Group 11 metal complexes of SPS-based pincer ligands: Syntheses, X-ray structures and reactivity

Article abstract of DOI:10.1039/b406792d

New group 11 d¹⁰ (Cu, Au) metal complexes with SPS pincer ligand were synthesized. Insoluble dimeric or oligomeric complexes [(SP(R)S)Cu] _n (R = Bu: 4, Me: 5) were readily cleaved by several two-electron donor ligands (phosphines, is

Full text of DOI:10.1039/b406792d

View Article Online / Journal Homepage / Table of Contents for this issue
F U L L P A P E R
Group 11 metal complexes of SPS-based pincer ligands: Syntheses,
X-ray structures and reactivity†
Marjolaine Doux, Louis Ricard, Pascal Le Floch and Nicolas Mézailles*
Laboratoire « Hétéroéléments et Coordination », UMR CNRS 7653 (DCPH),
Département de Chimie, Ecole Polytechnique, 91128 Palaiseau Cedex, France.
E-mail: mezaille@poly.polytechnique.fr; Fax: +33 1 69333990; Tel: +33 1 69334570
Received 6th May 2004, Accepted 22nd June 2004
First published as an Advance Article on the web 21st July 2004
New group 11 d10 (Cu, Au) metal complexes with SPS pincer ligand were synthesized. Insoluble dimeric or oligomeric
complexes [(SP(R)S)Cu]n (R = Bu: 4, Me: 5) were readily cleaved by several two-electron donor ligands (phosphines,
isocyanides, pyridine) to yield a range of new complexes (6–13). X-Ray crystal studies were performed on complexes 7,
8, 9, 11, which revealed distorted tetrahedral geometries and proved once again the flexibility of the SPS ligand, which
can accommodate square planar, tetrahedral, octahedral and trigonal bipyramidal geometries. A dimeric gold species with
an Au–Au interaction 16 was also synthesized. This dimer could be cleaved with two electron donor ligand (PPh3: 17,
RNC 18). Reactivity of complex 11 with ethyl diazoacetate yielded new 5-phosphinine 14.
Introduction
Since the pioneering work of Shaw and coworkers,1 the synthesis
and subsequent use of pincer ligands in coordination chemistry
and catalysis has been a rapidly growing field of investigations
over the past decade. These ligands share a common feature: an
aromatic central moiety. To this unit are attached, in the ortho
positions, two arms whose electronic and steric properties can be
Scheme 1 Reagents and conditions: (i) BuLi; (ii) [Pd(COD)Cl2].
Thus, anion 2 which is straightforwardly obtained by reacting
nBuLi with phosphinine 1 reacted with [Pd(COD)Cl2] to afford
complex 3. The latter was found to be particularly active (TON
up to 10000) in the Miyaura catalyzed cross-coupling process that
allows the synthesis of aryl boronic esters from halogeno aromatic
and dialkoxyboranes.11
The coordination chemistry of these 1-R-1-P-phosphahexadienyl
anions could be extended to group 9 metal centers (Rh), for which
activation of small molecules (CO, O2, CS2) was observed.12
Copper complexes are often used in catalysis in general and
in the cyclopropanation of olefins in particular.13 On the other
hand, gold complexes have only seldom been used for catalytic
transformations, but significant results have appeared lately.14 This
prompted us to start an investigation of the coordinating behaviour
of our SPS ligands towards group 11 metal centers (Cu, Au), which
we present here.
varied almost at will. These degrees of freedom in the tuning of
the overall coordinating behaviour of these pincer ligands resulted
in numerous applications in catalysis.2 Early work mainly focused
on PCP systems but CCC, CNC, CNS, NNN, NCN, PNP, OCO,
SCS, SNS have been reported in recent years.3 Sulfur-based ligands
are particularly appealing since they can also be employed in the
elaboration of synthetic models of enzymes.4 Mainly four types
of sulfur ligands are employed: classical thioethers,5 thiolates,6
sulfoxides,7 in which coordination occurs through the remaining
lone pair at sulfur, and phosphine sulfides.8 The latter, which have
not been employed in pincer type structures so far, have found an
interesting application in the rhodium catalyzed carbonylation of
methanol when they are incorporated in mixed P–PS bidentate
ligands.9
A different way of tuning the properties of the pincer ligand
would be to replace the nitrogen or carbon atoms of the central
unit by a phosphorus atom. Indeed, it is now well established that
phosphorus in multiple bonds essentially behaves as relatively poor
-donor ligands but display a powerful -accepting capacity.10
Therefore, their electronic properties markedly differ from those of
classical phosphines and nitrogen analogs. Very different systems
incorporating phosphaalkene (P analogs of alkenes) or heterocycles
can be anticipated but, in practice, only aromatic phosphorus
heterocycles (phosphinines) present sufficient stability.
We recently reported on the synthesis of mixed S–P–S ligands
featuring a phosphinine as the central unit and two phosphine sul-
fides as ancillary ligands.11 Such systems, like 1, are easily available
in large amounts (10 g scale) by sulfurization of the corresponding
2,6-bis(diphenylphosphino)-3-phosphinines. In a first study, these
systems were found to be particularly reactive towards nucleophilic
attack because of the highly electropositive phosphorus atom of the
phosphinine moiety, either as a free ligand or complexed to a pal-
ladium center. We thus devised a rational approach towards metal
complexes through the use of 1-R-1-P-phosphahexadienyl anions
(Scheme 1).
Results and discussion
Copper complexes
Synthesis. As previously described, reaction of 1 with equimolar
amounts of MeLi in THF at −78 °C readily yielded anion 4, which
was not isolated but whose complete formation was checked by
31P NMR (Scheme 2). It is characterized by a triplet at −65.7 ppm
(PMe moiety, 2JP–P = 155.5 Hz) and a doublet at 45.9 ppm (PPh2S,
2JP–P = 155.5 Hz). The tremendous variation of the chemical shift
of the central phosphorus atom from 1 to 4 of over −320 ppm is by
itself a proof of the change of sp2 hybridization to sp3 hybridiza-
tion. On the other hand, the chemical shift of the PPh2S moiety is
not altered to a great extent:  = 2.5 (1,  = 43.4). Anion 4 was
subsequently reacted with solid CuI (or [CuBr·SMe2]). The initially
deep red solution turned orange with concomitant formation of an
orange precipitate. After one night, a 31P NMR spectrum of the
crude mixture showed the absence of any signal indicating that all
the ligand had been consumed. The product is insoluble in usual
organic solvents, and is therefore isolated pure by washing and dry-
ing. Elemental analysis shows the composition to be a multiple of
the [(SP(Me)S)Cu] moiety: 5. In order to increase the solubility of
this precipitate, the same reaction was carried out with BuLi in place
† Electronic supplementary information (ESI) available: Colour ORTEP
views of complexes 8 and 9 (50% ellipsoids). See http://www.rsc.org/
suppdata/dt/b4/b406792d/
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 4
D a l t o n T r a n s . , 2 0 0 4 , 2 5 9 3 – 2 6 0 0
2 5 9 3

View Article Online
Table 1 31P NMR chemical shifts for 2, 4, 7–13
tronic nature of the ligand L added ( ranging from 5.50 for 11 to
5.66 for 8, compared to 5.13 in [D8]-thf for 4).16 The X-ray crystal
structures recorded all show similar tetrahedral like geometry
around the Cu(I) center. Complexes 7 and 11 are presented here, and
complexes 8 and 9 are reported in the ESI.† ORTEP views of one
molecule of 7 and 11 are presented in Fig. 1 and Fig. 2, respectively,
as well as the most relevant bond lengths and angles. Crystal data
and structural refinement details are given in Table 2.
