Unexpected anion effect in the alkoxylation of alkynes catalyzed by N-heterocyclic carbene (NHC) cationic gold complexes

Article abstract of DOI:10.1002/chem.201404539

The intermolecular alkoxylation of alkynes is the oldest application of cationic gold(I) catalysts; however, no systematic experimental data about the role of the anion are available. In this contribution, the role of the anion in this catalytic reaction as promoted by a N-heterocyclic carbene-based gold catalyst, [(NHC)AuX] (X = BARF^- BF₄^- , OTf^- ,OTs^-, TFA^- , or OAc^- ) is analyzed, through a combined experimental (NMR spectroscopy) and theoretical (DFT calculation) approach. The most important factor seems to be the ability to abstract the proton from the methanol during the nucleophilic attack, and such ability is related to the anion basicity. On the other hand, too high coordination power or basicity of the anion worsens the catalytic performance by preventing alkyne coordination or by forming too much free methoxide in solution, which poisons the catalyst. The intermediate coordinating power and basicity of the OTs^- anion provides the best compromise to achieve efficient catalysis.

Full text of DOI:10.1002/chem.201404539

DOI: 10.1002/chem.201404539
Communication
&
Gold Catalysis
Unexpected Anion Effect in the Alkoxylation of Alkynes Catalyzed
by N-Heterocyclic Carbene (NHC) Cationic Gold Complexes
[
a]
[b]
[c]
[d]
Luca Biasiolo, Marina Trinchillo, Paola Belanzoni, Leonardo Belpassi,
[b]
[d]
[b]
[c]
Vincenzo Busico, Gianluca Ciancaleoni, Angela D’Amora, Alceo Macchioni,
Francesco Tarantelli, and Daniele Zuccaccia*
[c]
[a]
[
5]
[6]
can influence the catalytic activity, the regioselectivity, and
Abstract: The intermolecular alkoxylation of alkynes is the
oldest application of cationic gold(I) catalysts; however,
no systematic experimental data about the role of the
anion are available. In this contribution, the role of the
anion in this catalytic reaction as promoted by a N-hetero-
[7]
even the stereoselectivity of the process. Moreover, it has
been observed that the anion can influence the structure of
[8]
[9]
the catalyst and of the intermediates in the catalytic cycle.
Although the importance of the anion is well recognized, its
role in each individual step of the reaction mechanism is still
largely unknown. One of the crucial pieces of information that
may contribute to the comprehension of such an effect is the
relative orientation of the anion/cation in solution. In the past
years, we studied the ion-pair structure of different [(L)Au(S)]X
salts, where S is an unsaturated organic substrate (alkene or
ꢀ
cyclic carbene-based gold catalyst, [(NHC)AuX] (X=BARF ,
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
BF4 , OTf , OTs , TFA , or OAc ) is analyzed, through
a combined experimental (NMR spectroscopy) and theo-
retical (DFT calculation) approach. The most important
factor seems to be the ability to abstract the proton from
the methanol during the nucleophilic attack, and such
ability is related to the anion basicity. On the other hand,
too high coordination power or basicity of the anion wor-
sens the catalytic performance by preventing alkyne coor-
dination or by forming too much free methoxide in solu-
tion, which poisons the catalyst. The intermediate coordi-
alkyne) and L represents a series of ligands commonly used in
[10,11]
catalysis.
Our results showed that the position of the anion
can be finely modulated by the choice both of the ligand L
[12]
and the substrate S.
The intermolecular alkoxylation of alkynes and similar reac-
tions (hydration and hydrophenoxylation) are among the
ꢀ
nating power and basicity of the OTs anion provides the
[13]
oldest applications of gold(I) catalyst. From the seminal work
best compromise to achieve efficient catalysis.
[14]
of Teles, several other studies on these classes of reaction
[15]
have been published, in order to find highly active catalysts
[3b,16]
and to shed more light on the reaction mechanism.
