Are β-H eliminations or alkene insertions feasible elementary steps in catalytic cycles involving gold(I) alkyl species or gold(I) hydrides?

Article abstract of DOI:10.1002/chem.201203043

The β-H-elimination in the (iPr)AuEt complex and its microscopic reverse, the insertion of ethene into (iPr)AuH, were investigated in a combined experimental and computational study. Our DFT-D3 calculations predict free-energy barriers of 49.7 and 36.4 kcal mol^-1 for the elimination and insertion process, respectively, which permit an estimation of the rate constants for these reactions according to classical transition-state theory. The elimination/insertion pathway is found to involve a high-energy ethene hydride species and is not significantly affected by continuum solvent effects. The high barriers found in the theoretical study were then confirmed experimentally by measuring decomposition temperatures for several different (iPr)Au^I-alkyl complexes which, with a slow decomposition at 180 °C, are significantly higher than those of other transition-metal alkyl complexes. In addition, at the same temperature, the decomposition of (iPr)AuPh and (iPr)AuMe, both of which cannot undergo β-H-elimination, indicates that the pathway for the observed decomposition at 180 °C is not a β-H-elimination. According to the calculations, the latter should not occur at temperatures below 200 °C. The microscopic reverse of the β-H-elimination, the insertion of ethene into the (iPr)AuH could neither be observed at pressures up to 8 bar at RT nor at 1 bar at 80 °C. The same is true for the strain-activated norbornene. Au^I reactions out of reach: A β-H elimination in an (iPr)Au-alkyl complex or an alkene insertion at an (iPr)AuH complex proceed through high-energy transition states and thus are inaccessible for gold catalysis (see figure). Copyright

Full text of DOI:10.1002/chem.201203043

DOI: 10.1002/chem.201203043
Are b-H Eliminations or Alkene Insertions Feasible Elementary Steps in
Catalytic Cycles Involving Gold(I) Alkyl Species or Gold(I) Hydrides?
Gꢀnter Klatt,[a] Rong Xu,[a] Markus Pernpointner,[a] Lise Molinari,[b]
Tran Quang Hung,[b] Frank Rominger,[b] A. Stephen K. Hashmi,[b] and Horst Kçppel*[a]
Abstract: The b-H-elimination in the
(iPr)AuEt complex and its microscopic
reverse, the insertion of ethene into
(iPr)AuH, were investigated in a com-
bined experimental and computational
study. Our DFT-D3 calculations predict
free-energy barriers of 49.7 and
36.4 kcalmolꢀ1 for the elimination and
insertion process, respectively, which
permit an estimation of the rate con-
stants for these reactions according to
classical transition-state theory. The
elimination/insertion pathway is found
to involve a high-energy ethene hy-
dride species and is not significantly af-
fected by continuum solvent effects.
The high barriers found in the theoreti-
cal study were then confirmed experi-
mentally by measuring decomposition
temperatures for several different
(iPr)AuI–alkyl complexes which, with a
slow decomposition at 1808C, are sig-
nificantly higher than those of other
transition-metal alkyl complexes. In ad-
dition, at the same temperature, the
decomposition of (iPr)AuPh and
(iPr)AuMe, both of which cannot un-
dergo b-H-elimination, indicates that
the pathway for the observed decom-
position at 1808C is not a b-H-elimina-
tion. According to the calculations, the
latter should not occur at temperatures
below 2008C. The microscopic reverse
of the b-H-elimination, the insertion of
ethene into the (iPr)AuH could neither
be observed at pressures up to 8 bar at
RT nor at 1 bar at 808C. The same is
true for the strain-activated norbor-
nene.
Keywords: alkenes · computational
chemistry
· elimination · gold ·
hydrides · insertion
Introduction
monohydride[6] and a dinuclear gold(I) hydride[7] in a spec-
tacular manner disprove a general low stability of gold hy-
dride complexes.
For a long time gold hydride complexes were assumed to
have a low stability.[1] This was supposed to be the reason
for not observing b-H-eliminations in homogeneous gold
catalysis,[2,3] whereas these reactions are commonplace in
catalysts of other transition metals. They often constitute el-
ementary steps in many catalytic processes, for example, the
isomerization of higher alkyls to metal hydrides. The inverse
reaction, ethene insertion into a metal–hydrogen bond, is
analogous to the rate-determining insertion step of transi-
tion-metal-catalyzed olefin polymerization.[4] The results ob-
tained in gold-catalyzed reactions[5] clearly indicate a lack of
b-H-elimination, but the recent isolation of a NHC-gold(I)
The reactivity of NHC-gold(I) monohydride with regard
to typical elementary steps of catalysis reactions has been
investigated;[6] normal alkynes such as 1-hexyne and diphe-
nylacetylene do not insert into the gold–hydrogen bond, but
dimethyl acetylenedicarboxylate does. Unexpectedly, for the
insertion product a crystal structure analysis showed an anti-
arrangement of the NHC-gold fragment and the hydride,
which excludes a concerted syn-insertion pathway (the mi-
croscopic reverse of a b-H-elimination). On the other hand,
the investigation neither could rule out a radical mechanism
nor a syn-addition/isomerization sequence. Previously, gold-
AHCTUNGTREG(NNNU III) monohydrides had already been suggested as inter-
mediates in homogeneous gold-catalyzed hydrogenation re-
actions, a pathway that was supported by DFT calculations[8]
and an active participation of gold(I) hydrides in the dehy-
drogenative silylation of alcohols had been proposed (the
reported hydrolysis of NHC-gold(I) monohydride would
nicely fit into that picture).[9] Another recent publication[10]
reports a gold(I) hydride species, which could be character-
[a] Dr. G. Klatt, Dr. R. Xu, Dr. M. Pernpointner, Prof. Dr. H. Kçppel
Physikalisch-Chemisches Institut
Ruprecht-Karls-Universitꢀt Heidelberg
Im Neuenheimer Feld 229
69120 Heidelberg (Germany)
Fax : (+49)6221-545-221
E-mail: Horst.Koeppel@pci.uni-heidelberg.de
1
ized by H and 31P NMR spectroscopic measurements and
[b] L. Molinari, T. Quang Hung, Dr. F. Rominger,+
Prof. Dr. A. S. K. Hashmi
ESI mass spectrometry. NMR and kinetic studies revealed
that the reaction mechanism of the dehydrogenative silyla-
tion indeed involves the gold(I) hydride species as a key in-
termediate.
