Synthesis, structure, and characterization of dinuclear copper(I) halide complexes with P^N ligands featuring exciting photoluminescence properties

Article abstract of DOI:10.1021/ic300979c

A series of highly luminescent dinuclear copper(I) complexes has been synthesized in good yields using a modular ligand system of easily accessible diphenylphosphinopyridine-type P^N ligands. Characterization of these complexes via X-ray crystallographic studies and elemental analysis revealed a dinuclear complex structure with a butterfly-shaped metal-halide core. The complexes feature emission covering the visible spectrum from blue to red together with high quantum yields up to 96%. Density functional theory calculations show that the HOMO consists mainly of orbitals of both the metal core and the bridging halides, while the LUMO resides dominantly on the heterocyclic part of the P^N ligands. Therefore, modification of the heterocyclic moiety of the bridging ligand allows for systematic tuning of the luminescence wavelength. By increasing the aromatic system of the N-heterocycle or through functionalization of the pyridyl moiety, complexes with emission maxima from 481 to 713 nm are obtained. For a representative compound, it is shown that the ambient-temperature emission can be assigned as a thermally activated delayed fluorescence, featuring an attractively short emission decay time of only 6.5 μs at φ_PL = 0.8. It is proposed to apply these compounds for singlet harvesting in OLEDs.

Full text of DOI:10.1021/ic300979c

Article
pubs.acs.org/IC
Synthesis, Structure, and Characterization of Dinuclear Copper(I)
Halide Complexes with P^N Ligands Featuring Exciting
Photoluminescence Properties
Daniel M. Zink,†,‡ Michael Bachle,‡ Thomas Baumann,*,‡ Martin Nieger,§ Michael Kuhn,∥ Cong Wang,∥
̈
̈
Wim Klopper,∥ Uwe Monkowius,*,⊥ Thomas Hofbeck,# Hartmut Yersin,*,# and Stefan Brase*,†
̈
†Institute of Organic Chemistry, Karlsruhe Institute of Technology, KIT Campus South, Fritz-Haber-Weg 6, D-76131 Karlsruhe,
Germany
‡cynora GmbH, Hermann-von-Helmholtz-Platz 1, D-76344 Eggenstein-Leopoldshafen, Germany
§Laboratory of Inorganic Chemistry, University of Helsinki, P.O. Box 55, FIN-00014, Germany
∥Institute of Physical Chemistry, Karlsruhe Institute of Technology, KIT Campus South, Fritz-Haber-Weg 2, D-76131 Karlsruhe,
Germany
⊥Institute of Inorganic Chemistry, Johannes Kepler University Linz, Altenbergerstrasse 69, A-4040 Linz, Austria
#Institute of Physical Chemistry, University of Regensburg, Universitatsstrasse 31, D-93053 Regensburg, Germany
̈
S
* Supporting Information
ABSTRACT: A series of highly luminescent dinuclear copper-
(I) complexes has been synthesized in good yields using a
modular ligand system of easily accessible diphenylphosphino-
pyridine-type P^N ligands. Characterization of these complexes
via X-ray crystallographic studies and elemental analysis
revealed a dinuclear complex structure with a butterfly-shaped
metal-halide core. The complexes feature emission covering the
visible spectrum from blue to red together with high quantum
yields up to 96%. Density functional theory calculations show
that the HOMO consists mainly of orbitals of both the metal
core and the bridging halides, while the LUMO resides
dominantly on the heterocyclic part of the P^N ligands.
Therefore, modification of the heterocyclic moiety of the bridging ligand allows for systematic tuning of the luminescence
wavelength. By increasing the aromatic system of the N-heterocycle or through functionalization of the pyridyl moiety,
complexes with emission maxima from 481 to 713 nm are obtained. For a representative compound, it is shown that the
ambient-temperature emission can be assigned as a thermally activated delayed fluorescence, featuring an attractively short
emission decay time of only 6.5 μs at ϕPL = 0.8. It is proposed to apply these compounds for singlet harvesting in OLEDs.
interest due to their rich structural and photophysical
properties.29−47 After the pioneering work of McMillin, Ford,
and Eisenberg on mostly cationic mononuclear Cu(I)
complexes, it is well known that these compounds frequently
feature metal-to-ligand charge-transfer (MLCT) absorption
bands in the visible part of the spectrum which can be regarded
as a formal oxidation of Cu(I) to Cu(II) in the excited
state.33−47 According to this type of electronic transition,
usually a geometry change occurs from a pseudotetrahedral
coordination with a flattening tendency toward a square-planar
coordination. In addition to this geometrical distortion being a
major quenching pathway of the emission, the complexes are
vulnerable to solvent coordination and exciplex formation,
resulting in additional quenching pathways. This frequently
1. INTRODUCTION
In the past decade, substantial research has been focused on
new energy-saving concepts for lighting and display applica-
tions. Organic light-emitting diodes (OLEDs) are the most
promising devices for these task, since they consume only a
fraction of the energy of conventional, inefficient light bulbs
and can be manufactured on thin, flexible, and lightweight
substrates.1,2 Significant progress has been achieved in both
increasing the efficiency and allowing for color tuning by use of
emitters based on Ir,3−11 Pt,10,12−18 and Os19−22 because with
these both triplet and singlet excitons formed upon charge
recombination in the emissive layer can be utilized.2,23 These
materials, however, are produced from rather expensive metals,
which is one reason why more abundant and cost-efficient
metals with d10 configuration such as Cu(I), Ag(I), or Au(I)
have received more attention recently.24−28 Out of this group
luminescent copper(I) complexes have attracted increasing
Received: May 11, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic300979c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
quaternary carbon. MS (EI) (electron impact mass spectrometry),
FAB (fast atom bombardment mass spectrometry), and HRMS (high-
resolution mass spectrometry) were carried out on a Finnigan MAT 90
(70 eV). Molecular fragments are quoted as the relation between mass
and charge (m/z) and the intensities as a percentage value relative to
the intensity of the base signal (100%). The abbreviation [M+] refers
to the molecular ion. Descriptions without nominated temperature
were done at room temperature (rt), and the following abbreviations
were used: calcd (theoretical value), found (measured value).
Information is given in mass percent. IR (infrared spectroscopy) was
carried out on a FT-IR Bruker IFS 88. IR spectra of solids were
recorded in KBr and as thin films on KBr for oils and liquids. The
leads to low quantum yields for these compounds in
solution.48−51
One way to address the aforementioned problems is the use
of trigonal-planar complexes with rigid ligands which hamper
the Jahn-Teller distortion from a Y- to a T-shaped geometry
upon excitation, yielding complexes with high emission
quantum yields.43 However, with respect to the emission of
the triplet states and the long phosphorescence decay times,
which are related to the relatively weak spin-orbit coupling of
the copper atom,42,52 application in OLEDs is problematic. On
the other hand, low singlet−triplet splitting found for several
Cu(I) compounds allows for emission from both the singlet
and the triplet excited state depending on the temper-
ature.44,46,47 The singlet excited state, which is slightly higher
in energy than the triplet state, can be thermally activated at the
expense of the triplet state, leading to the so-called thermally
activated delayed fluorescence (TADF).44,46,47,53−57 With
respect to electroluminescent applications, this concept is
termed “singlet harvesting”, in contrast to the “triplet
harvesting”, being of importance if Ir(III)- and Pt(II)-based
emitters are used.47 Even though the excited-state lifetime is
much shorter for emission from the singlet state than from the
triplet state, the equilibrium between those two states has to be
taken into account. Nevertheless, the overall emission decay
time is still much shorter than that of the triplet state.44,46,47
Moreover, it has been shown that increased structural rigidity
reduces radiationless deactivation pathways in mononuclear
copper(I) complexes leading to strongly enhanced quantum
yields,44,46,47 which may also be expected for rigid clusters.58
First examples confirm the suitability of such Cu(I) complexes
as emitters in OLEDs,31,43,51,59,60 whereas the processability is
frequently still a challenge and under investigation.61
In this study, we present a series of highly luminescent
dinuclear copper(I) complexes bearing bridging P^N ligands.
The ligands consist of N-heterocycles with a diphenylphosphi-
no group and allow for easy tuning of both the emission color
as well as the solubility of the complexes. Furthermore, the
electronic structures of these compounds are investigated using
ab initio calculations. In particular, the HOMO can be mostly
assigned to the copper(I)-halide core, while the LUMO mainly
resides on the chelating ligand and can therefore be tuned to
achieve different emission colors. For a representative
compound, it will be shown that the ambient-temperature
emission can be assigned as thermally activated delayed
fluorescence. This implies the high potential of these materials
for OLED applications.
deposit of the absorption band was given in wavenumbers ν
̃
in cm−1.
Forms and intensities of the bands were characterized as follows: s =
strong 10−40% transmission, m = medium 40−70% transmission, w =
weak 70−90% transmission, vw = very weak 90−100% transmission,
br = broad
Synthesis of P^N-Type Ligands. Spectroscopic data of ligands
1,63 2,62 6,64 8,65 and 1166 have already been published in the
literature; for spectroscopic data of ligands 3−5, 7, 9, and 12−17 see
Supporting Information.
I. General Procedure for Substitution of Halide Substituents (Cl,
Br) by Reaction with Lithium Diphenylphosphanide Yielding
Compounds 3−11. Lithium (21.6 mmol, 2.16 equiv) was suspended
under a nitrogen atmosphere in dry tetrahydrofuran (50 mL) at 0 °C.
Chlorodiphenylphosphane (11.0 mmol, 1.10 equiv) dissolved in
tetrahydrofuran (15 mL) was added dropwise via cannula, and the
mixture was stirred at room temperature overnight. To the resulting
deep red solution was added the respective aryl chloride, fluoride, or
bromide (10.0 mmol, 1.00 equiv) dropwise via cannula at room
temperature. The mixture was stirred for 8 h at room temperature and,
after quenching with water (2 mL), filtered through a plug of silica gel.
Purification via flash column chromatography yielded the product.
II. General Procedure for Alkylation of 4-Methyl-2-
(diphenylphosphino)pyridine (10) Yielding Compounds 13−17.
Under a nitrogen atmosphere 4-methyl-2-(diphenylphosphino)-
pyridine (10) (4.00 mmol, 1.00 equiv) was dissolved in dry
tetrahydrofuran and cooled to −78 °C. Lithium diisopropylamine
(1.8 M in tetrahydrofuran, 6.00 mmol, 1.50 equiv) was added dropwise
via cannula. The reaction mixture was stirred at −78 °C for 60 min,
afterward at room temperature for 1 h, and then cooled again to −78
°C. The electrophile (6.00 mmol, 1.50 equiv) was added dropwise via
cannula, and the resulting mixture was allowed to warm to room
temperature. After quenching with water (2 mL), the mixture was
filtered through a plug of silica gel. Purification via flash column
chromatography yielded the product.
