and Arrhenius plot of 𝐴𝑢𝐶𝑧
𝐶𝐴𝐴𝐶
, 𝐶𝑢𝐶𝑧
𝐶𝐴𝐴𝐶
, 𝐴𝑢𝐶𝑧
𝑀𝐴𝐶 and
𝐶𝑢𝐶𝑧
𝑀𝐴𝐶 doped PS thin film. In red is the fits and the fitting parameters are stated in the insets in each
figure.

86
50 100 150 200 250 300
0
100
200
300
400
500
600
700
MAC-Cu-dPhCz Lifetime (s)
Temperature (K)
Model TADFBoltzmann (User) Equation
(2+exp(ZFS/(-0.0000861733*x)
)+exp(dEst/(-0.0000861733*x)))
/((2/T12)+((exp(ZFS/(-0.000086
1733*x)))/T3)+((exp(dEst/(-0.00
00861733*x)))/S1)) Plot Lifetime ZFS 4.75489E-4 ± 2.35036E-4
dEst 0.067 ± 0 T12 829.79967 ± 166.76784 T3 276.83243 ± 33.09529 S1 0.016 ± 0 Reduced Chi-Sqr 0.56083 R-Square (COD) 0.99758 Adj. R-Square 0.99743
0.003 0.004 0.005 0.006 0.007 0.008 0.009 0.010 0.011
9
10
11
12
13
14
15
MAC-Cu-dPhCz ln k (s-1
)
1/T (K-1)
Equation y = a + b*x Plot ln k Weight No Weighting
Intercept 16.83562 ± 0.03281 Slope -780.48255 ± 5.64206 Residual Sum of Squares 0.04988 Pearson's r -0.9995 R-Square (COD) 0.99901 Adj. R-Square 0.99896
0.0032 0.0034 0.0036 0.0038 0.0040
13.7
13.8
13.9
14.0
14.1
14.2
14.3
14.4 BZAC-Au-Cz ln k (s-1
)
1/T (K-1)
Equation y = a + b*x Plot ln k Weight No Weighting
Intercept 16.73805 ± 0.08912 Slope -749.52203 ± 25.0028 Residual Sum of Squares 0.00685 Pearson's r -0.99339 R-Square (COD) 0.98682 Adj. R-Square 0.98572
10 100
0
20
40
60
80
100
120
140 BCzAuBZAC Lifetime (s)
TEMP (K)
ZFS (eV) 0.0013 ± 2.33417E-4
dEst (eV) 0.0699 ± 0 T12 (us) 147.09143 ± 4.83098 T3 (us) 51.22748 ± 2.21458 S1 (us) 0.014 ± 0 0.0034 0.0036 0.0038 0.0040 0.0042 0.0044 0.0046
13.2
13.4
13.6
13.8
14.0
14.2
14.4 BCzAuBZAC ln(k) (s-1
)
1/T (K-1)
Intercept 16.98419 ± 0.08154 Slope -807.82585 ± 20.90 Residual Sum of Squar 0.00884 Pearson's r -0.99634 R-Square (COD) 0.99269 Adj. R-Square 0.99202
Figure 5-8: Temperature dependent lifetime data and Arrhenius plot of 𝐴𝑢𝑃ℎ𝐶𝑧
𝑀𝐴𝐶
, 𝐴𝑢𝐶𝑧
𝐵𝑍𝐴𝐶 and 𝐴𝑢𝐵𝐶𝑧
𝐵𝑍𝐴𝐶
doped PS thin film. In red is the fits and the fitting parameters are stated in the insets in each figure.

87
50 100 150 200 250 300
0
25
50
75
100
125
150 PAC-Au-tBuCz Lifetime (us)
Temperature (K)
Model TADFBoltzmann (User) Equation
(2+exp(ZFS/(-0.0000861733*x))+ex p(dEst/(-0.0000861733*x)))/((2/T12) +((exp(ZFS/(-0.0000861733*x)))/T3) +((exp(dEst/(-0.0000861733*x)))/S1
)) Plot Lifetime ZFS 6.70102E-4 ± 3.64747E-5
dEst 0.07054 ± 0 T12 1.19985E16 ± 0 T3 32.19452 ± 0.3972 S1 0.0141 ± 0 Reduced Chi-Sqr 6.83389 R-Square (COD) 0.99767 Adj. R-Square 0.99752
0.003 0.004 0.005 0.006 0.007 0.008 0.009
10
11
12
13
14
15 PAC-Au-tBuCz ln k (s-1
)
1/T (K-1)
Equation y = a + b*x Plot ln k Weight No Weighting
Intercept 16.98071 ± 0.05716 Slope -818.59615 ± 10.54988 Residual Sum of Squares 0.05638 Pearson's r -0.99892 R-Square (COD) 0.99785 Adj. R-Square 0.99768
0 50 100 150 200 250 300
0
200
400
600
800
1000
1200 PAC-Ag-tBuCz Lifetime (us)
Temperature (K)
Model TADFBoltzmann (User) Equation
(2+exp(ZFS/(-0.0000861733*x)
)+exp(dEst/(-0.0000861733*x)))
/((2/T12)+((exp(ZFS/(-0.000086
1733*x)))/T3)+((exp(dEst/(-0.00
00861733*x)))/S1)) Plot Lifetime ZFS 1.01284E-4 ± 4.72146E-5
dEst 0.026 ± 0 T12 3.25E15 ± 4.84653E27 T3 281.95443 ± 68.94124 S1 0.035 ± 0 Reduced Chi-Sqr 2.14207 R-Square (COD) 0.99367 Adj. R-Square 0.99331
0.003 0.004 0.005 0.006 0.007 0.008 0.009 0.010 0.011
13.0
13.5
14.0
14.5
15.0 PAC-Ag-tBuCz ln k (s-1
)
1/T (K-1)
Equation y = a + b*x Plot ln k Weight No Weighting
Intercept 16.0697 ± 0.03781 Slope -302.35493 ± 6.29539 Residual Sum of Squares 0.03326 Pearson's r -0.99719 R-Square (COD) 0.9944 Adj. R-Square 0.99396
0 50 100 150 200 250 300
0
200
400
600
800 PAC-Cu-tBuCz Lifetime (us)
Temperature (K)
Model TADFBoltzmann (User) Equation
(2+exp(ZFS/(-0.0000861733*x))+ex p(dEst/(-0.0000861733*x)))/((2/T12) +((exp(ZFS/(-0.0000861733*x)))/T3) +((exp(dEst/(-0.0000861733*x)))/S1
)) Plot Lifetime ZFS 3.80336E-4 ± 2.8678E-4
dEst 0.07031 ± 0 T12 853.80906 ± 136.26163 T3 439.15138 ± 67.02002 S1 0.0176 ± 0 Reduced Chi-Sqr 262.37964 R-Square (COD) 0.99761 Adj. R-Square 0.99747
0.0030 0.0035 0.0040 0.0045 0.0050 0.0055 0.0060 0.0065 0.0070
11.0
11.5
12.0
12.5
13.0
13.5
14.0
14.5 PAC-Cu-tBuCz ln k (s-1
)
1/T (K-1)
Equation y = a + b*x Plot ln k Weight No Weighting
Intercept 16.75676 ± 0.03908 Slope -815.9533 ± 8.01276 Residual Sum of Squares 0.0105 Pearson's r -0.99947 R-Square (COD) 0.99894 Adj. R-Square 0.99884
Figure 5-9: Temperature dependent lifetime data and Arrhenius plot of 𝐴𝑢𝐵𝐶𝑧
𝑃𝐴𝐶
, 𝐴𝑔𝐵𝐶𝑧
𝑃𝐴𝐶 and 𝐶𝑢𝐵𝐶𝑧
𝑃𝐴𝐶
doped PS thin film. In red is the fits and the fitting parameters are stated in the insets in each figure.

88
0 50 100 150 200 250 300
0
50
100
150
200
250
300
350
400 PZI-Au-CZ Lifetime (us)
Temperature (K)
Model TADFBoltzmann (User) Equation
(2+exp(ZFS/(-0.0000861733*x)
)+exp(dEst/(-0.0000861733*x)))
/((2/T12)+((exp(ZFS/(-0.000086
1733*x)))/T3)+((exp(dEst/(-0.00
00861733*x)))/S1)) Plot Lifetime ZFS 8.88688E-4 ± 5.93085E-5
dEst 0.05377 ± 0 T12 381.72268 ± 10.49223 T3 82.29897 ± 1.18367 S1 0.0209 ± 0 Reduced Chi-Sqr 0.10934 R-Square (COD) 0.99897 Adj. R-Square 0.99892 0.003 0.004 0.005 0.006 0.007 0.008
11.5
12.0
12.5
13.0
13.5
14.0
14.5
15.0 PZI-Au-Cz ln k (s-1
)
1/T (K-1)
Equation y = a + b*x Plot ln k Weight No Weighting
Intercept 16.58374 ± 0.03321 Slope -624.00396 ± 6.77037 Residual Sum of Squares 0.02545 Pearson's r -0.99906 R-Square (COD) 0.99812 Adj. R-Square 0.998
0 50 100 150 200 250 300
0
20
40
60
80
100
120 BZI-Au-Bimbim Lifetime (s)
Temperature (K)
Model TADFBoltzmann (User) Equation
(2+exp(ZFS/(-0.0000861733*x))+ex p(dEst/(-0.0000861733*x)))/((2/T12) +((exp(ZFS/(-0.0000861733*x)))/T3) +((exp(dEst/(-0.0000861733*x)))/S1
)) Plot Lifetime ZFS 0.00125 ± 3.41948E-4
dEst 0.04113 ± 0 T12 130.40601 ± 19.89634 T3 18.79193 ± 1.93753 S1 0.0146 ± 0 Reduced Chi-Sqr 1.0786 R-Square (COD) 0.97366 Adj. R-Square 0.97164
0.002 0.003 0.004 0.005 0.006 0.007 0.008 0.009 0.010 0.011
11
12
13
14
15
16 BZI-Au-Bimbim ln k (s-1
)
1/T (K-1)
Equation y = a + b*x Plot ln k Weight No Weighting
Intercept 16.94568 ± 0.08623 Slope -477.30619 ± 14.19551 Residual Sum of Squares 0.05339 Pearson's r -0.99692 R-Square (COD) 0.99385 Adj. R-Square 0.99297
50 100 150 200 250 300
-20
0
20
40
60
80
100
120
140
160
180 BZAC-Au-Bimbim Lifetime (s)
Temperature (K)
Model TADFBoltzmann (User) Equation
(2+exp(ZFS/(-0.0000861733*x))+ex p(dEst/(-0.0000861733*x)))/((2/T12) +((exp(ZFS/(-0.0000861733*x)))/T3) +((exp(dEst/(-0.0000861733*x)))/S1
)) Plot Lifetime ZFS 0.00124 ± 5.2374E-5
dEst 0.04084 ± 0 T12 171.23194 ± 3.3678 T3 19.40224 ± 0.50036 S1 0.0186 ± 0 Reduced Chi-Sqr 4.61935 R-Square (COD) 0.99823 Adj. R-Square 0.99808
0.003 0.004 0.005 0.006 0.007 0.008 0.009 0.010 0.011
11.5
12.0
12.5
13.0
13.5
14.0
14.5
15.0
15.5 BZAC-Au-Bimbim ln k (s-1
)
1/T (K-1)
Equation y = a + b*x Plot ln k Weight No Weighting
Intercept 16.69986 ± 0.08168 Slope -473.99079 ± 13.36437 Residual Sum of Squares 0.09344 Pearson's r -0.99644 R-Square (COD) 0.9929 Adj. R-Square 0.99211
Figure 5-10: Temperature dependent lifetime data and Arrhenius plot of 𝐴𝑢𝐶𝑧
𝑃𝑍𝐼
, 𝐴𝑢𝐵𝑖𝑚
𝐵𝑍𝐼 and 𝐴𝑢𝐵𝑖𝑚
𝐵𝑍𝐴𝐶
doped PS thin film. In red is the fits and the fitting parameters are stated in the insets in each figure.