Product no.
 (P–R)a
2J (P–P)b
 (Ph2PS)a
 (P)a
2
4
7
8
9
10
11
12
13
−66.2
−65.7
−28.0
−25.5
−20.1
−19.6
−36.0
−15.5
−17.9
156.0
155.5
150.7
154.3
162.8
159.1
151.9
152.5
143.4
45.8
45.9
47.3
48.4
49.8
49.1
46.6
47.5
46.8
3.5 (PPh3)
113.7 (P(OPh)3)
a In ppm. b In Hz.
of MeLi. Similarly, the formation of the anion 2 was observed which
2
appeared as a triplet at −66.2 ppm (PBu moiety, JP–P = 156.0 Hz)
and a doublet at 45.8 ppm (PPh2S, 2JP–P = 156.0 Hz). Anion 2 was
also reacted with CuI (or [CuBr·SMe2]), and here again an insoluble
orange solid formed: 6.
Scheme 2 Reagents and conditions: (i) RLi, THF, −78 °C to rt; (ii) CuI,
THF, −78 °C to rt.
Fig. 1 ORTEP view of complex 7 (50% ellipsoids). The numbering is ar-
bitrary and different from that used in NMR data. Selected bond lengths (Å)
and angles (°): P1–C1 1.794(2), C1–C2 1.410(2), C2–C3 1.406(2), C1–P2
1.781(2), P2–S1 2.0021(6), P1–C6 1.827(2), P1–Cu1 2.2377(5), S1–Cu1
2.3694(5), S2–Cu1 2.4321(6), Cu1–C43 1.887(2), C43–N1 1.154(2), P1–
Cu1–C43 143.39(6), S2–Cu1–S1 112.64(2), N(1)–C(43)–Cu(1) 170.9(2),
C(43)–N(1)–C(44) 176.4(2).
Despite their insolubility, these complexes reacted with several
two-electron donor ligands, such as isocyanides, phosphines,
phosphites or even pyridine to give soluble tetrahedral Cu(I) orange
complexes in quasi-quantitative yields (Scheme 3).
Scheme 3 Reagents and conditions: n L, CH2Cl2, rt. 7: R = Me, L = 2,6-
dimethylphenylisocyanide; 8: R = Me, L = t-butylisocyanide; 9: R = Me,
L = PPh3; 10: R = Me, L = P(OPh)3; 11: R = Me, L = Py; 12: R = n-Bu,
L = 2,6-dimethylphenylisocyanide; 13: R = n-Bu, L = t-butylisocyanide.
These were fully characterized by NMR spectroscopy and
elemental analysis and X-ray crystal studies were performed for
7, 8, 9 and 11. In the complexes 7–11, the 31P NMR signals of the
SPS moiety all appeared as expected sets of triplet and doublet
at around −25 ppm (P–Me) and 49 ppm (PPh2S), respectively
(Table 1), with a large 2J(P–P) coupling constant of about 150 Hz,
and the complexes 12–13 as triplet and doublet at around −16 ppm
(P–Bu) and 47 ppm (PPh2S), respectively. One can note that the
signal of the PPh2S appeared as sharp lines whereas the other signals
appeared broad. The variation of the chemical shifts for the P(R)
moiety between 7 and 12 ( = 12.5), and 8 and 13 ( = 7.6) is
consistent with the change of chemical shifts between PPh2Me and
PPh2Bu ( = 10).15
Fig. 2 ORTEP view of complex 11 (50% ellipsoids). The numbering
is arbitrary and different from that used in NMR data. Phenyl groups
have been omitted for clarity. Selected bond lengths (Å) and angles (°):
P1–C1 1.797(2), C1–C2 1.403(3), C2–C3 1.413(3), C1–P2 1.776(2), P2–S1
2.0072(8), P1–C4 1.835(3), C1′–P1–C1 101.5(1), P1–Cu1 2.2088(8),
S1–Cu1 2.3621(6), Cu1–N1 2.026(3), P1–Cu1–N1 124.5(1), S1–Cu1–S1′
101.35(3).
The signal of the P(Me) for the pyridine complex 11 resonated at
higher frequency than in the other complexes, indicating a higher
partial charge at the phosphorus (compared to −65.7 for anion 4).
In 1H NMR, the chemical shift for H4 (para to the P atom in the 5
phosphinine subunit) was not modified significantly by the elec-
As noted above, the two structures show tetrahedral like geo-
metry. The angles at the copper center deviate from the ideal 109°
to a different extent in these two complexes. In complex 7, the angle
2 5 9 4
D a l t o n T r a n s . , 2 0 0 4 , 2 5 9 3 – 2 6 0 0

View Article Online
Table 2 Crystal data and refinement parameters for 7–9, 11, 16 and 17
7
8
9
11
16
17
Formula
Molecular weight
Crystal system
Space group
C51H43CuNP3S2
890.43
Triclinic
P1
C47H43CuNP3S2
842.39
Orthorhombic
P212121
C62H53.50ClCuP4S2
1085-54
Triclinic
P1
C62H54CuN4P3S2
1075.66
Monoclinic
Cm
C85H70Au2Cl2P6S4
1870.30
Monoclinic
C2/c
C60H49AuP4S2
1154.96
Triclinic
P1
a/Å
b/Å
c/Å
/°
/°
/°
10.8652(2)
13.2635(2)
15.6616(3)
86.741(7)
73.493(6)
87.071(8)
2
0.751
18811
9720
0.1096
8.773(5)
12.595(5)
39.454(5)
15.7730(10)
17.3780(10)
20.3450(10)
75.4440(10)
78.7570(10)
89.2700(10)
4
0.704
39341
21349
0.1197
15.4940(10)
17.1260(10)
11.6850(10)
13.631(1)
19.509(1)
28.894(1)
12.595(5)
13.589(5)
17.267(5)
70.440(5)
68.800(5)
70.900(5)
2
3.164
16814
8806
0.0811
121.4650(10)
91.8768(1)
Z
4
2
4
/cm−1
0.740
10282
9129
0.0964
0.0384
0.628
6931
6675
0.0986
0.0386
4.165
19109
8163
0.1038
0.0346
Reflections measured
Reflections used
wR2
R1
0.0403
0.0412
0.0401
Table 3 IR spectra for copper isocyanide complexes
S2–Cu1–S1 measures 112.64(2)° and P1–Cu1–C43 143.39(6)°
whereas in complex 11 S1–Cu1–S1′ measures 101.35(3)° and
P1–Cu1–N1 124.5(1)° which shows that 6 is more distorted. The
most interesting piece of data is given by the external P–C bond
lengths (C1–P2 1.781(2) Å C5–P3 1.769(2) Å in 7 and C1–P2
1.776(2) Å in 11) which are clearly shorter than in the free ligand
(P–C 1.826(2) Å),16 and therefore have gained significant double
bond character. Moreover, they are shorter than the internal P–C
bond lengths (P1–C1 1.794(2), Å C5–P1 1.797(2) Å in 7 and P1–C1
1.797(2) Å in 11). These internal P–C bonds are now much longer
than in the free ligand (1.742(2) Å) but compare to those in reported
square planar Pd(II) complexes (1.776(3) Å).11 Nevertheless, these
values are still much shorter than the P–Me bond length (P1–C6
1.827(2) Å in 7 and C4–P1 1.835(3) Å in 11). The PS bond
lengths (P2–S1 2.0021(6) Å, P3–S2 2.0014(6) Å for 7 and P2–S1
2.0072(8) Å for 11) are in turn slightly elongated when compared to
free ligand (1.956(1) Å).16 From this data and the data collected with
the Pd and Ni complexes reported earlier it is relatively difficult to
establish whether the ligand behaves as an anionic 4-phosphinine
(form A) or as classical tertiary phosphine (form B) (Scheme 4).16
On the one hand, short internal P–C bond lengths suggest that the
ylidic structure of the phosphinine has been restored upon complex-
ation (formA) but, on the other hand, relatively long P–S bonds and
short external P–C connections point out that delocalization takes
place within the unsaturated part of the ligand (form B). Reason-
ably, one may propose that the two forms can contribute to the bond-
ing, the respective contributions being determined by the nature of
the metal centre and its formal oxidation state.
