Copper salts as additives in gold(I)-Catalyzed reactions

Article abstract of DOI:10.1002/anie.201300600

The right combination: Cu^I and Cu^II salts can advantageously replace silver additives in Au^I-catalyzed reactions. On the basis of reactivity studies and NMR experiments, it is believed that anion metathesis between CuY_n (Y=OTf, BF₄, PF₆, SbF₆) and [R₃PAuCl] takes place to give [R ₃PAu]Y. As this process is slow, there is no fast decay of the active species, thus allowing large-scale reactions, even at high temperatures, with low loadings of the gold complex. Copyright

Full text of DOI:10.1002/anie.201300600

Angewandte
Chemie
DOI: 10.1002/anie.201300600
Gold Catalysis
Copper Salts as Additives in Gold(I)-Catalyzed Reactions**
Amandine Guꢀrinot, Weizhen Fang, Marie Sircoglou, Christophe Bour, Sophie Bezzenine-
Lafollꢀe,* and Vincent Gandon*
Over the past 15 years, gold catalysis has become an essential
tool in organic chemistry.[1] In particular, spectacular trans-
formations have been reported in the field of homogeneous
gold(I)-catalysis,[2] the usefulness of which has been demon-
strated through various total syntheses of natural products.[3]
The overwhelming majority of gold(I)-catalyzed reactions
involve [LAu][Y] as active species, where Y is a weakly
reasoned that a gradual (possibly reversible) delivery of
[LAu]+ from a reservoir of stable [LAuX] should prevent
rapid decomposition of all of the gold. This does not seem to
be possible with silver salts, as the precipitation of AgX makes
this step very fast and irreversible. We herein report that
copper salts address these issues.
During the course of our studies on gold-catalyzed
hydroalkylations of unactivated alkenes,[11] we decided to
evaluate the influence of some additives. Surprisingly, we
observed that a catalytic mixture of [(JohnPhos)AuCl] (A;
10 mol%) and Cu(OTf)·0.5C6H6 (10 mol%) was able to
catalyze the transformation of the ene-b-ketoamide 1 into the
cyclized product 2 (Table 1, entry 1). No silver salt was added
coordinating anion, L = phosphine, phosphite, phosphorami-
À
dite, carbene, or other species, and Y= TfOÀ, (RO)2PO2
,
Tf2NÀ, BF4À, PF6À, SbF6À, or other species (Tf = trifluoro-
methanesulfonyl). These electrophilic compounds are usually
generated by anion metathesis between LAuX (X = Cl, Br)
and a silver salt (AgY), and are not necessarily isolated.[4] In
spite of the possible interference of silver with the catalytic
process,[5] such two-component systems are widely used,
because both gold halide precursors and silver salts are
readily available and quite easy to handle. Although the vein
of gold-catalyzed synthesis seems inexhaustible, some prob-
lems associated with the use of fragile cationic complexes
remain to be solved. For example, the classical [Ph3PAu]+ (or
other phosphine–gold cationic intermediates) may rapidly
decay to give gold(0) (as mirror, precipitate, or nanoparticles)
and inactive [(Ph3P)2Au]+.[6] Although gold nanoparticles can
sometimes be extremely active,[7] in most cases, the reduction
of [LAu]+ means deactivation. As [LAu]+ is normally stable in
solution, the decomposition process is attributed to the
interaction between the cationic complex and a reductive
substrate such as an alkyne, allene, or alkene.[6–8] Thus, some
AuI-catalyzed reactions remain limited in terms of scalability,
catalyst loading, and temperature range. Most of them have
been carried out with a few milligrams of starting material and
attempts to work at larger scale have resulted in significant
erosion of the yields.[9] To ensure a sufficient concentration of
active species, a proportion of catalyst that exceeds 1 mol% is
usually necessary. Lastly, high temperatures accelerate the
decomposition process, especially above 808C, which is
detrimental to the development of energetically demanding
transformations. To circumvent these problems, work has
been focused on the ligands used in AuI catalysis.[6,10] On our
side, we envisaged to play on the anion metathesis itself. We
Table 1: Screening of the Au/Cu catalytic system in intramolecular
hydroalkylation of ene-b-ketoamide 1.