Organisch-Chemisches Institut
Ruprecht-Karls-Universitꢀt Heidelberg
Im Neuenheimer Feld 270
69120 Heidelberg (Germany)
Fax : (+49)711-685-4205
[+] Crystallographic investigation
Lately, a report stressed a b-H-elimination as an elemen-
tal step in a goldACTHNGUTERNNU(G III)-catalyzed reaction forming strained
3954
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 3954 – 3961

FULL PAPER
oxetenes.[11] The claim is supported by calculations at the
PCM(CH2Cl2)-B3LYP/def2-SVP level.
cently reported gold hydride complex (iPr)AuH[6] as well as
free ethene. The starting points for our work were the X-ray
crystal structures of the (iPr)AuEt (which was determined in
the course of this investigation)[25] and (iPr)AuH complexes
(which had been determined previously).[6] We subjected
these structures to DFT geometry optimization and arrived
at structures 1 and 2, which are given in Figure 1 together
with the atom numbering adopted in the following discus-
sion, and for which selected structural parameters are dis-
played in Table 1.
ACHTUNGTRENNUNG
Since this elemental step could be of significant impor-
tance for developing new catalytic cycles in homogeneous
gold catalysis, we now decided to conduct a detailed investi-
gation of the b-H-elimination (respectively its microscopic
reverse, the alkene insertion pathway which is of importance
for gold-catalyzed hydrogenations) for gold(I) as the most
frequently used oxidation state in homogeneous gold cataly-
sis.
The calculated bond lengths generally agree well (within
0.02 ꢃ) with experiment, except for bonds involving either
hydrogen atoms (which are difficult to pinpoint by X-ray
crystallography) or the Au center, which is probably not as
well described as the lighter atoms by the basis sets and
Computational Details
We performed initial geometry optimizations at the B3LYP/cc-pVDZ
level of theory for all calculated species using the Gaussian 09 software
package.[12] In this basis, the gold atom is described by the (8 s7p6d1f)/
AHCTUNG-
ꢀ
methods used in this work. Notable exceptions are the C C
bond length in the ethyl group of 1, which deviates by
TRENNUNG[4s4p3d1f] pseudopotential-based correlation consistent basis set of Pe-
terson and Puzzarini,[13] and the light atoms by Dunningꢂs cc-pVDZ basis
functions.[14] For gold a relativistic effective core potential (RECP) was
employed,[15] which replaced the innermost 60 electrons. This RECP was
designed to accurately account for relativistic effects, such as s orbital
contraction and d-orbital expansion, and has proven to be very successful
for calculations of organometallic compounds.[16] All stationary points in-
volved were fully optimized. Frequency calculations were undertaken to
confirm the nature of the stationary points, yielding one imaginary fre-
quency for transition states (TS) and zero for minima. Transition states
were obtained by either a combined scanning and transit-guided quasi-
Newton (STQN) method, or a combination of the two STQN approaches
LST[17] and QST3,[18] as implemented in Gaussian 09. The connectivity of
the transition states and their adjacent minima was confirmed by intrinsic
reaction coordinate (IRC) calculations. Zero-point energy (ZPE) correc-
tions were carried out for all computed energies. Free energies G298 were
calculated by using the harmonic approximation and standard textbook
procedures. To obtain estimates on the solvent effect, we made use of the
polarizable continuum model (PCM) method with the dielectric constant
for dichloromethane to generate single-point energies at the B3LYP/cc-
pVDZ-optimized geometries. These were then used in the calculation of
ZPE-corrected energies E0 and free energies G298 based on the B3LYP/
cc-pVDZ harmonic frequencies.
ꢀ
0.05 ꢃ (DFT: 1.54 ꢃ, X-ray: 1.49 ꢃ), and the C C bond
length in the NHC subunit, which deviates by 0.04 ꢃ (DFT:
1.36 ꢃ, X-ray: 1.32 ꢃ).
Likewise, the angles obtained with our DFT methods also
agree well (within 3–48) with the experimental result,
except for the larger deviations observed for several dihe-
drals, which is a likely consequence of their larger flexibility,
Very recently, an implementation of Grimmeꢂs D3 dispersion parame-
ters[19] into the Turbomole 6.4 program[20] has paved the way for calcula-
tions on large organometallic systems that include dispersive interactions;
see for example reference [21]. To account for these forces, which are
poorly described by standard DFT functionals, as well as to compare two
different methods, we re-optimized all stationary points at the BP86-D3
level of theory, which contains Grimmeꢂs D3 dispersion parameters.[19]
The double z-quality basis set def2-SVP[22] as well as the multipole-accel-
erated resolution of the identity approximation MARI-J with suitable fit-
ting basis sets[23] were used in these computations. Scalar relativistic ef-
fects were included by using a Stuttgart ECP for gold.[24]
For the complexes [(iPr)AuEt] and [(iPr)AuH], we additionally per-
formed MP2 single-point energy calculations at the B3LYP/cc-pVDZ-op-
timized geometries, that is, for the MP2/cc-pVDZ//B3LYP/cc-pVDZ
model chemistry. ZPE-corrections and free energies were obtained in the
same manner as for the PCM calculations.
Results and Discussion
Intermediates and transition states of the b-elimination
pathway: Taking into account the importance of the iPr
ligand for the stabilization of gold(I) hydride, we decided to
investigate the thermal decay of (iPr)AuEt to form the re-
Figure 1. Stationary points of the thermal decay of (iPr)AuEt. Com-
pounds 1, 1a, 2a, and 2 correspond to the ethyl complex, ethene hydride,
van der Waals adduct and gold hydride plus free ethene, respectively.
TS1 and TS2 represent the interconnecting transition states.
Chem. Eur. J. 2013, 19, 3954 – 3961
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3955

H. Kçppel et al.
Table 1. Selected calculated and experimental bond lengths [ꢃ] and angles [8] for stationary points of the ther-
mal decay of (iPr)AuEt.[a]
latter is 1.428 ꢃ and thus sub-
stantially longer than the
1.333 ꢃ calculated for free
C2H4, which indicates the don-
ation of electron density from
the Au center into the anti-
bonding p* orbital.
