Cationic iridium(iii) complexes bearing a bis(triazole) ancillary ligand

Article abstract of DOI:10.1039/c3dt50334h

Three new heteroleptic cationic iridium complexes of the form [Ir(C^N)(btl)]⁺, where btl = 1,1′-benzyl-4,4′-bi-1H-1,2, 3-triazolyl and C^N = 2-phenylpyridine (ppyH) (1), 1-benzyl-4-phenyl-1H-1,2,3- triazole (phtl) (2) or 1-benzyl-4-(2,4-difluorophenyl)-1H-1,2,3-triazole (dFphtl) (3), were synthesized and isolated as their hexafluorophosphate (PF₆^-) salts and fully characterized. The single crystal structure of 3 has been solved. Along the series from 1-3 the absorption spectra shift hypsochromically while the electrochemical gap increases from 3.25 to 3.54 to 3.88 V. Acetonitrile solutions of 1 and 2 are poorly luminescent, sky-blue emitters with predominant ligand-centered and charge transfer character, respectively. Theoretical calculations support these assignments. Complex 3 is not photostable and decomposes to solvento-based structures of the form [Ir(dFphtl)₂(ACN)_n]⁺ (n = 1, 2) through a dissociation and degradation of the btl ligand.

Full text of DOI:10.1039/c3dt50334h

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Cationic iridium(III) complexes bearing a bis(triazole)
ancillary ligand†
Cite this: Dalton Trans., 2013, 42, 8402
Loïc Donato, Philippe Abel and Eli Zysman-Colman*
Three new heteroleptic cationic iridium complexes of the form [Ir(C^N)(btl)]+, where btl = 1,1’-benzyl-
4,4’-bi-1H-1,2,3-triazolyl and C^N = 2-phenylpyridine (ppyH) (1), 1-benzyl-4-phenyl-1H-1,2,3-triazole
(phtl) (2) or 1-benzyl-4-(2,4-difluorophenyl)-1H-1,2,3-triazole (dFphtl) (3), were synthesized and isolated
as their hexafluorophosphate (PF6−) salts and fully characterized. The single crystal structure of 3 has
been solved. Along the series from 1–3 the absorption spectra shift hypsochromically while the electro-
chemical gap increases from 3.25 to 3.54 to 3.88 V. Acetonitrile solutions of 1 and 2 are poorly lumine-
scent, sky-blue emitters with predominant ligand-centered and charge transfer character, respectively.
Theoretical calculations support these assignments. Complex 3 is not photostable and decomposes to sol-
vento-based structures of the form [Ir(dFphtl)2(ACN)n]+ (n = 1, 2) through a dissociation and degradation
of the btl ligand.
Received 2nd February 2013,
Accepted 4th April 2013
DOI: 10.1039/c3dt50334h
www.rsc.org/dalton
bearing a btl N^N ligand was to (i) systematically investigate
the impact of triazole incorporation upon the optoelectronic
Introduction
The 1H-1,2,3-triazole (triazole) synthon, easily obtained in properties of the complex and (ii) promote a further blue-shift
high yield through a copper-catalyzed azide–alkyne cyclo- in the emission spectrum as part of our goal of developing
addition reaction,1 has been exploited in myriad applications, highly luminescent and stable deep-blue emitters.
including functional materials.2 Recently, we and others have
In this report, we describe the computational study, syn-
investigated the optoelectronic properties of iridium(III) com- thesis and characterization of three btl-containing cationic
plexes containing the triazole unit, which acts as a coordinat- iridium(III) complexes 1–3. In order to best assess the effect of
ing ligand. Cationic heteroleptic iridium(III) complexes the btl ancillary ligand, we contrast the properties of 1–3 with
incorporating 1N-alkyl(aryl)-4-aryl-1,2,3-triazoles (aryltriazoles, [(ppy)2Ir(bpy)]PF6 (4), [(phtl)2Ir(bpy)]PF6 (5) and [(dFphtl)2Ir-
atl) cyclometallating (C^N) ligands,3 1N-alkyl(aryl)-4-pyridyl- (bpy)]PF6 (6). While computations, UV-vis spectroscopy and
1,2,3-triazoles (pyridyltriazoles, pytl) ancillary (N^N) ligands,4 cyclic voltammetry (CV) indicate very large HOMO–LUMO
or both3b,5 have exhibited improved photoluminescent gaps, the photophysical investigation reveals that only blue-
quantum efficiencies (ΦPL) and bluer absorption and emission green to sky-blue emission is obtained, underscoring the
compared to their pyridine-containing congeners. The blue- difficulty in rationally designing deep-blue phosphorescent
shifting is due to a greater destabilization of the LUMO com- emitters (Chart 1).
pared to the HOMO upon replacement of the pyridine ring
with a triazole moiety. Still unexplored is the incorporation of
a 1,1′-alkyl(aryl)-4,4′-bi-1H-1,2,3-triazolyl (bistriazole, btl) unit
as an ancillary ligand. Indeed, the btl moiety has only been
Experimental section
General procedures
fleetingly explored as a bidentate ligand for Ru,6 Cu6a,7 and
Re.6a Our motivation for the study of iridium complexes
Commercial chemicals were used as supplied. All experiments
were carried out with freshly distilled anhydrous solvents
obtained from a Pure Solv™ solvent purification system from
Département de Chimie, Université de Sherbrooke, 2500 Boul. de l’Université,
Sherbrooke, QC, Canada, J1K 2R1. E-mail: Eli.Zysman-Colman@USherbrooke.ca;
Innovative Technologies except where specifically mentioned.
Triethylamine (Et3N) and diisopropylamine (iPr2NH) were dis-
tilled over CaH2 under a nitrogen atmosphere. Solvents
employed for the Click reaction were used without further
purification: these consisted of commercial lab grade metha-
nol and deionised water, which was accessible in house.
Fax: +1 819 821 8017; Tel: +1 819 821 7922
†Electronic supplementary information (ESI) available: 1H NMR spectra for btl,
1–3. 13C NMR spectra for btl, 1–2. 19F NMR for 1–3. Photophysical characteriz-
ation CV traces spectra for 1–3. Supplementary computational output, including
MO quantifications, ground and excited state dipole moment predictions. CCDC
922364 (3). For ESI and crystallographic data in CIF or other electronic format
see DOI: 10.1039/c3dt50334h
8402 | Dalton Trans., 2013, 42, 8402–8412
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
evaporated under reduced pressure to yield the product as a
white solid.
White solid. Yield: 64%. Rf: 0.25 (DCM–acetone 95/5; silica).
1H NMR (400 MHz, CD3CN) δ (ppm): 7.83 (s, 2H), 7.14–7.27
(m, 10H), 5.45 (s, 4H). 13C NMR (100 MHz, acetone-d6) δ
(ppm): 141.2, 137.1, 129.8, 129.2, 129.0, 121.8, 54.4. HRMS
(ES-Q-TOF) (C18H16N6Na+) Calculated: 339.1334; Experimental:
339.1332.
