Structural and solid state31P NMR studies of the four-coordinate copper(I) complexes [Cu(PPh3)3X] and [Cu(PPh 3)3(CH3CN)]X

Article abstract of DOI:10.1039/b505200a

The tris(triphenylphosphine)copper(I) complexes [(PPh₃) ₃CuX] for X = Cl (1), Br (2), I (3), ClO₄ (4), BF ₄ (5), [(PPh₃)₃CuCl]·CH₃CN (1a), [Cu(PPh₃)₃(CH₃CN)]X for X = ClO ₄ (6), BF₄ (7), and [Cu(PPh₃) ₃(CH₃CN)]X·CH₃CN for X = SiF₅ (8), PF₆ (9) have been studied by solid state ³¹P CP/MAS NMR spectroscopy together with single crystal X-ray diffraction for compounds (6)-(9), the latter completing the availability of crystal structure data for the series. Compounds (1)-(5) form an isomorphous series in space group P3 (a ～ 19, c ～ 11 A) with three independent molecules in the unit cell, all disposed about 3-fold symmetry axes. Average values (with estimated standard deviations) for the P-Cu-P, P-Cu-X bond angles and Cu-P bond lengths in compounds (1)-(3) are 110.1(6)°, 108.8(6)° and 2.354(8) A and 115.2(6)°, 102.8(9)° and 2.306(9) A for compounds (4) and (5). For the acetonitrile solvated compound (1a), the corresponding parameters are 115(4)°,103(3)° and 2.309(3) A. The solid state ³¹P CP/MAS NMR quadrupole distortion parameters, dv_Cu, for (1)-(3) and (1a) are all less than 1 × 10⁹ Hz², despite the changes in donor properties of the halide in (1)-(3), and the coordination geometry of the P₃CuX core in (1a). Change of anion to ClO ₄^- and BF₄^- in compounds (4) and (5) results in a significant increase of dv_Cu to 4.4-5.2 10⁹ Hz² and 5.2-6.0 × 10⁹ Hz², respectively. Compounds (6) and (7) crystallise as isomorphous [Cu(PPh₃) ₃(CH₃CN)]X salts in space group Pbca, (a ～ 17.6, b ～22.3, c ～24.2 A), while compounds (8) and (9) crystallize as isomorphous acetonitrile solvated salts [Cu(PPh₃)₃(CH ₃CN)]X·CH₃CN in space group P1 (a ～ 10.5, b ～ 13.0, c ～19.5 A, a ～ 104, β ～ 104, γ ～ 94°). The P₃CuN angular geometries in all four compounds are distorted from tetrahedral symmetry with average P-Cu-P, P-Cu-N angles and Cu-P bond lengths of 115(4)°, 103(4)° and 2.32(1) A, with dv _Cu ranging between 1.3 and 2.5 × 10⁹ Hz². The solid state ²⁹Si CP/MAS NMR spectrum of the pentafluorosilicate anion in compound (8) is also reported, affording ¹J(²⁹Si, ¹⁹F) = 146 Hz. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b505200a

View Article Online / Journal Homepage / Table of Contents for this issue
F U L L P A P E R
Structural and solid state 31P NMR studies of the four-coordinate
copper(I) complexes [Cu(PPh3)3X] and [Cu(PPh3)3(CH3CN)]X
John V. Hanna,*a Sue E. Boyd,b Peter C. Healy,*b Graham A. Bowmaker,c Brian W. Skeltond
and Allan H. Whited
a ANSTO NMR Facility, c/-Materials Science & Engineering, Private Mail Bag 1, Menai,
Australia 2234
b School of Science, Griffith University, Brisbane, Australia 4111
c Department of Chemistry, University of Auckland, Private Bag 92019, Auckland, New Zealand
d Chemistry M313, University of Western Australia, Crawley, Australia 6009
Received 13th April 2005, Accepted 7th June 2005
First published as an Advance Article on the web 27th June 2005
The tris(triphenylphosphine)copper(I) complexes [(PPh3)3CuX] for X = Cl (1), Br (2), I (3), ClO4 (4), BF4 (5),
[(PPh3)3CuCl]·CH3CN (1a), [Cu(PPh3)3(CH3CN)]X for X = ClO4 (6), BF4 (7), and [Cu(PPh3)3(CH3CN)]X·CH3CN
for X = SiF5 (8), PF6 (9) have been studied by solid state 31P CP/MAS NMR spectroscopy together with single
crystal X-ray diffraction for compounds (6)–(9), the latter completing the availability of crystal structure data for the
˚
series. Compounds (1)–(5) form an isomorphous series in space group P3 (a ∼ 19, c ∼ 11 A) with three independent
molecules in the unit cell, all disposed about 3-fold symmetry axes. Average values (with estimated standard
deviations) for the P–Cu–P, P–Cu–X bond angle◦s and Cu–P bond lengths in compounds (1)–(3) are 110.1(6)◦,
◦
◦
˚
˚
108.8(6) and 2.354(8) A and 115.2(6) , 102.8(9) and 2.306(9) A f◦or compounds (4) and (5). For the acetonitrile
◦
solvated compound (1a), the corresponding parameters are 115(4) , 103(3) and 2.309(3) A. The solid state 31P
CP/MAS NMR quadrupole distortion parameters, dmCu, for (1)–(3) and (1a) are all less than 1 × 109 Hz2, despite the
changes in donor properties of the halide in (1)–(3), and the coordination geometry of the P3CuX core in (1a). Change
˚
of anion to ClO4 and BF4 in compounds (4) and (5) results in a significant increase of dmCu to 4.4–5.2 109 Hz2 and
−
−
5.2–6.0 × 109 Hz2, respectively. Compounds (6) and (7) crystallise as isomorphous [Cu(PPh3)3(CH3CN)]X
˚
salts in space group Pbca, (a ∼ 17.6, b ∼22.3, c ∼24.2 A), while compounds (8) and (9) crystallize as isomorphous
¯
˚
acetonitrile solvated salts [Cu(PPh3)3(CH3CN)]X·CH3CN in space group P1 (a ∼ 10.5, b ∼ 13.0, c ∼19.5 A, a ∼ 104,
b ∼ 104, c ∼ 94◦ ). The P3CuN angular geometries in all four compounds are distorted from tetrahedral symmetry
◦
◦
˚
with average P–Cu–P, P–Cu–N angles and Cu–P bond lengths of 115(4) , 103(4) and 2.32(1) A, with dmCu ranging
between 1.3 and 2.5 × 109 Hz2. The solid state 29Si CP/MAS NMR spectrum of the pentafluorosilicate anion in
compound (8) is also reported, affording 1J(29Si, 19F) = 146 Hz.
to 97.56(5)–108.80(5)◦ (average 103(3)◦) (Tables 1 and 2).2–6
Introduction
Similar distortions are observed for complexes with hard oxy-
The adducts of tris(triphenylphosphine)copper(I) cations with
univalent anions comprise flexible three- or four-coordinate
arrays with structural and chemical properties that are de-
pendent on the nature of the bonding and steric interactions
between the [(PPh3)3Cu]+ acceptor cation, the anion donor and,
where present, solvate molecules in the crystal lattice. For the
unsolvated complexes, [(PPh3)3CuX] for X = Cl, Br or I, X-
ray structural data show the halide anions coordinate to the
copper to give monomeric [(PPh3)3CuX] (Scheme 1(a)) in which
the P–Cu–P and P–Cu–X angles lie close to the tetrahedral
angle of 109.5◦.1,2 However, recrystallization of these complexes
from acetone, acetonitrile or tetrahydrofuran yields solvated
complexes in which the P3CuX coordination sphere is trigonally
distorted with the P–Cu–P angles increasing to 111.3(3)–
119.99(4)◦ (average 115(3)◦) and the P–Cu–X angles decreasing
−
and fluoro-anions such as ClO4−, BF4−, HCO2 or NO3−, with
P–Cu–P 112.29(4)–121.37(6)◦ (average 116(3)◦) and P–Cu–X
94.4(1)–109.77(9)◦ (average 102(4)◦).7–9 Complexes with larger
−
anions, such as PF6 and V(CO)6−, afford three-coordinate
trigonal planar salts, [(PPh3)3Cu]X with non-coordinated anion
(Scheme 1(b)).10,11 Crystallization of the ClO4−, and PF6
−
complexes from acetonitrile solution does not result in sol-
vated [(PPh3)3CuX]·CH3CN complexes as observed for the
halide complexes, but rather four-coordinate mixed ligand salts,
[Cu(PPh3)3(CH3CN)]X, in which the anion is displaced from the
coordination sphere by acetonitrile (Scheme 1(c)).12,13 A room
temperature structure determination of the perchlorate complex
has shown the angular geometry to be trigonally distorted, with
P–Cu–P angles between 113.0(2) and 119.3(2)◦ and P–Cu–N
angles between 100.1(3) and 106.0(3)◦.12
Complementary information on the structural and molec-
ular properties of copper(I) phosphine complexes can also
be obtained from solid state 31P CP/MAS NMR spectra in
which 1J(31P,63Cu) and 1J(31P,65Cu) quartets for each crys-
tallographically independent phosphorus atom are observed.