1
TS1
1a
TS2
2a
2
Au-C9
Exp
2.133
2.086
2.085
3.030
3.055
3.094
0.980
1.112
1.104
3.128
3.245
3.296
1.999
2.064
2.080
1.373
1.370
1.369
1.489
1.542
1.540
1.320
1.372
1.359
1.520
1.541
1.540
1
1
1
1
1
1
1
1
1
BP86-D3/def2-SVP
B3LYP/cc-pVDZ
Exp
BP86-D3/def2-SVP
B3LYP/cc-pVDZ
Exp
BP86-D3/def2-SVP
B3LYP/cc-pVDZ
Exp
BP86-D3/def2-SVP
B3LYP/cc-pVDZ
exp
BP86-D3/def2-SVP
B3LYP/cc-pVDZ
exp
BP86-D3/def2-SVP
B3LYP/cc-pVDZ
exp
BP86-D3/def2-SVP
B3LYP/cc-pVDZ
exp
BP86-D3/def2-SVP
B3LYP/cc-pVDZ
exp
BP86-D3/def2-SVP
B3LYP/cc-pVDZ
exp
BP86-D3/def2-SVP
B3LYP/cc-pVDZ
exp
BP86-D3/def2-SVP
B3LYP/cc-pVDZ
exp
BP86-D3/def2-SVP
B3LYP/cc-pVDZ
2.289
2.285
2.143
2.136
2.348
2.364
4.025
4.538
Au-C12
2.280
2.270
2.209
2.208
2.462
2.490
4.819
5.346
TS1 corresponds to the
H8-C9
ꢀ
breaking of the C H bond in
1.613
1.593
3.133
3.053
2.900
2.868
3.626
3.748
the methyl group and exhibits
more structural similarity with
1a than with the preceding
alkyl structure 1; in particular,
Au-H8
1.750
1.778
1.779
1.674
1.676
1.658
1.662
1.622
1.619
1.621
1.618
1.999
2.078
2.091
1.357
1.370
1.368
Au-C2
ꢀ
the C H distance has in-
1.965
1.981
2.125
2.116
2.079
2.096
2.078
2.090
creased to 1.593 ꢃ, indicating
an already broken bond,
whereas it remains below 1.3 ꢃ
in agostic transition-metal
alkyls (see Table 2 in ref. [26]).
C2-N3
1.382
1.379
1.372
1.370
1.375
1.374
1.369
1.368
C9-C12
1.453
1.448
1.439
1.428
1.390
1.376
1.342
1.334
ꢀ
The C C bond has shortened
to 1.448 ꢃ, and the AuH bond
length is 1.779 ꢃ (B3LYP/cc-
pVDZ).
C4-C6
1.320
1.371
1.359
1.525
1.544
1.540
1.369
1.355
1.371
1.355
1.370
1.357
1.371
1.359
C21-C48
Au-C2-N3
C10-C18-C21-C48
C2-N3-C10-C18
We found 1a to be connect-
ed to 2a via the transition state
TS2, as confirmed by IRC cal-
culations. TS2 is characterized
by a C2-Au-H bond angle of
1.542
1.539
1.543
1.538
1.538
1.538
1.541
1.541
127.6
130.3
127.5
128.1
150.0
119.5
114.0
ꢀ72.9
ꢀ80.6
ꢀ90.2
128.1
128.0
127.5
127.6
126.7
128.4
128.1
128.0
128.1
128.0
151.3
90.6
115.6
ꢀ88.0
ꢀ96.5
ꢀ90.6
130.58 and
a
distance of
110.1
120.8
86.8
128.2
151.3
133.6
110.6
109.6
2.364 ꢃ between Au and the
nearest C atom of the ethene
unit (B3LYP/cc-pVDZ).
Comparing the B3LYP re-
sults to the parameters ob-
tained with the BP86-D3
method, we note in particular
the considerable decrease in
distance between the (iPr)AuH
ꢀ89.6
ꢀ90.1
ꢀ98.2
ꢀ90.7
ꢀ88.3
ꢀ95.4
ꢀ84.0
ꢀ90.9
[a] Compounds 1, 1a, 2a, and 2 correspond to the ethyl complex, ethene hydride, van der Waals adduct, and
gold hydride plus free ethene, respectively. TS1 and TS2 represent the interconnecting transition states. Exper-
imental data are available for structures 1 and 2 only. The atom numbering is given in Figure 1.
resulting in considerable crystal-packing effects in the solid
state.
and ethene fragments in the van der Waals complex 2a
when dispersion is included. For example, in 2a the Au···C9
distance shortens by over 0.5 ꢃ from 4.538 to 4.028 ꢃ, and
the Au···C12 distance from 5.364 (B3LYP) to 4.819 ꢃ
(BP86-D3). This demonstrates the importance of accounting
for dispersion when describing such weakly bonded species.
We also notice large discrepancies between some dihedral
angles obtained with these methods. For example, for the
C10-C18-C21-C48 angle in the gold hydride 2 we obtain a
value of 115.68 with B3LYP, and of 90.68 by using BP86-D3,
both significantly different from the experimental value of
151.38. On the other hand, the C2-N3-C10-C18 dihedral in
the gold ethyl 1 is calculated as ꢀ90.28 (B3LYP) and ꢀ80.68
(BP86-D3), compared with the experimental value of
ꢀ72.98. Whereas the inclusion of dispersive interactions
clearly has a marked influence on the location of the shal-
low minima of these torsional potentials, it is also evident
that this does not always lead to improvement with regard
to experiment, presumably because of the complex interplay
To determine a transition state for this reaction by using
the LST-QST3 method, we reoptimized 2 after introduction
of a free ethene molecule and obtained the van der Waals
complex 2a. It consists of the subunits (iPr)AuH and C2H4,
for which the shortest distance is 2.660 ꢃ, measured be-
tween the hydride and an ethene hydrogen atom (B3LYP/
cc-pVDZ).
Our initial attempt to determine a transition state for the
reaction 1–2a by using the LST-QST3 method yielded a
transition structure TS1, in which we found 1 to be connect-
ed to an ethylene-hydride isomer [(iPr)AuACTHNUGTRNEUNG(C2H4)H] 1a, in
which both an ethene and a hydride ligand are connected to
the metal center. The most striking feature of 1a is an
almost orthogonal arrangement of the NHC-Au-H structure
ꢀ
with qACHTUNGTRENNUNG(C2-Au-H)=96.88 (B3LYP/cc-pVDZ). The Au C
bond lengths for the carbon atoms of the ethene ligand are
ꢀ
2.135 ꢃ and 2.209 ꢃ, respectively. The C C distance in the
3956
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 3954 – 3961

Feasible Elementary Steps in Catalytic Cycles
FULL PAPER
of intra- and intermolecular nonbonded interactions that are
responsible for their values in the solid state.
The overall reaction scheme for the b-H-elimination in
the (iPr)AuEt complex according to our calculations is
given in Scheme 1, and the associated B3LYP energy profile
Scheme 1. Reaction pathway of the b-H-elimination in (iPr)AuEt.
in Figure 2. All electronic energies E, ZPE-corrected ener-
gies, and free energies G298, as well as energies Esol and Gsol
calculated with the PCM method with the dielectric constant
for the solvent dichloromethane, as well as relative quanti-
ties, are listed in Table 2.
Contrary to the situation in alkyl complexes of other tran-
sition metals, such as Co,[26,27] Rh,[26,27] or Cr,[28] no isomer
displaying an agostic interaction between an alkyl hydrogen
and the metal center was found along the reaction path.
