Gold acyls and imidoyls prepared from anionic Fischer-type carbene complexes

Article abstract of DOI:10.1016/j.ica.2005.03.029

Reaction of [(CO)₅WC(O)Ph]Li or [(CO)₅WC(O)Ph] NBu₄ with Ph₃PAuCl affords acyl complexes of gold. In the latter conversion, both the crystalline products [(CO)₅WCl]NBu ₄ (2) and Ph₃PAuC(O)Ph (3) have been isolated and fully characterised. Similarly, imidoyl gold compounds (4-8) result from deprotonated aminocarbene complexes, [(CO)₅MC(NR²)R¹]Li (M = Cr, W; R¹ = Ph, Me; R² = H, Me) and Ph₃PAuCl. Crystal and molecular structure determinations of dinuclear [Ph ₃PAuC(NH)Ph]·Cr(CO)₅ (6) show N-coordination of the chromium carbonyl unit that selectively affords a Z-isomer.

Full text of DOI:10.1016/j.ica.2005.03.029

Inorganica Chimica Acta 358 (2005) 4217–4228
www.elsevier.com/locate/ica
Gold acyls and imidoyls prepared from anionic Fischer-type
carbene complexes
1
*
Helgard G. Raubenheimer , Matthias W. Esterhuysen, Catharine Esterhuysen
Department of Chemistry and Polymer Science, University of Stellenbosch, Private Bag X1, Matieland 7602, Stellenbosch, South Africa
Received 26 January 2005; revised 2 March 2005; accepted 14 March 2005
Available online 12 April 2005
Dedicated to Professor Hubert Schmidbaur on the occasion of his 70th birthday.
Abstract
Reaction of [(CO)5WC(O)Ph]Li or [(CO)5WC(O)Ph]NBu4 with Ph3PAuCl affords acyl complexes of gold. In the latter conver-
sion, both the crystalline products [(CO)5WCl]NBu4 (2) and Ph3PAuC(O)Ph (3) have been isolated and fully characterised. Simi-
larly, imidoyl gold compounds (4–8) result from deprotonated aminocarbene complexes, [(CO)5MC(NR2)R1]Li (M = Cr, W;
R1 = Ph, Me; R2 = H, Me) and Ph3PAuCl. Crystal and molecular structure determinations of dinuclear [Ph3PAuC(NH)-
Ph] Æ Cr(CO)5 (6) show N-coordination of the chromium carbonyl unit that selectively affords a Z-isomer.
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: Carbene complexes; Acyl and imidoyl gold complexes; Organometallic synthons; Chloro(pentacarbonyl)tungstate; Gold–gold
interactions
1. Introduction
[4]. Furthermore, anionic transition metal–acyl com-
plexes are also important intermediates in many syn-
thetic routes to useful organometallic compounds,
including Fischerꢀs classical, and Semmelhackꢀs more re-
cent, routes to carbene complexes. Both these routes in-
volve the formation of anionic metal–acyl species that,
upon oxygen alkylation, yield a wide variety of carbene
complexes [5,6]. Due to the electronic properties of low-
valent metal centres in neutral metal–acyl complexes,
the acyl moiety is viewed as a close relative of ketones
or esters. This analogy has been utilised successfully in
developing much of the chemistry of metal–acyl com-
plexes, especially reduction and alkylation reactions [7].
Transition metal–acyl complexes are mostly prepared
via three main reaction pathways: the attack by nucleo-
philic carbon reagents at metal-bound CO; acylation at
a nucleophilic metal centre; and the migratory insertion
reaction of a CO ligand into an M–C bond [1,8]. From a
catalysis point of view the latter, also frequently referred
to as catalytic carbonylation, is the more interesting
Transition metal–acyl complexes comprise an extre-
mely important class of organometallic compounds. It
has long been known that metal–acyl complexes are a
ubiquitous feature in many homogenous catalytic pro-
cesses, especially ones in which carbon monoxide is in-
volved [1], and have hence been the subject of many
reactivity and theoretical studies [2]. Transition metal–
acyl complexes are also useful models for key catalytic
intermediates in heterogeneous catalytic processes [3],
such as the Fischer–Tropsch process operated by SA-
SOL, where Fe-surface alkyl and acyl species are be-
lieved to play an important role in the formation of
oxygenates (mainly formates, alcohols and ketones)
*
Corresponding author. Tel.: +27 21 808 3850; fax: +27 21 808 3849.
E-mail addresses: hgr@sun.ac.za (H.G. Raubenheimer), ce@sun.
ac.za (C. Esterhuysen).
1
Tel.: +27118083345; fax: +27118083360.
0020-1693/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2005.03.029

4218
H.G. Raubenheimer et al. / Inorganica Chimica Acta 358 (2005) 4217–4228
route. This route provides a mechanism for the incorpo-
ration of CO feedstock into organic compounds and
polymers (mostly co-polymers with ethylene) and is in-
deed unequalled in importance in catalytic conversions
involving CO. This transformation has many practical
applications and a great number of catalytic carbonyla-
tions are employed in both laboratory and industrial
syntheses [1]. The relationship between anionic carbene
complexes and acyl complexes of Group 6 metals is
shown in Scheme 1.
that are either situated in heterocyclic ring systems
[14] or contain heteroatoms (O or N) bonded to the
a-carbon atom [15].
Herein, we report the first successful method for the
isolation of unattached gold acyl complex systems as
well as a simple, related protocol for the preparation
of imidoylgold(I) compounds.
2. Experimental
There are no reports in the literature of simple inser-
tion reactions of CO into Au–C bonds. It has, however,
been observed that Au–C(O)OR derivatives can be ob-
tained from reactions of gold(III) complexes under con-
ditions where alkoxy (OR) complexes may be expected
as intermediates [9].
From the above-mentioned considerations, it be-
comes clear that the synthesis of gold–acyl complexes
along conventional reaction pathways is not possible.
We have reported previously [10] that alkoxycarbene
complexes deprotonated a to the carbene carbon react
with Ph3PAu+ (a fragment isolobal to H+) by transmet-
allation and ꢁaurolysisꢀ (a second formal metal ex-
change), to form alkoxyvinylgold compounds. Herein,
the C(carbene)–CH2 connectivity appears largely as a
double bond coordinated to expelled Cr(CO)5 (Eq. (1)).
2.1. General procedures and instruments
All solvents were dried and purified by conventional
methods and freshly distilled under nitrogen shortly be-
fore use. Unless otherwise stated, all common reagents
were used as obtained from commercial suppliers with-
out further purification. PhLi was freshly prepared by
the reaction of metallic Li with bromobenzene in
Et2O. MeLi, PhLi and BuLi were standardised before
use according to the procedure reported by Winkle
et al. [16]. Ph3PAuCl [17], the tetrabutylammonium ace-
tyl/benzoylpentacarbonyl chromates and tungstates [18],
and the group 6 metal pentacarbonyl Fischer-type
aminocarbene starting complexes [19] were prepared
according to the procedures described in the literature.
All reactions and manipulations involving organometal-
lic reagents were carried out under a dry nitrogen atmo-
sphere using standard Schlenk and vacuum-line
techniques. NMR spectra were recorded on Varian IN-
OMe
OMe
(CO)5Cr
Ph3PAuCl
Ph3PAu
+
CH2Li
CH2
ð1Þ
1
OVA 600 (600 MHz for H, 151 MHz for 13C{1H} and
Cr(CO)5
243 MHz for 31P{1H}) or Varian VXR 300 (300 MHz
for 1H, 75.4 MHz for 13C{1H} and 121.5 MHz for
31P{1H}) NMR spectrometers. 1H and 13C chemical
shifts are reported in ppm relative to the 1H and 13C res-
idue of the deuterated solvents. 31P chemical shifts are
reported in ppm relative to an 85% H3PO4 external stan-
dard solution. IR spectra (4000–400 cmꢀ1, resolution
4 cmꢀ1) were recorded on a Perkin–Elmer 1600 series
FTIR spectrometer. FAB-MS were recorded on a
Micromass DG 70/70E mass spectrometer using xenon
gas as bombardment atoms and m-nitrobenzylalcohol
as matrix. Flash column chromatography was per-
formed with ‘‘flash grade’’ silica (SDS 230–400 mesh).
Crystal structure data collections were carried out on a
Nonius KappaCCD diffractometer. Elemental analyses
were carried out at the Chemistry department, Univer-
sity of Cape Town.
The question we address in the present work is
whether it is possible to similarly express the C(O)R
and C(NR2)R1 double bonds in untrapped acyl and imi-
doyl complexes of gold by using deprotonated hydroxy-
carbene (that is acylate – compare Scheme 1) and
aminocarbene (imidoylate) complexes of group 6 metals
as starting materials. Haupt recently successfully
trapped the first acylgold(I) complexes along a similar
synthetic pathway as O-coordinated, axial ligands to
Re2(l-PPh2)(CO)7 fragments; the uncoordinated gold–
acyl species, however, remained elusive [11].
Previous work in our laboratory not only involved
the isolation and characterisation of anionic acylate
complexes [12], but also their utilisation as O-donor li-
gands towards harder metals such as Zr(IV) for catalytic
application [13].
Although many interesting imidoyl complexes of gold
have been reported, all of these involve imine groups
2.2. Preparation of 1
1.2 mol equiv. of standardised PhLi was added drop-
wise to 0.80 mmol (282 mg) W(CO)6 suspended in 20 ml
Et2O at room temperature over a period of 10–15 min.
After addition of the PhLi, the mixture was stirred at
room temperature for 30 min, after which it was cooled
-
O
R
O
R
-
(CO)5M
(CO)5M
Scheme 1.

