Gold-catalyzed homogeneous oxidative cross-coupling reactions

Full text of DOI:10.1002/anie.200900585

Communications
DOI: 10.1002/anie.200900585
Homogeneous Catalysis
Gold-Catalyzed Homogeneous Oxidative Cross-Coupling Reactions**
Guozhu Zhang, Yu Peng, Li Cui, and Liming Zhang*
Homogeneous gold catalysis has attracted much attention
lately owing to the exceptional capacity of gold complexes/
salts in activating alkynes and allenes toward nucleophilic
attack.[1] A variety of practically useful synthetic methods
using alkyne or allene substrates have been developed.
However, most of these reactions do not involve changes in
the oxidation state of the metal, which is commonly observed
in catalysis by late transition metals such as palladium, nickel,
and rhodium. The combination of a Au(I)/Au(III) catalytic
cycle with contemporary homogeneous gold chemistry based
on alkyne/allene activation would substantially broaden the
field of gold catalysis and offer more functionalized products.
The reduction and oxidation processes catalyzed by gold
nanoparticles or supported gold have been well studied;[2–4]
because of their heterogeneous nature, the reaction mecha-
nisms in many cases are not clear, however, in some studies a
Au(I)/Au(III) catalytic cycle was either proposed or highly
likely to be involved.[5] This catalytic cycle has also been
invoked in gold-catalyzed Suzuki and Sonogashira reactions
using soluble gold catalysts at rather high reaction temper-
atures (1308C), wherein the reaction homogeneity is likely
but not vigorously established.[6] These encouraging results
point to the feasibility of a homogeneous system involving
Au(I)/Au(III) catalytic cycles. Such homogeneous Au(I)/
Au(III) catalysis not only facilitates reaction mechanism
studies but also permits systematic modification of gold
catalysts to improve the reaction scope, selectivity, reactivity,
and efficiency.
reactions are most likely to be homogeneous but proof of
this is lacking. In the latter study by Wegner et al., the
reaction also incorporated a Au(III)-catalyzed cyclization of
aryl propiolates, thus combining a Au(I)/Au(III) catalytic
cycle with contemporary gold chemistry using alkyne sub-
strates. An earlier study by Hashmi et al. proceeded similarly
with Au(III)-catalyzed cyclization of allenyl carbinols and
subsequent dimerization albeit in a substoichiometric fash-
ion.[8a] Although elegant proofs of concepts, these reactions
often resulted in low yields and the dimerization reaction
seriously limited their synthetic potential. To date, no cross-
coupling reactions involving a combination of Au(I)/Au(III)
catalytic cycles and contemporary gold chemistry using
alkyne/allene substrates has been reported. Herein, we
disclose the first example of this type of gold-catalyzed
oxidative cross-coupling reaction, and synthetically useful
a-arylenones are formed in one step; moreover, the homo-
geneity of this chemistry is supported by dynamic light
scattering experiments.
In our attempts to develop a gold-catalyzed synthesis of
a-fluoroenones from propargylic acetates,[9] Selectfluor was
chosen as an electrophilic fluorinating reagent. To our
surprise, when propargylic acetate 1 was treated with
[Ph3PAu]NTf2 (5 mol%) and Selectfluor (1 equiv) in acetone
in a sealed vial at 808C, enone dimer 2 was formed in 19%
yield along with 11% of enone 3[10] and some unreacted
starting material [Eq. (1)]. Although the initial yield was low,
we were intrigued by the reaction and proposed a working
mechanism. As shown in Scheme 1, gold-catalyzed tandem
reactions of propargylic acetate 1, as established by previous
studies by us and others,[10,11] should lead to the formation of
intermediate A upon hydrolysis, which is then oxidized by
Selectfluor to a give a Au(III) species (i.e., B). The trans-
metalation from A to B could lead to the diorganogold(III)
intermediate C, which could then undergo reductive elimi-
nation to give dimer 2 and regenerate the gold(I) catalyst. The
essential oxidation of A to B by Selectfluor is most likely
facilitated by the 1-acylalkenyl ligand, which makes the gold
center relatively electron-rich. The alternative oxidation of
[Ph3PAu]NTf2 to Au(III) is less likely because of its cationic
nature. Notably, other electrophilic fluorine reagents such as
1-fluoropyridinium tetrafluoroborate and N-fluorobenzene-
sulfonimide did not promote this dimerization.
