Electrophosphorescent homo- and heteroleptic copper(I) complexes prepared from various bis-phosphine ligands

Article abstract of DOI:10.1039/b707398d

Homo- and heteroleptic copper(i) complexes obtained from various chelating bis-phosphine ligands and Cu(CH₃CN)₄BF₄ have been used for the preparation of light emitting devices. The Royal Society of Chemistry.

Full text of DOI:10.1039/b707398d

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Electrophosphorescent homo- and heteroleptic copper(I) complexes
prepared from various bis-phosphine ligands{
Omar Moudam,a Adrien Kaeser,a Be´atrice Delavaux-Nicot,a Carine Duhayon,a Michel Holler,b
Gianluca Accorsi,c Nicola Armaroli,*c Isabelle Se´guy,d Jose Navarro,d Pierre Destruel*d and
Jean-Franc¸ois Nierengarten*a
Received (in Cambridge, UK) 16th May 2007, Accepted 13th June 2007
First published as an Advance Article on the web 26th June 2007
DOI: 10.1039/b707398d
combining two different P–P ligands have never been reported to
the best of our knowledge. Finally, we also show that some of
these compounds are promising materials for OLED applications.
CuI(Phen)(P–P) derivatives are usually prepared by reaction of
1,10-phenanthroline with a 1 : 1 mixture of the bis-phosphine
ligand and a CuI salt such as Cu(CH3CN)4BF4. When this
standard procedure was applied to 1,2-bis(diphenylphosphino)-
benzene (dppb) and bis(diphenylphosphino)methane (dppm), the
formation of the expected CuI(Phen)(P–P) derivatives was not
observed. Actually, the homoleptic complexes CuI(dppb)2 (1) and
CuI(dppm)2 (2) were thus produced as their tetrafluoroborate salts.
The structural assignment of 1 and 2 thus obtained was deduced
from their 31P{H} NMR spectra displaying sharp singlets at d 8.1
and 27.3 ppm, respectively. The structures of both 1 and 2 were
further confirmed by mass spectrometry. The expected molecular
ion peaks were observed at m/z 955.7 for 1 ([M 2 BF42]+, calcd
for C60H48P4Cu: 955.20) and 831.2 for 2 ([M 2 BF42]+, calcd for
C50H44P4Cu: 831.17). Finally, it can be added that compounds 1
and 2 were obtained in good yields by reaction of dppb and dppm
(2 equiv.), respectively, with Cu(CH3CN)4BF4 (1 equiv.) in
CH2Cl2. For complex 1, crystals suitable for X-ray crystal-
structure analysis were obtained by slow diffusion of Et2O into a
CH2Cl2 solution of 1.{ As shown in Fig. 1, the CuI center adopts a
highly distorted tetrahedral geometry as a result of the small bite
angle of the dppb ligand (84.7u). It can be noted that the Cu atom
Homo- and heteroleptic copper(I) complexes obtained from
various chelating bis-phosphine ligands and Cu(CH3CN)4BF4
have been used for the preparation of light emitting devices.
The increasing market demand for low-cost flat-panel displays and
efficient lighting sources is a strong driving force for intensive
research on organic light-emitting diodes (OLEDs).1 Fluorescent
semi-conducting polymers have been widely used for such
applications.1 However, the internal quantum efficiency of the
resulting devices is limited to 25% by the non-emissive triplet
excitons produced in OLEDs based on singlet emitters.1 To
overcome this problem, devices employing triplet emitters have
been developed in the past few years.2 Such OLEDs using
phosphorescent heavy metal complexes can, in principle, reach a
quantitative internal quantum efficiency as both singlet and triplet
excitons participate in the emission owing to intersystem crossing
of the singlet excited states to the triplet states. To date, most of the
electrophosphorescent materials used for OLED applications are
complexes of noble metal ions such as IrIII, PtII, RuII and OsII.2
Recent reports have however shown that inexpensive and non-
toxic CuI coordination compounds are also promising candidates
for light emitting devices.3–6 As part of this research, we became
interested in electrophosphorescent heteroleptic CuI complexes
containing 1,10-phenanthroline (phen) and bis-phosphine (P–P)
ligands.6,7 By changing systematically the nature of the chelating
P-ligand, we found that some of them produce stable homoleptic
CuI complexes rather than the expected CuI(Phen)(P–P) derivative.