H.G. Raubenheimer et al. / Inorganica Chimica Acta 358 (2005) 4217–4228
4219
to ꢀ78 ꢁC. AuPPh3Cl (0.80 mmol; 395 mg) was then
added to the cooled reaction mixture. After stirring at
ꢀ78 ꢁC for 30–45 min, the reaction mixture was allowed
to warm to room temperature over a period of 212 h, dur-
ing which time the AuPPh3Cl dissolves and a white LiCl
precipitate forms. The mother liquor of the product mix-
ture was carefully removed with a syringe upon settling
of the LiCl precipitate. Carefully cooling the mother li-
quor upon removal of the solvent in vacuo resulted in
the formation of crystals of 1 in moderate yield [65%].
The yield was not optimised.
11.1 Hz, PPhCortho), 128.8 (s, PhCbenzoyl). 31P NMR
(CD2Cl2, d): 37.9 (s). Anal. Calc. for C25H20OPAu: C,
53.21; H, 3.57; P, 5.49. Found: C, 53.27; H, 3.61; P, 5.50%.
Complex 3: yellow crystals, yield 462 mg, 96%. Phys-
ical and spectroscopic data agree with those reported
previously [25].
2.4. Preparation of 4–8
0.4 ml of 1.6 M BuLi (1 mol equiv.) was added to a
solution of the Fischer-type aminocarbene complex
(0.80 mmol) in 15 ml thf cooled to ꢀ78 ꢁC. The mixture
was stirred at that temperature for 10 min, after which
1 mol equiv. Ph3PAuCl (395 mg) was added to the solu-
tion. After stirring this mixture at ꢀ78 ꢁC for 30 min, it
was allowed to warm to room temperature over a period
of 21/2 h. Removal of the solvent in vacuo resulted in
dark yellow to brown oily residues containing a mixture
of products (TLC). The desired products were isolated
in moderately low to high yields [32% (4), 41% (5),
56% (6), 85% (7) and 57% (8)] as broad yellow bands
by means of low temperature (ꢀ15 ꢁC) silica gel column
chromatography (hexane/diethyl ether, 2:1). Yields were
not optimised.
Complex 1: orange microcrystalline material, yield
839 mg, 65%. IR (KBr, cmꢀ1): m 1560 m, 1870 st, 1917
st, 1940 v st, 1972 m, 2065 w. FAB-MS, m/z (relative
intensity) for C59H50O8BrLiP2Au2W: 721 ((Ph3P)2Au+,
70), 565 (benzoyl-AuPPh3+, 25), 459 (Ph3PAu+, 100),
290 (W(CO)–Br, 10), 263 (W–Br, 15). 1H NMR
3
(CD2Cl2, d): 7.2–7.6 (m, 18H, Ph), 8.04 (d, JH–H
=
7.2 Hz, 2H, PhCbenzoyl). 13C NMR (CD2Cl2, d): 203.3
(br, s, C@O), 199.5 (s, COtrans), 192.1 (s, COcis),
146.9(s, PhCbenzoyl), 134.6 (d, 3JP–C =13.1 Hz, PPhCmeta),
131.9 (s, PPhCpara), 131.2 (d, 1JP–C = 53.0 Hz, PPhCipso),
2
130.7 (s, PhCbenzoyl), 129.6 (d, JP–C
= 11.1 Hz,
PPhCortho), 127.9 (s, PhCbenzoyl), 126.1 (s, PhCbenzoyl).
31P NMR (CD2Cl2, d): 39.8 (s). Anal. Calc. for
C59H50O8BrLiP2Au2W: C, 43.92; H, 3.12; P, 3.84.
Found: C, 43.52; H, 2.98; P, 3.95%.
Complex 4: yellow oil, yield 178 mg, 32%. IR (pen-
tane, cmꢀ1): m 1902 st, 1929 v st, 1950 w, 2060 w.
FAB-MS, m/z (relative intensity) for C25H19NO5PAuCr:
721 ((Ph3P)2Au+, 70), 459 (Ph3PAu+, 100), 694 (M+, 55).
1H NMR (CDCl3, d): 8.89 (br s, 1H, NH), 7.4–7.7 (m,
15H, Ph), 2.28 (s, 3H, CH3). 13C NMR (CDCl3, d):
2.3. Preparation of 2 and 3
2
Complexes 2 and 3 were prepared in much the same
way, except that 0.80 mmol Bu4N benzoylpentacarbonyl
tungstate, prepared and isolated separately, was em-
ployed in this reaction instead of the freshly prepared
lithium benzoylpentacarbonyl tungstate used previ-
ously. Instead of a LiCl precipitate forming in the reac-
tion mixture upon completion of the conversion, a
precipitate of 3 forms, which is isolated in near quanti-
tative yield (96%) by careful removal of the cooled
(ꢀ20 ꢁC) mother liquor, now containing only 2, with a
syringe. The isolated precipitate of 3 can be washed with
cooled 10 ml portions of Et2O to remove any remaining
traces of 2 if so required. 2 was isolated in high yield
(72%) by prolonged crystallisation (ꢁ10 weeks) at
ꢀ20 ꢁC from concentrated Et2O solutions, prepared di-
rectly from the mother liquor. Yields were not
optimised.
239.6 (d, JP–C = 122.1 Hz, NC), 221.8 (s, COtrans),
3
216.6 (s, COcis), 134.6 (d, JP–C = 13.1 Hz, PhCmeta),
132.0 (br s, PhCpara), 130.3 (d, 1JP–C = 51.4 Hz, PhCipso),
2
129.6 (d, JP–C = 11.0 Hz, PhCortho), 42.1 (br s, CH3).
31P NMR (CDCl3, d): 41.9 (s). Anal. Calc. for
C25H19NO5PAuCr: C, 43.31; H, 2.76; N, 2.02. Found:
C, 43.12; H, 2.61; N, 2.15%.
Complex 5: yellow oil, yield 232 mg, 41%. IR (pen-
tane, cmꢀ1): m 1904 st, 1931 v st br, 1953 w, 2058 w.
FAB-MS, m/z (relative intensity) for C26H21NO5PAuCr:
1
721 ((Ph3P)2Au+, 65), 459 (Ph3PAu+, 100). H NMR
(CD2Cl2, d): 7.4–7.7 (m, 15H, Ph), 3.24 (s, 3H,
NCH3), 2.25 (s, 3H, CH3). 13C NMR (CD2Cl2, d):
236.8 (d, JP–C = 129.8 Hz, NC), 221.8 (s, COtrans),
2
3
216.0 (s, COcis), 134.6 (d, JP–C = 13.4 Hz, PhCmeta),
1
131.9 (s, PhCpara), 130.6 (d, JP–C = 52.8 Hz, PhCipso),
2
3
129.5 (d, JP–C = 11.0 Hz, PhCortho), 49.4 (d, JP–C
=
Complex 2: orange crystals, yield 325 mg, 72%. IR
(KBr, cmꢀ1): m 1606 m. FAB-MS, m/z (relative intensity)
for C25H20OPAu: 721 ((Ph3P)2Au+, 75), 564 (M+, 10), 459
9.3 Hz, CH3), 32.0 (s, NCH3). 31P NMR (CD2Cl2, d):
41.9 (s). Anal. Calc. for C26H21NO5PAuCr: C,
44.15; H, 2.99; N, 1.98. Found: C, 44.01; H, 3.05; N,
1.96%.
(Ph3PAu+, 100). 1H NMR (CD2Cl2, d): 7.91 (d, 3JH–H
=
7.2 Hz, 2H PhbenzoylCortho), 7.2–7.6 (m, 18H, Ph). 13C
Complex 6: dark yellow oil, yield 338 mg, 56%. IR
(pentane, cmꢀ1): m 1904 st, 1930 v st br, 1951 w, 2058
w. FAB-MS, m/z (relative intensity) for C30H21NO5-
PAuCr: 721 ((Ph3P)2Au+, 55), 694 (M+, 100), 459
2
NMR (CD2Cl2, d): 198.1 (d, JP–C = 6.0 Hz, C@O),
140.1 (s, PhCbenzoyl), 134.6 (d, 3JP–C = 13.1 Hz, PPhCmeta),
132.2 (s, PhCbenzoyl), 131.9 (s, PPhCpara), 131.2 (d, 1JP–C
53.0 Hz, PPhCipso), 130.1 (s, PhCbenzoyl), 129.6 (d, 2JP–C
=
=
1
(Ph3PAu+, 80). H NMR (CD2Cl2, d): 9.59 (br s, 1H,