89
0 50 100 150 200 250 300
0
50
100
150
200
250
300 CAAC-Au-BB Lifetime (s)
Temperature (K)
Model TADFBoltzmann (User) Equation
(2+exp(ZFS/(-0.0000861733*x)
)+exp(dEst/(-0.0000861733*x)))
/((2/T12)+((exp(ZFS/(-0.000086
1733*x)))/T3)+((exp(dEst/(-0.00
00861733*x)))/S1)) Plot Lifetime ZFS 0.00112 ± 2.996E-5
dEst 0.053 ± 0 T12 344.31323 ± 11.92772 T3 15.93309 ± 0.25638 S1 0.018 ± 0 Reduced Chi-Sqr 0.15295 R-Square (COD) 0.9968 Adj. R-Square 0.99669 0.003 0.004 0.005 0.006 0.007 0.008 0.009 0.010 0.011
10
11
12
13
14
15 CAAC-Au-Bimbim ln k (s-1
)
1/T (K-1)
Equation y = a + b*x Plot ln k Weight No Weighting
Intercept 16.62181 ± 0.03875 Slope -601.24343 ± 6.02988 Residual Sum of Squares 0.0906 Pearson's r -0.99885 R-Square (COD) 0.99769 Adj. R-Square 0.99759
0 50 100 150 200 250 300
0
50
100
150
200
250
300 PAC-Au-Bimbim lifetime (us)
Temperature (K)
Model TADFBoltzmann (User) Equation
(2+exp(ZFS/(-0.0000861733*x)
)+exp(dEst/(-0.0000861733*x)))
/((2/T12)+((exp(ZFS/(-0.000086
1733*x)))/T3)+((exp(dEst/(-0.00
00861733*x)))/S1)) Plot TADFBoltzmann (User) Fit of S
heet1 B"lifetime" ZFS 9.84417E-4 ± 1.16221E-5
dEst 0.0447 ± 0 T12 297.79467 ± 2.77043 T3 38.4813 ± 0.11196 S1 0.01695 ± 0 Reduced Chi-Sqr 0.03689 R-Square (COD) 0.99869 Adj. R-Square 0.99869
0.003 0.004 0.005 0.006 0.007 0.008
12.5
13.0
13.5
14.0
14.5
15.0
15.5 PAC-Au-Bimbim ln k (s-1
)
1/T (K-1)
Equation y = a + b*x Plot ln k Weight No Weighting
Intercept 16.82721 ± 0.04291 Slope -518.36482 ± 8.16522 Residual Sum of Squares 0.01895 Pearson's r -0.99864 R-Square (COD) 0.99728 Adj. R-Square 0.99703
Figure 5-11: Temperature dependent lifetime data and Arrhenius plot of 𝐴𝑢𝐵𝑖𝑚
𝐶𝐴𝐴𝐶
, 𝐴𝑢𝐵𝑖𝑚
𝑀𝐴𝐶 and 𝐴𝑢𝐵𝑖𝑚
𝑃𝐴𝐶
doped PS thin film. In red is the fits and the fitting parameters are stated in the insets in each figure.
50 100 150 200 250 300 350
0
20
40
60
80
100
120
140
160
180
200 MAC-Au-Bimbim Lifetime (s)
Temperature (K)
Model TADFBoltzmann (User) Equation
(2+exp(ZFS/(-0.0000861733*x))+ex p(dEst/(-0.0000861733*x)))/((2/T12) +((exp(ZFS/(-0.0000861733*x)))/T3) +((exp(dEst/(-0.0000861733*x)))/S1
)) Plot Lifetime ZFS 0.00113 ± 3.529E-5
dEst 0.05075 ± 0 T12 206.08862 ± 2.52415 T3 33.29432 ± 0.42001 S1 0.014 ± 0 Reduced Chi-Sqr 4.40851 R-Square (COD) 0.99882 Adj. R-Square 0.99876
0.003 0.004 0.005 0.006 0.007 0.008 0.009 0.010 0.011
11
12
13
14
15 MAC-Au-Bimbim ln k (s-1
)
1/T (K-1)
Equation y = a + b*x Plot ln k Weight No Weighting
Intercept 16.98479 ± 0.05798 Slope -588.98955 ± 9.68083 Residual Sum of Squares 0.1206 Pearson's r -0.99771 R-Square (COD) 0.99543 Adj. R-Square 0.99516

90
50 100 150 200 250 300
0
50
100
150
200
250
300 PZI-Au-Bim Lifetime (us)
Temperature (K)
Model TADFBoltzmann (User) Equation
(2+exp(ZFS/(-0.0000861733*x)
)+exp(dEst/(-0.0000861733*x)))
/((2/T12)+((exp(ZFS/(-0.000086
1733*x)))/T3)+((exp(dEst/(-0.00
00861733*x)))/S1)) Plot Lifetime ZFS 0.00139 ± 3.32638E-4
dEst 0.03013 ± 6.55056E-4 T12 286.43167 ± 7.48748 T3 115.70378 ± 8.91613 S1 0.05036 ± 0.00568 Reduced Chi-Sqr 0.30015 R-Square (COD) 0.99663 Adj. R-Square 0.99629
0.0030 0.0035 0.0040 0.0045 0.0050 0.0055 0.0060
14.2
14.4
14.6
14.8
15.0
15.2
15.4 PZI-Au-Bim ln k (s-1
)
1/T (K-1)
Equation y = a + b*x Plot ln k Weight No Weighting
Intercept 16.32419 ± 0.0407 Slope -349.61031 ± 9.43521 Residual Sum of Squares 0.01378 Pearson's r -0.99494 R-Square (COD) 0.98991 Adj. R-Square 0.98919
0 50 100 150 200 250 300
0
20
40
60
80
100 BZI-Au-MeBimbim Lifetime (s)
Temperature (K)
Model TADFBoltzmann (User) Equation
(2+exp(ZFS/(-0.0000861733*x))+ex p(dEst/(-0.0000861733*x)))/((2/T12) +((exp(ZFS/(-0.0000861733*x)))/T3) +((exp(dEst/(-0.0000861733*x)))/S1
)) Plot Lifetime ZFS 0.00147 ± 1.16117E-4
dEst 0.044 ± 0 T12 111.16791 ± 5.46735 T3 10.96691 ± 0.43717 S1 0.0162 ± 0 Reduced Chi-Sqr 0.15993 R-Square (COD) 0.99437 Adj. R-Square 0.99403
0.003 0.004 0.005 0.006 0.007 0.008 0.009 0.010 0.011
11.5
12.0
12.5
13.0
13.5
14.0
14.5
15.0
15.5 BZI-Au-MeBimbim ln k (s-1
)
1/T (K-1)
Equation y = a + b*x Plot ln k Weight No Weighting
Intercept 17.33686 ± 0.1015 Slope -568.66943 ± 17.36296 Residual Sum of Squares 0.28606 Pearson's r -0.99354 R-Square (COD) 0.98712 Adj. R-Square 0.9862
50 100 150 200 250 300 350
0
20
40
60
80
100 BZI-Au-MeOBimbim Lifetime (s)
Temperature (K)
Model TADFBoltzmann (User) Equation
(2+exp(ZFS/(-0.0000861733*x))+ex p(dEst/(-0.0000861733*x)))/((2/T12) +((exp(ZFS/(-0.0000861733*x)))/T3) +((exp(dEst/(-0.0000861733*x)))/S1
)) Plot Lifetime ZFS 0.00124 ± 6.49961E-5
dEst 0.044 ± 0 T12 109.94625 ± 2.25118 T3 16.60029 ± 0.37244 S1 0.019 ± 0 Reduced Chi-Sqr 4.25967 R-Square (COD) 0.99582 Adj. R-Square 0.99563
0.003 0.004 0.005 0.006 0.007 0.008 0.009 0.010 0.011
11.5
12.0
12.5
13.0
13.5
14.0
14.5
15.0
15.5 BZI-Au-MeOBimbim ln k (s-1
)
1/T (K-1)
Equation y = a + b*x Plot ln k Weight No Weighting
Intercept 16.68038 ± 0.09323 Slope -471.52126 ± 15.56639 Residual Sum of Squares 0.31183 Pearson's r -0.99086 R-Square (COD) 0.98181 Adj. R-Square 0.98074
Figure 5-12: Temperature dependent lifetime data and Arrhenius plot of 𝐴𝑢𝐵𝑖𝑚
𝑃𝑍𝐼
, 𝐴𝑢𝑀𝐵𝑖𝑚
𝐵𝑍𝐼 and 𝐴𝑢𝑂𝐵𝑖𝑚
𝐵𝑍𝐼
doped PS thin film. In red is the fits and the fitting parameters are stated in the insets in each figure.

91
0 50 100 150 200 250 300
0
20
40
60
80
 / µs
T / K
AuBZAC
DMBim
Model TADFBoltzmann (User) Equation
(2+exp(ZFS/(-0.0000861733*x)
)+exp(dEst/(-0.0000861733*x)))
/((2/T12)+((exp(ZFS/(-0.000086
1733*x)))/T3)+((exp(dEst/(-0.00
00861733*x)))/S1)) Plot \i(\g(t)) / µs ZFS 3.56627E-4 ± 6.0106E-5
dEst 0.036 ± 0 T12 100 ± 0 T3 38.38903 ± 1.31072 S1 0.016 ± 0 Reduced Chi-Sqr 0.14016 R-Square (COD) 0.99607 Adj. R-Square 0.99595
0.0030 0.0035 0.0040 0.0045 0.0050 0.0055 0.0060 0.0065 0.0070
13.8
14.0
14.2
14.4
14.6
14.8
15.0
15.2
15.4
AuBZAC
DMBim ln k
1/T
Equation y = a + b*x Plot ln k Weight No Weighting
Intercept 16.86273 ± 0.0582 Slope -426.8891 ± 11.62977 Residual Sum of Squares 0.0112 Pearson's r -0.99741 R-Square (COD) 0.99483 Adj. R-Square 0.99409
10 100
0
20
40
60
80
100
120
140
160 BCzAuBBIAuBCz Lifetime (us)
Temperature (K)
ZFS (meV) 0.00109 ± 1E-4
dEst (meV) 0.04291 ± 0 T12 (us) 151.40386 ± 6 T3 (us) 33.05652 ± 1 S1 (us) 0.0126 ± 0
0.003 0.004 0.005 0.006 0.007
13.6
13.8
14.0
14.2
14.4
14.6
14.8
15.0
15.2
15.4
15.6 BCzAuBBIAuBCz ln k (s-1
)
1/T (K-1)
Weight No Weighting
Intercept 17.08899 ± 0.0 Slope -498.025 ± 5.8 Res Sum of Squ 0.01091 Pearson's r -0.9989 R-Square (COD) 0.9978 Adj. R-Square 0.99767
10 100
0
50
100
150
200
250 BCzAuBAZAuBCz Lifetime (s)
TEMP (K)
ZFS (eV) 8.4E-4 ± 5E-5
dEst (eV) 0.05353 ± 0 T12 (us) 320± 10 T3 (us) 32.9 ± 0.8 S1 (us) 0.0127 ± 0 0.003 0.004 0.005 0.006 0.007 0.008 0.009 0.010 0.011
11
12
13
14
15 BCzAuBAZAuBCz ln(k) (s-1
)
1/T (K-1)
Intercept 17.08678 ± 0.0767 Slope -621.20661 ± 12.47 Residual Sum of Squar 0.23592 Pearson's r -0.99679 R-Square (COD) 0.99359 Adj. R-Square 0.99319
Figure 5-13: Temperature dependent lifetime data and Arrhenius plot of 𝐴𝑢𝐷𝑀𝐵𝑖𝑚
𝐵𝑍𝐴𝐶
, BCzAuBBIAuBCz and
BCzAuBAZAuBCz doped PS thin film. In red is the fits and the fitting parameters are stated in the insets in
each figure.