Homogeneous gold catalysis has received considerable atten-
tion in recent years and great efforts have been made to opti-
Maier and Zhdanko have shown that when the typical N-
heterocyclic carbene [NHC, (1,3-bis(2,6-di-isopropylphenyl)-im-
idazol-2-ylidene)] is used as the ancillary ligand, the catalyzed
addition of methanol to 3-hexyne is pseudo-zero order in rela-
tion to 3-hexyne and pseudo-first order in relation to methanol
[1]
mize known reactions and find new catalytic applications. Re-
cently, in-depth kinetic and mechanistic studies on the gold(I)-
catalyzed nucleophilic addition to a carbonꢀcarbon unsaturat-
ed bond have been published, with the aim to understand the
[16a]
and the catalyst.
[
2]
ligand effect and to isolate and characterize the intermedi-
The gold metal fragment acts as a Lewis acid, coordinating
3-hexyne (Scheme 1, OSIP=outer sphere ion pair), which sub-
sequently undergoes a nucleophilic attack from MeOH, with
the formation of an organogold intermediate (Scheme 1, inter-
mediate I). The goldꢀcarbon bond in the intermediate is typi-
cally cleaved by a proton (protodeauration) to give the vinyl
[
3,4]
ates involved in the catalytic cycle.
However, also the anion
[a] L. Biasiolo, Dr. D. Zuccaccia
Dipartimento di Chimica, Fisica e Ambiente, Universitꢀ di Udine
Via Cotonificio 108, I-33100 Udine (Italy)
E-mail: daniele.zuccaccia@uniud.it
[b] M. Trinchillo, Prof. Dr. V. Busico, Dr. A. D’Amora
Dipartimento di Scienze Chimiche, Universitꢀ Federico II di Napoli
Via Cintia- Complesso di Monte S. Angelo- 80126 Napoli (Italy)
[
c] Prof. P. Belanzoni, Prof. A. Macchioni, Prof. F. Tarantelli
Dipartimento di Chimica, Biologia e Biotecnologie
Universitꢀ degli Studi di Perugia
Via Elce di sotto 8, I-06123, Perugia (Italy)
[d] Dr. L. Belpassi, Dr. G. Ciancaleoni
Istituto di Scienze e Tecnologie Molecolari del CNR (CNR-ISTM)
Universitꢀ degli Studi di Perugia
Via Elce di sotto 8, I-06123, Perugia (Italy)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201404539.
Scheme 1. Steps in the alkoxylation of 3-hexyne with [(NHC)AuX].
Chem. Eur. J. 2014, 20, 1 – 6
1
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
ether and regenerate the catalyst (Scheme 1, ISIP=inner
sphere ion pairs).
Table 1. Alkoxylation of 3-hexyne with [(NHC)AuX].
[
e]
[f]
ꢀ1
The formation of vinyl ether is the key gold-catalyzed step,
although the addition of the second molecule of methanol,
Entry
1
Conditions
Anion
Conv. [%]
t [min]
TOF [min
]
ꢀ
[a]
[a]
[a]
[a]
[a]
[a]
[a]
[a]
OTs
99
99
99
99
99
99
99
0
17
20
22
33
40
48
266
–
5.6
4.8
4.3
2.9
2.4
2.0
0.4
–
[
16b]
[c]
ꢀ
leading to the final product (acetal), seems to be very fast
2
3
OTs
[
16a]
[d]
ꢀ
OTs
and is considered to be a classical proton-catalyzed process.
ꢀ
4
5
6
7
8
OTf
BF
To the best of our knowledge, no systematic experimental data
about the role of the anion are available. In this contribution,
we analyze in detail the role of the anion in different steps of
this catalytic cycle through a combined experimental/theoreti-
cal approach. The mechanistic framework is largely simplified if
one chooses [(NHC)AuX] species as catalysts: the stability of
the species [(NHC)Au(S)]X and the absence of gem-diaurated
4ꢀ
ꢀ
BARF
TFA
OAc
ꢀ
ꢀ
ꢀ
9
[b]
[b]
[b]
[b]
OTs
OTf
BF4
99
99
99
99
18
47
64
89
5.3
2.0
1.5
1.1
ꢀ
10
ꢀ
11
ꢀ
12
BARF
species [(NHC-Au) S]X under the catalytic conditions are
2
[
16a]
known.