These quartets exhibit asymmetry in the line spacings which
emanate from quadrupolar perturbations of the 31P Zeeman
states. These perturbations reduce the overall efficiency of MAS
averaging when combined direct (dipolar, D) and indirect (scalar,
J) couplings between 31P (I = 1/2, natural abundance 100%) and
quadrupolar 63,65Cu (S = 3/2, natural abundances 69.09% and
30.91%, respectively) nuclei are taken into account.2,14–27
Scheme 1
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
D a l t o n T r a n s . , 2 0 0 5 , 2 5 4 7 – 2 5 5 6
2 5 4 7
©

View Article Online
˚
Table 1 Cu–P and Cu–X bond lengths (A) for Cu(PPh3)3X(·solvent) compounds
Anion
Cl
Cu–P1
Cu–P2
Cu–P3
Average
Cu–X
Ref.
1
2.348(2)
2.351(2)
2.355(2)
2.351(4)
2.340(4)
2.369(4)
2.362(3)
2.346(5)
2.357(5)
2.314(5)
2.313(7)
2.317(8)
2.296(2)
2.296(2)
2.303(2)
2.306(1)
2.327(8)
2.318(2)
2.341(2)
2.335(4)
2.351(3)
2.327(2)
2.332(1)
2.315(1)
2.312(2)
2.277(1)
2.293(1)
2.348(2)
2.351(2)
2.355(2)
2.351(4)
2.340(4)
2.369(4)
2.362(3)
2.346(5)
2.357(5)
2.314(5)
2.313(7)
2.317(8)
2.296(2)
2.296(2)
2.303(2)
2.312(1)
2.327(7)
2.318(2)
2.346(2)
2.335(4)
2.352(3)
2.340(2)
2.332(1)
2.329(1)
2.337(2)
2.286(1)
2.297(1)
2.348(2)
2.351(2)
2.355(2)
2.351(4)
2.340(4)
2.369(4)
2.362(3)
2.346(5)
2.357(5)
2.314(5)
2.313(7)
2.317(8)
2.296(2)
2.296(2)
2.303(2)
2.310(1)
2.326(7)
2.318(2)
2.352(2)
2.355(4)
2.340(3)
2.358(2)
2.341(1)
2.332(1)
2.339(2)
2.301(1)
2.296(1)
2.351(4)a
2.336(4)
2.320(4)
2.350(4)
2.494(3)
2.478(4)
2.478(5)
2.684(3)
2.661(4)
2.662(4)
2.32(3)
2.23(3)
2.23(3)
2.339(9)
2.261(9)
2.318(9)
2.363(1)
2.349(6)
2.347(3)
2.324(2)
2.510(3)
2.439(2)
2.686(1)
2.042(4)
2.085(3)
2.174(4)
—
Br
2.353(14)a
2.355(8)a
2.314(2)a
2.298(4)a
2
2
7
8
I
ClO4
BF4
Cl·MeCN
Cl·acetone
Cl·3THF
Cl·THF
2.309(3)
2.327(8)
2.318(2)
2.338(12)
2.342(11)
2.348(7)
2.342(15)
2.335(5)
2.323(9)
2.329(15)
2.288(12)
2.925(2)
3
2
4
5
2
6
2
9
9
7
10
11
Br·acetone
Br·THF
I·acetone
HCO2·EtOH
HCO2·HCO2H
NO3·EtOH
PF6
V(CO)6
—
a Values averaged over all the three independent trigonally symmetric molecules in the unit cell.
Table 2 P–Cu–P and P–Cu–X bond angles (◦) for Cu(PPh3)3X(·solvent) compounds
X
P1–Cu–P2
P1–Cu–P3
P2–Cu–P3
< >
P1–Cu–X
P2–Cu–X
P3–Cu–X
< >
Cl
109.83(6)
109.12(6)
110.51(6)
110.6(2)
109.5(2)
110.5(2)
110.6(2)
109.7(2)
110.5(2)
114.5(3)
114.5(3)
115.6(3)
115.89(6)
115.22(6)
115.80(5)
119.99(4)
111.1(3)
115.00(5)
115.28(8)
117.7(1)
115.7(1)
118.00(6)
112.29(4)
112.72(4)
114.15(7)
125.2(8)
117.47(5)
109.83(6)
109.12(6)
110.51(6)
110.6(2)
109.5(2)
110.5(2)
110.6(2)
109.7(2)
110.5(2)
114.5(3)
114.5(3)
115.6(3)
115.89(6)
115.22(6)
115.80(5)
112.29(4)
115.1(3)
115.00(5)
111.93(8)
111.9(1)
118.5(1)
111.80(7)
116.09(4)
115.42(4)
113.42(6)
120.0(8)
119.94(5)
109.83(6)
109.12(6)
110.51(6)
110.6(2)
109.5(2)
110.5(2)
110.6(2)
109.7(2)
110.5(2)
114.5(3)
114.5(3)
115.6(3)
115.89(6)
115.22(6)
115.80(5)
114.08(4)
118.9(2)
115.00(5)
118.03(8)
116.0(1)
112.0(1)
116.08(6)
117.98(4)
119.45(4)
121.37(6)
113.3(8)
122.59(5)
109.8(7)a
109.11(6)
109.83(6)
108.41(7)
108.3(1)
109.2(1)
108.3(1)
108.3(1)
109.2(1)
108.4(1)
103.9(2)
103.9(2)
102.3(2)
101.86(8)
102.82(8)
101.99(7)
100.26(3)
104.0(2)
103.13(5)
104.23(8)
100.9(1)
102.51(8)
102.03(4)
109.77(9)
106.4(2)
109.5(1)
109.11(6)
109.83(6)
108.41(7)
108.3(1)
109.2(1)
108.3(1)
108.3(1)
109.2(1)
108.4(1)
103.9(2)
103.9(2)
102.3(2)
101.86(8)
102.82(8)
101.99(7)
101.96(4)
102.5(2)
103.13(5)
103.34(8)
99.2(1)
109.11(6)
109.83(6)
108.41(7)
108.3(1)
109.2(1)
108.3(1)
108.3(1)
109.2(1)
108.4(1)
103.9(2)
103.9(2)
102.3(2)
101.86(8)
102.82(8)
101.99(7)
105.39(4)
102.7(2)
103.13(5)
101.47(8)
108.5(1)
102.51(8)
108.80(5)
97.35(2)
99.8(2)
108.8(6)a
Br
110.2(6)a
110.3(5)a
114.8(7)a
115.6(4)a
108.6(5)a
108.8(6)a
103.4(9)a
102.2(5)a
I
ClO4
BF4
Cl·MeCN
Cl·acetone
Cl·3THF
Cl·THF
115(4)
115(4)
115.00(5)
115(3)
115(3)
115(3)
115(3)
115(3)
116(4)
116(3)
120(6)
120(3)
103(3)
102.6(7)
103.12(5)
102.8(9)
103(5)
103(2)
103(5)
103(3)
103(3)
102(4)
Br·acetone
Br·THF
101.60(8)
97.56(5)
98.36(9)
101.3(1)
95.4(1)
I·acetone
HCO2·EtOH
HCO2·HCO2H
NO3·EtOH
PF6
97.9(1)
V(CO)6
a Values averaged over all the three independent trigonally symmetric molecules in the unit cell.
In the present work we correlate the structural and solid state
Experimental
31P CP/MAS NMR properties of the four-coordinate
tris(triphenylphosphine)copper(I) complexes [Cu(PPh3)3X] for
X = Cl (1), Br (2), I (3), ClO4 (4) and BF4 (5), [Cu(PPh3)3Cl]·
CH3CN (1a), [Cu(PPh3)3(CH3CN)]X for X = ClO4 (6), BF4 (7),
and [Cu(PPh3)3(CH3CN)]X·CH3CN for X = SiF5, (8) and PF6
(9). As part of this study we also recorded the solid state 29Si
CP/MAS spectra of the pentafluorosilicate anion in (8). The
results of this work form the basis of this present report.
