Structural and spectroscopic studies on three-coordinate complexes of copper(I) halides with tricyclohexylphosphine

Article abstract of DOI:10.1039/b111121n

Three-coordinate 1:1 and 1:2 compounds of copper(I) halides with tricyclohexylphosphine have been synthesised and characterized by single crystal X-ray structure determinations, ⁶³Cu NQR spectroscopy, solid state ³¹P CPMAS NMR spectr

Full text of DOI:10.1039/b111121n

View Article Online / Journal Homepage / Table of Contents for this issue
Structural and spectroscopic studies on three-coordinate complexes
of copper(I) halides with tricyclohexylphosphine
Graham A. Bowmaker,a Sue E. Boyd,b John V. Hanna,c Robert D. Hart,d Peter C. Healy,*b
Brian W. Skeltond and Allan H. Whited
a Department of Chemistry, University of Auckland, Private Bag 92019 Auckland,
New Zealand
b School of Science, Griffith University, Brisbane, Queensland, 4111, Australia
c ANSTO NMR Facility, Materials Division, Private Mail Bag 1, Menai, New South Wales,
2234, Australia
d Department of Chemistry, University of Western Australia, Crawley, Western Australia, 6009,
Australia
Received 5th December 2001, Accepted 22nd April 2002
First published as an Advance Article on the web 17th May 2002
Three-coordinate 1 : 1 and 1 : 2 compounds of copper() halides with tricyclohexylphosphine have been synthesised
and characterized by single crystal X-ray structure determinations, 63Cu NQR spectroscopy, solid state 31P CPMAS
NMR spectroscopy, and low frequency vibrational spectroscopy. The 1 : 1 compounds crystallize as isomorphous,
centrosymmetric [(PCy3)Cu(µ-X)2Cu(PCy3)] dimers with three-coordinate (PCuX2) copper. Solid state 31P CPMAS
NMR spectra show asymmetric quartets with the spectra of the chloride and bromide compounds broadened by
quadrupole relaxation effects. For X = I, 1J(31P–63Cu) is 1.74 kHz with the quadrupole distortion parameter, dνCu
11.5 × 109 Hz2. Far infrared spectra of the compounds are unusually complex with groups of strong bands in the
region expected for ν(CuX) vibrational modes (100–260 cmϪ1). The 1 : 2 compounds crystallize as [CuX(PCy3)2]
,
monomers. The solid state 31P NMR spectra of the bromide and iodide compounds show sharp asymmetric quartets
with 1J(P–Cu) 1.20 and 1.23 kHz and dνCu 8.7 and 10.3 × 109 Hz2 respectively. Relaxation effects collapse the
spectrum of the chloride to a broad doublet. Room temperature 63Cu NQR frequencies for [CuX(PCy3)2] are found
to be 34.5, 33.43 and 32.06 MHz for X = Cl, Br and I respectively and are of the order of 4 MHz greater than values
recorded for the 1 : 1 complexes. Far infrared spectra of the complexes exhibit strong bands due to ν(CuX)
vibrational modes at 253, 189 and 156 cmϪ1 for X = Cl, Br and I.
In our continuing investigations into the dependence of the
structural and spectroscopic properties of these compounds on
Introduction
Reactions of copper() halides with unidentate tertiary organo-
the choice of phosphine and halide, we have sought to extend
phosphine bases yield a diverse array of two-, three- and
the array of available data for three-coordinate complexes of
four-coordinate complexes with structural and spectroscopic
properties that are determined by the specific choices of
phosphine ligand and, to a lesser extent, by the choice of
halide anion. Thus, for compounds of 1 : 1 stoichiometry,
copper() halides with tricyclohexylphosphine. Single crystal
X-ray structures have been determined for the 1 : 1 adducts for
X = Br and I, and for the 1 : 2 adducts for X = Cl (two poly-
morphs) and Br. The structure of the analogous 1 : 2 nitrate
simple monomeric [CuX(PR3)] complexes with two-coordinate
compound was also redetermined. Spectroscopic data has been
(PCuX) copper are obtained for trimesitylphosphine (Pmes3)
obtained from low frequency infrared spectroscopy, solid state
(X = Br)1 and tris(2,4,6-trimethoxyphenyl)phosphine (Ptmp3)
31P CPMAS NMR spectroscopy and 63Cu NQR spectroscopy.
(X = Cl, Br, I).2,3 Dimeric [CuX(PR3)]2 complexes with bridging
The outcomes of this study, which are reported here, also com-
halides and three-coordinate (PCuX2) copper are obtained
plete a series of studies on the 1 : 1 and 1 : 2 two-, three- and
for a number of phosphines, including tricyclohexylphosphine
four-coordinate complexes of copper, silver and gold() halides
(PCy3) (X = Cl, I),4,5 tris(o-tolylphosphine (Po-tol3) (X = Cl, Br,
with tricyclohexylphosphine.30–33
I),6–8 PPh2mes, PPhmes2 (X = Cl, Br, I),9 tris(o-anisylphosphine
(Po-anis3) (X = Br)10 and tribenzylphosphine (PBn3) (X = I),11
Experimental
while the stellaquadrangula or “cubane” [CuX(PR3)]4 com-
plexes, with µ3 bridging halides and four-coordinate (PCuX3)
copper, are obtained for PEt3 (X = Cl, Br, I),12,13 PMePh2 (X =
I),14 PPh3 (X = Cl, Br, I)15–17 and PtBu3 (X = Br).18 A “step”
tetrameric structure which incorporates both three- (PCuX2)
and four- (PCuX3) coordinate copper has also been reported for
PPh3 (X = Br, I).19–21 Compounds of 1 : 2 stoichiometry,
CuX(PR3)2, crystallize either as [CuX(PR3)2]2 dimers with four-
coordinate (P2CuX2) copper for PPhH2 (X = I),22 PMePh2 (X =
I)23 and PMe3 (X = I),24 or as [CuX(PR3)2] monomers with
three-coordinate (P2CuX) copper for PPh3 (X = Cl, Br, I),25–27
PPh2o-tol (X = Cl, Br, I),28 PCy3 (X = Br, I)5,29 and PBn3 (X = Cl,
Br, I).11
Preparation of the complexes
[CuCl(PCy3)]2. CuCl (0.100 g, 1.0 mmol) and PCy3 (0.280 g,
1.0 mmol) were dissolved in 20 ml of hot dimethylformamide
(dmf ). The resultant solution was filtered and allowed to stand
overnight to give a white crystalline precipitate. (Found: C 56.9,
H 8.9; C18H33ClCuP requires C 56.98, H 8.77%), mp 240–245
ЊC (darkens 215–225 ЊC). (Lit.34 218 ЊC).
[CuBr(PCy3)]2. CuBr (0.150 g, 1.05 mmol) and PCy3 (0.280 g,
1.0 mmol) and were dissolved in 20 ml of hot dmf. The result-
ant solution was filtered and left to stand overnight to give a
2722
J. Chem. Soc., Dalton Trans., 2002, 2722–2730
This journal is © The Royal Society of Chemistry 2002
DOI: 10.1039/b111121n

View Article Online
᎐
white crystalline precipitate. (Found: C 51.1, H 8.0; C18H33-
BrCuP requires C 51.00, H 7.85%), mp 245–250 ЊC (Lit.34
237–240 ЊC). The compound was also prepared according to
published procedures34 by the addition of ethanolic solutions
of CuBr2 and PCy3 under an argon atmosphere in a 1 : 2 molar
ratio. The mixture was refluxed for four hours and the resulting
white precipitate removed by filtration. Crystals of the complex
suitable for the X-ray diffraction study were obtained by slow
evaporation of the filtrate. mp 240–245 ЊC.
Crystal data. [CuBr(PCy )] ᎐ C H Br Cu P . M = 847.8,
᎐
1
3
2
36 66
2
2
2
¯
triclinic, space group P1 (Ci , No. 2), a = 8.619(3), b = 9.321(2),
c = 13.351(4) Å, α = 101.23(2), β = 88.98(2), γ = 114.33(2)Њ, V =
956 Å3, Dc = 1.46 g cmϪ3 (Z = 1 dimer), µ = 33.1 cmϪ1. Crystal
size: 0.33 × 0.12 × 0.08 mm; A*min,max = 1.27, 1.47; 2θmax = 45Њ,
N = 2492, No = 1892; R = 0.043, Rw = 0.047.