92
0.0030 0.0035 0.0040 0.0045 0.0050 0.0055 0.0060 0.0065
13.6
13.8
14.0
14.2
14.4
14.6
BimAuBAZAuBim ln(K) (s-1
)
Temperature (K-1)
Intercept 15.13 ± 0.053 Slope -214.7987 ± 11.450 R-Square (COD) 0.95912
Figure 5-14: Temperature dependent lifetime Arrhenius plot of BimAuBAZAuBim doped PS thin film. In red
is the fits and the fitting parameters are stated in the insets in each figure.

93
Conclusion
In summary, this research has explored design strategies and synthesis of luminescent cMa complexes,
and shown their potential for high-efficiency photophysical properties for OLED applications. Chapter
2 delved into the optimization of radiative rates through the π-extension of both carbene and amide
ligands, demonstrating significant enhancements in TADF-controlled emission processes. Chapter 3
introduced bimetallic complexes utilizing Janus carbenes, which exhibited accelerated radiative rates
and broad luminescence spectra spanning from blue to green. These complexes showed promise as
efficient dopants in OLEDs, with MoOx emerging as a promising alternative hole injection material.
Chapter 4 introduced a novel approach by incorporating electron-withdrawing groups on the amide,
leading to blue-emitting complexes with enhanced stability and comparable or increased radiative rates
with lower HOMO and LUMO energies than their carbazole or bim analogs. The exploration of these
diverse strategies highlights the versatility and potential of cMa complexes in advancing optoelectronic
technologies.

94
Experimental Methods
Single Crystal Diffraction Analysis
All crystals were grown by recrystallization. Vapor diffusion of hexanes or pentane into a solution of
the compound in dichloromethane. A Cryo-Loop was used to mount the sample with Paratone oil.
All single crystal structures were determined at 100K with Rigaku Xta LAB Synergy S, equipped with
an HyPix-600HE detector and an Oxford Cryostream 800 low Temperature unit, using Cu K PhotonJetS X-ray source. The frames were integrated using the SAINT algorithm to give the hkl files. Data were
corrected for absorption effects using the multi-scan method (SADABS) with Rigaku CrysalisPro. The
structures were solved by intrinsic phasing and refined with the SHELXTL Software Package.172 If
necessary, the disordered solvent treatment method BYPASS for co-crystalizing solvent molecules, was
implemented and marked in the CCDC entry. All cif files and report data including atom position, bond
lengths and bond angle can be downloaded from the CCDC database, using the database number given
in the crystallographic tables.
Electrochemistry
Cyclic voltammetry and differential pulsed voltammetry were performed using a VersaSTAT
potentiostat measured at 100 mV/s scan. Anhydrous dimethylformamide was used as the solvent, with
0.1 M tetra(n-butyl)ammonium hexafluorophosphate as the supporting electrolyte. The redox
potentials are based on values measured from differential pulsed voltammetry and are reported relative
to the ferrocenium/ferrocene (Cp2Fe+
/Cp2Fe) redox couple using either ferrocene or
decamethylferrocene as an internal reference. Electrochemical reversibility was determined using
cyclic voltammetry.
Modeling methods
The electronic properties of the complexes were modelled using Density Functional Theory (DFT) and
Time Dependent DFT (TDDFT). Geometry optimization was performed using the B3LYP functional
and LACVP* basis set using crystallographic coordinates as starting points when possible. TDDFT
calculations were performed on the geometry-optimized structures using the CAM-B3LYP exchange,
LACVP effective core potential, with the random phase approximation enabled and the omega value
set to 0.2 arbitrary units. The full carbene ligand, but only the parent Cz and bim ligands were
investigated in these modeling studies. Natural transition orbitals (NTOs) were generated by
performing a singular value decomposition on the transition density matrix using the Q-Chem v5.0
software package.173
The spatial overlap of the hole and electron NTO (NTO) for a particular excited state is defined here by
Eqn 2-3.
𝛬𝑁𝑇𝑂 ≈ ∰|𝜑ℎ+||𝜑𝑒−|𝑑𝜏 (2-3)
where h+ and e- are the hole and electron NTOs, respectively, for the excited state. Both h+ and eare orthonormal which ensures that 𝛬𝑁𝑇𝑂 falls between 0 and 1, which represent 0% and 100% overlap
respectively. Values for 𝛬𝑁𝑇𝑂 were calculated for these compounds using a script that draws from the

95
integration methods developed by Herman, et al.,
174 and Castro, et al.,
175 using a method reported
previously.1, 93, 136

The center of charge for the hole and electron NTOs of a given excited state are extracted from the
expectation values of the respective wavefunction positions, Eqns. 2-4 and 2-5.
〈𝑟ℎ+〉 = ⟨𝛹ℎ+|𝑟̂|𝛹ℎ+⟩ (2-4)
〈𝑟𝑒−〉 = ⟨𝛹𝑒−|𝑟̂|𝛹𝑒−⟩ (2-5)
where |𝛹ℎ+⟩ and |𝛹𝑒−⟩ represent the hole and electron NTO wavefunctions. The centers of charge are
calculated component-wise to extract two vectors, whose termini are the center of charge for the hole
or electron, h+
xyz and exyz, respectively. The positions of these centers of charge are illustrated in Figure
2-10 by placing an atom at the coordinates of h+
xyz and exyz. The distance between the charges [d(h+
,
e
-
)] for each cMa are determined algebraically from h+
xyz and exyz.
Further details for the calculation of NTO and center of charge are given in the supporting information
for this manuscript.
Photophysical Measurements
Samples in fluid solution were both sparged and examined under N2. Doped polystyrene thin films
were prepared from a solution of polystyrene in toluene, drop cast onto a quartz substrate and measured
under N2. The UV-visible spectra were recorded on a Hewlett-Packard 4853 diode array spectrometer.
Steady state excitation and emission spectra were obtained using a Photon Technology International
QuantaMaster spectrofluorimeter. Photoluminescence quantum yields were recorded using a
Hamamatsu C9920 integrating sphere equipped with a xenon lamp. Luminescence lifetimes were
measured using Time-Correlated Single Photon Counting (TCSPC) on an IBH Fluorocube apparatus
interfaced to a Horiba FluoroHub+ controller.
Variable temperature photophysical measurements were carried out on a Janis SHI-4-2 (0.2 W 4K)
optical cryocooler. The IBH Fluorocube was used as a detector for luminescence lifetimes and the
Photon Technology International QuantaMaster spectrofluorimeter as a detector for steady state
emission spectra with 365 nm LED (Thorlabs M365L4, 880 mW) as excitation source. Doped
polystyrene thin films were spin coated onto a round sapphire substrate that was used to insure good
thermal conductivity at low temperatures.
OLED Device fabrication
For OLEDs in Chapter 2: Glass substrates with pre-patterned, 2 mm wide indium tin oxide (ITO) stripes
were cleaned by sequential sonication in deionized water, acetone, and isopropanol, followed by 10
min UV ozone exposure. Organic materials and metals were deposited at rates of 0.5−1 Å s−1
through
shadow masks in a Angstrom vacuum thermal evaporator with a base pressure of 10–7 Torr. A separate
shadow mask was used to deposit 1 mm wide stripes of 100 nm thick Al films perpendicular to the ITO
stripes to form the cathode, resulting in a 4 mm2 device area. A semiconductor parameter analyzer
(HP4156A) and a calibrated large area photodiode that collected all light exiting the glass substrate
were used to measure the current density–voltage–luminance (J–V–L) characteristics. The device
spectra were measured using a Photon Technology International QuantaMaster spectrofluorimeter.

96
For Devices in Chapter 3: OLED devices were fabricated on pre-patterned ITO-coated glass substrates
(20 ± 5 Ω cm2, Thin Film Devices, Inc.). Prior to deposition, the substrates were cleaned with soap,
rinsed with deionized water and sonicated for 10 minutes. Afterwards, two subsequent rinses and 15-
minute sonication baths were performed in acetone and isopropyl alcohol sequentially, followed by 15
min UV ozone exposure. After the MoOx deposition (5nm) with an EvoVac (Angstrom Engineering)
the substrates were transferred without contact to air into a particle free Nitrogen Glovebox. The EML
was spin coated from a toluene solution, which was stirred overnight at 65 C and filtered before usage.
A consistent thickness of 30 nm was achieved with 65 L of solution with concentration of 8mg per
mL toluene at a spin rate of 3000 rpm for 90s. If applicable substrates were annealed under N2
atmosphere for 10min @110C. Transfer into the deposition system occurred without contact to air.
TPBi (ETL) was deposited using a Vacuum Thermal Evaporation (Angstrom Engineering) and the
Cathode (Liq and Al) was deposited using a Vacuum Thermal Evaporation (Kurt J. Lesker). Currentvoltage-luminescence (J-V-L) curves were measured in an by using a Keithley power source meter
model 2400 and a Newport multifunction optical model 1835-C, PIN-220DP/SB blue enhanced silicon
photodiodes (OSI optoelectronics Ltd.). The sensor was set to measure power at an energy of 520 nm,
followed by correcting to the average electroluminescence wavelength for each individual device
during data process. Electroluminescence (EL) spectra of OLEDs were measured using the fluorimeter
(model C-60 Photon Technology International QuantaMaster) at several different voltages. Thicknesses
were determined on Silicon wafers using a Filmsense FS1 Ellipsometer.

97
References
1. Li, T.-y.; Schaab, J.; Djurovich, P. I.; Thompson, M. E., Toward rational design of TADF twocoordinate coinage metal complexes: understanding the relationship between natural transition orbital
overlap and photophysical properties. Journal of Materials Chemistry C 2022, 10 (12), 4674-4683.
2. Shi, S.; Jung, M. C.; Coburn, C.; Tadle, A.; Sylvinson M. R, D.; Djurovich, P. I.; Forrest,
S. R.; Thompson, M. E., Highly Efficient Photo- and Electroluminescence from Two-Coordinate Cu(I)
Complexes Featuring Nonconventional N-Heterocyclic Carbenes. Journal of the American Chemical
Society 2019, 141 (8), 3576-3588.
3. Tang, R.; Xu, S.; Du, L.; Hung, F.-F.; Lam, T.-L.; Cheng, G.; Low, K.-H.; Wan, Q.; Wu,
S.; Chen, Y.; Che, C.-M., Au(I)-TADF Emitters for High Efficiency Full-Color Vacuum-Deposited
OLEDs and TADF-Sensitized Fluorescent OLEDs with Ultrahigh Brightness and Prolonged
Operational Lifetime. Advanced Optical Materials 2023, 11 (23), 2300950.
4. Hamze, R.; Idris, M.; Muthiah Ravinson, D. S.; Jung, M. C.; Haiges, R.; Djurovich, P. I.;
Thompson, M. E., Highly Efficient Deep Blue Luminescence of 2-Coordinate Coinage Metal
Complexes Bearing Bulky NHC Benzimidazolyl Carbene. Frontiers in Chemistry 2020, 8.
5. Sworakowski, J.; Lipiński, J.; Janus, K., On the reliability of determination of energies of
HOMO and LUMO levels in organic semiconductors from electrochemical measurements. A simple
picture based on the electrostatic model. Organic Electronics 2016, 33, 300-310.
6. Brunen, S.; Grell, Y.; Steinlandt, P. S.; Harms, K.; Meggers, E., Bis-Cyclometalated Indazole
and Benzimidazole Chiral-at-Iridium Complexes: Synthesis and Asymmetric Catalysis. Molecules
2021, 26 (7).
7. Tang, R.; Xu, S.; Lam, T.-L.; Cheng, G.; Du, L.; Wan, Q.; Yang, J.; Hung, F.-F.; Low, K.-
H.; Phillips, D. L.; Che, C.-M., Highly Robust CuI-TADF Emitters for Vacuum-Deposited OLEDs
with Luminance up to 222 200 cd m−2 and Device Lifetimes (LT90) up to 1300 hours at an Initial
Luminance of 1000 cd m−2. Angewandte Chemie International Edition 2022, 61 (33), e202203982.
8. Romanov, A. S.; Bochmann, M., Synthesis, structures and photoluminescence properties of
silver complexes of cyclic (alkyl)(amino)carbenes. Journal of Organometallic Chemistry 2017, 847,
114-120.
9. Romanov, A. S.; Jones, S. T. E.; Yang, L.; Conaghan, P. J.; Di, D.; Linnolahti, M.;
Credgington, D.; Bochmann, M., Mononuclear Silver Complexes for Efficient Solution and VacuumProcessed OLEDs. Advanced Optical Materials 2018, 6 (24), 1801347.
10. Hamze, R.; Shi, S.; Kapper, S. C.; Muthiah Ravinson, D. S.; Estergreen, L.; Jung, M.-C.;
Tadle, A. C.; Haiges, R.; Djurovich, P. I.; Peltier, J. L.; Jazzar, R.; Bertrand, G.; Bradforth, S. E.;
Thompson, M. E., “Quick-Silver” from a Systematic Study of Highly Luminescent, Two-Coordinate,
d10 Coinage Metal Complexes. Journal of the American Chemical Society 2019, 141 (21), 8616-8626.
11. Schaab, J.; Djurovich, P.; Thompson, M. E., Two coordinate, monovalent copper complexes
as chromophores and luminophores. In Advances in Inorganic chemistry, Elsevier: 2024; Vol. 84.