4220
H.G. Raubenheimer et al. / Inorganica Chimica Acta 358 (2005) 4217–4228
NH), 7.3–7.7 (m, 20H, Ph). 13C NMR (CD2Cl2, d):
232.4 (d, JP–C = 131.9 Hz, NC), 221.6 (s, COtrans),
(CDCl3, d): 7.4–7.7 (m, 20H, Ph), 3.34 (s, 3H, NCH3).
2
2
13C NMR (CDCl3, d): 235.9 (d, JP–C = 124.6 Hz,
NC), 203.6 (s, COtrans), 199.8 (s, COcis), 148.8 (br, s,
3
216.1 (s, COcis), 149.1 (br, s, PhCipso), 134.5 (d, JP–C
=
13.4 Hz, PPhCmeta), 131.9 (s, PPhCpara), 130.9 (s,
3
PhCipso), 134.1 (d, JP–C = 10.5 Hz, PPhCmeta), 131.8
1
PhCo/m/p), 130.2 (d, JP–C = 51.2 Hz, PPhCipso), 129.5
1
(s, PPhCpara), 132.0 (s, PhCo/m/p), 130.0 (d, JP–C
=
2
2
(d, JP–C = 11.0 Hz, PPhCortho), 129.0 (s, PhCo/m/p),
53.2 Hz, PPhCipso) 129.2 (d, JP–C = 9.0 Hz, PPhCortho),
129.1 (s, PhCo/m/p), 125.8 (s, PhCo/m/p), 32.4 (s, NCH3).
31P NMR (CDCl3, d): 39.3 (s). Anal. Calc. for
C31H23NO5PAuW: C, 41.31; H, 2.57; N, 1.55. Found:
C, 41.41; H, 2.59; N, 1.60%.
126.2 (s, PhCo/m/p). 31P NMR (CD2Cl2, d): 42.3 (s). Anal.
Calc. for C30H21NO5PAuCr: C, 47.70; H, 2.80; N, 1.85.
Found: C, 47.65; H, 2.76; N, 2.01%.
Complex 7: yellow oil, yield 603 mg, 85%. IR (pen-
tane, cmꢀ1): m 1904 st, 1924 v st br, 1962 w, 2062 w.
FAB-MS, m/z (relative intensity) for C30H21NO5PAuW:
830 (M-2CO, 35), 721 ((Ph3P)2Au+, 70), 459 (Ph3PAu+,
2.5. X-ray structure determinations
1
100). H NMR (CD2Cl2, d): 10.05 (br s, 1H, NH), 7.4–
7.7 (m, 20H, Ph). 13C NMR (CD2Cl2, d): 238.4 (br s,
NC), 203.3 (s, COtrans), 199.6 (s, COcis), 149.4 (br, s,
The crystal data collection and refinement details for
complexes 1, 2, 3 and 6 are summarised in Table 1.
X-ray quality orange prismatic single crystals of 1 were
obtained by crystallisation of the cooled reaction mix-
ture. Orange platelet crystals of 2, suitable for X-ray dif-
fraction, were obtained by prolonged crystallisation
(ꢁ10 weeks) from a concentrated thf solution layered
with Et2O and cooled to ꢀ20 ꢁC. Good quality crystals
of 3 (yellow platelets) were obtained overnight by re-
crystallisation of the isolated compounds from a thf
solution layered with pentane and cooled to ꢀ20 ꢁC.
X-ray quality crystals (yellow needles) of 6 were only ob-
tained after an extended period (2 months) of crystallisa-
3
PhCipso), 134.3 (d, JP–C = 10.3 Hz, PPhCmeta), 132.0
1
(s, PPhCpara), 131.3 (s, PhCo/m/p), 130.1 (d, JP–C
=
2
54.3 Hz, PPhCipso) 129.4 (d, JP–C = 9.1 Hz, PPhCortho),
129.3 (s, PhCo/m/p), 126.5 (s, PhCo/m/p). 31P NMR
(CD2Cl2, d): 41.2 (s). Anal. Calc. for C30H21NO5PAuW:
C, 40.61; H, 2.39; N, 1.58. Found: C, 40.54; H, 2.46; N,
1.65%.
Complex 8: yellow oil, yield 411 mg, 57%. IR (pen-
tane, cmꢀ1): m 1905 st, 1926 v st br, 1964 w, 2064 w.
FAB-MS, m/z (relative intensity) for C31H23NO5PAuW:
1
721 ((Ph3P)2Au+, 75), 459 (Ph3PAu+, 100). H NMR
Table 1
Crystal data and structure refinements for 1, 2, 3 and 6
Compound
1
2
3
6
Chemical formula
MW (g/mol)
C
59H50O8BrLiP2Au2W
C25H20OPAu
564.37
C21H36O5NClW
601.81
C30H21O5NPAuCr
755.43
1613.61
monoclinic
P21/n
Crystal system
Space group
Unit cell dimensions
monoclinic
P21/n
monoclinic
C2/c
orthorhombic
Pc21b
˚
a (A)
15.4626(3)
19.2859(5)
19.7262(4)
90
12.8710(3)
10.3025(3)
16.6403(5)
90
15.7654(2)
19.5093(2)
18.0605(2)
90
11.0724(2)
12.6013(2)
20.4100(3)
90
˚
b (A)
˚
c (A)
a (ꢁ)
b (ꢁ)
c (ꢁ)
101.147(2)
90
109.771(1)
90
107.932(1)
90
90
90
3
˚
Volume (A )
5771.6(2)
4
1.857
2076.5(1)
4
1.805
5285.1(1)
8
1.513
2847.74(8)
4
1.762
Z
Dcalc (g/cm3)
Temperature (K)
lMo Ka (cmꢀ1
2hmax (ꢁ)
Crystal size (mm)
173(2)
7.859
24.71
188(2)
7.174
29.13
173(2)
4.499
27.00
173(2)
5.623
26.99
)
0.05 · 0.09 · 0.13
ꢀ18 6 h 6 18,
ꢀ22 6 k 6 21,
ꢀ23 6 l 6 22
17 356
0.06 · 0.08 · 0.15
ꢀ17 6 h 6 17,
ꢀ14 6 k 6 12,
ꢀ22 6 l 6 22
9883
0.17 · 0.21 · 0.25
ꢀ19 6 h 6 20,
ꢀ23 6 k 6 24,
ꢀ23 6 l 6 23
18 275
0.05 · 0.10 · 0.13
ꢀ14 6 h 6 13,
ꢀ15 6 k 6 16,
ꢀ26 6 l 6 23
15 131
Index range
Number of reflections collected/observed
Number of independent reflections
Number of observed reflections
Parameters
9760 (Rint = 0.0512)
6855
686
5507 (Rint = 0.0451)
4031
325
5762 (Rint = 0.0314)
4547
267
6056 (Rint = 0.0674)
4414
356
R1 (Fo > 2rFo)
wR2 (all data)
Goodness of fit (S)
0.0476
0.0930
1.081
0.0395
0.0867
1.057
0.0245
0.0573
1.028
0.0374
0.0711
0.964