This is likely a result of goldꢂs filled d-shell in the (iPr)AuEt
complex, which prevents the formation of a strong 3c–2e
Figure 2. Energy profiles for the thermal decay of (iPr)AuEt calculated
on the DFTACTHUNGTRNE(UGN B3LYP) level of theory. E, E0, and G298 refer to absolute
electronic energies, ZPE-corrected electronic energies, and free energies
at 298.15 K, respectively. Results including the effect of the solvent
CH2Cl2 according to the PCM method as well as relative MP2 energies
for (iPr)AuEt and (iPr)AuH are also displayed. Compound 1 was select-
ed as the isomer of zero energy. Compounds 1, 1a, 2a, and 2 correspond
to the ethyl complex, ethene hydride, van der Waals adduct and gold hy-
dride plus free ethene, respectively. TS1 and TS2 represent the intercon-
necting transition states.
bond involving the metal center and the electron pair of the
Another consequence of the much weaker bonding in 1a
compared with usual transition-metal p-complexes is its
easy decay into (iPr)AuH and free ethene; according to our
calculations this process is exergonic with DG298 =ꢀ27.6 kcal
molꢀ1 and a free energy barrier of only 2.5 kcalmolꢀ1. The
dissociation of ethene-hydride complexes of Cr, on the
[29]
ꢀ
C H bond that defines an agostic interaction. Shell-filling
also prevents the formation of a p-complex by the donation
of olefin p-electrons into empty metal d orbitals, which re-
sults in a comparatively high TS1 barrier between the alkyl
structure 1 and the ethene-hydride 1a, as well as a high rela-
tive energy of 1a compared
with 1. At the B3LYP/cc-
pVDZ level of theory the TS1
electronic energy is 57.1 kcal
molꢀ1, and the ethene-hydride
Table 2. Absolute electronic energies E, ZPE-corrected electronic energies E0, free energies G298 , PCM elec-
tronic energies Esol and PCM free energies Gsol [Hartree], as well as relative quantities DX [kcalmolꢀ1] of sta-
tionary points for the thermal decay of (iPr)AuEt.[a]
1
TS1
1a
TS2
2a
2
complex 1a has
a relative
BP86-D3/def2-SVP
B3LYP/cc-pVDZ
E
E0
ꢀ1374.4755 ꢀ1374.3901 ꢀ1374.4216 ꢀ1374.4174 ꢀ1374.4390 ꢀ1374.4317
ꢀ1373.8580 ꢀ1373.7788 ꢀ1373.8080 ꢀ1373.8051 ꢀ1373.8263 ꢀ1373.8201
ꢀ1373.9346 ꢀ1373.8553 ꢀ1373.8819 ꢀ1373.8811 ꢀ1373.9060 ꢀ1373.9133
energy of +34.5 kcalmolꢀ1.
This is in sharp contrast with
our findings for ethyl com-
plexes of Rh and Co, in which
the ethene-hydride complexes
are about 10–15 kcalmolꢀ1
lower in energy than the a-
agostic isomers, and the transi-
tion-state energies do not
G298
DE
DE0
DG298
E
0
0
0
53.6
49.7
49.7
33.8
31.3
33.0
36.5
33.2
33.6
22.9
19.9
17.9
27.5
23.7
13.3
ꢀ1375.1457 ꢀ1375.0554 ꢀ1375.0914 ꢀ1375.0863 ꢀ1375.1164 ꢀ1375.1152
ꢀ1374.5140 ꢀ1374.4297 ꢀ1374.4640 ꢀ1374.4602 ꢀ1374.4899 ꢀ1374.4895
ꢀ1374.5877 ꢀ1374.4995 ꢀ1374.5361 ꢀ1374.5319 ꢀ1374.5703 ꢀ1374.5802
ꢀ1375.1559 ꢀ1375.0643 ꢀ1375.1025 ꢀ1375.0918 ꢀ1375.1302 ꢀ1375.1303
ꢀ1374.5979 ꢀ1374.5085 ꢀ1374.5472 ꢀ1374.5444 ꢀ1374.5841 ꢀ1374.5953
E0
G298
Esol
Gsol
DE
DE0
DG298
DEsol
DGsol
E
E0
G298
DE
DE0
DG298
0
0
0
0
57.1
52.7
55.2
57.7
55.8
34.5
31.4
32.6
33.3
31.4
37.7
33.9
35.1
35.8
33.3
18.8
15.1
11.3
16.3
8.8
19.5
15.1
5.0
16.3
1.8
exceed
10 kcalmolꢀ1
(see
Figure 3 of ref. [21a]). Like-
wise, in Cr alkyl complexes the
ethene-hydride species are
only 5 kcalmolꢀ1 higher in
energy than the a-agostic spe-
cies, and the maximum transi-
tion-state barriers are about
15 kcalmolꢀ1 (see Figure 2 of
ref. [22]).
0
MP2/cc-pVDZ
ꢀ1370.6814
ꢀ1370.6363
ꢀ1370.0106
ꢀ1370.1013
28.3
ꢀ1370.0497
ꢀ1370.1234
0
0
0
24.5
13.9
[a] Compounds 1, 1a, 2a, and 2 correspond to the ethyl complex, ethene hydride, van der Waals adduct, and
gold hydride plus free ethene, respectively. TS1 and TS2 represent the interconnecting transition states. Com-
pound 1 was selected as the isomer of zero energy.
Chem. Eur. J. 2013, 19, 3954 – 3961
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3957

H. Kçppel et al.
#
other hand, is endergonic by about +5 to 15 kcalmolꢀ1 and
has a free energy barrier of about 15 kcalmolꢀ1 (see
Figure 5 in ref. [23]), because it corresponds to the energeti-
cally unfavorable formation of an empty coordination site in
a complex for which the high-spin analogue of the 18 VE
rule applies.
The inclusion of the solvent effect does not significantly
alter this picture; except for TS1 the relative electronic ener-
gies DE and free energies DG are lowered in dichlorome-
thane by up to 3.2 kcalmolꢀ1 and are of the same order of
magnitude as the ZPE corrections. As a result the Gsol
values for 1 and 2 become nearly identical, with 1 being
only 1.8 kcalmolꢀ1 more stable.
ꢀ
3=2
MkBT
2p
ð2Þ
QTrans
¼
V
Y
1
QVib
¼
ð3Þ
ð4Þ
1 ꢀ eE =k
BT
i
i
3=2pffiffiffiffiffiffiffiffiffiffiffiffiffi
ð2pkBTÞ
IAIBIC
QRot
¼
sp
in which M and V are the molecular mass and volume for an
ideal gas at p=0.1MPa, respectively, Ei is the energy spacing
of the i-th vibrational mode in the harmonic approximation,
and IA, IB, IC represent the moments of inertia. The symme-
try number s accounts for the number of rotations in the nu-
clear permutation group and has been set to unity for all
species. The standard scaling factor of 0.9914 for the BP86/
SVP method was used for all harmonic frequencies.
