Synthesis, structural characterisation and reactions of some vinylgold(I) phosphane complexes

Article abstract of DOI:10.1002/ejic.200500861

A series of vinylgold(I) complexes [Au(CR=CHR)L] (R = H, Me; L = PPh ₃, PPh₂Me, PPhMe₂) were prepared from the reaction of the Grignard reagents [MgBr(CR=CHR)] (R = H, Me) with the gold(I) phosphane complexes [AuCl(L)] (L = PPh₃, PPh₂Me, PPhMe₂) at low temperature. The complexes were characterised by various spectroscopic techniques and, in the case of [Au(CMe=CHMe)(PPh ₃)], by a single-crystal X-ray structure determination. The gold-carbon bonds of these vinylgold(I) complexes are easily cleaved by acids and, in the presence of potassium permanganate, by species containing acidic protons. Wiley-VCH Verlag GmbH & Co. KGaA, 2006.

Full text of DOI:10.1002/ejic.200500861

FULL PAPER
DOI: 10.1002/ejic.200500861
Synthesis, Structural Characterisation and Reactions of Some Vinylgold(
I
)
Phosphane Complexes
Fabian Mohr,[a] Larry R. Falvello,[a] and Mariano Laguna*[a]
Keywords: Gold / Vinyl ligands / Phosphane ligands
A series of vinylgold(
Me; L = PPh3, PPh2Me, PPhMe2) were prepared from the re-
action of the Grignard reagents [MgBr(CR=CHR)] (R = H,
I
) complexes [Au(CR=CHR)L] (R = H,
ray structure determination. The gold–carbon bonds of these
vinylgold( ) complexes are easily cleaved by acids and, in the
presence of potassium permanganate, by species containing
acidic protons.
I
Me) with the gold(I) phosphane complexes [AuCl(L)] (L =
PPh3, PPh2Me, PPhMe2) at low temperature. The complexes
were characterised by various spectroscopic techniques and,
in the case of [Au(CMe=CHMe)(PPh3)], by a single-crystal X-
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2006)
of L = PMe3, the crystal structure of the complex was ob-
tained and shown to be the mixed-valence gold()/gold()
complex [Me3PAuIII(Me)2C(CF3)=C(CF3)AuIPMe3] (2),
which contains a bridging CF3C=CCF3 unit.[11]
Introduction
Organometallic gold() complexes of the type [AuR(L)],
where R is an alkyl, aryl or alkynyl group and L is a neutral
donor ligand such as a tertiary phosphane or isocyanide,
have been studied in depth over the years and numerous
derivatives have been synthesised, many of which have been
structurally characterised.[1–4] In marked contrast, analo-
gous vinylgold() complexes [Au(CR=CR2)(L)] have re-
ceived very little attention to date. In the 1970s a group of
Russian workers reported the synthesis and some reactions
of [Au(CH=CH2)(PPh3)] but no X-ray structure was re-
ported and the published experimental procedure seems du-
bious (see below).[5,6] Around the same time, other groups
were studying reactions of [AuMe(L)] complexes with the
fluorinated hydrocarbons CF3CϵCCF3 and F2C=CF2.
Stone and Mitchell first observed that UV irradiation of a
benzene solution of CF3CϵCCF3 and [AuMe(PPh3)] af-
forded a colourless product, which, based on 19F NMR
spectroscopic data and elemental analysis, they formulated
as the binuclear complex [Au2(µ-CF3C=CCF3)(PPh3)2]
(1).[7,8] The structure was later confirmed by an X-ray dif-
fraction study, which showed that the complex consists of
a CF3C=CCF3 molecule bridging two AuPPh3 units.[9] Sim-
ilar observations were made by Puddephatt and Johnson,
who studied the reactions of [AuMe(PPhMe2)] and [Au-
Me(PMe3)] with CF3CϵCCF3. The initial reaction prod-
ucts, formulated as [AuMe2(CF3CϵCCF3)L] (L = PPhMe2,
PMe3), undergo ethane elimination in acetone to give the
binuclear complexes [Au2(µ-CF3C=CCF3)L2] (1), which is
analogous to those observed earlier by Stone.[10] In the case
More recently, Brisdon and co-workers have prepared
and structurally characterised the vinylgold() complexes
[AuCR=CF2(PPh3)] (R = Cl, F) by treating [AuCl(PPh3)]
with the organolithium reagent CF2=CRLi.[12] Puddephatt
and Treurnicht reported the synthesis of the very thermally
unstable vinylgold() complex [Au(CH=CH2)(CNMe)],
which could only be characterised by 1H NMR spec-
troscopy at –20 °C.[13] Other approaches for adding vinyl
groups to an AuPPh3 unit have also been investigated. The
vinyl(methyl)titanium complex [Cp*2TiMe(CH=CH2)] re-
acts with [AuCl(PPh3)] to give [Cp*2TiCl(µ-C=CH2)-
AuPPh3] (3), the structure of which contains a vinylidene
unit bridging the titanium and gold atoms in an unsym-
metrical
fashion.[14]
The
vinylgold()
complexes
[Au{C(OR)=CH2}(PPh3)M(CO)5] (4) (M = Cr, Mo, W; R
= Me, Et), in which the vinyl group is π-coordinated (η2)
to the M(CO)5 fragment, are formed by treating the depro-
tonated carbene species Li[(CO)5M=C(OR)CH2] with
[AuCl(PPh3)].[15] Addition of PPh3 displaces the coordi-
nated M(CO)5 unit from 4 and produces the vinylgold()
complex [Au{C(OR)=CH2}(PPh3)].