In this paper, we now report on a series of homo- and heteroleptic
CuI(P–P)2 compounds prepared on the basis of this initial finding.
Whereas examples of homoleptic CuI(P–P)2 derivatives have
been already described,8 analogous heteroleptic CuI complexes
aLaboratoire de Chimie de Coordination du CNRS (UPR 8241), 205
route de Narbonne, 31077, Toulouse Cedex 4, France.
E-mail: jfnierengarten@lcc-toulouse.fr; Fax: +33 (0) 5 61 55 30 03;
Tel: +33 (0) 5 61 33 31 51
bLaboratoire de Physico-Chimie Bioinorganique, ULP-CNRS (UMR
7177), Institut de Chimie, ECPM, 25 rue Becquerel, 67087, Strasbourg
Cedex 2, France
cIstituto per la Sintesi Organica e la Fotoreattivita`, Molecular
Photoscience Group, Consiglio Nazionale delle Ricerche, Via Gobetti
101, 40129, Bologna, Italy. E-mail: armaroli@isof.cnr.it;
Fax: +39 051 639 98 44; Tel: +39 051 639 98 20
Fig. 1 ORTEP plot of the structure of the cation in 1. Thermal ellipsoids
are drawn at the 30% probability level. The prime (9) characters in the
atom labels indicate that these atoms are at equivalent position (3/2 2 x, y,
dLaboratoire Plasma et Conversion d’Energie (LAPLACE), UPS-
CNRS (UMR 5213), 118 route de Narbonne, 31062, Toulouse Cedex 9,
France. E-mail: Pierre.destruel@laplace.ups-tlse.fr;
˚
3/2 2 z). Selected bond lengths and bond angles: Cu(1)–P(1): 2.3092(8) A;
Fax: +33 (0) 5 61 55 64 52; Tel: +33 (0) 5 61 55 62 61
{ Electronic supplementary information (ESI) available: Experimental
details for the preparation of the compounds. DOI: 10.1039/b707398d
˚
Cu(1)–P(2): 2.2907(9) A; P(1)–Cu(1)–P(2): 84.72(3)u; P(1)–Cu(1)–P(19):
122.59(5)u; P(1)–Cu(1)–P(29): 120.94(3)u; P(2)–Cu(1)–P(29): 127.85(5)u.
Chem. Commun., 2007, 3077–3079 | 3077
This journal is ß The Royal Society of Chemistry 2007

View Article Online
lies on a two fold axis. The Cu–P distances are in the range
expected from other known structures.8
Similarly, addition of dppb (1 equiv.) to 1 : 1 mixtures of various
bis-phosphine ligands (POP, dppe, dppp) and Cu(CH3CN)4BF4 in
CH2Cl2 gave the heteroleptic complexes 4–6 in good yields (Fig. 3).
Their structure was confirmed by 1H, 13C, and 31P NMR
spectroscopy, mass spectrometry and elemental analysis.
The attempted preparation of the CuI(Phen)(P–P) derivatives
from dppm and dppb revealed a peculiar behavior for these two
ligands. Effectively, the reaction of 1,10-phenanthroline (1 equiv.)
with a 1 : 1 mixture of other bis-phosphine ligands (bis[2-
(diphenylphosphino) phenyl] ether: POP; 1,2-bis(diphenylpho-
sphino)ethane: dppe; 1,3-bis(diphenylphosphino) propane: dppp)
and Cu(CH3CN)4BF4 in CH2Cl2 gave the expected CuI(Phen)(P–
P) complexes. This difference maybe be explained by the
differences in bite angles for the different chelating P-ligands.
When this angle is small enough (dppm and dppb), the metal
center can easily accommodate two ligands. In contrast, steric
factors resulting from the wider P–Cu–P angle for the other bis-
phosphines substantially destabilize the Cu(P–P)2 derivative, thus
preventing its formation under our experimental conditions.