General procedure for formation of iridium(III) complexes
C^N ligand (2.20 equiv.) was mixed with IrCl3·6H2O (1.00
equiv.) in a mixture of 2-ethoxyethanol and water (6/1) to reach
a concentration in IrCl3·6H2O of 0.04 M. The mixture was
degassed by multiple vacuum and N2 purging cycles. The sus-
pension was heated at 130 °C for 24 h. A yellow precipitate
formed and was collected by filtration and used without
further purification. To this solid was added the btl ligand
(1.10 equiv.) and 2-ethoxyethanol was added to reach a concen-
tration in iridium of 0.04 M. The mixture was degassed by mul-
tiple vacuum and N2 purging cycles. The suspension was
heated at 130 °C for 24 h. The reaction mixture was cooled to
room temperature and diluted with water. The aqueous sus-
pension was washed several times with Et2O. The aqueous
layer was heated at 70 °C for 15 min and cooled back down
to room temperature. A solution of NH4PF6 (10 equiv., 1.0 g/
10 mL) was added dropwise to the aqueous phase to cause the
precipitation of a yellow solid. The suspension was cooled to
0 °C for 1 h, filtered and the resulting solid was washed with
cold water. The crude solid was purified by flash chromato-
graphy on silica gel using DCM to DCM–acetone (9/1).
Chart 1 Complexes in study.
All reactions were performed using standard Schlenk tech-
niques under an inert (N2) atmosphere, save for the Click reac-
tions. Flash column chromatography was performed using
silica gel (Silia-P from Silicycle, 60 Å, 40–63 μm). Analytical
thin layer chromatography (TLC) was performed with silica
plates with aluminum backings (250 μm with indicator F-254).
Compounds were visualized under UV light. 1H, 19F and 13C
NMR spectra were recorded on a Varian INOVA spectrometer at
400 MHz, 376 MHz and 100 MHz respectively. The following
abbreviations have been used for multiplicity assignments: “s”
for singlet, “d” for doublet, “t” for triplet, “q” for quartet and
“m” for multiplet. Perdeuterated acetonitrile (CD3CN) or
perdeuterated acetone (acetone-d6) were used as the solvents
of record. Spectra were referenced to the solvent peak and 19F
spectra were referenced to trifluoroacetic acid (TFA). Exact
mass measurements were performed on a quadrupole time-of-
flight (ESI-Q-TOF), model Maxis from Bruker in positive electro-
spray ionisation mode and spectra were recorded at the
Université de Sherbrooke. 1,4-Bis(trimethylsilyl)butadiyne,8
benzyl azide,9 1-benzyl-4-phenyl-1H-1,2,3-triazole,1 and
1-benzyl-4-(2,4-difluorophenyl)-1H-1,2,3-triazole1 were syn-
thesized following a literature procedure and the charac-
terization matches that found in the literature. UV-vis and
NMR-based photolysis studies were conducted in a photolysis
reactor Rayonet (model RPR-200) using 16 × 35 W excitation
lamps (λexc = 300 nm).
Iridium(III) bis[2-phenylpyridinato-N,C2′]-4,4′-bi-1H-1,2,3-
triazole, 1,1′-bis(phenylmethyl) hexafluorophosphate (1)
1
Yellow solid. Yield: 15%. Rf: 0.3 (DCM–acetone 9/1; silica). H
NMR (400 MHz, CD3CN) δ (ppm): 8.28 (s, 2H), 8.05 (d, J = 7.2
Hz, 2H), 7.88 (td, J = 7.2, 2.8 Hz, 2H), 7.71 (td, J = 8.0, 1.2 Hz,
4H), 7.35–7.40 (m, 4H), 7.18 (m, 2H), 7.07 (td, J = 5.6, 1.2 Hz,
2H), 6.96 (td, J = 8.0, 1.2 Hz, 4H), 6.82 (td, J = 8.0, 2.8 Hz, 4H),
6.26 (d, J = 8.0 Hz, 2H), 5.55 (s, 4H). 13C NMR (100 MHz,
acetone-d6) δ (ppm): 169.1, 150.4, 147.6, 145.6, 141.5, 139.5,
135.2, 132.8, 130.5, 130.1, 129.9, 129.1, 125.3, 124.6, 124.1,
123.0, 120.5, 56.1. 19F NMR (376 MHz, acetone-d6) δ (ppm):
−72.4 (d, J = 707 Hz, 6F). HRMS (ES-Q-TOF) (C40H32N8Ir+)
Calculated: 817.2379; Experimental: 817.2387.
4,4′-Bi-1H-1,2,3-triazole, 1,1′-bis(phenylmethyl), btl
Iridium(III) bis[(1′-phenyl)1,2,3-triazolato-N,C2′]-4,4′-bi-1H-
1,2,3-triazole, 1,1′-bis(phenylmethyl) hexafluorophosphate (2)
To 1,4-bis-(trimethylsilyl)-1,3-butadiyne (800 mg, 4.11 mmol,
1
1.0 equiv.) in 1 : 1 isopropanol–water (80 mL) were added White solid. Yield: 35%. Rf: 0.3 (DCM–acetone 9/1; silica). H
CuSO4·5H2O (467 mg, 1.87 mmol, 0.45 equiv.), K2CO3 (2.53 g, NMR (400 MHz, CD3CN) δ (ppm): 8.32 (s, 2H), 8.12 (s, 2H),
23.6 mmol, 6 equiv.), ascorbic acid (701 mg, 4.0 mmol, 0.95 7.44 (dd, J = 7.6, 1.2 Hz, 2H), 7.37–7.40 (m, 12H), 7.25 (td, J =
equiv.), benzyl azide (410 mg, 6.3 mmol, 1.5 equiv.) and pyri- 5.2, 2.4 Hz, 4H), 7.18 (td, J = 5.2, 2.4 Hz, 4H), 6.90 (td, J = 7.6,
dine (1.6 mL, 19.8 mmol, 5 equiv.). The mixture was placed 1.2 Hz, 2H), 6.73 (td, J = 7.6, 1.2 Hz, 2H), 6.18 (d, J = 7.6, 2H),
under a N2 atmosphere and then stirred for 24 h at room temp- 5.58 (d, J = 4.0 Hz, 4H), 5.52 (s, 4H). 13C NMR (100 MHz,
erature. To the mixture was added 50 mL of dichloromethane CD3CN) δ (ppm): 158.0, 142.7, 141.7, 136.9, 135.4, 135.1,
and 50 mL of 28% aqueous NH4OH. The organic layer was sep- 133.7, 130.0, 129.9, 129.8, 128.8, 128.6, 128.4, 123.6, 123.4,
arated and then dried over Na2SO4, filtered and the solvent 123.0, 120.6, 118.2, 56.2, 55.9. 19F NMR (376 MHz, acetone-d6)
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 8402–8412 | 8403

View Article Online
Paper
Dalton Transactions
δ (ppm): −72.2 (d, J = 707 Hz, 6F). HRMS (ES-Q-TOF) experimental uncertainty in the emission quantum yields is
(C48H40N12Ir+) Calculated: 977.3128; Experimental: 977.3108.
conservatively estimated to be 10%, though we have found that
statistically we can reproduce PLQYs to 3% relative error.