Preparation of compounds
The halide complexes [Cu(PPh3)3X], for X = Cl (1), Br (2) and
I (3), were prepared as described in the literature by precipi-
tation from chloroform solutions of stoichiometric quantities
of CuX and PPh3.2,28 The acetonitrile solvated halide complex,
[Cu(PPh3)3Cl]·CH3CN (1a), was prepared by recrystallization
2 5 4 8
D a l t o n T r a n s . , 2 0 0 5 , 2 5 4 7 – 2 5 5 6

View Article Online
of [Cu(PPh3)3Cl] from acetonitrile. Unit cell parameters are
consistent with those of the previously reported structure.3
warmed with stirring to give a clear solution. On cooling and
standing, crystalline precipitates of the required complexes were
obtained.
[Cu(PPh3)3ClO4], (4). A mixture of 0.7 g Cu2O and 6.0 g
PPh3 was suspended in 200 ml ethanol and 2 ml conc. HClO4
carefully added. Heating the solution to reflux with stirring gave
a clear colourless solution which was filtered and left to stand
overnight to yield a crystalline precipitate of the complex. Yield
4.4 g, 47.3%. (Found: C 68.3, H 4.7%. C54H45CuP3ClO4 requires
C 68.3, H 4.8%), mp 278–280 ◦C. The potentially explosive
properties of perchlorate complexes should be remembered
when preparing and storing this and the other perchlorate
compounds described below.
Structure determinations
Full spheres of ‘low-temperature’ CCD area-detector diffrac-
tometer data were measured (Bruker AXS instrument, T ∼
˚
150 K, monochromatic Mo-Ka radiation, k = 0.71073 A; x-
scans) yielding Nt(otal) reflections, these merging to N unique
after ‘empirical’/multiscan absorption correction (Rint cited), No
with F > 4r(F) being employed in the full matrix least squares
refinements, refining anisotropic displacement parameter forms
for the non-hydrogen atoms, (x, y, z, Uiso)H being constrained at
estimates. Conventional residuals at convergence, R, Rw(weights:
r2(F) + 0.000nwF2)−1 are cited. Neutral atom complex scattering
factors were employed. Computation used the Xtal 3.7 program
system32 and PLATON33 and ORTEP-334 software. Pertinent
results are given below and in the tables and figures, the latter
showing 50% probability amplitude displacement ellipsoids for
the non-hydrogen atoms, hydrogen atoms having arbitrary radii
[Cu(PPh3)3BF4], (5). A mixture of 0.7 g Cu2O and 6.0 g PPh3
was suspended in 100 ml ethanol and 2 ml aqueous HBF4 (45%)
slowly added with stirring to give a clear solution, together with
a small quantity of fine black particulate matter. The solution
was filtered and left to stand overnight to yield a crystalline
precipitate of the complex. Yield 4.3 g, 46.9%. (Found: C 69.1, H
4.9%. C54H45CuP3BF4 requires C 69.2, H 4.8%), mp 225–228 ◦C.
[Cu(PPh3)3(CH3CN)]ClO4, (6). 1.0 ml of acetonitrile was
added to a suspension of 0.5 g [Cu(PPh3)3ClO4] in 5.0 ml
ethanol and the mixture gently warmed to give a clear solution.
Slow cooling and evaporation of solvent yielded well-formed
crystals of the complex. (Found: C 68.9, H 4.9, N 1.0%.
C56H48ClCuNO4P3 requires C 67.9. H 4.9, N 1.4%), mp 275–
280 ◦C.
˚
of 0.1 A.
Crystal data
(6) [Cu(PPh3)3(CH3CN)]ClO4 = C56H48ClCuNO4P3, M = 990.9.
Orthorhombic, space group Pbca (No. 61), a = 17.685(1), b =
3
−3
˚
˚
22.314(2), c = 24.289(2) A, U = 9585 A . Dc (Z = 8) 1.367 g cm .
lMo = 6.6 cm−1; specimen: 0.40 × 0.38 × 0.19 mm; Tꢀmin/max
=
0.81. 2hmax = 65◦; Nt = 152360, N = 17060 (Rint = 0.077), No =
[Cu(PPh3)3(CH3CN)]BF4, (7). 0.2 ml of acetonitrile was
added to a suspension of 0.5 g [Cu(PPh3)3BF4] in 2.0 ml ethanol
and the mixture gently warmed to give a clear solution. Slow
cooling and evaporation of the solvent yielded well-formed
crystals of the complex. (Found: C 69.0, H 4.9, N 1.3%.
C56H48BCuF4NP3 requires C 68.8, H 4.9, N 1.4%), mp 224–
227 ◦C.
−3
˚
11151; R = 0.051, Rw = 0.067 (nw = 20). |Dqmax| = 1.36(5) e A .
(7) [Cu(PPh3)3(CH3CN)]BF4 = C56H48BCuF4NP3, M = 978.3.
Orthorhombic, space group Pbca, a = 17.730(1), b = 22.280(1),
3
−3
˚
˚
c = 24.185(1) A, U = 9554 A . Dc (Z = 8) = 1.360 g cm ; lMo
=
6.1 cm−1; specimen: 0.32 × 0.18 × 0.09 mm; Tꢀmin/max = 0.91.
2hmax = 53◦; Nt = 126697, N = 9761, (Rint = 0.097) No = 7132;
−3
˚
R = 0.047, Rw = 0.064 (nw = 2.5). |Dqmax| = 0.67(4) e A .
[Cu(PPh3)3(CH3CN)]SiF5·CH3CN, (8). This compound was
initially prepared adventitiously as follows, with its identity
being established by the spectroscopic and X-ray diffraction
studies described below. A mixture of PPh3 (3.95 g) and
CuF2·2H2O (1.38 g) was boiled in 40 ml acetonitrile in a glass
beaker until all the solids had dissolved. The resultant clear
solution was filtered and allowed to cool and evaporate to give a
colourless microcrystalline precipitate. Recrystallization of this
precipitate from acetonitrile yielded crystals of the complex
suitable for X-ray diffraction studies. (Found: C 65.5, H 5.0,
N 1.6%. C58H51CuN2P3SiF5 requires C 66.0 H 4.9, N 2.6%), mp
278–283 ◦C.
(8) [Cu(PPh3)3(CH3CN)]SiF5·CH3CN = C58H51CuF5N2P3Si,
¯
M = 1055.6. Triclinic, space group P1, a = 10.485(2), b =
˚
13.189(2), c = 19.451(4) A, a = 104.136(4) b = 104.592(4), c =
◦
3
−3
˚
93.839(5) , U = 2500 A . Dc (Z = 2) = 1.402 g cm . lMo
=
6.2 cm−1; specimen: 0.42 × 0.32 × 0.26 mm; Tꢀmin/max = 0.93.
ꢀ
2hmax = 65◦; Nt = 51618, N = 17597 (Rint = 0.044), No = 14031;
−3
˚
R = 0.043, Rw = 0.055 (nw = 8). |Dqmax| = 0.79(4) e A .
Variata. (x, y, z, Uiso)H were refined throughout.
(9) [Cu(PPh3)3(CH3CN)]PF6·CH3CN = C58H51CuF6N2P4,
¯
M = 1077.49. Triclinic, space group P1, a = 10.5757(8), b =
˚
3
13.246(1), c = 19.597(2) A, a = 104.180(2) b = 104.139(2), c =
◦
−3
˚
93.938(2) , U = 2557 A . Dc (Z = 2) = 1.399 g cm . lMo
=
The complex was subsequently synthesized rationally by
adaptation of methods described in the literature for the
6.2 cm−1; specimen: 0.38 × 0.28 × 0.19 mm; Tꢀmin/max = 0.94.
ꢀ
2hmax = 70◦; Nt = 51243, N = 21697 (Rint = 0.024), No = 16753;
preparation of other SiF5 compounds.29,30 Thus, 2 ml aqueous
−
−3
˚
R = 0.042, Rw = 0.049 (nw = 5). |Dqmax| = 0.67(5) e A .
HF (48%)) was carefully added to a warm, stirred mixture of
PPh3 (4.0 g, 15.3 mmol), Cu2O (0.36 g, 5.0 mmol) and high
surface area silica (0.30 g, 5.0 mmol) in acetonitrile (40 ml). The
solids in the mixture rapidly dissolved to give a clear solution
which, on cooling and standing overnight yielded well-formed
crystals of the complex.
CCDC reference numbers 268666–268669.
See http://dx.doi.org/10.1039/b505200a for crystallographic
data in CIF or other electronic format.