We have considered a series of anions with increas-
3
[a] 308C, 3-hexyne (100 mL), 1X loading (1 mol%), CH OH (143 mL).
[b] 308C, 3-hexyne (100 mL), 1X loading (1 mol%), CH OD (143 mL) in
3
[
17]
ing coordination ability:
tetrakis(3,5-bis(trifluoro-methyl)-
ꢀ
ꢀ
CDCl3 (400 mL). [c] With 5% NBu OTs as additive. [d] With 15% pTsOH as
4
phenyl)-borate (BARF ), tetrafluoroborate (BF ), trifluorome-
4
1
additive. Conversions were determined by H NMR spectroscopy, using
ꢀ
ꢀ
thanesulfonate (OTf ), p-toluensolfonate (OTs ), trifluoroacetate
TMS as internal standard, as average of three runs. [e] Time for 95% con-
version. [f] TOF=([acetal]/[1X])/t at 95% of conversion.
ꢀ
ꢀ
(
TFA ) and acetate (OAc ).
All the catalysts [(NHC)AuX] (1X) were synthesized from
(NHC)AuCl] through anion metathesis procedures, but in the
[
ꢀ
ꢀ
[20]
case of BARF and BF4 the complexes are not stable enough
strongly coordinating anions reduce the catalytic activity.
Indeed, from the mechanism depicted in Scheme 1, we may
expect that non-coordinating anions should maximize the cat-
alytic rate. On the other hand, the trend of the TOF values
2
and so [(NHC)Au(h -3-hexyne)]X complexes (1BARF and
[
18]
1
BF4, respectively) have been used as catalysts (see the Sup-
porting Information for details and characterization of new
complexes).
[
21]
does not reflect the basicity of the anions towards the ab-
straction of the alcoholic proton of the methanol before or
after the nucleophilic attack, either. As a matter of fact, the
The results for the alkoxylation of 3-hexyne with MeOH are
[
19]
shown in Figure 1 and Table 1. The most active catalyst is
ꢀ
[22]
complex 1OTs, followed by 1OTf, 1BF , 1BARF, then 1TFA,
best performing anion, OTs , possesses intermediate coordi-
nating and basicity properties in the considered series of coun-
terions. To get a deeper insight into these intriguing results,
we analyzed the effect of the anion in the ISIP/OSIP pre-equi-
librium step (Scheme 1). The equilibrium [(NHC)AuX]+3-
4
2
hexyne$[(NHC)Au(h -3-hexyne)]X between the ISIP and the
OSIP (Scheme 1) is shifted to the right when weakly and non-
ꢀ
ꢀ
ꢀ
[18]
coordinating OTf , BF4 , and BARF anions are considered.
This means that the concentration of ISIP is constant during
the reaction, leading to a linear profile of the conversion
against time (Figure 1).
In contrast, for 1TFA (Figure 1), the reaction rate is not con-
1
9
stant, but decreases after about 90 min. Recording the F NMR
spectrum of the reaction mixture at different reaction times,
we observed the disappearance of a broad resonance at
ꢀ
74.8 ppm and the simultaneous formation of a sharp new
resonance at ꢀ76.2 ppm (assigned to trifluoroacetic acid,
TFAH, see the Supporting Information). Dissolving 1TFA in
Figure 1. Rate of alkoxylation of 3-hexyne with [(NHC)AuX].
CDCl with 400 equivalents of methanol, the same phenome-
3
which led to complete conversion of 3-hexyne in 17, 33, 40,
non is observed, with the total disappearance of the NMR
signal corresponding to 1TFA within one hour. If 3-hexyne is
added to the solution after three hours (at complete formation
of TFAH), no catalytic activity was observed. These experimen-
tal results can be explained by assuming an acid–base reaction
48, and 266 min, respectively (entries 1, 4, 5, 6 and 7 in
Table 1). The values of the turnover frequencies (TOF) are 5.6,
ꢀ
1
2
.9, 2.4, 2.0, and 0.4 min for 1OTs, 1OTf, 1BF , 1BARF, and
4
1
TFA, respectively (Table 1). 1OAc does not show any catalytic
activity.