Time-resolved excited-state lifetime measurements were
performed using the time-correlated single photon counting
(TCSPC) option of the Jobin Yvon Fluorolog-3 spectrofluoro-
Iridium(III) bis[1′-(4′,6′-difluorophenyl)1,2,3-triazolato-N,C2′]-
4,4′-bi-1H-1,2,3-triazole, 1,1′-bis(phenylmethyl)
hexafluorophosphate (3)
White solid. Yield: 36%. Rf: 0.3 (DCM–acetone 85/15; silica). meter. A pulsed NanoLED at 341 nm (pulse duration <1 ns;
1H NMR (400 MHz, CD3CN) δ (ppm): 8.31 (s, 2H), 8.16 (d, J = fwhm = 14 nm), mounted directly on the sample chamber at
1.6, 2H), 7.16–7.38 (m, 20H), 6.59 (td, J = 9.2, 2.0 Hz, 2H), 5.71 90° to the emission monochromator, was used to excite the
(dd, J = 8.8, 2.4 Hz, 2H), 5.58 (d, J = 2.8, 4H), 5.51 (d, J = 1.2, samples and photons were collected using a FluoroHub single-
4H). 19F NMR (376 MHz, acetone-d6) δ (ppm): −72.3 (d, J = 707 photon-counting detector from Horiba Jobin Yvon. The lumine-
Hz, 6F), −110.9 (q, J = 9 Hz, 2F), −113.2 (t, J = 10 Hz, 2F). scence lifetimes were obtained using the commercially avail-
HR-MS (ES-Q-TOF) (C48H36F4N12Ir+) Calculated: 1049.2749; able Horiba Jobin Yvon Decay Analysis Software version 6.4.1,
Experimental: 1049.2737. Due to the low solubility of the included within the spectrofluorimeter. Lifetimes were deter-
product, we were unable to obtain a 13C NMR spectrum of mined through an assessment of the goodness of its mono-
sufficient resolution.
exponential fit by minimizing the chi-squared function (χ2)
and by visual inspection of the weighted residuals.
Photophysical measurements
Electrochemistry
All samples were prepared in either HPLC grade acetonitrile
(ACN) or distilled 2-methyltetrahydrofuran (2-MeTHF), with Cyclic voltammetry (CV) analyses were performed on an
concentrations on the order of 25 μM. Absorption spectra were Electrochemical Analyzer potentiostat model 600D from CH
recorded at room temperature in a 1.0 cm capped quartz Instruments. Solutions for cyclic voltammetry were prepared
cuvette, using a Shimadzu UV-1800 double beam spectro- in distilled ACN and degassed with ACN-saturated nitrogen
photometer. Molar absorptivity determination was verified by bubbling for ca. 15 min prior to scanning. Tetra(n-butyl)-
a linear least squares fit of values obtained from at least three ammonium hexafluorophosphate (TBAPF6; ca. 0.1 M in ACN)
independent solutions at varying concentrations with absorp- was used as the supporting electrolyte. A non-aqueous Ag/Ag+
tions ranging from 0.01 to 0.5 a.u. Steady-state emission electrode (silver wire in a solution of 0.1 M AgNO3 in ACN) was
spectra were obtained by exciting at the longest wavelength used as the pseudo-reference electrode; a glassy-carbon elec-
absorption maxima using a Horiba Jobin Yvon Fluorolog-3 trode was used for the working electrode and a Pt electrode
spectrofluorometer equipped with double monochromators was used as the counter electrode. The redox potentials are
and a photomultiplier tube detector (Hamamatsu model reported relative to a saturated calomel electrode (SCE) with a
R955). Emission quantum yields were determined using the ferrocenium/ferrocene (Fc+/Fc) redox couple as an internal
optically dilute method.10 A stock solution for each complex reference (0.38 V vs. SCE).12
with an absorbance of ca. 0.5 was prepared and then four
dilutions were obtained with different dilution factors result-
Computations
ing in optical dilution absorbances of ca. 0.05–0.2. The Beer– Density functional theory (DFT) calculations. All calculations
Lambert law was assumed to remain linear at the concen- were performed with the Gaussian 0913 suite. The level of
trations of the solutions. The emission spectra were then theory for all DFT14 and TD-DFT15 calculations was B3LYP;
measured after the solutions were rigorously degassed with excited-state triplet geometries were calculated using the
solvent-saturated nitrogen gas (N2) for 10 minutes prior to unrestricted B3LYP method (UB3LYP).16 The 6-31G* basis set17
spectrum acquisition using septa-sealed quartz cells from was used for C, H and N directly linked to iridium while the
Starna. For each sample, linearity between absorption and other C, H, N and F atoms were undertaken with the 3-21G*
emission intensity was verified through linear regression basis set,18 and the VDZ (valence double ζ) with SBKJC
analysis and additional measurements were acquired until the effective core potential basis set18a,19 was used for iridium.
Pearson regression factor (R2) for the linear fit of the dataset The X-ray structure of 3 was used as the input geometry for the
surpassed 0.98. Individual relative quantum yield values were S0 optimization calculation for 3 and the structure modified
calculated for each solution and the values reported represent accordingly for 1 and 2. Frequency calculations performed on
the slope obtained from the linear fit of these results. each optimized structure resulted in only positive vibrational
The equation Φs = Φr(Ar/As)(Is/Ir)(ns/nr)2 was used to calculate frequencies. The predicted phosphorescence wavelengths were
the relative quantum yield of the sample, where Φr is the obtained by energy differences between the triplet and singlet
absolute quantum yield of the reference, n is the refractive optimized states at their respective optimized geometries
20
index of the solvent, A is the absorbance at the excitation wave- (E0,0
)
and also as the energy difference between the triplet
length, and I is the integrated area under the corrected emis- and singlet optimized states at the optimized geometry of the
sion curve. The subscripts s and r refer to the sample and triplet state (EAE). The energy, oscillator strength and related
reference, respectively. A solution of [Ru(bpy)3](PF6)2 in ACN MO contributions for the 100 lowest singlet–singlet and 5
(Φr = 0.095)11 was used as the external reference. The lowest singlet–triplet excitations were obtained from the
8404 | Dalton Trans., 2013, 42, 8402–8412
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
TD-DFT/singlets and the TD-DFT/triplets output files, respecti-
vely. The calculated absorption spectra were visualized with
GaussSum 2.1 (fwhm: 1000 cm−1).21 An ACN quantum mech-
anical continuum solvation model as implemented within
Gaussian 09 was employed.22
Results and discussion
Ligand and complex synthesis
The 1,1′-dibenzyl-4,4′-bi-1H-1,2,3-triazolyl (btl) ligand was
obtained in 64% yield through a tandem silyl-deprotection
and CuI-catalyzed azide–alkyne cycloaddition reaction between
benzyl azide and 1,4-bis(trimethylsilyl)butadiyne, itself
obtained via a Glaser-type coupling reaction (Scheme 1),
similar to a protocol reported by Fletcher and co-workers.6c All
the complexes were synthesized through cleavage of the
respective μ-dichloro-bridged iridium(III) dimers with 2.22
equivalents of btl in refluxing 2-ethoxyethanol.23 Following
anion metathesis of the chloride salt with NH4PF6, pure com-
plexes were obtained as their hexafluorophosphate salts.
Complex 1 is a pale yellow solid while 2 and 3 are white solids.
Each complex was purified on silica gel and satisfactory micro-
analysis was obtained for each. Additionally and fortuitously,
we were able to grow single crystals of sufficient quality of 3 for
the structure to be solved by X-ray single crystal diffraction
analysis.