Indeed, gold catalysis involving Au(I)/Au(III) catalytic
cycles have been realized lately in the oxidative dimerization
of non-activated arenes[7a] and aryl propiolates.[7b] These
[*] G. Zhang, Dr. Y. Peng, Dr. L. Cui, Prof. Dr. L. Zhang
Department of Chemistry, University of Nevada, Reno
1664 North Virginia Street, Reno, NV 89557 (USA)
Fax: (+1)775-784-6804
E-mail: lzhang@chem.unr.edu
Homepage: http://wolfweb.unr.edu/homepage/lzhang/
We reason that the organogold(I) A could be supplanted
with external organometallic reagents for transmetalation.
Consequently, cross-coupling product 4, instead of dimer 2,
would be formed by the reductive elimination of intermediate
D. This oxidative cross-coupling reaction would open up a
novel area for gold catalysis and bridge contemporary gold
catalysis, using alkyne/allene substrates, and the well-estab-
lished late transition metal catalyzed cross-coupling reactions.
[**] The authors thank NSF CAREER (CHE 0748484) and Amgen for
generous financial support and Meng Yu for some initial studies.
L.Z. would like to acknowledge the assistance of Jesse Ruppert and
Prof. Olive Graeva of the Nanomaterials Processing Laboratory
(http://people.alfred.edu/~graeve) for the dynamic light scattering
measurements.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200900585.
3112
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 3112 –3115

Angewandte
Chemie
(Table 1, entry 2). The formation of 2 suggested
that the phenyl transfer from the boronate was
not facile enough. While searching for a boro-
nate with a reactivity between that of PhBF3K
and 2-phenyl[1,3,2]dioxaborolane, other boro-
nates were used without much success (Table 1,
entries 3 and 4). Since tetracoordinated, nega-
tively charged boronates are well accepted as
the species undergoing transmetalation in many
transition metal catalyzed reactions,[12] includ-
ing the Suzuki–Miyaura reaction, we reasoned
that such species were the ones undergoing the
phenyl transfer to the Au(III) complex B, and
the low reactivity of dialkoxylphenylboronates
(Table 1, entries 2–4) might be a result of low
concentrations of [PhB(OR)2(OH)]À formed by
the reaction of PhB(OR)2 with a small amount
of H2O in acetonitrile. Consequently, we exam-
ined the effect of H2O using 2-phenyl-
[1,3,2]dioxaborinane. Gratifyingly,
improved yield of 5 was realized by using a
MeCN/H2O (100:1) solvent mixture (Table 1,
a
much
Scheme 1. Proposed reaction mechanism and design.
entry 5). We anticipated that commercially
We chose arylboronates/arylboronic acids as the external
organometallic reagent because of their stability, availability,
and broad spectra of nucleophilicity. With [Ph3PAu]NTf2 as
the catalyst, PhBF3K was highly active for transmetalation,
and as a result biphenyl[5a] was formed predominantly because
of premature phenyl transfer from PhBF3K to [Ph3PAu]NTf2
instead of transfer to B; 3-phenyl-oct-2-en-4-one (5) was not
available phenylboronic acid should also be effective as the
phenyl source as dialkoxyphenylboronates can hydrolyze to
boronic acids. A quick search for the optimal water concen-
tration using phenylboronic acid revealed that a MeCN/H2O
(20:1) solvent mixture offered the best yield as determined by
NMR methods (Table 1, entry 7). Other gold catalysts
including 6 (Table 1, entry 9) and [(4-CF3C6H4)3PAu]NTf2
(Table 1, entry 10) gave similar results. In entry 7 of Table 1,
the major side product was enone 3, likely formed by the
protodeauration of A or B by in situ generated HNTf2. We
reasoned that a weaker in situ gen-
observed
(Table 1,
entry 1).