98
12. Ziolo, R. F.; Lipton, S.; Dori, Z., The photoluminescence of phosphine complexes of d 10
metals. Journal of the Chemical Society D: Chemical Communications 1970, (17).
13. Antipas, A.; Dolphin, D.; Gouterman, M.; Johnson, E. C., Porphyrins. 38. Redox potentials,
charge transfer transitions, and emission of copper, silver, and gold complexes. Journal of the American
Chemical Society 1978, 100 (24), 7705-7709.
14. Beaudelot, J.; Oger, S.; Perusko, S.; Phan, T. A.; Teunens, T.; Moucheron, C.; Evano, G.,
Photoactive Copper Complexes: Properties and Applications. Chem Rev 2022, 122 (22), 16365-16609.
15. Hossain, A.; Bhattacharyya, A.; Reiser, O., Copper's rapid ascent in visible-light photoredox
catalysis. Science 2019, 364 (6439).
16. Sorrell, T. N.; Borovik, A. S., Absorption, emission, and photophysical properties of copper(I)
pyrazole complexes and their carbonyl adducts. Inorganic Chemistry 1987, 26 (12), 1957-1964.
17. Mercs, L.; Albrecht, M., Beyond catalysis: N-heterocyclic carbene complexes as components
for medicinal, luminescent, and functional materials applications. Chemical Society Reviews 2010, 39
(6), 1903-1912.
18. Visbal, R.; Gimeno, M. C., N-heterocyclic carbene metal complexes: photoluminescence and
applications. Chemical Society Reviews 2014, 43 (10), 3551-3574.
19. Wang, H. M. J.; Chen, C. Y. L.; Lin, I. J. B., Synthesis, Structure, and Spectroscopic Properties
of Gold(I)−Carbene Complexes. Organometallics 1999, 18 (7), 1216-1223.
20. Yam, V. W.-W.; Lai, T.-F.; Che, C.-M., Novel luminescent polynuclear gold(I) phosphine
complexes. Synthesis, spectroscopy, and X-ray crystal structure of [Au3(dmmp)2]3+[dmmp =
bis(dimethylphosphinomethyl)methylphosphine]. Journal of the Chemical Society, Dalton
Transactions 1990, (12), 3747-3752.
21. King, C.; Khan, M. N. I.; Staples, R. J.; Fackler, J. P., Luminescent mononuclear gold(I)
phosphines. Inorganic Chemistry 1992, 31 (15), 3236-3238.
22. Hong, X.; Cheung, K.-K.; Guo, C.-X.; Che, C.-M., Luminescent organometallic gold(I)
complexes. Structure and photophysical properties of alkyl-, aryl- and µ-ethynylene gold(I) complexes.
Journal of the Chemical Society, Dalton Transactions 1994, (13), 1867-1871.
23. Wang, H. M. J.; Vasam, C. S.; Tsai, T. Y. R.; Chen, S.-H.; Chang, A. H. H.; Lin, I. J. B.,
Gold(I) N-Heterocyclic Carbene and Carbazolate Complexes. Organometallics 2005, 24 (4), 486-493.
24. Gimeno, M. C.; Laguna, A.; Visbal, R., N-Heterocyclic Carbene Coinage Metal Complexes as
Intense Blue-Green Emitters. Organometallics 2012, 31 (20), 7146-7157.
25. Gomez-Suarez, A.; Nelson, D. J.; Thompson, D. G.; Cordes, D. B.; Graham, D.; Slawin, A.
M.; Nolan, S. P., Synthesis, characterization and luminescence studies of gold(I)-NHC amide
complexes. Beilstein J Org Chem 2013, 9, 2216-23.
26. Romanov, A. S.; Di, D.; Yang, L.; Fernandez-Cestau, J.; Becker, C. R.; James, C. E.; Zhu,
B.; Linnolahti, M.; Credgington, D.; Bochmann, M., Highly photoluminescent copper carbene

99
complexes based on prompt rather than delayed fluorescence. Chemical Communications 2016, 52
(38), 6379-6382.
27. Romanov, A. S.; Di, D.; Yang, L.; Fernandez-Cestau, J.; Becker, C. R.; James, C. E.; Zhu,
B.; Linnolahti, M.; Credgington, D.; Bochmann, M., Correction: Highly photoluminescent copper
carbene complexes based on prompt rather than delayed fluorescence. Chem Commun (Camb) 2018,
54 (29), 3672.
28. Hamze, R.; Jazzar, R.; Soleilhavoup, M.; Djurovich, P. I.; Bertrand, G.; Thompson , M. E.,
Phosphorescent 2-, 3- and 4-coordinate cyclic (alkyl)(amino)carbene (CAAC) Cu(I) complexes.
Chemical Communications 2017, 53, 9008-9011.
29. Shi, S.; Collins, L. R.; Mahon, M. F.; Djurovich, P. I.; Thompson, M. E.; Whittlesey, M. K.,
Synthesis and characterization of phosphorescent two-coordinate copper(i) complexes bearing
diamidocarbene ligands. Dalton Transactions 2017, 46 (3), 745-752.
30. Romanov, A. S.; Becker, C. R.; James, C. E.; Di, D.; Credgington, D.; Linnolahti, M.;
Bochmann, M., Copper and Gold Cyclic (Alkyl)(amino)carbene Complexes with Sub-Microsecond
Photoemissions: Structure and Substituent Effects on Redox and Luminescent Properties. Chemistry –
A European Journal 2017, 23 (19), 4625-4637.
31. Yersin, H.; Czerwieniec, R.; Shafikov, M. Z.; Suleymanova, A. F., TADF Material Design:
Photophysical Background and Case Studies Focusing on CuI and AgI Complexes. ChemPhysChem
2017, 18 (24), 3508-3535.
32. Gernert, M.; Balles-Wolf, L.; Kerner, F.; Müller, U.; Schmiedel, A.; Holzapfel, M.; Marian,
C. M.; Pflaum, J.; Lambert, C.; Steffen, A., Cyclic (Amino)(aryl)carbenes Enter the Field of
Chromophore Ligands: Expanded π System Leads to Unusually Deep Red Emitting CuI Compounds.
Journal of the American Chemical Society 2020, 142 (19), 8897-8909.
33. Li, T.-y.; Muthiah Ravinson, D. S.; Haiges, R.; Djurovich, P. I.; Thompson, M. E.,
Enhancement of the Luminescent Efficiency in Carbene-Au(I)-Aryl Complexes by the Restriction of
Renner–Teller Distortion and Bond Rotation. Journal of the American Chemical Society 2020, 142
(13), 6158-6172.
34. Li, T.-Y.; Zheng, S.-J.; Djurovich, P. I.; Thompson, M. E., Two-coordinate Thermally Assisted
Delayed Fluorescence (TADF) Coinage Metal Complexes: Molecular Design, Photophysical
Characters and Organic Light-emitting Diode (OLED) Application. Chemical Reviews 2024,submitted.
35. Beaudelot, J.; Oger, S.; Peruško, S.; Phan, T.-A.; Teunens, T.; Moucheron, C.; Evano, G.,
Photoactive Copper Complexes: Properties and Applications. Chemical Reviews 2022, 122 (22),
16365-16609.
36. Ying, A.; Gong, S., A Rising Star: Luminescent Carbene-Metal-Amide Complexes. Chemistry
– A European Journal 2023, 29 (59), e202301885.
37. Brown, R. E.; Singer, L. A.; Parks, J. H., Prompt and delayed fluorescence from benzophenone.
Chemical Physics Letters 1972, 14 (2), 193-195.
38. Deaton, J. C.; Switalski, S. C.; Kondakov, D. Y.; Young, R. H.; Pawlik, T. D.; Giesen, D. J.;
Harkins, S. B.; Miller, A. J. M.; Mickenberg, S. F.; Peters, J. C., E-Type Delayed Fluorescence of a

100
Phosphine-Supported Cu2(μ-NAr2)2 Diamond Core: Harvesting Singlet and Triplet Excitons in
OLEDs. Journal of the American Chemical Society 2010, 132 (27), 9499-9508.
39. Tsuboyama, A.; Kuge, K.; Furugori, M.; Okada, S.; Hoshino, M.; Ueno, K., Photophysical
Properties of Highly Luminescent Copper(I) Halide Complexes Chelated with 1,2-
Bis(diphenylphosphino)benzene. Inorganic Chemistry 2007, 46 (6), 1992-2001.
40. Kirchhoff, J. R.; Gamache, R. E., Jr.; Blaskie, M. W.; Del Paggio, A. A.; Lengel, R. K.;
McMillin, D. R., Temperature dependence of luminescence from Cu(NN)2+ systems in fluid solution.
Evidence for the participation of two excited states. Inorganic Chemistry 1983, 22 (17), 2380-2384.
41. Blasse, G.; McMillin, D. R., On the luminescence of bis (triphenylphosphine) phenanthroline
copper (I). Chemical Physics Letters 1980, 70 (1), 1-3.
42. Bergmann, L.; Hedley, G. J.; Baumann, T.; Bräse, S.; Samuel, I. D. W., Direct observation of
intersystem crossing in a thermally activated delayed fluorescence copper complex in the solid state.
Science Advances 2 (1), e1500889.
43. Cai, S.; Tong, G. S. M.; Du, L.; So, G. K.-M.; Hung, F.-F.; Lam, T.-L.; Cheng, G.; Xiao,
H.; Chang, X.; Xu, Z.-X.; Che, C.-M., Gold(I) Multi-Resonance Thermally Activated Delayed
Fluorescent Emitters for Highly Efficient Ultrapure-Green Organic Light-Emitting Diodes.
Angewandte Chemie International Edition 2022, 61 (52), e202213392.
44. Ravinson, D. S. M.; Thompson, M. E., Thermally assisted delayed fluorescence (TADF):
fluorescence delayed is fluorescence denied. Materials Horizons 2020, 7 (5), 1210-1217.
45. Jones, P. F.; Calloway, A. R., Temperature effects on the intramolecular decay of the lowest
triplet state of benzophenone. Chemical Physics Letters 1971, 10 (4), 438-443.
46. Scaltrito, D. V.; Thompson, D. W.; O'Callaghan, J. A.; Meyer, G. J., MLCT excited states of
cuprous bis-phenanthroline coordination compounds. Coordination Chemistry Reviews 2000, 208 (1),
243-266.
47. Bersuker, I. B., Modern Aspects of the Jahn−Teller Effect Theory and Applications To
Molecular Problems. Chemical Reviews 2001, 101 (4), 1067-1114.
48. Iwamura, M.; Takeuchi, S.; Tahara, T., Ultrafast Excited-State Dynamics of Copper(I)
Complexes. Accounts of Chemical Research 2015, 48 (3), 782-791.
49. Shaw, G. B.; Grant, C. D.; Shirota, H.; Castner, E. W.; Meyer, G. J.; Chen, L. X., Ultrafast
structural rearrangements in the MLCT excited state for copper(I) bis-phenanthrolines in solution.
Journal of the American Chemical Society 2007, 129 (7), 2147-2160.
50. Iwamura, M.; Takeuchi, S.; Tahara, T., Real-Time Observation of the Photoinduced Structural
Change of Bis(2,9-dimethyl-1,10-phenanthroline)copper(I) by Femtosecond Fluorescence
Spectroscopy: A Realistic Potential Curve of the Jahn−Teller Distortion. Journal of the American
Chemical Society 2007, 129 (16), 5248-5256.
51. Vorontsov, II; Graber, T.; Kovalevsky, A. Y.; Novozhilova, I. V.; Gembicky, M.; Chen, Y. S.;
Coppens, P., Capturing and Analyzing the Excited-State Structure of a Cu(I) Phenanthroline Complex