Entry
[Au]
[Cu]
Conv. [%][a]
d.r.[b]
1
2
3
4
5
6
7
8
A
CuOTf·0.5C6H6
CuOTf·0.5C6H6
(MeCN)4CuPF6
(MeCN)4CuBF4
Cu(OTf)2
(MeCN)5Cu(SbF6)2
CuCl2
Cu(OAc)2
(MeCN)4CuPF6
(MeCN)4CuPF6
(MeCN)4CuBF4
Cu(OTf)2
Cu(OTf)2
(MeCN)5Cu(SbF6)2
CuCl2
100
15
0
80:20
n.d.
–
n.d.
n.d.
–
none
none
none
none
none
none
none
A
B
A
A
B
20
15[c]
[d]
–
0
0
100
76
–
–
9
85:15
87:13
82:18
95:5
88:12
–
10
11
12
13
14
15
16
100
100
100
[d]
A
A
A
–
0
0
–
–
Cu(OAc)2
[a] Estimated by 1H NMR spectroscopy; 100% Conversion corresponds
to an 80% yield of isolated product in all cases. [b] Estimated by 1H NMR
spectroscopy, ratios shown are trans/cis. [c] No reaction if JohnPhos is
added. [d] Decomposition. Ac=acetyl, n.d.=not determined, OTf=tri-
fluoromethanesulfonate.
[*] Dr. A. Guꢀrinot, W. Fang, Dr. M. Sircoglou, Dr. C. Bour,
Dr. S. Bezzenine-Lafollꢀe, Prof. Dr. V. Gandon
ICMMO (UMR CNRS 8182), LabEx CHARMMMAT Universitꢀ
Paris-Sud, 91405 Orsay (France)
in the reaction mixture. To the best of our knowledge, no
report mentions the joint use of a neutral AuI complex and
a copper salt.[12,13] Intrigued by this finding, we decided to test
various catalytic systems. At first, all the experiments were
carried out at 1108C in toluene for 20 h on a 0.11 mmol scale
(30 mg). After checking that A alone was not able to catalyze
the cyclization, various CuI and CuII complexes were eval-
E-mail: vincent.gandon@u-psud.fr
[**] We thank ANR JCJC HAONA, UPS, CNRS, and IUF for financial
support. We also thank Dr. Mohamed Mellah for useful discussions.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201300600.
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
Table 3: Comparison of Au/Cu and Au/Ag catalytic systems.
uated. Without gold, no or low conversions were observed
(entries 2–8).[14,15] In contrast, when 10 mol% of gold complex
A or [Ph3PAuCl] (B) was used, the transformation worked
well in most cases (entries 9–13), except with (MeCN)5Cu-
(SbF6)2, which promoted decomposition (entry 14), and with
CuCl2 and Cu(OAc)2, which proved inactive (entries 15 and
16). In terms of diastereomeric ratio, the best result was
obtained with the A/Cu(OTf)2 system (entry 12).[16] The gold-
free JohnPhos/Cu(OTf)2 catalytic mixture does not promote
any reaction.
Entry
L
Y
Amount of 1 [g] Time [h] Conv. [%] d.r.[a]
1
2
3
4
5
6
A
A
B
A
A
B
Cu(OTf)2 0.030[b]
Cu(OTf)2 2[c]
Cu(OTf)2 1[e]
2
1
1.5
2
1
100
95:5
88:12
95:5
n.d.
n.d.
–
100[d]
100[f]
28
AgOTf
AgOTf
AgOTf
0.030[b]
2[c]
Focusing next on the [Ph3PAuCl]/Cu(OTf)2 catalytic
system, we noticed that, with the same catalytic loading, full
conversion of 1 could be reached in 1 h at 1108C in toluene
(Table 2, entry 1). When using 1 mol% of [Ph3PAuCl] and
20[g,h]
0[d]
1[e]
1.5
[a] Estimated by 1H NMR spectroscopy, ratios shown are trans/cis.
[b] 0.11 mmol. [c] 7.25 mmol. [d] 78% yield of isolated product (1.56 g).
[e] 3.63 mmol. [f] 70% yield of isolated product (0.70 g). [g] 25%, 28%,
32% after 4, 7, and 24 h, respectively. [h] Same conditions, but protected
from light.
Table 2: Effects of complex loading, temperature, and reaction time.
