Control of emission colour with N-heterocyclic carbene (NHC) ligands in phosphorescent three-coordinate Cu(i) complexes

Article abstract of DOI:10.1039/c4cc02037e

A series of three phosphorescent mononuclear (NHC)-Cu(i) complexes were prepared and characterized. Photophysical properties were found to be largely controlled by the NHC ligand chromophore. Variation of the NHC ligand leads to emission colour tuning over 200 nm range from blue to red, and emission efficiencies of 0.16-0.80 in the solid state.

Full text of DOI:10.1039/c4cc02037e

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Control of emission colour with N-heterocyclic
carbene (NHC) ligands in phosphorescent three-
coordinate Cu(I) complexes†
Cite this: Chem. Commun., 2014,
50, 7176
Received 19th March 2014,
Accepted 16th May 2014
Valentina A. Krylova, Peter I. Djurovich, Brian L. Conley, Ralf Haiges,
Matthew T. Whited, Travis J. Williams and Mark E. Thompson*
DOI: 10.1039/c4cc02037e
www.rsc.org/chemcomm
A series of three phosphorescent mononuclear (NHC)–Cu(I) com-
plexes were prepared and characterized. Photophysical properties
were found to be largely controlled by the NHC ligand chromo-
phore. Variation of the NHC ligand leads to emission colour tuning
over 200 nm range from blue to red, and emission efficiencies of
0.16–0.80 in the solid state.
Phosphorescent Cu(I) complexes are an emerging class of lumines-
cent materials based on an inexpensive and abundant metal.1 The
ability to tune chemical and photophysical properties in a desirable
and predictable way is highly important when considering potential
applications of Cu(I)-based phosphors. The typical strategy to
modulate the excited state properties of these and related lumines-
cent materials is usually achieved through variation of the coordi-
nating ligand(s).2 To date the types of ligands most commonly
used to prepare phosphorescent Cu(I) complexes are diimines or
organophosphines and their derivatives.3 Alternatively, N-heterocyclic
Fig. 1 Molecular structures (top) and perspective view at 50% probability
(bottom) of complexes 1–3. Only one of the unique structures for 2 found
in the unit cell is shown. Hydrogen atoms are omitted for clarity.
carbenes (NHC) are an attractive class of ligands as they are electro- Herein, we report a series of luminescent (NHC)Cu(N^N) com-
nically and sterically tunable and form strong bonds with transition plexes 1–3 (Fig. 1), where the NHC ligand is principally involved in
metals giving robust complexes.4 However, while NHCs have been the excited state and demonstrate a wide range emission colour
employed as either chromophoric or ancillary ligands in luminescent tunability through modification of carbene moiety. In particular,
Ir and Pt complexes5 they have been rarely used as chromophoric we systematically lowered the energy gap of 1 by benzannulation of
ligands in Cu(I) complexes.6
imidazolylidene ring to make 2 and further introduced nitrogen
We have recently used NHC ligands to prepare phosphorescent atoms to form the pyrazinyl moiety in 3. In addition, we utilize an
3-coordinate Cu(I) complexes (NHC)Cu(N^N), where N^N denotes anionic non-conjugated N^N ligand, i.e. di(2-pyridyl)dimethylborate
a neutral diimine or monoanionic pyridyl-azolate ligand.7 The (py2BMe2) that possesses high triplet energy to serve as an ancillary
monodentate NHC ligand, 1,3-bis(2,6-diisopropylphenyl)imidazol- ligand.8 To the best of our knowledge the py2BMe2 ligand, unlike the
2-ylidene (IPr), employed in these complexes has both a large p–p* isoelectronic di(1-pyrazolyl)borates (pz2BR2, R = H, alkyl, aryl),2c,9 has
energy gap and high triplet energy, therefore the emission energy in never been used to prepare luminescent transition metal complexes.
these derivatives is controlled by variations in the N^N ligand. We have found that (NHC)Cu complexes with the py2BMe2 ligand are
more robust and luminescent than the pyrazolyl-borate congeners.
