Luminescent Cyclometalated Gold(III) Alkyl Complexes: Photophysical and Photochemical Properties

Article abstract of DOI:10.1021/acs.inorgchem.7b00180

A series of luminescent cyclometalated gold(III) complexes having alkyls as auxiliary ligands has been prepared. The alkyl ligand was found to effectively increase the emission quantum yields and lifetimes of luminescent cyclometalated gold(III) complexes

Full text of DOI:10.1021/acs.inorgchem.7b00180

This is an open access article published under an ACS AuthorChoice License, which permits
copying and redistribution of the article or any adaptations for non-commercial purposes.
Article
pubs.acs.org/IC
Luminescent Cyclometalated Gold(III) Alkyl Complexes:
Photophysical and Photochemical Properties
Wai-Pong To,† Glenna So Ming Tong,† Chi-Wah Cheung,† Chen Yang,† Dongling Zhou,†
and Chi-Ming Che*,†,‡
†State Key Laboratory of Synthetic Chemistry, Institute of Molecular Functional Materials, HKU-CAS Joint Laboratory on New
Materials, Department of Chemistry, The University of Hong Kong, Pokfulam Road, Hong Kong, People’s Republic of China
‡HKU Shenzhen Institute of Research and Innovation, Shenzhen 518053, People’s Republic of China
S
* Supporting Information
ABSTRACT: A series of luminescent cyclometalated gold(III)
complexes having alkyls as auxiliary ligands has been prepared.
The alkyl ligand was found to effectively increase the emission
quantum yields and lifetimes of luminescent cyclometalated
gold(III) complexes by circumventing the population of LLCT
excited states that are found in complexes supported by
arylacetylide ligands. These gold(III) alkyl complexes exhibit
emission quantum yields and lifetimes of up to 0.40 and 180
μs, respectively, in solution at room temperature. The triplet
emission color of these complexes is tunable from yellow to sky
blue by modifying the cyclometalating ligand.
low-lying π and π* orbitals of the alkyl ligand for forming a
LLCT excited state that could complicate the excited state
dynamics, which occurs in gold(III) complexes with
arylacetylide ligands.2j,3 Herein we report the synthesis and
photophysical properties of a series of cyclometalated gold(III)
complexes with a doubly C deprotonated [C∧N∧C] chelating
ligand and an alkyl group as the auxiliary ligand (Chart 1).
These complexes exhibit strong ligand-centered phosphor-
escence in solution at room temperature with E0−0 emission
INTRODUCTION
■
In the development of phosphorescent gold(III) complexes, the
use of strong σ-donating carbon donor ligand(s) for
coordination is a common strategy used to achieve phosphor-
escence at room temperature.1 Coordination of a strong σ-
donating ligand to gold(III) can raise the energy of the
2
2
gold(III) 5dσ* (5dx −y ) orbital, thus minimizing the thermal
population of the nonemissive 3LMCT (ligand-to-metal charge
transfer) and 3dd excited state. In the case of [Au(C∧N∧C)C
CAr], this would also increase the energy of the 3LLCT
(ligand-to-ligand charge transfer) 3[π(CCAr)−π*-
(C∧N∧C)]) excited state due to the destabilization of the
LUMO on pyridine by the strong-field ligand. In the literature,
most phosphorescent gold(III) complexes are constructed with
a combination of a cyclometalating [C∧N] or [C∧N∧C] ligand
(e.g., HC∧N = 2-phenylpyridine, HC∧N∧CH = 2,6-diphenyl-
pyridine) together with auxiliary ligand(s).2 The function of the
cyclometalating ligand is to restrict the four-coordinated
gold(III) complex in a planar coordination geometry from
undergoing distortion, while the auxiliary ligands are employed
to provide a strong ligand field so as to destabilize the gold(III)
Chart 1. Chemical Structures of Gold(III) Complexes in
This Study
2
2
5dx −y orbital. At present, the most commonly used auxiliary
carbon donor ligands that can successfully switch on the
emission of gold(III) complexes include acetylide, N-hetero-
cyclic carbene (NHC), aryl, and cyanide.1 As the σ-donor
strength of the ligand is crucial to effect luminescence, we
conceive that alkyl ligands may be an appropriate choice in the
construction of phosphorescent gold(III) complexes since (1)
the sp3-hybridized carbon donor atom should provide a strong
Received: January 20, 2017
2
2
ligand field to destabilize the 5dx −y orbital and (2) there are no
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.7b00180
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
energies as high as 2.80 eV and quantum yields and lifetimes of
up to 0.40 and 180 μs, respectively. The two structurally similar
complexes [Au(C∧N∧C)R] (R = Me, Et) have been reported
by Bochmann and co-workers, but their photophysical
properties have not been studied in detail.2i
SEM and TEM images of most gold(III) complexes show
entangled fibrous structures on the order of micrometers.4b−e
Given the small size of the alkyl ligand, the use of the ligand
may allow intermolecular dispersive and/or π−π interactions
which induce the formation of nanostructures. The morphol-
ogies of the nanostructures obtained with 1a, 1c, 4, and 5 in
THF/hexane (v/v 1/9) mixture were therefore examined
(Figure 2a−d). Interestingly, rectangular blocks, a fibrous
RESULTS
■
Synthesis and Characterization. Cyclometalating
[C∧N∧C] ligands were synthesized by a Krohnke pyridine
̈
synthesis or Suzuki coupling similar to our previous work.2j
Coordination of these ligands to gold(III) ions was achieved by
transmetalation reactions between the mercury(II) chloride
complexes bearing monodeprotonated [HC∧N∧C] ligands and
KAuCl4 in refluxing MeCN.2j The chloro ligand of the gold(III)
complex was substituted with trifluoroacetate (OCOCF3) by
reaction with AgOCOCF3 in CH2Cl2.2g Alkylation was
accomplished by reaction of [Au(C∧N∧C) (OCOCF3)] with
alkyllithium (alkylmagnesium bromide for 1c) in THF at −78
°C and subsequent warming to room temperature.2h The
products were purified by column chromatography on SiO2,
1
and characterized by H NMR, 13C NMR, mass spectrometry,
and elemental analysis. The complexes are stable to air and
moisture.
Diffraction-quality crystals of 1a and 4 were obtained by slow
evaporation of THF/hexane solutions of the respective
complex at room temperature. Perspective views of 1a and 4
are depicted in Figure 1. Selected bond lengths and angles are
Figure 2. (a−d) SEM images of nanostructures obtained from 1a, 1c,
4, and 5, respectively, in THF/hexane (v/v 1/9). (e) SAED pattern of
nanowire formed from 5. (f) Photograph showing the gel of 1c (0.8 wt
%) in DMSO in a tilted vial.
network, rods, and wires were observed for 1a,c, 4, and 5,
respectively. The dimensions of these structures ranged from
100 nm to 15 μm. Selected area electron diffraction (SAED)
experiments revealed a crystalline nature of these structures.
For example, the SAED pattern of nanowires of 5 (Figure 2e)
showed a regular d spacing of 3.58 Å characteristic of π−π
interactions along the rod longitude. The gelation properties of
these complexes were also tested. Only 1c was found to form
an opaque, white gel upon cooling a hot DMSO solution of the
complex (0.8 wt %) to room temperature (Figure 2f). Gelation
was not observed for 1c in other solvents. Variable-temperature
1H NMR spectra of 1c in D6-DMSO (Figure 3) revealed an
upfield shift for H2 and downfield shifts for H1, H3, and H6,
Figure 1. Perspective views of 1a (left) and 4 (right). Hydrogen atoms
are not shown.
summarized in Table S1 in the Supporting Information.
Coordination of the tridentate cyclometalating [C∧N∧C] ligand
to the gold atom in both 1a and 4 leads to a distorted-square-
planar geometry with C(sp2)−Au−C(sp2) angles in the range
of 161.6(4)−162.0(2)°. The N, Au, and C(sp3) atoms are
almost collinear, with N−Au−C(sp3) angles of 178.5−178.65°.
The Au−C(sp2) distances are in the range of 2.050(8)−
2.076(5) Å, whereas the Au−C(sp3) distances vary from
2.051(12) to 2.121(5) Å, both of which are similar to those of
structurally comparable complexes.2i The [Au(C∧N∧C)]
moieties are essentially planar. The closest interplanar distances
between two [Au(C∧N∧C)] planes are 3.397 Å for 1a and
3.382 Å for 4, and the closest Au−Au distances between two
adjacent molecules of 1a and 4 are 5.271 and 5.036 Å,
respectively.
1
For the gold(III)−methyl complexes studied herein, the H
NMR signal of the Au−CH3 group appears as a sharp singlet
lying between 1.21 and 1.40 ppm; complex 5 is an exception, in
which the signal is a triplet and is more downfield at 1.86 ppm.
The 13C NMR signal of the methyl group ranges from 1.21 to
2.0 ppm.
Figure 3. Variable-temperature 1H NMR spectra showing the aromatic
protons of 1c in D6-DMSO (400 MHz, [1c] = 2 × 10−3 M). No NMR
signal of 1c was recorded at 25 °C due to low solubility.