[a] Departamento de Química Inorgánica, Instituto de Ciencia de
Materiales de Aragón, Universidad de Zaragoza–C.S.I.C.,
50009 Zaragoza, Spain
Eur. J. Inorg. Chem. 2006, 833–838
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
833

F. Mohr, L. R. Falvello, M. Laguna
FULL PAPER
1
derivatives are stable in solution for less than 12 h. The H
NMR spectra of the vinyl complexes 5a–5c consists of three
doublets of doublets displaying trans,cis and geminal coup-
ling constants of 20, 14 and 5 Hz, respectively. The 1H
NMR spectra of the dimethyl analogues 6a–6c display a
broad singlet at δ Ϸ 6.4 ppm due to the vinylic proton.
Furthermore, the two methyl groups attached to the vinyl
group resonate as a broad singlet (coupling unresolved) and
doublet of quartets for the α- and β-Me groups, respec-
tively; the coupling constants confirm that the methyl
groups are trans to each other. In the 13C NMR spectra of
all six complexes we observed, in addition to the phenyl
and Me signals, a singlet at δ Ϸ 129 ppm due to the β-C of
the vinyl group. The resonance of the α-C of the vinyl group
attached to the gold atom could not be observed, presum-
ably due to the large quadrupole moment of the proximal
gold atom. As expected, the 31P{1H} NMR spectra of all
six complexes show singlet resonances whose chemical shift
varies depending on the phosphane ligands. The FAB+
mass spectra of most of the complexes show weak molecu-
lar-ion peaks as well as peaks due to loss of the vinyl group
and formation of higher mass ions [Au(phosphane)2]+; the
latter are frequently observed in FAB mass spectra of
gold() phosphane complexes.[17] The vinylgold() structure
was confirmed by a partial X-ray diffraction study of 6a.
Due to the limited stability of the complex in solution, the
crystals obtained were of poor quality and, more import-
antly, signs of twinning were apparent. However, the data
were of sufficient quality to establish the connectivity and
to derive geometric parameters of middling accuracy but
wholly consistent with all other experimental data reported
here and elsewhere for this and analogous compounds. A
displacement ellipsoid plot of one molecule is shown in Fig-
ure 1.[18]
Our group has been studying the catalytic activity of
various gold complexes in the addition of water or MeOH
to terminal alkynes.[16] A vinylgold species has been sug-
gested as a possible intermediate in this catalytic process.
For this reason we were interested in preparing some vi-
nylgold() complexes and studying their structures and reac-
tivity. The results of this investigation are presented here.
Results and Discussion
Initially, we attempted to reproduce the Russian work to
prepare [Au(CH=CH2)(PPh3)] (5a). What seemed suspi-
cious to us was the claim that the reaction between the vinyl
Grignard reagent and [AuCl(PPh3)] was carried out at
+50 °C and that the complex was obtained in 90% yield![5]
In our hands, these conditions led to complete decomposi-
tion of the reaction mixture to metallic gold. However,
when vinylmagnesium bromide was added to an ethereal
suspension of [AuCl(PPh3)] at –78 °C the complex
[Au(CH=CH2)(PPh3)] (5a) could be isolated in 73% yield
after work-up. We found that when a large excess of Grig-
nard reagent was used in the reaction and the subsequent
hydrolysis was carried out with large volumes of water or
aqueous NH4Cl, significant decomposition to metallic gold
occurred. This decomposition can be avoided by using only
a small excess of Grignard reagent (ca. 10%), and simply
exposing the reaction mixture to air was sufficient to hydro-
lyse the excess Grignard reagent. Following this procedure
the vinylgold() complexes [Au(CR=CHR)(L)] (R = H, 5a:
L = PPh3, 5b: L = PPh2Me, 5c: L = PPhMe2; R = Me, 6a:
L = PPh3, 6b: L = PPh2Me, 6c: L = PPhMe2 ) were pre-
pared in good to moderate yields (Scheme 1).