Solid state CP/MAS and MAS NMR spectroscopy
[Cu(PPh3)3(CH3CN)]PF6·CH3CN, (9). [Cu(MeCN)4]PF6·
MeCN31 (0.8 g, 1.9 mmol) and PPh3 (1.6 g, 6.1 mmol) were
dissolved in a warm solution of 5 ml EtOH and 5 ml MeCN to
give a clear colourless solution from which well formed crystals
of the complex were obtained by cooling to room temperature
and partial evaporation of the solvent. (Found: C 64.6, H 4.8.
N 2.5%. C58H51CuN2P4F6 requires C 64.6 H 4.8, N 2.6%), mp
235–238 ◦C.
Compounds (6), (7) and (9) were also prepared by the addition
of HXaq to a 3 : 1 P : Cu molar ratio of PPh3 and Cu2O in a 2 : 1
mixture of ethanol/acetonitrile. Typically, 0.36 g Cu2O and 4.0 g
PPh3 were suspended in 40 ml ethanol and 20 ml acetonitrile.
1–2 ml concentrated HXaq was added drop-wise and the solution
31P CP/MAS and 29Si MAS NMR spectra were obtained
at room temperature using a Bruker MSL-400 spectrometer
operating at the 31P and 29Si frequencies of 161.98 MHz and
79.49 MHz, respectively. For the 31P experiments, conventional
cross-polarisation and magic angle spinning techniques, coupled
with low temperature alternation to eliminate spectral artefacts,
were implemented using a Bruker 4 mm double-air-bearing
probe in which MAS frequencies of 10–12 kHz were achieved.
A recycle delay of 30 s, contact period of 10 ms and an initial
1H p/2 pulse length of 3 ls were common to all 31P spectra.
1
This represents a nominal H decoupling field strength ∼ 80
kHz during data acquisition if power switching to higher 1H de-
coupling levels is not implemented. No spectral smoothing was
D a l t o n T r a n s . , 2 0 0 5 , 2 5 4 7 – 2 5 5 6
2 5 4 9

View Article Online
employed prior to Fourier transformation. The 31P CP/MAS
two-dimensional correlation spectroscopy (COSY) experiments
were implemented with the time proportional phase incremen-
tation (TPPI) method for acquisition of phase sensitive data in
both F1 and F2 dimensions. The recycle delay, contact period,
1H p/2 pulse length and MAS spinning rate were the same as
those implemented for the one-dimensional experiments, details
of which have been reported elsewhere.17 All 31P chemical shifts
were externally referenced to 85% H3PO4 via solid ammonium
dihydrogen phosphate (d 1.0) or solid triphenylphosphine
(d −9.9). For the 29Si MAS NMR experiments, conventional
single pulse or Bloch decay methods, with and without high
power 1H decoupling, were implemented using a Bruker 7 mm
double-air-bearing probe in which MAS frequencies of 5 kHz
were achieved. A recycle delay of 30 s and a 29Si p/4 pulse length
of 2.5 ls were used for these 29Si experiments. When high power
1H decoupling was introduced the nominal 1H decoupling field
strength during data acquisition was ∼50 kHz. All 29Si chemical
shifts were externally referenced to tetramethylsilane (TMS) via
solid kaolinite (d −91.2).
the experiment with methanol or ethanol as solvent resulted in
formation of fluoride complexes only, suggesting an active role
for the solvent in this reaction. Compound (8) was subsequently
synthesized by adaptation of literature procedures for preparing
−
SiF5 compounds,29,30 in which concentrated aqueous HF was
carefully added to a suspension of PPh3, Cu2O and finely divided
SiO2 in acetonitrile.
Structure determinations
The low temperature crystal structure determinations of (6) and
(7) show the two complexes to be isomorphous. The anion is
displaced from the copper coordination sphere to give discrete
[(PPh3)3(CH3CN)Cu]+ cations and ClO4−/BF4− anions in space
˚
group Pbca with a ∼17.6, b ∼ 22.3, c ∼ 24.2 A. A representative
−
view of the cation and anion for the ClO4 complex is shown in
Fig. 2 with relevant geometric parameters listed in Table 3.
Results and discussion
Syntheses
−
−
Previously the unsolvated ClO4 and BF4 compounds (4)
and (5) have been synthesized by reaction of [Cu(PPh3)2BH4]
with perchloric or tetrafluoroboric acid in ethanol contain-
ing excess PPh3, and by precipitation from solutions of
[Cu(PPh3)3(CH3CN)]X in ethanol.7,8,35 Here, the complexes were
prepared by careful addition of concentrated aqueous HClO4
or HBF4 to a 3 : 1 P/Cu molar ratio of PPh3 and Cu2O
suspended in warm ethanol. Stirring gave clear solutions from
which well-formed crystals of the complexes precipitated in
good yield. Compound (6) has been prepared by the reaction
of stoichiometric quantities of [Cu(CH3CN)4]ClO4 and PPh3
in acetonitrile.12 Here, both (6) and (7) were prepared on a
semi-micro scale by addition of a slight excess of acetonitrile
to solutions of (4) and (5) in ethanol and by the addition
of HXaq to a 3:1 P/Cu molar ratio of PPh3 and Cu2O
suspended in 2 : 1 ethanol/acetonitrile. The SiF5 complex
(8) was obtained adventitiously from an attempt to prepare
the fluoride complex.36 Observation of a trigonal bipyramidal
species in the structure determination of the crystals obtained
Fig. 2 The asymmetric unit of [Cu(PPh3)3(MeCN)]ClO4, (6). Displace-
ment ellipsoids are drawn at the 50% probability level. H atoms are
˚
shown as spheres of arbitrary radius 0.1 A.
The structural parameters for both compounds are similar
to those recorded for the room temperature structure determi-
nation of (6) reported previously.12 The anions are located in
general positions in the unit cell and are well ordered with a
−
and subsequent population refinement suggested SiF5 as the
˚
number of phenyl H · · · O, F contacts at distances of 2.6–3.0 A
anion. This was confirmed by chemical analysis, the presence of
strong, sharp bands in the infrared spectrum at 880 and 790 cm−1
indicative of pentafluorosilicate,29 and by the solid state 29Si
MAS NMR spectrum which exhibits a characteristic 29Si sextet
˚
for (6) and 2.4–2.9 A for (7). Unlike the unsolvated complexes (4)
and (5) in which the three PPh3 ligands are related by three-fold
crystallographic symmetry axes,7,8 no internal crystallographic
symmetry is observed and the PPh3 ligands in (6) and (7)
adopt a diverse array of conformational structures with Cu–P–
Cipso–Cortho torsion angles ranging between 10.6(2) and 78.5(2)◦
(Table 3) while the chirality of ligand 1 is of the opposite hand
to those of ligands 2 and 3. The PPh3 and CH3CN ligands
coordinate to the copper atom through the P and N atoms to
give trigonal pyramidal P3CuN coordination spheres with P–
Cu–P angles 113.55(2)–119.42(2)◦ (6), 112.66(3)–119.68(3)◦ (7)
(average 115(3)◦) and N–Cu–P angles of 100.80(6)–105.74(6)◦
(6), 101.11(8)–105.29(8)◦ (7) (average 103(2)◦). The Cu–N–C
angles are essentially linear with values of 178.0(2) and 178.3(3)◦
for (6) and (7), respectively. Only minor differences are observed
at −146.5 ppm with 1J(Si, F) = 146 Hz (Fig. 1) (the solution 19
F
NMR spectrum of [(C3H7)4N][SiF5] gives 1J(F, Si) = 148 Hz37).
−
The SiF5 anion is most likely generated from reaction of silica
of the glass reaction vessel with residual HF according to the
−
38
reaction: SiO2 + 5HF → 2H2O + H+ + SiF5
. Repetition of
˚
in the Cu–P bond lengths, with values of 2.3126(7)–2.3205(6) A
˚
(6) and 2.3016(6)–2.3151(8) A (7); these are similar to the values
˚
of 2.296(2)–2.317(8) A found for the unsolvated complexes (4)
and (5) (Tables 1 and 2).7,8
The structure determinations of the SiF5− and PF6− complexes
(8) and (9) show these to be solvated with acetonitrile and
¯
isomorphous, crystallizing in space group P1; with a ∼ 10.5, b ∼
◦
˚
13.0, c ∼19.5 A, a ∼104, b ∼ 104, c ∼ 94 . The asymmetric unit
of each consists of a discrete [(PPh3)3(MeCN)Cu]+ cation, a
−
−
Fig. 1 Solid state29Si CP/MAS NMR spectra of the SiF5− anion in (8).
SiF5 or PF6 anion and an acetonitrile molecule of solvation.