The (NHC)Cu(py2BMe2) complexes were obtained from their
respective (NHC)CuCl precursors upon addition of a stoichiometric
amount of sodium di(2-pyridyl)dimethylborate in tetrahydrofuran
at RT. Complexes 1–3 are stable in solid state and in solution under
anaerobic conditions. Complex 1 can be sublimed under vacuum
and is stable for hours in solution, while 2 and 3 decompose slowly
under aerobic conditions in solution and blacken in the solid state
Department of Chemistry, University of Southern California,
Los Angeles 90089-0744, USA. E-mail: met@usc.edu
† Electronic supplementary information (ESI) available: Experimental and com-
putational details, synthesis and characterization of new compounds, absorption
data for precursors and (IPr)Cu(pz2BH2) complex, photophysical data for 1–3
obtained at 77 K in the solid state and in methylcyclohexane glass and crystal-
lographic data. CCDC 976577–976579 and 990891. For ESI and crystallographic
data in CIF or other electronic format see DOI: 10.1039/c4cc02037e
7176 | Chem. Commun., 2014, 50, 7176--7179
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
after 24 h exposure to air. Evidently, the isopropyl groups at ortho in 1 (e B 7000–14 200 MÀ1 cmÀ1), 310 nm in 2 (e B 6500–
positions of phenyl groups of the NHC ligand in 1 impart greater 19 000 MÀ1 cmÀ1) and 340 nm in 3 (e B 4000–13 400 MÀ1 cmÀ1
)
stability of the (NHC)Cu(py2BMe2) complexes than in 2 and 3. We also are assigned to spin-allowed ligand centered (LC) transitions on
prepared complexes analogous to 1 and 2 using the pz2BH2 ligand both the NHC and py2BMe2 ligands. Lower energy bands, not
instead of py2BMe2. While the analog to 1, (IPr)Cu(pz2BH2), can be observed in absorption spectra of precursors (see ESI†), are
isolated and fully characterized (see ESI†), the congener to 2 decom- assigned to charge transfer (CT) transitions. In complex 1 the CT
posed rapidly upon exposure to air and was not examined further.
bands appear at 316 nm (e = 6100 MÀ1 cmÀ1) with a shoulder at
X-ray diffraction analyses confirmed monomeric three-coordinate 360 nm (e B 1300 MÀ1 cmÀ1). A comparison between the absorp-
structures for complexes 1–3. Complex 2 has two unique structures in tion spectrum of 1 to that of (IPr)Cu(pz2BH2) (see ESI†) shows the
the unit cell that have similar geometric parameters. The coordination LC band is unchanged in energy in both derivatives, whereas the
geometry in complexes 1–3 can be described as Y-shaped with the CT bands shift to higher energy and lower intensity (lmax = 305 nm,
sum of bond angles around copper close to 3601 (359.981 in 1, 359.721 e = 2000 MÀ1 cmÀ1 and lmax = 330 nm, e B 1000 MÀ1 cmÀ1) in the
in 2 and 358.641 in 3). The Cu–N–C–B–C–N ring formed upon latter complex. The bathochromic shift for the low energy bands in
chelation of the py2BMe2 ligand adopts a boat-shaped conformation 1 indicates that the py2BMe2 ligand participates in these transi-
similar to that reported in metal complexes bearing related tions, although some CT character involving the IPr ligand may
di(2-pyridyl)borate ligands.8,10 The relative orientations of NHC and contribute as well. Upon expansion of p-system of the NHC ligand
py2BMe2 ligands in crystals differ within the series. In complex 1 the in 2 the CT band becomes more distinct and intense (lmax
=
ligands are arranged with the pyridyl rings situated opposite the aryl 346 nm, e = 9100 MÀ1 cmÀ1). Substitution of the two CH-groups
rings of the NHC ligand across a crystallographic mirror plane that with nitrogens in 3 leads to a marked red shift and increase in
bisects the CNHC, Cu and B atoms. In contrast, the py2BMe2 ligand in molar absorptivity (lmax = 422 nm, e = 10 300 MÀ1 cmÀ1). Thus,
2 and 3 is oriented about CNHC–Cu bond so that the two pyridyl rings these bands in 2 and 3 are unambiguously assigned to CT transi-
are situated above and below a plane defined by the NNHC, NNHC and tions involving NHC ligands.