Entry
x
y
T [8C]
Time [h]
Conv. [%][a]
d.r.[b]
1
2
3
4
5
6
7
10
1
1
0.1
1
1
10
10
5
10
10
10
10
110
110
110
110
80
1
1
1
3
2
4
3
100
95
63
100
92
95
86:14
80:20
80:20
95:5
80:20
>95:5[c]
n.d.
Scheme 1. Enantioselective hydroalkylation of an unactivated olefin.
DTBM-Segphos=(4,4’-bi-1,3-benzodioxole)-5,5-diylbis(di(3,5-di-tert-
butyl-4-methoxyphenyl)phosphine).
80
50
1
<5
[a] Estimated by 1H NMR spectroscopy. [b] Estimated by 1H NMR
spectroscopy, ratios shown are trans/cis. [c] Minor diastereomer not
detected in the 1H NMR spectrum.
interest, the Au/Cu method is also compatible with the chiral
biarylphosphine digold complex [(R)-DTBM-Segphos-
(AuCl)2] (Scheme 1). A catalytic 1:2 mixture of this complex
and Cu(OTf)2 transformed ene-b-ketoamide 3 into the
diastereomeric mixture 4.[18] An enantiomeric excess of
59% was measured for the major diastereomer. To the best
of our knowledge, this unoptimized reaction represents the
first example of a catalytic asymmetric hydroalkylation of an
unactivated olefin.
How gold and copper activate the substrate is a puzzling
question. The formation of active nanoparticles is unlikely,
because neither the purple color that often occurs with the
[Ph3PAuCl]/AgX systems nor a precipitate were observed
when using Au/Cu mixtures. Also, as discussed below, the
presence of inactive [(Ph3P)2Au]+ could hardly be detected by
31P NMR. Besides, the formation of the enantioenriched
product 4 rules out the sole intervention of (Au)n nano-
particles. To gain some insight, the reaction of 1 under the
experimental conditions of Table 1 (toluene, 1108C,
[Ph3PAuCl]/Cu(OTf)2 10 mol% each), was monitored by
NMR spectroscopy (Figure 1). After 2 h, although full con-
version of 1 was reached, [Ph3PAuCl] remained the major
gold species (d31P = 33 ppm). Only a very small peak at d =
44 ppm could also be observed. After 5 days at 1108C, the two
peaks showed similar integration. The chemical shift of d =
44 ppm actually corresponds to the inactive [(Ph3P)2Au]-
[OTf].[7,19] Pertaining to the absence of reactivity shown in
Table 3, entry 6, this complex forms quickly and quantita-
10 mol% of Cu(OTf)2, 95% conversion was obtained after
1 h (entry 2). Reduction of the loading of the copper complex
to 5 mol% affected the rate of the transformation (entry 3).
The proportion of [Ph3PAuCl] could be reduced to 0.1 mol%,
with full conversion being reached in 3 h (entry 4). In this case
the diastereomeric ratio was greater than 95:5, which is due to
an equilibration between the trans and the cis isomers (see
below). The reaction still proceeded at 808C but was slower
(entries 5 and 6), and it virtually did not work at 508C
(entry 7).
Reduction of the loading of [(JohnPhos)AuCl] (A) was
also tested. With 0.1 mol%, complete conversion required 2 h
at 1108C (Table 3, entry 1). Again, decreasing the proportion
of Cu(OTf)2 to 5 mol% resulted in prolonged reaction time
(20 h). Interestingly, the transformation could be carried out
on a 2 g scale (entry 2). In this case, the concentration of 1 was
higher than that in the 30 mg scale experiment (0.1m vs. 0.5m),
resulting in a higher reaction rate.[17] Even with [Ph3PAuCl]
(B) and Cu(OTf)2, full conversion was reached after 1.5 h (1 g
scale, entry 3).
The transformations catalyzed by the Au/Cu system
reported in Table 3 could be carried in an open vessel using
undistilled toluene without affecting the yields or the
diastereoselectivities. When AgOTf was used instead of
Cu(OTf)2, low or no conversions were measured with A
(entries 4 and 5) or B (entry 6), respectively. Of particular
2
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
Figure 2. Formation and reactivity of the gold C-enolate 6. Chemical
shifts shown are for 31P NMR spectroscopy.