Complex
(CN)a
Complex
(CN)a
Free L (CN)a,b
7
8
2131
2174
12
13
2130
2158
2122
2140 (CH2Cl2)
2138
a In cm−1. b See ref. 18.
by DFT calculations.17 It was thus shown that in these complexes,
there was negligible -back donation from Cu(I) into the ligand.
This implied that the coordination is only based on  donation.
Reactivity. In a first step, based on literature precedence,
we focused our studies on the catalytic cyclopropanation of
alkenes. We had selected the cyclopropanation of styrene with
ethyldiazoacetate. In complexes 7–10 and 12–13, the coordination
sphere at the copper center is saturated and accordingly they proved
rather robust. We thus envisioned the use of either the insoluble
complex 5 or the pyridine complex 11 for further reactions.
Indeed, as shown above, complex 5 reacted with 2 electron
donors to yield mononuclear complexes and we had hoped that,
even though alkenes are weaker ligands than pyridine for copper
complexes, excess of alkenes would lead to a soluble mononuclear
complex. We also hoped that the pyridine ligand in 11 would be
easily displaced and therefore that this complex would serve as a
masked unsaturated complex, which is a prerequisite for catalytic
reactions.
However, addition of large amounts of styrene to suspensions
of complex 4 in various solvents did not lead to any dissolution
(even partial) as indicated by 31P NMR spectroscopy, nor did
styrene alone displaced pyridine in complex 10. This prompted us
to study the reaction of these complexes with the diazoacetate alone
(Scheme 5).
Scheme 4
Interestingly also in complex 7, the isocyanide ligand is not
perfectly linear: C43–N1–C44 measures 176.4° and N1–C43–Cu1
measures 170.9° which points towards a small change in hybridiza-
tion at C43. In this case a Cu1–C43 showing a partial double bond
character would be expected. In fact, the bond length Cu1–C43 at
1.887(2) Å is normal and comparable with the Cu1–C43 measured
in complex 7 (1.890(2) Å, see ESI†), for which the N1–C43–Cu1 is
almost linear at 177.4(3)°. These two results seem to contradict each
other and a close inspection of the IR CN stretches was warranted.
The results are given in Table 3. Both complexes 7 and 12 showed
only a marginal increase in the value when compared to the free iso-
cyanide. On the other hand, somewhat unexpectedly, the increase in
the stretch is significant for complex 13 (18 cm−1) and large for com-
plex 8 (36 cm−1). This type of increase has been observed previously
in the case of [Cu(CO)n]+ (n = 1,2) complexes, and was investigated
Scheme 5 Reagents and conditions: 2 eq. ethyldiazoacetate, CH2Cl2.
D a l t o n T r a n s . , 2 0 0 4 , 2 5 9 3 – 2 6 0 0
2 5 9 5

View Article Online
The first surprising point was the need of two equivalents of diazo
per copper center to obtain a clean reaction. Indeed, after addition
of the diazo, the solution turned dark orange and a unique product
formed characterized by a set of triplet at 18.2 ppm (PMe moiety,
2JP–P = 42.0 Hz) and doublet at 35.7 ppm (PPh2S, 2JP–P = 42.0 Hz) in
31PNMR.At this point, we had hoped to obtain a new stable copper–
carbene complex, which are quite rare species,19 but the very small
value of the coupling constant compared to the other complexes
perturbed us. This species proved to be quite stable and soluble in
all common organic solvents. It was purified by chromatography on
silica gel and fully characterized by NMR spectroscopy. The forma-
tion of the previously unknown 5-phosphinine 14 was proved by the
following data: presence of a new CH2 signal in 13C NMR coupled
with phosphorus ( 34.5, 1JP–C = 42.2 Hz). This moiety appears in 1H
NMR also as a doublet at 3.3 ppm (2JH–P = 16.7 Hz). The CH3–CH2
unit is also visible as an expected set of triplet-quartet at 1.2 and
4.1 ppm, respectively. The H4 (proton para to the P atom) appeared
as the usual weakly coupled triplet at 5.50 ppm (4JH–P = 4.8 Hz).
In short, the addition of the diazo resulted in a clean oxidation of
the Cu(I) species which resulted in the loss of the ligand. It is the
first time that we observed such a phenomenon with all of the SPS
complexes we had obtained so far. Indeed, for example, oxidation of
Rh(I) centers resulted in the isolation of Rh(III) species and heating
Pd(II) complexes in oxygenated and oxidizing solvents did not lead
to decomposition. The reactivity towards different oxidizing agents
was therefore studied. Dissolving complex 11 in CDCl3 and follow-
ing the reaction by both 31P NMR and 1H NMR provided the follow-
ing information. After one night at room temperature, about ꢀ of
the starting complex had evolved into two new products in 5/1 ratio
characterized by a doublet at 31.9 ppm (PPh2S, 2JP–P = 26.7 Hz) and
is observed for the methyl substituent and a triplet at 5.72 ppm is
observed for the H4 proton of the phosphinine ring. X-Ray quality
crystals were obtained after diffusion of hexanes into a CDCl3 solu-
tion of the complex. These were subjected to X-ray structure analy-
sis which definitely proved the complex to be a dimer. An ORTEP
view of one molecule of 16 is presented in Fig. 3 as well as the most
relevant bond lengths and bond angles. Crystal data and structural
refinement details are given in Table 2.
Scheme 7 Reagents and conditions: [AuCl·SMe2], THF, −78 °C to rt.
2
a triplet at 42.8 ppm (PMe moiety, JP–P = 26.7 Hz) for the major
2
species and a doublet at 35.9 ppm (PPh2S, JP–P = 43.7 Hz) and a
triplet at 66.2 ppm (PMe moiety, 2JP–P = 43.7 Hz) for the minor spe-
cies (15). After two days, all of the starting complex had evolved
to a mixture of four phosphorus containing products, which could
not be separated. One of the by products of this reaction, 15, is the
known 1-Me-1-Cl-5-phosphinine which was prepared alone by
oxidation of 11 with equimolar amounts of C2Cl6 (Scheme 6).