Electron Microscopy. Morphology studies of gold(III)
complexes are relatively sparse in the literature.4 The reported
B
DOI: 10.1021/acs.inorgchem.7b00180
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 1. Photophysical and Electrochemical Data of Gold(III) Complexes
emission
reduction
c
a
a
ab
,
complex
UV−vis absorption λmax (nm) (ε (103 dm3 mol−1 cm−1))
λmax (nm) (τ (μs))
Φ
Epc (V)
1a
1b
1c
2
260 (sh, 24.6), 280 (14.1), 308 (9.3), 317 (10.0), 354 (3.2), 370 (3.8), 389 (3.1)
259 (sh, 26.7), 280 (15.0), 307 (10.0), 317 (10.7), 352 (3.2), 369 (3.7), 387 (3.0)
259 (sh, 22.8), 280 (12.5), 307 (8.3), 317 (9.0), 353 (2.7), 370 (3.1), 387 (2.5)
260 (sh, 25.3), 279 (13.0), 306 (8.9), 316 (9.7), 349 (2.9), 365 (3.4), 383 (2.8)
257 (28.8), 301 (6.8), 313 (7.4), 356 (2.6), 371 (3.5), 389 (3.0)
465, 498, 530 (43)
465, 499, 526 (31)
465, 498, 526 (39)
464, 497, 530 (40)
463, 496, 524 (77)
462, 494, 523 (38)
464, 499, 528 (45)
0.11
0.09
0.09
0.10
0.17
0.10
0.11
−2.16
−2.22
−2.20
−2.32
−2.12
−2.28
−2.07
3
4
252 (sh, 41.0), 302 (7.7), 312 (9.2), 352 (2.7), 367 (3.8), 384 (3.3)
5
260 (sh, 20.1), 274 (sh, 15.0), 283 (sh, 13.0), 310 (12.0), 323 (11.6), 340 (sh, 3.7), 358 (3.9),
376 (2.9)
6
7
8
274 (10.3), 282 (sh, 9.9), 310 (sh, 3.0), 334 (sh, 1.5), 352 (2.3), 368 (3.5), 385 (3.1)
455, 485, 517 (56)
0.14
0.02
−2.10
−2.31
−2.04
258 (sh, 43.8), 286 (sh, 7.9), 294 (sh, 6.8), 302 (sh, 5.2), 318 (sh, 2.7), 332 (3.2), 347 (4.0), 364 (3.2) 443, 472, 504 (7)
276 (48.1), 299 (sh, 31.0), 320 (br, 26.3), 332(sh, 24.5), 348 (23.0), 377 (sh, 6.7), 398 (9.3),
416 (8.9)
531, 570, 618 (180) 0.40
a
b
Determined in a degassed CH2Cl2 solution at 2 × 10−5 M. Emission quantum yield (Φ) was estimated with BPEA (9,10-
c
bis(phenylethynyl)anthracene) in benzene as the standard (Φ = 0.85). Measured in CH2Cl2 under argon with 0.1 M tetra-n-butylammonium
hexafluorophosphate as the supporting electrolyte; values versus Ag/AgNO3 (0.1 M in MeCN). E(Cp2Fe+/0) ranges between 0.05 and 0.10 V.
whereas only upfield shifts were observed for all aliphatic
protons upon decreasing the temperature from 80 to 40 °C
(Figure S5 in the Supporting Information), suggesting that
hydrophobic−hydrophobic interactions between the long alkyl
chains and π−π interactions between the C∧N∧C plane of 1c
molecules are involved in the gelation process.4e
UV−Visible Absorption Spectroscopy. The UV−visible
absorption and emission spectral data of all these gold(III) alkyl
complexes in CH2Cl2 solutions at room temperature are given
in Table 1. Absorption spectra of selected complexes are shown
in Figure 4. Complexes 1a−c and 2−7 feature strong
acetylide counterparts, but with a significant blue shift of 14−16
nm for the absorption peak maxima at 398−416 nm (ε = (8.9−
9.3) × 103 dm3 mol−1 cm−1).2j
Luminescence Spectroscopy. With the use of an alkyl
ligand as a strong σ-donor, the as-prepared gold(III) complexes
are emissive in solutions at room temperature. As shown in
Table 1 and Figures 5 and 6, complexes with substituted 2,6-
Figure 5. Emission spectra of 1a, 2, and 6−8 in degassed CH2Cl2.
Figure 4. UV−vis absorption spectra of 1a, 2, and 6−8 in CH2Cl2.
absorption bands (ε = (1−4) × 104 dm3 mol−1 cm−1) at
250−330 nm and moderately intense, vibronic bands (ε = (2−
4) × 103 dm3 mol−1 cm−1) at 350−390 nm. The absorption
spectrum of 1a is insensitive to solvent polarity (maximum shift
4 nm, Figure S3 in the Supporting Information). On the basis
of the electrophilic nature of gold(III) and insensitivity of
absorption bands to solvent polarity, the absorptions of the
gold(III) alkyl complexes are attributable to metal perturbed
1IL transitions localized on the [C∧N∧C] ligand. Since the alkyl
ligands in the gold(III) complexes do not have a low-lying π
orbital, the contribution of a LLCT transition which is often
proposed in [Au(C∧N∧C)CCAr] complexes is excluded.2c
Complex 8, which bears a fluorene unit in the cyclometalating
ligand, displays an absorption profile similar to that of its
Figure 6. Absorption and emission spectra of 7 in CH2Cl2, showing
the Stokes shift and vibronic spacings.
C
DOI: 10.1021/acs.inorgchem.7b00180
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
diphenylpyridine as the cyclometalating ligand (1a−c, 2−7)
display vibronic structured emission with peak maxima at 443−
465 and 472−499 nm. The emission color ranges from green
(1a−c, 2−5) to blue-green (6) to sky blue (7). Their emission
quantum yields and lifetimes in deoxygenated CH2Cl2 solutions
are up to 0.17 and 77 μs, respectively, both of which are the
highest reported values for gold(III) complexes having
substituted C-deprotonated 2,6-diphenylpyridine as the cyclo-
metalating ligand.2c,d,g,o The radiative decay rate constants (kr)
of these complexes are in the range of (2.2−2.9) × 103 s−1,
whereas their nonradiative decay rate constants (knr) span from
1.1 × 104 to 2.9 × 104 s−1, except for 7, which is an outlier,
showing a relatively high knr value of 1.4 × 105 s−1. The
excitation spectra monitored at the emission maxima exhibit the
same vibronic structure (300−400 nm) as those observed in
the absorption spectra. The emission peak maxima of 1a are
insensitive to solvent polarity (464−466 nm) (Figure S3 in the
Supporting Information), but the emission quantum yield and
lifetime of the complex are significantly affected (toluene, Φ =
0.09, τ = 35 μs; DMF, Φ = 0.02, τ = 8 μs; Table S2 in the
Supporting Information). The vibronic structured emission,
large Stokes shift (4200−5000 cm−1), and small radiative decay
rate constant ((2−3) × 103 s−1) all point to an emission origin
dynamics. As presented in Figure 8, the emissions of 5 at
460−550 and ∼650 nm decay with lifetimes of 0.30 and 0.67
3
of a IL π−π* excited state of the [C∧N∧C] ligand. The
emission self-quenching rate constants of 5, 7, and 8 are 9.4 ×
108, 2.8 × 108, and 1.2 × 107 dm3 mol−1 s−1, respectively.
Complex 8 displays a yellow emission with a quantum yield and
lifetime of 0.40 and 180 μs in degassed CH2Cl2, respectively.
The emission of all these complexes in aerated solutions is
barely visible to the naked eye.
In glassy solution (MeOH/EtOH, 1/4) at 77 K, all of these
complexes display an intense, vibronic structured emission with
energies slightly blue shifted (2−4 nm) from those recorded in
dilute solution at room temperature (Figure 7). The emission
lifetimes are greatly lengthened to up to 400 μs. For example,
the emission lifetime of 1a increases from 43 μs at room
temperature to 381 μs at 77 K.
Figure 8. (a) Time-resolved emission spectra of 5 in the solid state at
room temperature. Emission kinetics of 5 in the solid state at 650 nm
at (b) room temperature and (c) 77 K. (d) Steady-state and (e, f)
time-resolved emission spectra of 7 in the solid state at room
temperature.
μs, respectively, suggesting that the emission comes from
different electronic excited states. Emission kinetics recorded
for the lower energy emission (Figure 8b) reveal time delays of
∼70 and ∼800 ns to reach the maximum emission intensity
after laser pulse excitation (355 nm) at room temperature and
at 77 K, respectively. In PMMA thin film, 5 displays a vibronic
structured emission similar to that in solution with a quantum
yield of up to 0.13 (Figure S4 in the Supporting Information).
However, the low-energy emission at ∼650 nm could not be
clearly observed in the film with up to 60 wt % of 5 in PMMA
thin film. For complexes 6 and 7, the emission was found to
comprise two components with different lifetimes, which is
similar to the case for 5. Complex 6 displays a short-lived
emission (τ ≈ 60 ns) with λmax at ∼435 nm and a structureless
emission with λmax at 555 nm and a lifetime of 4.1 μs. For
complex 7, a relatively short lived (τ = 0.5 μs) vibronic
emission band at 446−515 nm was observed from 0 to 2 μs.
After this emission faded out, a much longer lived (τ = 110 μs)
structureless emission with maximum at 515 nm was observed.
Concentration-Dependent 1H NMR Spectroscopy.
Concentration-dependent 1H NMR spectroscopic measure-
ments were conducted for 1a, 5, and 7 to elucidate the different
solid-state emission behaviors.4e Figure 9 shows the aromatic
Figure 7. Normalized emission spectra of 1a, 7 ,and 8 in glassy
solution (MeOH/EtOH, 1/4) at 77 K.
In the solid state, most of these complexes are weakly
emissive at room temperature with emission profiles similar to
those in solution but with greatly reduced emission lifetimes
(<1 μs). Interestingly, complexes 5−7 show a red-shifted
emission band in addition to the emission at 430−500 nm.
Time-resolved emission measurements conducted on these
complexes provide more information on the emission
1
region of the H NMR spectra of these complexes recorded at
2, 4, and 6 × 10−3 M at 298 K in CD2Cl2. The 1H NMR spectra
D
DOI: 10.1021/acs.inorgchem.7b00180
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 10. Cyclic voltammograms of 1a, 7, and 8 in CH2Cl2.
430 nm. The decay lifetime (27 μs) of the absorption at 340
nm is close to the emission lifetime (29 μs) under the same
conditions. A similar profile was also observed in deoxygenated
toluene. Under oxygen-saturated conditions, a similar absorp-
tion profile was observed, but it decayed quickly to the baseline
level with a lifetime of 0.13 μs. In the presence of 0.05 M
diisopropylethylamine (DIPEA), a different spectrum with
absorption peak maxima at 330, 425, and 600 nm was observed
(Figure 11b). These absorption peaks decayed much more
slowly with a time duration of ∼400 μs, which is attributed to
the decay of one-electron-reduced 1a via reductive quenching
of triplet excited 1a by DIPEA. The low stability of the as-
formed 1a− corroborates the irreversible nature of the
reduction wave of 1a observed in cyclic voltammetry
measurements.
1
Figure 9. Aromatic region of the H NMR spectra of 1a (top), 5
(middle), and 7 (bottom) recorded at 2, 4, and 6 × 10−3 M at 298 K in
CD2Cl2.
The excited state difference absorption spectrum of 5 in
CH2Cl2 at 8 × 10−5 M shares a profile similar to that observed
for 1a in THF (Figure 11c). The decay lifetime (11.9 μs) of the
absorption at ∼340 nm is close to the emission lifetime (11.4
μs) under the same conditions. Both lifetimes are much smaller
than that observed at 2 × 10−5 M due to emission self-
quenching. Complex 7 displayed absorption maxima at ∼320,
of both 5 and 7 show upfield shifts (up to 0.09 ppm) for all
aromatic proton signals upon increasing concentration. The
proton signals of the gold−methyl group of 5 also shifted from
1.89 to 1.82 ppm. However, only negligible shifts (<0.01 ppm)
were observed for all proton signals of 1a under the same
conditions. These data suggest that both 5 and 7 have a
stronger proclivity toward aggregation, which may account for
why excimeric emissions are observed in the cases of 5 and 7
but not 1a.