Very surprisingly, the trend of the TOF values does not re-
ꢀ
ꢀ
between TFA and MeOH, giving TFAH and MeO , which, in
turn, probably poisons the catalyst by forming the catalytically
[
17]
[16]
flect the coordinating power of the anions. Usually, when
the first catalytic step involves competition between the sub-
strate and the anion for the metal coordination vacancy,
inert [(NHC)Au(OMe)] species.
For 1OAc, which shows no catalytic activity (Figure 1), the
reaction with MeOH is very slow (30% of acetic acid in 48 h).
&
&
Chem. Eur. J. 2014, 20, 1 – 6
www.chemeurj.org
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communication
In this case, the ISIP is predominant in solution owing to the
strong coordinating ability of the anion (see the Supporting
Information).
study on the addition of the first methanol molecule, as we
mentioned the addition of the second molecule of methanol is
[16b]
shown to be fast.
We selected the complex [(NHC’)AuOTs],
1
Finally, for the H NMR spectra recorded during the catalysis
(NHC’=1,3-dimethylimidazol-2-ylidene) with 2-butyne and
methanol as models for the catalyst, substrate, and nucleo-
phile, respectively. In particular, we decided to investigate the
with 1OTs, the resonances of the tosylate anion closely resem-
ble those of NBu OTs (see the Supporting Information), which
4
ꢀ
indicates that the anion is not coordinated to the metal. As
the pre-equilibrium is completely shifted toward the ISIP in ab-
sence of methanol (see the Supporting Information), we can
surmise that methanol may help the de-coordination of the
OTs anion because, unexpectedly, it is the most efficient in
ꢀ
spite of its stronger coordinating ability with respect to BF
4
ꢀ
and BARF . According to recent benchmark papers, we used
density functional theory (DFT) in combination with the BP86
exchange-correlation functional to optimize the geometries
and the double-hybrid functional B2PLYP to compute the cor-
[23]
anion, probably through the formation of a hydrogen bond.
The importance of the pre-equilibrium step is demonstrated
by the fact that the addition of an external salt, such as
[30,31]
responding energy.
For the pre-equilibrium state, the electronic energy of the
ꢀ
1
NBu OTs (5%), reduces the TOF to 4.8 min , probably because
4
ꢀ
1
of the shift of the ISIP/OSIP equilibrium toward the ISIP.
We have determined the order of reaction with respect to
complex [(NHC’)AuOTs]·(2-butyne) is 3.6 kcalmol lower than
2
that of [(NHC’)Au(h -2-butyne)]OTs, a fact that indicates that
ꢀ
1
OTs as first-order (see the Supporting Information). Note that
the alkyne substitution of the coordinated OTs is somewhat
the same order has been measured in the case of non-coordi-
unfavorable. Interestingly, when one molecule of methanol is
introduced into the system, the difference between
[
16a]
nating anions.
Consequently, only one gold atom is in-
2
volved in the rate-determining step (RDS) of the reaction. How-
ever, the RDS can be either the nucleophilic attack or the pro-
todeauration (Scheme 1). Different catalytic conditions have
been tested. In the presence of HOTs (15%), the reaction is de-
[(NHC’)AuOTs]·(2-butyne)·(MeOH) (A_ISIP) and [(NHC’)Au(h -2-
ꢀ
1
butyne)]OTs·(MeOH) (A_NR) lowers to 1.8 kcalmol thanks to
hydrogen-bond formation between the methanol and the
anion (O ꢀH
=1.722 ꢂ). In the A_NR configuration, the
MeOH
OTs
ꢀ
1
celerated and the TOF changes from 5.6 to 4.3 min (entries 1
and 3, Table 1). In addition, by using CH OD instead of CH OH,
anion weakly interacts with the gold (AuꢀO =3.133 ꢂ) and
OTs
the methanol is positioned nearly over the gold. However, a dif-
ferent local minimum configuration has been found, with the
methanol still interacting with the anion, but located at the
opposite side of the alkyne with respect to the gold (A_RC,
3
3
we observe a slight reduction in the TOF, which shifts from 5.6
ꢀ
1
to 5.3 min (compare entries 1 and 9, Table 1), giving a kinetic
[
24]
isotopic effect (KIE)
equal to 1.1. For 1OTf, 1BF , and
4
1
BARF, the KIE was equal to 1.4, 1.6, and 1.8, respectively (en-
AuꢀO =3.283 ꢂ, O ꢀH
=1.780 ꢂ). The energy of A_RC is
MeOH
OTs
OTs
ꢀ
1
tries 4 vs. 10, 5 vs. 11, and 6 vs. 12, Table 1). These values of KIE
point out that, under our conditions, the turnover-limiting step
only 0.9 kcalmol higher than that of A_NR (Figure 2). A_RC is
a particularly suitable reactant complex configuration for the
anti-periplanar addition of the nucleophile, which is the most
[
25]
is the nucleophilic attack of the methanol.