Scheme 1.
All the complexes prepared are colourless, crystalline sol-
ids (except for 5c and 6c, which were isolated as pale-brown
oils) that are air- and thermally stable in the solid state; in
solution, however, decomposition occurs slowly (2–3 days)
Figure 1. ORTEP view of one of the independent molecules of 6a.
Ellipsoids show 30% probability levels; phenyl hydrogen atoms
for the PPh3 and PPh2Me complexes, whilst the PPhMe2 have been omitted for clarity.
834
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 833–838

Vinylgold() Phosphane Complexes
FULL PAPER
The structure was refined in the space group C2/c with of permanganate we treated CDCl3 solutions of complex 6a
two independent molecules in the asymmetric unit. Some with an equivalent amount of acetone or phenylacetylene
bond lengths and angles differ slightly between the two and examined the 1H and 31P NMR spectra over time. Even
molecules; but we do not wish to attribute chemical signifi- after a period of 48 h there was no evidence that any reac-
cance to this observation given that the differences are con- tion had occurred − only signals from the starting complex
sistent with the inaccuracies evident in the diffraction data were observed in the NMR spectra − thus confirming that
(vide infra) [wR2 = 0.1482, R1 = 0.0822, C–C bond pre- permanganate is indeed required for this reaction. If a solu-
cision 0.022 Å]. The structure of 6a consists of a dimethylvi- tion of KMnO4 in CD3CN is then added to the NNR tubes
nyl moiety (with the Me groups in a trans configuration) containing complex 6a and acetone or phenylacetylene a
and a PPh3 unit linearly [177.9(3)° and 175.4(4)°] coordi- brown precipitate is observed after a few minutes and after
nated to the gold() centre. The Au–P [2.3092(19) and about 30 min the main species in solution, determined by
2.291(2) Å] and Au–C [2.004(7) and 2.053(10) Å] distances NMR spectroscopy, are [Au{CH2C(O)CH3}(PPh3)] and
are similar to those in [Au(CF=CF2)(PPh3)] [Au–P = [Au(CϵCPh)(PPh3)], respectively. These results suggest that
2.272(3)
and
Au–C
=
2.028(9) Å]
and the permanganate reacts initially with the vinylgold() com-
[Au(CCl=CF2)(PPh3)] [Au–P = 2.266(2) and Au–C = plex, probably oxidising the carbon–carbon double bond,
2.026(10) Å].[12] The vinyl C=C distances [1.322(18) and via an intermediate gold()/manganese() cyclic ester 7
1.327(17) Å] are slightly longer than those in (Scheme 2). It is known that the oxidation of alkenes with
[Au(CF=CF2)(PPh3)] [C=C = 1.297(14) Å][12] but shorter permanganate proceeds via such cyclic MnV ester interme-
than that in [AuC(OMe)=CH2(PPh3)Cr(CO)5] [C=C = diates, which undergo decomposition reactions to give or-
1.345(11) Å].[15]
ganic products and MnO2.[21] Recently, this [3+2] mecha-
nism via cyclic MnV esters has been confirmed by DFT
calculations and kinetic isotope effect data.[22] In our case,
the evidence presented above suggests that the presumed
intermediate 7 causes deprotonation of acetone and phenyl-
Reactions of Vinylgold(I) Complexes
The Russian workers studied the reactions of acetylene, thus producing the acetonyl- and alkynylgold()
[Au(CH=CH2)(PPh3)] (5a) with HCl, HgCl2 Br2, (SCN)2, complexes as well as the diol, accompanied, perhaps even
benzoic acid and KMnO4.[6] In the first five cases the gold– driven, by precipitation of insoluble MnO2. We tried to
1
carbon bond is cleaved to give [AuX(PPh3)] (X = Cl, Br, identify the organic oxidation products by both H NMR
SCN, O2CPh) as the gold-containing product. However, in and GC/MS spectroscopy, but no conclusive results were
the case of KMnO4, neither metal–carbon bond cleavage obtained. In an attempt to observe the intermediate species
nor oxidation of the carbon–carbon double bond occurs; 7 we treated a CDCl3 solution of 6a with a CD3CN solution
instead,
the
acetonylgold()
phosphane
complex of KMnO4 and observed immediate formation of a brown