᎐
[CuI(PCy )] ᎐ C H I Cu P . M = 941.2, triclinic, space
᎐
3
2
36 66
2
2
2
¯
group P1, a = 8.802(3), b = 9.832(4), c = 12.935(3) Å, α =
100.31(3), β = 86.07(3), γ = 116.94(2)Њ, V = 982 Å3, Dc = 1.59 g
cmϪ3 (Z = 1 dimer), µ = 27.6 cmϪ1. Crystal size: 0.45 ×
0.36 × 0.44 mm; A*min,max = 2.17, 2.77; 2θmax = 55Њ, N = 3837,
[CuI(PCy3)]2. CuI (0.190 g, 1.0 mmol) and PCy3 (0.288 g,
1.0 mmol) were dissolved in 20 ml of hot dmf. The resultant
solution was filtered and left to stand overnight to give a crys-
talline precipitate of the complex suitable for X-ray diffraction
studies. (Found: C 45.9, H 7.1; C18H33ICuP requires C 45.91,
H 6.58%), mp 226–230 ЊC (Lit.34 232–235 ЊC).
No = 2969; R = 0.043, Rw = 0.057.
᎐
α-[CuCl(PCy ) ] ᎐ C H ClCuP . M = 659.9, monoclinic,
᎐
space group C2/c (C2h6, No. 15), a =218.00(2), b = 9.113(2), c =
3
2
36 66
22.23(1) Å, β = 96.32(8)Њ, V = 3625 Å3, Z = 4, Dc = 1.21 g cmϪ3
,
=
µ = 7.8 cmϪ1. Crystal size: 0.50 × 0.35 × 0.13 mm; A*min,max
1.10, 1.30; 2θmax = 50Њ, N = 3173, No = 1983; R = 0.059,
ꢀ-[CuCl(PCy3)2]. This compound was prepared according
to published procedures,34 by the addition of an ethanolic
solutions of CuCl2ؒ2H2O and PCy3 in a 1 : 3 molar ratio under
an argon atmosphere. The mixture was refluxed for four hours
and the resulting white precipitate removed by filtration. Crys-
tals of the complex suitable for X-ray diffraction studies were
obtained by slow evaporation of the filtrate. (Found C 65.6; H
10.3; C36H66CuClP2 requires C 65.53; H 10.08%), mp 170–75 ЊC
(lit.34 170 ЊC).
Rw = 0.061.
᎐
β-[CuCl(PCy ) ] ᎐ C H ClCuP . M = 659.8, monoclinic,
᎐
space group P21/n (C2h5, No. 14), a =29.964(3), b = 15.593(4), c =
3
2
36 66
23.987(4) Å, β = 97.36(2)Њ, V = 3696 Å3, Z = 4, Dc = 1.19 g cmϪ3
,
=
µ = 7.7 cmϪ1. Crystal size: 0.35 × 0.25 × 0.15 mm; A*min,max
1.12, 1.22; 2θmax = 50Њ, N = 6780, No = 3154; R = 0.042,
Rw = 0.043.
᎐
α-[CuBr(PCy ) ] ᎐ C H BrCuP . M = 704.3, monoclinic,
᎐
3
2
36 66
2
space group C2/c, a = 18.034(7), b = 9.104(11), c = 22.281(9) Å,
β = 96.62(3)Њ, V = 3634 Å3, Z = 4, Dc = 1.29 g cmϪ3, µ =
18.1 cmϪ1. Crystal size: 0.22 × 0.16 × 0.09 mm; A*min,max = 1.19,
ꢁ-[CuCl(PCy3)2]. CuCl (0.05 g, 0.51 mmol) and PCy3 (0.30 g,
1.07 mmol) were dissolved in 10 ml of warm dmf. The resultant
solution was filtered and allowed to stand overnight at room
temperature to give a crystalline precipitate of the complex
suitable for X-ray diffraction studies. (Found: C 66,4, H 10.4%),
mp 167–174 ЊC.
1.48; 2θmax = 46Њ, N = 2517, No = 1703; R = 0.057, Rw = 0.070.
᎐
[CuNO (PCy ) ] ᎐ C H CuNO P . M = 686.4, monoclinic,
᎐
3
3
2
36 66
3
2
space group C2/c, a = 18.201(5), b = 9.306(3), c = 22.167(6) Å,
β = 96.39(2)Њ, V = 3731 Å3, Z = 4, Dc = 1.22 g cmϪ3, µ = 7.0 cmϪ1
.
Crystal size: 0.40 × 0.40 × 0.20 mm; A*min,max = 1.14, 1.26;
2θmax = 50Њ, N = 3268, No = 1997; R = 0.045, Rw = 0.047. In this
complex, as modelled in the present space group, the nitrate
peripheral NO atoms are resolved as disordered over pairs of
sites related by the 2-axis, N ؒ ؒ ؒ N 0.65(3), O ؒ ؒ ؒ O 1.26(2) Å.
CCDC reference numbers 175631–175636.
ꢀ-[CuBr(PCy3)2]. CuBr (0.066 g, 0.046 mmol) and PCy3
(0.30 g, 1.07 mmol) were dissolved in 10 ml of warm dmf. The
resultant solution was filtered and allowed to stand overnight at
room temperature to give crystals of the complex suitable for
X-ray diffraction studies. (Found: C 61.4, H 9.7; C36H66BrCuP2
requires C 61.39, H 9.45%), mp 182–185 ЊC (lit. 172–174 ЊC,34
190–192 ЊC35).
See http://www.rsc.org/suppdata/dt/b1/b111121n/ for crystal-
lographic data in CIF or other electronic format.
Spectroscopy
[CuI(PCy3)2]. CuI (0.198 g 1.04 mmol) and PCy3 (0.602 g,
2.15 mmol) were dissolved in 25 ml of warm dmf. The resultant
clear solution was filtered and allowed to stand overnight at
room temperature to give a microcrystalline precipitate of the
complex. (Found: C 57.3, H 9.0; C36H66CuIP2 requires C 57.55,
H 8.85%), mp. 180–82 ЊC (lit.35 176–77 ЊC).
31P CPMAS solid state NMR spectra were acquired at room
temperature operating at a field strength of Bo = 9.40 T and a
31P Zeeman frequency of 162.92 MHz. Conventional cross
polarisation and magic angle spinning techniques, coupled with
spin temperature alternation to eliminate spectral artifacts were
implemented using a Bruker 4 mm double air bearing probe
in which MAS frequencies > 10 kHz were achieved. A
[CuNO3(PCy3)2]. This compound was prepared according to
published procedures36 by the addition of ethanolic solutions
of CuNO3ؒ3H2O and PCy3 in a 1 : 3 molar ratio. The mixture
was refluxed for 4 h and allowed to stand overnight to give well
formed crystals of the complex suitable for X-ray diffraction
studies. mp 220–225 ЊC (lit.36 225–227 ЊC).
recycle delay of 20 s, a H–31P contact period of 10 ms and a
1
1H π/2 pulse length of 3 µs were common to all spectra. No
spectral smoothing was invoked prior to Fourier transform-
ation. Chemical shifts were referenced to 85% H3PO4 via solid
triphenylphosphine (δ Ϫ9.9).
Far-infrared spectra were recorded at 2 cmϪ1 resolution as
polythene discs on a Digilab FTS-60 Fourier transform infrared
spectrometer employing an FTS-60V vacuum optical bench
with a 5 lines/mm wire mesh beam splitter, a mercury
lamp source and a pyroelectric triglycine sulfate detector. All
measurements were carried out at room temperature.
Crystallography
Structure determinations. Unique, room temperature, four
circle diffractometer data sets (2θ/θ scan mode; monochromatic
Mo-Kα radiation, λ = 0.7107 Å, T = 295 K) were measured for
all structures. N independent reflections were obtained, No with
I > 3σ(I ) being considered ‘observed’ and used in the full mat-
rix least squares refinements after absorption correction. Aniso-
tropic thermal parameters were refined for the non-hydrogen
atoms; (x, y, z, Uiso)H were included constrained at estimated
values. Conventional residuals at convergence, R, RЈ on |F| are
quoted, statistical reflection weights being used. Neutral atom
complex scattering factors were used; computation used XTAL
3.2,37 TEXSAN38 and ORTEP-339 software. Pertinent results
are presented in Tables 1–3.