Electrochemistry. The electrochemical properties of these
gold(III) alkyl complexes were investigated by cyclic
voltammetry in CH2Cl2 containing 0.1 M nBu4NPF6. The
data are summarized in Table 1, and the cyclic voltammograms
of 1a, 7, and 8 are shown in Figure 10. All of the complexes
show one irreversible reduction wave with Epc ranging from
−2.04 to −2.32 V vs Ag/AgNO3 (0.1 M in MeCN). These
reduction waves are assigned to reduction at the pyridine
moiety. An irreversible oxidation wave is observed for 8 with
Epa at +1.3 V. The potentials of the reduction waves are affected
by the substituents on the [C∧N∧C] ligand. For example, with
the addition of a tert-butyl group at the para position of the
pyridine, the Epc value shows a cathodic shift from −2.16 V
(1a) to −2.32 V (2).
430, and 500 nm with decay lifetimes of 6.0 μs at 1.2 × 10−4
M
(Figure 11d). However, it was noted that the absorption profile
recorded at 16 μs showed absorption peaks at ∼320 and 375
nm which are different from those recorded immediately after
the laser excitation, indicating that another species may have
been formed from the triplet excited state of 7. The solution
after time-resolved absorption measurement was then subjected
to light irradiation at 365 nm and monitored by UV−vis
absorption spectroscopy. Within 30 min, the signal showed a
gradual change to another absorption profile with isosbestic
points at 332 and 353 nm (Figure S6 in the Supporting
Information). The final species showed similar vibronic-
structured absorptions at 340, 359, and 377 nm, which are
red-shifted by 8−13 nm in comparison to 7. 1H NMR
spectroscopic analysis revealed that the chloride analogue of 7
was formed in the photolysis reaction.
To shed light on the reason for such reactivity, both 1a and
1b were subjected to photolysis in deoxygenated CH2Cl2. No
significant spectral change was observed in the UV−visible
absorption spectrum of 1a within 10 min, but a gradual red shift
was observed in the absorption of 1b. Figure 12 shows the
UV−visible absorption spectral change of 1b upon irradiation.
The vibronic structured absorption band of 1b at 352−387 nm
gradually shifted to 369−406 nm with isosbestic points at 349
and 376 nm. The absorption spectrum recorded at 90 min was
Excited State Properties and Photochemistry. The
excited state properties of 1a, 5, and 7 were examined by
nanosecond time-resolved absorption difference spectroscopy.
Figure 11a shows the excited state absorption difference
spectrum of 1a in deoxygenated THF. The spectrum features a
broad absorption from 325 to 600 nm with maxima at 340 and
E
DOI: 10.1021/acs.inorgchem.7b00180
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 12. UV−vis absorption spectral change of 1b recorded at 5 ×
10−5 M at 298 K in degassed CH2Cl2 upon light irradiation at 365 nm.
Arrows indicate the change in absorbance upon irradiation.
and the baseline level increased. After irradiation for 12 h, a
small amount of dark purple precipitate was observed at the
bottom of the cuvette, which suggests the decomposition of 1b
into aggregated gold nanoparticles.
In addition to photoinduced electron transfer, the capacity of
1a to undergo energy transfer was examined with sensitization
of triplet−triplet annihilation (TTA) as an example.2j,5 9,10-
Diphenylanthracene (DPA) was chosen as the triplet energy
acceptor because the TTA reaction between two 3DPA*
1
molecules produces DPA*, which decays promptly to give an
intense, blue fluorescence. Upon 355 nm laser pulse excitation,
a mixture of 1a (5 × 10−5 M) and DPA (5 × 10−5 M) in
degassed CH2Cl2 solution gave a strong delayed fluorescence
signal in addition to the vibronic-structured phosphorescence
of 1a (Figure 13a). Time-resolved emission measurements
Figure 13. Time-resolved emission spectra ((a) 0.1−50 μs, (b) 0.1−
60 μs with 10 μs interval) of a mixture of 1a (5 × 10−5 M) and DPA (5
× 10−5 M) in degassed CH2Cl2 upon light irradiation at 355 nm.
Figure 11. (a) Time-resolved absorption difference spectrum of 1a in
deoxygenated THF and (b) that in the presence of 0.05 M DIPEA.
Transient difference absorption spectra of (c) 5 and (d) 7 in
deoxygenated CH2Cl2.
(Figure 13b) revealed that while both delayed fluorescence of
DPA and phosphorescence of 1a were observed in the first 10
μs, only a delayed fluorescence of DPA was observed after 10
μs. This suggests that the triplet state of 1a was completely
quenched by DPA to form 3DPA* within 10 μs. The as-formed
3DPA* molecule then underwent a TTA reaction with another
molecule of DPA* to give one molecule of DPA in the
ground state and one molecule of 1DPA in the excited state, the
latter of which rapidly returned to the ground state by radiative
decay, resulting in the observed delayed fluorescence of DPA.
found to be identical with that of [Au(C∧N∧C)Cl], which has
2,6-diphenylpyridine as the cyclometalating ligand. The
solution also changed from emissive (green) to nonemissive.
When the photolysis was conducted in toluene, the absorption
spectrum of the solution after 1 h showed that the absorption
of 1b decreased without formation of a new absorption band
3
1
F
DOI: 10.1021/acs.inorgchem.7b00180
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 3. Energy Gap (ΔEST/cm−1) and the Associated Spin−
Orbit Coupling Matrix Element Squared (HSOC2/cm−2)
between the Lowest Singlet Excited State (S1) and the nth
Triplet Excited State (Tn) for 1a, 6, and 7 Evaluated at Their
Respective Optimized Ground State Geometries
The photochemical reactivity of 1a was also demonstrated by
using the oxidative cyanation of a tertiary amine as an
example.2f Initially, when a mixture of N-(4-chlorophenyl)-
1,2,3,4-tetrahydroisoquinoline (1a; 1 mol %), NaCN, and
HOAc in a THF/MeOH mixture (v/v 1/1) was subjected to
light irradiation (λ >350 nm) under aerobic conditions, the α-
cyanated tertiary amine was obtained with 86% conversion and
87% yield in 3 h (Table 2, entry 1). When the amount of 1a
a
2
complex
Tn
ΔEST (cm−1
)
HSOC (cm−2
)
1a
T1
T2
T1
T2
T1
T2
T3
T4
T5
T6
5895
5389
2.91 × 103
∼0
6
7
5333
1.66 × 103
3.64
Table 2. Conversion and Yield of Oxidative Cyanation of
Tertiary Amines by 1a
5272
6449
8.38 × 102
7.81
3.28 × 103
4.66 × 102
7.36 × 102
5.24 × 103
5535
−602
−1250
−1439
−1712
a
entry
amt of 1a (mol %)
R
conversion (%)
yield (%)
a
1
2
3
4
5
6
1
Cl
86
100
100
100
99
87
86
80
87
84
80
ΔEST = E(S1) − E(Tn); n = 1, 2, 3, ...
1.5
1.5
1.5
1.5
1.5
Cl
Br
large SOC (HSOC2 ≈ 103 cm−2) with the S1 excited state for 7:
namely, T3 and T6 (Table 3). These two triplet excited states
are at similar energies at their respective excited state minima
and are only ∼200 cm−1 above the S1 excited state minimum. It
is thus expected that S1 → T3 and S1 → T6 (maintaining the
state ordering at the FC geometry) are both probable pathways
for ISC in 7.
It should be noted that, except in the T6 excited state of 7,
the optimized geometries of the T1 excited state for 1a and 6
and the T3 excited state of 7 remain planar (Figure 14). The
H
Me
OMe
100
a
Conditions: 1a, amine (0.1 mmol), NaCN (0.2 mmol), HOAc (0.15
mmol), THF (0.8 mL), MeOH (0.8 mL), O2 bubbling, light (300 W
xenon lamp, λ >350 nm), reaction time 3 h. After the reaction, the
1
solvent was evaporated and the residue was subjected to H NMR
1
analysis. Conversions were determined by crude H NMR analysis
using 5,5′-dimethyl-2,2′-bipyridine as an internal standard, and yields
were calculated on the basis of the starting amount of substrate.
was increased to 1.5 mol %, the desired products could be
obtained with excellent conversions (99−100%) and yields
(80−87%, Table 2, entries 2−6), which are comparable to
those obtained by using a cyclometalated iridium(III) complex
as the photocatalyst.6
Computational Study. All of the complexes investigated in
this work have similar radiative and nonradiative decay rate
constants (kr ≈ (2.2−2.9) × 103 s−1 and knr ≈ (1.1−2.9) × 104
s−1) except for 7, which has a much higher nonradiative decay
rate of 1.40 × 105 s−1. This higher rate of 7 is contradictory to
the energy gap law, which predicts that knr should decrease with
an increase in the emission energy. DFT/TDDFT calculations
have thus been performed on 1a, 6, and 7 to help decipher why
the introduction of an alkoxy group at the pyridyl ring of the
C∧N∧C ligand results in an almost 10-fold increase of the
nonradiative decay rate constant.
Figure 14. Schematic energy level diagrams of 6 (left) and 7 (right)
and the optimized geometries of the relevant triplet excited states. It
should be noted that the optimized T6 excited state of 7 exhibits
significant deviation from planarity and may contribute to the much
higher knr value in 7.
Table 3 presents the energy gap (ΔEST) between the lowest
singlet excited state S1 and the nth triplet excited states (Tn; n =
1, 2, 3, ...) and the corresponding spin−orbit coupling matrix
elements (HSOC). For complexes 1a and 6, there are two triplet
excited states below their S1 excited state at their respective
optimized ground state geometries, of which there is significant
SOC between the S1 and T1 excited states (HSOC2 values are on
the order of 103 cm−2, comparable to the energy gap between
the S1 and T1 excited states; Table 3). Thus, intersystem
crossing (ISC) from the S1 to the T1 excited state is efficient for
both complexes. On the other hand, for 7, although there are
also two triplet excited states below its S1 excited state, the SOC
between S1 and T1 (or T2) is more than 1 order of magnitude
smaller than the corresponding S1−T1 (or S1−T2) energy gap
and, hence, ISC from S1 to T1 is not a facile path for 7.