However, the
[32]
small increase in the KIE on going from a more coordinating
favored mechanism according to previous studies.
ꢀ
ꢀ
anion (OTs ) to a non-coordinating one (BARF ) may indicate
that the importance of the protodeauration step in the mecha-
Starting from A_RC, the methanol can approach the carbon
atom coordinated to the gold, passing through a transition
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ1
nism increases from OTs to OTf , from OTf to BF , and from
state (TS_attack) with an activation barrier of 15.6 kcalmol
4
ꢀ
ꢀ
BF4 to BARF . In agreement with our findings, Zhdanko and
Maier showed clearly that the reaction rate of the hydro-
alkoxylation of 3-hexyne by methanol (in methanol) is inde-
pendent from the amount of acid, excluding the protodeaura-
(computed with respect to A_RC). Note that on removing the
anion from the catalytic cycle, the s-coordinated gold(I) vinyl
ether (B-like structure of Figure 2) is not stable. In the transi-
tion-state geometry, the distance between the methanol and
the carbon atom is 2.084 ꢂ (Figure 2, right), with the anion fa-
cilitating the attack in two ways: it acts as a template, keeping
the reactive methanol molecule in the right position for addi-
tion, and, simultaneously, it “activates” the nucleophile through
[
16a]
tion as the rate-limiting step.
We also note that the turn-
over-limiting step strongly depends on the reaction conditions
and on the reactants. In a recent paper by Straub and co-work-
[
26]
ers, a KIE of 3–5 is observed for the hydration of terminal
alkyne conducted in methanol, and the group of Gagnꢁ and
Widenhoefer reported a KIE of 5.3 for the gold-catalyzed intra-
molecular hydroalkoxylation of 2,2-diphenyl-4,5-hexadien-1-ol
a hydrogen bond that is shorter than in A_RC (O ꢀH
=
MeOH
OTs
1.538 ꢂ). The product of the attack is a metal s-coordinated
vinyl ether (B), with the hydrogen atom of the methanol com-
[
27]
to a 2-vinyltetrahydrofuran derivative. These authors suggest
that protonolysis of the goldꢀcarbon bond was the turnover-
limiting step in both cases.
pletely transferred to the anion (O ꢀH
=1.035 ꢂ), giving
MeOH
OTs
a neutral molecule of p-toluenesulfonic acid. The acid easily
donates its hydrogen atom to the carbon coordinated to the
gold (TS_proto), thereby acting as an efficient proton shuttle
between the methanol and the substrate (activation barrier
From a computational point of view, the role of the anion
has usually been recognized only in the protodeauration
[
28]
ꢀ1
step, where it acts as a “proton shuttle”, or forms weak inter-
actions with the substrate, in order to explain the enantioselec-
4.7 kcalmol with respect to B). The product of the proto-
deauration step is a p-coordinated vinyl ether (C). In C, the
gold is not equidistant from the two carbon atoms, but instead
is further away from the carbon that underwent the nucleo-
philic attack (AuꢀC2=2.209 ꢂ, AuꢀC1=2.430 ꢂ). The same
[
29]
tivity. However, as the nucleophilic attack is the rate-deter-
mining step under our catalytic conditions, an active role of
the anion is also expected within this step. We focused our
Chem. Eur. J. 2014, 20, 1 – 6
www.chemeurj.org
3
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
Figure 2. Left: DFT-calculated profile for the reaction between 2-butyne and methanol, catalyzed by (NHC’)AuOTs. The structures of the species are also
shown. The protons of 2-butyne have been omitted for clarity. Right: Geometric structure of TS_attack, with relevant distances [ꢂ].