Figure 1. 31P NMR spectra of the reaction of 1 catalyzed by [Ph3PAuCl]/
Cu(OTf)2 in toluene after 2 h and 5 days at 1108C, and reference
spectra of [Ph3PAuCl]/AgOTf and [Ph3PAuCl] in toluene recorded in the
presence of 1 after 1 h at 1108C.
The 31P spectrum recorded after 30 min showed some
remaining [Ph3PAuCl] and a new peak at d = 37 ppm. A
13C DEPT 135 experiment confirmed the disappearance of
a CH2 signal and showed a new CH doublet (2JCP = 53 Hz).
The 31P NMR peak of 6 disappeared after 20 h at room
temperature, replaced by two new peaks at d = 45 and
46 ppm, which were attributed to the two diastereomers of
the alkylgold complex 7.[26] As protodeauration did not occur
spontaneously, PTSA·H2O (PTSA = para-toluenesulfonic
acid) was added to the mixture, leading to 2 in poor yield
(28%).[27]
Next, we questioned the formation of 6 under the reaction
conditions. For this purpose, we used b-ketoamide 9, which
cannot cyclize (Scheme 3). To obtain the NMR data for the
Scheme 2. Neutral and cationic pathways to rationalize the Au/Cu-
catalyzed enantioselective hydroalkylation of 1.
tively when mixing [Ph3PAuCl], AgOTf, and 1. No other gold
complex, such as [Ph3PAu][OTf] (expected at d =
26 ppm),[19,20] could be detected.[21,22]
From this set of data, two mechanistic scenarios can be
proposed (Scheme 2), both starting from the copper O-
enolate 5.[23] Following the neutral path a, nucleophilic attack
on [Ph3PAuCl] gives rise to the gold C-enolate 6.[24] Insertion
Scheme 3. Unlikely formation of a gold C-enolate under the reaction
conditions. Chemical shifts shown are for 31P NMR spectroscopy.
À
of the alkene into the C Au bond furnishes 7 (6-exo-trig
cyclization) and then 2 after protodemetallation.[25] Alterna-
tively, 7 could be formed following the cationic path b, which
consists of the coordination of Ph3PAu+ to the alkene to give
8, which triggers the nucleophilic attack of the copper enolate.
As 6 could not be observed, we envisaged that its
cyclization and subsequent protodemetallation were fast
processes. To verify this hypothesis, 6 was prepared inde-
pendently by treating 1 with NaH and [Ph3PAuCl] (Figure 2).
corresponding C-enolate 10, compound 9 was treated with
NaH and [Ph3PAuCl]. The 31P NMR spectrum showed a peak
at d = 37 ppm, and the 13C DEPT 135 experiment clearly
indicated the disappearance of one CH2 and the appearance
of a CH doublet, as before. In a second experiment, 9 was
treated with one equivalent of Cu(OTf)2 (1 equiv) and then
[Ph3PAuCl] (1 equiv). After two hours at 1108C, only the
peak corresponding to [Ph3PAuCl] and [(Ph3P)2Au][OTf]
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
were observed, the latter species being a minor component of
the mixture.
Information for details). No reaction occurred in the sole
presence of Cu(OTf)2 or TfOH. [(JohnPhos)AuCl] could be
used jointly with (MeCN)4CuPF6, but full conversion required
a longer reaction time of 18 h. When (MeCN)4CuPF6 was
used alone, only 10% conversion was monitored after 18 h.