H.G. Raubenheimer et al. / Inorganica Chimica Acta 358 (2005) 4217–4228
4221
OLi
Ph
tion from a concentrated Et2O solution layered with
2W(CO)6
2PhLi(LiBr)
+
(CO) W
2
5
pentane at ꢀ20 ꢁC.
Low temperature (ꢀ100 ꢁC for 1, 3 and 6, and
ꢀ85 ꢁC for 2) diffraction data for all the crystals were
collected on an Enraf–Nonius KappaCCD diffractome-
ter [20] using graphite monochromated Mo Ka radia-
2Ph3PAuCl
Et2O
Ph
˚
tion (k = 0.71073 A). All the data sets were scaled,
AuPPh3
O
O
O
O
reduced and corrected for Lorentz and polarisation ef-
fects using DENZO-SMN [21]. The structures of 1 and
6 were solved with direct methods and refined anisotrop-
ically for all the non-hydrogen atoms, except those men-
tioned below, by full-matrix least squares calculations
on F2. Dynamic disorder of the C(3)–O(3) and C(4)–
O(4) CO ligands in the structure of 1 prevented the
anisotropic refinement of the C(3) and C(4) atomic
positions.
+
Li
AuPPh3
CO
W
C
Ph
OC
Br
CO
CO
1
Scheme 2.
The structures of 2 and 3 were solved by the interpre-
tation of Patterson syntheses, which yielded the posi-
tions of the metal atoms. The structures were
completed by full-matrix least squares calculations
(SHELXL-97) [22] on F2 and, unless stated otherwise, re-
fined anisotropically.
Extensive disorder, in the form of a complete mole-
cule of 2 orientated as a mirror image across a plane
through C1 and C11, was found in the crystal structure
of 2. The low site occupancy of the minor disorder com-
ponent [14.2(1)%] meant that it could only be refined
isotropically and that the phenyl ring carbon atom posi-
tions had to be restrained to idealised 6-membered rings.
In addition, the C(1b) atom was restrained to a reason-
able position by fixing the Au–C(1b) and C(1b)–C(11b)
bond lengths to sensible values.
Hydrogen atoms, with the exception of the imine
hydrogen atom in the structure of 6, were placed in idea-
lised positions with their displacement parameters fixed
at 1.5 times (for aliphatic hydrogen atoms) or 1.2 times
(for aromatic hydrogen atoms) the equivalent isotropic
displacement parameters of their parent atoms. The
aforementioned hydrogen atom in 6 was located on
the difference Fourier map and refined isotropically.
All calculations were performed using SHELX-97 [22]
in the X-SEED [23] environment. ORTEP-III for Win-
dows [24] was used to generate the various figures at
the 50% probability level.
tion could not be proven beyond all reasonable doubt.
Furthermore, decomposition of the complexes
–
indicated by rapid darkening of the reaction mixtures –
was frequently observed during the purification
procedure. One such product, however, crystallised in
moderate yield, in the form of an unusual lithium com-
plex of complexes and solvent (Scheme 2).
The formation of the tetrahedral lithium complex, 1,
is proposed to proceed as follows: LiBr, originally pres-
ent in the freshly prepared PhLi solution employed to
synthesise the lithium benzoylpentacarbonyl tungstate
in the first step of the synthesis, reacts with expelled
W(CO)5 in the reaction mixture. The ionic species,
[BrW(CO)5][Li], is formed, which precipitates, or rapidly
crystallises, from the cooled (ꢀ20 ꢁC) reaction mixture
with two already-formed benzoyl-AuPPh3 fragments
and an Et2O solvent molecule coordinated to the Li+-
cation.
A single crystal X-ray structure determination re-
vealed this unexpected molecular structure for the prod-
uct and also confirmed that the electrophilic addition of
Ph3PAu+ to lithium acylpentacarbonyl tungstates in-
deed results in the formation of the acylgold(I) organyls.
The formation and precipitation/crystallisation of 1 is
driven by its poor solubility in Et2O. Equilibria of halide
salts and coordinatively unsaturated M(CO)5 fragments
are well known [25]. The fact that simple acylgold–
M(CO)5 complexes could not be isolated successfully
in these reactions is thus probably as a result of the poor
stabilisation that the M(CO)5 fragments receive from the
acylgold complex (or Et2O solvent molecules) in
the reaction mixture, as shown by the nature of the
W(CO)5 fragment in 1. No precipitation or crystallisa-
tion of 1 could be achieved in the presence of thf.
The formation of LiBr adducts of W(CO)5, observed
in the crystal structure of 1, indicated a method for the
preparation of the first free acylgold(I) complex, ben-
zoyl-AuPPh3 (2) (Scheme 3). If, instead of lithium ben-
zoylpentacarbonyl tungstate, tetrabutylammonium
3. Results and discussion
3.1. Acyl gold compounds
Attempted syntheses of acyl gold compounds by reac-
tion of Ph3PAuCl and freshly prepared lithium acetyl/
benzoyl pentacarbonyl chromates and tungstates were
unsuccessful. Although evidence from IR analysis of
the product mixtures pointed to weak acylgold–oxygen
coordination to the M(CO)5 fragments, the coordina-