63Cu and 65Cu NQR frequencies were measured on ca. 50 mg
of powdered sample of the complex at room temperature using
a Bruker CXP console pulsing into a probe arrrangement that
was well removed from the magnet (> 5 m) and shielded from
extraneous magnetic and radio frequency interference by a
mumetal container. Solid-echo experiments were employed in
the measurements using hard pulses of 2 µs duration and
recycle delays of 0.5 s and with extended phase cycles to elimin-
ate baseline distortions and echo tails. The location of both
J. Chem. Soc., Dalton Trans., 2002, 2722–2730
2723

View Article Online
Table 1 PCuX2CuP core geometries for dimeric [PR3Cu(µ-X)2CuPR3] complexes
X
PR3
Cu–P
Cu–X
Cu–XЈ
X ؒ ؒ ؒ X
Cu ؒ ؒ ؒ Cu CuXCu
XCuX
PCuX
PCuXЈ
Ref.
Cl
Cl
Cl
Cl
Cl
Br
Br
Br
Br
Br
I
Po-tol3
PPh2mes
PPhmes2
PCy3
2.191(1)
2.196(2)
2.202(2)
2.183(2)
2.185(1)
2.208(1)
2.194(1)
2.198(3)
2.197(3)
2.191(2)
2.238(4)
2.225(5)
2.226(5)
2.201(4)
2.264(5)
2.215(2)
2.225(2)
2.289(2)
2.287(2)
2.281(2)
2.285(2)
2.302(1)
2.414(1)
2.356(1)
2.397(2)
2.393(2)
2.409(1)
2.525(2)
2.553(2)
2.544(3)
2.554(3)
2.552(4)
2.539(1)
2.562(1)
2.316(2)
2.329(3)
2.326(1)
2.332(2)
2.305(1)
2.431(1)
2.541(1)
2.441(2)
2.415(2)
2.430(1)
2.640(2)
2.569(3)
2.584(2)
2.589(3)
2.597(3)
2.563(1)
2.579(1)
3.402(2)
3.447(2)
3.442(2)
3.439(3)
3.439(1)
3.716(1)
3.807(1)
3.748(2)
3.684(2)
3.717(1)
4.116(1)
4.316(2)
3.103(1)
3.076(2)
3.086(1)
3.066(1)
3.065(1)
3.109(1)
3.085(1)
3.083(2)
3.052(3)
3.098(1)
3.122(3)
2.690(3)
84.74(6)
83.56(8)
84.08(5)
83.44(7)
83.41(4)
79.83(4)
78.01(3)
79.18(6)
78.80(7)
79.62(4)
74.34(5)
63.94(7)
95.26(5)
96.44(8)
95.9(1)
132.56(6)
135.40(8)
135.8(1)
136.05(7)
132.07(4)
130.00(6)
141.38(5)
132.2(1)
130.9(1)
132.88(6)
136.60(8)
121.9(1)
127.0(1)
128.3(1)
125.1(1)
125.43(5)
123.34(5)
131.98(6)
128.15(7)
128.0(1)
124.40(7)
131.14(4)
129.64(6)
116.53(5)
127.0(1)
128.7(1)
126.74(6)
117.66(7)
122.8(1)
118.3(1)
114.9(1)
118.3(2)
126.23(5)
124.93(5)
6
9
9
4
11
6
10
9
9
96.56(7)
96.59(4)
100.17(3)
101.99(3)
100.8(1)
100.1(1)
100.38(5)
105.66(6)
114.9(1)
114.6(1)
116.0(1)
115.8(1)
108.20(3)
111.52(3)
PBn3
Po-tol3
Po-anis3
PPh2mes
PPhmes2
PCy3
Po-tol3
PPh2mes
a
6
9
I
I
PPhmes2
4.361(3)
2.728(3)
63.49(7)
9
a
I
I
PCy3RT
PCy3LT
4.133(1)
4.498(9)
2.992(1)
2.893(2)
71.80(3)
68.48(3)
5
a This work.
Table 2 P2CuX core geometries for monomeric [CuX(PCy3)2] and [CuX(PR3)2] complexes, X = Cl, Br, I
X
PR3
Cu–P
Cu–PЈ
Cu–X
P–Cu–P
P–Cu–X
PЈ–Cu–X
Ref.
25a
Cl
Cl
Cl
Cl
Cl
Br
Br
Br
Br
PPh3
2.272(2)
2.257(2)
2.261(2)
2.261(2)
2.2399(5)
2.263(2)
2.255(2)
2.257(2)
2.2436(7)
2.2509(7)
2.273(2)
2.255(1)
2.275(1)
2.239(1)
2.260(2)
2.241(2)
2.261(2)
2.282(2)
2.2275(5)
2.282(3)
2.240(2)
2.257(2)
2.2127(8)
2.2396(7)
2.273(2)
2.255(1)
2.285(1)
2.2324(1)
2.208(2)
2.205(2)
2.243(3)
2.247(2)
2.2521(6)
2.346(2)
2.336(1)
2.392(3)
2.4034(4)
2.3995(4)
2.524(2)
2.507(1)
2.5913(7)
2.5865(8)
125.48(7)
126.96(7)
134.44(8)
134.54(6)
136.53(2)
126.0(1)
127.89(7)
135.6(1)
138.92(3)
139.86(3)
126.9(1)
113.76(7)
113.61(7)
112.78(8)
115.82(7)
105.34(2)
112.8(1)
112.52(6)
112.20(6)
100.76(2)
106.92(2)
116.61(5)
116.82(4)
114.79(3)
105.11(4)
120.74(6)
118.55(7)
112.78(8)
109.27(6)
117.76(2)
121.0(1)
118.25(6)
112.20(6)
119.83(2)
113.09(2)
116.61(5)
116.82(4)
110.99(3)
115.07(4)
PPh2o-tol
PCy3 (α)
PCy3 (β)
PBn3
28
b
b
11
PPh3
26a
PPh2o-tol
PCy3 (α)
PBn3 (i)
(ii)
28
b
11
I
I
I
I
PPh3
25
28
5
PPh2o-tol
PCy3 (γ) LT
PBn3
126.36(7)
134.06(4)
139.78(5)
11
a Hemibenzene solvate. b This work.
Table 3 P2MX core geometries for monomeric [MX(PCy3)2] complexes for M = Cu, Ag, Au and X = Cl, Br, I
MX
M–P
M–PЈ
M–X
P–M–P
P–M–X
PЈ–M–X
Ref.
a
a
a
CuCl (α)
CuCl (β)
CuBr (α)
CuI (γ)
AgCl (δ)
AgBr (δ)
AgI (δ)
2.261(2)
2.261(2)
2.257(2)
2.275(1)
2.471(1)
2.474(1)
2.478(1)
2.321(2)
2.323(2)
2.317(3)
2.302(3)
2.328(3)
2.344(2)
2.315(3)
2.261(2)
2.282(2)
2.257(2)
2.285(1)
2.471(1)
2.474(1)
2.478(1)
2.321(2)
2.323(2)
2.311(3)
2.292(3)
2.305(3)
2.328(2)
2.310(3)
2.243(3)
2.247(2)
2.392(4)
2.591(1)
2.489(1)
2.618(1)
2.778(1)
—
2.777(2)
2.894(1)
2.842(1)
3.765(2)
2.895(2)
3.008(1)
134.44(8)
134.54(6)
135.6(1)
134.06(4)
128.29(3)
129.42(3)
130.75(4)
180.0(—)
147.5(1)
162.06(9)
157.7(1)
178.4(1)
143.1(1)
159.1(1)
112.78(5)
115.82(7)
112.20(6)
114.79(3)
115.85(2)
115.29(2)
114.63(3)
—
106.27(6)
99.56(7)
101.65(8)
72.52(9)
106.26(7)
98.69(7)
112.78(5)
109.27(6)
112.20(6)
110.99(3)
115.85(2)
115.29(2)
114.63(3)
—
106.27(6)
97.69(7)
100.51(8)
109.0(1)
110.45(7)
101.87(8)
5
31
31
31
33
33
33
AuCl
AuBr (α)
AuBr (ζ1)b
AuBr (ζ2)
AuBr (ε)b
AuI (γ)
33
33
33
AuIؒPCy3
a This work. b Designated α and β respectively in ref. 33.