CNHC atoms. The Cu–Npy bond lengths in 1 are 2.0288(15) Å and
slightly shorter in 2 (1.9929(16) Å and 1.9997(16) Å) and 3 (2.010(9) Å solids of complexes 1–3 at room temperature are broad and
and 2.014(9) Å). The CNHC–Cu–Npy angles are 132.78(4)1 in 1 and vary featureless (Fig. 2). Complex 1 gives sky-blue emission (lmax
The emission spectra recorded for neat microcrystalline
=
from 134.32(7)1 and 129.27(7)1 in 2 to (135.0(6)1 and 128.1(6)1) in 3. 476 nm), complex 2 displays yellow emission (lmax = 570 nm) and 3
The CNHC–Cu distances in 1–3 (1.8678(19)–1.895(2) Å) are within the has orange-red emission (lmax = 638 nm). The bathochromic shift in
range for reported NHC–Cu(I) complexes.11
In solution H NMR data indicates rapid boat-to-boat intercon- energy excited state is governed largely by the NHC ligand. Solid
version of the py2BMe2 ligand as resonances of methyl groups powders of 1 and 2 glow brightly upon excitation with emission
attached to boron atom appear as one broad singlet both at room quantum yields (F) of 0.80 and 0.70 respectively, while 3 has
emission for complexes 2 and 3 further indicates that the lowest
1
1
temperature and at À40 1C in acetone-d6. Although the H NMR moderate emission efficiency (F = 0.16). In contrast, the quantum
data do not allow us to assess if there is free rotation about the efficiency of the (IPr)Cu(pz2BH2) derivative in the solid state is much
CNHC–Cu bonds in solution, the chemical shift for the protons ortho lower (l = 415 nm, F = 0.03). The observed luminescence for 1–3 is
to the pyridyl nitrogens gives insight into the preferred molecular phosphorescence as emission lifetimes (t) are in the microsecond
conformation. This Àresonance appears at d = 8.36 ppm in the regime. The radiative rate constants (kr) in the solid state vary within
protonated py2BMe2 ligand, whereas upon coordination to copper the small range of values (kr = (3.3–7.2) Â 104 sÀ1). In fluid solution
in 1 it is shifted markedly upfield to d = 7.3 ppm due to shielding by the emission efficiency is substantially lower than in the solid state.
the diamagnetic ring current from the adjacent aryl rings of the In particular, complex 1 has quantum yield of 0.15 (t = 2.3 ms) in
NHC ligand. In contrast, the same resonance undergoes a much cyclohexane, while emission from 2 and 3 is almost completely
smaller shift upon coordination in 2 and 3, appearing at d = quenched (F o 0.005).
1
7.97 ppm and d = 8.05 ppm, respectively. Thus, the H NMR data
The emission spectra of neat samples of 1–3 shift to lower
in solution correlate with the relative ligand orientation found in energies upon cooling to 77 K (Table 1, also see ESI†). Batho-
crystalline state; co-planar for 1, perpendicular for 2 and 3. chromic shifts in emission energy at low temperature are
Photophysical data for complexes 1–3 are summarized in common for Cu(I) complexes. This phenomenon is often attributed
Table 1. The UV-visible absorption spectra for complexes 1–3 in to suppression of thermally activated delayed fluorescence (TADF)
dichloromethane are shown in Fig. 2. High energy bands at 290 nm and usually accompanied by a marked increase in emission
Table 1 Photophysical data for complexes 1–3
Emission at room temperatureb
Emission at 77 Kb
Absorbance,a l (nm) e (103 MÀ1 cmÀ1
)
lmax (nm)
t (ms)
F
kr (sÀ1
)
knr (sÀ1
)
lmax (nm)
t (ms)
1
2
3
268 (14.2), 316 (6.1), 360 sh (1.3)
257 sh (19.3), 346 (9.1)
271 (13.3), 305 (7.4), 422 (10.3)
476
570
638
11
15
7.5
0.8
0.7
0.16
7.2 Â 104
4.7 Â 104
3.3 Â 104
1.8 Â 104
2.0 Â 104
1.0 Â 105
492
586
650
36
17
21
a
b
In dichloromethane. In solid state.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 7176--7179 | 7177

View Article Online
ChemComm
Communication
Fig. 2 Absorption (open symbols, CH2Cl2) and emission (closed symbols,
solid powder) spectra of complexes 1–3 at room temperature.