The MP2 calculations agree with the DFT results in pre-
dicting 1 to be thermodynamically more stable than the hy-
dride 2; for the latter we obtain a relative energy/free
energy of +28.3/+13.9 kcalmolꢀ1.
Comparing the BP86-D3/def2-SVP reaction energetics
with the respective B3LYP/cc-pVDZ values, we note a sig-
nificant destabilization of 2 relative to 1 with the former
method. One may be tempted to regard this solely as a con-
sequence of the loss of dispersion interactions between the
gold-hydride complex and the free ethene in 2. The nearly
identical relative energies of 1 and 2 obtained with the
BP86-D3 and MP2 methods seem to confirm this interpreta-
tion, since the latter also accounts for dispersion interac-
tions. However, a more detailed analysis reveals that this de-
stabilization is partly due to the different functional and
partly due to the inclusion of dispersion corrections in
BP86-D3.[30] The same combination of effects leads to a de-
stabilization of the van der Waals complex 2a relative to 1
by 4–6 kcalmolꢀ1, and a stabilization of TS1 by 3–5 kcal
molꢀ1.
Figure 3 summarizes the values of k(T) obtained for the
b-H-elimination and its inverse, ethene insertion into the
Au H bond, over the temperature range 350–550 K.
ꢀ
Kinetics of b-elimination: To assess the significance of the
b-H-elimination reaction for AuI alkyl complexes we used
the calculated energies E0 (BP86-D3/SVP) and thermo-
chemical data to estimate rate constants according to classi-
cal transition-state theory (TST).[31] The TST rate constant k
is given by Equation (1):
Figure 3. Calculated rate-constants for the thermal decay of (iPr)AuEt
(bottom line) and for the insertion of ethene into (iPr)AuH (top line) as
a function of temperature. For T=450 K half-lives based on first-order
kinetics are given.
It is noted that the high energetic barrier for the b-elimi-
nation of the AuI ethyl complex leads to rather small reac-
tion rates, for example at T=450 K the rate constant for b-
¼
DE0
kBT
h
Q
kB
T
ð1Þ
k ¼
k
e
Qr
elimination becomes
k
AuEt =7.28ꢄ10ꢀ12 sꢀ1, which corre-
¼
in which Q and Qr are the temperature-dependent parti-
tion functions of the transition state and reactant species,
and DE0 represents the ZPE-corrected transition-state
energy. The transmission coefficient k was assumed to be
unity.
sponds to a half-life of 3000 a assuming first-order kinetics.
The experimentally observed decomposition at 1808C there-
fore cannot possibly be due to b-elimination.
A quite different picture is obtained for the ethene-inser-
tion reaction, however. The BP86-D3 method leads to a sig-
nificantly lowered insertion barrier in comparison with our
B3LYP results, and the corresponding rate constant and
half-life for ethene insertion become kAuH =1.12ꢄ10ꢀ3 sꢀ1
and t1/2 =10 min, respectively.
The partition functions are calculated according to the
standard expressions for an ideal gas in the canonical en-
semble; for details see, for example, ref. [32] and other stat-
istical mechanics texts. The total partition function is ob-
tained as a product of translational, vibrational and rotation-
al contributions, QTot ¼ QTrans ꢁ QVib ꢁ QRot, which are in
turn evaluated in Equations (2)–(4) as follows (by using
atomic units):
Experimental studies: These previous calculations indicated
that the iPrAu–alkyl complexes rather than being sensitive
and unstable compounds should display good thermostabili-
3958
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 3954 – 3961

Feasible Elementary Steps in Catalytic Cycles
FULL PAPER
Next we turned to the exper-
imental investigation of the re-
verse reaction (Scheme 2).
When (iPr)AuH was subjected
to ethene, no conversion to 1
could be detected. This is true
for both the experiment at
8 bar of ethene and room tem-
perature and the experiment at
1 bar of ethene at tempera-
tures up to 808C. With norbor-
nene, a strained and thus more
reactive alkene, up to 508C no
insertion could be observed
either. This differs from the ad-
dition of the gold(I) hydride
Figure 4. Left: Solid-state molecular structure of 1. The fitting of the three independent molecules in the ele-
mental cell (middle) shows differences in the orientation of the ligand and the ethyl group. These differences
result mainly from the different environment of the three molecules in the crystal. Right: The packing in the
crystal lattice.
ties. To verify this prediction from the computational stud-
ies, we prepared the iPr-gold(I) complexes 1, 3, and 4 bear-
ing different alkyl groups.
Single crystals of 1 that were suitable for a crystal struc-
ture analysis could be grown. The structure is shown in
Figure 4[25]
.
The results of the thermolysis experiments are shown in
Table 3. Unlike the phosphane complexes reported in the
literature, all (iPr)Au-alkyl complexes are perfectly stable at
1108C in toluene, even after 24 h no decomposition could
be detected (Table 3, entries 1–3). To switch to higher tem-
peratures, DMSO was used as the solvent; now within 12 h
at 1808C a partial decomposition was detected, but there
was no indication that the gold(I) hydride was formed
(Table 3, entries 4–6). To re-evaluate the decomposition
temperatures of the phosphane alkyl-gold(I) complexes,
(Ph)3PAuEt and (Ph)3PAuPh were heated to 1108C in tol-
uene (Table 3, entries 7 and 8). After a short reaction time
of 30 min a gold mirror had formed and free (Ph)3P could
be detected by 31P NMR spectroscopy.
Scheme 2. No insertion of an unactivated alkene could be observed at
8 bar/RT and at 1 bar/808C; compound 1 is not formed.
mentioned in the introduction, which needs an alkyne with
two acceptors. On the other hand, the observation is in ac-
cordance with the fact that no direct insertion of alkenes
and Michael acceptors could be achieved with different or-
ganogold(I) compounds, even in stoichiometric reactions.[33]
This also strongly suggests that the previously observed re-
actions with a,b-unsaturated carbonyl compounds involve
an activation of the carbonyl group by gold acting as a
Lewis acid rather than an insertion into a gold–carbon
bond.[5b,34]
Table 3. Thermal decomposition of the gold(I) complexes.
According to the data from
the computational study, for
normal alkenes such an inser-
tion should be possible only at
temperatures above 1508C,
which is a temperature beyond
the decomposition temperature
of most known gold catalysts
in the presence of organic sub-
strates.