Nevertheless, there are two other triplet excited states that have
optimized T6 geometry of 7, on the contrary, deviates from
planarity, meaning that ISC from S1 to T6 will lead to a large
structural distortion. As both the optimized T3 and T6 excited
states of 7 are of similar energies, ISC from S1 may proceed
through either the T3 or the T6 excited state and, if the latter
state is populated, the large structural change may contribute to
the much higher nonradiative decay rate of 7.
The difference in SOC between the S1 and T1 excited states
among 1a and 6 (both have large SOC) and 7 (little SOC) is
explained as follows. Table 4 gives the contributions of one-
electron excitation configuration state functions (CSFs) to the
T1 excited states of 1a, 6, and 7. The MOs for 1a, 6, and 7 are
presented in Figures S24−S26 in the Supporting Information,
and selected MOs for 1a and 7 are shown in Figure 15.
G
DOI: 10.1021/acs.inorgchem.7b00180
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
To examine whether low-lying LMCT or dd excited states
contribute to the much higher nonradiative decay rate of 7, we
have also carried out TDDFT optimization of the LMCT/dd
excited states. The LMCT excited states are calculated to be
1.88, 1.77, and 1.75 eV higher lying than the respective T1
Table 4. Contributions of One-Electron Excitation
Configuration State Functions (CSF) to the T1 Excited State
of 1a, 6, and 7 at Their Respective Optimized S0
a
3
Geometries
major
minor
excited states of 1a, 6, and 7. Given the similar energy gap
3
3
1a
6
H-3 → L (17)
H-8 → L+9 (2), H-6 → L (2), H-4 → L (2), H-4
→ L+2 (3)
between the emitting IL and deactivating LMCT excited
3
states for 6 and 7, it is envisioned that LMCT is not likely to
H → L+1 (61)
H-2 → L (15)
H-2 → L+1 (6), H-2 → L+6 (2)
be the deactivating excited state leading to the much higher knr
observed for 7.
H-7 → L+9 (2), H-6 → L (2), H-4 → L (4), H-2
→ L+5 (2)
H → L+1 (61)
H-2 → L (16)
H → L (10)
H-1 → L+4 (2), H → L (5)
DISCUSSION
7
H-7 → L+9 (2), H-4 → L (4), H-2 → L+2 (2)
H-1 → L+4 (3), H-1 → L+4 (2), H → L (5)
■
The development of phosphorescent gold(III) complexes has
been gaining momentum in recent years. Much effort has been
directed to exploring new ligand systems and applications in
OLEDs and photocatalysis.2f,7−9 As summarized in a review
article by Wenger and co-workers,1b luminescent gold(III)
complexes are often constructed by the combined use of a
cyclometalating bidentate or tridentate ligand and an auxiliary
ligand(s) such as arylacetylide. Although arylacetylide was
found to provide a strong ligand field and provoke
phosphorescence at room temperature, our recent work
shows that the high-lying HOMO of arylacetylide may have a
detrimental effect on the emissive excited state by providing a
nonradiative 3LLCT decay channel (3[π(CCAr)−π*-
(C∧N∧C)]).2j,3 This problem is exacerbated in complexes
supported by C-deprotonated 2,6-diphenylpyridine because the
HOMO of the ligand is lower lying than that of arylacetylide by
0.64 eV. Therefore, we hypothesized that replacing the
arylacetylide with a simple alkyl would lead to an increase in
both emission quantum yield and lifetime for the complexes
supported by the aforementioned cyclometalating ligand.
Cyclometalating [C∧N∧C] ligands were obtained by
H → L+1 (57)
a
The values in parentheses indicate the percent contribution of each
CSF.
Krohnke pyridine synthesis or Suzuki coupling, depending on
̈
the availability of starting materials. Addition of peripheral
Figure 15. Selected frontier MO surfaces for 1a and 7 at the optimized
S0 geometry.
t
substituents (e.g., F, Bu, OEt) on the ligand is a strategy to
tune the triplet emission energy of the complexes. With the
exception of the fluorene-containing ligand, all of these ligands
formed complexes with mercury(II) and precipitated in
ethanol, allowing easy separation from unreacted ligands. The
solid obtained by filtration had to be redissolved in CH2Cl2 and
filtered to remove any other salts. The [Au(C∧N∧C)Cl]
complexes obtained by refluxing the as-formed mercury(II)
complexes with KAuCl4 in MeCN appeared pale yellow or
white. The chloride ligand was then substituted by trifluor-
oacetate because the trifluoroacetate analogue was reported to
give the corresponding alkyl complex with a shorter reaction
time.2h Insertion of alkyl ligand(s) into cyclometalated
gold(III) complexes can be achieved by a transmetalation
reaction with tetraalkyltin(IV),10 alkyl Grignard reagent,
alkyllithium,2h and trialkylaluminum.2i Taking both conven-
ience of preparation and safety into consideration, we chose to
use the corresponding commercially available alkyllithium
reagents for preparing complexes with methyl and butyl
groups, whereas an octadecyl Grignard reagent was freshly
prepared for the synthesis of 1c. Similar to the report by Tilset,
a darkening of the solution arising from formation of metallic
gold was observed when the reaction mixture was warmed to
room temperature.2h These gold(III) alkyl complexes could be
purified by chromatography on SiO2.
The S1 excited state for all three complexes is dominantly
derived from a HOMO → LUMO transition with the HOMO
being derived mainly from π(C∧N∧C) with a small % of
Au(dyz) character (<5%). For effective SOC with the S1 excited
state, the coupling triplet excited state should have the occupied
MOs derived from Au(d orbital) of a different orientation. For
both 1a and 6, there is a small contribution (2%) of the H-6 →
L transition to their T1 excited states and H-6 is composed of a
Au(dxz) orbital, i.e., a d orbital with an orientation different
from the HOMO (Au(dyz)). In addition, this H-6 has a
considerable amount of Au(dxz) character (Au(dxz): 35% in 1a
and 25% in 6), with the remainder mainly localized on the
pyridyl ring (Figures S24 and S25 in the Supporting
Information). Thus, though the contribution of this CSF is
small, the large percentage of Au(dxz) character in H-6 makes
this CSF the dominant contributor to a large SOC (>50%). On
the other hand, for 7, H-4 is of Au(dxz) orientation but there is
only a small percentage of it in H-4 (2%) and so the SOC
between S1 and T1 in 7 is only half that of 6 (Table 3).
Therefore, the introduction of the electron-donating group at
the pyridyl ring of the C∧N∧C ligand decreases the Au(dxz)
character and ISC has to be mediated by higher-lying triplet
excited states in 7, where these triplet excited states are prone
to distort.
1
In this study, most of the H NMR signals of the methyl
ligand range from 1.21 to 1.40 ppm, which are similar to those
found in cyclometalated gold(III) complexes with the methyl
H
DOI: 10.1021/acs.inorgchem.7b00180
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
ligand trans to the pyridine (1.33−1.62 ppm).2h,i The only
exception was found in the 1H NMR spectrum of 5, where the
1H resonance signal of the methyl group appears as a triplet and
is rather downfield at 1.86 ppm in comparison to others (1.21−
1.40 ppm). One possible reason is the existence of an
intramolecular H−F interaction between the proton of the
methyl group and the F atom on the C∧N∧C ligand, which
withdraws electron density from the proton, leading to a
deshielding effect. However, attempts to confirm this by
obtaining an X-ray crystal structure of 5 were unsuccessful.
Photophysical properties of luminescent gold(III) complexes
have been extensively studied in recent years.1 It is generally
conceived that the HOMO and LUMO of cyclometalated
gold(III) complexes mainly localize on phenyl (C) and pyridyl
(N) moieties, respectively. As a result, both the absorption and
emission energies of a gold(III) complex can be rationally
tuned via the utilization of suitable substituents on different
parts of the C∧N∧C ligand. With an eye toward achieving high-
energy (e.g., blue) emission, several complexes in this work
were prepared with the [C∧N∧C] ligands having a tert-butyl or
ethoxy group at the para position of the pyridyl ring and
fluorine atom(s) at different positions of the phenyl rings.
Comparison of the absorption of 1a with that of 2 reveals that
the latter shows blue-shifted absorption peaks at higher
energies (5−6 nm), a likely result with the expectation that
the LUMO residing on the pyridyl ring would be destabilized
by the additional tert-butyl group. In order to lower the energy
of the HOMO, fluorine atoms were added to the phenyl rings.
For example, substitution with fluorine atoms at the meta
positions of phenyl rings proved effective in further blue-
shifting the absorption of 2 at 383 nm to that of 5 at 376 nm.
Complex 7, which bears an ethoxy group on pyridine and two
fluorine atoms on each phenyl ring, shows the largest blue shift
in absorption energy with the last absorption peak at 364 nm.
In addition to varying substituents on the cyclometalating
ligand, since the pyridyl ring is trans to the auxiliary ligand,
auxiliary ligands having varying field strengths are expected to
destabilize the LUMO to different extents, thus affecting the
energy of the HOMO−LUMO gap. When 1a is compared to
the analogue with phenylacetylide (both auxiliary ligands are
monoanionic carbon donor ligands), it was found that the
vibronic absorption bands at 350−390 nm of 1a are blue-
shifted by about 12 nm.2c Such a large increase in energy
illustrates the much stronger field strength provided by alkyl
ligands, in comparison to acetylide, and this underscores the
importance of considering the orbital contribution (sp3 vs sp)
of the auxiliary ligand with the same donor atom.
The dramatic improvement in luminescence of the 3IL
excited state of the cyclometalating ligand upon changing the
auxiliary ligand from phenylacetylide to methyl deserves further
discussion. It should be stressed that both acetylide and alkyl
are strong σ-carbon donor ligands which effectively enlarge the
d−d field splitting to such a large extent that nonradiative decay
via a low-lying d−d excited state is not favorable. The main
reason for the low emission quantum yield of acetylide
counterparts is due to the presence of a π orbital localized on
the arylacetylide ligand. This π orbital, according to DFT
calculations in our previous work, is 0.64 eV higher in energy
than the π-orbital of the C∧N∧C ligand.2j Thus, the charge
transfer excited state from the π-orbital of arylacetylide to the
π*-orbital of the C∧N∧C ligand (LLCT) plays a significant role
during relaxation from the excited state. This facile relaxation
pathway involves bond lengthening leading to severe excited
state structural distortion and a subsequent dissipation of the
3
emissive IL excited state. Rotation of the aryl group of the
acetylide ligand along the pyridyl−Au−acetylide axis is not
sufficient to cause low luminescence efficiency because
gold(III) arylacetylide complexes supported by fluorene
functionalized C∧N∧C ligands also show high emission
quantum yields of 0.5−0.6.2j,8b The success of this class of
fluorene-functionalized gold(III) complexes lies in the use of
fluorene-based C∧N∧C ligands so that the increased con-
jugation at the fluorene destabilizes the HOMO of the
cyclometalating ligand and minimizes the energy difference
with the π-orbital of the phenylacetylide. Thus, in addition to
looking at the number of strong σ-donor ligands, the
compatibility of these ligands from an orbital-energy point of
view, namely the relative order of ligand frontier MOs, must be
taken into consideration.