III. General Procedure for Ortho-Lithiation of Pyridine Derivatives
to Synthesize Compounds 2 and 12. A solution of N,N-
dimethylethanolamine (27.4 mmol, 2.00 equiv) in n-hexane (25 mL)
was cooled to −5 °C, and n-butyllithium (54.9 mmol, 4.00 equiv) was
added dropwise. After stirring at this temperature for 15 min, a
solution of the pyridine derivate (13.7 mmol, 1.00 equiv) in n-hexane
(10 mL) was added dropwise via canula. The resulting solution was
stirred at −5 °C for another 5 min, and hereafter, the reaction mixture
was cooled to −80 °C and a solution of chlorodiphenylphosphane
(34.3 mmol, 2.50 equiv) was added dropwise. The reaction mixture
was allowed to warm to room temperature, and after quenching with
water (2 mL), the reaction mixture was filtered through a plug of silica
gel. Purification via flash column chromatography yielded the product.
General Procedure for the Syntheses of Complexes 1-I−17-I.
The copper(I) halide salt (2.00 equiv) and the ligand (3.00 equiv)
were suspended under nitrogen in dry dichloromethane (10 mL) and
stirred for 12 h at room temperature resulting in a clear solution.
Afterward the complex was purified through precipitation with diethyl
ether, pentane, or cyclohexane (100 mL). The solid was filtered off,
washed with the appropriate solvent, and dried in vacuum.
2. EXPERIMENTAL SECTION
General Procedures. All reactions were performed in oven-dried
glassware under an argon atmosphere. Solvents and chemicals used
were purchased from commercial suppliers. Solvents were dried under
standard conditions. All materials were used without further
purification. Thin-layer chromatography (TLC) was carried out on
silica gel plates (Silica gel 60, F254, Merck) with detection by UV.
Purification was performed with preparative chromatography using
normal-phase silica gel (Silica gel 60, 230−400 mesh, Merck). 1H
NMR, 13C NMR, and 31P NMR were recorded on Bruker AM 400 and
Bruker DRX 250. Chemical shifts are reported as δ values (ppm).
Signal abbreviations include s = singlet, bs = broad singlet, d = doublet,
t = triplet, q = quartet, quin = quintet, m = multiplet, Ar-H = aromatic
proton. Signals of 13C NMR spectra were allocated through DEPT
technology (DEPT = distortionless enhancement by polarization
transfer). The signal abbreviations include C-Ar = aromatic carbon, +
[(2-(Diphenylphosphino)pyridine)3Cu2I2] (1-I). Yield: 591 mg, 0.51
̃
mmol, 80%; yellow powder. IR (ATR): ν = 1570 (w), 1480 (w), 1452
(w), 1433 (w), 1419 (w), 1092 (w), 743 (m), 694 (m), 506 (m), 492
= primary or tertiary carbon, − = secondary carbon, Cquart
=
(m), 421 (w) cm−1. MS (FAB): Due to insolubility of the compound,
B
dx.doi.org/10.1021/ic300979c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
no mass spectra could be measured. Anal. Calcd for C51H42Cu2I2N3P3
(1168.9): C, 52.32; H, 3.62; N, 3.59. Found: C, 52.07; H, 3.57; N,
3.49.
[(2-(Diphenylphosphino)pyridine)3Cu2Br2] (1-Br). Yield: 507 mg,
̃
0.47 mmol, 75%; yellow powder. IR (ATR): ν = 1570 (w), 1480 (w),
1453 (w), 1434 (w), 1420 (w), 1094 (w), 767 (w), 742 (m), 693 (m),
504 (m), 491 (m), 437 (w), 421 (w) cm−1. MS (FAB), m/z (%): 996
(1), 733 (2), 589 (5), 470 (8), 326 (13), 154 (100). Anal. Calcd for
C51H42Cu2Br2N3P3 (1072.9): C, 57.04; H, 3.92; N, 3.78. Found: C,
56.89; H, 3.93; N, 3.90.
[(2-(Diphenylphosphino)pyridine)3Cu2Cl2] (1-Cl). Yield: 453 mg,
0.46 mmol, 73%; yellow powder. IR (ATR): ν = 1569 (w), 1479 (w),
̃
1452 (w), 1434 (w), 1420 (w), 1093 (w), 987 (w), 767 (w), 742 (m),
692 (m), 502 (m), 490 (m), 436 (w), 421 (w) cm−1. MS (FAB), m/z
(%): 787 (2), 687 (32), 589 (51), 426 (45) 326 (100). Anal. Calcd for
C51H42Cu2Cl2N3P3·1/4CH2Cl2 (1059.1): C, 61.00; H, 4.25; N, 4.16.
Found: C, 61.03; H, 4.29; N, 3.93.
[(2-(Diphenylphosphino)-N,N-dimethylpyridine-4-amine)3Cu2I2]
(2-I). Yield: 421 mg, 0.32 mmol, 60%; white powder. IR (ATR): ν =
̃
1588 (w), 1433 (vw), 1376 (vw), 1224 (vw), 1094 (vw), 999 (vw),
984 (vw), 812 (vw), 739 (vw), 691 (w), 525 (w), 501 (w), 467 (w),
425 (vw) cm−1. MS (FAB), m/z (%): 1170 (1), 1056 (6), 865 (35),
675 (34), 559 (42), 369 (100), 307 (31). Anal. Calcd for
C57H57Cu2I2N6P3 (1298.0): C, 52.67; H, 4.42; N, 6.47. Found: C,
52.39; H, 4.44; N, 6.38.
3047 (vw), 1596 (vw), 1482 (vw), 1464 (vw), 1433 (w), 1310 (vw),
1236 (vw), 1157 (vw), 1093 (vw), 1028 (vw), 1028 (vw) 999 (vw),
811 (w), 739 (w), 690 (w), 604 (vw), 505 (m), 457 (w), 428 (w)
cm−1. MS (FAB), m/z (%): 1380 (1), 1253 (9), 921 (43), 729 (11),
586 (42), 460 (7), 396 (28), 154 (100). Anal. Calcd for
C60H48Cu2I2N3P3S3·1/4CH2Cl2 (1378.9): C, 51.60; H, 3.49; N,
3.00; S, 6.86. Found: C, 51.38; H, 3.55; N, 2.75; S, 6.85.
[(2-(Diphenylphosphino)-4-methylpyridine)3Cu2I2] (10-I). Yield:
̃
516 mg, 0.43 mmol, 71%; yellow powder. IR (ATR): ν = 1775
(vw), 1585 (vw), 1432 (w), 1225 (w), 1191 (w), 1161 (w), 1090 (w),
826 (w), 741 (w), 691 (w), 510 (w), 495 (w), 460 (w), 426 (w) cm−1.
MS (FAB), m/z (%): 996 (32), 809 (23), 721 (31), 617 (33), 530
(55), 340 (100). Anal. Calcd for C54H48Cu2I2N3P3 (1210.9): C, 53.48;
H, 3.99; N, 3.46. Found: C, 53.09; H, 3.95; N, 3.39.
[(2-(Diphenylphosphino)-6-methylpyridine)3CuI] (11-I). Yield: 325
mg, 0.32 mmol, 53%; white powder. IR (ATR): ν = 1583 (w), 1557
̃
(vw), 1434 (w), 1094 (w), 810 (vw), 738 (w), 726 (w), 690 (m), 567
(w), 517 (w), 494 (w), 453 (w) cm−1. MS (FAB), m/z (%): 998 (6),
896 (3) [L3Cu+], 809 (36), 617 (100) [L2Cu+], 530 (38), 340 (96)
[LCu+], 294 (50). Anal. Calcd for C54H48CuIN3P3·2/3CH2Cl2
(1210.9): C, 60.85; H, 4.61; N, 3.89. Found: C, 60.88; H, 4.72; N,
3.89.
[(2-(Diphenylphosphino)-3,5-dimethylpyridine)Cu2I2] (12-I).
Yield: 149 mg, 0.12 mmol, 35%; yellow powder. IR (ATR): ν =
̃
1480 (w), 1434 (w), 1389 (w), 1094 (w), 1027 (w), 740 (m), 692
(m), 625 (w), 544 (w), 509 (m), 491 (m), 468 (w) cm−1. MS (FAB),
m/z (%): 1082 (8), 1005 (9), 893 (9), 814 (15), 764 (29), 737 (18),
624 (82), 547 (100), 495 (29), 382 (82), 242 (44). Anal. Calcd for
C57H54Cu2I2N3P3·1/2CH2Cl2 (1253.0): C, 53.23; H, 4.27; N, 3.24.
Found: C, 53.15; H, 4.40; N, 2.88.
[(2-(Diphenylphosphino)-6-fluoropyridine)3Cu2I2] (3-I). Yield: 548
mg, 0.45 mmol, 76%; yellow powder. IR (ATR): ν = 1584 (w), 1567
̃
(w), 1480 (w), 1433 (m), 1253 (w), 1093 (w), 988 (w), 896 (w), 800
(w), 742 (w), 691 (m), 617 (w), 518 (m), 496 (m) cm−1. MS (FAB),
m/z (%): 1006 (1), 816 (3), 723 (1), 625 (10), 534 (3), 344 (10).
Anal. Calcd for C51H39Cu2F3I2N3P3 (1222.9): C, 50.02; H, 3.21; N,
3.43. Found: C, 50.17; H, 3.29; N, 3.42.
[(2-(Diphenylphosphino)-4-propylpyridine)3Cu2I2] (13-I). Yield:
455 mg, 0.35 mmol, 80%; yellow powder. IR (ATR): ν = 1588 (w),
̃
1481 (w), 1434 (w), 1390 (w), 1094 (w), 989 (vw), 825 (vw), 741
(w), 693 (m), 513 (m), 497 (m), 470 (w) cm−1. MS (FAB), m/z (%):
1244 (4), 1055 (8), 865 (26), 750 (7), 673 (22), 558 (55), 368 (100).
Anal. Calcd for C60H60Cu2I2N3P3·1/4CH2Cl2 (1295.1): C, 54.90; H,
4.63; N, 3.19. Found: C, 54.90; H, 4.95; N, 3.01.
[(2-(Diphenylphosphino)-4-isobutylpyridine)3Cu2I2] (14-I). Yield:
̃
452 mg, 0.34 mmol, 65%; yellow powder. IR (ATR): ν = 2953 (w),
1586 (w), 1542 (w), 1480 (w), 1461 (w), 1433 (w), 1384 (w), 1091
(w), 739 (w), 690 (m), 508 (m), 494 (m), 481 (m) cm−1. MS (FAB),
m/z (%): 1082 (44), 893 (17), 763 (38), 701 (25), 572 (54), 382
(100). Anal. Calcd for C63H66Cu2I2N3P3 (1337.1): C, 56.51; H, 4.97;
N, 3.14. Found: C, 56.72; H, 4.96; N, 3.10.
[(2-(Diphenylphosphino)-4,6-dimethylquinoline)3Cu2I2] (4-I).