To
our
delight,
2-phenyl[1,3,2]dioxaborolane gave cross-coupling product 5
albeit in a low yield, and the major product was enone dimer 2
erated acid would minimize this
side reaction. The enone formation
Table 1: Gold-catalyzed cross-coupling reactions: Reaction conditions optimization.[a]
indeed
[(Ph3PAu)3O]H2F3
decreased
by
using
(Table 1,
[13]
entry 11) or Ph3PAuOBz (Table 1,
entry 12). Surprisingly, Ph3PAuCl
catalyzed this cross-coupling reac-
tion exceptionally well, and 5 was
formed in 72% yield as determined
Entry
Catalyst
(5 mol%)
PhBXn
Reaction conditions
Yield [%][c]
5
2
3
1[b]
2[b]
3[b]
4[b]
5
6
7
8
9
10
11
12
13
14[f]
15
[Ph3PAu]NTf2
[Ph3PAu]NTf2
[Ph3PAu]NTf2
[Ph3PAu]NTf2
[Ph3PAu]NTf2
[Ph3PAu]NTf2
[Ph3PAu]NTf2
[Ph3PAu]NTf2
6[e]
[(CF3Ph)3PAu]NTf2
[(Ph3PAu)3O]H2F3
Ph3PAuOBz
Ph3PAuCl
PhBF3K
PhB(OCH2)2
PhB(pin)
MeCN[d]
MeCN
MeCN
MeCN
0
0
60
66
44
17
14
9
10
10
8
0
0
30
21
30
50
25
53
39
54
50
64
59
72
50
25
by
NMR
analysis
(Table 1,
<5
<5
9
13
23
24
<5
21
11
5
entry 13). The reaction could not
be additionally improved by using
additives such as LiF or by using
fewer equivalents of the reagents
(Table 1, entry 14). The role of
Ph3PAuCl, which is normally cata-
lytically inert, in this reaction was
examined. The treatment of
PhB[O(CH2)3O]
PhB[O(CH2)3O]
PhB(OH)2
PhB(OH)2
PhB(OH)2
PhB(OH)2
PhB(OH)2
PhB(OH)2
PhB(OH)2
PhB(OH)2
PhB(OH)2
PhB(OH)2
MeCN/H2O 100:1
MeCN/H2O 100:1
MeCN/H2O 20:1
MeCN/H2O 5:1
MeCN:H2O 20:1
MeCN/H2O 20:1
MeCN/H2O 20:1
MeCN/H2O 20:1
MeCN/H2O 20:1
MeCN/H2O 20:1
MeCN/H2O 20:1
8
8
9
9
Ph3PAuCl
with
Selectfluor
6
15
14
(20 equiv) in MeCN/H2O (20:1) at
808C for 15 minutes gave a clear
Ph3PAuCl
AuCl3
16
yellow solution and, moreover,
[a] Reaction run in a flask using MeCN distilled over CaH2. The reaction concentration was 0.05m.
[b] Reaction run in a 7 mL vial with HPLC-grade MeCN. [c] Estimated by 1H NMR analysis using diethyl
phthalate as internal reference. [d] Reaction time: 40 min. [e] [(2-Biphenyl)Cy2PAu]NTf2. [f] Used
1.5 equiva of Selectfluor and 3 equiv of PhB(OH)2.
[14]
=
Ph3P O
was detected by both
31P NMR and ES+ MS analyses.
Although the nature of the gold
Angew. Chem. Int. Ed. 2009, 48, 3112 –3115
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3113

Communications
species generated needs to be studied additionally, this yellow
The side products for this oxidative cross-coupling reac-
tion were enones, enone dimers, and a-fluoroenones, all of
which can not be completely suppressed. Importantly, all the
cross-coupling reactions proceeded with excellent E selectiv-
ity, and the Z isomers were not detected.
solution was indeed catalytically active, affording 5 in 60%
yield (NMR). At this stage the gold catalyst for the initial
conversion of propargylic substrate 1 into intermediate A,
whether it be Au(I) or Au(III), is not clear as our previous
work[10,11] showed that both species were viable. Finally, AuCl3
did catalyze this reaction albeit with low efficiency (Table 1,
entry 15).