Fig. 3 ORTEP view of complex 16 (50% ellipsoids). The numbering is
arbitrary and different from that used in NMR data. Phenyl groups have
been omitted for clarity. Selected bond lengths (Å) and angles (°): P(1)–C(1)
1.759(3), C(1)–C(2) 1.431(4), C(2)–C(3) 1.369(4), C(3)–C(4) 1.441(4),
C(4)–C(5) 1.377(4), P(1)–C(5) 1.829(3), P(1)–C(6) 1.837(3), P(2)–C(1)
1.746(3), S(1)–P(2) 2.037(1), C(1)–P(1)–C(5) 102.6(2), Au(1)–Au(1′)
3.0481(2), P(1′)–Au(1)–S(1) 173.87(3).
Scheme 6 Reagents and conditions: C2Cl6, CH2Cl2, rt.
We then turned our attention to gold(I) complexes whose coordi-
nation chemistry towards 4-phosphinine anions was not known.
As envisioned, the geometry around the Au(I) centers is (almost)
linear dicoordinate, the sulfur of one ligand located trans to the
phosphorus atom of the other ligand. Apparent in the structure is the
aurophilic interaction between the two gold centers (Au(1)–Au(1′)
3.0481(2) Å). This interaction is strong enough to force the geometry
to deviate slightly from linearity (P(1′)–Au(1)–S(1) 173.87(3) Å).
Interesting results were provided by the external P–C bond lengths.
In fact, they are very different within the ligand itself: P(2)–C(1)
measures 1.746(3) Å whereas P(3)–C(5) measures 1.799(3) Å.
The first one has now gained a strong double bond character (com-
parable to delocalized internal PC bond in the free phosphinine
1: 1.742(2) Å) whereas the second one is only slightly shorter than
in the free ligand (P–C 1.826(2) Å). This desymmetrization is
also found in the P–S bonds: S(4)–P(3) 1.967(1) Å vs. S(1)–P(2)
2.037(1) Å. Concerning the internal P–C bonds, here also P(1)–C(1)
1.759(3) Å is much shorter than P(1)–C(5) 1.829(3) Å, which is in
turn similar to the P(1)–C(6) 1.837(3) Å bond and therefore can
be considered as a real single bond. Bond alternance is also found
for the C–C bonds: C(1)–C(2) 1.431(4) Å, C(2)–C(3) 1.369(4) Å,
Gold complexes
The anion 4 was prepared as above and reacted with equimolar
amounts of the [AuCl·SMe2] precursor at low temperature followed
by warming to room temperature (Scheme 7). The initial red solu-
tion turned orange and 31P NMR showed complete formation of a
single new product after one hour. Three sets of signals appeared
at 45 (doublet), 39 (doublet) and −12 ppm (doublet of doublet).
These coupling figures could only be accounted for by postulating
a structure in which one sulfur ligand would bind to a gold center.
In turn, knowing that the geometry of AuL2 complexes is linear, a
dimeric structure had to be envisioned. In fact, after isolation of
the pure complex, accumulation on a smaller spectral window in
31P revealed that the signal at −12 ppm is in fact not a doublet of
doublets but a much more complicated feature consistent with the
dimeric structure proposed for which an AA′MM′XX′ spin system
is expected. On the other hand, the two signals at 45 and 39 ppm
1
remain deceptively simple (doublets). The H NMR spectrum is
also more simple than expected and only a doublet at 1.20 ppm
2 5 9 6
D a l t o n T r a n s . , 2 0 0 4 , 2 5 9 3 – 2 6 0 0

View Article Online
C(3)–C(4) 1.441(4) Å, C(4)–C(5) 1.377(4) Å, which gives an over-
all electronic picture for the dimer as follows (Scheme 8).
This dimer can be cleaved by two electron donors as in the case
of copper complexes (Scheme 9).
Scheme 8
Fig. 4 Caption ORTEP view of complex 17 (50% ellipsoids). The num-
bering is arbitrary and different from that used in NMR data. Phenyl groups
have been omitted for clarity. Selected bond lengths (Å) and angles (°):
P(1)–C(1) 1.799(2), C(1)–C(2) 1.400(3), C(2)–C(3) 1.417(3), C(3)–C(4)
1.402(3), C(4)–C(5) 1.404(3), P(1)–C(5) 1.776(2), P(2)–C(1) 1.787(2),
P(3)–C(5) 1.771(3), S(1)–P(2) 1.983(1), S(2)–P(3) 1.9709(8), Au(1)–P(1)
2.3080(6), Au(1)–S(1) 2.7903(6), Au(1)–P(4) 2.2862(7), C(5)–P(1)–C(1)
102.3(1), P(4)–Au(1)–P(1) 162.93(2), P(4)–Au(1)–S(1) 108.90(2).
Scheme 9
1.776(2) Å, P(1)–C(1) 1.799(2) Å, respectively), and C–C bonds.
As already indicated by 31P NMR, the two phosphorus atoms are lo-
cated trans to each other (angle P(4)–Au(1)–P(1) 162.93(2)°). The
deviation from linear geometry is caused by the Au–S interaction.
Apart from that the structure does not deserve further comments.
The reaction of the dimer with a threefold excess of the iso-
cyanide, on the other hand, was sluggish and took several hours at
room temperature to yield an orange complex, 18. Once isolated,
the complex partially reverted (≈20%) to the starting complex when
redissolved, pointing towards a weak interaction between the iso-
cyanide ligand and the gold center. This reflects the poorer  donor
ability of isocyanides and confirms the negligible  back donating
ability of gold(I) centers. In the same vein, it is worth noting that the
dimer was not cleaved by excesses of pyridine, alkynes or alkenes.
When PPh3 was added to a CH2Cl2 solution, a new complex, 17,
formed within minutes at room temperature. This complex is char-
acterized by three sets of peaks in 31P NMR: a doublet of triplets at
3.0 ppm (PPh3, 2JP–P = 286 Hz, 3JP–P = 91 Hz), a doublet at 41.5 ppm
2
3
(PMe, JP–P = 286 Hz) and a doublet at 42.5 ppm (PPh2S, JP–P
=
91 Hz). The very large coupling constant of 286 Hz is indicative of
the two ligands located trans to each other. This complex was also
characterized by 1H and 13C NMR spectroscopy, elemental analysis
and its 3-D structure obtained by X-ray crystallographic methods.
1
As for the copper complexes, in H NMR, the H4 (para to the P
atom of the phosphinine moiety) is observed as a triplet at 5.53 ppm
(4JP–H = 10.9 Hz) and therefore is not coupled to the PMe.
This complex was also obtained via a second method starting
from the [AuPPh3Cl] precursor, as shown below (Scheme 10).
Conclusion
We have presented here the first examples of SPS pincer complexes
with group 11, d10 (Cu(I) and Au(I)) metal centers. All of the Cu(I)
complexes proved very stable except for complex 11. Because of
a weakly bound ligand (pyridine), this complex proved reactive
towards a variety of oxydizing agents (oxygen, chlorinated solvents,
C2Cl6). A quite unexpected increase in the (CN) stretches in IR for
isocyanide complexes (7, 8, 12, 13) was measured which implied a
negligible  back donation from filled d orbitals at the Cu(I) center
into the empty * orbitals of the isocyanide ligand, and therefore
a  donation only coordination. A dimeric complex of Au(I) was
synthesized, 16, for which a gold–gold interaction was observed by
X-ray crystallographic studies. This dimer was readily cleaved as
expected by two electron donors. The isocyanide complex 18 was
shown to revert back to the dimer which suggests a weak  donation
only type interaction between Au and CNR.
Scheme 10 Reagents and conditions: [AuClPPh3], rt.