Yield: 509 mg, 0.36 mmol, 74%; ochre powder. IR (ATR): ν =
̃
1582 (w), 1545 (vw), 1481 (vw), 1434 (w), 1377 (w), 1094 (w), 882
(vw), 823 (w), 742 (w), 692 (m), 534 (w), 508 (w), 451 (w), 420 (w)
cm−1. MS (FAB), m/z (%): 1316 (<1), 1127 (2), 937 (19), 845 (7),
745 (30), 594 (22), 404 (68), 342 (100). Anal. Calcd for
C69H60Cu2I2N3P3 (1403.1): C, 58.98; H, 4.30; N, 2.99. Found: C,
59.52; H, 4.78; N, 2.73.
[(3-(Diphenylphosphino)isoquinoline)3Cu2I2] (5-I). Yield: 303 mg,
0.23 mmol, 72%; yellow powder. IR (ATR): ν = 1569 (vw), 1481
̃
(vw), 1433 (w), 1276 (vw), 1095 (w), 885 (vw), 746 (w), 692 (m),
661 (w), 514 (w), 475 (w), 449 (w) cm−1. MS (FAB), m/z (%): 1193
(1), 1070 (3), 878 (4), 757 (1), 689 (3), 566 (5), 376 (10). Anal.
Calcd for C63H48Cu2I2N3P3·1/2CH2Cl2 (1320.9): C, 55.94; H, 3.62;
N, 3.08. Found: C, 55.51; H, 3.66; N, 2.98.
[(1-(Diphenylphosphino)isoquinoline)3Cu2I2] (6-I). Yield: 156 mg,
̃
0.12 mmol, 28%; red powder. IR (ATR): ν = 1545 (w), 1480 (w),
1434 (w), 1306 (w), 1097 (w), 827 (w), 741 (m), 692 (m), 540 (m),
521 (w), 487 (m), 461 (m) cm−1. MS (FAB), m/z (%): 1068 (8), 879
(4), 757 (16), 689 (32), 566 (19), 376 (100). Anal. Calcd for
C63H48Cu2I2N3P3·1.5CH2Cl2 (1444.9): C, 53.49; H, 3.55; N, 2.90.
Found: C, 53.99; H, 2.95; N, 3.23.
[(2-(Diphenylphosphino)quinoxaline)3Cu2I2] (7-I). Yield: 582 mg,
̃
0.44 mmol, 83%; orange powder. IR (ATR): ν = 1483 (w), 1434 (w),
1365 (vw), 1090 (w), 964 (w), 758 (m), 742 (m), 691 (m), 626 (w),
509 (m), 485 (m), 402 (m) cm−1. MS (FAB), m/z (%): 1263 (1),
1073 (3), 881 (8), 691 (18), 567 (15), 377 (31), 155 (100). Anal.
Calcd for C60H45Cu2I2N6P3·CH2Cl2 (1409.9): C, 52.01; H, 3.36; N,
5.97. Found: C, 51.78; H, 3.42; N, 5.84.
[(2-(Diphenylphosphino)benzo[d]thiazole)3Cu2I2] (8-I). Yield: 598
mg, 0.45 mmol, 86%; yellow powder. IR (ATR): ν = 1450 (w), 1432
̃
(w), 1311 (w), 1092 (w), 985 (w), 849 (vw), 758 (m), 739 (m), 690
(m), 601 (w), 580 (w), 506 (m), 487 (m) cm−1. MS (FAB), m/z (%):
1211 (4), 1082 (4), 892 (15), 763 (4), 701 (5), 572 (21), 382 (18).
Anal. Calcd for C57H42Cu2I2N3P3S3 (1336.8): C, 51.13; H, 3.16; N,
3.14; S, 7.18. Found: C, 51.19; H, 3.34; N, 3.05; S, 6.57.
[(2-(Diphenylphosphino)-4-heptylpyridine)3Cu2I2] (15-I). Yield:
343 mg, 0.23 mmol, 51%; yellow powder. IR (ATR): ν = 2919
̃
(vw), 2850 (vw), 1586 (vw), 1432 (w), 1389 (vw), 1095 (vw), 989
(vw), 739 (w), 692 (w), 513 (w), 497 (w), 460 (w) cm−1. MS (FAB),
m/z (%): 1355 (2), 1166 (11), 976 (17), 805 (17), 785 (44), 613
(59), 424 (100), 362 (16). Anal. Calcd for C72H84Cu2I2N3P3 (1463.2):
C, 59.02; H, 5.78; N, 2.87. Found: C, 58.91; H, 5.64; N, 2.85.
[(2-(Diphenylphosphino)-4-cyclopentylmethylpyridine)3Cu2I2]
̃
(16-I). Yield: 387 mg, 0.27 mmol, 57%; yellow powder. IR (ATR): ν =
2948 (w), 1588 (w), 1434 (w), 1094 (w), 753 (w), 740 (w), 695 (w),
518 (w), 496 (m), 518 (m), 496 (m), 429 (w) cm−1. MS (FAB), m/z
(%): 1134 (15), 943 (14), 787 (33), 753 (25), 598 (62), 408 (100).
Anal. Calcd for C69H72Cu2I2N3P3 (1415.1): C, 58.48; H, 5.04; N, 2.93.
Found: C, 58.05; H, 5.04; N, 2.93.
[(2-(diphenylphosphino)-4-phenethylpyridine)3Cu2I2] (17-I).
Yield: 334 mg, 225 mmol, 62%; yellow powder. IR (ATR): ν =
̃
3025 (vw), 1587 (w), 1546 (vw), 1495 (vw), 1480 (vw), 1453 (vw),
1434 (w), 1387 (vw), 1094 (vw), 741 (w), 692 (m), 500 (w), 466 (w)
cm−1. MS (FAB), m/z (%): 1270 (2), 1180 (8), 990 (8), 812 (12),
798 (24), 620 (45), 430 (100), 339 (41). Anal. Calcd for
C75H66Cu2I2N3P3·1/2CH2Cl2 (1481.1): C, 59.44; H, 4.43; N, 2.75.
Found: C, 59.29; H, 4.48; N, 2.38.
X-ray Crystallography. Crystal structure determinations of 1-I, 2-
I, 3-I, 5-I, 8-I, 9-I, 10-I, 11-I, 11-I-MTBE, 14-I, and 15-I were
performed on a Bruker-Nonius Kappa-CCD diffractometer at 123(2)
K and for 1-Cl on a SMART X2S diffractometer at 200(2) K using Mo
[(2-(Diphenylphosphino)-6-methylbenzo[d]thiazole)3Cu2I2] (9-I).
Yield: 530 mg, 0.38 mmol, 61%; yellow powder. IR (ATR): ν =
̃
C
dx.doi.org/10.1021/ic300979c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Kα radiation (λ = 0.71073 Å). Crystal data, data collection parameters,
and results of the analyses are listed in Table S1, Supporting
Information. Direct methods (for 15-I Patterson methods) were used
for structure solution (SHELXS-97),67 and refinement was carried out
using SHELXL-97 (full-matrix least-squares on F2).67 Hydrogen atoms
were refined using a riding model. A semiempirical absorption
correction using equivalent reflections was applied. Single crystals
suitable for X-ray diffraction analysis were obtained by slow diffusion
of diethyl ether into a solution of the complexes in dichloromethane.
In the cases of sparingly soluble complexes crystals were grown by
diffusion of diethyl ether into the filtered reaction mixture.
Chart 1. Type I P^N Ligands Featuring Different Electronic
Characteristics
Photophysical Measurements. Steady-state emission spectra
were recorded on a FluoroMax-4 spectrofluorometer by Horiba
Scientific equipped with a 150 W xenon-arc lamp, excitation and
emission monochromators (1200 grooves/nm blazed at 330
(excitation) and 500 nm (emission)), and a Hamamatsu R928
photomultiplier tube. Emission and excitation spectra were corrected
for source intensity (lamp and grating) by standard correction curves.
Luminescence quantum yields were measured with a Hamamatsu
Photonics absolute PL quantum yield measurement system (C9920-
02G) equipped with a L9799-01 CW Xenon light source (150 W),
monochromator, C7473 photonic multichannel analyzer, integrating
sphere, and employing U6039-05 PLQY measurement software
(Hamamatsu Photonics, Ltd., Shizuoka, Japan). All solvents used
were of spectrometric grade. Degassed samples were prepared by
purging with argon for 30 min. UV−vis absorption spectra were
measured on a Thermo Scientific Evolution 201 UV−vis spectropho-
tometer.
Computational Methods. Density functional theory (DFT) using
the BP8668,69 and B3LYP70,71 functionals was applied to complexes 2-
I, 5-I, 6-I, and 10-I. The def2-SV(P) basis set72,73 was used. The
resolution-of-identity approximation74−76 was employed in the BP86
calculations. The m4 grid was used for numerical integration.
Analytical harmonic vibrational frequency calculations were conducted
to verify that the optimized ground-state structures are minima on the
potential energy surface. For the excited states, time-dependent
density functional theory (TD-DFT) and the spin-flip Tamm-Dancoff
approximation (SF-TDA)77 approach were applied. All calculations
were performed with the Turbomole program package version 6.3.78
Chart 2. Type II P^N Ligands Featuring Different Alkyl
Substituents for Better Solubility and Processability
3. RESULTS AND DISCUSSION
3.1. Synthesis. Copper(I) halide complexes exhibit many
coordination motifs depending on the coordinating ligand as
well as the molar ratios between ligand and metal salt.42,79 With
pyridine-type ligands L and an equimolar stoichiometry the
complexes frequently form [CuXL]4 cubane-like structures or
one-dimensional ladder chains [CuXL]∞ consisting of fused
Cu2X2 subunits.42,52,80 Changing the stoichiometry to a ratio of
1:1:2 or even 1:1:3 results in either a [(CuX)2L4] form of
complex with an isolated rhombohedron Cu2X2 core or a
mononuclear complex [CuIL3].42,81,82 Another dimeric struc-
ture with a rhombic Cu2X2 core featuring the composition
[Cu2X2L2(R3P)2] can be obtained using a mixed ligand system
with bulky ligands like triphenylphosphane blocking further
aggregation.83−86 Nevertheless, due to their di-, oligo-, or
polymeric structure, once precipitated from solution, these
complexes are usually only sparingly soluble in organic solvents.
Our approach involves phosphane-substituted pyridine
heterocycles, combining phosphine and imine functionalities
in a single ligand to achieve a straightforward synthetic pathway
for preparation of copper(I) halide cluster complexes. For that,
N-heterocycles with a diphenylphosphino group in the ortho
position to the coordinating imine nitrogen atom were used.
The ligands synthesized and examined on their complexation
behavior in this work can be divided into two groups (shown in
Charts 1 and 2). The first group (type I) includes an
unsubstituted 2-diphenylphosphinopyridine ligand 1 as refer-
ence ligand which is then modified at the heterocycle moiety
with electron donor or acceptor groups, such as dimethylamino
or fluorine, in order to study the electronic effects.