lifetimes of an order of magnitude or more.9,12 Complexes 1–3,
however, show only a modest increase in emission lifetimes
upon cooling as emission lifetimes measured at 77 K are in the
range of 17–36 ms (Table 1). The relatively small increase in lifetime
at 77 K is inconsistent with processes typically associated with
TADF and suggests instead that emission measured both at room
temperature and 77 K is from a state that is principally triplet in
character. The behavior also implies that the radiative rate constant
for the lowest triplet state is significantly enhanced in 1–3. This
unusual temperature dependence on the emission lifetime is being
currently investigated in greater detail at lower temperatures.
Fig. 3 (A) HOMO and LUMO plots and energies for 1–3. (B) Optimized triplet
geometries and triplet spin density contour plots (isovalue: 0.004 e a0À3).
Hydrogen atoms are omitted for clarity.
The observed bathochromic shift of emission energy in 1–3 by 2 (ELUMO = À0.98 eV) and 3 (ELUMO = À2.05 eV). The HOMO
upon expanding the size of the p-system of a ligand chromo- energies show a similar trend in stabilization, albeit to a lesser
phore and N-substitution is consistent with a decrease in degree (EHOMO = À4.85 eV, À4.96 eV, À5.20 eV, for 1–3,
separation between the highest occupied molecular orbital respectively). The HOMO–LUMO gap is thus progressively
(HOMO) and the lowest unoccupied molecular orbital (LUMO). smaller for 1 (DEH–L = 4.27 eV), 2 (DEH–L = 3.98 eV) and 3
Computational analyses of the ground and excited state proper- (DEH–L = 3.15 eV).
ties performed using density functional theory (DFT) and time-
The lowest vertical singlet and triplet excitations obtained
dependent DFT (TD-DFT) calculations compare favorably with from TD-DFT calculations are mainly HOMO - LUMO transi-
the experimental observations. The calculated wavelength and tions (see ESI†). On the basis of the MO description given above
oscillator strength of the lowest singlet transitions progres- the lowest lying transition for complex 1 can be ascribed as
sively increase for 1 (l = 381 nm, f = 0.0028), 2 (l = 400 nm, (M + L)LCT admixed with intraligand p - p* (py2BMe2) (ILCT)
f = 0.1440) and 3 (l = 522 nm, f = 0.1645). This result follows the character, whereas for complexes 2 and 3 the transition is
trend observed in absorption spectra, i.e. a decrease in energy principally metal–ligand to NHC–ligand charge transfer ((M + L)LCT).
and increase of molar absorption for the CT bands when going The calculated spin density surfaces for the triplet electronic
from 1 to 2 to 3. The frontier molecular orbitals for 1–3 are configuration further support this assignment (Fig. 3B). For
shown in Fig. 3A. For all three complexes the calculated complexes 1–3 the spin contours are localized along the CNHC
–
HOMOs have essentially identical spatial contours, consisting Cu bond axis. Both ligands also contribute to the triplet spin
predominantly of d orbitals on copper (39–48%) mixed with density; however, while complex 1 has a significant contribution
orbitals on di(2-pyridyl)dimethylborate ligand (41–45%). The from the borate ligand (31% NHC, 31% py2BMe2), the spin distribu-
LUMO in 1–3 is localized on the NHC ligand (85–94%) with tion is shifted toward NHC ligand for complexes 2 (52% NHC, 16%
minimal metal character (4–6%). A small contribution (8%) py2BMe2) and 3 (54% NHC, 16% py2BMe2).
from the py2BMe2 orbitals appears in the LUMO of complex 1;
To emphasize the importance of proper ligand design
however, there is less (4% and 2%) in both 2 and 3. Noteworthy to achieve efficient room temperature phosphorescence from
is a substantial contribution (8–23%) from the carbene carbon this family of Cu(I) compounds, we prepared complex 4 where
2pz orbital in the LUMO of all three complexes. Congruent with the p-system of the NHC ligand was expanded by annulation of
the orbital composition, variations of the carbene ligand have imidazolylidene with a peri-naphthyl moiety (Fig. 4, full char-
pronounced effect on LUMO energies. Complex 1 has the acterization is given in ESI†). For this derivative, the intensity of
highest LUMO energy in the series (ELUMO = À0.58 eV) followed the lowest lying CT absorption band centered at 450 nm is low
7178 | Chem. Commun., 2014, 50, 7176--7179
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
these findings introduce a new versatile method to control the
photophysical properties of luminescent Cu(I) complexes.