4222
H.G. Raubenheimer et al. / Inorganica Chimica Acta 358 (2005) 4217–4228
R2
R2
R2
-
O
-
O
N
N
BuLi
-
Bu4N+
N
H
+
(CO)5W
+
Li
(CO)5M
Li
Ph3PAuCl
Ph3PAu
(CO)5M
W(CO)5
+
(CO)5M
Ph
R'
R'
R'
Ph
[Bu N]Cl
BuH
4
O
Ph3PAuCl
R'
-
Bu4N+[(CO)5W Cl ]
Ph3PAu
+
LiCl
Ph
R2
N
4: M = Cr, R' = Me, R2 = H,
5: M = Cr, R' = Me, R2 = Me
6: M = Cr, R' = Ph, R2 = H
7: M = W, R' = Ph, R2 = H
8: M = W, R' = Ph, R2 = Me
2
3
(CO)5M
Scheme 3.
AuPPh3
Scheme 4.
benzoylpentacarbonyl tungstate was employed as start-
ing material, the reaction with Ph3PAuCl led to the
formation of benzoyl-AuPPh3, [Bu4N][Cl] and an unsat-
urated W(CO)5 fragment. The latter rapidly reacted with
the [Bu4N][Cl] to yield the ionic species, [ClW(CO)5]-
[Bu4N] (3), which is sparingly soluble in Et2O and rap-
idly precipitates from the cooled (0 ꢁC) reaction mixture.
Since the Bu4N+-cation in 3 cannot accommodate acyl
coordination, as is observed for the Li-cation in the crys-
tal structure of 1, the free benzoyl-AuPPh3 complex (2)
remains in solution and can be isolated in high (72%)
yields by crystallisation from concentrated Et2O solu-
tions. X-ray diffraction studies unequivocally confirmed
the formation of both 2 and 3 in this reaction.
The mechanism involved in the formation of 1 and 2
is not clear. For their reaction, Haupt and co-workers
[11] suggested a concerted mechanism in which a
hypothetical ion pair complex, [Re2{l-PPh2}2(CO)7-
{C(O)R}]ꢀ[AuPPh3]+, is converted to the acylgold(I)-
coordinated rhenium complexes via an assumed
bimetallic g2-acyl transition state. The thermodynamic
considerations they rely on are, however, not valid in
the present situation.
Fischer, Schubert and Sarkar have reported the rear-
rangement of certain Fischer-type carbene complexes to
ketones or aldehydes upon thermolysis [26]. In a recent
paper, Sarkar and co-workers [27] suggests a probable
mechanism for such a reaction involving electrophilic
addition of a carbon electrophile onto the metal moiety,
followed by a formal reductive elimination of a ketone
fragment.
Based on these reports, a mechanism related to the
aurolysis of deprotonated carbene complexes by
Ph3PAu+ [10] should also be considered; W–Au interac-
tion, for example, could occur prior to acylgold
formation.
lowed to slowly warm up to room temperature over a
period of 3 h. Removal of the solvent in vacuo yielded
dark-yellow oily residues containing 4–8 (32–85% yields
after separation). Separation of the complexes was car-
ried out by low temperature (ꢀ15 ꢁC) silica gel column
chromatography with hexane/diethyl ether (2:1) as elu-
ant, followed by crystallisation from a diethyl ether solu-
tion layered with pentane at ꢀ20 ꢁC.
The novel formation of the {acetimidoyl-
AuPPh3}M(CO)5 and {benzimidoyl-AuPPh3}M(CO)5
complexes (Scheme 4), 4–8, probably proceeds along a
reaction path similar to that of 1 and 2. Perhaps the
term aurolysis can therefore also be applied to describe
the current reaction. Furthermore, it is interesting to
note that this conversion appears to be stereospecific,
since only the Z-isomers of 4–8 were isolated.
Coordination of the acetimidoyl-AuPPh3 and benz-
imidoyl-AuPPh3 moieties to the M(CO)5 fragments
through the nitrogen atoms in 4–8 is much stronger than
the potential oxygen coordination of M in {acyl-
AuPPh3}M(CO)5 (2). This is reflected by the fact that
4–8 are, unlike their oxygen analogues, stable in all com-
mon laboratory solvents (under inert conditions), are
not involved in any ligand substitution reactions with
solvent molecules or halide salts, and can be successfully
isolated by column chromatography (SiO2). A single
crystal X-ray structure of 6 confirms the unique molec-
ular structure of these compounds.
The closest example of a similar conversion in the lit-
erature is the thermolysis reaction of Fischer-type
aminocarbene complexes in the presence of CO, which
also yields imines [28].
We have reported that a second deprotonation-
auration reaction with complexes 4 and 6 yields a novel
triangular Au2Cr cluster compound with an extraordi-
˚
3.2. Imidoyl complexes of gold
narily short Auꢂ ꢂ ꢂAu separation (2.69 A) as the only iso-
lable product [29].
The syntheses of complexes 4–8 (Scheme 4) were car-
ried out by irreversible N-deprotonation of the respective
group 6 metal aminocarbene complexes with 1 mol
equiv. n-butyllithium at ꢀ78ꢁC, followed by the addition
of 1 mol equiv. Ph3PAuCl to the reaction mixture. After
addition of the Ph3PAuCl, the reaction mixture was al-
3.3. Spectroscopic characterisation
3.3.1. NMR spectroscopy
Since 1 is an ionic compound, it is unlikely that the
Li-coordination found in the single crystal X-ray

H.G. Raubenheimer et al. / Inorganica Chimica Acta 358 (2005) 4217–4228
4223
structure will persist in solution. However, small vari-
ances in the NMR data of the acylgold fragments in 1
and 2, most notably of the acyl carbon atom 13C chem-
ical shift (d 203.3 in 1, d 198.1 in 2), were observed, sug-
gesting that a weak interaction between the W(CO)5
species and the acylgold complex does occur. It is inter-
esting to note that this resonance in 2 is a small doublet
(2JP–C = 6.0 Hz), whereas it manifests as a broadened
singlet in the 13C NMR spectrum of 1. The acyl carbon
resonance in 2 at d 198.1 is comparable to a value of d
192.0 for such an atom in benzaldehyde. This suggests
that there is not much electron loss to the gold atom
after auration. The spectroscopic analysis of 3 agrees
with that reported previously [25].
13C resonances of the Ph groups bonded to the C@N
imine carbon atom in 6–8 are characterised by a down-
field shifted signal (compared to the starting material)
representing the ipso carbon atom of this ring (d 149.1,
149.4 and 148.8). This signal is, furthermore, broadened
3
as a result of weak JC–P coupling. The 13C resonances
for the methyl carbon atoms in this position in 4 and
5 are found at d 42.1 and 49.4, as a broadened singlet
and a doublet (3JP–C = 9.3 Hz), respectively.
3.3.2. Infrared spectroscopy
The infrared (IR) spectra of 1 and 2 were recorded as
KBr pellets from crystals. The spectrum of 1 exhibits six
clearly separated absorption bands in the carbonyl re-
gion. These values are in good agreement with reported
The 1H NMR spectra of 4, 6 and 7 are characterised, in
particular, by broad imine proton resonances between d
8.89 and d 10.05. These chemical shifts suggest a great deal
of electron donation from the imine nitrogen atom to the
CO vibrational modes for the BrWðCOÞ ꢀ-anion in
5
[BrW(CO)5][R4N] [2064 cmꢀ1 (w, A1ð1Þ), 1904 (st, E),
1868 ðA1ð2ÞÞ] [25], although the ꢁEꢀ vibrational modes
in 1 differ somewhat. The crystal structure of 1 shows
that the BrW(CO)5-anion is coordinated to a Li-cation
through the CO ligand trans to the Br. Although the
Li-coordination to O(5) (Fig. 1) seems to have little ef-
1
M(CO)5 fragments. H and 13C{1H} resonances for the
methyl groups on the imine nitrogen atoms in 5 and 8
are also found shifted downfield from their usual posi-
tions in examples of uncoordinated imines, at d 3.24 and
32.0 (complex 5), and d 3.34 and 32.4 (complex 8) [30].
A good indicator for the amount of electron donation
from the imine groups to the M(CO)5 fragments in 4–8
is the chemical shift of the C@N imine 13C resonances.
These appear as doublets (2JC–P = 122.1–131.9 Hz) be-
tween d 232.4 and 239.6 for 4–6 and 8. Curiously this
resonance in 7 appears in a similar region of the 13C
NMR spectrum (d 238.4), but as a greatly broadened
singlet. These resonances show that the imidoyl carbon
atoms are substantially deshielded and do not exclude
the possibility that they could have a carbene-type
character.
ð2Þ
fect on the A1 vibrational mode, it may be responsible
for substantial distortion within the W(CO)4 coordina-
tion plane because it brings the anion into closer prox-
imity to the rest of the molecules (particularly the
phenyl ring of one of the benzoyl-AuPPh3 fragments
in the structure) in the solid state. This is also apparent
in the observation of a relatively strong B1 absorption
mode, which is only IR active upon distortion of the
C4v local symmetry in the BrW(CO)5 moiety. Instead
of the single twofold degenerate E-vibrational mode
observed for the BrWðCOÞ ꢀ-anion in the [BrW(CO)5]-
5
[R4N] compound, two non-degenerate, clearly separated
Fig. 1. ORTEP view of 1, showing the numbering scheme (disordered CO and Et2O atom positions, hydrogen atoms and anisotropic displacement
parameters of carbon atoms of the PPh3 group and Et2O omitted for clarity). Displacement ellipsoids are shown at the 50% level.