By using the optimized conditions (Table 1, entry 13) the
reaction scope was first studied by varying propargylic
acetates. As shown in Table 2, a range of acetates having
various functionalized or nonfunctionalized alkyl groups on
Importantly, no precipitates were observed during these
reactions, and the reaction mixtures were either colorless or
slightly yellow solutions, indicating the nonexistence of gold
nanoparticles, which tend to make the solution purple. To
provide additional support for the homogeneous nature of the
reaction, we performed dynamic light scattering experiments
on the reaction mixture using 1 as the substrate.[15] Although a
Microtrac Nanotrac ULTRA having a particle size detection
limit of 0.8 nm was used, the filtered reaction solution[16]
which was still catalytically active,[17] could not be effectively
measured because of the lack of light scattering.[18] This
negative result strongly supports the homogeneous nature of
this oxidative cross-coupling reaction.
Table 2: Gold-catalyzed oxidative cross-coupling reactions: Scope.[a]
In summary, an unprecedented homogeneous gold-cata-
lyzed oxidative cross-coupling of propargylic acetates and
arylboronic acids has been developed, leading to a one-step
synthesis of a-arylenones. Whereas the reaction mechanism
deserves additional investigation, this chemistry strongly
suggests the feasibility of Au(I) and Au(III) catalytic cycles,
and indicates that oxidants such as Selectfluor can readily
oxidize Au(I) into Au(III). Moreover, this cross-coupling
reaction reveals for the first time the synthetic potential of
incorporating Au(I)/Au(III) catalytic cycles into gold chemis-
try and promises a new area of gold research by merging
powerful contemporary gold catalysis, using alkyne/allene
substrates, and oxidative metal-catalyzed cross-coupling reac-
tions.
Entry
R
R’
ArB(OH)2
8 (Yield [%])
1
2
3
4
5
6
7
8
Ph
iPr
Me
Me
n-butyl
n-butyl
Ph
PhB(OH)2
PhB(OH)2
PhB(OH)2
8a (62)
8b (65)
8c (59)
8d (60)
8e (68)
8 f (70)
8g (70)
8h (59)
8i (61)
8j (61)
8k (72)
MeOCH2CH2 PhB(OH)2
Me
cyclohexyl
cyclohexyl
n-butyl
PhB(OH)2
PhB(OH)2
PhB(OH)2
PhB(OH)2
PhB(OH)2
PhB(OH)2
p-MePhB(OH)2
cyclohexyl
PhCH2CH2
p-BrC6H4
AcOCH2CH2 n-butyl
H
cyclohexyl
cyclohexyl
cyclohexyl
cyclohexyl
n-butyl
9
10
11
12
13
14
cyclohexyl
n-butyl
n-butyl
n-butyl
n-butyl
p-MeO2CPhB(OH)2 8l (57)[b]
p-ClPhB(OH)2 8m (58)
m-MeO2CPhB(OH)2 8n (45)[c]
[a] Reactions run in flasks using MeCN distilled over CaH2. The reaction
concentration was 0.05m. [b] CH3CN/H2O=100:1. [c] MeCN/H2O=
200:1.
Experimental Section
General procedure for gold-catalyzed oxidative cross-coupling reac-
tion: Ph3PAuCl (0.0075 mmol, 3.7 mg) was added into a solution of
the propargylic acetate substrate (0.15 mmol), Selectfluor (2 equiv),
and an arylboronic acid (4 equiv) in MeCN and water (anhydrous
MeCN/water= 20:1 unless otherwise specified) under N2. The
reaction was heated at 808C for the indicated time and then cooled
to room temperature. The reaction mixture was treated with aqueous
Na2S2O3 (5% w/w, 10 mL) and extracted using diethyl ether (3 ꢀ
10 mL). The combined organic layers were washed with brine
(10 mL), dried with MgSO4, and filtered. The filtrate was concen-
trated under vacuum, and the residue was purified by silica gel flash
column chromatography to afford the cross-coupling product.
both ends of the propargyl moiety were tolerated, yielding
2-phenylenones 8b, 8d–8g, and 8i in respectful yields.