X-Ray quality crystals were obtained after diffusion of hexanes
into a CH2Cl2 solution of the complex. An ORTEP view of one mol-
ecule of 17 is presented in Fig. 4 as well as the most relevant bond
lengths and bond angles. Crystal data and structural refinement
details are given in Table 2.
Experimental
The first interesting point of this structure is the presence of a
S–Au bond which gives an overall T-shape geometry. This bond is
rather long at 2.7903(6) Å, which already points towards its weak-
ness. Obviously in solution this geometry is not preserved as it
would lead to electronically different PPh2S moieties and therefore
different signals which we did not observe. The two Au–P bond
distances of 2.2862(7) and 2.3080(6) Å are normal. Unlike what
was observed in the dimer 16, the 4-phosphinine subunit is almost
symmetrical with nearly identical external and internal P–C bonds
(P(2)–C(1) 1.787(2) Å and P(3)–C(5) 1.771(3) Å, and P(1)–C(5)
General
All reactions were routinely performed under an inert atmosphere
of argon or nitrogen by using Schlenk and glove-box techniques
and dry deoxygenated solvents. Dry THF and hexanes were
obtained by distillation from Na/benzophenone and dry ether
from CaCl2 and then NaH and dry CH2Cl2 from P2O5. CDCl3
was dried from P2O5 and stored on 4 Å Linde molecular sieves.
CD2Cl2, [D5]-pyridine were used as purchased and stored in the
glovebox. Nuclear magnetic resonance spectra were recorded on
D a l t o n T r a n s . , 2 0 0 4 , 2 5 9 3 – 2 6 0 0
2 5 9 7

View Article Online
a Bruker AC-200 SY spectrometer operating at 300.0 MHz for
1H, 75.5 MHz for 13C and 121.5 MHz for 31P. Solvent peaks are
used as internal reference relative to Me4Si for 1H and 13C chemi-
cal shifts (ppm); 31P chemical shifts are relative to a 85% H3PO4
external reference. Coupling constants are given in hertz. The
following abbreviations are used: s, singlet; d, doublet; t, triplet; q,
quadruplet; p, pentuplet; m, multiplet; v, virtual; b, broad. Mass
spectra were obtained at 70 eV with a HP 5989B spectrometer
coupled to a HP 5980 chromatograph by the direct inlet method.
Elemental analyses were performed by the ‘Service d’analyse du
CNRS’, at Gif sur Yvette, France. Phosphinine 1,11 anions 211 and
416 and [AuClPPh3],20 [AuCl·SMe2],21 were prepared according to
reported procedures.
2J (PA–PX) = 150.7, PXPh2). 13C NMR (75.5 MHz, CDCl3, 298 K):
1
3
 = 12.2 (dt, J (C–P) = 12.2, J (C–P) = 8.2, CH3 of P–Me), 30.5
(s, CH3 of t-Bu), 56.0 (s, C of t-Bu), 66.41 (ddd, J (C–P) = 84.8,
3
J (C–P) = 26.5, J (C–P) = 2.9, C2,6), 119.7 (td, J (C–P) = 10.9,
3J (C–P) = 5.5, C4H), 127.1–132.8 (m, CH of Ph), 136.3 (d, J
(C–P) = 80.5, C of Ph), 137.4 (d, J (C–P) = 85.2, C of Ph), 143.6 (m,
C of Ph), 153.0 (m, C3,5). C of isocyanide was not seen. IR (KBr): 
2175 cm−1. C47H43P3S2NCu (842.46): calcd. C 67.01, H 5.14; found
C 67.25, H 5.35%.
Copper complex 9
To a suspension of complex 5 (100 mg, 0.13 mmol) in CH2Cl2
(5 mL) was added solid PPh3 (34.1 mg, 0.13 mmol) which resulted
in the dissolution of the starting complex. The 31P NMR spectrum
of the crude mixture showed the formation of a unique product. The
solution was then taken to dryness. The product was washed with
hexanes and dried to yield an orange solid. Yield: 93%, 123 mg.
Copper complex 4
A solution of MeLi in Et2O (1.25 mL, C = 0.16 M, 2.0 mmol) was
syringed into a solution of 1 (1.36 g, 2.0 mmol) in THF (30 mL)
at −78 °C. The solution was warmed to room temperature and
stirred for 20 min. 31P NMR indicated then the complete formation
of anion 4, to which solid CuI (380 mg, 2.0 mmol) was added.
The initially red solution turned rapidly orange with concomitant
formation of an orange precipitate. The mixture was stirred for an
additional 15 h. The solid was then filtered, washed with THF and
dried under vacuum. Yield: 92%, 1.40 g. [C42H34P3S2Cu]n (759.33)n:
calcd. C 66.43, H 4.51; found C 66.63, H 4.75%.
2
1H NMR (300 MHz, CDCl3, 298 K):  = 1.30 (d, J (H–P) = 2.0,
3H, CH3), 5.60 (t, 2J (H–P) = 4.0, 1H, H4), 6.70–7.9 (m, 30H, CH
of Ph). 31P NMR (121.5 MHz, CDCl3, 298 K):  = −20.1 (b, PA),
2
3.5 (b, PPh3), 49.8 (AX2, d, J (PA–PX) = 162.8, PXPh2). 13C NMR
1
3
(75.5 MHz, CDCl3, 298 K):  = 11.8 (dt, J (C–P) = 11.8, J (C–
P) = 9.9, CH3 of P–Me), 66.4 (m, C2,6), 119.7 (m, C4H), 127.1–134.4
(m, CH of Ph), 135.1 (dd, J (C–P) = 29.0, J (C–P) = 2.9, C of Ph),
136.4 (d, J (C–P) = 86.9, C of Ph), 137.6 (d, J (C–P) = 86.2, C of
Ph), 144.1 (m, C of Ph), 152.8 (m, C3,5). C60H48P4S2Cu (1021.62):
calcd. C 70.54, H 4.73; found C 70.90, H 5.03%.
Copper complex 6
A solution of BuLi in hexanes (1.25 mL, C = 0.16 M, 2.0 mmol)
was syringed into a solution of 1 (1.36 g, 2.0 mmol) in THF (30 mL)
at −78 °C. The solution was warmed to room temperature and stirred
for 20 min. 31PNMR indicated then the complete formation of anion
2, to which solid CuI (380 mg, 2.0 mmol) was added. The initially
red solution turned rapidly orange with concomitant formation of
an orange precipitate. The mixture was stirred for an additional
15 h. The solid was then filtered, washed with thf and dried under
vacuum. Yield: 89%, 1.40 g. [C42H34P3S2Cu]n (801.40)n: calcd. C
67.44, H 5.03; found C 67.69, H 5.31%.