[Au{CH2C(O)CH3}(PPh3)] is formed in high yield.[6,19] The precipitate. The 31P NMR spectrum of the supernatant
X-ray crystal structure of this complex was reported later, solution shows a singlet resonance at δ = 44.50 ppm,
but neither the method of preparation nor source of the slightly shifted relative to that of the starting complex (δ =
sample was mentioned.[20] In order to examine the general- 44.77 ppm), which may be due to the intermediate. How-
1
ity of these reactions we studied the reaction of hydrochlo- ever, the presence of a multiplet at δ = 6.4 ppm in the H
ric acid and KMnO4 with the new vinylgold() complex 6a. NMR spectrum, due to the vinylic proton, suggests that the
Treatment of an acetone solution of 6a with a solution of C=C double bond in the vinyl ligand is still intact; in fact,
HCl or DCl in D2O (the deuterated acid was more conve- the overall spectrum is almost identical to that of 6a. When
nient to study by 1H NMR spectroscopy) results in electro- excess PhCϵCH is added to this solution the 31P NMR
philic cleavage of the gold–carbon bond and formation of spectrum shows, in addition to the original peak at δ =
[AuCl(PPh3)] and trans-MeDC=CMeH, as was evident 44.50 ppm, a new signal at δ = 42.15 ppm. The latter
1
from the H and 31P NMR spectra; the chemical shift ob- resonance is consistent with the presence of
served in the 31P NMR spectrum is identical to that of an [Au(CϵCPh)(PPh3)], which was confirmed by comparison
authentic sample of [AuCl(PPh3)]. Treatment of an acetone of chemical shifts of an authentic sample. Over a period
solution of complex 6a at 0 °C with KMnO4 resulted in of about 10 h the intensity of the signal due to
formation of a brown precipitate (MnO2). After about [Au(CϵCPh)(PPh3)] increases, whilst that of the starting
30 min reaction time a pale-yellow solid could be isolated complex 6a decreases. After about 24 h the major species in
from the mixture after filtration and removal of the solvent. solution is [Au(CϵCPh)(PPh3)]. If the same experiment is
1
On the basis of H and 31P{1H} NMR spectroscopic data, carried out with acetone similar observations are made, na-
this product was identified as the acetonylgold() complex mely the intensity of the 31P NMR signal due to the ace-
[Au{CH2C(O)CH3}(PPh3)], identical to that reported by tonylgold() complex [Au{CH2C(O)CH3}(PPh3)] increases
Nesmeyanov.[19] We were interested in investigating this re- over time, while that due to 6a decreases. These results show
action in more detail in order to gain some understanding that KMnO4 is required for the vinylgold() complex to re-
of the mechanism and the exact role of the permanganate. act with species containing acidic protons and, although we
To test whether species containing acidic protons, such as failed to detect any intermediate, the simple fact that ad-
acetone or phenylacetylene, can react with 6a in the absence dition of KMnO4 to a solution of 6a causes precipitation of
Eur. J. Inorg. Chem. 2006, 833–838
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
835

F. Mohr, L. R. Falvello, M. Laguna
FULL PAPER
NMR (100 MHz, CDCl3, 25 °C): δ = 134.3 (d, 3JP,C = 13 Hz, meta-
Ph), 131.4 (s, AuC=C), 131.2 (s, para-Ph), 129.0 (d, JP,C = 10 Hz,
MnO2 suggests that some reaction yielding an undetectable
intermediate must be taking place. More detailed investi-
gations are necessary to fully understand this process. Other
oxidants including dichromate and H2O2 either do not react
at all or, in the case of peroxide, oxidation of the phosphane
occurs to give Ph3P=O and decomposition of the starting
complex to gold metal.
2
ortho-Ph) ppm. 31P{1H} NMR (162 MHz, CDCl3, 25 °C): δ =
44.40 ppm ppm. FAB MS: m/z (%) = = 486 (5) [M]+, 459 (95)
[Au(PPh3)]+, 721 (100) [Au(PPh3)2]+. C20H18AuP (486.3): calcd. C
49.40, H 3.73; found C 49.30, H 3.50.
Compounds 5b and 5c were prepared in a similar manner.