Compound Conditions
(iPr)AuMe [D8]Toluene/
Result
1
2
3
4
5
6
7
8
G
Starting material unchanged
1108C/24 h
[D8]Toluene/
1108C/24 h
N
Starting material unchanged
Starting material unchanged
ACHTUNGTRENNUNG(iPr)AuBu [D8]Toluene
/1108C/24 h
1
A
Partial decomposition detectable in the H NMR spectrum
1808C/12 h
[D6]DMSO/
1808C/12 h
1
Conclusion
(iPr)AuEt
Partial decomposition detectable in the H NMR spectrum, under these
conditions no indication for gold(I) hydride formation
1
(iPr)AuBu [D6]DMSO/
Partial decomposition detectable in the H NMR spectrum, under these
We have carried out a com-
bined experimental and theo-
retical study on the possibility
of b-elimination or alkene in-
sertion in AuI alkyl or AuI hy-
dride complexes, respectively.
1808C/12 h
(Ph)3PAuEt [D8]Toluene/
1108C/12 h
(Ph)3PAuPh [D8]Toluene/
1108C/12 h
conditions no indication for gold(I) hydride formation
After 30 min, formation of a gold mirror. 31P NMR spectrum shows only
one peak at d=ꢀ4 ppm corresponding to the free (Ph)3P.
After 30 min, formation of a gold mirror. 31P NMR shows only one peak at
d=ꢀ4 ppm corresponding to the free (Ph)3P. Biphenyl formation was de-
tected by GC-MS
Chem. Eur. J. 2013, 19, 3954 – 3961
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3959

H. Kçppel et al.
ꢀ788C. The organolithium reagent (4 equiv) was added slowly. The mix-
ture was allowed to warm to RT and stirred for 3 h. The reaction mixture
was then quenched with 2–3 drops of water and stirred for 10 min, then
dried over Na2SO4, filtered through a plug of Celite, topped by a layer of
neutral aluminium oxide. The solution was concentrated in vacuo to
afford the gold complex.
The barrier height for the b-elimination calculated by using
density functional theory clearly rules out the feasibility of
this elementary step in catalytic cycles involving gold(I)
complexes with NHC ligands. Transition-state theory pre-
dicts significant reaction rates for this process only at tem-
peratures above 2008C. The experimentally observed de-
composition of gold(I) alkyl complexes at 1808C is therefore
caused by a different process.
AHCTUNGTRENNUNG
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
Although considerably lower, the barrier to alkene inser-
tion nevertheless remains significant. The reaction of AuI
hydrides with ethene, for example, should not proceed at
temperatures below 1008C. This theoretical result is also
supported by the lack of observing any insertion of even ac-
tivated alkenes into the gold(I)-hydride bond at moderate
temperatures.
We believe the increased barrier height for the b-elimina-
tion/migratory-insertion reaction, as well as the decreased
relative stability of the gold(I) olefin hydride compared with
other early and late transition metals, to be a consequence
of the filled d-shell of gold(I). This also results in a conspic-
uous absence of agostic interactions in AuI alkyl complexes,
which are near-ubiquitous in early as well as late transition-
metal complexes of higher alkyls.
dure A on a 0.322 mmol scale in 95% yield.[35] 1H NMR (301 MHz,
CD2Cl2): d=7.52 (t, J=7.5 Hz, 2H), 7.30–7.37 (m, 4H), 7.10 (s, 2H), 2.64
(sept, J=6.9 Hz , 4H), 1.35 (d, J=6.9 Hz, 12H), 1.22 (d, J=6.9 Hz,
12H), 0.86 (t, J=7.9 Hz, 3H), 0.54 ppm (q, J=7.9 Hz, 2H); 1H NMR
(301 MHz, DMSO): d=0.35 (q, J=7.9 Hz, 2H), 0.76 (t, J=7.9 Hz, 3H),
1.18 (d, J=6.9 Hz, 12H), 1.27 (d, J=6.8 Hz, 12H), 2.57 (sep, J=6.8 Hz,
4H), 7.35 (d, J=7.7 Hz, 4H), 7.50–7.55 (m, 2H), 7.72 ppm (s, 2H);
13C NMR (126 MHz, CD2Cl2): d=13.0, 16.8, 24.0 (4C), 24.5 (4C), 29.1
(4C), 123.0 (2C), 124.2 (4C), 130.3 (2C), 135.4 (2C), 146.4 (4C),
201.9 ppm; IR (Film): n˜ =3163, 3137, 3074, 2965, 2869, 3803, 2301, 2196,
1678, 1593, 1552, 1472, 1413, 1385, 1256, 1214, 1181, 1106, 1059, 992, 954,
711, 676, 532, 451 cmꢀ1
{1,3-bis[2,6-bis(1-methylethyl)phenyl]-1,3-dihydro-2H-imidazol-2-
ylidene}CATHNUGTENR(NUG hydride)gold (2): Compound 2 was prepared according to the
.
A
ACHTUNGTRENNUNG
literature procedure.[6a] 1H NMR (301 MHz, CD2Cl2): d=1.22 (d, J=
6.9 Hz, 6H), 1.34 (d, J=6.9 Hz, 6H), 2.60 (sep, J=6.8 Hz, 2H), 3.39 (s,
1H), 7.16 (s, 2H), 7.30–7.37 (m, 4H), 7.54 ppm (t, J=7.2 Hz, 2H).
1H NMR spectroscopy data matched with that reported previously.[6a]
Complexes of gold in higher oxidation-states are expected
to behave differently, because empty d orbitals may become
available for bonding.
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
A
CD2Cl2): d=ꢀ0.28 (s, 3H), 1.22 (d, J=6.9 Hz, 12H), 1.35 (d, J=6.8 Hz,
12H), 2.65 (sep, J=6.9 Hz, 4H), 7.11 (s, 2H), 7.33 (d, J=7.7 Hz, 4H),
7.53 ppm (t, J=7.8 Hz, 2H); 1H NMR (500 MHz, DMSO): d=ꢀ0.46 (s,
3H), 1.18 (d, J=6.9 Hz, 12H), 1.28 (d, J=6.9 Hz, 12H), 2.58 (sep, J=
6.8 Hz, 4H), 7.36 (d, J=7.8 Hz, 2H), 7.50–7.54 (m, 2H), 7.79 ppm (s,
2H).