Similarly, substrates having aryl substituents such as phenyl or
4-bromophenyl groups at either end of the propargyl moiety
reacted smoothly, yielding 8a, 8c, and 8h in fairly good yields.
In addition, b-unsubstituted-a-phenylenone 8j was obtained
in 61% yield from a substrate without substitution at the
propargylic position. The scope of arylboronic acids was
examined next. Electron-withdrawing groups including ester
and chloro moieties were tolerated in the meta (Table 2,
entry 14) or para (Table 2, entries 12 and 13) positions of the
phenyl ring, affording a-arylenones 8l–8n in mostly accept-
able yields; notably, less water was used for the ester
substrates so as to minimize enone formation (Table 2,
entries 12 and 14). In addition, p-tolylboronic acid (Table 2,
entry 11) reacted very well. The more electron-rich
4-methoxyphenylboronic acid did not afford any of the
desired product, likely because of the incompatibility of the
anisole ring with strongly oxidative Selectfluor. Ortho-sub-
stituents including methyl, MeO, and Cl were generally
detrimental to this reaction, suggesting that this reaction is
sensitive to the steric bulk on the aromatic ring.
Received: February 1, 2009
Published online: March 25, 2009
Keywords: homogeneous catalysis · cross-coupling · gold ·
.
oxidation · alkynes
[1] For selected reviews on gold catalysis, see: a) A. S. K. Hashmi,
M. Rudolph, Chem. Soc. Rev. 2008, 37, 1766; b) A. Arcadi,
Chem. Rev. 2008, 108, 3266; c) Z. G. Li, C. Brouwer, C. He,
Chem. Rev. 2008, 108, 3239; d) D. J. Gorin, B. D. Sherry, F. D.
Toste, Chem. Rev. 2008, 108, 3351; e) R. Skouta, C.-J. Li,
Tetrahedron 2008, 64, 4917; f) E. Jimꢁnez-Nfflæez, A. M. Echa-
3114
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 3112 –3115

Angewandte
Chemie
varren, Chem. Rev. 2008, 108, 3326; g) A. Fürstner, P. W. Davis,
Angew. Chem. 2007, 119, 3478; Angew. Chem. Int. Ed. 2007, 46,
3410; h) A. S. K. Hashmi, Chem. Rev. 2007, 107, 3180; i) L.
Zhang, J. Sun, S. A. Kozmin, Adv. Synth. Catal. 2006, 348, 2271.
[2] For a review on Au-catalyzed oxidations and reductions includ-
ing homogeneous examples, see reference [1h]; for a review on
gold-catalyzed oxidation, see: D. T. Thompson, Top. Catal. 2006,
38, 231.
[3] For recent studies, see: a) A. Grirrane, A. Corma, H. Garcia,
Science 2008, 322, 1661; b) B. P. Taduri, S. M. A. Sohel, H.-M.
Cheng, G.-Y. Lin, R.-S. Liu, Chem. Commun. 2007, 2530; c) J.
Gong, C. B. Mullins, J. Am. Chem. Soc. 2008, 130, 16458.
[4] For selected gold-catalyzed oxidations, see: a) B. Guan, D. Xing,
G. Cai, X. Wan, N. Yu, Z. Fang, L. Yang, Z. Shi, J. Am. Chem.
Soc. 2005, 127, 18004 – 18005; b) Y. Liu, F. Song, S. Guo, J. Am.
Chem. Soc. 2006, 128, 11332 – 11333; c) B. P. Taduri, S. M. A.
Sohel, H.-M. Cheng, G.-Y. Lin, R.-S. Liu, Chem. Commun. 2007,
2530 – 2532.
[10] a) M. Yu, G. Li, S. Wang, L. Zhang, Adv. Synth. Catal. 2007, 349,
871; b) N. Marion, P. Carlqvist, R. Gealageas, P. de Fremont, F.
Maseras, S. P. Nolan, Chem. Eur. J. 2007, 13, 6437.
[11] For the very first study of this tandem process, see: a) L. Zhang,
J. Am. Chem. Soc. 2005, 127, 16804 – 16805; for our additional
studies on this system, see: b) L. Zhang, S. Wang, J. Am. Chem.