Fig. 1 ORTEP diagram at 30% probability of the [(dFphtl)2Ir(btl)]+ cation of 3
(unit cell parameters: a 12.729(2) Å; b 15.350(3) Å; c 15.924(3) Å; α 116.212(4)°;
β 93.122(4)°; γ 98.548(5)°). Counterions, solvent molecules and hydrogen atoms
have been removed for clarity. Selected bond parameters: Ir–CdFphtl = 2.020(5),
2.028(5); Ir–NdFphtl = 2.009(5), 2.030(5); Ir–Nbtl = 2.152(5), 2.156(4) Å; CdFphtl–Ir–
N
dFphtl = 80.2(2), 80.3(2); Nbtl–Ir–Nbtl = 75.99(17); (N–C–C–N)btl = 0.7(8)°.
interactions are not structure directing (or at least two confor-
mations are near equi-energetic).
The structure of 3 adopts a distorted octahedral coordi-
nation environment (Fig. 1). The coordinating nitrogen atoms
of the dFphtl ligands are disposed in a mutually trans con-
figuration while the cyclometallating carbon atoms are found
trans to the datively bound nitrogen atoms of the btl ligand.
The benzyl groups of the dFPhtl ligands are rotated away from
the btl while the benzyl groups of the latter ligand adopt a
mutually anti conformation. The bite angles about iridium are
Crystal structure
Single crystals of sufficient quality of 3 were grown from a
mixed solution of chloroform and diisopropyl ether by slow
very similar to those of [(bpy)2Ru(btl)]Cl2 [(N–Ru–N)btl
=
ˉ
diffusion. Complex 3 crystallizes in the triclinic space group P1
77.68(10)°],6b and the average CdFphtl–Ir–NdFphtl bond angle in
6 (79.6°).3a No twisting of the central C–C bond of the btl unit
is observed in 3. The bond lengths found in 3 are very similar
to the average bond lengths in the crystals structure of 6
(Ir–CdFphtl = 2.027, Ir–NdFPhtl = 1.988 and Ir–Nbpy = 2.129 Å; the
longer Ir–Nbtl bonds suggest a weaker coordination of the
ancillary ligand in 3 than in 6.
as a racemic mixture with one molecule (Λ-isomer) in the
asymmetric unit with the Δ-isomer related to the Λ-isomer via
the crystallographic inversion center. The structure of the
Λ-isomer is shown in Fig. 1. The structure additionally con-
tains two chloroform solvate molecules. The [PF6]− anion is
disordered over two sites in a 65 : 35 ratio with the two orien-
tations related via rotation about one of the molecular 4-fold
axes with the axial F atoms weakly hydrogen bonded to one of
the CHCl3 solvate molecules. In addition there are numerous
short CH–triazole⋯Fanion contacts in 3 though the disorder in
the anion position would indicate that either of these
Electrochemical characterization
The ground state electronics of 1–3 were probed through electro-
chemical studies by means of cyclic voltammetry (CV) and
are shown in Fig. 2. The CV data are summarized in Table 1.
All oxidation and reduction waves are irreversible. The oxi-
dation waves of both 1 and 2 are essentially identical at
ca. 1.27 V and similar to that found for [(ppy)2Ir(bpy)], 4 and
Ox.
[(phtl)2Ir(bpy)], 5 (E = 1.28 V for 4 and 1.29 V for 5).3a Thus,
pa
as with 4 and 5, replacement of the ppy ligands with phtl does
not influence the stability of the HOMO. The increased irrever-
sibility of the oxidation wave points to an increased contri-
bution from the C^N ligands to the mixed oxidation, which
also implicates an IrIII/IrIV redox couple. The oxidation in 3 is
shifted anodically by 370 mV compared to 2 due to the pres-
ence of the electron-withdrawing fluorine atoms on the dFphtl
Scheme 1 Synthesis of the btl ligand and complexes 1–3.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 8402–8412 | 8405

View Article Online
Paper
Dalton Transactions
C^N ligands, similar to that observed for [(dFphtl)2Ir(bpy)], 6 (1LL′CT) charge transfers. From the time-dependent DFT
Ox.
pa
(E = 1.64 V). The first reduction waves in 2 and 3 are separ- (TDDFT) calculations (vide infra), these CT transitions in 1 are
ated by only 30 mV and are assigned to the reduction of the btl mixed with LC transitions as well.
ligand, the reduction of 3 also modestly influenced by the
The photoluminescent emission spectra at 298 K in ACN
fluorine atoms on the C^N ligands. Analogously, the second solutions and at 77 K in 2-MeTHF rigid glass matrixes are also
reduction in 1 is assigned to reduction of the btl. The first shown in Fig. 3 and the photophysical data are summarized in
reduction event in 1 at −1.99 V is 280 mV less negative than 2 Table 2. The 298 K emission in 1 is somewhat structured with
and corresponds to the addition of an electron to the ppy a high-energy shoulder at 481 nm and a maximum at 511 nm.
ligands.24 The electrochemical gap increases from 3.25 This structured emission, combined with the absence of a
through to 3.88 V from 1 through 3, which are all substantially rigidochromic shift upon cooling to 77 K, is indicative of
larger than that found for sky-blue emitter 6 (ΔEredox = 3.09 V) emission with strong LC character. The 298 K emission
and sandwich the ΔEredox = 3.68 V measured for deep centred at 495 nm in 2 is unstructured, which implies a strong
blue emitter Ir(4-(t-butyl)-2′,6′-difluoro-2,3′-bipyridine)2(1,10- CT character to the emission. Upon cooling, there is
dimethyl-3,3′-methylenediimidazolium
carbene),
recently a rigidochromic shift of 4368 cm−1 and the 77 K emission
reported by Baranoff and co-workers.25 The trends observed in spectrum, as with 1, becomes structured. The 77 K emission
the CV traces and assignments are corroborated in their spectrum of 5 by contrast remains unstructured. Thus, due to
entirety by DFT calculations (vide infra).
the use of the more electron-rich btl ligand, the energy of the
LUMOs increases and results in increased LC character to the
emission. The assignments of the nature of the emission for 1
and 2 are supported by theoretical calculations (vide infra).
The emission maxima for 1 and 2 are each significantly blue-
shifted compared to 4 (λem = 602 nm) and 5 (λem = 580 nm).
Complex 1 in fact shows a similar photophysical profile to
[Ir(ppy)2(phpzpy)]PF6 (λem = 480 nm, ΦPL = 3%, τe = 0.18 μs),
where phpzpy = 2-(1-phenyl-1H-pyrazol-3-yl)pyridine.27 Un-
fortunately, 3 was found to not be photostable at ambient
Photophysical characterization
The absorption spectra for 1–3 in ACN show similar features
(Fig. 3). There is a progressive blue shift in the absorption
spectra from 1–3, which mirrors the increasing electrochemi-
cal gap observed in the CV studies. The high intensity absorp-
tion bands above 250 nm are assigned to ligand-based π→π*
transitions while the low intensity bands beyond 300 nm are
mixed charge-transfer (CT) transition involving both metal-to-
ligand (1MLCT), intra-ligand (1ILCT) and ligand-to-ligand
Fig. 2 CV traces for 1–3 in deaerated ACN with 0.1 M nBu4PF6 as the support-
Fig. 3 Absorption in ACN solution at 298 K, emission in ACN at 298 K and
emission in 2-MeTHF at 77 K for 1–3.
ing electrolyte at 298 K. ν = 0.1 V s−1
.