Furthermore, by synthesizing several quinoline-, isoquinoline-,
or quinoxaline-diphenylphosphanes, the conjugated pyridine
system has been extended at different positions. Additionally,
the pyridine moiety was also replaced with a benzothiazole unit
D
dx.doi.org/10.1021/ic300979c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 1. Synthesis of P^N-Type Ligands 1−17
in order to compare it with the pyridine derivatives, and its
processability was modified as in the examples below.
by mixing the P^N ligands with the respective copper(I) halide
in a 3:2 ratio in dichloromethane at room temperature
following a literature procedure (Scheme 2).89 The initially
The second group (type II) only involves alkyl substituents
on the pyridine ring mostly in the 4 position (Chart 2). These
are used to adjust the solubility and therefore the processability
of these new materials as well as to investigate steric influences.
Various alkyl substituents with different chain lengths or
branching were introduced to fine tune the solubility of the
copper complexes in nonpolar solvents like toluene, xylene, or
chlorobenzene which are expected to be the solvents of choice
used to process these new materials.
Scheme 2. General Procedure for Synthesis of Dinuclear
Copper(I) Complexes
Unsubstituted 2-diphenylphosphinopyridine (1) has been
prepared via a substitution reaction from 2-chloropyridine and
a diphenylphosphino-nucleophile (Scheme 1). During this
reaction, the nucleophile NaPPh2 is generated in situ from
triphenylphosphane and sodium in liquid ammonia.64,87,88
A
similar procedure is applied for ligands 3−11 starting from the
appropriate precursor halides using LiPPh2 as nucleophile in
tetrahydrofuran at 0 °C. Alkyl-substituted ligands 13−17 are
accessible in a further reaction sequence, in which the methyl
group of 2-(diphenylphosphino)-4-methylpyridine (10) can be
deprotonated by lithium diisopropylamide and subsequently
reacted with different alkyl halides. In a third reaction pathway,
substituted 2-diphenylphosphinopyridine ligands can be
synthesized via direct C2 lithiation of the corresponding
substituted pyridine precursor, followed by addition of
chlorodiphenylphosphane as electrophile, which was used for
syntheses of ligands 2 and 12 (Scheme 1).62 All P^N ligands
were obtained in moderate to good yields and fully
characterized by standard chemical analysis (NMR, MS, IR).
It should be noted that synthesis of the ligands as well as
synthesis of the complexes (vide infra) does not show any
yield-lowering effects when performed in multigram scale.
For the complexation reactions, mostly copper(I) iodide was
used as metal salt due to its better oxidation stability compared
to copper(I) bromide and copper(I) chloride. However, in
terms of investigating the influence of the ligand field strength
of different halide anions, copper(I) iodide, copper(I) bromide,
and copper(I) chloride were used to obtain the corresponding
complexes of ligand 1, yielding complexes 1-I, 1-Br, and 1-Cl,
respectively. The syntheses of the complexes were carried out
cloudy reaction mixture became clear within hours, indicating
the progress of the reaction. Pure materials were obtained upon
precipitation with diethyl ether from the filtered reaction
mixture and subsequent washing with several organic solvents.
The usually high yields dropped with increasing solubility of the
complexes. Crystals suitable for X-ray analysis were obtained by
slow diffusion of diethyl ether into the reaction mixture as well
as into a saturated dichloromethane solution of the appropriate
complex. All complexes have been characterized by mass
spectroscopy, elemental analysis,90 and in many cases single-
crystal X-ray diffraction, which confirmed a 3:2 ligand to copper
stoichiometry for all compounds except for complex 11-I,
which has a 3:1 stoichiometry.
3.2. Structural Studies. The molecular structures of the
compounds obtained by X-ray diffraction revealed a butterfly-
shaped Cu2I2 core surrounded by three ligands. One of the
three P^N ligands acts as bridging unit concerning the two
copper atoms. The other two ligands coordinate the copper
ions only via their phosphorus atom, resulting in a tetrahedral
coordination for each metal center. The crystal structure of
compound 1-I is shown in Figure 1. The tetrahedral
coordination is typical for mononuclear Cu(I) complexes as
E
dx.doi.org/10.1021/ic300979c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
unprecedented for copper(I), whereas the silver congener
Ag2Cl2(pyPPh2)2 was already published in 1982.93 In general, a
coordination geometry with a folded {Cu2(μ-X)2} moiety is
rare and could be found for only a few chelating ligands such as
naphthyridine,94 bis(diphenylphosphino)ethane,95 and
(phenylamino)bis(phosphate).96
Introduction of various substituents, e.g., methyl (10-I),
isobutyl (14-I), n-heptyl (15-I), or cyclopentylmethylene (16-
I), in the 4 position of the pyridine ring does not affect the
complexation behavior of these ligands nor does it remarkably
change the structural characteristics of the obtained complexes
(Figure 2). Also, extension of the aromatic system (e.g.,
quinoline, quinoxaline, quinaxoline) or alteration of the type of
N-heterocycle (e.g., benzothiazole) in the P^N ligands lead to
complexes with the prominent butterfly-shaped copper-halide
̂
motif with one bridging and two monodentate NP ligands.
Notably, substitution only influences the solubility and also the
photophysical properties of the synthesized complexes (vide
infra).
Figure 1. Molecular structure of complex 1-I (hydrogen atoms are
omitted for clarity; displacement parameters are drawn at 50%
probability level).
The copper-copper distance of all measured complexes
(2.69−2.83 Å) is in the order of the sum of the van der Waals
radii or slightly below (2.80 Å)97 and therefore too large for
significant metallophilic interactions. Furthermore, the distance
is basically not affected by changes of the ligands, such as
extension of the π systems (4-I−9-I) or solubility enhancing
alkyl groups (10-I, 12-I−17-I). The bond distances between
the copper and the iodine atoms are nearly identical within one
well as for polynuclear Cu(I) clusters, while the butterfly-
shaped metal-halide core is in contrast to the typical planar
Cu2X2 geometry observed for many halogen-bridged Cu2X2-
type structures.85,91,92 Interestingly, no intramolecular ππ-
stacking interactions between any aromatic rings can be
observed. The structure of the presented complexes is
Figure 2. Molecular structures of complexes 5-I, 9-I, 14-I, and 15-I (hydrogen atoms and minor disordered parts are omitted for clarity;
displacement parameters are drawn at 50% probability level).
F
dx.doi.org/10.1021/ic300979c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
Table 1. Selected Bond Lengths (Angstroms) of Structurally Characterized (P^N)3Cu2X2 Complexes
1-I
1-Cl
2-I
5-I
9-I
Cu(1)−Cu(2)
Cu(1)−X(1)
Cu(1)−X(2)
Cu(2)−X(1)
Cu(2)−X(2)
P(2)−Cu(1)
P(1)−Cu(1)
P(3)−Cu(2)
N(1)−Cu(2)
2.7694(5)
2.6733(7)
2.6803(5)
2.7280(6)
2.6446(5)
2.2514(6)
2.2522(7)
2.2507(7)
2.1035(17)
2.7521(4)
2.6383(4)
2.6834(4)
2.6876(5)
2.6818(4)
2.7714(8) {2.823}
2.7034(8) {2.746}
2.6717(7) {2.776}
2.6140(7) {2.727}
2.8121(7) {2.854}
2.2524(12) {2.3578}
2.2645(13) {2.3573}
2.2443(13) {2.3416}
2.7204(6) {2.751}
2.6930(5) {2.762}
2.6954(5) {2.781}
2.6277(5) {2.727}
2.6802(5) {2.786}
2.2404(9) {2.356}
2.2555(8) {2.356}
2.2468(9) {2.348}
2.829(1)
2.658(1)
2.648(1)
2.706(1)
2.698(1)
2.2486(1)
2.2424(16)
2.2429(16)
b
2.242(1)
c
2.248(1)
d
2.224(1)
e
2.106(3)
2.069 (4) {2.1737}
2.076(2) {2.172}
2.069(5)
e
a
b
c
d
Calculated bond lengths at the B3LYP/def2-SV(P) level are shown in curly brackets. Cu(2)−P(1). Cu(2)−P(2). Cu(1)−P(3). Cu(1)−N(1).
a
Table 2. Selected Bond Angles (degrees) of Structurally Characterized (P^N)3Cu2X2 Complexes
1-I
1-Cl
2-I
5-I
9-I
X(1)−Cu(1)−X(2)
X(1)−Cu(2)−X(2)
Cu(1)−X(1)−Cu(2)
Cu(1)−X(2)−Cu(2)
P(1)−Cu(1)−X(1)
P(1)−Cu(1)−X(2)
P(2)−Cu(1)−P(1)
P(1)−Cu(1)−Cu(2)
107.63(2)
107.07(1)
61.68(1)
62.67(1)
106.31(2)
104.34(2)
119.87(2)
87.07(2)
108.13(1)
106.74(1)
62.34(1)
61.85(1)
106.91(4)
114.12(4)
119.04(3)
87.55(2)
110.82(2) {109.42}
109.26(2) {107.69}
62.80(2) {62.12}
60.65(2) {61.84}
107.18(4) {102.2}
99.07(4) {104.9}
125.39(5) {119.2}
108.86(2) {110.2}
111.32(2) {111.1}
61.48(1) {60.2}
108.76(3)
105.93(3)
63.67(2)
63.90(2)
102.13(5)
104.98(5)
118.55(6)
87.39(5)
60.80(1) {59.2}
b
c
106.37(2) {101.7}
103.37(2) {104.7}
118.16(3) {119.7}
86.96(4) {84.0}
89.46(2) {84.8}
c
a
b
Calculated bond angles at the B3LYP/def2-SV(P) level are shown in curly brackets. P(3)−Cu(1)−Cl(1). P(3)−Cu(1)−Cl(1).
particular complex with values ranging from 2.62 to 2.81 Å.
Analysis of the relevant bond angles reveals an only slightly
distorted tetrahedral geometry around both copper ions,
demonstrating the fitting bite angle of this bidentate P^N
ligand structure. Bond angles are approaching the tetrahedral
standard and reflect the small steric requirements of this ligand
system as well as the small constraints that are imposed on the
complex structure. Interestingly, the Cu2X2 butterfly core
almost exhibits a perfectly equilateral Cu2X triangle with
corresponding bond angles very close to 60°. Selected bond
lengths and angles are given in Tables 1 and 2.
However, the only modification of the ligands which leads to
a completely different coordination behavior is introduction of
a methyl group in the 6 position of the pyridine moiety. This
can be seen in the case of ligand 11, which leads to formation of
the mononuclear Cu(I) complex 11-I (Figure 3). Most
probably due to the steric demand of this methyl group, a
dinuclear structure cannot be formed. In the mononuclear
complex, Cu is coordinated by three ligands solely via their
phosphorus atom and an iodide ligand yielding a neutral
tetrahedral L3CuI-type complex. Its structural parameters are
very similar to the literature-known compound (Ph3P)3CuI.98
The corresponding bond angles closely match the tetrahedral
angle of 109.5°, and the Cu−P distances are within the range of
2.30−2.31 Å, which is slightly longer than the corresponding
distances observed in the dinuclear complexes. For more
detailed information about bond lengths or angles of complex
11-I, see Figure S10 in the Supporting Information.