Copper complex 10
To a suspension of complex 5 (100 mg, 0.13 mmol) in CH2Cl2
(5 mL) was syringed P(OPh)3 (34.6 L, 0.13 mmol) which resulted
in the dissolution of the starting complex within 5 min. The 31P
NMR spectrum of the crude mixture showed the formation of
a unique product. The solution was then taken to dryness. The
product was washed with hexanes and dried to yield an orange
1
solid. Yield: 93%, 129 mg. H NMR (300 MHz, CDCl3, 298 K):
 = 0.91 (d, 2J (H–P) = 2.2, 3H, CH3), 5.53 (t, 2J (H–P) = 4.2, 1H,
H4), 6.70–7.9 (m, 30H, CH of Ph). 31P NMR (121.5 MHz, CDCl3,
298 K):  = −19.6 (b, PA), 49.8 (AM2X, dd, 2J (PA–PM) = 159.1, 2J
(PM–PX) = 19.5, PMPh2), 113.7 (b, P(OPh)3). 13C NMR (75.5 MHz,
CDCl3, 298 K):  = 10.8 (m, CH3), 66.34 (ddd, J (C–P) = 86.5, J
(C–P) = 25.5, J (C–P) = 2.2, C2,6), 115.9 (s, CH of Ph), 119.5 (m,
C4H), 121.5–133.1 (m, CH of Ph), 135.9 (d, J (C–P) = 77.5, C of
Ph), 136.8 (d, J (C–P) = 84.7, C of Ph), 143.6 (d, J (C–P) = 2.8,
C of Ph), 151.4.6 (bs, C of Ph), 152.8 (m, C3,5). C60H48P4S2O3Cu
(1069.62): calcd. C 67.37, H 4.52; found C 67.43, H 4.72%.
Copper complex 7
To a suspension of complex 5 (100 mg, 0.13 mmol) in CH2Cl2
(5 mL) was syringed xylyl-isocyanide (17 mg, 0.13 mmol) resulting
in an instantaneous dissolution of the solid. The 31P NMR spectrum
of the crude mixture showed the formation of a unique product. The
solution was then taken to dryness. Yield: 95%, 110 mg. 1H NMR
(300 MHz, CDCl3, 298 K):  = 1.44 (d, 2J (H–P) = 2.5, 3H, CH3),
2.19 (s, 6H, CH3), 5.65 (t, 2J (H–P) = 4.1, 1H, H4), 6.70–7.9 (m, 33H,
CH of Ph). 31P NMR (121.5 MHz, CDCl3, 298 K):  = −25.5 (AX2,
t, 2J (PA–PX) = 154.3, PA), 48.4 (AX2, d, 2J (PA–PX) = 154.3, PXPh2).
13C NMR (75.5 MHz, CD2Cl2, 298 K):  = 11.2 (dt, 1J (C–P) = 12.1,
3J (C–P) = 9.1, CH3 of P–Me), 18.8 (s, CH3 of xylyl), 65.6 (dd, J
(C–P) = 85.31, J (C–P) = 23.4, C2,6), 119.6 (m, C4H), 127.0–132.6
(m, CH of Ph), 135.3 (s, C of Ph), 135.7 (d, J (C–P) = 86.7, C of
Ph), 136.7 (d, J (C–P) = 78.8, C of Ph), 143.5 (d, J (C–P) = 3.8, C
of Ph), 152.9 (d, J (C–P) = 8.3, C3,5). C of isocyanide was not seen.
IR (KBr):  2131 cm−1. C51H43P3S2NCu (890.51): calcd. C 68.79, H
4.87; found C 69.01, H 5.10%.
Copper complex 11
Complex 5 (50 mg, 0.065 mmol) was dissolved in pyridine to give
a dark orange solution. The 31P NMR spectrum of the crude mixture
shows the formation of a unique product. The solution was then
taken to dryness. The product was washed with hexanes and dried
to yield an orange solid. Yield: 100%, 129 mg. 1H NMR (300 MHz,
2
CDCl3, 298 K):  = 1.31(s, 3H, CH3), 5.50 (t, J (H–P) = 4.3, 1H,
H4), 6.70–7.9 (m, 30H, CH of Ph). 31P NMR (121.5 MHz, CDCl3,
298 K):  = −36.0 (b, PA), 46.6 (AX2, d, 2J (PA–PX) = 151.9, PXPh2).
13C NMR (75.5 MHz, C5D5N, 298 K):  = 11.4 (dt, 1J (C–P) = 9.2,
3J (C–P) = 7.9, CH3), 68.1 (ddd, J (C–P) = 86.6, J (C–P) = 26.0, J
(C–P) = 3.3, C2,6), 119.0 (td, 3J (C–P) = 10.6, 3J (C–P) = 5.0, C4H),
123.7 (s, CH of Py), 126.9–132.7 (m, CH of Ph), 135.7 (s, CH of
Py), 136.6 (d, J (C–P) = 81.4, C of Ph), 137.8 (d, J (C–P) = 85.8, C
of Ph), 144.0 (vq, J (C–P) = 3.4, C of Ph), 149.9 (s, CH of Py), 153.1
(m, ∑ J = 12.4, C3,5). C47H39P3S2NCu (838.43): calcd. C 67.33, H
4.68; found C 67.63, H 4.72%.
Copper complex 8
To a suspension of complex 5 (100 mg, 0.13 mmol) in CH2Cl2
(5 mL) was syringed tert-butylisocyanide (14 L, 0.13 mmol)
resulting in an instantaneous dissolution of the solid. The 31P NMR
spectrum of the crude mixture showed the formation of a unique
product. The solution was then taken to dryness. The product was
washed with hexanes and dried to yield an orange solid. Yield:
96%, 110 mg. 1H NMR (300 MHz, CDCl3, 298 K):  = 1.37 (s, 9H,
CH3), 1.41 (d, 2J (H–P) = 2.2, 3H, CH3), 5.66 (t, 2J (H–P) = 4.0, 1H,
H4), 6.70–7.9 (m, 30H, CH of Ph). 31P NMR (121.5 MHz, CDCl3,
Copper complex 12
To a suspension of complex 6 (104 mg, 0.13 mmol) in CH2Cl2
(5 mL) was syringed xylyl-isocyanide (17 mg, 0.13 mmol) resulting
2
298 K):  = −28.0 (AX2, t, J (PA–PX) = 150.7, PA), 47.3 (AX2, d,
2 5 9 8
D a l t o n T r a n s . , 2 0 0 4 , 2 5 9 3 – 2 6 0 0

View Article Online
in an instantaneous dissolution of the solid. The 31P NMR spectrum
of the crude mixture showed the formation of a unique product. The
solution was then taken to dryness. Yield: 96%, 116 mg. 1H NMR
(300 MHz, CDCl3, 298 K):  = 0.91 (t, 3J (H–H) = 7.1, 3H, CH3 of
Bu), 1.52 (m, ∑ J = 33.9, 4H, CH2), 1.84 (m, ∑ J = 19.7, 2H, CH2),
Method B: A solution of MeLi in Et2O (125 L, C = 0.16 M,
0.2 mmol) was syringed into a solution of 1 (136 mg, 0.2 mmol) in
THF (3 mL) at −78 °C. The solution was warmed to room tempera-
ture and stirred for 20 min. 31P NMR indicates then the complete
formation of anion 4, to which [AuClPPh3] (99 mg, 0.2 mmol)
dissolved in THF (2 mL) was added at −78 °C. The solution was
warmed back to room temperature to yield an orange solution which
was then stirred for 1 h. The volatiles were removed in vacuo and the
compound extracted in CH2Cl2. After filtration and drying, the title
complex was recovered as an orange powder. Yield 92%, 212 mg.