We are grateful to the Universal Display Corporation for
financial support. B.L.C. and T.J.W. also thank National Science
Foundation CHE-1054910 grant for support. Theoretical calcu-
lations were supported by the University of Southern California
Center for High-Performance Computing and Communications
(www.usc.edu/hpcc). The X-ray diffractometer is sponsored by
National Science Foundation CRIF Grant 1048807.
Notes and references
1 A. Barbieri, G. Accorsi and N. Armaroli, Chem. Commun., 2008, 2185.
2 (a) J. Brooks, Y. Babayan, S. Lamansky, P. I. Djurovich, I. Tsyba,
R. Bau and M. E. Thompson, Inorg. Chem., 2002, 41, 3055; (b) Y. Chi
and P.-T. Chou, Chem. Soc. Rev., 2010, 39, 638; (c) J. Li,
P. I. Djurovich, B. D. Alleyne, M. Yousufuddin, N. N. Ho,
J. C. Thomas, J. C. Peters, R. Bau and M. E. Thompson, Inorg. Chem.,
2005, 44, 1713; (d) Y. You and S. Y. Park, Dalton Trans., 2009, 1267.
3 (a) G. F. Manbeck, W. W. Brennessel and R. Eisenberg, Inorg. Chem.,
2011, 50, 3431; (b) C.-W. Hsu, C.-C. Lin, M.-W. Chung, Y. Chi,
G.-H. Lee, P.-T. Chou, C.-H. Chang and P.-Y. Chen, J. Am. Chem. Soc.,
2011, 133, 12085; (c) M. Hashimoto, S. Igawa, M. Yashima, I. Kawata,
M. Hoshino and M. Osawa, J. Am. Chem. Soc., 2011, 133, 10348;
(d) S. B. Harkins and J. C. Peters, J. Am. Chem. Soc., 2005, 127, 2030;
(e) D. G. Cuttell, S.-M. Kuang, P. E. Fanwick, D. R. McMillin and
R. A. Walton, J. Am. Chem. Soc., 2002, 124, 6; ( f ) M. G. Crestani,
G. F. Manbeck, W. W. Brennessel, T. M. McCormick and R. Eisenberg,
Inorg. Chem., 2011, 50, 7172; (g) N. Armaroli, G. Accorsi, F. Cardinali
and A. Listorti, Top. Curr. Chem., 2007, 280, 69; (h) S. Igawa,
M. Hashimoto, I. Kawata, M. Yashima, M. Hoshino and M. Osawa,
J. Mater. Chem. C, 2013, 1, 542–551; (i) M. Wallesch, D. Volz, D. M. Zink,
U. Schepers, M. Nieger, T. Baumann and S. Braese, Chem. – Eur. J.,
2014, 20, 6578–6590.
4 (a) D. Bourissou, O. Guerret, F. P. Gabbaie and G. Bertrand, Chem.
Rev., 2000, 100, 39; (b) H. Jacobsen, A. Correa, A. Poater, C. Costabile
and L. Cavallo, Coord. Chem. Rev., 2009, 253, 687.
5 (a) Y. Unger, D. Meyer, O. Molt, C. Schildknecht, I. Muenster,
G. Wagenblast and T. Strassner, Angew. Chem., Int. Ed., 2010,
49, 10214; (b) K.-Y. Lu, H.-H. Chou, C.-H. Hsieh, Y.-H. O. Yang,
H.-R. Tsai, H.-Y. Tsai, L.-C. Hsu, C.-Y. Chen, I. C. Chen and
C.-H. Cheng, Adv. Mater., 2011, 23, 4933; (c) T. Sajoto, P. I. Djurovich,
A. Tamayo, M. Yousufuddin, R. Bau, M. E. Thompson, R. J. Holmes
and S. R. Forrest, Inorg. Chem., 2005, 44, 7992.