Soc. 2006, 128, 1442; c) S. Wang, L. Zhang, J. Am. Chem. Soc.
2006, 128, 8414; d) S. Z. Wang, L. M. Zhang, Org. Lett. 2006, 8,
4585; for related studies by others, see: e) T. Schwier, A. W.
Sromek, D. M. L. Yap, D. Chernyak, V. Gevorgyan, J. Am.
Chem. Soc. 2007, 129, 9868; f) J. Zhao, C. O. Hughes, F. D. Toste,
J. Am. Chem. Soc. 2006, 128, 7436; g) N. Marion, S. Diez-
Gonzalez, P. de Fremont, A. R. Noble, S. P. Nolan, Angew.
Chem. 2006, 118, 3729; Angew. Chem. Int. Ed. 2006, 45, 3647;
h) A. Buzas, F. Gagosz, J. Am. Chem. Soc. 2006, 128, 12614.
[12] D. G. Hall, Boronic Acids: Preparation and Applications in
Organic Synthesis and Medicine, Wiley-VCH, Weinheim, 2005.
[13] T. Mathieson, A. Schier, H. Schmidbaur, Z. Naturforsch. B 2000,
55, 1000.
[5] a) S. Carrettin, J. Guzman, A. Corma, Angew. Chem. 2005, 117,
2282; Angew. Chem. Int. Ed. 2005, 44, 2242; b) R. de Souza,
M. S. Bittar, L. V. P. Mendes, C. M. F. da Silva, V. T. da Silva,
O. A. C. Antunes, Synlett 2008, 1777.
[14] Ph3PO is likely formed by hydrolysis of the product of Ph3P
oxidation by Selectfluor.
[15] For details, see the Supporting Information.
[6] For gold-catalyzed cross-coupling reactions with substrates as
oxidants, see: a) C. Gonzµlez-Arellano, A. Corma, M. Iglesias, F.
Sanchez, J. Catal. 2006, 238, 497; b) C. Gonzµlez-Arellano, A.
Abad, A. Corma, H. Garcia, M. Iglesias, F. Sanchez, Angew.
Chem. 2007, 119, 1558; Angew. Chem. Int. Ed. 2007, 46, 1536;
c) J. Han, Y. Liu, R. Guo, J. Am. Chem. Soc. 2009, 131, 2060 –
2061.
[7] a) A. Kar, N. Mangu, H. M. Kaiser, M. Beller, M. K. Tse, Chem.
Commun. 2008, 386; b) H. A. Wegner, S. Ahles, M. Neuburger,
Chem. Eur. J. 2008, 14, 11310.
[8] For dimerizations without external oxidants with substoichio-
metric or stoichiometric amount of Au(III), see: a) A. S. K.
Hashmi, M. C. Blanco, D. Fischer, J. W. Bats, Eur. J. Org. Chem.
2006, 1387; b) A. K. Sahoo, Y. Nakamura, N. Aratani, K. S. Kim,
S. B. Noh, H. Shinokubo, D. Kim, A. Osuka, Org. Lett. 2006, 8,
4141.
[16] Although the reaction mixture is clear, filtration before the
measurement using a Millex syringe-driven filter unit (450 nm
diameter) is necessary to get rid of the contaminating dust;
otherwise, the measurement gave particle sizes of approximately
700 nm, which must result from the dust particles, as gold
particles of this size should be visible.
[17] Importantly, the filtrate was still catalytically active, and an
additional portion of propargylic acetate 1 was converted into
product 2 in the presence of additional Selectfluor and phenyl-
boronic acid without the addition of more of the gold catalyst.
[18] Upon forcing the instrument to perform the measurement, the
result was 2.50 nm. As a control experiment, we performed the
same measurement of the reaction solvent (filtered), which was
also used for zeroing, and an identical number (i.e., 2.50 nm) was
obtained, suggesting that the instrument could not effectively
give negative results.
[9] M. Yu, G. Zhang, L. Zhang, Org. Lett. 2007, 9, 2147.
Angew. Chem. Int. Ed. 2009, 48, 3112 –3115
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3115