Experimental Section
Material and methods: All reagents were used without further purifica-
tion unless otherwise noted. Dry solvents were dispensed from solvent
purification system MB SPS-800. The preparation of air- and moisture-
sensitive materials was carried out in flame dried flasks under an atmos-
phere of dinitrogen using Schlenk-techniques. Reactions were performed
in dry and degassed solvents. Thin-layer chromatography (TLC) was per-
formed using Polygram pre-coated plastic sheets SIL G/UV254 (SiO2,
0.20 mm) and Alugram pre-coated aluminium sheets SIL G/UV254
(SiO2, 0.20 mm) from Macherey–Nagel. Column chromatography was
performed by using silica gel (40.0–63.0 nm particle size) from Macher-
ey–Nagel. NMR spectra were recorded on Bruker Avance 500, Bruker
Avance 300 and Bruker ARX-250 spectrometers at RT. Chemical shifts
(in ppm) were referenced to residual solvent proton/carbon peak or
using external standard 85% H3PO4 for 31P NMR spectroscopy. Signal
multiplicity was determined as s (singlet), d (doublet), t (triplet), q (quar-
tet) or m (multiplet). 13C assignment was achieved DEPT135 spectra. MS
spectra were recorded on a Vakkum Generators ZAB-2F, Finnigan MAT
TSQ 700 or a JEOL JMS-700 spectrometer. GC-MS spectra were record-
ed on an Agilent 5890 Series II Plus with a HP 5972 mass analyzer. IR
spectra were recorded on a Bruker Vector 22 FT-IR. Crystal structure
analysis was accomplished on Bruker Smart CCD or Bruker APEX dif-
fractometers.
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
B
a
CD2Cl2): d=0.59 (t, J=7.3 Hz, 3H), 0.66 (t, J=7.3 Hz, 2H), 0.85 (qu, J=
7.4 Hz, 2H), 1.21 (d, J=6.9 Hz, 12H), 1.28 (qu, J=7.4 Hz, 2H), 1.35 (d,
J=6.9 Hz, 12H), 2.64 (sep, J=6.8 Hz, 4H), 7.11 (s, 2H),7.31 (d, J=
7.8 Hz, 4H), 7.51 (t, J=7.8 Hz, 2H); 1H NMR (301 MHz, DMSO): d=
0.43–0.59 (m, 5H), 0.80 (t, J=7.3 Hz, 2H), 1.20 (d, J=6.9 Hz, 12H), 1.25
(m, 2H), 1.29 (d, J=6.8 Hz, 12H), 2.57 (sep, J=6.8 Hz, 4H), 7.36 (d, J=
7.7 Hz, 4H), 7.49–7.55 ppm (m, 2H), 7.73 ppm (s, 2H); 13C NMR
(101 MHz, CD2Cl2): d=14.2, 21.5, 24.1 (4C), 24.4 (4C), 29.0, 29.1 (4C),
34.9, 122.9 (2C), 124.2 (4C), 130.3 (2C), 135.4 (2C), 146.4 (4C), 202.6; IR
(KBr): n˜ =3161, 2963, 2926, 2868, 2790, 2303, 2197, 1471, 1412, 1384,
1365, 1342, 1106, 1058955, 803, 756, 711, 548, 449 cmꢀ1; MS (FAB+): m/z:
641 [MꢀH]+
;
HRMS (FAB+): calcd for C31H44AuN2: 641.3170
[C31H44AuN2]+; found 641.3190.
Thermolysis experiments: The gold complex (24.0 mmol) was dissolved in
deuterated toluene (600 mL) or deuterated DMSO (600 mL) in a NMR
tube and heated to 1108C for 24 h or 1808C for 12 h, respectively.
Insertion experiments: a) In a flame-dried Schlenk flask under an atmos-
phere of ethene, the gold hydride complex (51.0 mmol) was dissolved in
deuterated benzene. A stream of ethene was bubbled through the reac-
tion mixture for 30 min. Then the reaction was heated to 808C for 12 h.
Procedure A: iPrAuCl (200 mg, 1 equiv) was dissolved in THF (0.04m) in
a flame-dried Schlenk flask in an atmosphere of dinitrogen and cooled to
ꢀ208C. The Grignard reagent (4 equiv) was added slowly. The mixture
was allowed to warm to RT and stirred for 3 h. The reaction mixture was
then quenched with 2–3 drops of water and stirred for 10 min, then dried
over Na2SO4, filtered through a plug of Celite, topped by a layer of neu-
tral aluminium oxide. The solution was concentrated in vacuo to afford
the gold complex.
b) In a pressure-resistant NMR tube the gold complex was dissolved in
deuterated dichloromethane under a pressure of 8 bar of ethene for 24 h.
Procedure B: iPrAuCl (1 equiv) was dissolved in THF (0.04m) in a
flame-dried Schlenk flask in an atmosphere of dinitrogen and cooled to
3960
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 3954 – 3961

Feasible Elementary Steps in Catalytic Cycles
FULL PAPER
perts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Po-
melli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzew-
ski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D.
Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J.
Fox, Gaussian, Inc., Wallingford CT, 2010.
Acknowledgements
This work was supported by the Deutsche Forschungsgemeinschaft
through Sonderforschungsbereich 623 “Molekulare Katalysatoren: Struk-
tur und Funktionsdesign”. The authors wish to express their gratitude to
Mikko Muuronen for very helpful advice and assistance concerning the
DFT-D3 calculations.
[13] K. A. Peterson, C. Puzzarini, Theor. Chem. Acc. 2005, 114, 283–296.
[14] T. H. Dunning, J. Chem. Phys. 1989, 90, 1007–1023.
[15] D. Figgen, G. Rauhut, M. Dolg, H. Stoll, Chem. Phys. 2005, 311,
227–244.
[16] S. Niu, M. B. Hall, Chem. Rev. 2000, 100, 353–405.
[17] T. A. Halgren, W. N. Lipscomb, Chem. Phys. Lett. 1977, 49, 225–
232.
[18] C. Peng, H. B. Schlegel, Isr. J. Chem. 1993, 33, 449–454.
[19] S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010,
132, 154104.
[20] TURBOMOLE V6.4 2012, a development of University of Karls-
ruhe and Forschungszentrum Karlsruhe GmbH, 1989–2007; TUR-
BOMOLE GmbH, since 2007; available from http://www.turbomo-
le.com.
[21] M. Muuronen, J. E. Perea-Buceta, M. Nieger, M. Patzschke, J.
Helaja, Organometallics 2012, 31, 4320–4330.
[1] a) X. Wang, L. Andrews, J. Am. Chem. Soc. 2001, 123, 12899–
12900; b) X. Wang, L. Andrews, Angew. Chem. 2003, 115, 5359–
5364; Angew. Chem. Int. Ed. 2003, 42, 5201–5206; c) L. Andrews,
X. Wang, J. Am. Chem. Soc. 2003, 125, 11751–11760; for a detailed
discussion, see: d) M.-J. Crawford, T. M. Klapçtke, Angew. Chem.
2002, 114, 2373–2375; Angew. Chem. Int. Ed. 2002, 41, 2269–2271.