Owing to the strong destabilization of the LUMO offered by
the alkyl ligand, the emission of 1a is at a higher energy (465
nm) than that of the acetylide (476 nm) and NHC (479 nm)
analogues.2c,d The emission energy of this class of complex can
also be tuned in a way similar to the shifts in absorption energy
observed in complexes with different substituents. With the
addition of a tert-butyl group and a fluoride at specific positions
of the C∧N∧C ligand, the emission of 4 is blue-shifted by 3 nm
relative to 1a. For 6 and 7, the presence of fluorine atoms at the
2,4-positions of the phenyl ring seems more effective in
augmenting the emission energy. With an ethoxy group on
pyridine, 7 exhibits the highest energy emission with the first
emission peak maximum at 443 nm, whereas 6 has the first
maximum at 455 nm. However, the relative intensity of the
second emission peak maximum (472 nm) is about 30% higher
than that of the first (443 nm) in 7, suggesting that the
structural difference between the excited state and ground state
is quite severe; this may account for the low emission quantum
yield (0.02) of 7 among this series of complexes. The use of
butyl instead of methyl as the auxiliary ligand is not likely to be
the reason for the drop in emission quantum yield because the
butyl-containing 1b has an emission quantum yield similar to
that of 1a (1a, 0.11; 1b, 0.09). The relatively high intensity of
the second emission maximum of 7 also renders the emission
color sky blue (CIE coordinates = (0.17, 0.24)) instead of pure
blue.
Replacing arylacetylide with alkyl indeed boosted the
emission quantum yield and lifetime of these [Au(C∧N∧C)]
complexes. Comparison of 1a with the phenylacetylide
analogue revealed a drastic increase in emission quantum
yield and lifetime by more than 100-fold.2c There are two
possible reasons accounting for this: (i) the alkyl ligand pushes
the dσ* orbital of gold(III) to such a high energy that thermal
population of electrons from π* of the cyclometalating ligand
3
to dσ* of the gold(III), i.e., changing from emissive IL to
3
nonemissive LMCT excited state, is impossible and (ii) the
alkyl ligand does not possess any π orbital, thus excluding any
Of particular interest is the observation that both 5 and 7
show an additional low-energy emission band in the solid state.
A similar phenomenon has been observed for the [Au-
(C∧N∧C)CCR] complexes; the low-energy emission was
assigned to dimeric or excimeric emission arising from π···π
3
possible low-lying LLCT excited state that may be present in
analogous systems with acetylide. When these nonradiative
pathways are shut down, the long-lived nature of the 3IL excited
state (up to 77 μs) can be measured at room temperature.
I
DOI: 10.1021/acs.inorgchem.7b00180
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
stacking of the [C∧N∧C] ligand.2c The emission kinetics of 5 at
room temperature revealed a grow-in period of ∼70 ns before
reaching the maximum emission intensity after excitation,
which corroborates the finding of the kinetics of an excimeric
emission of a platinum(II) complex at 700 nm.11 Thus, we
assign the high-energy and low-energy emissions of 5 as 3IL and
excimeric emissions, respectively. At 77 K, the excimeric
emission of 5 is red-shifted by 10 nm and the grow-in time
lengthened from ∼70 to ∼800 ns, presumably due to a slower
excimer formation rate in the low-temperature environment.
which showed absorption peak maxima at 330, 440, and 560
nm.7
In the literature, examples of energy transfer reactions
demonstrated for gold(III) complexes include photoinduced
sensitization of singlet oxygen for the oxidation of amines and
TTA for energy upconversion.2f,j In this study, while 1a was
found to sensitize the delayed fluorescence of DPA, this process
cannot be regarded as energy upconversion because the
fluorescence energy of DPA (λmax 410−430 nm) is lower
than that of the light photon (355 nm) required to excite 1a.
For gold(III) complexes in general, the lack of absorption of
light beyond 450 nm is in part due to the absence of low-lying
1MLCT (metal-to-ligand charge transfer) transitions which are
responsible for the absorption in the blue to green spectral
region in platinum(II) and iridium(III) complexes. This
characteristic arises from the electrophilicity of gold(III),
which stabilizes the occupied 5dπ orbital to such an extent
that the latter is lower lying than the highest occupied dπ
orbital of the cyclometalating ligand, rendering the 1IL
transition the lowest lying absorption transition in most
1
The upfield shifts observed in the H NMR spectra of 5 and 7
upon increasing concentration in CD2Cl2 at room temperature
also suggested their tendency to undergo aggregation.
Electrochemical measurements offered additional insight into
how the LUMO energy was affected by the substituents.
Addition of an electron-donating group such as tert-butyl at the
para position of the pyridine induced a cathodic shift of 0.16 V
(1a vs 2), supporting the assignment that the LUMO is
localized in the pyridine moiety. Although the addition of
fluorine atoms on the phenyl ring was aimed at lowering the
HOMO energy, it was found that substitution with fluorine at
the 3- and 5-positions of the phenyl ring induces an anodic shift
of the reduction wave from −2.32 V (2) to −2.07 V (5),
presumably because the electron-deficient phenyl rings with-
draw electron density from the pyridine by an inductive effect.
In contrast, substitution with fluorine at the 2- and 4-positions
of the phenyl ring had only a minor effect on the reduction
wave (1, −2.16 V; 6, −2.10 V). Thus, the substitution pattern is
also an important consideration when it comes to the tuning of
photophysical and electrochemical properties.
1
cases. Attempts to narrow the IL transition by increasing the
conjugation of the cyclometalating ligand with the use of two
fluorene moieties have been made, but the absorption of the
resultant complex only shows vibronic structured absorption
peak maxima at 423 and 446 nm with a weak absorption tail
ending at 470−480 nm.8b Thus, the development of non-
porphyrin-type phosphorescent gold(III) complexes showing
strong absorption beyond 450 nm remains a challenge.
Studies on the application of gold(III) complexes in light-
induced reactions are relatively rare in comparison to those of
iridium(III), ruthenium(II), and platinum(II) complexes,12
presumably due to the limited number of gold(III) complexes
displaying decent emission quantum yields at room temper-
ature. As the photochemical reactivities of gold(III) complexes
with NHC, acetylide, or chloride as auxiliary ligands have been
explored by us and other groups,2a,f,7,9 we are interested in
whether these gold(III) alkyl complexes possess similar
properties. In this study, photoinduced oxidative cyanation of
tertiary amines was used as an example to illustrate the
photochemical reactivity of gold(III) alkyl complexes with 1a as
a representative. It was found that 1a was capable of catalyzing
the formation of α-cyanated tertiary amines under light
irradiation with excellent conversions and yields at 1.5 mol %
loading. Although the amount of catalyst required is more than
that needed in our previous study with [Au(C∧N∧C)NHC]+ as
catalyst,2f this finding reveals that gold(III) alkyl complexes
could have sufficient stability to act as catalysts for light-induced
reactions, and this also suggests that other metal alkyl
complexes may possess rich photochemical properties which
are not yet discovered because of a general expectation of their
poor stability.
The nanosecond time-resolved absorption difference spec-
trum of 1a shows absorption peak maxima at 340 and 430 nm
which decay with a lifetime (27 μs) close to the emission
3
lifetime, suggesting that the absorption originates from the IL
excited state of 1a. This is also supported by the spectrum’s
high similarity to the nanosecond time-resolved spectral
characteristics of the NHC analogue,7 which showed
absorption maxima at 340 and 440 nm. When diisopropylethyl-
amine (DIPEA) was present, the absorption profile showed
absorptions at 330, 425, and 600 nm, which decayed with a
much longer duration of 400 μs. Since the thermodynamic
driving force of the reaction of *1a with DIPEA is +0.27 V
(E(*1a/1a−) = +0.62 V, E(DIPEA+/0) = +0.35 V, vs Ag/
AgNO3), the observed long-lived species is assigned as one-
electron-reduced 1a produced via reductive quenching of *1a
by DIPEA (Figure 16). The absorption profile of this long-lived
species is also highly similar to that observed for the NHC
analogue of 1a with 0.1 M dibenzylamine in degassed DMF,
CONCLUSION
■
In summary, this study demonstrates the use of the alkyl ligand
as an excellent alternative to acetylide, NHC, and aryl ligands in
pincer-type gold(III) complexes. With the advantages of
absence of a low-lying π-orbital in the alkyl ligands and their
strong σ-donating ability, population of both the LLCT and the
highly structurally distorted, deactivating LMCT excited state
can be prevented. When these nonemissive decay pathways are
kept at bay, these complexes display emissive and long-lived 3IL
excited states with emission quantum yields and lifetimes of up
to 0.40 and 180 μs, respectively, in solution at room
Figure 16. Schematic diagram of the reductive quenching and energy
transfer processes of 1a in this study.
J
DOI: 10.1021/acs.inorgchem.7b00180
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
operating at 3.0 kV. SEM samples were prepared by drop-casting
suspensions onto silicon wafers and air-dried. TEM and SAED were
performed on a Philips Tecnai G2 20 S-TWIN transmission electron
microscope with an accelerating voltage of 200 kV. The TEM images
and SAED patterns were taken by a Gatan Model 794 MultiScan
Camera. TEM samples were prepared by depositing a few drops of
suspensions on the Formvar-coated copper grids, and the excess
solvent was removed by a piece of filter paper.
Crystal Structure Determination. X-ray diffraction data were
collected on a Bruker X8 Proteum diffractometer. The diffraction
images were interpreted and the diffraction intensities were integrated
by using the program SAINT. The crystal structure was solved by
direct methods employing the SHELXS-97 program and refined by
full-matrix least squares using the program SHELXL-97.16
Computational Methods. In this work, the hybrid density
functional PBE017 was employed for all calculations using the program
package G09.18 The 6-31G* basis set19 was used for all atoms except
Au, which was described by the Stuttgart relativistic pseudopotential
and its accompanying basis set (ECP60MWB).20 Solvent effects were
also included by means of the polarizable continuum model (PCM),21
and default parameters were used for the solvent, dichloromethane.