101
by Time-Resolved Diffraction and Theoretical Calculations. Journal of the American Chemical Society
2009, 131 (18), 6566-6573.
52. Garakyaraghi, S.; Danilov, E. O.; McCusker, C. E.; Castellano, F. N., Transient Absorption
Dynamics of Sterically Congested Cu(I) MLCT Excited States. The Journal of Physical Chemistry A
2015, 119 (13), 3181-3193.
53. Miller, M. T.; Gantzel, P. K.; Karpishin, T. B., A Highly Emissive Heteroleptic Copper(I)
Bis(phenanthroline) Complex: [Cu(dbp)(dmp)]+ (dbp = 2,9-Di-tert-butyl-1,10-phenanthroline; dmp =
2,9-Dimethyl-1,10-phenanthroline). Journal of the American Chemical Society 1999, 121 (17), 4292-
4293.
54. Miller, M. T.; Gantzel, P. K.; Karpishin, T. B., Effects of Sterics and Electronic Delocalization
on the Photophysical, Structural, and Electrochemical Properties of 2,9-Disubstituted 1,10-
Phenanthroline Copper(I) Complexes. Inorganic Chemistry 1999, 38 (14), 3414-3422.
55. Barakat, K. A.; Cundari, T. R.; Omary, M. A., Jahn-Teller distortion in the phosphorescent
excited state of three-coordinate Au(I) phosphine complexes. Journal Of The American Chemical
Society 2003, 125 (47), 14228-14229.
56. Lee, K.; Lai, P.-N.; Parveen, R.; Donahue, C. M.; Wymore, M. M.; Massman, B. A.;
Vlaisavljevich, B.; Teets, T. S.; Daly, S. R., Modifying the luminescent properties of a Cu(i)
diphosphine complex using ligand-centered reactions in single crystals. Chemical Communications
2020, 56 (64), 9110-9113.
57. Ford, P. C.; Cariati, E.; Bourassa, J., Photoluminescence Properties of Multinuclear Copper(I)
Compounds. Chemical Reviews 1999, 99 (12), 3625-3648.
58. Cuttell, D. G.; Kuang, S.-M.; Fanwick, P. E.; McMillin, D. R.; Walton, R. A., Simple Cu(I)
Complexes with Unprecedented Excited-State Lifetimes. Journal of the American Chemical Society
2002, 124 (1), 6-7.
59. Breddels, P. A.; Berdowski, P. A. M.; Blasse, G.; McMillin, D. R., Luminescence of some CuI
complexes. Journal of the Chemical Society, Faraday Transactions 2: Molecular and Chemical Physics
1982, 78 (3), 595-601.
60. Blaskie, M. W.; McMillin, D. R., Photostudies of copper(I) systems. 6. Room-temperature
emission and quenching studies of bis(2,9-dimethyl-1,10-phenanthroline)copper(I). Inorganic
Chemistry 1980, 19 (11), 3519-3522.
61. Ruthkosky, M.; Castellano, F. N.; Meyer, G. J., Photodriven Electron and Energy Transfer
from Copper Phenanthroline Excited States. Inorganic Chemistry 1996, 35 (22), 6406-6412.
62. Castellano, F. N.; Ruthkosky, M.; Meyer, G. J., Photodriven Energy Transfer from Cuprous
Phenanthroline Derivatives. Inorganic Chemistry 1995, 34 (1), 3-4.
63. Zhang, Y.; Schulz, M.; Wächtler, M.; Karnahl, M.; Dietzek, B., Heteroleptic diimine–
diphosphine Cu(I) complexes as an alternative towards noble-metal based photosensitizers: Design
strategies, photophysical properties and perspective applications. Coordination Chemistry Reviews
2018, 356, 127-146.

102
64. Osawa, M.; Hoshino, M., Molecular Design and Synthesis of Metal Complexes as Emitters for
TADF-Type OLEDs. In Highly Efficient OLEDs, 2018; pp 119-176.
65. Harkins, S. B.; Peters, J. C., A Highly Emissive Cu2N2 Diamond Core Complex Supported by
a [PNP]- Ligand. Journal of the American Chemical Society 2005, 127 (7), 2030-2031.
66. Miller, A. J. M.; Dempsey, J. L.; Peters, J. C., Long-Lived and Efficient Emission from
Mononuclear Amidophosphine Complexes of Copper. Inorganic Chemistry 2007, 46 (18), 7244-7246.
67. Czerwieniec, R.; Yu, J.; Yersin, H., Blue-Light Emission of Cu(I) Complexes and Singlet
Harvesting. Inorganic Chemistry 2011, 50 (17), 8293-8301.
68. Ravaro, L. P.; Zanoni, K. P. S.; de Camargo, A. S. S., Luminescent Copper(I) complexes as
promising materials for the next generation of energy-saving OLED devices. Energy Reports 2020, 6,
37-45.
69. Leitl, M. J.; Krylova, V. A.; Djurovich, P. I.; Thompson, M. E.; Yersin, H., Phosphorescence
versus Thermally Activated Delayed Fluorescence. Controlling Singlet-Triplet Splitting in Brightly
Emitting and Sublimable Cu(I) Compounds. Journal of the American Chemical Society 2014, 136 (45),
16032-16038.
70. Krylova, V. A.; Djurovich, P. I.; Aronson, J. W.; Haiges, R.; Whited, M. T.; Thompson, M.
E., Structural and Photophysical Studies of Phosphorescent Three-Coordinate Copper(I) Complexes
Supported by an N-Heterocyclic Carbene Ligand. Organometallics 2012, 31 (22), 7983-7993.
71. Krylova, V. A.; Djurovich, P. I.; Whited, M. T.; Thompson, M. E., Synthesis and
characterization of phosphorescent three-coordinate Cu(I)-NHC complexes. Chemical
Communications 2010, 46 (36), 6696-6698.
72. Krylova, V. A.; Djurovich, P. I.; Conley, B. L.; Haiges, R.; Whited, M. T.; Williams, T. J.;
Thompson, M. E., Control of emission colour with N-heterocyclic carbene (NHC) ligands in
phosphorescent three-coordinate Cu(I) complexes. Chemical Communications 2014, 50 (54), 7176-
7179.
73. Shi, S.; Djurovich, P. I.; Thompson, M. E., Synthesis and characterization of phosphorescent
three-coordinate copper(I) complexes bearing bis(amino)cyclopropenylidene carbene (BAC).
Inorganica Chimica Acta 2018, 482, 246-251.
74. Osawa, M.; Hoshino, M.; Hashimoto, M.; Kawata, I.; Igawa, S.; Yashima, M., Application
of three-coordinate copper(I) complexes with halide ligands in organic light-emitting diodes that exhibit
delayed fluorescence. Dalton Transactions 2015, 44 (18), 8369-8378.
75. Osawa, M., Highly efficient blue-green delayed fluorescence from copper(I) thiolate
complexes: luminescence color alteration by orientation change of the aryl ring. Chemical
Communications 2014, 50 (15), 1801-1803.
76. Lotito, K. J.; Peters, J. C., Efficient luminescence from easily prepared three-coordinate
copper(I) arylamidophosphines. Chemical Communications 2010, 46 (21), 3690-3692.

103
77. Hashimoto, M.; Igawa, S.; Yashima, M.; Kawata, I.; Hoshino, M.; Osawa, M., Highly
Efficient Green Organic Light-Emitting Diodes Containing Luminescent Three-Coordinate Copper(I)
Complexes. Journal of the American Chemical Society 2011, 133 (27), 10348-10351.
78. Herzberg, G.; Teller, E., Schwingungsstruktur der Elektronenübergänge bei mehratomigen
Molekülen. In Zeitschrift fur Physikalische Chemie, 1933; Vol. 21B, p 410.
79. Renner, R., Zur Theorie der Wechselwirkung zwischen Elektronen- und Kernbewegung bei
dreiatomigen, stabförmigen Molekülen. Zeitschrift für Physik 1934, 92 (3), 172-193.
80. Lin, S.; Peng, Q.; Ou, Q.; Shuai, Z., Strong Solid-State Fluorescence Induced by Restriction
of the Coordinate Bond Bending in Two-Coordinate Copper(I)–Carbene Complexes. Inorganic
Chemistry 2019, 58 (21), 14403-14409.
81. Ma, J.; Kapper, S. C.; Ponnekanti, A.; Schaab, J.; Djurovich, P. I.; Thompson, M. E.,
Dynamics of rotation in two‐coordinate thiazolyl copper(I) carbazolyl complexes. Applied
Organometallic Chemistry 2022, e6728.
82. Jazzar, R.; Soleilhavoup, M.; Bertrand, G., Cyclic (Alkyl)- and (Aryl)-(amino)carbene
Coinage Metal Complexes and Their Applications. Chemical Reviews 2020, 120 (9), 4141-4168.
83. Gaillard, S.; Cazin, C. S. J.; Nolan, S. P., N-Heterocyclic Carbene Gold(I) and Copper(I)
Complexes in C-H Bond Activation. Accounts of Chemical Research 2012, 45 (6), 778-787.
84. Lin, J. C. Y.; Huang, R. T. W.; Lee, C. S.; Bhattacharyya, A.; Hwang, W. S.; Lin, I. J. B.,
Coinage Metal-N-Heterocyclic Carbene Complexes. Chemical Reviews 2009, 109 (8), 3561-
3598.
85. Boehme, C.; Frenking, G., N-Heterocyclic Carbene, Silylene, and Germylene Complexes of
MCl (M = Cu, Ag, Au). A Theoretical Study1. Organometallics 1998, 17 (26), 5801-5809.
86. Nemcsok, D.; Wichmann, K.; Frenking, G., The Significance of π Interactions in Group 11
Complexes with N-Heterocyclic Carbenes. Organometallics 2004, 23 (15), 3640-3646.
87. Gernert, M.; Müller, U.; Haehnel, M.; Pflaum, J.; Steffen, A., A Cyclic Alkyl(amino)carbene
as Two-Atom π-Chromophore Leading to the First Phosphorescent Linear CuI Complexes. Chemistry
– A European Journal 2017, 23 (9), 2206-2216.
88. Di, D.; Romanov, A. S.; Yang, L.; Richter, J. M.; Rivett, J. P.; Jones, S.; Thomas, T. H.;
Abdi Jalebi, M.; Friend, R. H.; Linnolahti, M.; Bochmann, M.; Credgington, D., High-performance
light-emitting diodes based on carbene-metal-amides. Science 2017, 356 (6334), 159-163.
89. Brütting, W.; Frischeisen, J.; Schmidt, T. D.; Scholz, B. J.; Mayr, C., Device efficiency of
organic light-emitting diodes: Progress by improved light outcoupling. physica status solidi (a) 2013,
210 (1), 44-65.
90. Hamze, R.; Peltier, J. L.; Sylvinson, D.; Jung, M.; Cardenas, J.; Haiges, R.; Soleilhavoup,
M.; Jazzar, R.; Djurovich, P. I.; Bertrand, G.; Thompson, M. E., Eliminating nonradiative decay in
Cu(I) emitters: > 99% quantum efficiency and microsecond lifetime. Science 2019, 363 (6427), 601-
606.