1H NMR (300 MHz, CDCl3, 298 K):  = 1.55 (d, 2J(H–P) = 7.7,
2
2.22 (s, 6H, CH3), 5.59 (t, J (H–P) = 4.0, 1H, H4), 6.70–7.75 (m,
33H, CH of Ph). 31P NMR (121.5 MHz, CD2Cl2, 298 K):  = −15.5
(AX2, t, 2J (PA–PX) = 152.5, PA), 47.5 (AX2, d, 2J (PA–PX) = 152.5,
PXPh2). 13C NMR (75.5 MHz, CD2Cl2, 298 K):  = 13.1 (s, CH3 of
Bu), 17.6 (s, CH3 of xylyl), 22.4 (dd, J (C–P) = 15.9, J (C–P) = 8.3,
CH2), 23.2 (d, J (C–P) = 14.3, CH2), 25.6 (d, J (C–P) = 10.6, CH2),
63.6 (m, C2,6), 119.5 (m, C4H), 126.0–131.5 (m, CH of Ph), 127.8 (s,
C of Ph), 134.3 (s, C of Ph), 135.2 (d, J (C–P) = 81.5, C of Ph), 135.9
(d, J (C–P) = 85.7, C of Ph), 142.2 (s, C of Ph), 152.4 (s, C3,5). C of
isocyanide was not seen. IR (KBr):  2129.9 cm−1. C54H49P3S2NCu
(932.57): calcd. C 69.55, H 5.30; found C 69.34, H 5.07%.
2
3H, CH3), 5.53 (t, J (H–P) = 10.9, 1H, H4), 6.70–7.9 (m, 30H,
CH of Ph). 31P NMR (121.5 MHz, CDCl3, 298 K):  = 2.17 (dt,
AMX2, 2J (PA–PM) = 286.0, 2J (PA–PX) = 90.9, PA), 41.46 (d, AMX2,
2
2J (PA–PM) = 286.0, PM), 42.55 (d, AMX2, J (PA–PX) = 90.9, PX).
13C NMR (75.5 MHz, CDCl3, 298 K):  = 15.5 (s, CH3), 69.5 (m,
C2, 6), 118.8.6 (td, 3J (C–P) = 9.6, 3J (C–P) = 5.8, C4H), 128.8–135.0
(m, CH of Ph), 137.7 (d, J (C–P) = 85.8, C of Ph), 137.8 (d, J
(C–P) = 85.8, C of Ph), 144.1 (t, J (C–P) = 6.4, C of Ph), 155.7
(bs, C3, 5). C60H49P4S2Au (1155.04): calcd. C 62.39, H 4.28; found
C 62.60, H 4.62%.
Copper complex 13
To a suspension of complex 6 (104 mg, 0.13 mmol) in CH2Cl2
(5 mL) was syringed tert-butylisocyanide (14 L, 0.13 mmol) re-
sulting in an instantaneous dissolution of the solid. The 31P NMR
spectrum of the crude mixture showed the formation of a unique
product. The solution was then taken to dryness. The product was
washed with hexanes and dried to yield an orange solid. Yield:
Gold complex 18
Dimeric complex 16 (200 mg, 0.11 mmol) was dissolved in CH2Cl2
(5 mL) and tert-butylisocyanide (73 L, 0.66 mmol) was syringed
in. The solution was then stirred at room temperature for several
hours. 31P NMR indicated the complete formation of the desired
complex 15. The volatiles were then removed under vacuum. The
solid was dissolved in CHCl3 and hexanes was layered on top of
the solution. The orange crystals which deposited were dissolved
in CDCl3 showing the partial reformation of the starting material
(20%). The 1H NMR spectrum was thus recorded with the mixture.
1H NMR (300 MHz, CDCl3, 298 K):  = 1.35 (s, 9H, CH3),
1
94%, 110 mg. H NMR (300 MHz, CDCl3, 298 K):  = 0.91 (t,
3J (H–H) = 7.0, 3H, CH3 of Bu), 1.24 (bs, 2H, CH2), 1.42 (m, ∑
J = 41.2, 20H, CH2 of Bu and CH3 of t-Bu), 1.79 (m, ∑ J = 16.0,
2H, CH2), 5.56 (t, 2J (H–P) = 4.2, 1H, H4), 6.70–7.72 (m, 30H, CH
of Ph). 31P NMR (121.5 MHz, CD2Cl2, 298 K):  = −17.9 (AX2, t, 2J
(PA–PX) = 143.4, PA), 46.8 (AX2, d, 2J (PA–PX) = 143.4, PXPh2). 13
C
NMR (75.5 MHz, CD2Cl2, 298 K):  = 13.3 (s, CH3 of Bu), 22.4 (dd,
J (C–P) = 14.5, J (C–P) = 7.9, CH2), 23.2 (d, J (C–P) = 14.0, CH2),
25.6 (d, J (C–P) = 10.3, CH2), 29.4 (s, CH3 of t-Bu), 52.5 (s, C of
t-Bu), 64.1 (m, C2,6), 119.6 (m, C4H), 126.0–131.8 (m, CH of Ph),
135.2 (d, J (C–P) = 81.6, C of Ph), 135.9 (d, J (C–P) = 85.5, C of
Ph), 142.2 (s, C of Ph), 152.4 (s, C3,5). C of isocyanide was not seen.
IR (KBr):  2158.6 cm−1. C50H49P3S2NCu (884.32): calcd. C 67.89,
H 5.58; found C 67.51, H 5.23%.
2
2
1.45 (d, J (H–P) = 9.5, 3H, CH3), 5.52 (t, J (H–P) = 10.9, 1H,
H4), 6.70–7.9 (m, 30H, CH of Ph). 31P NMR (121.5 MHz, CDCl3,
298 K):  = −7.18 (t, AX2, 2J (PA–PX) = 83.7, PA), 41.30 (d, AX2, 2J
(PA–PX) = 83.7, PM).
Crystallography
Gold complex 16
Data were collected at 150.0(1) K on a Nonius Kappa CCD
diffractometer using a Mo K ( = 0.71070 Å) X-ray source and
a graphite monochromator. All data were measured using phi and
omega scans. Experimental details are described in Table 2. The
crystal structures were solved using SIR 9722 and Shelxl-97.23
ORTEP drawings were made using ORTEP III for Windows.24
CCDC reference numbers 238093 to 238098.
A solution of MeLi in Et2O (625 L, C = 0.16 M, 1.0 mmol) was
syringed into a solution of 1 (680 mg, 1.0 mmol) in THF (15 mL)
at −78 °C. The solution was warmed to room temperature and
stirred for 20 min. 31P NMR indicates then the complete formation
of anion 4, to which solid [AuCl·SMe2] (294 mg, 1.0 mmol) was
added at −78 °C. The initially red solution turned rapidly orange.
The mixture was stirred for an additional hour. The 31P NMR spec-
trum of the crude mixture shows the formation of a unique product.
The volatiles were then removed under vacuum. The product was
washed with hexanes and dried to yield an orange solid. Yield: 95%,
848 mg.
See http://www.rsc.org/suppdata/dt/b4/b406792d/ for crystallo-
graphic data in CIF or other electronic format.
Acknowledgements
This work was supported by the CNRS, the Ecole Polytechnique
and the DGA. M. D. thanks the DGA for financial support.