No symmetry constraints were applied in geometry optimizations. For
the singlet ground state (S0), the restricted density functional theory
(RDFT) formalism was employed while the triplet excited states (T1
and 3LMCT/3dd) were optimized using time-dependent density
functional theory (TDDFT).22 Frequency calculations were performed
on the optimized structures to ensure that they are minimum energy
structures by the absence of imaginary frequency (i.e., NImag = 0).
Stability calculations were also performed for all the optimized
structures to ensure that all the wave functions obtained were stable.
Vertical excitation energies of the singlet and triplet excited states were
calculated at the optimized ground state geometries on the basis of
TDDFT using the linear response approximation (LR-PCM). Direct
spin−orbit coupling matrix elements (HSOC) between the singlet and
triplet excited states were evaluated at the optimized ground state
geometries, and the detailed procedures were reported in our previous
work.3,23
temperature. This work highlights that, in addition to
considering the donor strength of the auxiliary ligand, the
nature of its contributing frontier molecular orbitals is an
important yet largely overlooked factor that may have a decisive
influence on the luminescent properties.
EXPERIMENTAL SECTION
Materials and Reagents. All chemicals, unless otherwise noted,
were purchased from commercial sources. All solvents for reaction and
photophysical studies were of HPLC grade. The ligands (except 2,6-
■
diphenylpyridine) were prepared by methods developed by Krohnke
̈
from the respective acetophenones and benzaldehydes or by Suzuki
coupling, and the gold(III) complexes were synthesized according to
literature methods.2b,13,14 4-Ethoxy-2,6-dibromopyridine (starting
material for ligand of 7) was synthesized according to a reported
procedure.15
Physical Measurements and Instrumentation. Nuclear mag-
netic resonance spectra were recorded on DPX-300, -400, and -600
Bruker FT-NMR spectrometer with chemical shifts (in ppm) relative
to tetramethylsilane (for CDCl3) or nondeuterated solvent residual
signal. Unless otherwise stated, all NMR spectra were recorded at
room temperature. Mass spectra (FAB and EI) were recorded on a
Finnigan MAT 95 mass spectrometer. Elemental analyses were
performed at the Institute of Chemistry of the Chinese Academy of
Sciences, Beijing, People’s Republic of China. All absorption spectra
were recorded on a Hewlett-Packard 8453 diode array spectropho-
tometer. Steady-state emission spectra were recorded on a SPEX
Fluorolog-3 spectrophotometer. Solutions for photophysical studies
were degassed by using a high-vacuum line in a two-compartment cell
with five freeze−pump−thaw cycles. Low-temperature (77 K)
emission spectra for glassy-state and solid-state samples were recorded
in quartz tubes (4 mm internal diameter) placed in a liquid nitrogen
Dewar flask with quartz windows. The emission quantum yield was
measured with 9,10-bis(phenylethynyl)anthracene (Φ = 0.85) in
benzene as reference and calculated by Φs = Φr(Br/Bs)(ns/nr)2(Ds/Dr),
in which the subscripts s and r refer to sample and reference standard
solutions, respectively, n is the refractive index of the solvents, D is the
integrated emission intensity, and Φ is the luminescence quantum
yield. The excitation intensity B is calculated by B = 1 − 10−AL, where
A is the absorbance at the excitation wavelength and L is the optical
path length (λ = 1 cm in all cases). The refractive indices of the
solvents at room temperature were taken from standard sources.
Emission quantum yields of thin film samples were measured with a
Hamamatsu C11347 Quantaurus-QY Absolute PL quantum yield
measurement system. The thin films were prepared by drop-casting
from a chlorobenzene solution containing the gold(III) complex with
PMMA. The solvent was evaporated at 60 °C, and translucent films
were obtained. The emission lifetime measurements were performed
on a Quanta Ray GCR 150-10 pulsed Nd:YAG laser system. Errors for
λ values ( 1 nm), τ ( 10%), and Φ ( 10%) are estimated.
Nanosecond time-resolved emission measurements were performed
on a LP920-KS Laser Flash Photolysis Spectrometer (Edinburgh
Instruments Ltd., Livingston, U.K.). The excitation source was the 355
nm output (third harmonic) of a Nd:YAG laser (Spectra-Physics
Quanta-Ray Lab-130 Pulsed Nd:YAG laser). The signals were
processed by a PC plug-in controller with L900 software. The
preparation of samples for the measurements was the same as that for
steady-state emission measurements. Cyclic voltammetric measure-
ments were performed on a Princeton Applied Research electro-
chemical analyzer (Model 273 A Potentiostat/Galvanostat) with a
General Procedure for the Synthesis of Gold(III) Alkyl
Complexes. For all complexes except 1c, an alkyllithium solution
(1.5 equiv) was added dropwise to an anhydrous THF solution/
suspension of the corresponding [Au(C∧N∧C)(OCOCF3)] (∼100
mg) at −78 °C and stirred for 1 h. The reaction mixture was warmed
to room temperature and stirred for another 1 h. After removal of
solvent in vacuo, the product was purified by column chromatography
on SiO2 using CH2Cl2 as eluent.
1
1a: yield 58 mg (71%). MS (+FAB) m/z: 441.9 [M+]. H NMR
(400 MHz, CD2Cl2): δ 7.82 (t, 1H, J = 7.7 Hz), 7.64 (d, 4H, J = 8.0
Hz), 7.52 (d, 2H, J = 8.0 Hz), 7.36 (t, 2H, J = 6.5 Hz), 7.23 (t, 2H, J =
8.1 Hz), 1.27 (s, 3H). 13C NMR (100.6 MHz, CD2Cl2): δ 166.8,
162.4, 151.3, 141.3, 133.1, 130.6, 126.1, 125.3, 116.6, 1.8. Anal. Calcd
for C18H14AuN: C, 48.99; H, 3.20; N, 3.17. Found: C, 49.22; H, 3.19;
N, 3.12.
1
1b: yield 48 mg (54%). MS (+FAB) m/z: 484.0 [M+]. H NMR
(300 MHz, D6-DMSO): δ 8.08 (t, 1H, J = 7.9 Hz), 7.92 (m, 4H), 7.61
(d, 2H, J = 7.2 Hz), 7.41 (t, 2H, J = 7.3 Hz), 7.27 (t, 2H, J = 7.5 Hz),
1.90 (t, 2H, J = 6.7 Hz), 1.80 (quint, 2H, J = 7.3 Hz), 1.53 (sextet, 2H,
J = 7.3 Hz), 0.92 (t, 3H, J = 7.3 Hz). 13C NMR (151 MHz, CD2Cl2): δ
168.2, 162.7, 151.3, 141.8, 133.9, 131.2, 126.7, 125.8, 117.1, 33.9, 27.7,
26.7, 14.6. Anal. Calcd for C21H20AuN: C, 52.18; H, 4.17; N, 2.90.
Found: C, 51.89; H, 4.10; N, 2.88.
1c: a freshly prepared octadecyl Grignard reagent was used instead
of alkyllithium solution; yield 95 mg (75%). MS (+FAB) m/z: 680.1
[M+]. 1H NMR (400 MHz, D6-DMSO, 50 °C): δ 8.06 (t, 1H, J = 8.4
Hz), 7.89 (m, 4H), 7.63 (d, 2H, J = 7.2 Hz), 7.40 (t, 2H, J = 7.3 Hz)
7.27 (t, 2H, J = 7.6 Hz), 1.94 (t, 2H, J = 6.8 Hz), 1.84 (quint, 2H, J =
7.1 Hz), 1.48 (quint, 2H, J = 6.8 Hz), 1.35 (m, 2H), 1.23 (m, 26H),
0.85 (t, 3H, J = 7.0 Hz). 13C NMR (151 MHz, CD2Cl2): 168.2, 162.8,
151.3, 141.8, 133.9, 131.2, 126.7, 125.8, 117.1, 34.7, 32.5, 31.5, 30.4,
30.3, 30.2, 29.9, 23.3. Anal. Calcd for C35H48AuN: C, 61.84; H, 7.12;
N, 2.06. Found: C, 62.10; H, 7.22; N, 1.95.
n
conventional three-compartment cell. A solution of 0.1 M Bu4NPF6
in CH2Cl2 was used as a supporting electrolyte for the electrochemical
measurements. All solutions used in electrochemical measurements
were deaerated by bubbling with argon, and the measurements were
done at room temperature. Ag/AgNO3 (0.1 M in MeCN), a glassy-
carbon electrode, and a platinum wire were used as reference
electrode, working electrode, and counter electrode, respectively;
ferrocene was used as an internal reference. SEM images were taken
on a Hitachi S-4800 field emission scanning electron microscope
K
DOI: 10.1021/acs.inorgchem.7b00180
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
2: yield 72 mg (57%). MS (+FAB) m/z: 497.9 [M+]. 1H NMR (400
MHz, CD2Cl2): δ 7.71 (d, 2H, J = 7.6 Hz), 7.66 (d, 2H, J = 7.2 Hz),
7.57 (s, 2H), 7.36 (t, 2H, J = 7.3 Hz), 7.25 (t, 2H, J = 7.6 Hz), 1.43 (s,
9H), 1.26 (s, 3H). 13C NMR (151 MHz, CD2Cl2): δ 167.4, 167.0,
162.5, 151.6, 133.7, 131.0, 126.6, 125.6, 114.5, 36.5, 30.8, 2.0. Anal.
Calcd for C22H22AuN: C, 53.13; H, 4.46; N, 2.82. Found: C, 53.44; H,
4.61; N, 2.80.
AUTHOR INFORMATION
Corresponding Author
*E-mail for C.-M.C.: cmche@hku.hk.
ORCID
Wai-Pong To: 0000-0002-6820-243X
Notes
The authors declare no competing financial interest.
■
3: yield 43 mg (52%). MS (+FAB) m/z: 478.0 [M+]. 1H NMR (400
MHz, CD2Cl2): δ 7.82 (t, 1H, J = 8.0 Hz), 7.65 (m, 2H), 7.43 (d, 2H, J
= 8.0 Hz), 7.32 (d, 2H, J = 8.0 Hz), 6.89 (t, 2H, J = 8.6 Hz), 1.21 (s,
3H); 19F NMR (376 MHz, CD2Cl2): δ = −109.4. 13C NMR (151
MHz, CD2Cl2): δ 169.8 (d, J = 3.4 Hz), 165.6, 163.9, 161.7, 146.9 (d, J
= 1.8 Hz), 127.8 (d, J = 8.1 Hz), 120.2 (d, J = 18.1 Hz), 116.9, 113.4
(d, J = 23.0 Hz), 2.8. Anal. Calcd for C18H12AuF2N: C, 45.30; H, 2.53;
N, 2.93. Found: C, 45.20; H, 2.77; N, 3.05.