104
91. Ma, J.; Schaab, J.; Paul, S.; Forrest, S. R.; Djurovich, P. I.; Thompson, M. E., Luminescent
Bimetallic Two-Coordinate Gold(I) Complexes Utilizing Janus Carbenes. Journal of the American
Chemical Society 2023, 145 (36), 20097-20108.
92. Muniz, C. N.; Archer, C. A.; Applebaum, J. S.; Alagaratnam, A.; Schaab, J.; Djurovich, P.
I.; Thompson, M. E., Two-Coordinate Coinage Metal Complexes as Solar Photosensitizers. Journal of
the American Chemical Society 2023, 145 (25), 13846-13857.
93. Muniz, C. N.; Schaab, J.; Razgoniaev, A.; Djurovich, P. I.; Thompson, M. E., π-Extended
Ligands in Two-Coordinate Coinage Metal Complexes. Journal of the American Chemical Society
2022, 144 (39), 17916-17928.
94. Hall, C. R.; Romanov, A. S.; Bochmann, M.; Meech, S. R., Ultrafast Structure and Dynamics
in the Thermally Activated Delayed Fluorescence of a Carbene–Metal–Amide. The Journal of Physical
Chemistry Letters 2018, 9 (19), 5873-5876.
95. Di, D.; Romanov, A. S.; Yang, L.; Richter, J. M.; Rivett, J. P. H.; Jones, S.; Thomas, T. H.;
Abdi Jalebi, M.; Friend, R. H.; Linnolahti, M.; Bochmann, M.; Credgington, D., High-performance
light-emitting diodes based on carbene-metal-amides. Science 2017, 356 (6334), 159-163.
96. Reponen, A.-P. M.; Chotard, F.; Lempelto, A.; Shekhovtsev, V.; Credgington, D.; Bochmann,
M.; Linnolahti, M.; Greenham, N. C.; Romanov, A. S., Donor N-Substitution as Design Principle for
Fast and Blue Luminescence in Carbene-Metal-Amides. Advanced Optical Materials 2022, 10 (15),
2200312.
97. Weinstein, C. M.; Junor, G. P.; Tolentino, D. R.; Jazzar, R.; Melaimi, M.; Bertrand, G., Highly
Ambiphilic Room Temperature Stable Six-Membered Cyclic (Alkyl)(amino)carbenes. J Am Chem Soc
2018, 140 (29), 9255-9260.
98. Hamze, R.; Idris, M.; Muthiah Ravinson, D. S.; Jung, M. C.; Haiges, R.; Djurovich, P. I.;
Thompson, M. E., Highly Efficient Deep Blue Luminescence of 2-Coordinate Coinage Metal
Complexes Bearing Bulky NHC Benzimidazolyl Carbene. Frontiers in Chemistry 2020, 8 (401), 1-9.
99. Paley, M. S.; McGill, R. A.; Howard, S. C.; Wallace, S. E.; Harris, J. M., Solvatochromism.
A New Method for Polymer Characterization. Macromolecules 1990, 23 (23), 4557-4564.
100. Mizukami, M.; Fujimori, H.; Oguni, M., Glass Transitions and the Responsible Molecular
Motions in 2-Methyltetrahydrofuran. Progress of Theoretical Physics Supplement 1997, 126, 79-82.
101. Qi, F.; El Goresy, T.; Böhmer, R.; Döß, A.; Diezemann, G.; Hinze, G.; Sillescu, H.;
Blochowicz, T.; Gainaru, C.; Rössler, E.; Zimmermann, H., Nuclear magnetic resonance and dielectric
spectroscopy of a simple supercooled liquid: 2-methyl tetrahydrofuran. The Journal of Chemical
Physics 2003, 118 (16), 7431-7438.
102. Richert, R., Molecular probing of dielectric relaxation in the glass-transition region. Chemical
Physics Letters 1992, 199 (3), 355-359.
103. Wang, H.-J.; Liu, Y.; Yu, B.; Song, S.-Q.; Zheng, Y.-X.; Liu, K.; Chen, P.; Wang, H.; Jiang,
J.; Li, T.-Y., A Configurationally Confined Thermally Activated Delayed Fluorescent Two-Coordinate
CuI Complex for Efficient Blue Electroluminescence. Angewandte Chemie International Edition 2023,
62 (7), e202217195.

105
104. Gu, Q.; Chotard, F.; Eng, J.; Reponen, A.-P. M.; Vitorica-Yrezabal, I. J.; Woodward, A. W.;
Penfold, T. J.; Credgington, D.; Bochmann, M.; Romanov, A. S., Excited-State Lifetime Modulation
by Twisted and Tilted Molecular Design in Carbene-Metal-Amide Photoemitters. Chemistry of
Materials 2022, 34 (16), 7526-7542.
105. Zou, S.-J.; Shen, Y.; Xie, F.-M.; Chen, J.-D.; Li, Y.-Q.; Tang, J.-X., Recent advances in
organic light-emitting diodes: toward smart lighting and displays. Materials Chemistry Frontiers 2020,
4 (3), 788-820.
106. Nayak, D.; Choudhary, R. B., A survey of the structure, fabrication, and characterization of
advanced organic light emitting diodes. Microelectronics Reliability 2023, 144, 114959.
107. l’Éclairage, C. I. d., Commission internationale de l’eclairage proceedings. Cambridge
University Press Cambridge: 1931.
108. Giebink, N. C.; D'Andrade, B. W.; Weaver, M. S.; Mackenzie, P. B.; Brown, J. J.; Thompson,
M. E.; Forrest, S. R., Intrinsic luminance loss in phosphorescent small-molecule organic light emitting
devices due to bimolecular annihilation reactions. Journal of Applied Physics 2008, 103 (4).
109. Zhang, Y.; Lee, J.; Forrest, S. R., Tenfold increase in the lifetime of blue phosphorescent
organic light-emitting diodes. Nat Commun 2014, 5.
110. Zhan, L. S.; Ying, A.; Qi, Y. Y.; Wu, K. L.; Tang, Y.; Tan, Y.; Zou, Y.; Xie, G. H.; Gong, S.
L.; Yang, C. L., Copper(I) Complex as Sensitizer Enables High-Performance Organic Light-Emitting
Diodes with Very Low Efficiency Roll-Off. Advanced Functional Materials 2021, 31 (48).
111. Baldo, M. A.; Thompson, M. E.; Forrest, S. R., High-efficiency fluorescent organic lightemitting devices using a phosphorescent sensitizer. Nature 2000, 403 (6771), 750-753.
112. Di, D.; Romanov, A. S.; Yang, L.; Richter, J. M.; Rivett, J. P. H.; Jones, S.; Thomas, T. H.;
Abdi Jalebi, M.; Friend, R. H.; Linnolahti, M.; Bochmann, M.; Credgington, D., High-performance
light-emitting diodes based on carbene-metal-amides. Science 2017, 356 (6334), 159.
113. Conaghan, P. J.; Menke, S. M.; Romanov, A. S.; Jones, S. T. E.; Pearson, A. J.; Evans, E.
W.; Bochmann, M.; Greenham, N. C.; Credgington, D., Efficient Vacuum-Processed Light-Emitting
Diodes Based on Carbene–Metal–Amides. Advanced Materials 2018, 30 (35), 1802285.
114. Romanov, A. S.; Yang, L.; Jones, S. T. E.; Di, D.; Morley, O. J.; Drummond, B. H.; Reponen,
A. P. M.; Linnolahti, M.; Credgington, D.; Bochmann, M., Dendritic Carbene Metal Carbazole
Complexes as Photoemitters for Fully Solution-Processed OLEDs. Chemistry of Materials 2019, 31
(10), 3613-3623.
115. Chotard, F.; Sivchik, V.; Linnolahti, M.; Bochmann, M.; Romanov, A. S., Mono- versus
Bicyclic Carbene Metal Amide Photoemitters: Which Design Leads to the Best Performance?
Chemistry of Materials 2020, 32 (14), 6114-6122.
116. Romanov, A. S.; Jones, S. T. E.; Gu, Q.; Conaghan, P. J.; Drummond, B. H.; Feng, J.;
Chotard, F.; Buizza, L.; Foley, M.; Linnolahti, M.; Credgington, D.; Bochmann, M., Carbene metal
amide photoemitters: tailoring conformationally flexible amides for full color range emissions
including white-emitting OLED. Chemical Science 2020, 11 (2), 435-446.

106
117. Feng, J.; Yang, L.; Romanov, A. S.; Ratanapreechachai, J.; Reponen, A.-P. M.; Jones, S. T.
E.; Linnolahti, M.; Hele, T. J. H.; Köhler, A.; Bässler, H.; Bochmann, M.; Credgington, D.,
Environmental Control of Triplet Emission in Donor–Bridge–Acceptor Organometallics. Advanced
Functional Materials 2020, 30 (9), 1908715.
118. Hamze, R.; Idris, M.; Muthiah Ravinson, D. S.; Jung, M. C.; Haiges, R.; Djurovich, P. I.;
Thompson, M. E., Highly Efficient Deep Blue Luminescence of 2-Coordinate Coinage Metal
Complexes Bearing Bulky NHC Benzimidazolyl Carbene. Frontiers in Chemistry 2020, 8 (401).
119. Conaghan, P. J.; Matthews, C. S. B.; Chotard, F.; Jones, S. T. E.; Greenham, N. C.;
Bochmann, M.; Credgington, D.; Romanov, A. S., Highly efficient blue organic light-emitting diodes
based on carbene-metal-amides. Nature Communications 2020, 11 (1), 1758.
120. Yersin, H., Triplet emitters for OLED applications. Mechanisms of exciton trapping and control
of emission properties. In Transition Metal And Rare Earth Compounds III, 2004; Vol. 241, pp 1-26.
121. Thompson, M. E.; Djurovich, P. E.; Barlow, S.; Marder, S., Organometallic Complexes for
Optoelectronic Applications. In Comprehensive Organometallic Chemistry III, 1 ed.; O'Hare, D., Ed.
Elsevier Ltd.: Oxford, UK, 2007; Vol. 12, pp 102-194.
122. Bolink, H. J.; Coronado, E.; Costa, R. D.; Lardiés, N.; Ortí, E., Near-Quantitative Internal
Quantum Efficiency in a Light-Emitting Electrochemical Cell. Inorganic Chemistry 2008, 47 (20),
9149-9151.
123. Mahoro, G. U.; Fernandez-Cestau, J.; Renaud, J.-L.; Coto, P. B.; Costa, R. D.; Gaillard, S.,
Recent Advances in Solid-State Lighting Devices Using Transition Metal Complexes Exhibiting
Thermally Activated Delayed Fluorescent Emission Mechanism. Advanced Optical Materials 2020, 8
(16), 2000260.
124. Kalyanasundaram, K., Photophysics, photochemistry and solar energy conversion with
tris(bipyridyl)ruthenium(II) and its analogues. Coordination Chemistry Reviews 1982, 46, 159-244.
125. Kim, S.; Toste, F. D., Mechanism of Photoredox-Initiated C–C and C–N Bond Formation by
Arylation of IPrAu(I)–CF3 and IPrAu(I)–Succinimide. Journal of the American Chemical Society
2019, 141 (10), 4308-4315.
126. Keefe, M. H.; Benkstein, K. D.; Hupp, J. T., Luminescent sensor molecules based on
coordinated metals: a review of recent developments. Coordination Chemistry Reviews 2000, 205 (1),
201-228.
127. Lo, K. K.-W.; Louie, M.-W.; Zhang, K. Y., Design of luminescent iridium(III) and rhenium(I)
polypyridine complexes as in vitro and in vivo ion, molecular and biological probes. Coordination
Chemistry Reviews 2010, 254 (21), 2603-2622.
128. Liu, Q.; Wu, H.; Zhao, Z.; Wei, D., Macrometallocycle binuclear NHC silver(I) complex with
bridging azobenzene: Synthesis, structure and recognition for hydrogen sulfate. Tetrahedron 2019, 75
(23), 3128-3134.
129. Yang, Z.; Mao, Z.; Xie, Z.; Zhang, Y.; Liu, S.; Zhao, J.; Xu, J.; Chi, Z.; Aldred, M. P.,
Recent advances in organic thermally activated delayed fluorescence materials. Chemical Society
Reviews 2017, 46 (3), 915-1016.