63Cu and 65Cu isotope resonances (related by a ratio νQ(63Cu)/
νQ(65Cu) = 1.081) verified that the true copper NQR frequencies
were being observed.
molar ratios for X = Cl, Br and from CuI and PCy3 in 1 : 1 and
1 : 2 molar ratios for X = I. Colourless crystals suitable for
X-ray studies were obtained by slow evaporation of the filtrates
obtained from these reaction mixtures. Spectroscopic data on
these samples revealed on occasion, however, co-crystallization
of both 1 : 1 and 1 : 2 complexes. We found that dissolution of a
suspension of appropriate molar ratios of CuX and PCy3 in an
aprotic solvent such as dimethylformamide, followed by cooling
to room temperature and slow evaporation of solvent, resulted
in crystallization of pure crystalline products in all cases except
Results and discussion
Synthesis
The 1 : 1 and 1 : 2 chloride and bromide complexes were initially
prepared according to literature procedures,34 by precipitation
from ethanolic solutions of CuX2 and PCy3 in 1 : 2 and 1 : 3
2724
J. Chem. Soc., Dalton Trans., 2002, 2722–2730

View Article Online
for the 1 : 2 iodide where both the 1 : 1 and 1 : 2 complexes
crystallized. Precipitation of the 1 : 1 compound was prevented
in this case by addition of a small excess of PCy3 to the reaction
mixture.
Structures
The single crystal structure determinations of this series of 1 : 1
and 1 : 2 complexes are consistent with formulation of the
complexes as centrosymmetric [(PCy3)Cu(µ-X)2Cu(PCy3)]
dimers and [CuX(PCy3)2] monomers respectively; the latter
compounds crystallizing as a number of polymorphs. The
cyclohexyl rings adopt chair conformations in all ligands with
P equatorial. One ring in each ligand (designated ring 1) lies
orthogonal to the Cu–P–C(11) plane with the P–Cu and C(11)–
H(11) bonds antiperiplanar. Rings 2 and 3 are disposed to
either side of the Cu–P–C(11) plane with the C(n1)–H(n1)
bonds (n = 2 or 3) variously (ϩ, Ϫ) synclinal, (ϩ, Ϫ) anticlinal
or orthogonal to the P–Cu bond. The conformational disposi-
tions for each ligand fall into one of two groups, designated
here A and B (Scheme 1). In the A (‘meso’) group, C(n1)–H(n1)
Fig. 1 ORTEP diagram of [CuBrPCy3]2 projected normal to the
central Cu2Br2 plane. Displacement ellipsoids are drawn at the 30%
probability level.
5Њ, with a corresponding reduction in the asymmetry of the
PCu(µ-X)2CuP molecular fragment.
(a) The geometric data listed in Table 1 for these and related
[CuX(PR3)]2 complexes show that the Cu–P bondlengths and
the Cu ؒ ؒ ؒ Cu distances are essentially the same for both the
chloride and bromide complexes with average values and
standard uncertainties for Cu–P of 2.190(8) and 2.198(6) Å and
for Cu ؒ ؒ ؒ Cu of 3.09(2) Å and 3.09(2) Å respectively. Signifi-
cant differences are observed, however, for the iodide com-
plexes, with the average Cu–P bondlength for these compounds
increasing to 2.23(2) Å and the average Cu ؒ ؒ ؒ Cu distance
decreasing to 2.9(2) Å. The spread of values for these distances
is also considerably greater than that found for either the
chloride or bromide complexes, reflecting the increased size
and polarizability of iodine by comparison with chlorine or
bromine.
(b) The iodide has also been the subject of an independent
low temperature study5 wherein it crystallized in a reduced cell
very similar to that reported here for the room temperature
structure, with the conformational dispositions of the PCy3 lig-
ands the same in both cases. Small but significant differences
are observed, however, in the Cu(µ-I)2Cu geometries. The most
noticeable differences are decreases in the Cu–I, Cu–IЈ
bondlengths and increases in the Cu–I–CuЈ angle and
Cu ؒ ؒ ؒ Cu distance with increasing temperature. In the dis-
cussion of the results of the low temperature study,5 it was
suggested that the unusually short Cu ؒ ؒ ؒ Cu distance is due to
a Cu ؒ ؒ ؒ Cu bonding interaction, and that the Cu(µ-I)2Cu
rhomb has a broad potential energy well that is easily distorted
by external influences. This latter hypothesis is supported by the
large variability observed in the values of the Cu–I and
Cu ؒ ؒ ؒ Cu distances recorded for the complexes listed in Table
1, and also the quite large increases in the Cu–I and Cu ؒ ؒ ؒ Cu
distances which occur upon the addition of a fourth ligand to
each copper atom in [(PR3)Cu(µ-X)2Cu(PR3)].6,41 Perhaps a
more fundamental indicator of the “softness” of the potential
energy surface of the Cu(µ-l)2Cu rhomb, however, is the tem-
perature dependence of the Cu(µ-I)2Cu geometry observed in
the present study, as it involves no significant variation in the
specific chemical environment of the unit. The reason for the
direction of the observed change – an increase in the Cu ؒ ؒ ؒ Cu
distance with increasing temperature – is not so obvious,
however. One possible explanation is that the interactions
which involve a small degree of orbital overlap result in a
shallow, anharmonic potential surface which is more sensi-
tive to thermal excitation. This would be the case for any
Cu ؒ ؒ ؒ Cu interaction which might be present, suggesting that
the Cu ؒ ؒ ؒ Cu distance should be sensitive to temperature
in a manner consistent with thermal occupancy of the
higher vibrational levels. This suggests that further structural
studies of [(PR3)Cu(µ-X)2Cu(PR3)] complexes over a range of
Scheme 1
(n = 2, 3) point in the same direction (either towards (Aϩ)
or away (AϪ) from ring 1). In the B group, the two C–H
bonds point in opposite directions, forming a pair of possible
enantiomers.
[CuX(PCy3)]2. The room temperature structures of the
bromide and iodide complexes are isomorphous with the
chloride, crystallizing in the triclinic space group P1 with a ≈
4
¯
8.5, b ≈ 9.5, c ≈ 13.0 Å, α ≈ 100, β ≈ 90, γ ≈ 115Њ. This lattice,
which has also been observed in the analogous silver complexes
[AgX(PCy3)]2 (X = Cl, Br)30 and [AgCl(AsCy3)]2,40 contains one
centrosymmetric dimer of the form [(ECy3)M(µ-X)2M(ECy3)].
A representative view of the dimer is shown in Fig. 1. The
PCu(µ-X)2CuP unit is planar within experimental error. The
two phosphine ligands are staggered and adopt B dispositions
of opposite chirality. Ring 1 is orthogonal to the Cu(µ-X)2Cu
plane. The detailed structural features of this dimer are found
to be influenced by the halide. For the chloride and bromide
complexes, the P–C(11) bond rotates out of the Cu(µ-X)2Cu
plane with torsion angles XЈ–Cu–P–C(11) of 13 and 14Њ
respectively. This rotation results in unsymmetrical interactions
between the halide anion and neighbouring cyclohexyl group
hydrogen atoms such that Cu–X is smaller than Cu–XЈ and
P–Cu–X is greater than P–Cu–XЈ (Table 1). This distortion is
accommodated by dislocation of the two Cy3PCuX fragments
within the PCu(µ-X)2CuP plane. In the case of the iodide com-
plex, weaker C–H ؒ ؒ ؒ I interactions and the greater covalency
of the Cu–I bonds result in a decrease of XЈ–Cu–P–C(11) to
J. Chem. Soc., Dalton Trans., 2002, 2722–2730
2725

View Article Online
temperatures may assist development of a better understanding
of the origins and nature of Cu ؒ ؒ ؒ Cu bonding interactions in
these polynuclear copper() complexes.