3.3. Photophysical Characterization at Ambient
Temperature. Figure 4 displays the absorption spectra of
complexes 6-I and 10-I in dichloromethane, along those of the
corresponding free ligands 6 and 10. Complex 6-I shows
Figure 3. Molecular structure of complex 11-I (hydrogen atoms,
solvent molecule, and minor disordered part are omitted for clarity;
displacement parameters are drawn at 50% probability level).
Figure 4. Absorption spectra of complexes 6-I and 10-I as well as their
corresponding ligands 6 and 10 at ambient temperature in
dichloromethane solution.
G
dx.doi.org/10.1021/ic300979c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
maxima around 272 (ε = 3400 M−1 cm−1) and 319 nm (ε =
2000 M−1 cm−1). These are assigned to π−π* transitions within
the phenyl rings on the phosphane moieties as well as to
transitions localized on the isoquinoline moiety, also of π−π*
character. These assignments are confirmed by comparison
with the spectrum of 10-I. Additionally, a weak shoulder
observed for 6-I at 345 nm, which is not present for the free
ligand, indicates an electronic change on one of the
isoquinoline moieties. Such a change correlates with the
structure of the complex with one of the heterocycles being
coordinated. The spectrum of 10-I is rather unstructured with a
broad shoulder around 252 nm, which is assigned to ligand-
centered (LC) π−π* transitions, being in accordance with the
occurrence of a corresponding peak in the free ligand spectrum.
The lack of structure is due to a smaller energetic separation
between transitions localized on the phenyl rings and the
heterocycle, respectively, which is the pyridine versus the
energetically lower-lying extended π system of the isoquinoline.
In contrast to the structure of the free ligand 10, the absorption
of 10-I does not show a steep decrease in the absorption
coefficient around 300 nm, which is most probably due to the
coordinating pyridine moiety of the bridging ligand similar to
the discussion of 6-I. Furthermore, both complexes show tails
which are not visible for the free ligands, ranging from
approximately 375 to 485 nm and from 371 to 416 nm for
complexes 6-I and 10-I, respectively (ε = 262−154 M−1 cm−1).
These low-energy absorptions are assigned to metal-to-ligand
charge-transfer (MLCT) transitions mixed with halogen-to-
ligand charge-transfer (XLCT) transitions and are therefore
assigned to (X+M)LCT states. Electronic absorption maxima
and extinction coefficients of compounds 6, 6-I, 10, and 10-I
are summarized in Table 3.
Table 3. Extinction Coefficients of Compounds 6, 6-I, 10,
and 10-I
Figure 5. HOMO and LUMO frontier orbital plots of complexes 2-I,
5-I, 6-I, and 10-I based on DFT calculations.
a
compound λabs [nm] ε [M−1 cm−1
]
transition
LC, Ph, π−π*
LC, isoquinoline, π−π*
LC, isoquinoline, π−π*
LC, Ph, π−π*
LC, isoquinoline, π−π*
LC, bridging isoquinoline, π−π*
(X+M)LCT
a
Table 4. Calculated Orbital Energies
6
274
996
2-I
5-I
6-I
10-I
289 (sh)
321 (sh)
272
712
526
HOMO
LUMO
−4.40
−1.01
−4.46
−1.83
−4.55
−2.04
−4.46
−1.83
6-I
3400
2000
1078
262
a
319
B3LYP/def2-SV(P) HOMO and LUMO orbital energies (in eV, 1
eV ≙ 8067 cm−1) at the ground-state geometry.
345 (sh)
375 (sh)
256
10
984
Ph, π−π*
LC, Ph, π−π*
(X+M)LCT
be classified as being largely of (X+M)LCT character.
Moreover, it appears that the differences in the observed
emission maxima may be explained by the variations in the
ligand LUMOs. Hence, a fine tuning of the luminescence
properties seems possible by making small chemical mod-
ifications on the ligands.
10-I
252 (sh)
371 (sh)
4470
154
a
In dichloromethane solution. sh = shoulder.
These assignments are further supported by theoretical
calculations of complexes 2-I, 5-I, 6-I, and 10-I using density
functional theory (DFT) with functionals BP8668,69 and
B3LYP.70,71 The frontier orbitals for the investigated
compounds are shown in Figure 5.
All complexes show a similar trend, with the HOMO being
largely localized on the Cu2X2 core with some participation of
the phosphorus atoms of the monocoordinated ligands. On the
other hand, the LUMO is mainly located on the bridging P^N
ligand. Therefore, as shown in Table 4, the modifications on
the N-heterocycle mainly affect the LUMO energy, while the
HOMO energy remains basically unchanged, lying roughly
around −4.50 eV. On the basis of these results, the type of
electronic transition that leads to the lowest excited states may
Room-temperature powder emission spectra of compounds
1-I, 2-I, 4-I−8-I, and 10-I are shown in Figure 6. Emission
spectra of complexes 3-I, 9-I, and 12-I−17-I are omitted for
clarity, since they are similar to the profiles of complexes 1-I, 2-
I, 4-I−8-I, and 10-I and can be found in Figure S13 in the
Supporting Information. For a collection of the photophysical
properties see Table 5.
In accordance with the assumption of a charge-transfer
nature of the emitting state, emission spectra of all investigated
complexes are relatively broad and unstructured. This correlates
well with our calculations, i.e., with the HOMO being on the
Cu2X2 cluster and the LUMO on the bridging ligand.32
Furthermore, the energetic variation of the LUMO, as seen in
H
dx.doi.org/10.1021/ic300979c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
nm). This becomes also apparent for complex 5-I, which has
also an isoquinoline moiety instead of a pyridine moiety but has
the phosphino substituent in the 3 instead of the 1 position,
resulting in exhibition of a smaller red shift (λem = 572 nm) as
compared to 6-I. Therefore, it can be assumed that a phosphino
substituent in the 1 position of an isoquinoline moiety stabilizes
the LUMO more than a substituent in the 3 position. Using
quinoline instead of pyridine also results in a red shift of the
emission as can be seen for compound 4-I (λem = 605 nm). Use
of benzothiazoles as coordinating moieties leads to emission
maxima at 558 and 546 nm for complexes 8-I and 9-I,
respectively.
These findings gained from powder measurements together
with the results of DFT and TD-DFT calculations show that
electron-donating substituents on the pyridine ring, e.g., a
dimethylamino group, destabilize the LUMO, consequently
leading to a hypsochromic shift of the emission. Conversely, an
electron-withdrawing substituent, e.g., a fluorine atom, lowers
the LUMO energy, thus shifting the emission to longer
wavelengths. Consequently, lowering the energy level of the
LUMO by extending the aromatic system also results in a red-
shifted emission.
The emission quantum yields in powder are higher than the
values reported for similar dinuclear complexes, reaching up to
96% for the alkylated compounds.32,61,85 These larger quantum
yields of 14-I to 16-I compared to compound 1-I with 81% can
tentatively be related to a more distinct hindrance due to the
alkyl substituents with respect to geometry changes upon
excitation. Furthermore, the low emission quantum yield of
complex 6-I (16%) or 7-I (3%) is probably related to the
energy gap law, i.e., a decrease in emission quantum efficiency
with decreasing emission energy, since competing nonradiative
processes become more relevant.99
Figure 6. Powder emission spectra of 1-I, 2-I, 4-I−8-I, and 10-I (λexc
=
375 nm, rt). For clarity, emission spectra of complexes 3-I, 9-I, and 12-
I−17-I can be found separately in Figure S13 in the Supporting
Information.
a
Table 5. Photoluminescence Characteristics
compound
λem [nm]
ΦPL
b
1-I
537
0.81
0.36
0.24
0.48
0.59
0.06
0.33
0.16
0.03
0.70
0.94
0.70
0.51
0.74
0.96
0.96
0.96
0.60
b
1-Br
1-Cl
2-I
550
b
569
481
553
605
572
657
713
558
546
523
546
509
497
522
507
3-I
4-I
5-I
6-I
7-I
8-I
9-I
In Figure 7, the calculated LUMO energies of the ligands are
shown as a function of the emission wavelengths of the
10-I
12-I
13-I
14-I
15-I
16-I
17-I
545
a
b
Neat powders at rt, λexc = 375 nm; λexc = 366 nm
the calculated values (Table 4), is reflected in the emission
wavelengths. As discussed on the basis of DFT calculations,
modifications of the ligands result mainly in energy alterations
of the ligand LUMOs and also of the LUMOs of the
corresponding complexes. This characteristic is clearly dis-
played in the emission spectra of compounds 2-I, 10-I, 14-I,
15-I, and 16-I. The corresponding ligands carry electron-
donating groups in the 4 position of the pyridine moiety,
ranging from weak donors, such as alkyl groups of different
length (10-I, 14-I, 15-I, 16-I), to a dimethylamino group as
strong donor for compound 2-I. This results in a moderate blue
shift for 10-I and 15-I of about 15 nm compared to 1-I and to a
sky-blue emission with a band maximum at 481 nm for 2-I.
Consequently, introduction of electron-withdrawing groups
such as fluorine or extension of the conjugation on the bridging
ligand leads to stabilization of the LUMO and hence a red shift
of the emission relative to 1-I which is visible for complexes 3-
I−6-I and 8-I.
Figure 7. Correlation between ligand LUMO energies and emission
maxima of the corresponding complexes (for exact values of the
LUMO energy see Table S2 in the Supporting Information).
complexes. The LUMO energies were obtained at the B3LYP/
def2-SV(P) level for the optimized free ligand. As visible from
the data, there is a linear correlation between those two: a
lower, i.e., a more negative, LUMO energy of the ligand is
related to a longer emission wavelength. Consequently, for the
presented Cu2X2 family, the emission color can be roughly
estimated based on the calculated LUMO energy of the
Depending on the position of the extension of the π system
on the pyridine ring, the red shift can be rather moderate as for
8-I (λem = 558 nm) or large as in the case of 6-I (λem = 657
I
dx.doi.org/10.1021/ic300979c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
bridging ligand, enabling a simple design of differently colored
emitters.
(which were only computed with the B3LYP functional), with
absorption maxima at 452, 496, 598, and 660 nm for 2-I, 10-I,
5-I, and 6-I, respectively. The triplet excitation energies differ
from the singlet excitation energies only by a few 10−2 eV
(Table 6).
In order to get further insight into the electronic transitions
of this kind of dinuclear complexes, TD-DFT calculations were
performed on complexes 2-I, 5-I, 6-I, and 10-I. A comparison
between theoretical and experimental structures for complexes
2-I and 5-I is provided in Tables 1 and 2. As is seen, selected
calculated bond lengths and angles are in reasonable agreement
with X-ray data (vide supra). In Table 6, the HOMO−LUMO
All calculated excitations show a very distinct charge-transfer
(CT) character, and thus, the TD-B3LYP approach is unable to
yield quantitatively correct results. CT excited states cannot be
̈
described by naive application of TD-DFT. Nevertheless, the
computed absorption wavelengths follow the experimental
trend, as indicated by the HOMO−LUMO gaps.