[2] a) A. S. K. Hashmi, G. J. Hutchings, Angew. Chem. 2006, 118, 8064–
8105; Angew. Chem. Int. Ed. 2006, 45, 7896–7936; b) A. Fꢅrstner,
P. W. Davis, Angew. Chem. 2007, 119, 3478–3519; Angew. Chem.
Int. Ed. 2007, 46, 3410–3449; c) A. S. K. Hashmi, Chem. Rev. 2007,
107, 3180–3211; d) A. Arcadi, Chem. Rev. 2008, 108, 3266–3325;
e) E. Jimꢆnez-NfflÇez, A. M. Echavarren, Chem. Rev. 2008, 108,
3326–3350; f) Z. G. Li, C. Brouwer, C. He, Chem. Rev. 2008, 108,
3239–3265; g) D. J. Gorin, B. D. Sherry, F. D. Toste, Chem. Rev.
2008, 108, 3351–3378; h) A. S. K. Hashmi, M. Bꢅhrle, Aldrichimica
Acta 2010, 43, 27–33.
[22] a) A. Schꢀfer, C. Huber, R. Ahlrichs, J. Chem. Phys. 1994, 100,
5829–5835; b) F. Weigend, M. Hꢀser, H. Patzelt, R. Ahlrichs, Chem.
Phys. Lett. 1998, 294, 143–152; c) F. Weigend, R. Ahlrichs, Phys.
Chem. Chem. Phys. 2005, 7, 3297–3305.
[23] a) M. Sierka, A. Hogekamp, R. Ahlrichs, J. Chem. Phys. 2003, 118,
9136–9148; b) K. Eichkorn, F. Weigend, O. Treutler, R. Ahlrichs,
Theor. Chem. Acc. 1997, 97, 119–124; c) F. Weigend, Phys. Chem.
Chem. Phys. 2006, 8, 1057–1065; d) A. Hellweg, C. Hꢀttig, S. Hçfen-
er, W. Klopper, Theor. Chem. Acc. 2007, 117, 587–597.
[24] D. Andrae, U. Haeussermann, M. Dolg, H. Stoll, H. Preuss, Theor.
Chim. Acta 1990, 77, 123.
[25] CCDC-897806 (1) contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
[26] R. Xu, G. Klatt, H. Wadepohl, H. Kçppel, Inorg. Chem. 2010, 49,
3289–3296.
[27] R. Xu, M. Bittner, G. Klatt, H. Kçppel, J. Phys. Chem. A 2008, 112,
13139–13148.
[28] R. Xu, G. Klatt, M. Enders, H. Kçppel, J. Phys. Chem. A 2012, 116,
1077–1085.
[29] For a recent review of theoretical aspects of the agostic interaction,
see: E. Clot, O. Eisenstein, Struct. Bonding 2004, 113, 1–36.
[30] G. Klatt, unpublished results.
[31] For modern discussions of transition-state theory, see: a) P. Pechu-
kas, in Dynamics of Molecular Collisions, Part B, (Ed.: W. H.
Miller), Plenum, New York, 1976; Chapter 6; b) W. H. Miller, Acc.
Chem. Res. 1976, 9, 306–312; c) D. G. Truhlar, W. L. Hase, J. T.
Hynes, J.Phys. Chem. 1983, 87, 2664–2682.
[32] D. A. McQuarrie, Statistical Thermodynamics, Harper & Row, New
York, 1976.
[33] A. S. K. Hashmi, T. D. Ramamurthi, F. Rominger, J. Organomet.
Chem. 2009, 694, 592–597.
[3] For a detailed study of the thermal decomposition of alkyl-gold(I)
complexes, see: A. Tamaki, J. K. Kochi, J. Organomet. Chem. 1973,
61, 441–450.
[4] a) H. H. Brintzinger, D. Fischer, R. Mꢅlhaupt, B. Rieger, R. M.
Waymouth, Angew. Chem. 1995, 107, 1255–1283; Angew. Chem. Int.
Ed. Engl.. 1995, 34, 1143–1170; b) M. Delferro, T. J. Marks, Chem.
Rev. 2011, 111, 2450–2485.
[5] a) R. T. Baker, P. Nguyen, T. B. Marder, S. A. Westcott, Angew.
Chem. 1995, 107, 1451–1452; Angew. Chem. Int. Ed. Engl. 1995, 34,
1336–1338; b) A. S. K. Hashmi, L. Schwarz, J.-H. Choi, T. M. Frost,
Angew. Chem. 2000, 112, 2382–2385; Angew. Chem. Int. Ed. 2000,
39, 2285–2288.
[6] a) E. Y. Tsui, P. Mꢅller, J. P. Sadighi, Angew. Chem. 2008, 120, 9069–
9072; Angew. Chem. Int. Ed. 2008, 47, 8937–8940; before, gold(I)
hydride had only been detected in the gas phase: b) U. Ringstrçm,
Nature 1963, 198, 981.
[7] a) A. L. Escalle, G. Mora, F. Gagosz, N. Mꢆzailles, X. F. Le Goff, Y.
Jean, P. Le Floch, Inorg. Chem. 2009, 48, 8415–8422; for the previ-
ous description of mixed-metal dinuclear gold hydrides with bridg-
ing hydrides, see: H. Lehner, D. Matt, P. S. Pregosin, L. M. Venanzi,
A. Albinati, J. Am. Chem. Soc. 1982, 104, 6825–6827.
[8] A. Comas-Vives, C. Gonzµlez-Arellano, A. Corma, M. Iglesias, F.
Sµnchez, G. Ujaque, J. Am. Chem. Soc. 2006, 128, 4756–4765.
[9] H. Ito, K. Takagi, T. Miyahara, M. Sawamura, Org. Lett. 2005, 7,
3001–3004.
[10] H. Ito, T. Saito, T. Miyahara, C. Zhong, M. Sawamura, Organometal-
lics 2009, 28, 4829–4840.
[11] B. Alcaide, P. Almendros, T. M. del Campo, I. Fernandez, Chem.
Commun. 2011, 47, 9054–9056.
[34] a) G. Dyker, E. Muth, A. S. K. Hashmi, L. Ding, Adv. Synth. Catal.
2003, 345, 1247–1252; b) A. S. K. Hashmi, L. Schwarz, P. Rubenba-
uer, M. C. Blanco, Adv. Synth. Catal. 2006, 348, 705–708.
[35] V. J. Scott, J. A. Labinger, J. E. Bercaw, Organometallics 2010, 29,
4090–4096.
[12] Gaussian 09, Revision B.01, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V.
Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X.
Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Son-
nenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven,
J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J.
Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, T. Keith, R. Ko-
bayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant,
S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene,
J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gom-
[36] N. P. Mankad, F. D. Toste, J. Am. Chem. Soc. 2010, 132, 12859–
12861.
Received: August 27, 2012
Revised: December 10, 2012
Published online: February 10, 2013
Chem. Eur. J. 2013, 19, 3954 – 3961
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3961