[Au(CH=CH2)(PMePh2)] (5b): Colourless solid (0.252 g, 64%
yield). 1H NMR (400 MHz, CDCl3, 25 °C): δ = 7.60–7.70 (m, 4 H,
Ph), 7.41–7.48 (m, 6 H, Ph), 7.14 (dd, Jtrans-H,H = 21, Jcis-H,H
=
14 Hz, 1 H, AuCH), 6.00 (dd, Jcis-H,H = 14, Jgem-H,H = 5 Hz, 1 H,
=CH2), 5.48 (dd, Jtrans-H,H = 21, Jgem-H,H = 5 Hz, 1 H, =CH2), 2.01
2
(d, JP,H = 8 Hz, 3 H, PCH3) ppm. 13C NMR (100 MHz, CDCl3,
3
1
25 °C): δ = 132.8 (d, JP,C = 13 Hz, meta-Ph), 132.7 (d, JP,C
=
50 Hz, ipso-Ph), 131.5 (s, AuC=C), 131.0 (s, para-Ph), 129.0 (d,
1
2JP,C = 10 Hz, ortho-Ph), 13.9 (d, JP,C = 32 Hz, PCH3) ppm.
31P{1H} NMR (162 MHz, CDCl3, 25 °C): δ = 30.10 ppm ppm.
FAB MS: m/z (%) = = 379 (60) [Au(PMePh2)]+, 597 (100) [Au(P-
MePh2)2]+ 821 (35) [M + Au(PPh2Me)]+. C15H16AuP (424.2): calcd.
C 42.47, H 3.80; found C 42.84, H 3.71.
[Au(CH=CH2)(PPhMe2)] (5c): Pale-brown oil (0.108 g, 55% yield).
1H NMR (400 MHz, CDCl3, 25 °C): δ = 7.70–7.78 (m, 2 H, Ph),
7.45–7.50 (m, 3 H, Ph), 7.10 (dd, Jtrans-H,H = 21, Jcis-H,H = 14 Hz,
1 H, AuCH), 5.98 (dd, Jcis-H,H = 12, Jgem-H,H = 5 Hz, 1 H, =CH2),
5.46 (dd, Jtrans-H,H = 21, Jgem-H,H = 5 Hz, 1 H, =CH2), 1.73 (br. s,
6 H, PCH3) ppm. 13C NMR (100 MHz, CDCl3, 25 °C): δ = 133.8
Scheme 2.
In summary, we have shown that vinylgold() complexes
can be prepared in good yields from the reaction of a vinyl
Grignard reagent with [AuCl(PPh3)] at low temperatures.
The gold–vinyl–C bond is easily cleaved by acids or, in the
presence of KMnO4, by species containing an acidic proton
such as acetone and phenylacetylene. Further work is cur-
rently in progress to further elucidate the mechanism of this
reaction and to study other reactions of vinylgold() com-
pounds.
1
3
(d, JP,C = 49 Hz, ipso-Ph), 131.77 (d, JP,C = 13 Hz, meta-Ph),
131.5 (s, AuC=C), 131.1 (s, para-Ph), 128.9 (d, 2JP,C = 10 Hz, ortho-
1
Ph), 15.34 (d, JP,C
=
31 Hz, PCH3) ppm. 31P{1H} NMR
(162 MHz, CDCl3, 25 °C): δ = 18.65 ppm. FAB MS: m/z (%) = 335
(62) [Au(PPhMe2)]+, 473 (100) [Au(PPhMe2)2]+.
The following complexes were prepared as described above using
1-methyl-1-propenylmagnesium bromide.
[Au(CMe=CHMe)(PPh3)] (6a): Colourless, crystalline solid
(0.206 g, 63% yield). 1H NMR (400 MHz, CDCl3, 25 °C): δ = 7.51–
7.60 (m, 6 H, Ph), 7.40–7.51 (m, 9 H, Ph), 6.45 (m, 1 H, =CHMe),
3
5
Experimental Section
2.06 (br. s, 3 H, AuCMe), 1.95 (dq, JH,H = 6, JH,H = 1 Hz, 3 H,
=CHMe) ppm. 13C NMR (100 MHz, CDCl3, 25 °C): δ = 134.3 (d,
3JP,C = 13 Hz, meta-Ph), 131.4 (d, 1JP,C = 48 Hz, ipso-Ph), 130.8 (s,
General: All reactions were performed under an atmosphere of dry
argon using Schlenk techniques. Solvents were dried by standard
2
para-Ph), 129.2 (AuC=C), 128.9 (d, JP,C = 10 Hz, ortho-Ph), 30.4
methods and freshly distilled under argon before use. H, 13C and
1
(s, AuCMe), 20.9 (s, CHMe) ppm. 31P{1H} NMR (162 MHz,
CDCl3, 25 °C): δ = 44.77 ppm. FAB MS: m/z (%) = 514 (15) [M]+,
459 (100) [Au(PPh3)]+, 721 (65) [Au(PPh3)2]+. C22H22AuP (514.4):
calcd. C 51.37, H 4.31; found C 51.48, H 4.12. X-ray quality crys-
tals were selected from the bulk sample.