ACKNOWLEDGMENTS
■
This work was funded by the National Key Basic Research
Program of China (No. 2013CB834802), the Hong Kong
Research Grants Council (HKU 17330416), the University
Grants Committee Areas of Excellence Scheme (AoE/P-03/
08), and Guangdong Special Project of the Introduction of
Innovative R&D Teams and GuangDong Aglaia Optoelectronic
Materials Co., Ltd. We thank Dr. X. Chang for solving the X-
ray crystal structure of 4.
4: yield 38 mg (45%). MS (+FAB) m/z: 533.9 [M+]. 1H NMR (400
MHz, CD2Cl2): δ 7.73 (d, 2H, J = 8.5 Hz), 7.49 (s, 2H), 7.36 (d, 2H, J
= 8.0 Hz), 6.91 (t, 2H, J = 8.6 Hz), 1.42 (s, 9H), 1.26 (s, 3H). 19F
NMR (376 MHz, CD2Cl2): δ −109.8. 13C NMR (151 MHz, CD2Cl2):
δ 169.7, 165.5, 163.8, 161.2, 147.2 (d, J = 2.6 Hz), 127.5 (d, J = 7.9
Hz), 120.2 (J = 18.0 Hz), 114.3, 113.3 (d, J = 23.0 Hz), 36.5, 30.8, 2.3.
Anal. Calcd for C22H20AuF2N: C, 49.54; H, 3.78; N, 2.63. Found: C,
49.76; H, 4.01; N, 2.36.
REFERENCES
■
(1) (a) Yam, V. W.-W.; Cheng, E. C.-C. Highlights on the recent
advances in gold chemistrya photophysical perspective. Chem. Soc.
Rev. 2008, 37, 1806−1813. (b) Bronner, C.; Wenger, O. S.
Luminescent cyclometalated gold(III) complexes. Dalton Trans.
2011, 40, 12409−12420. (c) Nevado, C.; Kumar, R. Cyclometalated
Au(III) complexes: synthesis, reactivity and physicochemical proper-
ties. Angew. Chem., Int. Ed. 2017, 56, 1994−2015.
5: yield 46 mg (54%). MS (+FAB) m/z: 569.9[M+]. 1H NMR (400
MHz, CDCl3): δ 7.72 (s, 2H), 7.17 (d, 2H, J = 8.9 Hz), 6.73 (t, 2H, J
= 8.7 Hz), 1.86 (s, 3H), 1.44 (s, 9H). 19F NMR (376 MHz, CDCl3): δ
−93.4, −113.4. 13C NMR not recorded due to low solubility. Anal.
Calcd for C22H18AuF4N: C, 46.41; H, 3.19; N, 2.46. Found: C, 46.09;
H, 3.20; N, 2.41.
(2) (a) Chan, C.-W.; Wong, W.-T.; Che, C.-M. Gold(III)
photooxidants, Photophysical, photochemical properties, and crystal
structure of a luminescent cyclometalated gold(III) complexes of 2,9-
diphenyl-1,10-phenanthroline. Inorg. Chem. 1994, 33, 1266−1272.
(b) Wong, K.-H.; Cheung, K.-K.; Chan, M. C.-W.; Che, C.-M.
Application of 2,6-diphenylpyridine as a tridentate [C∧N∧C] dianionic
ligand in organogold(III) chemistry. Structural and spectroscopic
properties of mono- and binuclear transmetalated gold(III) complexes.
Organometallics 1998, 17, 3505−3511. (c) Wong, K. M.-C.; Hung, L.-
L.; Lam, W. H.; Zhu, N.; Yam, V. W.-W. A class of luminescent
cyclometalated alkynylgold(III) complexes: synthesis, characterization,
and electrochemical, photophysical, and computational studies of
[Au(C∧N∧C)(CC-R)] (C∧N∧C = κ3C,N,C bis-cyclometalated 2,6-
diphenylpyridyl). J. Am. Chem. Soc. 2007, 129, 4350−4365. (d) Au, V.
K.-M.; Wong, K. M.-C.; Zhu, N.; Yam, V. W.-W. Luminescent
cyclometalated N-heterocyclic carbene-containing organogold(III)
complexes: synthesis, characterization, electrochemistry, and photo-
physical studies. J. Am. Chem. Soc. 2009, 131, 9076−9085. (e) Garg, J.
A.; Blacque, O.; Fox, T.; Venkatesan, K. Stable and tunable
phosphorescent neutral cyclometalated Au(III) diaryl complexes.
Inorg. Chem. 2010, 49, 11463−11472. (f) To, W.-P.; Tong, G. S.-
M.; Lu, W.; Ma, C.; Liu, J.; Chow, A. L.-F.; Che, C.-M. Luminescent
organogold(III) complexes with long-lived triplet excited states for
light-induced oxidative C−H bond functionalization and hydrogen
6: yield 34 mg (35%). MS (+FAB) m/z: 513.2 [M+]. 1H NMR (400
MHz, D6-DMSO, 60 °C): δ 8.17 (t, 1H, J = 6.5 Hz), 7.95 (d, 2H, J =
6.1 Hz), 7.21 (d, 2H, J = 5.3 Hz), 7.07 (t, 2H, J = 8.5 Hz), 1.24 (s,
3H). 19F NMR (376 MHz, CD2Cl2): δ −105.4, −107.5. 13C NMR not
recorded due to low solubility. Anal. Calcd for C18H10AuF4N: C,
42.12; H, 1.96; N, 2.73. Found: C, 42.00; H, 2.01; N, 2.76.
7: yield 38 mg (37%). MS (+FAB) m/z: 599.9 [M+]. 1H NMR (300
MHz, CD2Cl2): δ 7.20 (s, 2H), 7.11 (d, 2H, J = 7.1 Hz), 6.61 (td, 2H,
J = 10.6, 2.4 Hz), 4.13 (q, 2H, J = 7.0 Hz), 1.72 (m, 4H), 1.47 (m,
5H), 0.94 (t, 3H, J = 7.3 Hz). 19F NMR (376 MHz, CD2Cl2): δ
−107.0, −109.0. 13C NMR (151 MHz, CD2Cl2): δ 170.4 (d, J = 73.0
Hz), 165.5 (d, J = 10.3 Hz), 163.5 (dd, J = 69.8, 10.3 Hz), 161.6 (d, J =
10.8 Hz), 160.1 (d, J = 5.8 Hz), 133.8, 115.8 (d, J = 17.8 Hz), 107.5 (d,
J = 19.0 Hz), 102.6 (t, J = 26.6 Hz), 63.4, 30.2, 28.5, 27.4, 14.7, 14.5.
Anal. Calcd for C23H20AuF4NO: C, 46.09; H, 3.36; N, 2.34. Found: C,
46.40; H, 3.44; N, 2.30.
8: yield 60 mg (57%). MS (+FAB) m/z: 718.0 [M+]. 1H NMR (400
MHz, CD2Cl2): δ 8.06 (s, 1H), 7.83 (m, 4H), 7.72 (m, 4H), 7.56−
7.63 (m, 3H), 7.34−7.42 (m, 4H), 7.26 (t, 1H, J = 7.34 Hz), 2.04 (t,
4H, J = 8.3 Hz), 1.40 (s, 3H), 1.08 (m, 4H), 0.56−0.68 (m, 10H). 13C
NMR (151 MHz, CD2Cl2): δ 167.6, 166.3, 163.3, 163.2, 154.7, 152.0,
151.3, 150.0, 149.3, 144.2, 141.8, 138.9, 133.8, 131.2, 130.5, 129.8,
128.0, 127.9, 127.4, 126.7, 125.8, 124.8, 123.4, 120.7, 120.5, 115.6,
115.1, 55.5, 40.9, 26.6, 23.7, 14.2, 2.5. Anal. Calcd for C39H38AuN: C,
65.27; H, 5.34; N, 1.95. Found: C, 65.33; H, 5.53; N, 2.15.
production. Angew. Chem., Int. Ed. 2012, 51, 2654−2657. (g) Rosç a,
D.-A.; Smith, D. A.; Bochmann, M. Cyclometalated gold(III)
hydroxides as versatile synthons for Au−N, Au−C complexes and
luminescent compounds. Chem. Commun. 2012, 48, 7247−7249.
ASSOCIATED CONTENT
(h) Langseth, E.; Gorbitz, C. H.; Heyn, R. H.; Tilset, M. Versatile
̈
■
methods for preparation of new cyclometalated gold(III) complexes.
S
* Supporting Information
Organometallics 2012, 31, 6567−6571. (i) Smith, D. A.; Rosç a, D.-A.;
Bochmann, M. Selective Au−C cleavage in (C∧N∧C)Au(III) aryl and
alkyl pincer complexes. Organometallics 2012, 31, 5998−6000. (j) To,
W.-P.; Chan, K. T.; Tong, G. S.-M.; Ma, C.; Kwok, W.-M.; Guan, X.;
Low, K.-H.; Che, C.-M. Strongly luminescent gold(III) complexes
with long-lived excited states: high emission quantum yields, energy
up-conversion, and nonlinear optical properties. Angew. Chem., Int. Ed.
2013, 52, 6648−6652. (k) Herbst, A.; Bronner, C.; Dechambenoit, P.;
Wenger, O. S. Gold complexes with tridentate cyclometalating and
NHC ligands: a search for new photoluminescent gold(III)
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.7b00180.
Synthesis details, characterization data, photophysical
properties, and computational details (PDF)
Crystallographic data (CIF)
Crystallographic data (CIF)
L
DOI: 10.1021/acs.inorgchem.7b00180
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
compounds. Organometallics 2013, 32, 1807−1814. (l) Hung, F.-F.;
To, W.-P.; Zhang, J.-J.; Ma, C.; Wong, W.-Y.; Che, C.-M. Water-
soluble luminescent cyclometalated gold(III) complexes with cis-
chelating bis(N-heterocyclic carbene) ligands: synthesis and photo-
physical properties. Chem. - Eur. J. 2014, 20, 8604−8614.
(m) Szentkuti, A.; Bachmann, M.; Garg, J. A.; Blacque, O.;
Venkatesan, K. Monocyclometalated gold(III) monoaryl complexes
− a new class of triplet phosphors with highly tunable and efficient
emission properties. Chem. - Eur. J. 2014, 20, 2585−2596. (n) David,
B.; Monkowius, U.; Rust, J.; Lehmann, C. W.; Hyzak, L.; Mohr, F.