107
130. Li, W.; Li, B.; Cai, X.; Gan, L.; Xu, Z.; Li, W.; Liu, K.; Chen, D.; Su, S.-J., Tri-Spiral
Donor for High Efficiency and Versatile Blue Thermally Activated Delayed Fluorescence Materials.
Angewandte Chemie International Edition 2019, 58 (33), 11301-11305.
131. Li, P.; Chan, H.; Lai, S.-L.; Ng, M.; Chan, M.-Y.; Yam, V. W.-W., Four-Coordinate Boron
Emitters with Tridentate Chelating Ligand for Efficient and Stable Thermally Activated Delayed
Fluorescence Organic Light-Emitting Devices. Angewandte Chemie International Edition 2019, 58
(27), 9088-9094.
132. Hu, J.; Li, Q.; Wang, X.; Shao, S.; Wang, L.; Jing, X.; Wang, F., Developing Through-Space
Charge Transfer Polymers as a General Approach to Realize Full-Color and White Emission with
Thermally Activated Delayed Fluorescence. Angewandte Chemie International Edition 2019, 58 (25),
8405-8409.
133. Luo, Y.; Li, S.; Zhao, Y.; Li, C.; Pang, Z.; Huang, Y.; Yang, M.; Zhou, L.; Zheng, X.; Pu,
X.; Lu, Z., An Ultraviolet Thermally Activated Delayed Fluorescence OLED with Total External
Quantum Efficiency over 9%. Advanced Materials 2020, 32 (32), 2001248.
134. Izumi, S.; Higginbotham, H. F.; Nyga, A.; Stachelek, P.; Tohnai, N.; Silva, P. d.; Data, P.;
Takeda, Y.; Minakata, S., Thermally Activated Delayed Fluorescent Donor–Acceptor–Donor–Acceptor
π-Conjugated Macrocycle for Organic Light-Emitting Diodes. Journal of the American Chemical
Society 2020, 142 (3), 1482-1491.
135. Strickler, S. J.; Berg, R. A., Relationship between Absorption Intensity and Fluorescence
Lifetime of Molecules. The Journal of Chemical Physics 1962, 37 (4), 814-822.
136. Tadle, A. C.; El Roz, K. A.; Soh, C. H.; Sylvinson Muthiah Ravinson, D.; Djurovich, P. I.;
Forrest, S. R.; Thompson, M. E., Tuning the Photophysical and Electrochemical Properties of AzaBoron-Dipyridylmethenes for Fluorescent Blue OLEDs. Advanced Functional Materials 2021, 31 (27),
2101175.
137. Chen, T.; Zheng, L.; Yuan, J.; An, Z.; Chen, R.; Tao, Y.; Li, H.; Xie, X.; Huang, W.,
Understanding the Control of Singlet-Triplet Splitting for Organic Exciton Manipulating: A Combined
Theoretical and Experimental Approach. Sci Rep 2015, 5, 10923.
138. Yang, J.-G.; Song, X.-F.; Wang, J.; Li, K.; Chang, X.; Tan, L.-Y.; Liu, C.-X.; Yu, F.-H.;
Cui, G.; Cheng, G.; To, W.-P.; Yang, C.; Che, C.-M.; Chen, Y., Highly Efficient Thermally Activated
Delayed Fluorescence from Pyrazine-Fused Carbene Au(I) Emitters. Chemistry – A European Journal
2021, 27 (71), 17834-17842.
139. Karon, K.; Lapkowski, M., Carbazole electrochemistry: a short review. Journal of Solid State
Electrochemistry 2015, 19 (9), 2601-2610.
140. Tzouras, N. V.; Martynova, E. A.; Ma, X.; Scattolin, T.; Hupp, B.; Busen, H.; Saab, M.;
Zhang, Z.; Falivene, L.; Pisanò, G.; Van Hecke, K.; Cavallo, L.; Cazin, C. S. J.; Steffen, A.; Nolan,
S. P., Simple Synthetic Routes to Carbene-M-Amido (M=Cu, Ag, Au) Complexes for Luminescence
and Photocatalysis Applications. Chemistry – A European Journal 2021, 27 (46), 11904-11911.
141. Hilborn, R. C., Einstein coefficients, cross sections, f values, dipole moments, and all that.
American Journal of Physics 1982, 50 (11), 982-986.

108
142. Greff, M.; Peter, B.; Stumpf, R. PROCESS FOR THE PREPARATION OF BENZIMIDAZO[
1,2-AJBENZIMIDAZOLES. 2017.
143. Ziolo, R. F.; Lipton, S.; Dori, Z., The photoluminescence of phosphine complexes of d10
metals. Journal of the Chemical Society D: Chemical Communications 1970, (17), 1124-1125.
144. Feng, J.; Reponen, A.-P. M.; Romanov, A. S.; Linnolahti, M.; Bochmann, M.; Greenham,
N. C.; Penfold, T.; Credgington, D., Influence of Heavy Atom Effect on the Photophysics of Coinage
Metal Carbene-Metal-Amide Emitters. Advanced Functional Materials 2021, 31 (1), 2005438.
145. Li, T.-y.; Shlian, D. G.; Djurovich, P. I.; Thompson, M. E., A Luminescent Two-Coordinate
AuI Bimetallic Complex with a Tandem-Carbene Structure: A Molecular Design for the Enhancement
of TADF Radiative Decay Rate. Chemistry – A European Journal 2021, 27 (20), 6191-6197.
146. Ma, J.; Kapper, S. C.; Ponnekanti, A.; Schaab, J.; Djurovich, P. I.; Thompson, M. E.,
Dynamics of rotation in two-coordinate thiazolyl copper(I) carbazolyl complexes. Applied
Organometallic Chemistry n/a (n/a), e6728.
147. Di, D.; Romanov, A. S.; Yang, L.; Richter, J. M.; Rivett, J. P. H.; Jones, S.; Thomas, T. H.;
Jalebi, M. A.; Friend, R. H.; Linnolahti, M.; Bochmann, M.; Credgington, D., High-performance
light-emitting diodes based on carbene-metal-amides. Science 2017, 356 (6334), 159-163.
148. Hamze, R.; Peltier, J. L.; Sylvinson, D.; Jung, M.; Cardenas, J.; Haiges, R.; Soleilhavoup,
M.; Jazzar, R.; Djurovich, P. I.; Bertrand, G.; Thompson, M. E., Eliminating nonradiative decay in
Cu(I) emitters: >99% quantum efficiency and microsecond lifetime. Science 2019, 363 (6427), 601-
606.
149. Shen, W.; Li, J.; Zhang, C.; Shi, M.; Zhang, J., Copper, Silver and Sodium Salt-Mediated
Quaternization by Arylation: Syntheses of N-Heterocyclic Carbene Precursors and 6-H-Phenanthridine
Derivatives. Chemistry – An Asian Journal 2016, 11 (13), 1883-1886.
150. Khramov, D. M.; Boydston, A. J.; Bielawski, C. W., Synthesis and Study of Janus
Bis(carbene)s and Their Transition-Metal Complexes. Angewandte Chemie International Edition 2006,
45 (37), 6186-6189.
151. Telfer, S. G.; McLean, T. M.; Waterland, M. R., Exciton coupling in coordination compounds.
Dalton Transactions 2011, 40 (13), 3097-3108.
152. Schinabeck, A.; Chen, J.; Kang, L.; Teng, T.; Homeier, H. H. H.; Suleymanova, A. F.;
Shafikov, M. Z.; Yu, R.; Lu, C.-Z.; Yersin, H., Symmetry-Based Design Strategy for Unprecedentedly
Fast Decaying Thermally Activated Delayed Fluorescence (TADF). Application to Dinuclear Cu(I)
Compounds. Chemistry of Materials 2019, 31 (12), 4392-4404.
153. Sagara, Y.; Shizu, K.; Tanaka, H.; Miyazaki, H.; Goushi, K.; Kaji, H.; Adachi, C., Highly
Efficient Thermally Activated Delayed Fluorescence Emitters with a Small Singlet–Triplet Energy Gap
and Large Oscillator Strength. Chemistry Letters 2015, 44 (3), 360-362.
154. Shafikov, M. Z.; Martinscroft, R.; Hodgson, C.; Hayer, A.; Auch, A.; Kozhevnikov, V. N.,
Non-Stereogenic Dinuclear Ir(III) Complex with a Molecular Rack Design to Afford Efficient
Thermally Enhanced Red Emission. Inorganic Chemistry 2021, 60 (3), 1780-1789.

109
155. Marciniak, H.; Auerhammer, N.; Ricker, S.; Schmiedel, A.; Holzapfel, M.; Lambert, C.,
Reduction of the Fluorescence Transition Dipole Moment by Excitation Localization in a Vibronically
Coupled Squaraine Dimer. The Journal of Physical Chemistry C 2019, 123 (6), 3426-3432.
156. Reichardt, C., Solvatochromic Dyes as Solvent Polarity Indicators. Chemical Reviews 1994,
94 (8), 2319-2358.
157. Shafikov, M. Z.; Zaytsev, A. V.; Kozhevnikov, V. N., Cyclometalation Geometry of the
Bridging Ligand as a Tuning Tool for Photophysics of Dinuclear Ir(III) Complexes. The Journal of
Physical Chemistry C 2021, 125 (37), 20531-20537.
158. Shafikov, M. Z.; Pander, P.; Zaytsev, A. V.; Daniels, R.; Martinscroft, R.; Dias, F. B.;
Williams, J. A. G.; Kozhevnikov, V. N., Extended ligand conjugation and dinuclearity as a route to
efficient platinum-based near-infrared (NIR) triplet emitters and solution-processed NIR-OLEDs.
Journal of Materials Chemistry C 2021, 9 (1), 127-135.
159. Huang, F.; Liu, H.; Li, X.; Wang, S., Enhancing hole injection by processing ITO through
MoO3 and self-assembled monolayer hybrid modification for solution-processed hole transport layerfree OLEDs. Chemical Engineering Journal 2022, 427, 131356.
160. Zhou, D.; Wu, S.; Cheng, G.; Che, C.-M., A gold(iii)–TADF emitter as a sensitizer for highcolor-purity and efficient deep-blue solution-processed OLEDs. Journal of Materials Chemistry C
2022, 10 (12), 4590-4596.
161. Teng, T.; Li, K.; Cheng, G.; Wang, Y.; Wang, J.; Li, J.; Zhou, C.; Liu, H.; Zou, T.; Xiong,
J.; Wu, C.; Zhang, H.-X.; Che, C.-M.; Yang, C., Lighting Silver(I) Complexes for Solution-Processed
Organic Light-Emitting Diodes and Biological Applications via Thermally Activated Delayed
Fluorescence. Inorganic Chemistry 2020, 59 (17), 12122-12131.
162. Weseliński, Ł. J.; Luebke, R.; Eddaoudi, M., A Convenient Preparation of 9H-Carbazole-3,6-
dicarbonitrile and 9H-Carbazole-3,6-dicarboxylic Acid. Synthesis 2014, 46 (05), 596-599.
163. Chang, B.; Shao, H.; Yan, P.; Qiu, W.; Weng, Z.; Yuan, R., Quinone-Mediated
Trifluoromethylation of Arenes and Heteroarenes with Visible Light. ACS Sustainable Chemistry &
Engineering 2017, 5 (1), 334-341.
164. Ma, J.; Kapper, S. C.; Ponnekanti, A.; Schaab, J.; Djurovich, P. I.; Thompson, M. E.,
Dynamics of rotation in two-coordinate thiazole copper (I) carbazole complexes Manuscript, 2022.
165. Li, T. Y.; Schaab, J.; Djurovich, P. I.; Thompson, M., Towards Rational Design of TADF Two
coordinate Coinage Metal Complexes: Understanding the Relationship between Natural Transition
Orbital Overlap and Photophysical Properties. Journal of Materials Chemistry C 2022.
166. Morimoto, H.; Tsubogo, T.; Litvinas, N. D.; Hartwig, J. F., A Broadly Applicable Copper
Reagent for Trifluoromethylations and Perfluoroalkylations of Aryl Iodides and Bromides. Angewandte
Chemie International Edition 2011, 50 (16), 3793-3798.
167. Larsen, M. A.; Hartwig, J. F., Iridium-Catalyzed C–H Borylation of Heteroarenes: Scope,
Regioselectivity, Application to Late-Stage Functionalization, and Mechanism. Journal of the
American Chemical Society 2014, 136 (11), 4287-4299.