In principle, by replacing the phen ligand with dppm or dppb in
the reaction sequence used to prepare the CuI(Phen)(P–P)
derivatives, heteroleptic CuI(P–P)2 complexes should be obtained.
This hypothesis was first tested by adding dppm (1 equiv.) to a
1 : 1 mixture of POP and Cu(CH3CN)4BF4 in CH2Cl2. Complex
[Cu(POP)(dppm)]BF4 (3) was effectively thus obtained in a good
yield (80%). The 31P{H} NMR spectrum of 3 revealed two signals
at d 24.4 and 28.9 ppm in good agreement with the proposed
structure. It is worth noting that no detectable changes were
observed in the 31P{H} NMR spectrum of a CD3CN solution of
3 after one week, thus showing that there is no ligand exchange
in solution. X-Ray quality crystals of 3 were obtained by vapor
diffusion of Et2O into a CH2Cl2 solution of the complex.§ The
asymmetric unit of the crystal structure contains two complexes of
3 in two different conformations along with one unit-occupancy
and one 0.5-occupancy CH2Cl2 molecules. As shown in Fig. 2, the
structure of 3 reveals a highly distorted tetrahedral coordination
environment about CuI with P–Cu–P bond angles ranging from
73.28(12) to 125.43(13)u. It can be noted that the POP ligand is
bound to the metal only through its pair of P donor atoms, the
ether O atom being at a nonbonding distance from the Cu(I) center
The dppm containing derivatives 2 and 3 are oxygen sensitive.
Indeed, when these compounds were stored without particular
precaution, oxidation of the dppm ligand was observed. In
contrast, all the dppb-based complexes are stable under normal
laboratory conditions. For this reason, only compounds 1 and 4–6
were further investigated. The electrochemical properties of these
complexes were determined by cyclic voltammetry (CV) and
Osteryoung Square Wave Voltammetry (OSWV). All the experi-
ments were performed at room temperature in CH2Cl2 solutions
containing tetra-n-butylammonium tetrafluoroborate (0.1 M) as
supporting electrolyte, with a Pt wire as the working electrode and
a saturated calomel electrode (SCE) as a reference. Potential data
for all of the compounds are collected in Table 1. In the cathodic
region, a ligand-centered reduction is observed at ca. 22.2 V vs.
SCE for all the compounds. In the anodic region, all the CuI
complexes show a first quasi-reversible one-electron process
followed by an irreversible oxidation. The first wave is assigned
to the oxidation of the CuI cation while the second one corresponds
to a ligand centered process. The quite high potentials observed for
the oxidation of the metal center indicate a good stabilization of
these CuI complexes. The positive shift seen for the first oxidation
of 4 when compared to 1, 5 and 6 is ascribed, at least in part, to the
electron withdrawing effect of the diphenyl ether subunits of the
POP ligand. However, it is also believed that complex 4 is less
flexible owing to a more entangled structure resulting from the
large bite angle of the rigid POP ligand. As a result, distortion
towards a square planar coordination geometry upon oxidation of
the CuI center becomes more difficult in this particular case and
the corresponding CuII complex is thus more destabilized.
The photoluminescence (PL) properties of CH2Cl2 solutions of
1, 4, 5 and 6 are summarized in Table 2. The emission is tentatively
assigned to metal-to-ligand charge transfer (MLCT) excited states5
and appears to be substantially influenced by conformational
changes from a pseudo-tetrahedral coordination geometry to a
flattened structure in the excited state that may facilitate the
formation of non-emissive penta-coordinated exciplexes as typi-
cally observed for CuI complexes.9 Hence, the lifetime of
compound 4, for which this distortion is expected to be more
difficult, is the longest and its emission quantum yield the highest.
These observations are in agreement with the differences observed
in the first oxidation potential of 4 when compared to 1, 5 and 6
(vide supra). Whereas the emission quantum yields in solutions are
˚
(.3.1 A).
Fig. 2 Left: plot of the structure of the cation in 3 (only one conformer is
shown). Right: details of the coordination sphere around the CuI cation.