[
33]
asymmetry has been experimentally observed, too. Finally,
the vinyl ether can de-coordinate and be substituted by anoth-
Keywords: alkoxylation · anion effects · density functional
calculations · gold catalysis · reaction mechanisms
[
34]
er molecule of alkyne.
From the structure of TS_attack, the experimentally ob-
served anion effect (Figure 1) can be rationalized on the basis
of the coordinating and basicity properties of the different
anions. The most important factor seems to be the ability to
abstract the proton from the methanol, and this ability is pro-
portional to the anion basicity. Consistent with this picture, the
[
1] A. S. K. Hashmi, Chem. Rev. 2007, 107, 3180.
[2] W. Wang, G. B. Hammond, B. Xu, J. Am. Chem. Soc. 2012, 134, 5697.
[
[
[
3] a) A. S. K. Hashmi, Angew. Chem. 2010, 122, 5360–5369; Angew. Chem.
Int. Ed. 2010, 49, 5232–5241; b) C. Obradors, A. M. Echavarren, Chem.
Commun. 2014, 50, 16–28.
4] a) H. Schmidbaur, A. Schier, Organometallics 2010, 29, 2–23 and refer-
ence therein; b) R. E. M. Brooner, R. Widenhoefer, Angew. Chem. 2013,
1
25, 11930–11941; Angew. Chem. Int. Ed. 2013, 52, 11714–11724.
performances of 1BF and 1BARF are worse than that of 1OTs
4
5] a) C. Brouwer, C. He, Angew. Chem. 2006, 118, 1776–1779; Angew.
Chem. Int. Ed. 2006, 45, 1744–1747.
because such anions have more difficulty forming the corre-
sponding acids HBF and HBARF. On the other hand, too high
4
[6] Y. Xia, A. S. Dudnik, V. Gevorgyan, Y. Li, J. Am. Chem. Soc. 2008, 130,
coordination or basicity power of the anion worsens the cata-
lytic performance, preventing alkyne coordination (as in the
6940.
[
7] G. L. Hamilton, E. J. Kang, M. Mba, F. D. Toste, Science 2007, 317, 496.
[8] K. Aikawa, M. Kojima, K. Mikami, Angew. Chem. 2009, 121, 6189–6193;
Angew. Chem. Int. Ed. 2009, 48, 6073–6077.
9] D. Weber, T. D. Jones, L. Adduci, M. R. Gagnꢁ, Angew. Chem. 2012, 124,
2502–2506; Angew. Chem. Int. Ed. 2012, 51, 2452–2456.
ꢀ
ꢀ
case of OAc ) or forming too much free MeO in solution,
ꢀ
which poisons the catalyst (as in the case of TFA ).
[
In conclusion, we have shown that the anion properties,
both coordination ability and basicity, have a great and under-
evaluated impact on the catalytic performance of gold com-
[
10] a) D. Zuccaccia, L, Belpassi, F. Tarantelli, A. Macchioni, J. Am. Chem. Soc.
009, 131, 3170–3171; b) N. Salvi, L. Belpassi, D. Zuccaccia, F. Tarantelli,
2
A. Macchioni, J. Organomet. Chem. 2010, 695, 2679–2686.
plexes, and the “classical” non-coordinating anions (such as
[
11] a) G. Ciancaleoni, L. Belpassi, F. Tarantelli, D. Zuccaccia, A. Macchioni,
Dalton Trans. 2013, 42, 4122–4131; b) D. Zuccaccia, L. Belpassi, L. Roc-
chigiani, F. Tarantelli, A. Macchioni, Inorg. Chem. 2010, 49, 3080–3082.