Table 1 Electrochemical data of 1–3a
Ox.
pa
Red.
pc
Red.
pc
Red.
c
Complex
E
E
E
E
pc 3
EHOMO b/eV
ELUMO b/eV
ΔEredox
1
2
1
2
3
1.26
1.27
1.64
−1.99
−2.27
−2.24
−2.23
—
−2.43
−2.41
—
—
−6.06
−6.07
−6.44
−2.79
−2.53
−2.56
3.25
3.54
3.88
a Measured in N2-saturated ACN (ca. 1.5 mM) with nBu4NPF6 (ca. 0.1 M), 298 K, glassy carbon 3 mm as the working electrode, ν = 0.05 V s−1
.
Potentials are reported in V versus SCE and were calibrated with an internal standard Fc/Fc+ redox couple (0.38 V vs. SCE in ACN).12 Epa = anodic
potential. Epc = cathodic potential. b The HOMO and LUMO energies were calculated using the relation EHOMO/LUMO = −(Eonset Ox./Red. vs. Fc + 4.8)
ox
red
eV.26 c ΔEredox = E − E
.
pc
pa
8406 | Dalton Trans., 2013, 42, 8402–8412
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Table 2 Spectroscopic and photophysical data for 1–3a
λem/nm
τe/μs
λabs (ε) /nm (×103 M−1 cm−1
)
298 K
77 K
ΦPL b/% 298 K
77 K
kr/×105 s−1
knr/×105 s−1
1
2
3
252 (45.5); 330 (8.55); 375 (4.50) 481 (0.86); 511 (1.0) 470 (1.0); 506 (0.88) 2.8
2.17
4.77 0.129
5.66
4.48
204
229 (65.5); 300 (8.16)
226 (20.8); 294 (8.80)
495
407 (1.0); 434 (0.79) 2.7
393 (1.0); 419 (0.84)
4.77 × 10−2 15.8
c
a Measurements at 298 K in ACN and at 77 K in the 2-MeTHF glass state. Absorption measurements in aerated solution and 298 K emission
measurements in N2-saturated solution. Relative intensities of structured emission spectra in parentheses. b Measured using [Ru(bpy)3]PF6 (ΦPL
=
9.5% in ACN) as the reference.11 c Complex not photostable. See text.
temperature (vide infra) but did exhibit structured emission at
77 K that is further blue-shifted compared to 2.
Photoluminescent quantum yields for 1 and 2 were modest
at ∼3%. However, their excited-state decay kinetics are quite
distinct. Compound 1 shows microsecond 298 K lifetimes that
are approximately doubled at 77 K. By contrast, the 298 K τe
for 2 is only 48 ns while the emission is very long-lived at
15.8 μs and 77 K. This translates to a 44 fold increase in the
radiative rate constant, kr, offset by a 46 fold increase in the
non-radiative rate constant, knr.
Photodecomposition studies of 3
Upon excitation of a dilute ACN solution of single crystals of 3
at 290 nm (concentrations below 1 μM), we observed unusual
behaviour in the emission spectrum with maxima at 352 and
505 nm. Over successive scans, the peak at 505 nm decreased
in intensity (Fig. 4a). LC-HRMS analysis of an aliquot after 20
scans revealed several unidentified decomposition products,
including two solvento iridium complexes as the major pro-
ducts, [Ir(dFphtl)2(ACN)2]+ and [Ir(dFphtl)2(ACN)(S)]+, where S
is another weakly bound solvent molecule.
In order to investigate the nature of this decomposition, we
conducted photolysis experiments of 3 (λexc = 300 nm, Rayonet
photochemical reaction model RPR-200 operating with 16
lamps of 35 W) over 20 min. Decomposition progression was
monitored firstly by UV-vis absorption at μM concentrations
(Fig. 4b). An isosbestic point at ca. 210 nm was observed with
the 1LC absorption band at 229 nm decreasing in intensity.
1
The CT band at 296 nm also decreases in intensity and shifts
hypsochromically over the course of irradiation. Finally, in a
subsequent experiment, decomposition of 3 was monitored by
1H NMR in CD3CN at mM concentrations (Fig. 4c). Signals
characteristic of the btl in 3, such as the singlet at 8.33 ppm
associated with the proton of the triazole, disappear during
the photolysis, suggesting decomplexation and subsequent
degradation of the btl ligand. The proton of the triazole frag-
ment of the dFphtl ligands at 8.19 ppm disappears and is
replaced by a singlet at 8.15 ppm. A similar analysis can be
made for the benzylic protons wherein the doublet of doublets
at 5.73 and the doublet at 5.54 ppm disappear over the course
Fig. 4 Evidence of photodecomposition of 3. (a) Emission spectra obtained in
deaerated sub-μM ACN solution at 298 K (λexc = 290 nm); (b) absorption spectra
obtained in aerated μM ACN solution during photolysis experiments (see text);
(c) 1H NMR spectra in mM CD3CN solution during photolysis experiments (see
of the photolysis while the doublet at 5.61 ppm converges into
1
a singlet. These characteristic H NMR signals are reminiscent
of those found for N,N-trans-[Ir(dFphtl)2(CD3CN)Cl] (1H
NMR in CD3CN), reported by Fernández-Hernández and co- text).
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 8402–8412 | 8407

View Article Online
Paper
Dalton Transactions
workers.3d LC-HRMS analysis conducted after 20 min
Fig. 5 shows compiled Kohn–Sham energy diagrams of the
irradiation revealed several unassigned low molecular weight five highest-energy occupied and the five lowest-energy
decomposition products along with the previously detected unoccupied molecular orbitals (MOs) for 1–3 along with elec-
[Ir(dFphtl)2(ACN)2]+ and [Ir(dFphtl)2(ACN)(S)]+. From these experi- tron density contour plots of selected MOs. Quantification of
ments we can conclude that the energy required to excite 3 the localization of the electron density within each of
also promotes a dissociation and decomposition of the btl these MOs along with their relative energies are shown in
ligand, with acetonitrile solvent molecules coordinating the Fig. S7–S9† for 1–3, respectively. For each of the complexes, the
vacant sites about iridium. This decomposition pathway is dis- HOMO is distributed on the iridium and the C^N ligands with
tinct from the more commonly observed one in blue-emitting decreasing iridium contribution from 43% to 39% to 32%
iridium complexes, which involves photochemical cleavage of along the series. The HOMO − 1 is localized almost exclusively
one of the two Ir–NC^N bonds, necessitating a geometric on the C^N ligands for all three complexes. For 1, the electron
change and thus thermally accessing a metal-centered (3MC) density in HOMO − 2 to HOMO − 4 is shared between the Ir
state.28 This non-emissive state is then prone to subsequent and C^N. The HOMO − 2 and HOMO − 3 for 2 and the HOMO
adventitious attack by a small molecule.29
− 2 and HOMO − 4 for 3 possess topologies with significant
C^N contributions. The HOMO − 4 in 2 and the HOMO − 3 in
3 include contributions from the btl of 18 and 32%, respecti-
vely. The LUMO for 1 is localized exclusively (94%) on the ppy
while for 2 and 3 it is mainly (75%) on the btl. The LUMO + 1
for all three complexes is located in the C^N ligand. The
LUMO + 2 in 1 is mainly localized on the btl while in 2 and 3 it
is localized on the C^N ligands.