Table 6. HOMO−LUMO Gaps (in eV) Computed at the
BP86/def2-SV(P) and B3LYP/def2-SV(P) Levels for
Complexes 2-I, 5-I, 6-I, and 10-I at Their Ground- and
Excited-State Equilibrium Geometries
In the case of CT states, it is reported that the spin-flip
Tamm-Dancoff approximation (SF-TDA) provides much more
accurate phosphorescence energies than TD-DFT.77 Indeed,
the SF-TDA values in Table 7 (given in bold) agree very well
with the experimentally observed emission maxima. The
computed energies differ only by 5−10% from the
experimentally derived values and follow the observed trend.
The lowest singlet excited state cannot be described by the SF-
TDA approach, but because of the large spatial separation of
the two unpaired electrons, we expect this state to be very close
to the corresponding triplet state (i.e., within 0.01−0.07 eV as
found at the TD-DFT level). Indeed, as can be seen below for
compound 1-I, the singlet-triplet energy separation ΔE(S1−T1)
amounts to 400 cm−1 (0.05 eV).
compound
functional
S0 [eV]
S1 [eV]
T1 [eV]
2-I
BP86
1.72
3.39
1.06
2.63
0.99
2.50
1.52
3.15
0.80
2.26
0.59
1.90
0.51
1.81
0.73
2.16
0.80
2.21
0.59
a
B3LYP
BP86
5-I
B3LYP
BP86
6-I
0.51
1.80
0.73
2.11
B3LYP
BP86
10-I
B3LYP
a
value omitted because of inconsistent data.
In order to examine the influence of halogens on the
photoluminescence, the series using 2-(diphenylphosphino)-
pyridine as ligand (1-I, 1-Br, 1-Cl) was examined. As shown in
Figure 8, the sequence of the emission maxima (1-I 537 nm, 1-
gaps are summarized, computed for the geometries optimized
for the ground state (S0) as well as for the first excited singlet
(S1) and triplet states (T1). Concerning the ground state, the
largest gap is found for complex 2-I, followed by 10-I, 5-I, and
6-I with smaller gaps. According to Table 7, this trend is also
observed for the TD-DFT vertical singlet excitation energies
Table 7. Vertical Absorptions and Emissions Computed at
the TD-B3LYP/def2-SV(P) and SF-TDA/def2-SV(P) Levels
a
for Complexes 2-I, 5-I, 6-I, and 10-I
λabs
HOMO →
LUMO
compound state [nm]
λem [nm]
character
2-I
S1
452
98%
783
99% HOMO →
LUMO
T1
455
97%
807
98% HOMO →
LUMO
b
495
(481)
100% LUMO →
HOMO
5-I
S1
598
600
99%
98%
873
99% HOMO →
LUMO
T1
891
99% HOMO →
LUMO
Figure 8. Emission spectra of neat powder of complexes 1-I, 1-Br, and
1-Cl, λexc = 366 nm, rt.
b
547
(572)
100% LUMO →
HOMO
6-I
S1
660
665
99%
98%
1028
99% HOMO →
Br 550 nm, 1-Cl 569 nm) corresponds to the order of ligand
field strengths of the halides (I− < Br− < Cl−) leading to a
greater splitting of the d orbitals, destabilization of the HOMO,
and thus to a reduction of the energy gap.32 This trend is
contrary to the electronegativity of the halides, which should
result in a stabilization of the d orbitals due to an increased
electron-withdrawing effect. Obviously, the effect of the ligand
strength overcompensates the electronegativity. A similar
behavior was also found for tetranuclear copper clusters
Cu4X4L4 (L = substituted pyridine).100
LUMO
T1
1064
98% HOMO →
LUMO
b
599
100% LUMO →
HOMO
(657)
10-I
S1
496
500
98%
97%
834
99% HOMO →
LUMO
T1
875
98% HOMO →
LUMO
b
512
100% LUMO →
HOMO
(523)
Influences of different environments on the emission
properties have been investigated with the alkylated complexes
10-I, 14-I, 15-I, and 16-I. Ambient-temperature emission
spectra of powders, compounds doped in PMMA, neat thin
a
Experimentally observed emission maxima are given in parentheses.
Spin-flip Tamm-Dancoff approximation (SF-TDA/def2-SV(P))
b
single-point calculation at the TD-DFT-optimized triplet geometry.
J
dx.doi.org/10.1021/ic300979c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
films, and dichloromethane solutions, respectively, are shown in
Figures 9 and 10 for complexes 10-I and 16-I (for complexes
concentration is found (10%, 50% weight percent of the
emitting material: 540−550 nm). The observed red shift is
largest when measured in solution exhibiting maxima in the
range of 590−610 nm (Table 8).
For neat films as well as for different dopant concentrations
in PMMA matrices, the emission maxima almost overlap and
exhibit only slight variations in the emission maxima. The
emission quantum yields drop considerably when going from
neat powders, with ϕPL values above 80%, to degassed
dichloromethane solutions, featuring only about 2% quantum
yields (Table 8). The values of the neat films of the complexes
10-I, 14-I, 15-I, and 16-I lie between 30% and 50% and, thus,
are also significantly lower than those measured for the
corresponding powders.
Obviously, both the emission wavelengths as well as the
photoluminescence quantum yields strongly depend on the
environment of the complexes. By influencing the possible
extent of structural changes upon excitation, i.e., comparing
rigid media like crystals, neat films, and polymer matrices,
drastic effects can be observed. In particular, the probabilities of
radiationless deactivation are reduced, resulting in significantly
higher emission quantum yields for more rigid environments.44
Solutions, in which those rearrangements upon excitation can
occur easily, enhance radiationless deactivation processes
leading to a considerable decrease in emission quantum yields.
A corresponding behavior being also connected with a drastic
decrease of the emission decay time has been described in ref
44. Moreover, the very high emission quantum yields up to
96% in powders may indicate that also quenching processes
such as triplet-triplet annihilations (TTA)102 between adjacent
emitter molecules are negligible.102 This may be explained by
relatively small geometry changes in the lowest excited states
with subsequent energy reductions even in such rigid systems.
For more detailed discussions of these effects see refs 44 and
47.
Figure 9. Photoluminescence spectra of complexes 10-I in different
environments: powder, solution, neat film, PMMA doped with 10%
emitting compound, PMMA doped with 50% emitting compound: λexc
= 375 nm (powder, solution); 330 nm (films), rt.101
3.4. Thermally Activated Delayed Fluorescence and
Singlet Harvesting with Dinuclear Cu(I) Complexes.
Frequently, Cu(I) complexes represent singlet emitters at
ambient temperature, since in these cases a thermally activated
delayed fluorescence (TADF) dominates.44,46,47 In such a
situation, the emitter material is potentially highly attractive for
application in OLEDs by taking advantage of the singlet-
harvesting effect, which is shortly explained: Light generation in
an OLED occurs usually in the emission layer after formation of
one singlet and three triplet excitons, referring to the ensemble
of processes. From these exciton states, fast relaxations lead to
populations of the lower excited singlet S1 and triplet T1 states
of the emitter molecules. In organotransition metal emitters,
spin-orbit coupling is usually large enough to induce fast
intersystem crossing (ISC) from the S1 state to the T1 state.
Thus, for large energy separations between the singlet and the
Figure 10. Photoluminescence spectra of complexes 16-I in different
environments: powder, solution, neat film, PMMA doped with 10%
emitting compound, PMMA doped with 50% emitting compound: λexc
= 375 nm (powder, solution); 330 nm (films), rt.101
14-I and 15-I see Figures S14 and S15 in the Supporting
Information). In Table 8 the corresponding emission maxima
as well as the corresponding quantum yields are summarized
for compounds 10-I, 14-I, 15-I, and 16-I. The emission maxima
for these complexes investigated as neat powders range from
501 to 522 nm, while the maxima are shifted to longer
wavelengths when measured in neat films or doped into PMMA
matrices. For the latter case, also a dependence of the doping
Table 8. Luminescence Properties of Complexes 10-I, 14-I, 15-I, and 16-I in Different Environments
10-I
14-I
15-I
16-I
a
a
a
a
complex
powder
λmax [nm]
ΦPL
λmax [nm]
ΦPL
λmax [nm]
ΦPL
λmax [nm]
ΦPL
515
548
547
543
608
0.83
0.35
0.34
0.28
0.02
514
550
540
542
590
0.96
0.28
0.47
0.40
0.02
522
548
549
550
595
0.96
0.31
0.31
0.37
0.02
501
546
545
546
594
0.96
0.50
0.31
0.34
0.02
neat film
b
b
90% PMMA
50% PMMA
CH2Cl2 solution
a
b
λexc = 375 nm (powder, solution); 330 nm (films), rt. Weight percent.
K
dx.doi.org/10.1021/ic300979c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
triplet states, ΔE(S1−T1), for example amounting to several 103
cm−1, only a triplet emission is observed at ambient
temperature. Applied in OLEDs, this leads to the well-known
triplet-harvesting effect.23 On the other hand, for small energy
separations, for example, of ΔE(S1−T1) being only several 102
cm−1, thermal activation or up-ISC processes will occur at
ambient temperature from the T1 state to the S1 state.
Subsequently and under suitable conditions, almost all light
emission will stem from the singlet S1 state, representing the
singlet-harvesting effect.44,46,47 The corresponding emission is
not a spontaneous fluorescence, since the S1 state population is
fed by the T1 reservoir. Therefore, the resulting emission
exhibits a longer decay time than that of the prompt
fluorescence, representing the thermally activated delayed
fluorescence.
thermally activated additional decay channel via the S1 → S0
emission.
In conclusion, the thermally activated singlet harvesting
renders Cu(I) complexes to be attractive OLED emitters. The
very long emission decay time of the triplet state would not
allow for an application as triplet emitters, since the roll-off of
the OLED efficiency would be too large, for example, due to
significant saturation effects. On the other hand, a decay time of
the delayed fluorescence of τ(DF) = 6.5 μs for the discussed
example is already similarly short as found for Pt(II) complexes
which can be applied with good efficiencies in OLEDs15,18
Moreover, Cu(I) complexes offer the future potential for
designing materials with even smaller ΔE(S1−T1) energy
separations. Possibly, this might lead to materials that would
exhibit even larger radiative decay rates (or shorter emission
decay times) than Ir(ppy)3, exhibiting a value of kr(T1) ≈ 7 ×
105 s−1 (τ(T1) = 1.4 μs).11 In summary, the low-cost copper-
based complexes seem to develop to interesting OLED emitter
materials.