12] D. Zuccaccia, L, Belpassi, F. Tarantelli, A. Macchioni, Eur. J. Inorg. Chem.
ꢀ
BF ) may be not always the best choice. Intermediate coordi-
4
ꢀ
nating and basicity power of the anion OTs provides, under
[
[
[
[
the specific experimental conditions employed, the best com-
promise to achieve an efficient catalyst: the pre-equilibrium
with this anion is shifted toward the OSIP, deactivation of cata-
lyst to a gold–methoxide is prevented, and its characteristic
basicity promotes the nucleophilic attack. This study clearly
demonstrates that the anion can play an effective role in differ-
ent steps of the catalytic cycle, including the overall important
2
013, 4121–4135.
13] J. H. Teles in Modern Gold Catalyzed Synthesis (Eds.: A. S. K. Hashmi, F. D.
Toste) Wiley-VCH, Weinheim, 2012, pp. 201–236.
14] J. H. Teles, S. Brode, M. Chabanas, Angew. Chem. 1998, 110, 1475–1478;
Angew. Chem. Int. Ed. 1998, 37, 1415–1418.
15] a) E. Mizushima, K. Sato, T. Hayashi, M. Tanaka, Angew. Chem. 2002, 114,
4745–4747; Angew. Chem. Int. Ed. 2002, 41, 4563–4565; b) N. Marion,
R. S. Ramꢃn, S. P. Nolan, J. Am. Chem. Soc. 2009, 131, 448–449; c) J.
Oliver-Meseguer, J. R. Cabrero-Antonino, I. Domꢄnguez, A. Leyva-Pꢁrez,
A. Corma, Science 2012, 338, 1452–1455.
[
35]
nucleophilic attack. A more comprehensive study, both com-
putational and experimental (including high-throughput exper-
imentation methods), of the anion’s role in the mechanistic
steps are clearly required and are underway in our laboratory.
[
16] a) A. Zhdanko, M. M. Maier, Chem. Eur. J. 2014, 20, 1918; b) J. Roithovꢅ,
S. Jankovꢆ, L. JaSꢄkovꢆ, J. Vꢆ nˇ a, S. Hybelbauerovꢆ, Angew. Chem. 2012,
124, 8503–8507; Angew. Chem. Int. Ed. 2012, 51, 8378–8382; c) Y. Oo-
nishi, A. Gꢇmez-Suꢅrez, A. R. Martin, S. P. Nolan, Angew. Chem. 2013,
ˇ
ˇ
1
25, 9949–9953; Angew. Chem. Int. Ed. 2013, 52, 9767–9771.
[
17] A. Macchioni, Chem. Rev. 2005, 105, 2039–2074.
[
18] G. Ciancaleoni, L. Biasiolo, G. Bistoni, A. Macchioni, F. Tarantelli, D. Zuc-
caccia, L. Belpassi, Organometallics 2013, 32, 4444–4447.
Acknowledgements
[
[
19] Under our conditions, the formation of 3-hexanone, owing to the pres-
ence of trace water, is less than 2% with respect to the initial concen-
tration of 3-hexyne (see the Supporting Information).
This work was supported by grants from the Italian MIUR and
I
the FIRB-futuro-in-ricerca project: “Novel Au -based molecular
20] E. P. Kꢈndig, C. M. Saudan, G. Bernardinelli, Angew. Chem. 1999, 111,
catalysts: from know-how to know-why (AuCat),” RBFR1022UQ.
1298–1301; Angew. Chem. Int. Ed. 1999, 38, 1219–1223.
&
&
Chem. Eur. J. 2014, 20, 1 – 6
www.chemeurj.org
4
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communication
[
21] A. Kꢈtt, T. Rodima, J. Saame, E. Raamat, V. Memets, I. Kaljurand, I. A.
Koppel, R. Y. Garlyauskayte, Y. L. Yagupolskii, L. M. Yagupolskii, E. Bern-
hardt, H. Willner, I. Leito, J. Org. Chem. 2011, 76, 391–395.
c) R. H. Rang, W. Lai, J. Yao, S. Shaik, H. Chen, J. Chem. Theory Comput.