115, γ ≈ 95Њ). Here, the PCy3 ligands adopt Aϩ and B con-
formations respectively. Cu–P bondlengths and P–Cu–P angles
for these complexes (Table 2) are not significantly influenced by
either the choice of halide or differences in the conformational
dispositions of the phosphine ligands, with values lying
between 2.257(2) and 2.285(1) Å for Cu–P and 134.06(4) and
[CuX(PCy3)2]. The structural studies conducted on these
complexes both here and previously,5,29 show a tendency for the
complexes to crystallize in a number of polymorphic forms,
with three identified to date. The first (designated α), obtained
from ethanol for X = Cl, Br, dimethylformamide for X = Br, and
from hexane for X = Br, I,29 crystallizes in space group C2/c
with a ≈ 18.1, b ≈ 9.2, c ≈ 22.2 Å, β ≈ 96Њ. This polymorph is
found also for the nitrate complex36 (redetermined herein), the
perchlorate complex,42,43 and the analogous gold complex
[AuBr(PCy3)2].33 The molecules in this lattice are disposed
about a crystallographic two-fold axis coincident with the Cu–
X bonds so that one-half of the molecule represents the asym-
metric unit (Fig. 2(a)). The two ligands are eclipsed, and adopt
135.7(1)Њ for P–Cu–P. Comparison with data for complexes
Ϫ
with more weakly coordinating anions: NO3
, 2.245(1),
2.245(1) Å, 140.00(6)Њ (this work); ClO4Ϫ, 2.247(2), 2.247(2) Å,
144.49(9)Њ;43 and BF4Ϫ, 2.235(2), 2.237(2) Å, 159.98(8)Њ,44 how-
ever, reveal a significant increase in the P–Cu–P bond angles
with corresponding small but observable decreases in the Cu–P
bondlengths.
P2CuX geometric data available for a number of three-
coordinate [CuX(PR3)2] complexes are listed in Table 2. These
results show a greater dispersion of the Cu–P bond lengths than
found for the PCy3 complexes alone, with values ranging from
2.213(1) Å for [CuBr(PBn3)2] to 2.285(1) Å for γ-[CuI(PCy3)2].
The P–Cu–P angles divide into two distinct groups with values
ranging between 125.48(7) and 127.89(7)Њ for the PPh3 and
PPh2o-tol ligands and 134.06(4)–139.78(5)Њ for the bulkier PBn3
and PCy3 ligands, with the Cu–X bond lengths for this second
group of complexes some 0.04–0.08 Å longer than those of the
first group.
Comparison of these results with those obtained for the
analogous silver() and gold() halide complexes,31,33 show the
silver complexes also crystallize as [AgX(PCy3)2] monomers
with three-coordinate silver atoms, but in a different polymorph
(designated δ). Interestingly, these complexes crystallize in a
C2/c lattice with half the molecule forming the asymmetric unit
as found for the α complexes. However, the phosphine ligands
adopt B rather than A conformations. The gold() complexes
show a greater diversity in their structural chemistry. The chlor-
ide complex crystallizes as a two-coordinate ionic complex, the
bromide as three distinct forms (α, ζ, and ε) with both coord-
inated and semi-coordinated bromide, while the iodide crystal-
lizes as the γ polymorph with coordinated iodide. The results
of these studies yield comprehensive structural data for
[MX(PCy3)2] complexes for M = Cu, Ag and Au and X = Cl, Br
and I. Table 3 details the P2MX core geometries for these com-
plexes and shows the M–P bond lengths to increase in the order
Cu < Au < Ag, in accord with previously established trends,25,45
with average values and standard uncertainties of 2.27(2),
2.33(2) and 2.47(1) Å respectively. Comparison of these results
with the corresponding M–P bondlengths of 2.242(2), 2.353(1)
and 2.441(1) Å for the two-coordinate complexes [M(Pmes3)2]-
BF4,45 reveal a small increase of the order of 0.03 Å on
coordination of halide to the copper and silver complexes, but a
decrease of ca. 0.02 Å for the gold complexes, reflecting the
strong preference for the formation of only very weak bonds
with a third ligand in the case of gold() complexes.
Fig.
2
Representative views of the molecular structures of (a)
α-[CuCl(PCy3)2] and (b) β-[CuCl(PCy3)2].
Copper NQR spectroscopy
63Cu nuclear quadrupolar resonance frequencies were measured
at room temperature for the 1 : 2 complexes [CuX(PCy3)2] for
X = Cl, Br, I as well as for the analogous copper iodide tri-
phenylphosphine complex [CuI(PPh3)2] in order to complete
the availability of data for both series of complexes. Literature
data are available for the resonant frequencies of the 1 : 1
[CuX(PCy3)]2 complexes for X = Cl and I.8 These results,
together with room temperature (298 K) data taken or derived
from literature results for CuX2Ϫ, PCuX2, PCuX2Ϫ and P2CuX
complexes are listed in Table 4. For 63Cu (nuclear spin I =
3/2) there is a single NQR frequency ν(1/2 ↔ 3/2) whose
value is determined by the nuclear quadrupole coupling con-
stant e2qQ/h (= χCu) and the electric field gradient (EFG)
asymmetry parameter η (0 ≤ η ≤ 1 by definition) according to
eqn. (1):
Aϩ dispositions. The Cu–X bond is anticlinal to P–C(11). Ring
2 on each ligand lies almost coplanar with the P2CuX plane,
forming a pocket around the anion. The second copper poly-
morph (designated β), obtained from dmf for X = Cl, and from
benzene for X = Br,29 crystallizes in space group P21/n with a ≈
10.0, b ≈ 15.5, c ≈ 24.0 Å, β ≈ 95Њ. Here, the entire molecule
constitutes the asymmetric unit (Fig. 2(b)). The conformational
dispositions of the two ligands lie between eclipsed and stag-
gered with respect to each other, with the Cu–X bond cis to
P(1)–C(111) and synclinal to P(2)–C(211). Both ligands adopt
B conformations of the same chirality. The third copper poly-
morph (designated γ) has been obtained in a low temperature
study5 of the iodide complex obtained from THF–pentane and
is isomorphous with the azide complex,35 crystallizing in the
ν(1/2 ↔ 3/2) = (e2qQ/2h)(1 ϩ η2/3)1/2
(1)
¯
triclinic space group P1 with a ≈ 9.5, b ≈ 23, c ≈ 9Å, α ≈ 95, β ≈
2726
J. Chem. Soc., Dalton Trans., 2002, 2722–2730

View Article Online
Table 4 63Cu NQR frequencies (MHz, 298 K) for CuX2Ϫ, PCuX2,
Solid state 31P CPMAS NMR spectroscopy
PCuX2Ϫ and P2CuX complexes
Solid state 31P CPMAS NMR spectra recorded on the 1 : 1
and 1 : 2 complexes are presented in Figs. 3 and 4 respectively.
Chemical shift and coupling constant data are listed in Table 5.
The spectra of copper() phosphine complexes recorded under
high field (B0 տ 5 T) and slow 63,65Cu quadrupolar relaxation
conditions yield signals for each crystallographically independ-
ent phosphine ligand which are split into pairs of closely spaced
asymmetric quartets as a result of the quadrupole perturbed
Type
Compound
Cl
Br
I
Ref.
Ϫ
CuX2
PCuX2
[NBu4][CuX2]
[(CuX)2(PPh3)3]
[CuX(PPh3)]4
30.05
29.95
—
30.14
30.79
29.18
33.89
—
28.25
28.36
28.09
—
29.34
27.75
31.88
31.15
31.81
32.16
33.43
—
49
47
50a
8
27.35
26.09
28.17
—
—
—
28.54
—
31.17
32.06
[CuX(PCy3)]2
[CuX(Po-tol3)]2
[CuX(Po-anis)3]2
[NEt4][PPh3CuX2]
[NPr4][PPh3CuX2]
[PEt4][PPh3CuX2]
[CuX(PPh3)2]b
[CuX(PCy3)2]
8
10
51
51
51
Ϫ
PCuX2
P2CuX
34.57
33.17b
34.5
50c
c
a Chloroform solvates. b Hemibenzene solvates for X = Cl, Br. c This
work.
From the single observed pure quadrupole resonance
frequency it is not possible to determine both χCu and η; single
crystal Zeeman studies are required to determine these par-
ameters separately. Such studies have been reported for the
P2CuX site in [CuBr(PPh3)2] and for the trigonal PCuX2 sites in
[(PPh3)CuX2Cu(PPh3)2] (X = Cl, Br, I).46,47 which show the
asymmetry parameters to be about 0.1 and 0.3 respectively for
these two types of site. It is clear from eqn. (1) that η values
of this magnitude do not affect the value of the resonance
frequency to a large extent, and that changes in the reson-
ance frequency are therefore largely a reflection of changes in
the coupling constant. We assume that this is also true for the
P2CuX sites and the other compounds listed in Table 4.