2 5 5 0
D a l t o n T r a n s . , 2 0 0 5 , 2 5 4 7 – 2 5 5 6

View Article Online
Table 3 Core geometries (A, ◦) for [Cu(PPh3)3(MeCN)Cu]X, (6) and (7), and [Cu(PPh3)3(MeCN)]X·MeCN, (8) and (9)
˚
X = ClO4 (6)
X = BF4 (7)
X = SiF5 (8)
X = PF6 (9)
Cu–P1
Cu–P2
Cu–P3
Cu–N
2.3126(7)
2.3131(6)
2.3204(6)
2.055(2)
2.3135(8)
2.3151(8)
2.3227(8)
2.054(3)
2.3323(6)
2.3115(7)
2.3016(6)
2.080(2)
2.3460(5)
2.3311(5)
2.3206(5)
2.097(1)
P1–Cu–P2
P1–Cu–P3
P2–Cu–P3
N1–Cu–P1
N1–Cu–P2
N1–Cu–P3
Cu–N1–C1
Cu–P1–C111–C112
Cu–P1–C121–C122
Cu–P1–C131–C132
Cu–P2–C211–C212
Cu–P2–C221–C222
Cu–P2–C231–C232
Cu–P3–C311–C312
Cu–P3–C321–C322
Cu–P3–C331–C332
113.55(2)
119.42(2)
112.96(2)
100.80(6)
101.42(6)
105.74(6)
178.0(2)
28.5(2)
113.46(3)
119.68(3)
112.66(3)
101.11(8)
101.73(8)
105.29(8)
176.6(3)
28.6(3)
121.66(2)
111.57(2)
112.35(2)
102.43(5)
97.40(5)
109.20(4)
165.8(1)
33.2(2)
68.7(2)
54.8(2)
65.8(2)
25.3(2)
121.89(2)
111.67(2)
112.07(2)
102.39(4)
97.49(4)
109.08(4)
166.1(1)
33.0(1)
69.2(1)
53.7(1)
65.2(1)
23.8(1)
31.8(2)
32.7(2)
30.4(3)
31.8(3)
−10.6(2)
−78.5(2)
−28.2(2)
−39.8(2)
−43.2(2)
−71.6(2)
−11.7(3)
−77.9(3)
−28.5(3)
−40.1(3)
−42.1(3)
−73.0(3)
34.8(2)
11.4(2)
82.0(2)
29.4(2)
36.0(1)
12.37(1)
82.4(1)
29.9(1)
A representative view of the cation, anion and solvate acetoni-
trile are shown in Fig. 3 for compound (8) with geometric
parameters listed in Table 3. The structure determination of
(8) augments the small number of structures incorporating
preparation of PF6− complexes under appropriate reaction con-
ditions. In such cases, ‘disorder’, as often routinely assigned to
PF6− anions in structure determinations, may be a consequence
−
−
of co-crystallization of PF6 and SiF5 anions.
−
well ordered SiF5 anions.39,40 In accord with these previous
In these complexes all three phosphine ligands adopt the
same chirality. The change in anion–cation interactions and the
presence of the non-coordinated acetonitrile solvate results in
significant differences in the conformation of the PPh3 ligands
relative to those for (6) and (7) (Table 3). The geometric parame-
ters of the P3CuN coordination sphere show a greater dispersion
of values than do compounds (6) and (7) with Cu–P bond lengths
studies, the axial Si–F bond lengths of 1.651(2) and 1.652(2)
˚
˚
A are ∼ 0.05–0.10 A longer than the equatorial Si–F bond
−
˚
lengths of 1.560(2), 1.562(2) and 1.575(2) A. The SiF5 and
−
PF6 anions both crystallize with a number of H · · · F contact
˚
distances of 2.5–2.8 A between the fluorine atoms, and (i) phenyl
hydrogens on the PPh3 ligands, (ii) methyl hydrogens on the coor-
dinated acetonitrile, and (iii) methyl hydrogens on the solvated
acetonitrile. These two structures represent the first reported
˚
˚
of 2.3016(6)–2.3323(6) A (8), 2.3111(4)–2.3460(5) A (9); P–Cu–P
angles of 112.35(2)–121.66(2)◦ (8), 112.07(2)–121.89(1)◦ (9); and
P–Cu–N angles of 97.20(4)–109.20(4)◦ (8), 97.49(4)–109.08(4)◦
(9). The Cu–N–C angles deviate significantly from linearity
with values of 165.7(1) and 166.1(1)◦ while the Cu–N bond
−
−
example of isomorphous SiF5 and PF6 complexes and in this
context, we note that as HPF6 can decompose with release
−
of HF,31 the possibility exists of generating SiF5 during the
˚
˚
lengths of 2.080(2) and 2.096(1) A are ∼0.03 A longer than the
˚
values of 2.055(2) and 2.054(3) A in (6) and (7), possibly as a
consequence of the relatively strong interactions between the
−
−
methyl hydrogens and the PF6 or SiF5 anions.
Solid State 31P NMR spectroscopy
Solid state 31P CP/MAS spectra of copper(I) phosphine com-
plexes recorded at high magnetic fields (> 7 T) show overlap of
1J(31P–63Cu) and 1J(31P–65Cu) quartets for each crystallographi-
cally independent phosphorus atom.2,14,16–23 The two quartets are
not completely resolved and the observed spectra are dominated
by the 1J(31P–63Cu) coupled spectrum, with the 1J(31P–65Cu)
spectrum identified (when the line width is sufficiently narrow)
as a splitting of the outer resonances of the quartet due to the
differences in the gyromagnetic ratios for the two Cu isotopes
(c 63Cu/c 65Cu = 0.933). If the spectra involved only isotropic J
coupling to the copper nucleus, four equally spaced lines (line
spacing = J) would be observed (Scheme 2(a)). In practice the
line spacings in these spectra are field dependent and exhibit a
distinct asymmetry arising from a quadrupole perturbation to
the 31P Zeeman states (Scheme 2). Theoretical treatments of this
phenomenon have been approached in exact24 and first-order
perturbation25–27 calculations. For the 31P–63,65Cu case a first-
order analysis predicts that the outer two resonances (d1, d4) shift
upfield while the inner two resonances (d2, d3) shift downfield by
1
a magnitude d such that J(P–Cu) = J = D32 and d = (D32–
D21)/2 = (D43–D32)/2 = (D43–D21)/4 where D21, D32, D43 are the
three lines spacings for the quartet (Scheme 2(b)). To a first
order approximation, d is linearly dependent on field strength
Fig. 3 The asymmetric unit of [Cu(PPh3)3(MeCN)] SiF5·MeCN, (8).
D a l t o n T r a n s . , 2 0 0 5 , 2 5 4 7 – 2 5 5 6
2 5 5 1

View Article Online
coordinate linear complexes,22,23 reflecting primarily the increase
in the copper nuclear quadrupolar coupling constants, vCu,
as the coordination number decreases. For four-coordinate
copper(I) triphenylphosphine complexes, values of dmCu are
generally less than 10 × 109 Hz2 with, for example, values
of <1 × 109 Hz2 for [(PPh3)3CuX] (X = halide),2 ∼2–4 ×
109 Hz2 for [Cu(PPh3)2(CH3CN)2]X,18 ∼1–4 × 109 Hz2 for
[Cu(PPh3)(bipyridyl)X], (X = halide),15 4.5 × 109 Hz2 for
[Cu(PPh3)2(S2CPh)],20 5–9 × 109 Hz2 for [Cu(PPh3)X]4 (X =
halide),16 ∼7–9 × 109 Hz2 for [Cu(PPh3)(piperidine)]2 (X =
halide),16 and 9.7 × 109 Hz2 for [Cu(PPh3)2(NO3)].14
The solid state 31P CP/MAS NMR spectra of the unsolvated
[(PPh3)3CuX] compounds for X = Cl, Br and I, (1)–(3), were
amongst the first to be recorded for copper(I) phosphine
complexes.2,14 The spectra of these compounds consist of two
quartets with an intensity ratio of 2:1. Their structures are
consistent with the presence of three 31P quartets arising from
each of the three crystallographically independent, trigonally
symmetric molecules in the unit cell. Two of these quartets
exhibit coincident isotropic chemical shifts and J-coupling char-
acteristics which differ from the third. The NMR parameters for
these complexes are listed in Table 4 with the spectra presented
for the I, Br and Cl compounds in Figs. 4(a)–4(c). The isotropic
chemical shifts for the three compounds show a significant
downfield shift of diso in the order I (−15.4, −18.8 ppm) < Br
(−11.6, −14.4 ppm)) < Cl (−8.9, −11.2 ppm). The measurement
Scheme 2
and multiplication of d by the 63Cu Zeeman frequency mCu yields a
field independent parameter dmCu that is suitable for comparison
of results from data recorded at different field strengths. This first
order treatment yields expression (1) which relates the magnitude
of dmCu and appropriate spin-Hamiltonian parameters:
dmCu = 0.15 (D − DJ/3)(3cos2bD − 1 + gsin2bDcos2aD) vCu (1)
where vCu = e2qQ/h is the 63Cu quadrupole coupling constant, g
is the asymmetry parameter of the electric field gradient (EFG)
tensor, Deff = (D − DJ/3) which incorporates the P–Cu dipolar
coupling constant, D (= 12 900/(Cu–P)3), the anisotropy in the
J tensor, DJ, and the polar angles defining the orientation of
the P–Cu internuclear vector with respect to the principal axis
system (PAS) of the EFG tensor, aD, bD.20,27 Higher order effects
cause the line spacings to deviate from this first order model
with (D32 − D21) consistently greater than (D43 − D32). These
effects, however, are usually small and manifest themselves as
upfield shifts of lines 1 and 3 and a downfield shift of lines 2
and 4 by a small correction term of magnitude d1 (Scheme 2(c)).