2012, 8, 3119–3127.
[31] If other functionals, such as M-06 or B2PLYP-D3, are used to compute
the energies, only small differences can be found. Also, the inclusion of
the solvent (COSMO) does not alter the path significantly. For the de-
tails, see the Supporting Information.
[22] T. M. Schmid, G. Consiglio, Chem. Commun. 2004, 2318–2319.
23] M. Raducan, M. Moreno, C. Bour, A. M. Echavarren, Chem. Commun.
[
2012, 48, 52–54.
[
[
24] M. Gꢇmez-Gallego, M. A. Sierra, Chem. Rev. 2011, 111, 4857–4963.
25] E. M. Simmons, J. F. Hartwig, Angew. Chem. 2012, 124, 3120–3126;
Angew. Chem. Int. Ed. 2012, 51, 3066–3072.
[32] A. Zhdanko, M. M. Maier, Angew. Chem. 2014, 126, 7894–7898; Angew.
Chem. Int. Ed. 2014, 53, 7760–7764.
[33] Y. Zhu, C. S. Day, A. C. Jones, Organometallics 2012, 31, 7332–7335.
[34] Notably, all the complexes considered in Figure 2 have the same
number of atoms, minimizing the entropy problem. An estimation of
the entropic contribution can be found in the Supporting Information.
[35] Functionalized ligands also may direct and promote nucleophilic attack.
For example, see: Y. Wang, Z. Wang, Y. Li, G. Wu, Z. Cao, L. Zhang, Nat.
Commun. 2014, 5, 3470.
[26] S. G. Weber, D. Zahner, F. Rominger, B. F. Straub, ChemCatChem 2013, 5,
2
330–2335.
27] T. J. Brown, D. Weber, M. R. Gagnꢁ, R. A. Widenhoefer, J. Am. Chem. Soc.
012, 134, 9134–9137.
[
2
[
[
28] G. Kovꢅcs, G. Ujaque, A. Lledꢇs, J. Am. Chem. Soc. 2008, 130, 853–864.
29] a) M. Bandini, A. Bottoni, M. Chiarucci, G. Cera, G. P. Miscione, J. Am.
Chem. Soc. 2012, 134, 20690–20700; b) J. Kim, S. W. Park, S. R. Park, S.
Lee, E. J. Kang, Chem. Asian J. 2011, 6, 1982–1986.
[
30] a) G. Ciancaleoni, S. Rampino, D. Zuccaccia, F. Tarantelli, P. Belanzoni, L.
Belpassi, J. Chem. Theory Comput. 2014, 10, 1021–1034; b) R. H. Kang,
H. Chen, S. Shaik, J. Yao, J. Chem. Theory Comput. 2011, 7, 4002–4011;
Received: July 23, 2014
Published online on && &&, 0000
Chem. Eur. J. 2014, 20, 1 – 6
www.chemeurj.org
5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
COMMUNICATION
Goldilocks and the six anions: The role
of the anion in the intermolecular alkox-
ylation of alkynes catalyzed by
&
Gold Catalysis
L. Biasiolo, M. Trinchillo, P. Belanzoni,
L. Belpassi, V. Busico, G. Ciancaleoni,
A. D’Amora, A. Macchioni, F. Tarantelli,
D. Zuccaccia*
[
(NHC)AuX] is analyzed with a combined
experimental and theoretical approach.
The anion plays a role in the abstraction
of the proton from the methanol during
the nucleophilic attack, in the coordina-
tion of alkyne, and in the deactivation
of the catalyst. Intermediate coordinat-
&
& – &&
Unexpected Anion Effect in the
Alkoxylation of Alkynes Catalyzed by
N-Heterocyclic Carbene (NHC) Cationic
Gold Complexes
ꢀ
ing power and basicity of the OTs
anion gives the best catalytic per-
formance.
&
&
Chem. Eur. J. 2014, 20, 1 – 6
www.chemeurj.org
6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!