For all of the complexes listed in Table 4, the frequencies of
the observed signals follow a well-established trend, decreasing
monotonically from the chlorides to the iodides. The likely
cause of this effect is the influence of the halogen atoms on the
electric field gradient at the copper nucleus. Electronic structure
calculations have shown that the increase in the 63Cu resonance
frequency from CuBr2Ϫ to CuCl2Ϫ is due to a contraction of the
Cu 4p orbital involved in the bonding, and that the magnitude
of this effect varies with the Cu–X bond length.48 It is reason-
able to assume that the same explanation applies to the other
compounds in Table 4. This effect is greater if there are two
halide ligands present (PCuX2, CuX2Ϫ, PCuX2Ϫ) than if there is
only one (P2CuX), which is consistent with the calculated effect
Fig. 3 Solid state 31P CPMAS NMR spectra of the 1 : 1 complexes
(a) [CuCl(PCy3)]2, (b) [CuBr(PCy3)]2 and (c) [CuI(PCy3)]2.
of halide on the copper p orbital field gradient in CuX and
Ϫ
CuX2
.
Comparison of the results in Table 4 for the monomeric
Ϫ
PCuX2 and P2CuX complexes shows that the copper reson-
Ϫ
ance frequencies increase on going from the PCuX2 to the
P2CuX environments for the bromide and iodide compounds.
This can be ascribed to be a consequence of replacing a weak
σ-donor (the halide) with a stronger one (the phosphine). The
increase in the amount of charge transferred to the Cu 4p
orbitals causes an increase in the copper quadrupole coupling
constant and hence the resonance frequency. For the chloride
compounds, this effect appears to be counteracted by the effect
of decreasing the number of chloride ligands from two to one
(see discussion above), so that the frequencies for these two
environments are almost the same in this case. Also evident in
the results in Table 5 is a decrease in the resonance frequency
for the PCuX2 environment (involving bridging halide ligands
Ϫ
X) relative to the PCuX2 environment (involving terminal
halide). This is attributed to the fact that a bridging halide does
not donate as much charge to the Cu atom as a terminal halide
does, since the bridging halide is shared by two Cu atoms.
Within the P2CuX complexes, the observed increase in the
resonance frequencies of the PCy3 complexes compared to
those for the PPh3 complexes can be ascribed to the increased
σ donor capacity of the more basic PCy3 ligand and to the
increase in the P–Cu–P angle.
Fig. 4 Solid state 31P CPMAS NMR spectra of the 1 : 2 complexes:
(a) [CuCl(PCy3)2], (b) [CuBr(PCy3)2] and (c) [CuI(PCy3)2].
J. Chem. Soc., Dalton Trans., 2002, 2722–2730
2727

View Article Online
Table 5 31P CP-MAS NMR data for dimeric [CuX(PR3)]2 and monomeric [CuX(PR3)2] complexes
Type
X
PR3
Bo
δiso a/ppm
∆21/kHz
∆32/kHz
∆43/kHz
J/kHz
dνCu b/109 Hz2
Ref.
[CuX(PR3)]2
Cl
Cl
Br
Br
I
I
I
Po-tol3
PPhmes2
Po-tol3
PPhmes2
PCy3
Po-tol3
9.40
9.40
9.40
9.40
9.40
9.40
9.40
7.02
Ϫ19.6
1.63
1.64
1.58
1.62
1.45
1.45
1.43
0.90
0.95
1.15
1.00
1.07
1.00
1.05
1.30
1.03
1.09
1.00
0.88
1.09
1.97
1.99
1.90
1.93
1.81
1.73
1.71
1.25
1.30
1.43
1.27
1.16
1.15
1.30
1.46
1.26
1.39
1.26
1.20
1.37
2.12
2.15
2.05
2.05
1.91
1.88
1.84
1.40
1.60
1.64
1.37
1.40
1.40
1.55
1.60
1.50
1.58
1.39
1.43
1.56
1.93
1.94
1.86
1.88
1.74
1.70
1.67
1.20
1.29
1.41
1.23
1.20
1.17
1.30
1.46
1.26
1.36
1.23
1.17
1.34
12.9
13.5
12.5
11.4
11.5
11.7
10.6
9.9
12.9
13.0
9.8
8.7
8.0
9.9
8.0
12.4
13.0
10.3
10.9
12.4
41
41
41
Ϫ20.6
Ϫ17.5
Ϫ21.2
8.4
Ϫ15.2
Ϫ15.4
6
41
c
41
41
28
PPhmes2
PPh2o-tol
[CuX(PR3)2]
Cl
1
Cl
PBn3
9.40
Ϫ3.1
Ϫ13.4
5.2
11
c
Br
Br
PCy3
PPh2o-tol
9.40
7.02
7
1
28
Br
PBn3
9.40
Ϫ3.6
Ϫ8.1
Ϫ5.7
4.7
Ϫ5
Ϫ3.9
11
c
I
I
I
PCy3
PPh3
PBn3
9.40
7.02
9.40
25
11
a δiso is the average chemical shift with respect to 85% H3PO4 of the centre points of the four lines of the quartet. b The estimated error in
determination of dνCu is 1.0 × 109 Hz2. c This work.
scalar (J) and dipolar coupling between the spin 3/2 63Cu
(γ = 7.0965 × 107 rad TϪ1 Ϫ1, natural abundance 69.09%) and
shift of peaks 2 and 4 by a correction term of magnitude d1 such
that ∆21 = J Ϫ 2d Ϫ 2d1, ∆32 = J ϩ 2d1 and ∆43 = J ϩ 2d Ϫ 2d1;
from which d = (∆43 Ϫ ∆21)/4 as before, d1 = (∆32 Ϫ J)/2) while
J = (∆21 ϩ 2∆32 ϩ ∆43)/4. Multiplication of d by the 63Cu
Zeeman frequency yields a field independent parameter dνCu
that is suitable for comparison of results obtained from spectra
recorded at different field strengths. The first order theory
s
65Cu (γ = 7.6018 × 107 rad TϪ1 sϪ1, natural abundance 30.91%)
and the spin 1/2 31P nuclei.52–55 The line widths of the peaks are
such that the two quartets are not generally resolved and the
observed spectra are dominated by the 63Cu spectrum, with the
65Cu spectrum identified where the linewidth is sufficiently
narrow as a splitting of the outer peaks of the quartet. Under
fast 63,65Cu quadrupole relaxation conditions, the lines broaden
with the inner two moving outward and the outer two moving
inwards to afford a broad doublet. Under very fast relaxation
conditions, a single peak is observed.56 In this present study,
a well defined quartet for the 1 : 1 complexes was observed for
the 1 : 1 [CuI(PCy3)]2 complex, while in the spectra of [CuCl-
(PCy3)]2 and [CuBr(PCy3)]2, relaxation effects cause both a
broadening and shifting of the lines of the quartet (Fig. 3).
Single quartets were observed for the 1 : 2 iodide and bromide
complexes. These are assigned to the α form of the complex on
the basis that this is the only polymorph that contains just one
crystallographically independent ligand in the unit cell. The
presence of a broad weak doublet can also be identified in the
spectrum of the bromide complex, indicating the likely presence
of more than one polymorph in the sample used. Strong relax-
ation effects are apparent in the spectrum of chloride, which
consists of a broad doublet (Fig. 4). It is interesting to note that
these quadrupole relaxation effects are even stronger in the case
of the analogous nitrate complex, [CuNO3(PCy3)2], the spec-
trum of which is found to consist of a single broad peak centred
at δ 0.5 ppm with ν1/2 of 1000 Hz. The increasing influence of
relaxation effects on the spectra of these complexes along the
sequence I < Br < Cl < NO3 suggests a possible correlation with
increasing ionic character of the copper–anion bond. We are
exploring this phenomenon further through a study of the
structural, NQR, and solid state 31P NMR properties of this
system with a range of semi-and non-coordinating anions.