Lastly, enyne 15 was transformed into the formal [4+2]
cycloadduct 16 using [(JohnPhos)AuCl] (0.1 mol%) and
Cu(OTf)2 (10 mol%) at 508C in 1,2-dichloroethane. Yield
and diastereoselectivity are comparable to those reported for
the same reaction using [(JohnPhos)Au(MeCN)][SbF6]
(2 mol%).[33]
These results indicate that path b is more likely. We have
seen that with the Au/Ag system, the quantitative instant
production of catalytically active [Ph3PAu][OTf] is rapidly
followed by its deactivation, which is caused by the organic
substrate in the form of [(Ph3P)2Au][OTf] and (Au)n. This is
probably the reason why the conversion of 1 is low and the
active species not even visible in the 31P spectrum. With the
Au/Cu catalytic mixture, we assume that the concentration of
[Ph3PAu][OTf] remains very low, actually too low to allow this
complex to be detected (Figure 1). If anion metathesis
eventually occurs as we suppose, reactions that involve
activation of a single chemical function should also be
possible with the Au/Cu system. Thus, a few other typical
gold-catalyzed transformations of alkynes were studied
(Scheme 4). We first turned our attention to hydration:[28]
In conclusion, the Au/Cu system allows the use of
standard gold complexes such as [Ph3PAuCl] and avoids fast
decay of the active species. On the basis of NMR studies, we
suggest that the replacement of the strongly coordinating
chlorine anion for a weakly coordinating one (OTfÀ, BF4
,
À
À
PF6 ) is possible with copper salts, yet it is slow and thus the
production of active gold species is gradual. Therefore,
scalable reactions at high temperature with low loading of
the gold complex become possible. Although we do not
pretend that the Au/Cu catalytic system is systematically the
best choice, we have shown that it can outperform the Au/Ag
system in some cases and that various solvents can be used
(toluene, MeOH/H2O, 1,2-dichloroethane). New opportuni-
ties now await, notably cascade reactions that would exploit
the two metals. Also, the promise of enantioselective hydro-
alkylations of unactivated olefins, as shown in Scheme 1,
represents an interesting new tool in asymmetric synthesis.
Our findings in these areas will be reported in due course.
Received: January 23, 2013
Revised: February 28, 2013
Published online: && &&, &&&&
Keywords: catalysis · copper · cyclization · gold · hydroalkylation
.
Scheme 4. Au/Cu-catalyzed transformations of alkynes. DCE=1,2-
dichloroethane, E=CO2Me.
[1] Modern Gold-Catalyzed Synthesis (Eds.: A. S. K. Hashmi, F. D.
Toste), Wiley-VCH, Weinheim, 2012.
p-tolylacetylene 11 was treated with [(JohnPhos)AuCl]
(0.1 mol%) and Cu(OTf)2 (5 mol%) for 18 h at 808C in
a 2:1 MeOH/H2O mixture. Gratifyingly, the desired methyl-
ketone 12 was obtained in a 93% yield. No reaction took
place with the copper salt alone. As TfOH could also be
formed in appreciable amount under these experimental
conditions,[29] it was tested as a catalyst (10 mol%), but only
14% of the alkyne was converted in this case. Thus, the two
metals are needed for the effective formation of the hydration
product. When the proportion of [(JohnPhos)AuCl] was
decreased to 0.01 mol%, a good conversion (73%) was
measured after 18 h.[30] On the other hand, a poor conversion
(5%) was observed when using AgOTf instead of Cu(OTf)2.
We next evaluated the Au/Cu system in enyne cycloisomeri-
zation.[31] Compound 13 was treated with [(JohnPhos)AuCl]
(0.1 mol%) or [Ph3PAuCl] (0.1 mol%) and Cu(OTf)2
(5 mol%), giving rise to the five-membered ring product 14
in 80% yield after 3 h in 1,2-dichloroethane.[32] Although this
reaction takes place at room temperature, 19F NMR studies
support the anion metathesis scenario (see the Supporting
[2] a) M. Rudolph, A. S. K. Hashmi, Chem. Commun. 2011, 47,
6536; b) A. S. K. Hashmi, Angew. Chem. 2010, 122, 5360; Angew.
Chem. Int. Ed. 2010, 49, 5232; c) S. Dꢀez-Gonzꢁlez, N. Marion,
S. P. Nolan, Chem. Rev. 2009, 109, 3612; d) A. Fꢂrstner, Chem.
Soc. Rev. 2009, 38, 3208; e) D. J. Gorin, F. D. Toste, Nature 2007,
446, 395.
[3] See Chapter 15 in Ref. [1] and a) M. Rudolph, A. S. K. Hashmi,
Chem. Soc. Rev. 2012, 41, 2448; b) S. Sengupta, X. Shi,
ChemCatChem 2010, 2, 609.
[4] The treatment of alkyl-gold complexes such as [LAuMe] with
strong Brønsted acid is an alternative method; see Chapter 8 in
Ref. [1].
[5] Silver can play a positive or a negative role, see: a) D. Wang, R.