Thermal ellipsoids are drawn at the 30% probability level, only the Cu and
P atoms were refined anisotropically. Selected bond lengths and bond
angles: Cu(1)–P(1): 2.279(4); Cu(1)–P(2): 2.304(3); Cu(1)–P(3): 2.333(3);
Cu(1)–P(4): 2.425(4); P(1)–Cu(1)–P(2): 110.51(13)u; P(1)–Cu(1)–P(3):
124.44(13)u; P(1)–Cu(1)–P(4): 125.43(13)u; P(2)–Cu(1)–P(3): 109.77(13)u;
P(2)–Cu(1)–P(4): 108.26(13)u; P(3)–Cu(1)–P(4): 73.28(12)u.
Fig. 3 Heteroleptic CuI complexes 4–6.
3078 | Chem. Commun., 2007, 3077–3079
This journal is ß The Royal Society of Chemistry 2007

View Article Online
Table 1 Electrochemical data of 1, 4, 5, and 6 determined by OSWV
on a Pt working electrode in CH2Cl2 + 0.1 M nBu4NBF4 at room
temperaturea
Eox1/V
Eox2/V
Ered1/V
1
4
5
6
a
+1.19
+1.50
+1.19
+1.17
+1.66
+1.64
+1.63
+1.30
22.20
22.23
22.18
22.22
OSWVs were obtained using a sweep width of 20 mV, a frequency
of 10 Hz, and a step potential of 5 mV.
Table 2 Luminescence data in CH2Cl2 (air free solutions) at 298 K
l
max/nma
Wem (%)b
t/nsc
1
4
5
6
a
556
494
546
544
1 6 1022
2
244
2440
231
Fig. 4 I–V–B characteristics of the device obtained from 1. Inset: EL
spectra of 1 at a concentration of 12.5 wt% in a PVK matrix.
2 6 1022
8 6 1022
2.5–5.8
Emission maxima from uncorrected spectra, lexc
Emission quantum yields in air free solutions. Excited state
=
330 nm.
OLLA). We further thank A. Saquet for the CV and OSWV
measurements.
b
c
lifetimes.
Notes and references
quite low, compounds 1 and 4 exhibit a bright luminescence in the
solid state at room temperature as well as in rigid frozen CH2Cl2
solutions at 77 K. Indeed, under these conditions, geometric
distortions prompting non-radiative deactivation of the MLCT
excited states10 are prevented. This is further confirmed by the
50 nm blue shift of the emission maxima when going from room
temperature to 77 K for both 1 and 4. A similar trend is observed
for 5 and 6, however the effect is less dramatic. Indeed, the less
rigid dppe and dppp ligands may facilitate some excited-state
distortion.
{ C60H48P4CuBF4 (Mr = 1043.29), monoclinic space group P 2/n, Z = 2,
˚
a = 13.5721(9), b = 12.5841(8), c = 14.9692(12) A, a = 90, b = 100.516(6),
3
˚
c = 90u, V = 2513.7(3) A , 308 parameters, 11 658 reflections measured,
5719 unique (Rint = 0.04), 3534 reflections used in the calculations
[I . 2.5s], R = 0.0509, wR = 0.0579. CCDC 645319. For crystallographic
data in CIF or other electronic format see DOI: 10.1039/b707398d
¯
§ C61.75H51.50B1Cl1.50Cu1F4O1P4 (Mr = 1137.01), triclinic space group P1,
˚
Z = 4, a = 14.6080(14), b = 19.213(2), c = 21.0316(17) A, a = 80.126(11),
3
˚
b = 80.686(11), c = 89.676(12)u, V = 5737.2(10) A , 651 parameters, 57 446
reflections measured, 20986 unique (Rint = 0.09), 6820 reflections used in
the calculations [I . 2s], R = 0.0714, wR = 0.0789. CCDC 645318. For
crystallographic data in CIF or other electronic format see DOI: 10.1039/
b707398d
Based on the PL data, complexes 1 and 4 were selected for light
emitting device fabrication. Poly(vinyl carbazole) (PVK) was
chosen as the host material because of its good hole-transport
ability, broad band-gap and overlap of the emission spectra with
the absorption spectra of 1 and 4. The light emitting devices were
fabricated by spin coating a thin film of a 1–PVK or a 4–PVK
blend (ca. 120 nm) on an ITO substrate. The concentration of 1
and 4 in the PVK matrix was 12.5% in weight. After the film had
been dried under vacuum at room temperature for 2 h, the cathode
was fabricated by thermal evaporation of an Al layer (100 nm).