Fig. 4 Absorption (room temperature, CH2Cl2) and emission (77 K, 2-MeTHF)
spectra of complex 4. (inset) Molecular structure, optimized triplet geometry
and spin density contour plot (isovalue: 0.004 e a0À3).
(e = 450 MÀ1 cmÀ1) and the complex is nonemissive in the solid
state at room temperature and at 77 K. Very weak, structured
emission (F o 0.01, t o 10 ns) is observed from a dilute
solution of 4 in frozen 2-methyltetrahydrofuran (2-MeTHF)
glass at 77 K (Fig. 4). This emission is tentatively assigned as
phosphorescence since the radiative rate constant (kr o 106 sÀ1
)
and presence of vibronic features are inconsistent with fluores-
cence from the CT state. The HOMO for 4 calculated using DFT
is essentially identical to that of complexes 1–3. The LUMO,
localized primarily on the aromatic p-system of NHC ligand, has
no electron density on the 2pz orbital of CNHC atom in strong
contrast to what is found in complexes 1–3. Such an electronic
distribution leads to poor overlap between frontier orbitals and
thus a low oscillator strength for the lowest lying CT transitions
(l = 597 nm, f = 0.0044). The triplet spin density of 4 is localized
on the acenaphthyl moiety and, unlike that of complexes 1–3, has
a node across the CNHC–Cu bond axis (Fig. 4). The spin distribu-
tion, together with structured emission spectrum, suggests that
the luminescence is ligand centered in character. Thus, extension
of the p-system in this manner, while shrinking the HOMO–
LUMO gap (DEH–L = 2.55 eV), decreases the energy of the 3LC state
on the NHC ligand and reduces electronic coupling to such a
degree that it can no longer effectively interact with the MLCT
states responsible for promoting fast radiative decay.3b
6 (a) K. Matsumoto, N. Matsumoto, A. Ishii, T. Tsukuda, M. Hasegawa
and T. Tsubomura, Dalton Trans., 2009, 6795; (b) R. Visbal and
M. C. Gimeno, Chem. Soc. Rev., 2014, 43, 3551.
7 (a) V. A. Krylova, P. I. Djurovich, J. W. Aronson, R. Haiges,
M. T. Whited and M. E. Thompson, Organometallics, 2012,
31, 7983; (b) V. A. Krylova, P. I. Djurovich, M. T. Whited and
M. E. Thompson, Chem. Commun., 2010, 46, 6696.
8 T. G. Hodgkins and D. R. Powell, Inorg. Chem., 1996, 35, 2140.
9 R. Czerwieniec, J.-B. Yu and H. Yersin, Inorg. Chem., 2011, 50, 8293.
10 B. L. Conley and T. J. Williams, J. Am. Chem. Soc., 2010, 132, 1764.
In conclusion, we report a series of (NHC)–Cu(I) complexes
that show phosphorescence associated primarily with NHC 11 S. Diez-Gonzalez, E. C. Escudero-Adan, J. Benet-Buchholz,
E. D. Stevens, A. M. Z. Slawin and S. P. Nolan, Dalton Trans., 2010,
39, 7595.
12 (a) G. Blasse and D. R. McMillin, Chem. Phys. Lett., 1980, 70, 1;
ligand chromophore. Judicious modification of the NHC ligand
allows the emission colour to be tuned over 200 nm from blue
to orange-red while retaining high emission efficiencies. The
estimated triplet radiative rate constants are comparable with
those of third row transition metal complexes. Taking into
account electronic and steric tunability of the NHC ligands,
(b) J. C. Deaton, S. C. Switalski, D. Y. Kondakov, R. H. Young,
T. D. Pawlik, D. J. Giesen, S. B. Harkins, A. J. M. Miller,
S. F. Mickenberg and J. C. Peters, J. Am. Chem. Soc., 2010,
132, 9499; (c) M. J. Leitl, F.-R. Kuechle, H. A. Mayer, L. Wesemann
and H. Yersin, J. Phys. Chem. A, 2013, 117, 11823.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 7176--7179 | 7179