The asymmetric spacing between the lines of each quartet
arises from perturbation of the simple J spectrum by dipolar
interactions between the phosphorus and the quadrupolar
copper nuclei. First order analysis57 predicts the outer two lines
to shift upfield and the inner two lines (δ2, δ3) downfield by a
magnitude d such that (∆32 Ϫ ∆21) = (∆43 Ϫ ∆32) = 2d, from
which d = (∆43 Ϫ ∆21)/4. Higher order quadrupole effects result
in the line spacings deviating from this first order model with
(∆32 Ϫ ∆21) > 2d > (∆43 Ϫ ∆32). These effects are generally small
and manifest as upfield shifts of peaks 1 and 3 and a downfield
generates eqn. (2) for dνCu
:
dνCu = (3χCuDeff/20)(3cos2βD Ϫ 1 ϩ ηsin2βDcos2αD) (2)
where χCu = e2qQ/h is the 63Cu quadrupolar coupling constant,
Deff = (D Ϫ ∆J/3) (D = (µo/4π)γPγCuh/4π2r3 is the Cu–P dipolar
coupling constant and ∆J is the anisotropy in the J tensor), η is
the asymmetry parameter of the electric field gradient (EFG)
tensor, and αD, βD are the polar angles defining the direction of
the Cu–P internuclear vector with respect to the principal axial
system (PAS) of the EFG tensor.
The value of 1.74 kHz for 1J(31P–63Cu) (J) in [CuI(PCy3)]2 is
in accord with the values obtained for the other PCuX2 com-
plexes listed in Table 5 and supports the general trends reported
previously where the magnitude of J decreases along the series
Cl > Br > I. The values of 1.20 and 1.23 kHz respectively for
1 : 2 bromide and iodide complexes are also in accord with the
values listed in Table 5 for P2CuX complexes; the decrease in
magnitude of ca. 30% by comparison with the PCuX2 com-
plexes reflecting the increase in the number of coordinated
phosphorus atoms from one to two.
For [CuI(PCy3)]2, dνCu is 11.5 × 109 Hz2 which is of the same
order of magnitude as values reported for the other [CuI(PR3)]2
complexes listed in Table 5. These data also show a discernable
trend towards increasing dνCu along the sequence I < Br < Cl,
paralleling the trend observed for the copper NQR frequencies.
From eqn. (2) it can be seen that dνCu is dependent on a number
of geometric and molecular parameters, in addition to the
quadrupole coupling constant. However, the similar trends
observed for both the NQR and NMR results here suggest that
these factors are not significantly influenced by changes in the
halide.
For the P2CuX complexes, dνCu varies over a much wider
range of values (8–13 × 109 Hz2) compared to the PCuX2 com-
plexes, reflecting the increased complexity of the spectra arising
from contributions to the spectra from two crystallographically
independent phosphine ligands rather than one. There are
no obvious halide dependent trends in these results. This is
2728
J. Chem. Soc., Dalton Trans., 2002, 2722–2730

View Article Online
reflected also in the values of 8.7 and 10.3 × 109 Hz2 recorded
for the present bromide and iodide complexes. Consideration of
the results obtained for [CuI(PCy3)]2 and [CuI(PCy3)2], the
spectra of which each consisting of single well defined quartet,
shows that dνCu for [CuI(PCy3)2] is less than that observed for
the [CuI(PCy3)]2, despite the larger quadrupole resonance fre-
quency (Table 4). It seems likely in this case, that the expected
increase in dνCu, arising from an increase in the quadrupole
coupling constant, is counteracted by decreases in the magni-
tude of some of the other structural and molecular par-
ameters contained in eqn. (2); in particular, changes in the
orientation of the principal axial system of the EFG with
respect to the Cu–P vectors,57 and a decrease in the asymmetry
parameter, η.46,47
Infrared spectroscopy
The far-IR spectra of the halide complexes are shown in Figs. 5
and 6. Uncomplexed PCy3 shows no bands of significant inten-
sity below 370 cmϪ1. The spectra of the complexes show bands
above 370 cmϪ1 that are almost identical in wavenumber to
those in PCy3, and these are assigned to the coordinated phos-
phine. The spectra of the 1 : 1 complexes (Fig. 5) show halogen-
sensitive bands in the region below 300 cmϪ1. The spectrum of
the chloride compound is unexpectedly complex in this region.
Fig. 6 Far infrared spectra of the 1 : 2 complexes: (a) [CuCl(PCy3)2],
(b) [CuBr(PCy3)2] and (c) [CuI(PCy3)2]. The ν(CuX) bands are labelled
with their wavenumbers.
The far-IR spectrum of the corresponding silver() complex
shows only two strong bands in this region, at 225 and 143
cmϪ1, and these are readily assigned to the two ν(MX) modes
that are expected for the C2h M2X2 core that is present in the
structure.30 The CuCl complex shows two groups of bands
centred at 240 and 190 cmϪ1, and the increase in frequency of
these relative to the two bands in the AgCl complex supports
the assignment of these as ν(CuCl) modes. The splitting of
these bands should not be due to factor group effects, as there
is only one dimer in the crystallographic unit cell. Similar
unexplained splittings of the ν(AgBr) modes in the correspond-
ing AgBr complex have been noted previously.30 The far-IR
spectra of the CuBr and CuI complexes, while simpler than that
of the CuCl complex, nevertheless show more than the expected
two ν(CuX) bands (Fig. 5). Three bands are observed in the
100–220 cmϪ1 region, which probably consist mainly of ν(CuX)
vibrations. Comparison of the present far-IR spectra for
6
[CuX(PCy3)]2 with those for [CuXPo-tol3]2 reveals a remark-
able similarity in the ν(CuX) region, and suggests that not all of
the bands with ν(CuX) character were identified in the latter
case. In particular, three bands at very similar positions in the
range 100–220 cmϪ1 are observed for the bromide and the
iodide complexes, suggesting that the IR band patterns arising
from vibrations of the Cu2X2 core are fairly characteristic for
this structure.
By analogy with the corresponding silver complexes,30 strong
bands in the spectra of the 1 : 2 complexes due to ν(CuX), the
stretching mode of the terminal Cu–X bonds, are expected in
the far-IR region. We have shown previously that a relationship
exists between the ν(CuX) wavenumber and the bond length
r(CuX) for a wide range of copper() halide complexes contain-
ing a terminal Cu–X bond and from one to three phosphine or
amine ligands. This relationship has the form given in eqn. (3)
below:
Fig. 5 Far infrared spectra of the 1 : 1 complexes: (a) [CuCl(PCy3)]2,
(b) [CuBr(PCy3)]2 and (c) [CuI(PCy3)]2. The ν(CuX) bands are labelled
with their wavenumbers.
ν/cmϪ1 = b(r/Å)Ϫm
(3)
J. Chem. Soc., Dalton Trans., 2002, 2722–2730
2729

View Article Online
21 J. C. Dyason, L. M. Engelhardt, C. Pakawatchai, P. C. Healy and
A. H. White, Aust. J. Chem., 1985, 38, 1243.
22 R. J. Batchelor, T. Birchall and R. Faggiani, Can. J. Chem., 1985, 63,
928.
with b = 13800, 18000, 32300; m = 4.9, 5.2, 5.6.58 Using the
values of the Cu–X bond lengths for [CuX(PCy3)2] in Table 2,
this equation yields ν(CuX) = 265, 192, 156 cmϪ1 for X = Cl, Br,
I respectively. This permits the assignments ν(CuX) = 253, 189,
156 cmϪ1 for X = Cl, Br, I from the experimental spectra (Fig.
6). The spectrum of the chloride has been reported previously
and a band at 255 cmϪ1 was assigned to ν(CuCl), but no
assignment was reported for the bromide.34 The spectrum of the
iodide is more complex than those of the other two halides;
bands at 185 and 131 cmϪ1, whose intensities are comparable to
that of the ν(CuI) band, are also present.
23 P. G. Eller, G. J. Kubas and R. R. Ryan, Inorg. Chem., 1977, 16,
2454.
24 G. A. Bowmaker, R. D. Hart, B. E. Jones, B. W. Skelton and
A. H. White, J. Chem. Soc., Dalton Trans., 1995, 3063.
25 G. A. Bowmaker, J. C. Dyason, P. C. Healy, L. M. Engelhardt,
C. Pakawatchai and A. H. White, J. Chem. Soc., Dalton Trans., 1987,
1089.