The electroluminescence (EL) spectra of 1 and 4 in PVK films are
displayed in Fig. 4 and S4 respectively.{
In both cases, the EL spectra are red shifted and significantly
broader when compared to the PL spectra recorded in CH2Cl2
solutions (see ESI{). As a result, almost white light is produced by
both devices. The current–voltage–brightness (I–V–B) character-
istics (Fig. 4 and S4{) of the devices prepared from 1 and 4 indicate
both turn-on voltages of 15 V and brightnesses up to 490 (for 1)
and 330 cd m22 (for 4) at 20 V. The device efficiencies have not
been yet optimized, but they should be improved by using
appropriate hole blocking and electron transfer layers in the device
configuration. Work in this direction is currently under way in our
laboratories.
1 M. G. Harrison and R. H. Friend, in Electronic Materials: the Oligomer
Approach, ed. K. Mu¨llen and G. Wegner, Wiley-VCH, Weinheim, 1998,
pp. 515–558; Organic Light Emitting Devices, ed. K. Mu¨llen and
U. Scherf, Wiley-VCH, Weinheim, 2005.
2 For recent reviews, see: E. Holder, B. M. W. Langeveld and
U. S. Schubert, Adv. Mater., 2005, 17, 1109; P.-T. Chou and Y. Chi,
Chem.–Eur. J., 2007, 13, 380.
3 Y.-G. Ma, W.-H. Chan, X.-M. Zhou and C.-M. Che, New J. Chem.,
1999, 23, 263; Y. Ma, C.-M. Che, H.-K. Chao, X. Zhou, W. H. Chan
and J. Shen, Adv. Mater., 1999, 11, 852.
4 Q. Zhang, Q. Zhou, Y. Cheng, L. Wang, D. Ma, X. Jing and F. Wang,
Adv. Mater., 2004, 16, 432; Q. Zhang, Q. Zhou, Y. Cheng, L. Wang,
D. Ma, X. Jing and F. Wang, Adv. Funct. Mater., 2006, 16, 1203.
5 A. Tsuboyama, K. Kuge, M. Furugori, S. Okada, M. Hoshino and
K. Ueno, Inorg. Chem., 2007, 46, 1992.
6 N. Armaroli, G. Accorsi, M. Holler, O. Moudam, J.-F. Nierengarten,
Z. Zhou, R. T. Wegh and R. Welter, Adv. Mater., 2006, 18, 1313.
7 D. G. Cuttell, S. M. Kuang, P. E. Fanwick, D. R. McMillin and
R. A. Walton, J. Am. Chem. Soc., 2002, 124, 6.
8 J. R. Black, W. Levason, M. D. Spicer and M. Webster, J. Chem. Soc.,
Dalton Trans., 1993, 3129; S. Kitagawa, M. Kondo, S. Kawata,
S. Wada, M. Maekawa and M. Munakata, Inorg. Chem., 1995, 34,
1455; P. Comba, C. Katsichtis, B. Nuber and H. Pritzkow, Eur. J. Inorg.
Chem., 1999, 777; E. Szlyk, R. Kucharek, I. Szymanska and
L. Pazderski, Polyhedron, 2003, 22, 3389.
9 N. Armaroli, Chem. Soc. Rev., 2001, 30, 113 and references therein.
10 D. Felder, J. F. Nierengarten, F. Barigelletti, B. Ventura and
N. Armaroli, J. Am. Chem. Soc., 2001, 123, 6291.
This work was supported by the CNR (commessa PM.P04.010,
MACOL), the CNRS and the EU (contract n. IST-2002-004607,
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 3077–3079 | 3079