DFT computations
A detailed density functional theory (DFT) and time-dependent
DFT (TDDFT) study was undertaken to model the opto-
electronic properties of 1–3. We have previously shown that
a
similar computational methodology incorporating the
SBKJC-VDZ basis set on iridium accurately predicts both
ground- and excited-state properties for triazole-containing
iridium complexes.3a,g In order to assess the quality of the pre-
dictions, we contrasted the computed ground state structures
of 1–3 with those, respectively, of the reference X-ray structures
for 4–6 along with the X-ray structure of 3 obtained in this
study. An analysis of Table 3 reveals a good accord between the
predicted geometries and those from the X-ray structures.
Computed Ir–CC^N and Ir–NC^N bond lengths are overesti-
mated by less than 3%. Replacement of a bpy ligand with a btl
results in a slight elongation of the computed Ir–NN^N bond of
0.069, 0.064, 0.048 Å, respectively in 1 versus 4, 2 versus 5 and 3
versus 6; the predicted Ir–NN^N bond length in 3 overestimates
the average bond lengths in the crystal structure by only 0.3%.
The computations thus reproduce the weaker coordination of
the btl compared to bpy. The bite angle for the btl is compar-
able to that of bpy. The computations also accurately predict
the CC^N–Ir–NC^N bond angle. The geometry in the triplet state
remains essentially unchanged compared to that in the
ground state. Therefore, we believe that the agreement
between computed geometries and those of the crystal struc-
tures of the reference complexes 4–6, along with a more direct
comparison with 3, is sufficient to ensure a proper description
of their optoelectronic properties.
Replacement of ppy with phtl destabilizes the LUMO in 2,
resulting in a 0.41 eV increase in the HOMO–LUMO gap com-
pared to 1. Addition of fluorine atoms onto the C^N ligands in
3 promotes stabilization of both the HOMO and LUMO. The
stabilization is more pronounced for the occupied MOs result-
ing in a 0.24 eV increase in the HOMO–LUMO gap from 2 to 3.
While the HOMO is relatively isolated from the other lower-
energy occupied MOs, the same is not the case for the LUMO.
For 1, the LUMO + 1 is very close in energy to the LUMO and is
destabilized by only 0.07 eV. For 2, LUMO + 1 through LUMO +
3 are essentially electronically degenerate and are only 0.19 eV
higher in energy than the LUMO. Similarly observed for 3,
these same MOs are separated by only 0.06 eV with the LUMO
+ 1 destabilized by 0.15 eV compared to the LUMO. The phtl
and dFphtl ligands possess higher energy π* orbitals than ppy,
permitting less mixing in the excited state and conserving
greater CT character in the emission.
We have calculated the 100 lowest-energy vertical singlet–
singlet and 5 lowest-energy singlet–triplet transitions using
TDDFT to permit assignment of the nature of the different
absorption bands and to provide insight into the nature of the
emission. The most important low energy vertical transitions
are summarized in Table 4. Unlike the case found in most
Table 3 Selected computed S0 geometric parameters, bond lengths in Å and bond angles in deg (°), for 1–3a
1
4b
2
5c
3
3d
6c
(Ir–CC^N avg
(Ir–NC^N avg
(Ir–NN^N avg
(CC^N–Ir–NC^N avg
N^N–Ir–NN^N
)
)
)
2.022
2.076
2.202
80.4
2.014
2.045
2.133
80.4
2.045
2.042
2.192
79.9
2.010
1.994
2.128
78.3
2.045
2.049
2.177
80.4
2.024
2.020
2.154
80.3
2.027
1.988
2.129
79.6
)
N
74.8
76.2
75.3
76.3
75.5
76.0
77.0
a Average distances and angles reported. b Data from ref. 30. c Data from ref. 3a. d Data taken from the Λ isomer found in the crystal structure
(vide supra).
8408 | Dalton Trans., 2013, 42, 8402–8412
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Fig. 5 Calculated energy level scheme for the Kohn–Sham orbitals between HOMO − 4 to LUMO + 4 for 1–3, including contour plots (0.02 isovalue) of selected
most relevant MOs and the associated DFT calculated HOMO–LUMO energy gap (in eV).
Table 4 Principal theoretical low energy singlet–singlet and singlet–triplet electronic transitions with corresponding oscillator strengths (f) and assignments for
1–3a
Energy
Oscillator strength
State Nature of transition (contribution in %)
Primary character
(eV) (nm) (f)
1
2
3
S1
S5
S7
S9
T1
T2
HOMO→LUMO (98%)
HOMO→L + 4 (92%)
H − 4→LUMO (10%), H − 2→LUMO (71%)
H − 2→LUMO (14%), H − 1→L + 1 (74%)
H − 1→L + 1 (18%), HOMO→LUMO (63%)
H − 1→LUMO (26%), HOMO→L + 1 (53%)
MLCTppy/LCppy/ILCTppy
ML′CT/LL′CT/ILCTppy
MLCTppy/LCppy/ILCTppy
LCppy
LCppy/MLCTppy/ILCTppy
LCppy/MLCTppy/ILCTppy
3.25 382
3.88 320
3.98 312
4.04 307
2.88 431
2.92 425
0.1128
0.0442
0.0653
0.1246
0
0
S1
S5
S8
S10
T1
T2
HOMO→LUMO (94%)
MLCTbtl/LL′CT
3.69 336
4.14 299
4.31 288
4.45 279
3.32 374
3.34 371
0.0020
0.1351
0.0322
0.0486
0
HOMO→L + 4 (45%), HOMO→L + 5 (48%)
H − 1→LUMO (86%)
H − 1→L + 1 (77%)
H − 1→L + 6 (17%), HOMO→L + 1 (12%), HOMO→L + 5 (21%)
H − 1→L + 5 (21%), HOMO→L + 1 (12%), HOMO→L + 6 (22%)
LCphtl/MLCTbtl/LL′CT
MLCTphtl/LL′CT
MLCTphtl/ILCTphtl
MLCTphtl/LCphtl
MLCTphtl/LCphtl
0
S1
S5
S6
T1
HOMO→LUMO (93%)
H − 1→LUMO (91%)
HOMO→L + 4 (47%), HOMO→L + 5 (42%)
H − 1→L + 1 (14%), H − 1→L + 6 (17%), HOMO→L + 2 (25%),
HOMO→L + 5 (16%),
MLCTbtl/LL′CT
MLCTbtl/LLCT
MLCT/LL′CT/LCdFphtl
3.94 315
4.36 284
4.42 281
3.39 366
0.0018
0.0344
0.1307
0
ILCTdFphtl/LCdFphtl
MLCTdFphtl
/
T2
H − 1→L + 2 (16%), H − 1→L + 5 (18%), HOMO→L + 1 (22%),
HOMO→L + 6 (18%),
ILCTdFphtl/LCdFphtl
MLCTdFphtl
/
3.41 364
0
a H = HOMO; L = LUMO; MLCT = metal-to-ligand charge transfer; LL′CT = ligand-to-ligand charge transfer; ILCT = intra-ligand charge transfer;
LC = ligand-centered. For the primary character, the order indicated reflects the predominance of the different transitions (threshold >10%).
cationic iridium complexes, the oscillator strength for the increases markedly for higher-energy transitions (λabs
<
HOMO→LUMO transition in 1 is large ( f = 0.1128) and 300 nm). Overall, the calculations reproduce the principal fea-
involves redistribution of the electron density about the ppy tures of the experimental absorption spectra, however this
ligands. This transition is best described as possessing mixed comparison is rendered difficult due to the dearth of identifi-
charge-transfer (CT)/ligand-centered (LC) character. By con- able bands and structure.
trast, the principal low-energy and low-intensity transitions for
For each complex, the T1 and T2 states reside within 0.04 eV
2 and 3 are mixed metal-to-ligand/ligand-to-ligand MLCT/LL′ of one another. Ascription of the nature of emission is difficult
CT and implicate the btl unit. The proportion of LC character given the great number of distinct transitions required to
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 8402–8412 | 8409

View Article Online
Paper
Dalton Transactions
The excited-state dipole in 1 remains essentially unchanged
compared to the ground state dipole. There is a slight decrease
of 3.2 D in the magnitude of the excited state dipole in 2 while
its orientation is not significantly impacted. By contrast, both
the orientation and the magnitude of the excited state dipole
in 3 are significantly perturbed with a calculated increase of
7.9 D.
Conclusion
Fig. 6 Calculated spin density distribution (0.0004 isovalue) for 1–3.