31P{1H} NMR spectra were recorded with a 400 MHz Bruker Av-
ance spectrometer. Chemical shifts are quoted relative to external
TMS (1H and 13C) or 85% H3PO4 (31P). FAB mass spectra were
measured on a VG Autospec spectrometer in positive ion mode
using NBA as matrix. Elemental analyses were obtained in-house
using a Perkin–Elmer 240B microanalyser. The chlorogold() pre-
cursor complexes were prepared by treating equivalent amounts of
[AuCl(tht)] with the tertiary phosphanes in CH2Cl2. The vinyl
Grignard reagents were obtained commercially (Aldrich).
[Au(CMe=CHMe)(PPh2Me)] (6b): Colourless solid (0.118 g, 73%
yield). 1H NMR (400 MHz, CDCl3, 25 °C): δ = 7.61–7.71 (m, 4 H,
Ph), 7.39–7.48 (m, 6 H, Ph), 6.47 (m, 1 H, =CHMe), 2.06 (br. s, 3
2
3
H, AuCMe), 2.01 (d, JP,H = 6 Hz, 3 H, PMe) 1.96 (dq, JH,H = 6,
5JH,H = 1 Hz, 3 H, =CHMe) ppm. 13C NMR (100 MHz, CDCl3,
25 °C): δ = 133.2 (d, JP,C = 47 Hz, ipso-Ph), 132.7 (d, JP,C
[Au(CH=CH2)(PPh3)] (5a): A suspension of [AuCl(PPh3)] (0.402 g,
0.929 mmol) in dry Et2O (10 mL) was cooled to –78 °C and treated
with a solution of vinylmagnesium bromide (1.0 mL 1  solution
in THF, 1.0 mmol). The mixture was warmed to room temperature
over a period of 4 h. After this time the reaction vessel was exposed
to air for about 10 min. to hydrolyze any unreacted Grignard rea-
gent. The mixture was then filtered through anhydrous MgSO4 and
the filtrate evaporated to dryness in vacuo to afford 0.288 g (73%)
of colourless product. 1H NMR (400 MHz, CDCl3, 25 °C): δ =
1
3
=
13 Hz, meta-Ph), 130.8 (s, para-Ph), 129.3 (s, AuC=C), 128.9 (d,
2JP,C = 10 Hz, ortho-Ph), 30.3 (s, AuCMe), 20.9 (s, CHMe), 13.7
1
(d, JP,C = 39 Hz, PMe) ppm. 31P{1H} NMR (162 MHz, CDCl3,
25 °C): δ = 30.96 ppm. FAB MS: m/z (%) = 452 (10) [M]+, 397
(100) [Au(PPh2Me)]+, 597 (55) [Au(PPh2Me)2]+. C17H20AuP
(452.3): calcd. C 45.14, H 4.46; found C 45.28, H 4.32.
[Au(CMe=CHMe)(PPhMe2)] (6c): Pale-brown oil (0.170 g, 40%
7.42–7.61 (m, 15 H, Ph), 7.20 (dd, Jtrans-H,H = 20, Jcis-H,H = 14 Hz, yield). 1H NMR (400 MHz, CDCl3, 25 °C): δ = 7.68–7.80 (m, 2 H,
1 H, AuCH), 6.04 (dd, Jtrans-H,H = 20, Jgem-H,H = 5 Hz, 1 H, =CH2),
Ph), 7.41–7.50 (m, 3 H, Ph), 6.45 (m, 1 H, =CHMe), 2.01 (br. s, 3
H, AuCMe), 1.92 (dq, JH,H = 6, JH,H = 1 Hz, 3 H, =CHMe),
3
5
5.53 (dd, Jcis-H,H = 14, Jgem-H,H = 5 Hz, 1 H, =CH2) ppm. 13C
836
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 833–838

Vinylgold() Phosphane Complexes
FULL PAPER
2
1.70 (d, JP,H = 6 Hz, 6 H, PMe) ppm. 13C NMR (100 MHz,
residuals, some 20 difference peaks above 1 eÅ–3, and unusual
3
2
CDCl3, 25 °C): δ = 131.8 (d, JP,C = 13 Hz, meta-Ph), 131.1 (s,
analysis of variance with the worst-fit reflections largely having Fo
para-Ph), 129.4 (s, AuC=C), 129.0 (d, 2JP,C = 11 Hz, ortho-Ph), 30.4
greater than Fc . Several twin models were tested, each of which
2
1
(s, AuCMe), 20.9 (s, CHMe), 15.3 (d, JP,C = 32 Hz, PMe) ppm. produced some improvement in the crystallographic indicators but
31P{1H} NMR (162 MHz, CDCl3, 25 °C): δ = 19.96 ppm. FAB
MS: m/z (%) = 390 (11) [M]+, 335 (100) [Au(PPhMe2)]+, 473 (90)
[Au(PPhMe2)2]+.
did not influence the chemical implications of the results. Concur-
rent twin models were also tested, with similar results. The results
reported here are those obtained with no twin model; more infor-
mation on other models can be obtained from the authors. Com-
monly employed structure validation tools (PLATON[25]) did not
reveal the presence of missed crystallographic symmetry elements.