This produces D21 = J − 2d − 2d1, D32 = J + 2d1 and D43 = J +
1
of J(31P–63Cu) for these complexes shows that they fall within
a narrow range of 0.92–0.93 kHz; this result contrasts with a
trend previously reported for triarylphosphine copper(I) halide
1
complexes (e.g. ref. 21), where the magnitude of J(31P–63Cu)
decreases markedly along the series Cl < Br < I. As observed
previously,2,14 the distortions of the line spacings in the quartets
for these compounds are small with values of dmCu all less than
1 × 109 Hz2. In these compounds, the z axis of the electric field
gradient is coincident with the crystallographic three-fold axis
along the Cu–X bond and the asymmetry parameter, g, is zero,
resulting in simplification of eqn. (1) to eqn. (2).
dmCu = 0.15(D − DJ/3)(3cos2bD −1)vCu
(2)
2d − 2d1 from which d = (D43 − D21)/4 as before, d1 = (D32
J)/2 while J = (D21 + 2D32 + D43)/4.
−
The measured values of dmCu copper phosphine com-
plexes range from zero for four-coordinate complexes with
exact tetrahedral symmetry41–43 to 25 × 109 Hz2 for two-
Substitution of values of dmCu obtained from the NMR ◦spectra
(Table 4) and calculation of (3cos2bD − 1) for bD = 108.7 , D for
23,26,27
˚
Cu–P = 2.35 A and values of DJ ranging from 0.5–1.0 kHz
Table 4 31P CP/MAS NMR data for [Cu(PPh3)3X] and [Cu(PPh3)3(MeCN)]X complexes (1)–(9) measured at B = 9.40 T
Compound
disoa/ppm
D12/kHz
D23/kHz
D34/kHz
1J(Cu,P)
d/Hz
dmCu/109Hz2
(1) [Cu(PPh3)3Cl]
(2) [Cu(PPh3)3Br]
(3) [Cu(PPh3)3I]
(4) [Cu(PPh3)3ClO4]
−8.9
−11.2
−11.6
−14.1
−15.4
−18.8
−2.5
−4.7
−5.4
−1.2
−3.2
−4.3
−2.5
−3.9
1.9
0.91
0.91
0.91
0.91
0.90
0.91
0.98
0.98
0.98
0.97
0.98
0.98
0.92
0.96
0.98
0.91
0.95
0.98
0.91
0.94
0.94
0.94
0.94
0.93
0.95
0.94
0.94
0.93
1.11
1.11
1.10
1.13
1.12
1.12
0.94
0.99
1.04
0.94
1.00
1.06
0.94
1.00
1.00
1.00
0.91
0.91
0.93
0.94
0.94
0.93
1.17
1.15
1.15
1.20
1.19
1.17
0.94
0.99
1.07
0.97
1.01
1.07
0.96
1.03
1.01
1.01
0.93
0.92
0.93
0.93
0.93
0.92
1.09
1.09
1.08
1.11
1.10
1.10
0.93
0.99
1.03
0.94
1.00
1.04
0.94
0.99
0.99
0.99
0.0
0.0
4.8
7.0
9.5
0.0
0.0
0.5
0.7
1.0
0.5
5.2
4.7
4.4
6.0
5.7
5.2
0.5
0.7
2.5
1.6
1.6
2.6
1.3
2.3
1.9
1.9
4.8
48.8
44.3
42.8
56.5
56.3
41.8
5.0
(5) [Cu(PPh3)3BF4]
(1a) [Cu(PPh3)3Cl]·MeCN
(6) [Cu(PPh3)3(MeCN)]ClO4
7.5
23.3
15.3
5.3
24.5
12.3
22.0
18.3
18.3
0.6
−1.5
(7) [Cu(PPh3)3(MeCN)]BF4
1.4
0.3
−1.7
(8) [Cu(PPh3)3(MeCN)]SiF5·MeCN
(9) [Cu(PPh3)3(MeCN)]PF6·MeCN
1.3
1.2
a Referenced against 85% H3PO4; for all multiplet structures; calculated from the centre position of each line comprising that multiplet as diso = (d1 +
d2 +. . .+dN)/N. Estimated error in Dij = 20 Hz.
2 5 5 2
D a l t o n T r a n s . , 2 0 0 5 , 2 5 4 7 – 2 5 5 6

View Article Online
and makes a negative contribution to the total field gradient.
The extent of this contribution is strongly dependent on the
ability of the ligand to donate electron density into the Cu
4pz orbital. If the X ligand is electronically equivalent to the
P ligands, such as would occur in an exactly tetrahedral P4Cu
environment, the positive contribution of the three off-axis P
ligands to the field gradient is exactly balanced by the negative
contribution of the axial ligand, resulting in a total field gradient
of zero, and hence in a quadrupole coupling constant of zero.
If the X ligand is a weaker donor than the P ligands, then its
negative contribution to the field gradient would be reduced
in magnitude, resulting in a net positive field gradient for the
complex. Normally, the incorporation of a weaker donor ligand
X would be expected to result in an increase in the P–Cu–P angle,
which would further increase in the field gradient, as described
above. Neither of these things appear to happen in the case of
the X = halide complexes, all of which have P–Cu–P angles
close to the tetrahedral angle (Tables 1 and 2), and very small
quadrupole coupling constants as discussed above. A possible
explanation for this behaviour may be found in the results of a
previously published computational study of the copper nuclear
quadrupole coupling in diatomic CuX molecules.44 This showed
that, despite the fact that iodide is a stronger r-donor than
chloride, the copper nuclear quadrupole coupling constant in
CuCl is almost double that in CuI. This was shown to be due
to a contraction of the Cu 4p orbital involved in the bonding
to the halogen atom, the extent of this contraction increasing
substantially from X = I to X = Cl. A similar phenomenon in the
P3CuX complexes would result in the Cu 4pz orbital contracting
to a greater extent than the corresponding px, py orbitals, since
the former is the only one involved in the CuX bonding. In
the case of the X = Cl complex, this contraction is apparently
sufficient to allow the negative Cu–X contribution to the field
gradient to balance the positive Cu–P contribution, resulting
in a total field gradient and quadrupole coupling constant of
zero. According to the previously reported results for CuX
molecules,44 the negative contribution of the Cu–X bond in
P3CuX should decrease in magnitude from X = Cl to X = I, so
that the total field gradient and quadrupole coupling constant
should increase along this series. The observed progressive
increase in dmCu values (Table 4) indicates that this is the case.
The 31P CP/MAS NMR spectrum of the acetonitrile solvated
chloride complex, [Cu(PPh3)3Cl]·CH3CN (1a), is shown in
Fig. 4(d) with the associated NMR parameters listed in Table 4.
The spectrum of (1a) consists of two 31P broad quartets in an
approximately 2 : 1 ratio of intensities which can be assigned to
three overlapping quartets from the three crystallographically
independent phosphorus atoms bound to the same copper
atom.3 The average chemicals shifts of −2.5 and −3.9 ppm are
∼7 ppm downfield from the unsolvated choride complex (1).
The broadness and complex structure of the lines, which can be
ascribed to both the overlap of the quartets and second order
2J(31P–31P) coupling between three phosphorus atoms, reduces
the precision with which the spacings between the lines can be
Fig. 4 Solid state 31P CP/MAS NMR spectra of (a) [(PPh3)3CuI], (3),
(b) [(PPh3)3CuBr], (2), (c) [(PPh3)3CuCl], (1), (d) [(PPh3)3CuCl]·MeCN,
(1a) (e) [(PPh3)3CuClO4], (4) and (f) [(PPh3)3CuBF4], (5).
yields estimates for the copper quadrupole coupling constants,
vCu, of zero for the chloride, 6–10 MHz for the bromide and 6–
14 MHz for the iodide, suggesting a symmetric electron charge
distribution about the copper position for all three complexes.