The Cu(I) complex 1-I, as displayed in Figure 1, has been
investigated as a representative example.103 Figure 11
4. CONCLUSION
Highly luminescent dinuclear Cu(I) complexes have been
synthesized employing a modular approach, consisting of a
butterfly-shaped Cu2X2 core and three P^N-type ligands. One
of the ligands acts as a bridge between the two Cu(I) centers,
whereas the other two only coordinate via the phosphorus
atom as revealed by single-crystal X-ray diffraction. The
photoluminescence properties depend distinctly on the
substitution pattern at the pyridine moiety. Neat complex
powders feature photoluminescence maxima from the blue to
the red region of the visible spectrum depending on the
electronic characteristics of the corresponding ligand. It is
shown that an electron-donating substituent on the pyridine
ring, e.g., a dimethylamino group, raises the LUMO to a higher
energy level, consequently leading to a blue shift of the
photoluminescence maximum. Conversely, an electron-with-
drawing substituent stabilizes the LUMO, thus shifting the
emission to longer wavelengths. Similarly, by enlarging the
aromatic system of the ligands an energy reduction of the
LUMO is obtained, which also leads to a red shift of the
photoluminescence maximum. These changes were further
investigated and confirmed by TD-DFT calculations, revealing
that the HOMO energy stays roughly at −4.5 eV while changes
in the LUMO energies are more pronounced.
Moreover, emission of the copper(I) complexes exhibits a
strong dependence on the rigidity of their immediate
environment, as shown by investigating the compounds as
neat powders, neat films, in solution, and doped in PMMA
matrices. This behavior is related to a geometry change that can
occur upon excitation and thus depends on the available space
in a matrix or on the flexibility of the environment.
Most remarkably are the high emission quantum yields of
some of the investigated complexes of up to ϕPL = 96% and the
relatively short emission decay times of the thermally activated
delayed fluorescence of only a few microseconds. Together
with the calculated and experimentally determined small energy
gap between the S1 and T1 states of a few 102 cm−1, these
complexes may qualify as highly attractive candidates for
singlet-harvesting emitters in OLEDs.
Figure 11. Emission spectra, quantum yields, and decay times of the
dinuclear Cu(I) complex 1-I (powder) as displayed in Figure 1. Inset
symbolizes that the emission at ambient temperature represents a
thermally activated delayed fluorescence (TADF), exhibiting a decay
time of τ(DF) = 6.5 μs. kBT stands for the thermal energy, being
available at a given temperature T.
summarizes a series of emission data of the powder material.
At T = 77 K, a broad emission band, peaking at λmax (77 K) =
552 nm, is observed. The emission decay, exhibiting a time
constant of τ(77 K) = 32 μs, is monoexponential and ascribed
to a phosphorescence stemming from the T1 state. With this
decay time and the measured emission quantum yield of
ϕPL(77 K) = 0.9, a radiative decay of kr(T1) = ϕPL/τ(T1) = 2.8
× 104 s−1 is calculated.
With temperature increase to T = 300 K, a blue shift of the
emission maximum of ∼13 nm is observed, which corresponds
to about 4 × 102 cm−1. We assign this ambient-temperature
emission to result from the thermally activated S1 state. Almost
the same activation energy is determined from the temperature
dependence of the emission decay time. It decreases monotoni-
cally over the whole temperature range from 77 to 300 K and
thus excludes a possible misinterpretation according to an
occurrence of a phase transition in that range. The emission
quantum yield at T = 300 K of ϕPL = 0.8 is only slightly lower
than that at 77 K. With the measured decay time of τ(300 K) =
τ(DF) = 6.5 μs, a radiative decay rate of kr(DF) = 1.2 × 105 s−1
is obtained. The growing in of the radiative rate by more than a
factor of 4 with temperature increase is a consequence of the
L
dx.doi.org/10.1021/ic300979c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(18) Kalinowski, J.; Fattori, V.; Cocchi, M.; Williams, J. A. G. Coord.
Chem. Rev. 2011, 255, 2401−2425.
ASSOCIATED CONTENT
■
S
* Supporting Information
(19) Kim, J. H.; Liu, M. S.; Jen, A. K.-Y.; Carlson, B.; Dalton, L. R.;
Shu, C.-F.; Dodda, R. Appl. Phys. Lett. 2003, 83, 776−778.
(20) Lu, J.; Tao, Y.; Chi, Y.; Tung, Y. Synth. Met. 2005, 155, 56−62.
(21) Chou, P.-T.; Chi, Y. Eur. J. Inorg. Chem. 2006, 3319−3332.
(22) Breu, J.; Kratzer, C.; Yersin, H. J. Am. Chem. Soc. 2000, 122,
2548−2555.
X-ray crystallography data for complexes 1-I, 2-I, 3-I, 5-I, 8-I,
9-I, 10-I, 11-I, 14-I, 15-I, 11-I-MTBE, and 1-Cl in CIF format;
spectroscopic data of compounds 3−5, 7, 9, 10, 12−17,
additional photophysical spectra of complexes 3-I, 9-I, and 12-
I−17-I, X-ray crystallographic data for 1-I, 2-I, 3-I, 5-I, 8-I, 9-I,
10-I, 11-I, 14-I, 15-I (CCDC 838846−838855), 11-I-MTBE
(CCDC 878649), and 1-Cl (CCDC 876712) together with
Cartesian coordinates (in Angstroms) of the optimized BP86/
def2-SV(P) and B3LYP/def2-SV(P) ground-state equilibrium
geometries of compounds 2-I, 5-I, 6-I, and 10-I. This material
is available free of charge via the Internet at http://pubs.acs.org.
(23) Yersin, H. Top. Curr. Chem. 2004, 241, 1−26.
(24) Barbieri, A.; Accorsi, G.; Armaroli, N. Chem. Commun. 2008, 19,
2185−2193.
(25) Hsu, C.-W.; Lin, C.-C.; Chung, M.-W.; Chi, Y.; Lee, G.-H.;
Chou, P.-T.; Chang, C.-H.; Chen, P.-Y. J. Am. Chem. Soc. 2011, 133,
12085−12099.
(26) Czerwieniec, R.; Hofbeck, T.; Crespo, O.; Laguna, A. M.;
Gimeno, M. C.; Yersin, H. Inorg. Chem. 2010, 49, 3764−3767.
(27) Barakat, K. A.; Cundari, T. R.; A., M. J. Am. Chem. Soc. 2003,
125, 14228−14229.
(28) Igawa, S.; Hashimoto, M.; Kawata, I.; Hoshino, M.; Osawa, M.
Inorg. Chem. 2012, 51, 5805−5813.
(29) Zink, D. M.; Baumann, T.; Nieger, M.; Barnes, E. C.; Klopper,
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: baumann@cynora.com; uwe.monkowius@jku.at;
hartmut.yersin@chemie.uni-regensburg.de; braese@kit.edu.
Notes
W.; Brase, S. Organometallics 2011, 30, 3275−3283.
̈
(30) Zhang, L.; Li, B.; Su, Z. J. Phys. Chem. C. 2009, 113, 13968−
13973.
The authors declare no competing financial interest.
(31) Che, G.; Su, Z.; Li, W.; Chu, B.; Li, M.; Hu, Z.; Zhang, Z. Appl.
ACKNOWLEDGMENTS
■
Phys. Lett. 2006, 89, 103511/1−103511/3.
We acknowledge financial support from KIT. We also thank the
Deutsche Forschungsgemeinschaft (DFG) for support, through
projects B2 and C1 of SFB/TRR 88 and the Bundesministe-
(32) Tsuboyama, A.; Kuge, K.; Furugori, M.; Okada, S.; Hoshino, M.;
Ueno, K. Inorg. Chem. 2007, 46, 1992−2001.
(33) Manbeck, G. F.; Brennessel, W. W.; Eisenberg, R. Inorg. Chem.
2011, 50, 3431−3441.
rium fur Bildung und Forschung (BMBF). Furthermore, we
̈
(34) McMillin, D. R.; Buckner, M. T.; Ahn, B. T. Inorg. Chem. 1977,
16, 943−945.
also thank Lena-Maria Fleig, Dominik Pfaff (Institute of
Organic Chemistry), and Dr. Mathias Mydlak (cynora
GmbH) for their support.
(35) Buckner, M. T.; McMillin, D. R. J. Chem. Soc., Chem. Commun.
1978, 759−761.
(36) Blaskie, M. W.; McMillin, D. R. Inorg. Chem. 1980, 19, 3519−
3522.
REFERENCES
■
(1) So, F.; Kido, J.; Burrows, P. MRS Bull. 2008, 33, 663−669.
(2) Yersin, H., Ed. Highly Efficient OLEDs with Phosphorescent
Materials; Wiley-VCH: Weinheim, 2008.
(3) He, L.; Qiao, J.; Duan, L.; Dong, G.; Zhang, D.; Wang, L.; Qiu, Y.
Adv. Funct. Mater. 2009, 19, 2950−2960.
(4) Tsuzuki, T.; Tokito, S. Adv. Mater. 2007, 19, 276−280.
(5) Yang, C.-H.; Cheng, Y.-M.; Chi, Y.; Hsu, C.-J.; Fang, F.-C.;
Wong, K.-T.; Chou, P.-T.; Chang, C.-H.; Tsai, M.-H.; Wu, C.-C.
Angew. Chem., Int. Ed. 2007, 46, 2418−2421.
(6) Lowry, M. S.; Bernhard, S. Chem.Eur. J. 2006, 12, 7970−7977.
(7) Ulbricht, C.; Beyer, B.; Friebe, C.; Winter, A.; Schubert, U. S.
Adv. Mater. 2009, 21, 4418−4441.
(37) McMillin, D. R.; Kirchhoff, J. R.; Goodwin, K. V. Coord. Chem.
Rev. 1985, 64, 83−92.
(38) McMillin, D. R.; McNett, K. M. Chem. Rev. 1998, 98, 1201−
1219.
(39) Crane, D. R.; DiBenedetto, J.; Palmer, C. E. A.; McMillin, D. R.;
Ford, P. C. Inorg. Chem. 1988, 27, 3698−3700.
(40) Simon, J. A.; Palke, W. E.; Ford, P. C. Inorg. Chem. 1996, 35,
6413−6421.
(41) Crestani, M. G.; Manbeck, G. F.; Brennessel, W. W.;
McCormick, T. M.; Eisenberg, R. Inorg. Chem. 2011, 50, 7172−7188.
(42) Ford, P. C.; Cariati, E.; Bourassa, J. Chem. Rev. 1999, 99, 3625−
3647 and references therein.
(43) Hashimoto, M.; Igawa, S.; Yashima, M.; Kawata, I.; Hoshino,
M.; Osawa, M. J. Am. Chem. Soc. 2011, 133, 10348−10351.
(44) Czerwieniec, R.; Yu, J.; Yersin, H. Inorg. Chem. 2011, 50, 8293−
8301.
(45) Czerwieniec, R.; Yu, J.; Yersin, H. Inorg. Chem. 2012, 51, 1975.
(46) Yersin, H.; Czerwieniec, R.; Hupfer, A. Singlet harvesting with
brightly emitting Cu(I) and metal-free organic compounds. Proc. SPIE
2012, 843508.