Visual examination of the structural results suggests pseudo-trans-
lational symmetry, but at the same time reveals that the two inde-
pendent molecules have very different environments and participate
in quite distinct fashions in the crystal packing. One of the phenyl
groups (C50–C55) shows prolate displacement ellipsoids and a
short average C–C distance. While these are normally taken to be
definitive symptoms of libration, we cannot rule out an association
between these effects and the presence of a putative minor twin
component.
Reaction of 6a with Acid: A few drops of an approx. 1  solution
of DCl in D2O was added to a sample of 6a dissolved in [D6]
1
acetone in an NMR tube. H and 31P NMR spectra were recorded
1
after about 10 min. H NMR (400 MHz, [D6]acetone, 25 °C): δ =
7.53–7.66 (m, 15 H, Ph from Ph3PAuCl), 5.35 (br. s, 1 H, CHMe),
3
1.54 (q, JH,D = 1 Hz, 3 H, CDMe), 1.53 (s, 3 H, CHMe) ppm.
31P{1H} NMR (162 MHz, [D6]acetone, 25 °C): δ = 33.97 ppm. This
31P{1H} NMR chemical shift is identical to that of an authentic
sample of [AuCl(PPh3)].
Reaction of 6a with KMnO4: A sample of 6a was dissolved in ace-
tone and treated with an acetone solution of KMnO4 at about 0 °C.
After around 10 min a brown precipitate (MnO2) was observed.
The mixture was filtered and the filtrate was evaporated to dryness.
1H and 31P NMR spectra of the pale-yellow solid were recorded.
1H NMR (400 MHz, CDCl3, 25 °C): δ = 7.40–7.59 (m, 15 H, Ph),
[1] R. J. Puddephatt, The Chemistry of Gold, Elsevier Scientific
Publishing Co., New York, 1978.
[2] R. Usón, A. Laguna, Coord. Chem. Rev. 1986, 70, 1–50.
[3] H. Schmidbaur, A. Grohmann, M. E. Olmos, in Gold Progress
in Chemistry, Biochemistry and Technology (Ed.: H. Schmid-
baur), John Wiley & Sons, Chichester, 1999.
[4] E. J. Fernández, A. Laguna, M. E. Olmos, Adv. Organomet.
Chem. 2005, 52, 77–141.
[5] A. N. Nesmeyanov, E. G. Perevalova, V. V. Krivykh, A. N. Ko-
sina, K. I. Grandberg, E. I. Smyslova, Izv. Akad. Nauk. SSSR,
Ser. Khim. 1972, 653–654.
[6] K. I. Grandberg, E. I. Smyslova, A. N. Kosina, Izv. Akad.
Nauk. SSSR, Ser. Khim. 1973, 2787–2789.
[7] C. M. Mitchell, F. G. A. Stone, J. Chem. Soc., Chem. Commun.
1970, 1263–1264.
[8] C. M. Mitchell, F. G. A. Stone, J. Chem. Soc., Dalton Trans.
1972, 102–107.
[9] J. C. Gilmore, P. Woodward, J. Chem. Soc., Chem. Commun.
1971, 1233–1234.
3
2.71 (d, JH,P = 1 1 Hz, 2 H, AuCH2), 2.22 (s, 3 H, CMe) ppm.
31P{1H} NMR (162 MHz, CDCl3, 25 °C): δ = 42.20 ppm. In ad-
dition, there are trace amounts of starting complex 6a and a further
unidentified species present (δP = 44.7 and 29.2 ppm, respectively).
NMR Reactions of 6a with KMnO4 and Acetone or Phenylacetylene:
A solution of KMnO4 in CD3CN was added to a CDCl3 solution
of 6a. A brown solid (MnO2) precipitated immediately. Acetone or
phenylacetylene (4 µL) was then added to the NMR tube and 31P
1
and H NMR spectra were recorded before and after the addition.
Reaction of 6a with H2O2: A sample of 6a dissolved in acetone was
treated with 30% H2O2. After stirring for about 30 min a gold mir-
ror had formed on the inside of the flask. The solvents were re-
moved in vacuo and the residue was analysed by 31P spectroscopy.