The narrow range of these results is surprising in view of the very
different donor properties of the phosphine and halide ligands
and this question is explored further below.
The electric field gradient, eq, at the copper nucleus in an
axially symmetric (C3) P3CuX environment can be divided
into two contributing parts: one from the three phosphine
ligands, and one from the halide ligand. Considering first the
contribution from the three phosphine ligands, this will be
dependent on the P–Cu–P angle h. The resulting field gradient
would be a maximum for a trigonal planar arrangement with h =
120◦, and would be zero for the trigonal pyramidal arrangement
with h = 90◦. In complexes with a tetrahedral coordination
environment with h = 109.5◦, the three phosphine ligands make
a substantial positive contribution to the field gradient. The
X ligand lies on the threefold axis (the z axis) of the complex
1
measured. The values of J(31P–63Cu) for the two quartets are
0.93 and 0.99 kHz which are only slightly larger than those
for the unsolvated complexes, despite the observed decrease in
the magnitude of the Cu–P bond length. The values of dmCu
for the two quartets of 0.5 and 0.7 × 109 Hz2 are greater than
the values of zero observed for the unsolvated chloride and,
utilizing eqn. (2), yield estimates for vCu of 4–8 MHz which
are similar to the range of values estimated for the unsolvated
bromide and iodide complexes. This result also agrees with
previous results obtained at B = 7.05T for the acetone solvated
complexes[Cu(PPh3)3X]·acetone (X = Cl, Br, I)2 where the dmCu
values are found to range from 0.4–1.2 × 109 Hz2. However, the
increase in the P–Cu–P angle, the decrease in the Cu–P bond
length, and the increase in the Cu–Cl bond length in going from
(1) to (1a) might each be expected to contribute to an increased
imbalance in the electric field gradient about the copper and
D a l t o n T r a n s . , 2 0 0 5 , 2 5 4 7 – 2 5 5 6
2 5 5 3

View Article Online
Fig. 5 Two-dimensional solid state 31P CP/MAS NMR spectra of (a) [(PPh3)3CuClO4], (4) and (b) [Cu(PPh3)3(MeCN)]ClO4, (6).
hence larger values of dmCu than are observed. That this does not
occur may be a reflection of different values of DJ for the solvated
and unsolvated complexes, and/or that the effects of variation
in geometric parameters for complexes with the same donor
ligands may have a much smaller influence on the magnitude
of the copper quadrupolar coupling constant than predicted by
the first order theory encapsulated in eqn. (1). The answers to
these questions are dependent on direct determination of the
magnitude of vCu for these compounds.
The 31P CP/MAS NMR spectra for [Cu(PPh3)3ClO4] (4) and
[Cu(PPh3)3BF4] (5) are shown in Figs. 4(e) and 4(f) and the
NMR parameters are listed in Table 4. As both these compounds
are isomorphous with compounds (1)–(3), three independent
quartets are expected and, unlike (1)–(3), these are resolved.
That these quartets are derived from phosphorus atoms located
on three crystallographically independent, trigonally symmetric
[(PPh3)3CuX] molecules, rather than from phosphorus atoms
coordinated to the same copper atom, is verified by the 2D 31
P
CPCOSY result for (4), shown in Fig. 5(a), in which off-diagonal
correlations indicating connectivity (through homonuclear J
coupling) are absent. The average chemical shifts of the quartets
in these complexes are similar to those for the solvated chloride
complex (1a) and, as for (1a), significantly shifted downfield
Fig. 6 Solid state 31P CP/MAS NMR spectra of (a) [Cu(PPh3)3-
(MeCN)]ClO4, (6) and (b) [Cu(PPh3)3(MeCN)]BF4, (7).
1
from the unsolvated halide complexes. The values of J(31P–
63Cu) increase to 1.09–1.11 kHz, which is consistent with the
observed shortening of the Cu–P bond lengths. The magnitude
of dmCu for these complexes, increases significantly to 4.4–5.2 ×
109 Hz2 for (4) and 5.2–6.0 × 109 Hz2 for (5). Substitution
of the relevant parameters in eqn. (1) yields estimates of vCu
between 40 and 60 MHz for (4) and 45 and 65 MHz for (5).
The small effect of changes in the geometry of the P3CuX core
on the magnitude of dmCu for the unsolvated and solvated halide
complexes, suggests that this increase is best ascribed to the
increased size and significantly weaker r donor strength of the
oxy- and fluoro-anion donor atoms by comparison with the
halide anions. This reduction in r donor strength results in a
decrease in the magnitude of the negative Cu–X contribution
and an increase in the total field gradient, as discussed above.
The 1D 31P CP/MAS NMR spectra of the acetonitrile coordi-
nated complexes [Cu(PPh3)3(CH3CN)]ClO4 (6) and [Cu(PPh3)3-
(CH3CN)]BF4 (7) and the solvated complexes [Cu(PPh3)3-
(CH3CN)]SiF5·CH3CN (8) and [Cu(PPh3)3(CH3CN)]PF6·
CH3CN (9) are shown in Figs. 6 and 7. In compounds (6) and
(7), three overlapping 31P quartets are observed for each of
the PPh3 ligands (Fig. 6), with the similarities in the spectra
reflecting the structural isomorphism of the two complexes. As
observed for complex (1a), further complexity is introduced
into the analysis of these spectra by virtue of each P nucleus
2
being strongly J(31P–31P) coupled to the other two P nuclei
Fig. 7 Solid State 31P CP/MAS NMR spectra of [Cu(PPh3)3(MeCN)]·
X·MeCN (a) X = PF6 (9), (b) X = SiF5 (8).
within the three inequivalent PPh3 ligands. The fine structure
2 5 5 4
D a l t o n T r a n s . , 2 0 0 5 , 2 5 4 7 – 2 5 5 6

View Article Online
from this 2J(31P–31P) coupling in these complexes is not
well resolved in the 1D experiment. However, the 2D 31P
CPCOSY result for (6) (see Fig. 5(b)) resolves and verifies
this coupling by identifying the appropriate off-diagonal
correlations in what appears to be a strongly coupled type ABC
spin system. The spectra of compounds (8) and (9) are shown
in Fig. 7. Each spectrum consists of a single quartet (reflecting
the isomorphism of the two complexes) which is assigned as
consisting of three coincident quartets arising from each of
the crystallographically independent phosphorus atoms. The
average chemical shifts for the quartets in all four compounds
lie in the same range as the values observed for compounds (1a),
(4) and (5), while the values of 1J(31P–63Cu) range between 0.94
and 1.04 kHz in accord with the shortened Cu–P bond lengths
observed for these compounds. The estimated magnitudes of
dmCu for the compounds range from 1.3–2.6 × 109 Hz2 which lie
Recrystallization of the ClO4−, BF4−, PF6 and SiF5
−
−
complexes from acetonitrile yields complexes of the type
[Cu(PPh3)3(CH3CN)]X. The average P3CuN coordination ge-
ometry of these complexes is comparable to that for the
−
unsolvated ClO4−, BF4 and solvated chloride complexes. The
values of dmCu lie between those for the halide and ClO4−, BF4
−
complexes which is in line with the chemistry of these complexes.
Overall, this present study has extended our knowledge of the
structural and molecular properties of tris(triphenylphosphine)-
copper(I) complexes. [(PPh3)3CuX], [(PPh3)3CuX]·MeCN and
[(PPh3)3Cu(MeCN)]X and [(PPh3)3Cu(MeCN)]X·MeCN, illus-
trating the complementary nature of the structural and molecu-
lar information obtained from single crystal X-ray diffraction
and solid state 31P CP/MAS NMR studies and how small
changes in the donor and steric properties of the non-phosphine
ligand and solvated molecules can result in significant variations
in the molecular parameters.
−
between the values for the halide complexes and for the ClO4
−
and BF4 complexes, suggesting the donor strength of the non-
phosphorus donor groups in the present series of complexes
Acknowledgements
−
−
increase in the order: ClO4 ∼ BF4 < CH3CN < halide.