4224
H.G. Raubenheimer et al. / Inorganica Chimica Acta 358 (2005) 4217–4228
vibrational modes are observed. Disorder found in the
atomic positions of two CO ligands as well as the rela-
tively long O(5)–Li separation in the crystal structure
of 1 (possibly due to steric considerations), reported be-
low, supports this interpretation of the IR data. The acyl
C@O stretching frequency at 1560 cmꢀ1 suggests a
stronger acyl bond in 1 than in the literature reported
Re-coordinated benzoyl-AuPPh3 complex [t(C@O)
1508 cmꢀ1].
The IR spectrum of 2 exhibits a single, medium inten-
sity, CO stretching frequency (1606 cmꢀ1) situated next
to two weaker C@Cconj IR absorption modes (1588
and 1573 cmꢀ1). This result suggests, as would be ex-
pected for an uncoordinated benzoyl-AuPPh3 moiety,
substantially greater double bond character for the acyl
C@O bond in 2 than in 1 and the Re-coordinated ben-
zoyl-AuPPh3 complex reported in the literature [11].
The crystal structure of 2, reported below, confirms this
suggestion as it reports the shortest C@O separation of
the three compounds. The isolobal hydrogen analogue
of complex 2, benzaldehyde, exhibits a C@O stretching
frequency at 1702 cmꢀ1, which suggests that there is
either some measure of p-electron back-donation from
the gold atom to the p*-orbital of the acyl carbon atom,
or electron loss from the acyl carbon to the gold atom
that weakens the C@O acyl bond in 2.
The IR spectra of 4–8 were recorded as pentane solu-
tions of the complexes in a NaCl cell. The IR spectra
measured for these complexes each exhibit four clearly
separated absorption bands in the carbonyl region,
which represent the A1ð1Þ, A1ð2Þ, B1 and E vibrational
modes for molecules of the type M(CO)5L. The C@N
stretching frequency, expected to appear between 1500
and 1700 cmꢀ1, could, due to the presence of many
C@Cconj vibrational modes in the same region, unfortu-
nately not be unambiguously identified in the IR spectra
of these compounds.
for 7 the fragment M-2CO was found. Peaks corre-
sponding to free benzoyl-AuPPh3 and some WBr species
were identified in the FAB-mass spectrum of 1.
3.4. X-ray structure determinations
The low temperature crystal and molecular structures
of 1 (173 K), 2 (173 K), 3 (188 K) and 6 (173 K) (Figs. 1,
3–5) were determined by X-ray diffraction techniques.
Selected bond lengths and angles for 1, 2, 3 and 6 are
listed in Table 2.
The molecular structure of 1 (Fig. 1) revealed a novel
electrostatically associated ionic structure for this
compound. It consists of a Li-cation that is roughly
tetrahedrally coordinated to an Et2O molecule, a
BrW(CO)5-anion and two benzoyl-AuPPh3 complexes.
Coordination of the Li-cation to the BrW(CO)5 anion
is through the oxygen atom of the CO ligand trans to
the Br atom, while coordination to the two benzoyl-
AuPPh3 complexes is through the oxygen atoms of the
acyl moieties. The benzoyl-AuPPh3 complexes are,
furthermore, linked to each other by weak aurophilic
˚
Au–Au interactions [Au(1)ꢂ ꢂ ꢂAu(2) = 3.1213(5) A] with
a torsion angle [C(6)–Au(1)–Au(2)–C(7)] of 103.0(4)ꢁ
between the two pseudo-linear Au(I) coordination
modes. C(6)–Au(1)–P(1) and C(7)–Au(2)–P(2) angles of
175.6(3)ꢁ and 172.5(3)ꢁ describe how the linear coordina-
tion mode of Au(I) is distorted towards the neighbouring
gold atom due to the aforementioned aurophilic interac-
tions. An intramolecular non-bonded face-to-edge phenyl-
interaction, involving a phenyl ring from each PPh3 group
in the benzoyl-AuPPh3 moieties [C(13X) and C(22X)
rings], further stabilises the Au–Au linked configuration.
The tetrahedral coordination of the Li-cation is
somewhat distorted from the ideal, with the largest an-
gle between two of the coordinating oxygen atoms being
124.2(9)ꢁ [O(6)–Li–O(7)] between the oxygen atoms of
the two coordinating benzoyl-AuPPh3 complexes. The
BrW(CO)5-anion exhibits dynamic disorder (not shown
in Fig. 1) in the positions of two of its CO ligands [C(3)–
O(3) and C(4)–O(4)]. This could be partially modelled
by two different positions for these CO groups identified
from the difference Fourier map with site occupancies of
3.3.3. Mass spectroscopy
FAB-mass spectra of 1–8 were characterised by high
intensity peaks for PPh3Au+ and (PPh3)2Au+ fragments,
the latter being an artifact of the analytical method.
Molecular ions could be identified for 2, 4 and 6, while
Table 2
˚
Selected bond lengths (A) and angles (ꢁ)
Compound
1
2
3
6
Au–P
Au–Csp2
2.312(3), 2.317(3)
2.046(10), 2.079(9)
1.238(11), 1.234(10)
3.1213(5)
2.313(1)
2.085(5)
1.200(7)
2.301(2)
2.034(7)
1.281(9)
Csp2–O/N
Au. . .Au
W/Cr–Br/Cl
W/Cr–Ctrans
W/Cr–Ccis (average)
2.6913(12)
1.909(12)
2.01(2)
2.5602(8)
1.939(4)
2.033(4)
1.853(8)
1.898(9)
Csp2–Au–P
Au–Csp2–O/N
175.6(3), 172.5(3)
122.3(8), 121.1(7)
176.4(1)
123.4(4)
171.2(2)
124.2(6)

H.G. Raubenheimer et al. / Inorganica Chimica Acta 358 (2005) 4217–4228
4225
0.63:0.37. Elongated anisotropic displacement parame-
ters on O(3) and O(4) [C(3) and C(4) could not be aniso-
tropically refined] indicated the presence of the dynamic
disorder. A small degree of dynamic disorder (not
shown in Fig. 1) in the ethyl groups of the Et2O mole-
cule, as indicated by enlarged anisotropic displacement
parameters for the atoms involved, is also observed.
The reason for this disorder can be seen in the packing
of 1 (Fig. 2), where the molecules pack in regular rows in a
structure that leaves two open channels parallel to the b-
axis per unit cell. Both the disordered CO ligands and
the Li-coordinated Et2O solvent molecules border di-
rectly on these channels, allowing freedom of movement
of these groups within the crystal lattice. Furthermore, ra-
pid crystal degradation was observed upon isolation of
the crystalline mass of 1 and is ascribed to the loss of
Et2O from the crystal lattice along these channels.
The crystal and molecular structure of the free ben-
zoyl-AuPPh3 complex, 2 (Fig. 3), represents the first
example of its kind.
˚
The Au(1a)–C(1a) bond [2.085(5) A] in 2 is within the
expected range for Au(I)–C(sp2) single bonds [31]. The
Fig. 2. Packing diagram of 1, viewed along the b-axis of the unit cell showing the channels and the positions of the Et2O molecules and disordered
CO ligands within the unit cell.
Fig. 3. ORTEP view of 2 (a-orientation with 85.8% site occupancy), showing the numbering scheme. Displacement ellipsoids are shown at the 50%
level.

4226
H.G. Raubenheimer et al. / Inorganica Chimica Acta 358 (2005) 4217–4228
˚
Au(1a)–P(1a) bond [2.313(1) A] is also normal [11,32].
O(1a) is unmistakably double bonded to the acyl carbon
The structure of 6 (Fig. 5) clearly exhibits a benzimi-
doyl-AuPPh3 moiety coordinated through its imine
nitrogen atom to a Cr(CO)5 group and demonstrates
that it is the Z-isomer that forms. Compound 6 is the
first organoimidoyl complex of gold for which the crys-
tal and molecular structures have been determined. This
complex, furthermore, acts as a neutral N-donor ligand
to a Cr(CO)5 fragment, thereby also being the first
example of a bimetallic imidoylgold(I) complex that is
coordinated through the imine nitrogen atom to a metal
other than gold.
˚
atom [C(1a)–O(1a) = 1.200(7) A]. This bond is also, as
would be expected, slightly shorter than the correspond-
ing Li-coordinated and Re-coordinated C@O bonds in
the structures of 1 and the only other structurally char-
acterised acylgold(I) adduct referred to above [11]. The
sp2 hybridisation of C(1a) is confirmed by its nearly per-
fectly flat bonding geometry [the greatest deviation from
the least-squares plane defined by Au(1a), C(1a), O(1a),
˚
C(11a) is 0.012(4) A for C(1a)]. Despite the fact that no
˚
intermolecular metal–metal interactions are observed in
the crystal structure of 2, the P(1a)–Au(1a)–C(1a) angle
deviates from 180ꢁ [P(1a)–Au(1a)–C(1a) = 176.4(1)ꢁ].
This is most likely due to packing effects, since the acyl
oxygen atom, O(1a), is situated close to one of the phe-
nyl rings of the PPh3 group of a neighbouring molecule.
The other two phenyl rings are involved with face-to-
edge p interactions with neighbouring molecules.
The Au–C(1) bond length in 6 [2.034(7) A] is typical
for an Au(I)–C(sp2) r-bond [31]. The C(1)–N separation
˚
[1.281(9) A] confirms the double bond character of this
bond. There is a near planar bonding geometry around
C(1) [deviations from the least squares plane calculated
through C(1), Au, C(41) and N are 0.024(5), 0.006(1),
0.008(2) and 0.010(2) A, respectively].
Deviations from planarity of the least squares plane
˚
The crystal and molecular structure of [ClW(CO)5]-
[Bu4N] (3), which precipitated from the same reaction
mixture as 2, is shown in Fig. 4. Although 3 is a well-
known and stable organometallic compound and has
often been applied as starting material in reaction se-
quences [25], its crystal structure has not yet been
reported. Bond lengths and angles within the
W(CO)5Cl-anionic and the Bu4N-cationic components
of 3 are normal for such moieties [33]. A W–C(4) separa-
through Cr, C(2), C(3), C(4) and C(5) of 0.045(3),
˚
ꢀ0.077(4), 0.055(4), ꢀ0.075(4) and 0.052(4) A describe
its distortion due to the coordination of the sterically
large benzimidoyl-AuPPh3 moiety. A small Au–C(1)–
Cr–C(2) torsion angle of 13.3(5)ꢁ shows that the
Au–C(1) bond is surprisingly close to parallel to the
Cr–C(2) bond. This is also reflected in the 170.3(6)ꢁ
and 171.2(2)ꢁ Cr–C(2)–O(2) and P–Au–C(1) angles that
are distorted from linearity (vide infra).
˚
˚
tion of 1.939(4) A, which is ꢁ0.1A shorter than the other
W–C distances in the structure, shows the influence of the
weaker (compared to CO) trans directing chlorine ligand.
Compound 3 exhibits an efficient packing arrange-
ment that probably aids its crystallisation from the reac-
tion mixture. Layers of cations and anions are closely
packed, roughly parallel to the a-axis, with no solvent-
accessible or other significant voids or channels present.
The position of the imine hydrogen atom (H) could
be determined from the difference Fourier map and re-
fined isotropically. This hydrogen atom position is dis-
torted toward the Cr(CO)5 fragment, as can be seen in
the C(1)–N–H and Cr–N–H angles [127(6)ꢁ and 96(6)ꢁ]
˚
and the short Cr–H separation [2.35(8) A].
Fig. 4. ORTEP view of 3, showing the numbering scheme. Displace-
ment ellipsoids are shown at the 50% level.
Fig. 5. ORTEP view of 6, showing the numbering scheme. Displace-
ment ellipsoids are shown at the 50% level.

H.G. Raubenheimer et al. / Inorganica Chimica Acta 358 (2005) 4217–4228
4227
(c) F. Calderazzo, Angew. Chem., Int. Ed. Engl. 16 (1977) 299;
(d) E.J. Kuhlman, J.J. Alexander, Coord. Chem. Rev. 33 (1980)
195;
Short non-bonded intermolecular interactions be-
tween molecules of 6 occur along the a-axis of the unit
cell, involving the oxygen atom [O(6)] of the CO ligand
trans to the imine coordination, a hydrogen atom from
the benzimidoyl ring [H(43)] and a hydrogen atom from
a phenyl ring in the PPh3 group [H(33)]. It is likely that
this interaction and an offset herringbone packing con-
figuration of the P–Au–C(1) bonds are also partly
responsible for the high degree of distortion from the
usually linear P–Au–C(1) coordination mode.
(e) J.P. Collman, L.S. Hegedus, J.R. Norton, R.G. Finke,
Principles and Applications of Organotransition Metal Chemistry,
second ed., University Science Books, Mill Valley, CA, 1980, p.
619.
[2] (a) L.D. Durfee, I.P. Rithwell, Chem. Rev. 88 (1988) 1059;
(b) A.R. Cutler, P.K. Hanna, J.C. Vites, Chem. Rev. 88 (1988)
1363;
(c) T. Kinnunen, K.J. Laason, Organomet. Chem. 628 (2001) 222.
[3] J.R. Moss, J. Mol. Catal. A: Chem. 107 (1996) 169.
[4] M.E. Dry, Catal. Today 6 (1990) 183.
[5] (a) E.O. Fischer, A. Maasbo¨l, Angew. Chem., Int. Ed. Engl. 3
(1964) 580;
4. Conclusions
(b) E.O. Fischer, A. Maasbo¨l, Chem. Ber. 100 (1967) 2445.
[6] M.F. Semmelhack, G.R. Lee, Organometallics 6 (1987) 1839.
[7] (a) See for example: J.A. Van Doorn, C. Masters, H.C. Volger, J.
Organomet. Chem. 105 (1976) 245;
In this paper, we have described how formally
deprotonated pentacarbonyl(group 6)hydroxy- and
aminocarbene (alternatively, acyl and imidoyl) com-
plexes may function as synthons in transmetallation
for the preparation of unprecedented free acyl and
(CO)5M-N-coordinated (M = Cr, W) imidoyl com-
pounds of gold. The mechanism of these reactions is
not known, but metal–metal (group 6ꢂ ꢂ ꢂAu) interac-
tions deserve consideration. It is clear, however, that
acyl complexes of gold do not form by CO-insertion
into the gold–carbon single bond. In the presence of
other acidic and basic units, the gold acyl complex en-
gages in various forms of intramolecular interaction in
the solid state. The crystal and molecular structures of
a well-known member of the (CO)5MXꢀ family,
[(CO)5WCl]NBu4, have also been determined.
(b) J.A. Gladysz, J.C. Selover, C.E. Strouse, J. Am. Chem. Soc.
100 (1978) 6766;
(c) E.O. Fischer, Adv. Organomet. Chem. 14 (1976) 1.
[8] (a) J.E. Ellis, J. Organomet. Chem. 86 (1975) 1;
(b) R.B. King, Acc. Chem Res. 3 (1970) 417.
[9] S. Komiya, T. Sone, S. Ozaki, M. Ishikawa, N.J. Kasuga,
Organomet. Chem. 428 (1992) 303.
[10] H.G. Raubenheimer, M.W. Esterhuysen, A. Timoshkin, Y. Chen,
G. Frenking, Organometallics 21 (2002) 3173.
[11] H.-J. Haupt, D. Petters, U. Flo¨rke, J. Organomet. Chem. 553
(1998) 497.
[12] M.W. Esterhuysen, H.G. Raubenheimer, Eur. J. Inorg. Chem.
(2003) 3061.
[13] N. Luruli, V. Grumel, R. Brull, A. duToit, H.G. Raubenheimer,
¨
J. Polym. Sci., Part A: Polym. Chem. 42 (2004) 5121.
[14] (a) A. Burini, R. Bravi, J.P. Fackler Jr., R. Galassi, T.A. Grant,
M.A. Omary, B.R. Pietroni, R.J. Staples, Inorg. Chem. 39
(2000) 3158;
(b) W.P. Fehlhammer, L.F. Dahl, J. Am. Chem. Soc. 94 (1972)
3370;
Acknowledgements
(c) H.G. Raubenheimer, F. Scott, G.J. Kruger, J.G. Toerien, J.R.
Otte, W. van Zyl, T. Taljaard, P.J. Olivier, L. Linford, J. Chem.
Soc., Dalton Trans. (1994) 2091;
We thank the NRF, the Volkswagen-Stiftung and the
University of Stellenbosch for financial support.
(d) H.G. Raubenheimer, P.J. Olivier, L. Lindeque, M. Desmet, J.
Hrusak, G.J. Kruger, J. Organomet. Chem. 544 (1997) 91;
(e) K.M. Lee, C.K. Lee, I.J.B. Lin, Angew. Chem., Int. Ed. Engl.
36 (1997) 1850.
Appendix A. Supplementary data
[15] (a) A.L. Balch, M.M. Olmstead, J.C. Vickery, Inorg. Chem. 38
(1999) 3494;
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic
Data Centre, CCDC Nos. 263482 (1), 263483 (2),
263484 (3) and 263485 (6). Copies of this information
may be obtainded free of charge from the Director,
CCDC, 12 Union Road, Cambridge, CB2 1E2, UK
(fax: +44-1223-336-033; e-mail: deposit@ccdc.ca.ac.uk,
www: http://www.ccdc.cam.ac.uk). Supplementary data
associated with this article can be found, in the online
version, at doi:10.1016/j.ica.2005.03.029.
(b) M. Lanfranchi, M.A. Pellinghelli, A. Tiripicchio, F. Bonati,
Acta Crystallogr., Sect. C (Cr. Str. Comm.) 41 (1985) 52;
(c) A. Tiripicchio, M. Camelini, G. Minghetti, J. Organomet.
Chem. 171 (1979) 399;
(d) J.C. Vickery, A.L. Balch, Inorg. Chem. 36 (1997) 5978.
[16] M.R. Winkle, J.M. Lansinger, R.C. Roland, J. Chem. Soc.,
Chem. Commun. (1980) 87.
[17] J.P. Fackler Jr., Inorg. Synth. 21 (1982) 325.
[18] (a) L.S. Hegedus, M.A. McGuire, L.M. Schultz, Org. Synth. 65
(1987) 140;
(b) E.O. Fischer, A. Maasbo¨l, Chem. Ber. 100 (1967) 2445.
[19] E.O. Fischer, M. Leupold, Chem. Ber. 105 (1972) 599.
[20] COLLECT, Data Collection Software, Nonius BV Delft, The
Netherlands, 1998.
References
[21] Z. Otwinowski, W. Minor, Method Enzymol. 276 (1997) 307.
[22] G.M. Sheldrick, SHELX SHELX-97. Program for Crystal Structure
Analysis, University of Go¨ttingen, Germany, 1997.
[23] L.J. Barbour, J. Supramol. Chem. 1 (2001) 189.
[24] L.J. Farrugia, J. Appl. Cryst. 30 (1997) 565.
[1] (a) G.E. Coates, M.L.H. Green, K. Wade, Organometallic
compounds, Methuen, London, 1968, p. 257;
(b) A. Wojcicki, Adv. Organomet. Chem. 11 (1973) 87;

4228
H.G. Raubenheimer et al. / Inorganica Chimica Acta 358 (2005) 4217–4228
[25] J.L. Atwood, S.G. Bott, J.C. Junk, M.T. May, J. Organomet.
Chem. 487 (1995) 7, and references therein.
[30] J.-J. Brunet, A. Capperrucci, R. Chauvin, Organometallics 15
(1996) 5254.
[26] (a) E.O. Fischer, U. Schubert, J. Organomet. Chem. 170 (1979) C13;
(b) U. Schubert, H. Ho¨rnig, J. Organomet. Chem. 273 (1984) C11;
(c) M. Nandi, K.M. Sathe, A. Sarkar, J. Chem. Soc., Chem.
Commun. (1992) 793.
[31] (a) H.G. Raubenheimer, G.J. Kruger, C.F. Marais, J.T.Z.
Hattingh, L. Linford, P.H. van Rooyen, J. Organomet. Chem.
355 (1988) 337;
(b) H.G. Raubenheimer, M. Desmet, G.J. Kruger, J. Chem. Soc.,
Dalton Trans. (1995) 2067.
[27] K.M. Sathe, M. Nandi, S.R. Amin, V.G. Puranik, A. Sarkar,
Organometallics 15 (1996) 2881.
[32] (a) P.G. Jones, Gold Bull. 14 (1981) 102, 159;
(b) P.G. Jones, Gold Bull. 16 (1983) 114;
(c) P.G. Jones, Gold Bull. 19 (1986) 46.
[28] J.A. Connor, P.D. Rose, J. Organomet. Chem. 46 (1972)
329.
[29] M.W. Esterhuysen, H.G. Raubenheimer, Acta Crystallogr. Sect.
C. 59 (2003) m286.
[33] C. Burschka, W.A. Schenk, Z. Anorg. Allg. Chem. 477 (1981)
149.