Gold(III) compounds containing a chelating, dicarbanionic ligand
derived from 4,4′-di-tert-butylbiphenyl. Dalton Trans. 2014, 43,
W.-W. Bipolar gold(III) complexes for solution-processable organic
light-emitting devices with a small efficiency roll-off. J. Am. Chem. Soc.
2014, 136, 17861−17868.
(9) (a) Zehnder, T. N.; Blacque, O.; Venkatesan, K. Luminescent
monocyclometalated cationic gold(III) complexes: synthesis, photo-
physical characterization and catalytic investigations. Dalton Trans.
2014, 43, 11959−11972. (b) Xue, Q.; Xie, J.; Jin, H.; Cheng, Y.; Zhu,
C. Highly efficient visible-light-induced aerobic oxidative C−C, C−P
coupling from C−H bonds catalyzed by a gold(III)-complex. Org.
Biomol. Chem. 2013, 11, 1606−1609. (c) Yu, Z.-T.; Liu, X.-L.; Yuan,
Y.-J.; Li, Y.-H.; Chen, G.-H.; Zou, Z.-G. Evaluation of bis-cyclo-
metalated alkynylgold(III) sensitizers for water photoreduction to
hydrogen. Dalton Trans. 2016, 45, 17223−17232.
11059−11066. (o) Rosç a, D.-A.; Fernandez-Cestau, J.; Romanov, A.
S.; Bochmann, M. Synthesis, C−N cleavage and photoluminescence of
gold(III) isocyanide complexes. J. Organomet. Chem. 2015, 792, 117−
122. (p) Browne, A. R.; Deligonul, N.; Anderson, B. L.; Zeller, M.;
Hunter, A. D.; Gray, T. G. Cyclometalated (boroxinato)gold(III)
complexes from arrested transmetalation. Chem. Commun. 2015, 51,
(10) Venugopal, A.; Shaw, A. P.; Tornroos, K. W.; Heyn, R. H.;
̈
Tilset, M. Synthesis of a coordinately labile gold(III) methyl complex.
Organometallics 2011, 30, 3250−3253.
(11) Williams, J. A. G.; Beeby, A.; Davies, E. S.; Weinstein, J. A.;
Wilson, C. An alternative route to highly luminescent platinum(II)
́ ́
15800−15803. (q) Fernandez-Moreira, V.; Camara, J.; Smirnova, E. S.;
̂
̂
complexes: cyclometalation with NCN-coordinating dipyridylbenzene
ligands. Inorg. Chem. 2003, 42, 8609−8611.
Koshevoy, I. O.; Laguna, A.; Tunik, S. P.; Blanco, M. C.; Gimeno, M.
C. Tuning the energy emission from violet to yellow with bidentate
phosphine gold(III) complexes. Organometallics 2016, 35, 1141−1150.
(r) Nilakantan, L.; McMillin, D. R.; Sharp, P. R. Emissive biphenyl
cyclometalated gold(III) diethyl dithiocarbamate complexes. Organo-
metallics 2016, 35, 2339−2347. (s) Currie, L.; Fernandez-Cestau, J.;
Rocchigiani, L.; Bertrand, B.; Lancaster, S. J.; Hughes, D. L.;
Duckworth, H.; Jones, S. T. E.; Credgington, D.; Penford, T. J.;
Bochmann, M. Luminescent gold(III) thiolates: supramolecular
interactions trigger and control switchable photoemissions from
bimolecular excited states. Chem. - Eur. J. 2017, 23, 105−113.
(3) Tong, G. S. M.; Chan, K. T.; Chang, X.; Che, C.-M. Theoretical
studies on the photophysical properties of luminescent pincer
gold(III) arylacetylide complexes: the role of π−conjugation at the
C-deprotonated [C∧N∧C] ligand. Chem. Sci. 2015, 6, 3026−3037.
(4) (a) Lu, W.; Chan, K. T.; Wu, S.-X.; Chen, Y.; Che, C.-M. Quest
for an intermolecular Au(III)•••Au(III) interaction between cyclo-
metalated gold(III) cations. Chem. Sci. 2012, 3, 752−755. (b) Zhang,
J.-J.; Lu, W.; Sun, R. W.-Y.; Che, C.-M. Organogold(III) supra-
molecular polymers for anticancer treatment. Angew. Chem., Int. Ed.
2012, 51, 4882−4886. (c) Au, V. K.-M.; Zhu, N.; Yam, V. W.-W.
Luminescent metallogels of bis-cyclometalated alkynylgold(III)
complexes. Inorg. Chem. 2013, 52, 558−567. (d) Xiao, X.-S.; Kwong,
W.-L.; Guan, X.; Yang, C.; Lu, W.; Che, C.-M. Platinum(II) and
gold(III) allenylidene complexes: phosphorescence, self-assembled
nanostructures and cytotoxicity. Chem. - Eur. J. 2013, 19, 9457−9462.
(e) Yim, K.-C.; Lam, E. S.-H.; Wong, K. M.-C.; Au, V. K.-M.; Ko, C.-
C.; Lam, W. H.; Yam, V. W.-W. Synthesis, characterization, self-
assembly, gelation, morphology and computational studies of
alkynylgold(III) complexes of 2,6-bis(benzimidazol-2′-yl)pyridine
derivatives. Chem. - Eur. J. 2014, 20, 9930−9939.
(12) Prier, C. K.; Rankic, D. A.; MacMillan, D. W. C. Visible light
photoredox catalysis with transition metal complexes: applications in
organic synthesis. Chem. Rev. 2013, 113, 5322−5363.
(13) Krohnke, F. The specific synthesis of pyridines and
̈
oligopyridines. Synthesis 1976, 1976, 1−24.
(14) Wilkinson, A. J.; Puschmann, H.; Howard, J. A. K.; Foster, C. E.;
Williams, J. A. G. Luminescent complexes of iridium(III) containing
̂
̂
NCN-coordinating terdentate ligands. Inorg. Chem. 2006, 45, 8685−
8699.
(15) Liu, Z.; Yao, Y.; Kogiso, M.; Zheng, B.; Deng, L.; Qiu, J. J.;
Dong, S.; Lv, H.; Gallo, J. M.; Li, X.-N.; Song, Y. Inhibition of cancer-
associated mutant isocitrate dehydrogenases: synthesis, structure-
activity relationship, and selective antitumor activity. J. Med. Chem.
2014, 57, 8307−8318.
(16) Sheldrick, G. M. A short history of SHELX. Acta Crystallogr.,
Sect. A: Found. Crystallogr. 2008, 64, 112−122.
(17) Adamo, C.; Barone, V. Toward reliable density functional
methods without adjustable parameters: the PBE0 model. J. Chem.
Phys. 1999, 110, 6158−6169.
(18) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
(5) Cui, X.; Zhao, J.; Mohmood, Z.; Zhang, C. Accessing the long-
lived triplet excited states in transition-metal complexes: molecular
design rationales and applications. Chem. Rec. 2016, 16, 173−188.
(6) Rueping, M.; Zhu, S.; Koenigs, R. M. Visible-light photoredox
catalyzed oxidative Strecker reaction. Chem. Commun. 2011, 47,
12709−12711.
̈
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
Revision D.01; Gaussian, Inc., Wallingford, CT, 2009.
(19) (a) Francl, M. M.; Petro, W. J.; Hehre, W. J.; Binkley, J. S.;
Gordon, M. S.; DeFree, D. J.; Pople, J. A. Self-consistent molecular
orbital methods. XXIII. A polarization-type basis set for second-row
elements. J. Chem. Phys. 1982, 77, 3654−3665. (b) Hariharan, P. C.;
Pople, J. A. The influence of polarization functions on molecular
orbital hydrogenation energies. Theor. Chim. Acta 1973, 28, 213−222.
(20) Andrae, D.; Haeussermann, U.; Dolg, M.; Stoll, H.; Preuss, H.
Energy-adjusted ab initio pseudopotentials for the second and third
row transition elements. Theor. Chim. Acta 1990, 77, 123−141.
(b) Martin, J. M. L.; Sundermann, A. Correlation consistent valence
basis sets for use with the Stuttgart−Dresden−Bonn relativistic
effective core potentials: the atoms Ga−Kr and In−Xe. J. Chem. Phys.
2001, 114, 3408−3420.
(7) Sun, C.-Y.; To, W.-P.; Wang, X.-L.; Chan, K.-T.; Su, Z.-M.; Che,
C.-M. Metal-organic framework composites with luminescent gold-
(III) complexes. Strongly emissive and long-lived excited states in
open air and photo-catalysis. Chem. Sci. 2015, 6, 7105−7111.
(8) (a) Au, V. K.-M.; Wong, K. M.-C.; Tsang, D. P.-K.; Chan, M.-Y.;
Zhu, N.; Yam, V W.-W. High-efficiency green organic light-emitting
devices utilizing phosphorescent bis-cyclometalated alkynylgold(III)
complexes. J. Am. Chem. Soc. 2010, 132, 14273−14278. (b) Cheng, G.;
Chan, K. T.; To, W.-P.; Che, C.-M. Color tunable organic light-
emitting devices with external quantum efficiency over 20% based on
strongly luminescent gold(III) complexes having long-lived emissive
excited states. Adv. Mater. 2014, 26, 2540−2546. (c) Tang, M.-C.;
Tsang, D. P.-K.; Wong, Y.-C.; Chan, M.-Y.; Wong, K. M.-C.; Yam, V.
M
DOI: 10.1021/acs.inorgchem.7b00180
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(21) Cossi, M.; Scalmani, G.; Rega, N.; Barone, V. New
developments in the polarizable continuum model for quantum
mechanical and classical calculations on molecules in solution. J. Chem.
Phys. 2002, 117, 43−54.
(22) Scalmani, G.; Frisch, M. J.; Mennucci, B.; Tomasi, J.; Cammi, R.;
Barone, V. Substitution and chemical environment effects on the
absorption spectrum of indigo. J. Chem. Phys. 2006, 124, 094107.
(23) Tong, G. S. M.; Che, C.-M. Emissive or nonemissive? A
theoretical analysis of the phosphorescence efficiencies of cyclo-
metalated platinum(II) complexes. Chem. - Eur. J. 2009, 15, 7225−
7237.
N
DOI: 10.1021/acs.inorgchem.7b00180
Inorg. Chem. XXXX, XXX, XXX−XXX