31P{1H} NMR (162 MHz, CDCl3, 25 °C): δ = 29.69 ppm.
X-ray Crystallography of Complex 6a: Data were gathered twice,
from two different crystals and using two different CCD-based dif-
fractometers, with similar results in both cases. The results reported
here were derived from data gathered using
diffractometer (Xcalibur2 with a Sapphire3 detector and sealed
tube Mo Enhance source).[23]
crystal of dimensions
0.17×0.16×0.05 mm, and with a slight yellow tint as viewed under
an optical microscope, was mounted with epoxy at the end of a
quartz fibre and kept at 123 K for the measurements. A preliminary
unit-cell determination yielded the same results that had been de-
termined earlier using a Bruker Apex diffractometer, namely a mo-
noclinic cell with one unusually long axis (ca. 45.8 Å). The cell
dimensions were verified on the four-circle diffractometer using
normal-beam axial photography. Because of the long axis, for the
present data the crystal-to-detector distance was set to 90 mm and
the data collection strategy was optimised accordingly. The faces of
the crystal were indexed and absorption corrections were calculated
numerically and applied to the integrated data. The structure was
[10] A. Johnson, R. J. Puddephatt, J. M. Quirk, J. Chem. Soc.,
Chem. Commun. 1972, 938–939.
[11] J. A. J. Jarvis, A. Johnson, R. J. Puddephatt, J. Chem. Soc.,
Chem. Commun. 1973, 373–374.
[12] N. A. Barnes, A. K. Brisdon, W. I. Cross, J. G. Fay, J. A.
Greenall, R. G. Pritchard, J. Sherrington, J. Organomet. Chem.
2000, 616, 96–105.
a four-circle
A
[13] R. J. Puddephatt, I. Treurnicht, J. Organomet. Chem. 1987, 319,
129–137.
[14] R. Backhaus, J. Oster, R. Wang, U. Böhme, Organometallics
1998, 17, 2215–2221.
[15] H. G. Raubenheimer, M. W. Esterhuysen, A. Timoshkin, Y.
Chen, G. Frenking, Organometallics 2002, 21, 3173–3181.
[16] R. Casado, M. Contel, M. Laguna, P. Romero, S. Sanz, J. Am.
Chem. Soc. 2003, 125, 11925–11935.
[17] M. I. Bruce, M. J. Liddell, J. Organomet. Chem. 1992, 427, 263–
274.
[18] C22H22AuP, monoclinic, C2/c, a = 15.4392(6), b = 10.8911(5),
c = 45.7928(18) Å, β = 98.404(4)°, V = 7617.4(5) Å3, Z = 16,
ρ = 1.794 gcm–3, µ = 7.809 mm–1, 31300 data measured, 8639
unique, 98.6% completeness, Rint = 0.0440, Rσ = 0.0653, wR2
= 0.1482 (437 parameters), R1 = 0.0822 [6532 data with I Ͼ
2σ(I)], quality-of-fit = 1.232. CCDC-286877 contains the sup-
plementary crystallographic data for this paper. These data can
be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
2
solved by direct methods and refined against Fo using full-matrix
least-squares.[24] Non-hydrogen atoms were refined anisotropically.
Non-methyl hydrogen atoms were placed at calculated positions
and refined as riders. Methyl hydrogen atoms were located success-
fully in local slant-Fourier calculations and refined as riding atoms
but with a variable torsion angle about the local C–C bond. The
refinement was stable and convergent, but the results showed effects
that are normally considered characteristic of twinning, i.e. high
[19]
A. N. Nesmeyanov, K. I. Grandberg, E. I. Smyslova, E. G. Per-
evalova, Izv. Akad. Nauk. SSSR, Ser. Khim. 1972, 2375.
Eur. J. Inorg. Chem. 2006, 833–838
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
837

F. Mohr, L. R. Falvello, M. Laguna
FULL PAPER
[20] L. G. Kuzmina, Koord. Khim. Koordinatsionnaya Khimiya 1994,
20, 540–546.
[24] G. M. Sheldrick, SHELXTL-NT 6.1, University of Göttingen,
1998.
[21] D. G. Lee, J. R. Brownridge, J. Am. Chem. Soc. 1974, 96, 5517–
[25] A. L. Spek, Acta Crystallogr., Sect. A 1990, 46, C34.
5523.
Received: September 20, 2005
[22] K. N. Houk, T. Strassner, J. Org. Chem. 1999, 64, 800–802.
[23] CrysAlis CCD and CrysAlis RED: V 1.171.27p8, Oxford Dif-
fraction, Ltd. 1995–2005.
Published Online: December 20, 2005
838
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 833–838