Cai, S. Sharma, J. Jirak, S. K. Thummanapelli, N. G. Akhmedov,
H. Zhang, X. Liu, J. L. Petersen, X. Shi, J. Am. Chem. Soc. 2012,
134, 9012; b) H. Schmidbaur, A. Schier, Z. Naturforsch. B 2011,
66, 329.
[6] See Chapter 8 in Ref. [1] and W. Wang, G. B. Hammond, B. Xu,
J. Am. Chem. Soc. 2012, 134, 5697.
[7] J. Oliver-Meseguer, J. R. Cabrero-Antonino, I. Domꢀnguez, A.
Leyva-Pꢃrez, A. Corma, Science 2012, 338, 1452.
4
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
[8] G. Lemiꢄre, V. Gandon, N. Agenet, J.-P. Goddard, A. de Kozak,
C. Aubert, L. Fensterbank, M. Malacria, Angew. Chem. 2006,
118, 7758; Angew. Chem. Int. Ed. 2006, 45, 7596.
[18] Although 4 is also effectively formed with the [Ph3PAuCl]/
Cu(OTf)2 catalytic system, no reaction takes place when 3 is
treated with [Ph3PAuCl] (5 mol%), Cu(OTf)2 (10 mol%), and
(R)-DTMB-Segphos (15 mol%). The gold-free Cu(OTf)2/(R)-
DTMB-Segphos promotes no reaction either.
[9] For a rare example of scalable AuI-catalyzed transformation,
see: E. Mizushima, D.-M. Cui, D. C. D. Nath, T. Hayashi, M.
Tanaka, Org. Synth. 2006, 83, 55.
[19] J. Zhang, C.-G. Yang, C. He, J. Am. Chem. Soc. 2006, 128, 1798.
[20] M. Preisenberger, A. Schier, H. Schmidbaur, J. Chem. Soc.
Dalton Trans. 1999, 1645.
[10] See inter alia: a) A. S. K. Hashmi, J. P. Weyrauch, M. Rudolph,
E. Kurpejovic, Angew. Chem. 2004, 116, 6707; Angew. Chem. Int.
Ed. 2004, 43, 6545; b) N. Marion, R. S. Ramꢅn, S. P. Nolan, J.
Am. Chem. Soc. 2009, 131, 448; c) H. Teller, M. Corbet, L.
Mantilli, G. Gopakumar, R. Goddard, W. Thiel, A. Fꢂrstner, J.
Am. Chem. Soc. 2012, 134, 15331; d) V. Lavallo, J. H. Wright II,
F. S. Tham, S. Quinlivan, Angew. Chem. 2013, 125, 3254; Angew.
Chem. Int. Ed. 2013, 52, 3172.
[21] Although oxidation by CuII could have occurred, no species
corresponding to a AuIII complex could be detected.
[22] 31P NMR study in the JohnPhos series was not informative
because of the narrow range in the chemical shifts (see the
Supporting Information).
[23] The presence of the paramagnetic CuII complex made studies by
1H NMR spectroscopy difficult. However, directly after the
addition of Cu(OTf)2 and [Ph3PAuCl], the formation of the
copper enolate could be observed (see the Supporting Informa-
tion). The NMR spectrum recorded after 15 min at 508C showed
that no cyclized product was formed yet.
[24] a) M. Murakami, M. Inouye, M. Suginome, Y. Ito, Bull. Chem.
Soc. Jpn. 1988, 61, 3649; b) Y. Ito, M. Inouye, M. Suginome, M.
Murakami, J. Organomet. Chem. 1988, 342, C41.
[25] The protodeauration step can be promoted by D2O (10 equiv).
See the Supporting Information for details.
[26] Without the copper salt, heating the mixture at 1108C promoted
decomposition.
[27] R. L. LaLonde, W. E. Brenzovich Jr., D. Benitez, E. Tkatchouk,
K. Kelley, W. A. Goddard III, F. D. Toste, Chem. Sci. 2010, 1,
226.
[28] See Chapter 8 in Ref. [1].
[29] M. J.-L. Tschan, C. M. Thomas, H. Strub, J.-F. Carpentier, Adv.
Synth. Catal. 2009, 351, 2496.
[30] We know no other reaction catalyzed by [(JohnPhos)AuCl] with
such a low catalyst loading. It must be noted, however, that even
lower amounts of catalyst have been reported with the
[(IPr)AuCl]/AgSbF6 system, see Ref. [10b]. However, for similar
substrates, 50–1000 ppm were used (0.005–0.1 mol%).