26 P. H. Davis, R. L. Belford and I. C. Paul, Inorg. Chem., 1973, 12, 213.
27 T. Krauter and B. Neumuller, Polyhedron, 1996, 15, 2851.
28 G. A. Bowmaker, L. M. Engelhardt, P. C. Healy, J. D. Kildea,
R. I. Papasergio and A. H. White, Inorg. Chem., 1987, 26, 3533.
29 Z. S. Seddigi, P. Durand and E. M. Holt, 217th ACS National
Meeting, Anaheim, CA, 1999, p. 532.
Conclusions
This present study shows that the reaction of copper() halides
with tricyclohexylphosphine in aprotic solvents yields well
defined air stable crystalline samples of 1 : 1 dimeric and 1 : 2
monomeric complexes with three-coordinate PCuX2 and
P2CuX copper environments respectively. The structural and
spectroscopic data recorded for these complexes, together with
those for the analogous silver and gold complexes, reflect the
significant differences in the steric and electronic properties of
the metal cation and the the halide anion.
30 G. A. Bowmaker, Effendy, P. J. Harvey, P. C. Healy, B. W. Skelton
and A. H. White, J. Chem. Soc., Dalton Trans., 1996, 2459.
31 G. A. Bowmaker, Effendy, P. J. Harvey, P. C. Healy, B. W. Skelton
and A. H. White, J. Chem. Soc., Dalton Trans., 1996, 2449.
32 R. C. Bott, G. A. Bowmaker, R. W. Buckley, P. C. Healy and
M. C. Senake Perera, Aust. J. Chem., 1999, 52, 271.
33 G. A. Bowmaker, C. L. Brown, R. D. Hart, P. C. Healy, C. E. F.
Rickard and A. H. White, J. Chem. Soc., Dalton Trans., 1999, 881.
34 F. G. Moers and P. H. Op Het Veld, J. Inorg. Nucl. Chem., 1970, 22,
3225.
35 J. Green, E. Sinn and S. Woodward, Inorg. Chim. Acta, 1995, 230,
231.
Acknowledgements
36 W. A. Anderson, A. J. Carty, G. J. Palenik and G. Schreiber, Can. J.
Chem., 1971, 49, 761.
37 S. R. Hall, H. D. Flack and J. M. Stewart, The Xtal 3.2 Reference
Manual, Universities of Western Australia, Geneva and Maryland,
1992.
38 teXsan, Single Crystal Structure Analysis Software, version 1.8,
Molecular Structure Corporation, 9009 New Trails Drive, The
Woodlands, TX, 77381, USA, 1997.
39 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
40 G. A. Bowmaker, Effendy, P. C. Junk and A. H. White, J. Chem.
Soc., Dalton Trans., 1998, 2131.
We acknowledge support of this work by grants from the
Australian Research Grants Scheme and the University of
Auckland Research Committee.
References
1 E. C. Alyea, G. Ferguson, J. Malito and B. L. Ruhl, Inorg. Chem.,
1985, 24, 3720.
2 G. A. Bowmaker, J. D. Cotton, P. C. Healy, J. D. Kildea, S. B. Silong,
B. W. Skelton and A. H. White, Inorg. Chem., 1989, 28, 1462.
3 L.-J. Baker, G. A. Bowmaker, R. D. Hart, P. J. Harvey, P. C. Healy
and A. H. White, Inorg. Chem., 1994, 33, 3925.
4 M. R. Churchill and F. J. Rotella, Inorg. Chem., 1979, 18, 166.
5 G. L. Soloveichik, O. Eisenstein, J. T. Poulton, W. E. Streib,
J. C. Huffman and K. G. Caulton, Inorg. Chem., 1992, 31, 3306.
6 G. A. Bowmaker, J. V. Hanna, R. D. Hart, P. C. Healy and A. H.
White, Aust. J. Chem., 1994, 47, 25.
41 G. A. Bowmaker, J. V. Hanna, R. D. Hart, P. C. Healy and A. H.
White, J. Chem. Soc., Dalton Trans., 1994, 2621.
42 R. J. Restivo, A. Costin and G. Ferguson, Can. J. Chem., 1975, 53,
1949.
43 C.-M. Che, Z. Mao, V. M. Miskowski, M.-C. Tse, C.-K. Chan,
K.-K. Cheung, D. L. Philips and K.-H. Leung, Angew. Chem., Int.
Ed., 2000, 39, 4084.
44 J. Green, E. Sinn, S. Woodward and R. Butcher, Polyhedron, 1993,
12, 991.
7 S. K. Hadjikakou, P. D. Akrivos, P. Karagiannidis, E. Raptopoulou
and A. Terzis, Inorg. Chim. Acta, 1993, 210, 27.
8 S. Ramaprabhu, N. Amstutz, E. A. C. Lucken and G. Bernardinelli,
J. Chem. Soc., Dalton Trans., 1993, 871.
45 A. Bayler, A. Schier, G. A. Bowmaker and H. Schmidbaur, J. Am.
Chem. Soc., 1996, 118, 7006.
46 T. Okuda, M. Hiura and H. Negita, Bull. Chem. Soc. Jpn., 1981, 54,
1920.
9 G. A. Bowmaker, D. Camp, R. D. Hart, P. C. Healy, B. W. Skelton
and A. H. White, Aust. J. Chem., 1992, 45, 1155.
10 S. Ramaprabhu, N. Amstutz, E. A. C. Lucken and G. Bernardinelli,
Z. Naturforsch., Teil A, 1998, 53, 625.
47 H. Negita, M. Hiura, K. Yamada and T. Okuda, J. Mol. Struct.,
1980, 58, 205.
48 G. A. Bowmaker, P. D. Boyd and R. J. Sorrenson, J. Chem. Soc.,
Faraday Trans., 1985, 81, 1627.
11 E. W. Ainscough, A. M. Brodie, A. K. Burrell, G. H. Freeman,
G. B. Jameson, G. A. Bowmaker, J. V. Hanna and P. C. Healy,
J. Chem. Soc., Dalton Trans., 2000, 144.
12 M. R. Churchill, B. G. DeBoer and S. J. Mendak, Inorg. Chem.,
1975, 14, 2041.
13 M. R. Churchill and K. L. Kalra, Inorg. Chem., 1974, 13, 1899.
14 M. R. Churchill and F. J. Rotella, Inorg. Chem., 1977, 16, 3267.
15 P. F. Barron, J. C. Dyason, L. M. Engelhardt, P. C. Healy and
A. H. White, Inorg. Chem., 1984, 23, 3766.
49 G. A. Bowmaker, L. D. Brockliss and R. Whiting, Aust. J. Chem.,
1973, 26, 29.
50 T. Okuda, M. Hiura, K. Yamada and H. Negita, Chem. Lett., 1977,
367.
51 S. Ramaprabhu, N. Amstutz and E. A. C. Lucken, Z. Naturforsch.,
Teil A, 1993, 49, 199.
52 E. M. Menger and W. S. Veeman, J. Magn. Reson., 1982, 46, 257.
53 R. K. Harris and A. C. Olivieri, Prog. NMR Spectrosc., 1992, 24,
435.
16 M. R. Churchill and K. L. Kalra, Inorg. Chem., 1974, 13, 1065.
17 J. C. Dyason, P. C. Healy, L. M. Engelhardt, C. Pakawatchai,
V. A. Patrick, C. L. Raston and A. H. White, J. Chem. Soc., Dalton
Trans., 1985, 831.
18 R. G. Goel and A. L. Beauchamp, Inorg. Chem., 1983, 22, 395.
19 M. R. Churchill and K. L. Kalra, Inorg. Chem., 1974, 13, 1427.
20 M. R. Churchill, B. G. DeBoer and D. J. Donovan, Inorg. Chem.,
1975, 14, 617.
54 A. C. Olivieri, J. Magn. Reson., 1989, 81, 201.
55 J. V. Hanna, R. D. Hart, P. C. Healy, B. W. Skelton and A. H. White,
J. Chem. Soc., Dalton Trans., 1998, 2321.
56 J. Bacon, R. J. Gillespie and J. W. Quail, Can. J. Chem., 1963, 41,
3063.
57 A. Olivieri, J. Am. Chem. Soc., 1992, 114, 5758.
58 G. A. Bowmaker, P. C. Healy, J. D. Kildea and A. H. White,
Spectrochim. Acta: Part A, 1988, 44, 1219.
2730
J. Chem. Soc., Dalton Trans., 2002, 2722–2730