(47) Yersin, H.; Rausch, A. F.; Czerwieniec, R.; Hofbeck, T.; Fischer,
T. Coord. Chem. Rev. 2011, 255, 2622−2652.
(48) Gushurst, A. K. I.; McMillin, D. R.; Dietrich-Buchecker, C. O.;
Sauvage, J. P. Inorg. Chem. 1989, 28, 4070−4072.
(49) Miller, M. T.; Gantzel, P. K.; Karpishin, T. B. J. Am. Chem. Soc.
1999, 121, 4292−4293.
(50) Cuttel, D. G.; Kuang, S.-M.; Fanwick, P. E.; McMillin, D. R.;
Walton, R. A. J. Am. Chem. Soc. 2002, 124, 6−7.
(51) Zhang, Q.; Ding, J.; Cheng, Y.; Wang, L.; Xie, Z.; Jing, X.; Wang,
F. Adv. Funct. Mater. 2007, 17, 2983−2990.
(52) Kyle, K. R.; Ryu, C. K.; DiBenedetto, J. A.; Ford, P. C. J. Am.
Chem. Soc. 1991, 113, 2954−2965.
(8) Lamansky, S.; Djurovich, P. I.; Murphy, D.; Abdel-Razzaq, F.;
Kwong, R.; Tsyba, I.; Bortz, M.; Mui, B.; Bau, R.; Thompson, M. E.
Inorg. Chem. 2001, 40, 1704−1077.
(9) Flamigni, L.; Barbieri, A.; Sabatini, C.; Ventura, B.; Barigelletti, F.
Top. Curr. Chem. 2007, 281, 143−203.
(10) Yersin, H.; Rausch, A. F.; Czerwieniec, R. In Physics of Organic
Semiconductors; Bruetting, W., Adachi, C., Holmes, R. J., Eds.; Wiley-
VCH: Weinheim, 2012; p 371.
(11) Hofbeck, T.; Yersin, H. Inorg. Chem. 2010, 49, 9290−9299.
(12) Zhou, G.-J.; Wang, X.-Z.; Wong, W.-Y.; Yu, X.-M.; Kwok, H.-S.;
Lin, Z. J. Organomet. Chem. 2007, 692, 3461−3473.
(13) Wong, W.-Y.; He, Z.; So, S.-K.; Tong, K.-L.; Lin, Z.
Organometallics 2005, 24, 4079−4082.
(14) Che, C.-M.; Kwok, C.-C.; Lai, S.-W.; Rausch, A., F.;
Finkenzeller, W., J.; Zhu, N.; Yersin, H. Chem.Eur. J. 2010, 16,
233−247.
(15) Rausch, A., F.; Murphy, L.; Williams, J. A. G.; Yersin, H. Inorg.
Chem. 2012, 51, 312−319.
(16) Williams, J. A. G. Top. Curr. Chem. 2007, 281, 205−268.
(17) Williams, J. A. G.; Develay, S.; Rochester, D. L.; Murphy, L.
Coord. Chem. Rev. 2008, 252, 2596−2611.
M
dx.doi.org/10.1021/ic300979c | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(53) Parker, C. A.; Hatchard, C. G. Trans. Faraday Soc. 1961, 57,
1864−1904.
(89) Yersin, H.; Fischer, T.; Monkowius, U.; Hofbeck, T. DE 10
2009 030 475 A1.
(90) The complexes tend to include solvent molecules, which cannot
be removed completely. In addition, some of the complexes have a low
solubility in most common organic solvents which hampers
purification by recrystallisation. Therefore, a highly accurate elemental
analysis cannot be achieved.
(54) Parker, C. A.; Hatchard, C. G. J. Phys. Chem. 1962, 66, 2506−
2511.
(55) Saltiel, J.; Curtis, H. C.; Metts, L.; Miley, J. W.; Winterle, J.;
Wrighton, M. J. Am. Chem. Soc. 1970, 92, 410−411.
(56) Endo, A.; Sato, K.; Yoshimura, K.; Kai, T.; Kawada, A.; Miyazaki,
H.; Adachi, C. Appl. Phys. Lett. 2011, 98, 083302/1−083302/3.
(57) Goushi, K.; Yoshida, K.; Adachi, C. Nat. Photonics 2012, 6, 253−
258.
(58) Felder, D.; Nierengarten, J.-F.; Barigelletti, F.; Ventura, B.;
Armaroli, N. J. Am. Chem. Soc. 2001, 123, 6291−6299.
(59) Zhang, Q.; Zhou, Q.; Cheng, Y.; Wang, L.; Ma, D.; Jing, X.;
Wang, F. Adv. Mater. 2004, 16, 432−436.
(91) Aslanidis, P.; Cox, P. J.; Divanidis, S.; Tsipis, A. C. Inorg. Chem.
2002, 41, 6875−6886.
(92) Henary, M.; Wootton, J. L.; Khan, A. I.; Zink, J. I. Inorg. Chem.
1997, 36, 796−801.
(93) Alcock, N. W.; Moore, P.; Lampe, P. A.; Mok, K. F. J. Chem.
Soc., Dalton Trans. 1982, 207−210.
(94) Araki, H.; Tsuge, K.; Sasaki, Y.; Ishizaka, S.; Kitamura, N. Inorg.
Chem. 2007, 46, 10032−10034.
(60) Deaton, J. C.; Switalski, S. C.; Kondakov, D. Y.; Young, R. H.;
Pawlik, T. D.; Giesen, D. J.; Harkins, S. B.; Miller, A. J. M.;
Mickenberg, S. F.; Peters, J. C. J. Am. Chem. Soc. 2010, 132, 9499−
9508.
(95) Lobana, T. S.; Sharma, R.; Bermejo, E.; Castineiras, A. Inorg.
Chem. 2003, 42, 7728−7730.
(96) Ganesamoorthy, C.; Balakrishna, M. S.; George, P. P.; Mague, J.
T. Inorg. Chem. 2007, 46, 848−858.
(61) Liu, Z.; Qayyum, M. F.; Wu, C.; Whited, M. T.; Djurovich, P. I.;
Hodgson, K. O.; Hedman, B.; Solomon, E. I.; Thompson, M. E. J. Am.
Chem. Soc. 2011, 133, 3700−3703.
(97) Bondi, A. J. Phys. Chem. 1964, 68, 441−451.
(98) Barron, P. F.; Dyason, J. C.; Healy, P. C.; Engelhardt, L. M.;
Pakawatchai, C.; Patrick, V. A.; White, A. H. J. Chem. Soc., Dalton
Trans. 1987, 1099−1106.
(62) Cuperly, D.; Gros, P.; Fort, Y. J. Org. Chem. 2002, 67, 238−241.
(63) Bruck, A.; Ruhland, K. Organometallics 2009, 28, 6383−6401.
̈
(99) Cunningham, C. T.; Cunningham, K. L. H.; Michalec, J. F.;
McMillin, D. R. Inorg. Chem. 1999, 38, 4388−4392.
(100) Ryu, C. K.; Vitale, M.; Ford, P. C. Inorg. Chem. 1993, 32, 869.
Vitale, M.; Ryu, C. K.; Palke, W. E.; Ford, P. C. Inorg. Chem. 1994, 33,
561.
(64) Monkowius, U.; Zabel, M.; Fleck, M.; Yersin, H. Z. Naturforsch.
2009, 64b, 1513−1524.
(65) Moore, S. S.; Whitesides, G. M. J. Org. Chem. 1982, 47, 1489−
1493.
(66) Grotjahn, D. B.; Lev, D. A. J. Am. Chem. Soc. 2004, 126, 12232−
12233.
(101) NMR spectra of only a few complexes with sufficient solubility
could be measured and feature very broad signals indicative of a
fluxional behavior of the ligand sphere. In addition, in the mass spectra
no signals representing the ions of the complete complexes could be
detected. Therefore, the identity of the solution species is not without
doubt. However, due to the similar emission peaks we assume that the
complex molecules are identical both in solution and in the solid state.
(102) Tanaka, Y.; Azumi, T. Inorg. Chem. 1986, 25, 247−248.
(103) Complexes 1-Cl and 1-Br show similar properties as 1-I.
Corresponding data will be presented elsewhere.
(67) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112−122.
(68) Becke, A. D. Phys. Rev. A 1988, 38, 3098−3100.
(69) Perdew, J. P. Phys. Rev. B 1986, 33, 8822−8827.
(70) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785−789.
(71) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J.
Phys. Chem. 1994, 98, 11623−11627.
(72) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7,
3297−3305.
(73) Rappoport, D.; Furche, F. J. Chem. Phys. 2010, 133, 134105/1−
134105/11.
(74) Haser, M.; Ahlrichs, R. J. Comput. Chem. 1989, 10, 104−111.
̈
(75) Weigend, F.; Haser, M. Theor. Chem. Acc. 1997, 97, 331−340.
̈
(76) Sierka, M.; Hogekamp, A.; Ahlrichs, R. J. Chem. Phys. 2003, 118,
9136−9148.
(77) Kuhn, M.; Weigend, F. Chem. Phys. Chem. 2011, 12, 3331−
̈
3336.
(78) TURBOMOLE V6.3 2011, a development of University of
Karlsruhe and Forschungszentrum Karlsruhe GmbH, 1989−2007,
TURBOMOLE GmbH, since 2007; available from http://www.
turbomole.com.
(79) Eisler, D. J.; Kirby, C. W.; Puddephatt, R. J. Inorg. Chem. 2003,
42, 7626−7634 and references therein.
(80) Fitchett, C. M.; Steel, P. J. Inorg. Chem. Commun. 2007, 10,
1297−1300.
(81) Schramm, V.; Pierre, A.; Hiller, W. Acta Crystallogr. 1984, C40,
1840−1841.
(82) Hiller, W. Acta Crystallogr. 1986, C42, 149−150.
(83) Engelhardt, L. M.; Healy, P. C.; Kildea, J. D.; White, A. H. Aust.
J. Chem. 1989, 42, 913−922.
(84) Maeyer, J. T.; Johnson, T. J.; Smith, A. K.; Borne, B. D.; Pike, R.
D.; Pennington, W. T.; Krawiec, M.; Rheingold, A. L. Polyhedron 2003,
22, 419−431.
(85) Araki, H.; Tsuge, K.; Sasaki, Y.; Ishizaka, S.; Kitamura, N. Inorg.
Chem. 2005, 44, 9667−9675.
(86) Hirtenlehner, C.; Monkowius, U. Inorg. Chem. Commun. 2012,
15, 109−112.
(87) Mackewitz, T.; Volland, M.; Paciello, R.; Schafer, A.; Breit, B.;
̈
Seiche, W. DE 10 2004 014 474 A1, 2004.10.21, 2004.
(88) Feltham, R. D.; Gary, H. J. Organomet. Chem. 1971, 33, 347−
355.
N
dx.doi.org/10.1021/ic300979c | Inorg. Chem. XXXX, XXX, XXX−XXX