In conclusion, a family of cationic iridium complexes incorpor-
ating a bis(triazole) ancillary ligand have been synthesized and
their optoelectronic properties characterized. When the cyclo-
metallating ligand is ppy, the experimental data and compu-
tations point to a mostly ligand-centered, blue-green emission.
When the main ligands are based on an aryltriazole motif, the
characterization and theoretical studies imply an emission
with much greater charge-transfer character. With phtl as the
C^N ligand, emission is sky-blue but very short-lived while
with dFphtl the complex is not photostable at room temp-
erature. Their photophysical properties coupled with electro-
chemical irreversibility thus do not render these complexes as
prime candidates for incorporation into electroluminescent
devices.
describe each excited state and it is a challenge to deconvolute
the ratio of LC/CT character in the emission. However, for each
of the complexes, these triplet states largely implicate the C^N
ligands and the iridium but not the btl ligand, a behavior dis-
tinct from that calculated for 5 and 6.3a Thus, based on the
TDDFT analysis, emission in each involves a combination of
LCC^N, ILCTC^N and MLCTC^N. Complex 1 possesses the great-
est amount of 3LC character while emissions for 2 and 3
possess significant 3CT character. These assignments are
closely aligned with the experimental findings.
In order to obtain a better description of the nature of the
lowest-energy triplet state, the geometry and energy of the T1
state for each complex were fully optimized using unrestricted
B3LYP (UB3LYP). The spin densities for each of 1–3 (Fig. 6)
are localized on the iridium and one of the C^N ligands.
From an analysis of the spin density distribution, the
Acknowledgements
EZ-C acknowledges CFI (Canadian Foundation for Innovation),
NSERC (the National Sciences and Engineering Research
Council of Canada), FQRNT (Le Fonds québécois de la
recherche sur la nature et les technologies) and the Université
de Sherbrooke for financial support. The authors wish to
thank Sébastien Ladouceur and Dr Martina Sandroni for a
critical reading of the manuscript and Prof. Jeremy Rawson for
helpful discussions concerning the crystal structure of 3.
emission results from
a mixed LCC^N/ILCTC^N/MLCTC^N
state, corroborating the TDDFT calculations. Importantly, from
these calculations the btl moiety is not involved in the emis-
sion. The calculations in concert with the experimental results
3
suggest that in 1 the LC character is more pronounced than
in 2 or 3.
The predicted emission energies were assessed using two
approaches: Method 1 is the energy difference between the T1
and S0 states at their respective optimized geometries and best
reflects the E0,0 band of the 77 K emission; Method 2 is an
adiabatic energy difference between the T1 and S0 states at the
optimized geometry of the T1 state and best reflects the emis-
sion energy at 298 K. The predicted emission energies employ-
ing method 1 (E0,0) are 426, 403 and 413 nm while those
employing method 2 (EAE) are 468, 508 and 501 nm, respecti-
vely, for 1–3. While the predictions match very well with 2
(λem(77 K) = 407 nm; λem(298 K) = 495 nm), they are somewhat
less reliable for 1 (λem(77 K) = 470 nm; λem(298 K) = 511 nm); 3
is photochemically unstable at 298 K but luminesces at low
temperature (λem(77 K) = 393 nm) rendering a comparison
within method 2 impossible (vide supra). The overestimation
of the emission energy of 1 regardless of the method employed
suggests that the calculations are underestimating the
importance of the 3CT character to the mixed 3LC/3CT
emission of 1. For each complex, the ground-state dipoles
are oriented as to bifurcate the btl ligand (Fig. S10–S12†).
References
1 S. Ladouceur, A. M. Soliman and E. Zysman-Colman, Syn-
thesis, 2011, 2011, 3604, 3611.
2 M. Juríček, P. H. J. Kouwer and A. E. Rowan, Chem.
Commun., 2011, 47, 8740–8749.
3 (a) S. Ladouceur, D. Fortin and E. Zysman-Colman, Inorg.
Chem., 2011, 50, 11514–11526; (b) B. Beyer, C. Ulbricht,
D. Escudero, C. Friebe, A. Winter, L. González and
U. S. Schubert, Organometallics, 2009, 28, 5478–5488;
(c) M. de Barros e Silva Botelho, J. M. Fernandez-Hernan-
dez, T. B. de Queiroz, H. Eckert, L. De Cola and A. S. S. de
Camargo, J. Mater. Chem., 2011, 21, 8829–8834;
(d) J. M. Fernández-Hernández, C.-H. Yang, J. I. Beltrán,
V. Lemaur, F. Polo, R. Frõhlich, J. Cornil and L. De Cola,
J.
Am.
Chem.
Soc.,
2011,
133,
10543–10558;
(e) K. N. Swanick, S. Ladouceur, E. Zysman-Colman and
Z. Ding, Chem. Commun., 2012, 48, 3179–3181;
8410 | Dalton Trans., 2013, 42, 8402–8412
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
(f) K. N. Swanick, S. Ladouceur, E. Zysman-Colman and
Z. Ding, Angew. Chem., Int. Ed., 2012, 51, 11079–11082;
(g) K. Hasan and E. Zysman-Colman, Inorg. Chem., 2012,
51, 12560–12564.
4 (a) M. Felici, P. Contreras-Carballada, Y. Vida,
J. M. M. Smits, R. J. M. Nolte, L. De Cola, R. M. Williams
and M. C. Feiters, Chem.–Eur. J., 2009, 15, 13124–13134;
(b) E. Orselli, R. Q. Albuquerque, P. M. Fransen,
J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene,
J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas,
J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaus-
sian Inc., Wallingford, CT, 2009.
R. Frohlich, H. M. Janssen and L. De Cola, J. Mater. Chem., 14 (a) P. Hohenberg and W. Kohn, Phys. Rev. Sect. B, 1964,
2008, 18, 4579–4590; (c) M. Juríček, M. Felici, P. Contreras-
Carballada, J. Lauko, S. Rodríguez Bou, P. H. J. Kouwer,
A. M. Brouwer and A. E. Rowan, J. Mater. Chem., 2011, 21,
2104–2111; (d) S. Liu, P. Müller, M. K. Takase and
T. M. Swager, Inorg. Chem., 2011, 50, 7598–7609;
(e) M. Mydlak, C. Bizzarri, D. Hartmann, W. Sarfert,
136, 864; (b) W. Kohn and L. Sham, Phys. Rev. Sect. A, 1965,
140, 1133; (c) The Challenge of d and f Electrons, ed.