This result is in line with the chemistry of these compounds
where acetonitrile displaces the oxy- and fluoro-anions, but
not the halide anions from the copper coordination sphere. It
is interesting to note that similar values of dmCu are obtained
for a number of tetrahedral bis(triphenylphosphine)copper(I)
P2CuN2 systems: [Cu(PPh3)2(CH3CN)2]X (X = ClO4, BF4,
PF6), (2.5–3.5 × 109 Hz2)18 and [Cu(PPh3)2(4-XC6H4NCN)]2X
(X = H, Me, Cl) (0.6–2.5 × 109 Hz2).14 Further comparison can
be drawn from recently reported results for triphenylphosphine
poly(pyrazolyl)borate copper(I) complexes which exhibit
PCuN3 coordination and for which dmCu values range from
2.5–5.7 × 109 Hz2.19
We acknowledge support of this work through funding from
the Australian Research Council, the Australian Institute of
Nuclear Science and Engineering (AINSE) for funding of
AINSE project No. 02/188, Griffith University, the University
of Western Australia, and the University of Auckland. The
authors gratefully acknowledge the contribution of Ms Anita
Chappell to the initial synthesis of compound (8).
References
1 J. T. Gill, J. J. Mayerle, P. S. Welcker, D. F. Lewis, D. A. Ucko, D. J.
Barton, D. Stowens and S. J. Lippard, Inorg. Chem., 1976, 15, 1155.
2 P. F. Barron, J. C. Dyason, P. C. Healy, L. M. Engelhardt, C.
Packawatchai, V. A. Patrick and A. H. White, J. Chem. Soc., Dalton
Trans., 1987, 1099.
3 T. Krauter and B. Neumuller, Polyhedron, 1996, 15, 2851.
4 K. Folting, J. Huffman, W. Mahoney, J. M. Stryker and K. G.
Caulton, Acta Crystallogr., Sect. C, 1987, 43, 1490.
5 T. F. Carlson, J. P. Fackler and R. A. Kresinski, Acta Crystallogr.,
Sect. C, 1996, 52, 1117.
6 H. Eriksson, M. Ortendahl and M. Hakansson, Organometallics,
1996, 15, 4823.
7 J. C. Dyason, L. M. Engelhardt, P. C. Healy, H. L. Klich and A. H.
White, Aust. J. Chem., 1986, 39, 2003.
8 A. P. Gaughan, Z. Dori and J. A. Ibers, Inorg. Chem., 1974, 13, 1657.
9 G. A. Bowmaker, Effendy, J. V. Hanna, P. C. Healy, J. C. Reid, C. E. F.
Rickard and A. H. White, J. Chem. Soc., Dalton Trans., 2000, 753.
10 P. C. Healy and J. V. Hanna, Acta Crystallogr., Sect. E, 2003, 59,
m384.
Conclusions
The present study has investigated the manner in which changes
in the steric and donor properties of the ligands impact on the
structural and solid state 31P CP/MAS NMR parameters for a
series of solvated and unsolvated four-coordinate [(Cu(PPh3)3X]
and [(Cu(PPh3)3CH3CN]X copper(I) complexes. For the trig-
onally symmetric isomorphous series [(Cu(PPh3)3X] for X =
Cl, Br, I, ClO4, BF4, the results show that changes in halide
have only minimal effect on the coordination geometry and
the quadrupolar distortion parameter dmCu. The relatively small
changes in dmCu observed for these three complexes, despite the
fact that the r-donor strength of the anions decrease in the order
I > Br > Cl, is ascribed to contraction of the Cu 4p orbital
involved in bonding to the halogen atoms, with the extent of this
contraction increasing in the order I < Br < Cl. Changing anions
11 G. Doyle, K. A. Eriksen and D. V. Engen, Organometallics, 1985, 4,
2201.
12 P. F. Barron, J. C. Dyason, L. M. Engelhardt, P. C. Healy and A. H.
White, Aust. J. Chem., 1985, 38, 261.
13 A. M. Leiva, L. Rivera and B. Loeb, Polyhedron, 1991, 10, 347.
14 J. W. Diesveld, E. M. Menger, H. T. Edzes and W. S. Veeman, J. Am.
Chem. Soc., 1980, 102, 7935.
15 P. F. Barron, L. M. Engelhardt, P. C. Healy, J. D. Kildea and A. H.
White, Inorg. Chem., 1988, 27, 1829.
16 G. A. Bowmaker, J. V. Hanna, R. D. Hart, P. C. Healy and A. H.
White, J. Chem. Soc., Dalton Trans., 1994, 2621.
17 J. V. Hanna, M. E. Smith, S. N. Stuart and P. C. Healy, J. Phys.
Chem., 1992, 96, 7560.
18 J. V. Hanna, R. D. Hart, P. C. Healy, B. W. Skelton and A. H. White,
J. Chem. Soc., Dalton Trans., 1998, 2321.
19 G. G. Lobbia, J. V. Hanna, M. Pellei, C. Pettinari, C. Santini, B. W.
Skelton and A. H. White, Dalton Trans., 2004, 951.
20 F. Asaro, A. Camus, R. Gobetto, A. C. Olivieri and G. Pellizer, Solid
State NMR, 1997, 8, 81.
21 G. A. Bowmaker, S. E. Boyd, J. V. Hanna, R. D. Hart, P. C. Healy,
B. W. Skelton and A. H. White, J. Chem. Soc., Dalton Trans., 2002,
2722.
−
−
to ClO4 and BF4 results in significant trigonal distortion of
the P3CuX coordination geometry, with the P–Cu–P angles
increasing, the P–Cu–X angle decreasing by ca. 5–6◦ and the
˚
Cu–P bond length decreasing by ca. 0.05 A. The quadrupolar
distortion parameter dmCu increases ten-fold, which is consistent
with an increase in the estimated copper quadrupolar coupling
constant for these complexes from ca. 0–14 MHz for the halide
complexes to ca. 40–65 MHz for the ClO4− and BF4− complexes.
This result is in accord with the larger size and poorer r-donor
properties of the oxy- and fluoro-anions by comparison with the
halide anions.
Recrystallization of the chloride complex from acetonitrile
yields the solvated complex [(Cu(PPh3)3Cl]·CH3CN. The struc-
tural parameters for the P3CuX core in this compound are
similar to those for the ClO4 and BF4 complexes. The
magnitude of dmCu is, however, only slightly greater than that
of the unsolvated chloride complex. This result suggests that the
effects of variation in geometric parameters on this parameter
in circumstances where the coordinating ligands are unchanged
may be smaller that expected.
−
−
22 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., 2001, 144.
23 S. Kroeker, J. V. Hanna, R. E. Wasylishen, E. W. Ainscough and
A. M. Brodie, J. Magn. Reson., 1998, 135, 208.
24 E. M. Menger and W. S. Veeman, J. Magn. Reson., 1982, 46, 257.
D a l t o n T r a n s . , 2 0 0 5 , 2 5 4 7 – 2 5 5 6
2 5 5 5

View Article Online
25 R. K. Harris and A. C. Olivieri, Prog. NMR Spectrosc., 1992, 24,
435.
35 F. Cariati and L. Naldini, Gazz. Chim. Ital., 1965, 95, 3.
36 D. J. Gulliver, W. Levason and M. Webster, Inorg. Chim. Acta, 1981,
52, 153.
37 F. Klanberg and E. L. Muetterties, Inorg. Chem., 1968, 7, 155.
38 A. C. Cooper, J. C. Bollinger, J. C. Huffman and K. G. Caulton,
New J. Chem., 1998, 473.
39 D. Schomburg and R. Krebs, Inorg. Chem., 1984, 23, 1378.
40 F. Olbrich and R. J. Lagow, Z. Anorg. Allg. Chem., 1995, 621,
1929.
26 A. C. Olivieri, J. Magn. Reson., 1989, 81, 201.
27 A. Olivieri, J. Am. Chem. Soc., 1992, 114, 5758.
28 G. Costa, E. Reisenhofer and L. Stefani, J. Inorg. Nucl. Chem., 1965,
27, 2581.
29 H. C. Clark, K. R. Dixon and J. G. Nicolson, Inorg. Chem., 1969, 8,
450.
30 P. Bird, J. F. Harrod and K. A. Than, J. Am. Chem. Soc., 1974, 96,
1222.
31 G. J. Kubas, Inorg. Synth., 1979, 19, 90.
32 S. R. Hall, D. J. du Boulay and R. Olthof-Hazekamp, in The Xtal 3.7
system, 2001.
33 A. L. Spek, J. Appl. Crystallogr., 2003, 36, 7.
34 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
41 R. W. King, T. J. Hutteman and J. G. Verkade, Chem. Commun.,
1965, 561.
42 W. McFarlane and D. S. Rycroft, J. Magn. Reson., 1976, 24, 95.
43 A. Marker and M. J. Gunter, J. Magn. Reson., 1982, 47, 118.
44 G. A. Bowmaker and P. D. W. Boyd, J. Mol. Struct. (THEOCHEM),
1985, 122, 299.
2 5 5 6
D a l t o n T r a n s . , 2 0 0 5 , 2 5 4 7 – 2 5 5 6