[31] V. Michelet, P. Y. Toullec, J.-P. GenÞt, Angew. Chem. 2008, 120,
4338; Angew. Chem. Int. Ed. 2008, 47, 4268.
[32] C. Nieto-Oberhuber, M. P. MuÇoz, E. BuÇuel, C. Nevado, D. J.
Cꢁrdenas, A. M. Echavarren, Angew. Chem. 2004, 116, 2456;
Angew. Chem. Int. Ed. 2004, 43, 2402.
[11] For examples of gold(I)-catalyzed hydroalkylation of unacti-
vated alkenes with pronucleophiles, see: a) R.-V. Nguyen, X.-Q.
Yao, D. S. Bohle, C.-J. Li, Org. Lett. 2005, 7, 673; b) C.-Y. Zhou,
C.-M. Che, J. Am. Chem. Soc. 2007, 129, 5828; c) Y.-P. Xiao, X.-
Y. Liu, C.-M. Che, Angew. Chem. 2011, 123, 5039; Angew. Chem.
Int. Ed. 2011, 50, 4937.
[12] CuCl2 has been used as an oxidant to stabilize AuIII in AuCl3 –
CuCl2 catalyzed reactions; see: X. Zhang, A. Corma, Dalton
Trans. 2008, 397.
[13] For rare examples of cooperative Au/Mo or Au/Zn catalysis, see:
a) Y. Wang, L. Liu, L. Zhang, Chem. Sci. 2013, 4, 739; b) A. S.
Demir, M. Emrullahoglu, K. Buran, Chem. Commun. 2010, 46,
8032; c) L. Ye, L. Zhang, Org. Lett. 2009, 11, 3646; d) M. Egi, Y.
Yamaguchi, N. Fujiwara, S. Akai, Org. Lett. 2008, 10, 1867.
[14] For examples of Cu(OTf)2-catalyzed hydroalkylation of alkenes
or alkynes with activated methylene compounds, see: a) Y. Li, Z.
Yu, S. Wu, J. Org. Chem. 2008, 73, 5647; b) C.-L. Deng, T. Zou,
Z.-Q. Wang, R.-J. Song, J.-H. Li, J. Org. Chem. 2009, 74, 412.
[15] For (MeCN)4CuBF4-catalyzed cycloisomerizations, see: C. Fehr,
B. Winter, I. Magpantay, Chem. Eur. J. 2009, 15, 9773.
[16] Cyclic voltammetry experiments indicated no reduction of CuII
into CuI. Additional experiments also show that CuII cannot
rescue the catalytic system once Au0 is formed (see Supporting
Information).
[17] The diastereomeric mixture 2 was composed of trans and cis
compounds in 88:12 ratio, instead of 95:5 for the 0.11 mmol scale
reaction. This is actually not due to the amount of starting
material engaged, but to the different reaction time between the
two experiments. When 2 was treated once again with [(John-
Phos)AuCl] (0.1 mol%) and Cu(OTf)2 (10 mol%), the ratio
reached 95:5 after 2 h at 1108C.
[33] E. Jimꢃnez-NfflÇez, C. K. Claverie, C. Bour, D. J. Cꢁrdenas,
A. M. Echavarren, Angew. Chem. 2008, 120, 8010; Angew.
Chem. Int. Ed. 2008, 47, 7892.
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
Communications
Gold Catalysis
A. Guꢁrinot, W. Fang, M. Sircoglou,
C. Bour, S. Bezzenine-Lafollꢁe,*
V. Gandon*
&&&& — &&&&
Copper Salts as Additives in Gold(I)-
Catalyzed Reactions
The right combination: CuI and CuII salts
can advantageously replace silver addi-
tives in AuI-catalyzed reactions. On the
basis of reactivity studies and NMR
experiments, it is believed that anion
metathesis between CuYn (Y=OTf, BF4,
PF6, SbF6) and [R3PAuCl] takes place to
give [R3PAu]Y. As this process is slow,
there is no fast decay of the active
species, thus allowing large-scale reac-
tions, even at high temperatures, with low
loadings of the gold complex.
6
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
These are not the final page numbers!