Angewandte
Chemie
DOI: 10.1002/anie.201206468
Gold(III) Hydrides
A Thermally Stable Gold(III) Hydride: Synthesis, Reactivity, and
Reductive Condensation as a Route to Gold(II) Complexes**
Dragos¸-Adrian Ros¸ca, Dan A. Smith, David L. Hughes, and Manfred Bochmann*
Transition-metal hydride complexes are well-established as
key intermediates in numerous homogeneously and hetero-
geneously catalyzed reactions.[1] Hydride complexes of plat-
inum(II), for example, have been known for over 50 years.[2]
Analogous complexes of the isoelectronic gold(III), on the
other hand, appear to be unknown.[3] This is all the more
surprising as gold complexes have attracted much attention as
catalysts and catalyst precursors in recent years.[4] For
example, heterogeneous gold catalysts are used for selective
hydrogenations;[5] for zirconia-supported gold hydrogenation
catalysts, surface Au3+ ions have been proposed as active
sites.[6] Both gold(I) and gold(III) hydrides have been
postulated as intermediates in numerous homogeneously
catalyzed hydrogenations,[7,8] hydrosilylations,[9] dehydrogen-
ative alcohol silylations,[10] hydroborations,[11] and other
organic transformations.[12] Corma and co-workers used
immobilized gold(III) chelate complexes as hydrogenation
catalysts and explored the possible involvement of AuIII
hydride intermediates in detail by kinetic and computational
methods.[13] On the other hand, the highly positive standard
redox potentials for gold (Au3+/Au+ = 1.36 V; Au3+/Au0 =
1.52 V)[14] might suggest facile reduction of AuIII in the
presence of hydride. Binary hydrides, including AuH,
(H2)AuH, (H2)AuH3, and [AuH4]À, have been detected in
frozen gas matrices below 5 K.[15] We report herein the
synthesis and reactions of a thermally stable gold(III) hydride,
[(C^N^C)*AuH], wherein (C^N^C)* is a doubly cyclometa-
lated 2,6-bis(4’-tert-butylphenyl)pyridine ligand.
sponding gold deuteride complex 2-D by treating 1 with
LiDBEt3. The H NMR spectrum of 2-D shows the resonan-
1
ces for the C^N^C* ligand that are identical to those of 2-H,
without the hydride resonance, and the phenyl b proton shows
the expected doublet multiplicity (see the Supporting Infor-
mation). The 2H NMR spectrum of 2-D in CH2Cl2 shows
a singlet at d = À6.58.
The IR spectrum of 2-H shows a strong, sharp band at
À1
À
2188 cm , which is higher than the Au H stretch in
[(IPr)AuH] (1976 cmÀ1)[3a] but comparable to that found for
AuH and (H2)AuH in a frozen argon matrix (2226.6 and
2173.6 cmÀ1, respectively).[15] The Au D stretching frequency
À
of 2-D (expected at ca. 1550 cmÀ1) was not found and is likely
to be obscured by other ligand bands in this region.
As a solid, 2-H is stable to air and moisture at room
temperature. Although the hydride is cis to two phenyl
À
ligands, and although we do observe Au C cleavage reactions
of this pincer system under acidic conditions,[17] the rigidity of
the C^N^C pincer implies that reductive elimination is not
favorable. On the other hand, toluene and CH2Cl2 solutions of
2-H darkened rapidly when exposed to light, even though
little change was noticeable in the 1H NMR spectrum.
However, prolonged exposure to sunlight of CH2Cl2 solutions
of 2-H gave [(C^N^C)*AuCl] (t1/2 ꢀ 4 h), while heating 2-H in
C6D6 for 12 h generated a mixture of products consisting of
about 30% free ligand, with the loss of the hydride resonance.
The structure of 2-H was confirmed by X-ray diffraction
(Figure 1).[18]
The reaction of [(C^N^C)*AuOH][16] (1) with LiHBEt3 in
toluene at À788C followed by stirring at room temperature
for 15 min rapidly yielded [(C^N^C)*AuH] (2-H), which was
isolated as a yellow crystalline complex [Eq. (1)]. The
1H NMR spectrum showed a broad singlet resonance at d =
À6.51 in CD2Cl2 and at d = À5.73 in C6D6, while the protons
attached to the carbon atom in the b position with respect to
the gold center give rise to a pseudo triplet multiplicity,
4
suggesting a J coupling (1 Hz) to the hydride ligand. This
assignment was confirmed by the preparation of the corre-
Complex 2-H proved unreactive towards ethylene, 3-
hexyne, phenylacetylene, and even dimethyl acetylenedicar-
boxylate. There was also no reaction with CO2 or with
benzaldehyde in the dark or under photolysis. Also no
reaction was observed between 2-H and weak acids such as
acetic acid; however, 2-H reacts instantaneously with the
stronger trifluoroacetic acid (HOAcF) to give a mixture of
products, with the expected complex [(C^N^C)*AuOAcF]
not being observed among the reaction products. The results
suggest that the hydridic character of 2-H is less pronounced
than in the case of AuI hydrides stabilized by strong NHC
donor ligands.[3a] The hydride does however react with 1,1-
[*] D.-A. Ro¸sca, Dr. D. A. Smith, Dr. D. L. Hughes, Prof. M. Bochmann
Wolfson Materials and Catalysis Centre, School of Chemistry
University of East Anglia
Norwich, NR4 7TJ (UK)
E-mail: m.bochmann@uea.ac.uk
[**] This work was supported by the Leverhulme Trust. We thank
Johnson Matthey plc for a loan of gold salts and Dr. Yimin Chao
(UEA) for access to spectrophotometric facilities. D.A.R. thanks the
University of East Anglia for a studentship.
Supporting information for this article (synthetic, spectroscopic,
and crystallographic details) is available on the WWW under http://
Angew. Chem. Int. Ed. 2012, 51, 1 – 5
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!