1H NMR (300 MHz, CDCl3, 298 K):  = 1.20 (d, 2J(H–P) = 8.6,
3H, CH3), 5.72 (t, J (H–P) = 4.4, 1H, H4), 6.70–7.9 (m, 30H,
2
CH of Ph). 31P NMR (121.5 MHz, CDCl3, 298 K):  = −12.6 (m,
AA′MM′XX′, PA,A′), 39.2 (d, AA′MM′XX′, ∑ J (PA–PM) = 72.9,
PMPh2), 45.2 (d,AA′MM′XX′, ∑ J (PA–PX) = 86.3, PXPh2). 13C NMR
(75.5 MHz, CDCl3, 298 K):  = 13.7 (s, CH3), 72.0 (m, C2 or 6), 90.5
(m, C6 or 2), 120.6 (m, C4H), 127.0–134.3 (m, CH of Ph), 135.1 (m, C
of Ph), 135.6 (m, C of Ph), 142.9 (bs, C of Ph), 145.0 (bs, C of Ph),
155.7 (bs, C3 or 5), 157.6 (s, C5 or 3). (C42H34P3S2Au)2 (892.75)2: calcd.
C 56.51, H 3.84; found C 56.80, H 4.02%.
References
1 C. J. Moulton and B. L. Shaw, J. Chem. Soc., Dalton Trans., 1976,
1020.
2 M. Albrecht and G. Van Koten, Angew. Chem., Int. Ed., 2001, 40, 3750;
J. T. Singleton, Tetrahedron Lett., 2003, 59, 1837; M. E. van der Boom
and D. Milstein, Chem. Rev., 2003, 103, 1759.
3 E. Diez-Barra, J. Guerra, I. Lopez-Solera, S. Merino, J. Rodriguez-
Lopez, P. Sanehez-Verdu and J. Tejeda, Organometallics, 2003, 22, 541
and references therein.
4 J. D. Niemoth-Anderson, K. A. Clark, T. A. George and C. R. Ross,
J. Am. Chem. Soc., 2000, 122, 3977.
Gold complex 17
5 M. Tschoerner, G. Trabesinger, A. Albinati and P. S. Pregosin,
Organometallics, 1997, 16, 3447; D. A. Evans, K. R. Campos,
J. S. Tedrow, F. E. Michael and M. R. Gagne, J. Am. Chem. Soc., 2000,
122, 7905; D. R. Evans, M. Huang, W. M. Seganish, E. W. Chege,
Y. F. Lam, J. C. Fettinger and T. L. Williams, Inorg. Chem., 2002,
41, 2633; X. Verdaguer, M. A. Pericàs, A. Riera, M. A. Maestro and
J. Mahía, Organometallics, 2003, 22, 1868.
MethodA: Dimeric complex 16 (200 mg, 0.11 mmol) was dissolved
in CH2Cl2 (5 mL) and PPh3 (57 mg, 0.22 mmol) was added solid. 31
P
NMR indicates then the complete formation of monomeric complex
17 within minutes. The volatiles were then removed under vacuum.
The product was washed with hexanes and dried to yield an orange
solid. Yield: 97%, 129 mg.
D a l t o n T r a n s . , 2 0 0 4 , 2 5 9 3 – 2 6 0 0
2 5 9 9

View Article Online
6 K. Ortner, L. Hilditch, Y. F. Zheng, J. R. Dilworth and U. Abram,
Inorg. Chem., 2000, 39, 2801; N. Brugat, A. Polo, A. Alvarez-Larena,
J. F. Piniella and J. Real, Inorg. Chem., 1999, 38, 4829.
15 M. M. Crutchfield, C. H. Dungan, J. H. Letcher, V. Mark and J. R.
van Wazer, P31 Nuclear Magnetic Resonance, Interscience Publishers,
New York, 1967.
7 D. R. Evans, M. S. Huang, W. M. Seganish, J. C. Fettinger and
T. L. Williams, Organometallics, 2002, 21, 893.
16 M. Doux, N. Mézailles, L. Ricard and P. Le Floch, Eur. J. Inorg. Chem.,
2003, 3878.
8 S. M. Aucott, A. M. Z. Slawin and J. D. Woolins, Eur. J. Inorg.
Chem., 2002, 2408; H. Brunner, H. J. Lautenschlager, W. A. König
and R. Krebber, Chem. Ber., 1990, 123, 847; H. Z. Liang, S. Ito and
M. Yoshifuji, Org. Lett., 2004, 6, 425.
9 M. J. Baker, M. F. Giles, A. G. Orpen, M. J. Taylor and R. J. Watt,
J. Chem. Soc., Chem. Commun., 1995, 197; L. Gonsalvi, H. Adams,
G. J. Sunley, E. Ditzel and A. Haynes, J. Am. Chem. Soc., 2002, 124,
13597.
17 I. Antes, S. Dapprich, G. Frenking and P. Schwerdterfeger, Inorg.
Chem., 1996, 35, 2089; A. J. Lupinetti, V. Jonas, W. Thiel, S. H. Strauss
and G. Frenking, Chem. Eur. J., 1999, 5, 2573.
18 P. C. J. Kamer, R. J. M. Nolte and W. Drenth, J. Am. Chem. Soc.,
1988, 110, 6818; M. Hanack and R. Thies, Chem. Ber., 1988, 121,
1225.
19 B. F. Straub and P. Hoffmann, Angew. Chem., Int. Ed., 2001, 40,
1288.
10 K. B. Dillon, F. Mathey and J. F. Nixon, Phosphorus: The Carbon Copy,
John Wiley & Sons, Chichester, 1998.
11 M. Doux, N. Mézailles, M. Melaimi, L. Ricard and P. Le Floch, Chem.
Commun., 2002, 1566.
12 M. Doux, N. Mézailles, L. Ricard and P. Le Floch, Organometallics,
2003, 22, 4624.
13 D. A. Evans, K. A. Woerpel, M. M. Hinman and M. M. Faul, J. Am.
Chem. Soc., 1991, 113, 726; A. V. Malkov, D. Pernazza, M. Bell,
M. Bella, A. Massa, F. Teply, P. Meghani and P. Kočovský, J. Org.
Chem., 2003, 68, 4727 and references therein.
14 A. S. K. Hashmi, T. M. Frost and J. W. Bats, J. Am. Chem. Soc., 2000,
122, 11553; R. Casado, M. Contel, M. Laguna, P. Romero and S. Sanz,
J. Am. Chem. Soc., 2003, 125, 11925; C. Wei and C.-J. Li, J. Am. Chem.
Soc., 2003, 125, 9584; A. S. K. Hashmi, L. Schwarz, J.-H. Choi and
T. M. Frost, Angew. Chem., Int. Ed. Engl., 2000, 39, 2285.
20 P. Braunstein, H. Lehner and D. Matt, Inorg. Synth., 1990, 27, 218.
21 R. Uson andA. Laguma, in Organometallic Syntheses, Elsevier Science,
ed. R. B. King and J. Eisch, Amsterdam, 1986, p. 324.
22 A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C. Giacovazzo,
A. Guagliardi,A. G. G. Moliterni, G. Polidori and R. Spagna, SIR97, an
integrated package of computer programs for the solution and refinement
of crystal structures using single crystal data, J. Appl. Crystallogr., 1999,
32, 115.
23 G. M. Sheldrick, SHELXL-97, Program for refinement of crystal
structures, University of Göttingen, Germany, 1997.
24 M. N. Burnett and C. K. Johnson, ORTEP-III: Oak Ridge Thermal
Ellipsoid Plot Program for Crystal Structure Illustrations, Report
ORNL-6895, Oak Ridge National Laboratory, Oak Ridge, TN, USA,
1996; L. J. Farrugia, ORTEP-3, Department of Chemistry, University of
Glasgow.
2 6 0 0
D a l t o n T r a n s . , 2 0 0 4 , 2 5 9 3 – 2 6 0 0