110
168. Lishchynskyi, A.; Novikov, M. A.; Martin, E.; Escudero-Adán, E. C.; Novák, P.; Grushin, V.
V., Trifluoromethylation of Aryl and Heteroaryl Halides with Fluoroform-Derived CuCF3: Scope,
Limitations, and Mechanistic Features. The Journal of Organic Chemistry 2013, 78 (22), 11126-11146.
169. Hofbeck, T.; Yersin, H., The Triplet State of fac-Ir(ppy)3. Inorganic Chemistry 2010, 49 (20),
9290-9299.
170. Schaab, J.; Shi, S.; Jung, M. C.; Coburn, C.; Tadle, A.; M. R, D. S.; Djurovich, P. I.; Forrest,
S. R.; Thompson, M. E., Correction to “Highly Efficient Photo- and Electroluminescence from TwoCoordinate Cu(I) Complexes Featuring Nonconventional N-Heterocyclic Carbenes”. Journal of the
American Chemical Society 2023, 145 (1), 746-747.
171. Schaab, J.; Hamze, R.; Shi, S.; Kapper, S. C.; Muthiah Ravinson, D. S.; Estergreen, L.;
Jung, M.-C.; Tadle, A. C.; Haiges, R.; Djurovich, P. I.; Peltier, J. L.; Jazzar, R.; Bertrand, G.;
Bradforth, S. E.; Thompson, M. E., Correction to “ ‘Quick-Silver’ from a Systematic Study of Highly
Luminescent, Two-Coordinate, d10 Coinage Metal Complexes. Journal of the American Chemical
Society 2023, 145 (2), 1470-1471.
172. Sheldrick, G., Crystal structure refinement with SHELXL. Acta Crystallographica Section C
2015, 71 (1), 3-8.
173. Epifanovsky, E.; Gilbert, A. T. B.; Feng, X.; Lee, J.; Mao, Y.; Mardirossian, N.; Pokhilko,
P.; White, A. F.; Coons, M. P.; Dempwolff, A. L.; Gan, Z.; Hait, D.; Horn, P. R.; Jacobson, L. D.;
Kaliman, I.; Kussmann, J.; Lange, A. W.; Lao, K. U.; Levine, D. S.; Liu, J.; McKenzie, S. C.;
Morrison, A. F.; Nanda, K. D.; Plasser, F.; Rehn, D. R.; Vidal, M. L.; You, Z. Q.; Zhu, Y.; Alam,
B.; Albrecht, B. J.; Aldossary, A.; Alguire, E.; Andersen, J. H.; Athavale, V.; Barton, D.; Begam,
K.; Behn, A.; Bellonzi, N.; Bernard, Y. A.; Berquist, E. J.; Burton, H. G. A.; Carreras, A.; CarterFenk, K.; Chakraborty, R.; Chien, A. D.; Closser, K. D.; Cofer-Shabica, V.; Dasgupta, S.; de
Wergifosse, M.; Deng, J.; Diedenhofen, M.; Do, H.; Ehlert, S.; Fang, P. T.; Fatehi, S.; Feng, Q.;
Friedhoff, T.; Gayvert, J.; Ge, Q.; Gidofalvi, G.; Goldey, M.; Gomes, J.; Gonzalez-Espinoza, C. E.;
Gulania, S.; Gunina, A. O.; Hanson-Heine, M. W. D.; Harbach, P. H. P.; Hauser, A.; Herbst, M. F.;
Hernandez Vera, M.; Hodecker, M.; Holden, Z. C.; Houck, S.; Huang, X.; Hui, K.; Huynh, B. C.;
Ivanov, M.; Jasz, A.; Ji, H.; Jiang, H.; Kaduk, B.; Kahler, S.; Khistyaev, K.; Kim, J.; Kis, G.;
Klunzinger, P.; Koczor-Benda, Z.; Koh, J. H.; Kosenkov, D.; Koulias, L.; Kowalczyk, T.; Krauter,
C. M.; Kue, K.; Kunitsa, A.; Kus, T.; Ladjanszki, I.; Landau, A.; Lawler, K. V.; Lefrancois, D.;
Lehtola, S.; Li, R. R.; Li, Y. P.; Liang, J.; Liebenthal, M.; Lin, H. H.; Lin, Y. S.; Liu, F.; Liu, K. Y.;
Loipersberger, M.; Luenser, A.; Manjanath, A.; Manohar, P.; Mansoor, E.; Manzer, S. F.; Mao, S.
P.; Marenich, A. V.; Markovich, T.; Mason, S.; Maurer, S. A.; McLaughlin, P. F.; Menger, M.;
Mewes, J. M.; Mewes, S. A.; Morgante, P.; Mullinax, J. W.; Oosterbaan, K. J.; Paran, G.; Paul, A.
C.; Paul, S. K.; Pavosevic, F.; Pei, Z.; Prager, S.; Proynov, E. I.; Rak, A.; Ramos-Cordoba, E.;
Rana, B.; Rask, A. E.; Rettig, A.; Richard, R. M.; Rob, F.; Rossomme, E.; Scheele, T.; Scheurer,
M.; Schneider, M.; Sergueev, N.; Sharada, S. M.; Skomorowski, W.; Small, D. W.; Stein, C. J.; Su,
Y. C.; Sundstrom, E. J.; Tao, Z.; Thirman, J.; Tornai, G. J.; Tsuchimochi, T.; Tubman, N. M.;
Veccham, S. P.; Vydrov, O.; Wenzel, J.; Witte, J.; Yamada, A.; Yao, K.; Yeganeh, S.; Yost, S. R.;
Zech, A.; Zhang, I. Y.; Zhang, X.; Zhang, Y.; Zuev, D.; Aspuru-Guzik, A.; Bell, A. T.; Besley, N.
A.; Bravaya, K. B.; Brooks, B. R.; Casanova, D.; Chai, J. D.; Coriani, S.; Cramer, C. J.; Cserey,
G.; DePrince, A. E., 3rd; DiStasio, R. A., Jr.; Dreuw, A.; Dunietz, B. D.; Furlani, T. R.; Goddard,
W. A., 3rd; Hammes-Schiffer, S.; Head-Gordon, T.; Hehre, W. J.; Hsu, C. P.; Jagau, T. C.; Jung, Y.;
Klamt, A.; Kong, J.; Lambrecht, D. S.; Liang, W.; Mayhall, N. J.; McCurdy, C. W.; Neaton, J. B.;
Ochsenfeld, C.; Parkhill, J. A.; Peverati, R.; Rassolov, V. A.; Shao, Y.; Slipchenko, L. V.; Stauch,
T.; Steele, R. P.; Subotnik, J. E.; Thom, A. J. W.; Tkatchenko, A.; Truhlar, D. G.; Van Voorhis, T.;

111
Wesolowski, T. A.; Whaley, K. B.; Woodcock, H. L., 3rd; Zimmerman, P. M.; Faraji, S.; Gill, P. M.
W.; Head-Gordon, M.; Herbert, J. M.; Krylov, A. I., Software for the frontiers of quantum chemistry:
An overview of developments in the Q-Chem 5 package. J Chem Phys 2021, 155 (8), 084801.
174. Hermann, G.; Pohl, V.; Tremblay, J. C.; Paulus, B.; Hege, H.-C.; Schild, A., ORBKIT: A
modular python toolbox for cross-platform postprocessing of quantum chemical wavefunction data.
Journal of Computational Chemistry 2016, 37 (16), 1511-1520.
175. Castro, S. G. P. L., A.; et al. 2021 Python wrapper for Cubature: adaptive multidimensional
integration, Version 0.14.8; 2021.

Abstract (if available)

Abstract Two-coordinate carbene-MI-amide (cMa, MI = Cu, Ag, Au) complexes have emerged as highly efficient luminescent materials for use in a variety of photonic applications due to their extremely fast radiative rates through thermally activated delayed fluorescence (TADF) from an interligand charge transfer (ICT) process. This thesis presents a series of highly efficient luminescent 2-coordinate carbene-Gold-amide (cMa) complexes, to achieve very high radiative rates (kr=4x106 s-1) and near unity photoluminescence efficiencies. Temperature dependent photophysics allowed the determination of the singlet and triplet gap (ΔEST) and the singlet radiative rates. Theoretical calculations on hole and electron separation are used to explain the high radiative rates and offer a general design approach, to further improve this class of emitter.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Molecular design strategies for blue organic light emitting diodes

PDF

Synthesis, photo- and electroluminescence of three- and two-coordinate coinage metal complexes featuring non-N-heterocyclic carbene and non-conventional N-heterocyclic carbene ligands

PDF

Simple complexes: synthesis and photophysical studies of luminescent, monovalent, 2-coordinate carbene-coinage metal complexes and higher coordination geometries

PDF

Computational materials design for organic optoelectronics

PDF

Development of N-type chromophores for organic photovoltaics, and thermally activated delayed fluorescence NHC complexes for organic light-emitting diodes

PDF

Molecular design of pyridine and quinoline based compounds for organic optoelectronics

PDF

Photophysical properties of luminescent iridium and coinage metal complexes

PDF

On the pursuit of pushing the limits: from the isolation of highly reactive intermediates to the synthesis of energetic materials

PDF

Molecular design for organic photovoltaics: tuning state energies and charge transfer processes in heteroaromatic chromophores

PDF

Ir(III) and Pt(II) phosphorescent emitters in organic light emitting diodes: from materials development to light out-coupling

PDF

Predictive electronic structure methods for strongly correlated systems: method development and applications to singlet fission

PDF

Studies of molecular orientation using iridium phosphors and integration of corannulene into organic light emitting diodes (OLEDs)

PDF

Integration of cost effective luminescent materials into organic light emitting diodes (OLEDs) and molecular approaches to improve the device efficiency

PDF

Film deposition and optoelectronic properties of low-dimensional hybrid lead halides

PDF

Two-coordinate coinage metal complexes for solar fuels and organic LED chemistry

PDF

Effects of cationic species on the structure and properties of hybrid organic-inorganic materials

PDF

The converb -j in Alashan Mongolian

PDF

From first principles to machine intelligence: explaining charge carrier behavior in inorganic materials

PDF

Photoexcitation and nonradiative relaxation in molecular systems: methodology, optical properties, and dynamics

PDF

Cultivating seeds of support: growing the capacity of educators to create meaningful learning opportunities in math

Asset Metadata

Creator Schaab, Jonas (author)

Core Title Molecular modulation to fine-tune optoelectronic properties of OLED materials

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Chemistry

Degree Conferral Date 2024-05

Publication Date 06/12/2024

Defense Date 06/12/2024

Publisher Los Angeles, California (original), University of Southern California (original), University of Southern California. Libraries (digital)

Tag 2-coordinate,Amide,BIM,coordination,Emitter,gold,OAI-PMH Harvest,OLED,TADF

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Thompson, Mark E. (committee chair), Christe, Karl Otto (committee member), Nakano, Aiichiro (committee member)

Creator Email jonas.schaab92@gmail.com,jonassch@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC1139963AH

Unique identifier UC1139963AH

Identifier etd-SchaabJona-13093.pdf (filename)

Legacy Identifier etd-SchaabJona-13093

Document Type Dissertation

Format theses (aat)

Rights Schaab, Jonas

Internet Media Type application/pdf

Type texts

Source 20240614-usctheses-batch-1168 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu