Synthesis, isolation and characterization of cationic gold(I) N-heterocyclic carbene (NHC) complexes

Article abstract of DOI:10.1039/b601547f

A number of cationic gold(I) complexes have been synthesized and found to be stabilized by the use of N-heterocyclic carbene ligands. These species are often employed as in situ-generated reactive intermediates in gold catalyzed organic transformations. An isolated, well-defined species was tested in gold-mediated carbene transfer reactions from ethyl diazoacetate. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b601547f

View Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Synthesis, isolation and characterization of cationic gold(I)
N-heterocyclic carbene (NHC) complexes{
Pierre de Fre´mont,a Edwin D. Stevens,a Manuel R. Fructos,b M. Mar D´ıaz-Requejo,b Pedro J. Pe´rezb and
Steven P. Nolan*a
Received (in Berkeley, CA, USA) 8th February 2006, Accepted 16th March 2006
First published as an Advance Article on the web 10th April 2006
DOI: 10.1039/b601547f
examples of gold–phosphines16 and gold–NHC17-mediated trans-
A number of cationic gold(I) complexes have been synthesized
formations have recently appeared. Gold(I) halide complexes are
especially efficient at activating alkyne moieties towards nucleo-
philic addition under mild reaction conditions.16,17 A recent
example by He et al. also shows these complexes to be excellent
co-catalysts in the intra- and intermolecular hydroamination of
unsaturated olefins.18 The use of silver salts, with an accompanying
non-coordinating anion, is usually required to generate the active
catalyst. It is commonly accepted that silver assists in halide
abstraction from the gold center, generating a highly electrophilic
monoligated cationic gold complex.19 While Ferrer and
Echavarren have reported the isolation of a monoligated complex,
with a very bulky phosphine, [(2-(di-tert-butylphosphino)biphenyl)
Au+(NCMe)][SbF62], as an active catalyst for cycloisomeriza-
tion,19 attempts to synthesize or isolate ItBuAuBF4 by Baker
et al.20 and PPh3AuPF6 by Gagosz et al.21 have so far failed, due
to the rapid decomposition of these complexes into colloidal
gold(0). In this communication, we report the isolation and
characterization of such complexes by using a NHC ligand of
sufficient bulk and a weakly coordinating solvent, such as
acetonitrile or tetrahydrofuran (THF), leading to relatively stable
yet reactive cationic gold(I) complexes.
and found to be stabilized by the use of N-heterocyclic carbene
ligands. These species are often employed as in situ-generated
reactive intermediates in gold catalyzed organic transforma-
tions. An isolated, well-defined species was tested in gold-
mediated carbene transfer reactions from ethyl diazoacetate.
As part of an ongoing program aimed at examining the role of
N-heterocyclic carbenes (NHC) in transition metal-mediated
reactions, we have recently studied the stabilizing effects of
NHCs surrounding unsaturated and ‘‘reactive’’ metal centers.
Since the isolation of the first free stable NHC, bearing two
sterically demanding adamantyl groups on the nitrogens of an
imidazolyl framework, by Arduengo et al.,1 sterically encumbering
NHCs have allowed the isolation of unusual three-coordinate
(NHC)Ni(CO)2 complexes,2 highly unsaturated 14 electron Ir(I)
species,3 a number of orthometalated ruthenium4 and iridium5
species, well-defined monomeric copper(I) species6 and formally 16
electron second generation ruthenium-based olefin metathesis
catalysts.7 In view of the steric and electronic properties of this
ligand class, NHCs have been employed to prepare efficient and
robust catalysts for transformations such as palladium-catalyzed
cross-coupling reactions,8 platinum-mediated hydrosilylation,9
palladium telomerization of butadiene and methanol,10 copper-
catalyzed hydrosilylation11 and ruthenium-based olefin meta-
thesis,12 to name a few.
The previously reported IPrAuCl13 (IPr = 1,3-bis(di-iso-
propylphenyl)imidazol-2-ylidene), IMesAuCl13 (IMes
= 1,3-
bis(2,4,6-trimethylphenyl)imidazol-2-ylidene) and ItBuAuCl20
(ItBu = 1,3-di-tert-butylimidazol-2-ylidene) were dissolved in
acetonitrile, and AgPF6 or AgBF4 was added in stoichiometric
amounts, leading to the rapid formation of a precipitate (AgCl).
After stirring the solutions for one minute, the suspensions were
filtered through Celite to give the novel complexes in solution
(Scheme 1).22 The appearance of colloidal gold(0) was noticeable
after a few hours for all solutions. Attempts to obtain solid
materials for all complexes by simply removing the solvent under
vacuum only led to rapid decomposition of the materials, obvious
as the white material turns to a greyish powder. Carrying out these
We recently became involved in the synthesis and isolation of
well-defined NHC–gold(I) complexes.13 The first NHC–gold(I)
complexes were reported in 1989,14 and these usually bore two
strongly bound ligands arranged in a linear fashion around a gold
cation. These can be neutral or cationic and have either
[(NHC)AuX]13 or [(NHC)2Au+][X2]15 composition. Until recently,
catalytic organogold chemistry appeared to have been somewhat
forgotten. The ‘‘noble’’ character of the metal was possibly the
origin of the misconception that it would perform poorly in
catalysis. This misconception has now been shattered, as numerous
aDepartment of Chemistry, University of New Orleans, New Orleans,
Louisiana70148,USA.E-mail:snolan@uno.edu;Fax:(+1)5042806860;
Tel: (+1) 504 280 6445
bLaboratorio de Cata´lisis Homoge´nea, Departamento de Qu´ımica y
Ciencia de los Materiales, Unidad Asociada al CSIC, Universidad de
Huelva, Campus de El Carmen, Huelva 21007, Spain.
E-mail: perez@dqcm.uhu.es; Fax: (+34) 959 219942;
Tel: (+34) 959 219956
{ Electronic Supplementary Information (ESI) available: Detailed experi-
mental details with full NMR characterization data and catalytic
procedures.
Scheme 1 Synthesis route to cationic NHC–Au(I) complexes.
Chem. Commun., 2006, 2045–2047 | 2045
This journal is ß The Royal Society of Chemistry 2006

View Online
reactions in the presence or absence of oxygen neither improved
nor decreased the stability of the complexes.
1
Under these synthetic conditions, H and 13C NMR spectra of
both the neutral precursors and novel products were recorded.
While the pattern of the 1H NMR spectra remained the same, with
no trace of decomposition products between neutral and expected
complexes, we observed a slight downfield shift for the backbone
protons of the imidazole fragment. We attribute this change to a
loss of electron density in the aromatic heterocyclic system, due a
delocalization of the p-electrons towards the more acidic gold
centre. Moreover, the 13C NMR spectra present signals for the
carbenic carbons that are significantly shifted upfield for all
complexes. Once again, this observation confirms a more acidic
gold center.20 Both 1H and 13C NMR studies support the presence
of an electron deficient gold centre, confirming the very likely
presence of monoligated gold(I) complexes.
We were interested in further testing the stability of such
complexes. Attempts to synthesize the complexes in dichloro-
methane (DCM) or chloroform lead to the rapid appearance of
Fig. 1 Ball-and-stick representations of [IPrAu+(NCMe)][PF6
] (1).
2
Hydrogens are omitted for clarity. Selected bond lengths (s) and angle
(u): C1–Au = 1.952(2), N1–C1 = 1.38(2), N2–C1 = 1.34(2), Au–N3 =
2.022(2) s; C1–Au–N3 = 177.9(8).
1
large amounts of colloidal gold. The H NMR spectra of these
reactions indicate the existence of two different NHC environ-
ments, attributed to at least two NHC–gold species in solution. We
confirmed in this manner the necessity of a coordinating solvent to
stabilize the cationic gold centre. Ample precedent exists for
coordinating solvent stabilization of Pt and Pd complexes.23,24
THF was employed as an alternative to acetonitrile to generate
IPrAu(S)PF6 (S = coordinating solvent). In THF, no decomposi-
tion was observed, even after 24 hours in solution in air, but
surprisingly a gel was obtained, attributed to the ring opening
polymerization of THF.25 The 13C NMR spectrum indicated a
gold species in THF that is even more acidic than in acetonitrile,
with a carbenic carbon appearing at a more upfield position (159.7
vs. 167.6 ppm in CD3CN). The complex in THF also displayed a
second, more downfield signal of low intensity (after a few hours)
for the deuterated THF, which confirmed the formation of poly-
THF.25 It is worthy to note that the cationic polymerization of
THF by ring opening in the presence of a Lewis acid, such as
FeCl3 or the trityl cation, is well known.26 In the present case, this
polymerization behaviour confirms the presence of a cationic gold-
centered complex that is capable of acting as a Lewis acid, an
interesting reaction profile which we are presently examining. We
also noticed that adding IPrAu(S)PF6, synthesized from acetoni-
trile, into THF led to THF polymerization. While the acetonitrile
needs to be displaced by the THF to initiate its polymerization,
this result shows that, in solution, the molecules of solvent are
weakly bound, labile and can easily be displaced from the gold.
To unambiguously establish the solid state structure of one of
these complexes, X-ray quality crystals suitable for single crystal
diffraction studies were grown from a saturated acetonitrile
solution of IPrAu(NCMe)PF6 (1).27 While the complex slowly
decomposes over several hours in solution with appearance of
colloidal gold(0), suitable X-ray quality crystals could be grown in
this manner. It is noteworthy that the appearance of decomposi-
tion in solution or the solid state for these complexes is related to
the nature of the carbene and counterion used. Complex 1 is the
most stable complex observed so far.
NHC–Au–NCCH3 arrangement, with a C–Au–N angle of
177.9(78)u. The C–Au bond distance (1.952(2) s) is similar to
that found for IPrAuCl.13 The N–Au bond distance of 2.022(2) s
is in the range of reported gold complexes with nitrogen donor
ligands,28 but slightly longer than known gold(I) complexes with
coordinated acetonitrile such as [Au+(NCMe)2][X2],29 with an
Au–N bond distance equal to 1.96 s. Finally, the NMC bond
distances between coordinated and non-coordinated molecules of
acetonitrile remain the same, with a value of 1.12(3) s,30 explaining
the observed lability of the bound acetonitrile molecule.
Previous work from our laboratories has shown a very
interesting catalytic behaviour of IPrAuCl in the presence of
NaBAr4 (BAr4 = tetrakis(3,5-bis(trifluoromethyl)phenyl)borate)
for the decomposition of ethyl diazoacetate (N2LCHCO2Et, EDA)
and subsequent transfer of the :CHCO2Et unit to organic
substrates.17 This procedure has led to the functionalization of
aromatic sp2 and primary sp3 C–H bonds of alkanes in moderate
to high yields, in a process that requires the assistance of a halide
scavenger. Attempts to fully characterize a well-defined complex
with a BAr4 counterion has proven so far unsuccessful. Therefore,
the availability of 1, in solution or as an isolated solid, now allows
for the study of its catalytic properties in this transformation. In
reactions with catalytic amounts of 1, ethyl diazoacetate was
reacted with several substrates (Table 1). In the case of good
Table 1 Reaction of ethyl diazoacetate and several substrates in the
presence of IPrAu(NCMe)PF6 (1) as catalyst
Substrate
Product
Time/h Yield (%)a
Methanol
Ethanol
Aniline
CH3OCH2CO2Et
CH3CH2OCH2CO2Et
PhN(H)CH2CO2Et
t-BuN(H)CH2CO2Et
Cyclopropanes
0.2
0.2
24
24
20
.99
.99
.99
55
.99
—
tert-Butylamine
Styrene
Benzene
No reactionb
120
120
2,3-dimethylbutane
No reactionb
—
Results from the diffraction study confirm the NMR deter-
mined structure and the coordination of one acetonitrile to the
gold centre. (Fig. 1). The metrical parameters reveal a nearly linear
a
Determined by GC, diethyl fumarate and maleate accounted for
the remaining of the mass balance. EDA not consumed.
b
2046 | Chem. Commun., 2006, 2045–2047
This journal is ß The Royal Society of Chemistry 2006

View Online
B. Tinant, J.-P. Declercq and E. I. Marko, Chem. Commun., 2005, 30,
3856–3858.
10 R. Jackstell, S. Harkal, H. Jiao, A. Spannenberg, C. Borgmann,
D. Roettger, F. Nierlich, M. Elliot, S. Niven, K. Cavell, O. Navarro,
M. S. Viciu, S. P. Nolan and M. Beller, Chem.–Eur. J., 2004, 10,
3891–3900.
11 S. Diez-Gonza´lez, H. Kaun, F. K. Zinn, E. D. Stevens and S. P. Nolan,
J. Org. Chem., 2005, 70, 4784–4796.
12 L. Jafarpour, H.-J. Schanz, E. D. Stevens and S. P. Nolan,
Organometallics, 1999, 18, 5416–5425; A. Fu¨rstner, O. R. Thiel,
L. Ackermann, H.-J. Schanz and S. P. Nolan, J. Org. Chem., 2000,
65, 2204–2207.
13 P. de Fre´mont, N. M. Scott, E. D. Stevens and S. P. Nolan,
Organometallics, 2005, 24, 2411–2418.
14 F. Bonati, A. Burini, B. R. Pietroni and B. J. Bovio, J. Organomet.
Chem., 1989, 375, 147–160.
15 H. M. K. Wang and I. J. B. Lin, Organometallics, 1998, 17, 972–975;
P. J. Barnard, M. V. Baker, S. J. Berners-Price, B. W. Skelton and
A. H. White, Dalton Trans., 2004, 1038–1047.
16 For recent examples, see: L. Zhang and S. A. Kozmin, J. Am. Chem.
Soc., 2005, 127, 6962–6963; M. J. Johansson, D. J. Gorin, S. T. Staben
and F. D. Toste, J. Am. Chem. Soc., 2005, 127, 18002–18003.
17 M. R. Fructos, T. R. Belderrain, P. de Fre´mont, N. M. Scott,
S. P. Nolan, M. M. D´ıaz-Requejo and P. J. Pe´rez, Angew. Chem., Int.
Ed., 2005, 44, 5284–5288; N. Marion, S. Diez-Gonza´lez, P. de Fre´mont,
A. R. Noble and S. P. Nolan, Angew. Chem., Int. Ed., 2006, 45, in press.
18 J. Zhang, C.-G. Yang and C. He, J. Am. Chem. Soc., 2006, 128,
1798–1799.
nucleophiles such as alcohols, quantitative conversion was
obtained within minutes. Longer times were required for aniline,
whereas for tert-butylamine, incomplete conversion was observed
even after 24 hours. A similar result was found with styrene, for
which three days were required for complete conversion into
cyclopropanes. This trend could be attributed to the ease of
replacement of the coordinated acetonitrile in 1 by the substrate. In
accordance with this, the use of more weakly coordinating
molecules such benzene or 2,3-dimethylbenzene has led to
undetectable yields; only very minor amounts of diethyl fumarate
and maleate were detected by GC after several days, with most of
the initial EDA remaining in solution. However, an additional
experiment strongly suggests that this coordination of the substrate
is not the only factor at play. The use of an equimolar mixture of
IPrAuCl and NaBAr4 as the catalyst in the reaction of styrene and
EDA, with a five-fold excess of added acetonitrile, did not affect
the course of the reaction, and led to the same mixture of products
found in the absence of acetonitrile. We strongly suspect at this
point that the counterion plays an important role in this catalytic
transformation, a feature that is currently under investigation.
In conclusion, we have isolated, characterized by NMR
spectroscopy, and for one example by X-ray diffraction, well-
defined cationic (NHC)Au(I)(S)X complexes which are postulated
as active catalysts in numerous gold-mediated organic transforma-
tions. The well-defined, isolated species IPrAu(NCMe)PF6 (1) has
been tested as catalyst for the carbene transfer reaction from EDA.
The results suggest a large effect of the counterion on this
transformation when compared with the already reported in situ-
generated IPrAuCl + NaBAr4 system.17 Studies aimed at exploring
this relative stability issue, as well as investigations focusing on the
reactivity of NHC–Au complexes in organic chemistry, are
presently ongoing in our laboratories.
19 C. Ferrer and A. M. Echavarren, Angew. Chem., Int. Ed., 2006, 45,
1105–1109.
20 M. V. Baker, P. J. Barnard, S. K. Brayshaw, J. L. Hickey, B. W. Skelton
and A. H. White, Dalton Trans., 2005, 37–43. The existence of a cationic
species such as [NHCAu(NCMe]Cl cannot be excluded at this point,
and counterion and solvent effects are presently being examined.
21 N. Me´zaille, L. Ricard and F. Gagosz, Org. Lett., 2005, 7, 4133–4136.
22 In a scintillation vial in air, (NHC)AuCl (1 equiv.) is dissolved in 2 mL
of acetonitrile and AgX (1 equiv.) added. The reaction is stirred for one
minute and filtered over Celite to give a colorless solution. Complete
removal of the solvent under vacuum leads to the isolation of a white
powder that rapidly decomposes (turning grey or purple). Slow solvent
evaporation leads to formation of crystals of the desired product,
[(IPrAu)+(NCMe)][PF62] (1): 1H NMR (400 MHz, CD3CN, ppm) 7.66
(s, 2H, CH-imidazole), 7.63 (t, J = 8.0 Hz, 2H, CH-aromatic), 7.45 (d,
J = 8.0 Hz, 2H, CH-aromatic), 2.51 (sept, J = 6.8 Hz, 4H, CH(CH3)2),
1.31 (d, J = 6.8 Hz, 12H, CH(CH3)2) and 1.26 (d, J = 6.8 Hz, 12H,
CH(CH3)2);13C NMR (100 MHz, CDCl3, ppm) 167.6 (s, C-carbene),
148.1 (s, CH-aromatic), 135.4 (s, CH-aromatic), 132.3 (s, CH-aromatic),
127.4 (s, CH-aromatic), 126.6 (s, CH-imidazole), 28.9 (s, CH(CH3)2),
23.9 (s, CH(CH3)2) and 23.3 (s, CH (CH3)2).
The National Science Foundation is gratefully acknowledged
for the financial support of this work, as are Umicore AG, Eli Lilly
and Boehringer Ingelheim Pharmaceuticals for materials support
and unrestricted grants. We wish to thank the University of
Ottawa and its Chemistry Department for hosting our group
during our time away from the University of New Orleans.
23 R. Bertani, R. A. Michelin, M. Mozzon, A. Sassi, M. Basato, A. Biffis,
G. Martinati and Z. Zecca, Inorg. Chem. Commun., 2001, 4, 281–284;
R. Bertani, M. Biasioli, K. Darini, R. A. Michelin, M. Mozzon,
F. Visentin and L. Zanotto, J. Organomet. Chem., 2002, 642, 32–39.
24 M. S. Viciu, F. Kauer-Zinn, E. D. Stevens and S. P. Nolan,
Organometallics, 2003, 22, 3175–3177.
Notes and references
1 A. J. Arduengo, III, R. L. Harlow and M. Kline, J. Am. Chem. Soc.,
1991, 113, 361–363.
2 R. Dorta, E. D. Stevens, C. D. Hoff and S. P. Nolan, J. Am. Chem.
Soc., 2003, 125, 10490–10491.
25 S. Kobayashi, K. Morikawa and T. Saegusa, Macromolecules, 1975, 8,
954–956.
3 N. M. Scott, R. Dorta, E. D. Stevens, A. Correa, L. Cavallo and
S. P. Nolan, J. Am. Chem. Soc., 2005, 127, 3516–3526.
4 R. F. R. Jazzar, M. Varrone, A. D. Burrows, S. A. McGregor,
M. F. Mahon and M. K. Whittlesey, Inorg. Chim. Acta, 2006, 359,
815–820.
5 R. Dorta, E. D. Stevens and S. P. Nolan, J. Am. Chem. Soc., 2004, 126,
5054–5055.
6 H. Kaur, F. Kauer-Zinn, E. D. Stevens and S. P. Nolan,
Organometallics, 2004, 23, 1157–1160.
7 J. Huang, E. D. Stevens and S. P. Nolan, J. Am. Chem. Soc., 1999, 121,
2674–2678.
8 M. S. Viciu, R. F. Germaneau, O. Navarro-Ferna´ndez, E. D. Stevens
and S. P. Nolan, Organometallics, 2002, 21, 5470–5472; O. Navarro,
N. Marion, N. M. Scott, J. Gonza´lez, D. Amoroso, A. Bell and
S. P. Nolan, Tetrahedron, 2005, 61, 9716–9722.
9 J. W. Sprengers, M. J. Maayke, M. A. Duin, J. C. Kingsley and
J. C. Elsevier, J. Organomet. Chem., 2003, 679, 149–152; O. Busine,
G. Berthon-Gelloz, J.-F. Briere, S. Sterin, G. Mignani, P. Branlard,
26 I. Kuntz, J. Polym. Sci., Part B: Polym. Lett., 1966, 4, 427–430;
V. M. Siderko and V. P. Mardykin, Vestn. Beloruss. Gos. Univ. Ser. 2,
1970, 2, 35–37.
27 Crystal data for 1: C31H42AuN4PF6, M = 812.62, monoclinic, space
group P21/n, a = 8.9662(6), b = 18.347(1), c = 21.693(2) s, b =
96.841(1)u, V = 3544.0(4) s3, T = 298(2) K, Z = 4, m = 4.254 mm21
,
53909 reflections measured using a Bruker SMART 1K CCD
diffractometer, 4626 unique (Rint = 0.062), R1 = 0.0493 for all data,
wR2 (all data) = 0.1101. CCDC 296436. For crystallographic data in
CIF or other electronic format see DOI: 10.1039/b601547f.
28 W. Conzelmann, W. Hiller and J. Stra¨hle, Z. Anorg. Allg. Chem., 1982,
485, 81–87; P. W. R. Corfield and H. M. M. Shearer, Acta Crystallogr.,
1967, 23, 156–162.
29 D. M. P. Mingos and J. Yau, J. Organomet. Chem., 1994, 479,
C16–C17; S. Ahrland, K. Nilsson, I. Person, A. Yushi and J. E. Penner-
Hahn, Inorg. Chem., 1989, 28, 1833–1838.
30 Two molecules of acetonitrile are present in the asymmetric unit cell.
One is bound to the gold, the other is a solvent of co-crystalization.
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 2045–2047 | 2047