D. R. Salahub and M. C. Zerner, ACS, Washington, DC,
1989; (d) R. G. Parr and W. Yang, Density-functional
theory of atoms and molecules, Oxford Univ. Press, Oxford,
1989.
G. Schmid and L. De Cola, Adv. Funct. Mater., 2010, 20, 15 (a) R. E. Stratmann, G. E. Scuseria and M. J. Frisch,
1812–1820; (f) S. Zanarini, M. Felici, G. Valenti,
M. Marcaccio, L. Prodi, S. Bonacchi, P. Contreras-Carbal-
lada, R. M. Williams, M. C. Feiters, R. J. M. Nolte, L. De
Cola and F. Paolucci, Chem.–Eur. J., 2011, 17, 4640–4647.
J. Chem. Phys., 1998, 109, 8218; (b) R. Bauernschmitt and
R. Ahlrichs, Chem. Phys. Lett., 1996, 256, 454;
(c) M. E. Casida, C. Jamorski, K. C. Casida and
D. R. Salahub, J. Chem. Phys., 1998, 108, 4439.
5 M. Felici, P. Contreras-Carballada, J. M. M. Smits, 16 (a) A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652;
R. J. M. Nolte, R. M. Williams, L. De Cola and M. C. Feiters,
Molecules, 2010, 15, 2039–2059.
6 (a) U. Monkowius, S. Ritter, B. König, M. Zabel and
(b) C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens.
Matter, 1988, 37, 785–789; (c) B. Miehlich, A. Savin, H. Stoll
and H. Preuss, Chem. Phys. Lett., 1989, 157, 200–206.
H. Yersin, Eur. J. Inorg. Chem., 2007, 2007, 4597–4606; 17 V. A. Rassolov, J. A. Pople, M. A. Ratner and T. L. Windus,
(b) C. E. Welby, S. Grkinic, A. Zahid, B. S. Uppal, J. Chem. Phys., 1998, 109, 1223–1229.
E. A. Gibson, C. R. Rice and P. I. P. Elliott, Dalton Trans., 18 (a) J. S. Binkley, J. A. Pople and W. J. Hehre, J. Am. Chem.
2012, 41, 7637–7646; (c) J. T. Fletcher, B. J. Bumgarner,
N. D. Engels and D. A. Skoglund, Organometallics, 2008, 27,
5430–5433; (d) A. Mattiuzzi, I. Jabin, C. Moucheron and
A. Kirsch-De Mesmaeker, Dalton Trans., 2011, 40, 7395–
7402.
7 U. Monkowius, S. Ritter, B. Konig, H. Yersin and M. Zabel,
Acta Crystallogr., Sect. E: Struct. Rep. Online, 2008, 64, m195.
8 I. J. S. Fairlamb, P. S. Bauerlein, L. R. Marrison and
J. M. Dickinson, Chem. Commun., 2003, 632–633.
9 (a) T. N. Hoheisel and H. Frauenrath, Org. Lett., 2008, 10,
4525–4528; (b) L. S. Campbell-Verduyn, L. Mirfeizi,
R. A. Dierckx, P. H. Elsinga and B. L. Feringa, Chem.
Commun., 2009, 2139–2141.
10 G. A. Crosby and J. N. Demas, J. Phys. Chem., 1971, 75, 991–
1024.
11 H. Ishida, S. Tobita, Y. Hasegawa, R. Katoh and K. Nozaki,
Coord. Chem. Rev., 2010, 254, 2449–2458.
12 V. V. Pavlishchuk and A. W. Addison, Inorg. Chim. Acta,
2000, 298, 97–102.
13 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
Soc., 1980, 102, 939; (b) M. S. Gordon, J. S. Binkley,
J. A. Pople, W. J. Pietro and W. J. Hehre, J. Am. Chem. Soc.,
1982, 104, 2797; (c) W. J. Pietro, M. M. Francl, W. J. Hehre,
D. J. Defrees, J. A. Pople and J. S. Binkley, J. Am. Chem. Soc.,
1982, 104, 5039; (d) K. D. Dobbs and W. J. Hehre,
J. Comput. Chem., 1986, 7, 359; (e) K. D. Dobbs and
W. J. Hehre, J. Comput. Chem., 1987, 8, 861; (f) K. D. Dobbs
and W. J. Hehre, J. Comput. Chem., 1987, 8, 880.
19 (a) W. J. Stevens, W. J. Basch and M. Krauss, J. Chem. Phys.,
1984, 81, 6026; (b) W. J. Stevens, M. Krauss, H. Basch and
P. G. Jasien, Can. J. Chem., 1992, 70, 612; (c) T. R. Cundari
and W. J. Stevens, J. Chem. Phys., 1993, 98, 5555–5565.
20 (a) S. Ladouceur, D. Fortin and E. Zysman-Colman, Inorg.
Chem., 2010, 49, 5625–5641; (b) M. S. Lowry, W. R. Hudson,
R. A. Pascal Jr. and S. Bernhard, J. Am. Chem. Soc., 2004,
126, 14129–14135.
21 N. M. O’Boyle, GaussSum 2.0, Dublin City University,
Dubin, Ireland, 2006.
22 J. Tomasi, B. Mennucci and R. Cammi, Chem. Rev., 2005,
105, 2999–3094.
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, 23 M. Nonoyama, Bull. Chem. Soc. Jpn., 1974, 47, 767–768.
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, 24 A. A. Rachford, R. Ziessel, T. Bura, P. Retailleau and
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
F. N. Castellano, Inorg. Chem., 2010, 49, 3730–3736.
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, 25 S. B. Meier, W. Sarfert, J. M. Junquera-Hernandez,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J.A. Montgomery, Jr., J. E. Peralta,
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers,
M. Delgado, D. Tordera, E. Orti, H. J. Bolink, F. Kessler,
R. Scopelliti, M. Gratzel, M. K. Nazeeruddin and
E. Baranoff, J. Mater. Chem., 2013, 1, 58–68.
K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, 26 C. M. Cardona, W. Li, A. E. Kaifer, D. Stockdale and
K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
G. C. Bazan, Adv. Mater., 2011, 23, 2367–2371.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 8402–8412 | 8411

View Article Online
Paper
Dalton Transactions
27 L. He, L. Duan, J. Qiao, D. Zhang, L. Wang and Y. Qiu,
Chem. Commun., 2011, 47, 6467–6469.
28 Y. You and W. Nam, Chem. Soc. Rev., 2012, 41, 7061–7084.
R. D. Costa, E. Ortì, M. Sessolo, S. Graber, K. Doyle,
M. Neuburger, C. E. Housecroft and E. C. Constable, Adv.
Mater., 2008, 20, 3910–3913.
29 (a) J. D. Slinker, J.-S. Kim, S. Flores-Torres, J. H. Delcamp, 30 R. D. Costa, E. Ortì, H. J. Bolink, S. Graber, S. Schaffner,
H. D. Abruna, R. H. Friend and G. G. Malliaras, J. Mater.
Chem., 2007, 17, 76–81; (b) H. J. Bolink, E. Coronado,
M. Neuburger, C. E. Housecroft and E. C. Constable, Adv.
Funct. Mater., 2009, 19, 3456–3463.
8412 | Dalton Trans., 2013, 42, 8402–8412
This journal is © The Royal Society of Chemistry 2013