Gold (I) and (III) catalyze Suzuki cross-coupling and homocoupling, respectively

Article abstract of DOI:10.1016/j.jcat.2005.12.015

Trinuclear triphenylphosphine 1,2Au(I) complexes with the [N,N,O]-tridentate unsymmetrical Schiff bases (1,2) catalyze the Suzuki cross-coupling reaction to afford nonsymmetrical biaryls in good yields, whereas the 1,2Au(III) complexes give only arylboronic homocoupling.

Full text of DOI:10.1016/j.jcat.2005.12.015

Journal of Catalysis 238 (2006) 497–501
www.elsevier.com/locate/jcat
Priority Communication
Gold (I) and (III) catalyze Suzuki cross-coupling
and homocoupling, respectively
C. González-Arellano a,c, A. Corma b,∗, M. Iglesias c, F. Sánchez a
a
Instituto de Química Orgánica General, CSIC, C/Juan de la Cierva 3, 28006 Madrid, Spain
b
Instituto de Tecnología Química, UPV-CSIC, Universidad Politécnica de Valencia, Avda. de los Naranjos s/n, 46022 Valencia, Spain
c
Instituto de Ciencia de Materiales de Madrid, CSIC, C/Sor Juana Inés de la Cruz s/n, Cantoblanco, 28049 Madrid, Spain
Received 25 October 2005; revised 7 December 2005; accepted 12 December 2005
Available online 18 January 2006
Abstract
Trinuclear triphenylphosphine 1,2Au(I) complexes with the [N,N,O]-tridentate unsymmetrical Schiff bases (1,2) catalyze the Suzuki cross-
coupling reaction to afford nonsymmetrical biaryls in good yields, whereas the 1,2Au(III) complexes give only arylboronic homocoupling.
© 2005 Elsevier Inc. All rights reserved.
Keywords: Gold catalysts for carbon–carbon bond formation; Gold(I) catalysts for Suzuki cross-coupling; Gold(III) catalysts for homocoupling C–C formation
1. Introduction
boron reagent—the Suzuki reaction—has emerged as an in-
teresting reaction because of the mild reaction conditions
required [11]. Pd–phosphine complexes [12] have been the
most commonly used catalysts for the Suzuki reaction. How-
ever, alternative ligands such as N-heterocyclic carbenes [13],
imidazol-2-ylidenes [14], and diazabutadienes [15] have been
used in Suzuki coupling reactions. Very recently, we have
shown that phenolic Shiff-base complexes of Au(III) catalyze
the homocoupling of aryl boronic acids, but were unable to cat-
alyze the Suzuki cross-coupling [10]. Because Pd(0) is quite
an effective catalysts for the above reaction, and Au(I) has the
same d10 electronic configuration that Pd(0), we considered
that Au(I) complexes may be active catalysts for performing
the Suzuki cross-coupling.
Gold was believed to be chemically inert, and with very few
exceptions [1] the possibilities of homogeneous and heteroge-
neous gold catalysts for organic reactions were not considered.
Recent reports have shown that gold salts, specifically Au(III),
can act as a Lewis acid catalyst for a large variety of reac-
tions [2]. Solid gold catalysts, on the other hand, can be recy-
cled, and when prepared in the form of smaller clusters (particle
size ꢀ5 nm) are highly active and selective for reactions such as
CO oxidation in the presence of H2 [3,4], water–gas shift [5],
alcohol oxidations [6], and some C–C forming reactions [7].
Gold in homogeneous complexes [8] has shown possibilities
for several catalytic reactions, including asymmetric aldol con-
densation, and the asymmetric hydrogenation of alkenes and
imines [9]. More specifically, it has been recently shown [10]
that Au(III) complexes were active and selective catalysts for
carrying out the homocoupling of different aryl boronic acids.
However, Au(III), unlike Pd complexes, were not able to cat-
alyze cross-coupling reactions.
In the present work, new types of Au(I) complexes were
prepared and characterized. They were found to be active and
selective for catalyzing Suzuki cross-coupling, whereas Au(III)
complexes with same ligands can only catalyze the arylboronic
homocoupling.
Transition metal-catalyzed cross-coupling reactions are use-
ful tools in organic synthesis for constructing C–C bonds.
Among these, the coupling of an aryl halide with an organo-
2. Experimental
2.1. Catalyst preparation
*
2-tert-Butyl-4-methyl-6-[({[(2S)-1-(2-naphthylmethyl)pyr-
rolidinyl])methyl}-imino)methyl]-phenol (1) was used as lig-
Corresponding author. Fax: +34 96 3877809.
E-mail address: acorma@itq.upv.es (A. Corma).
0021-9517/$ – see front matter © 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2005.12.015

498
C. González-Arellano et al. / Journal of Catalysis 238 (2006) 497–501
Fig. 1. Ligands and proposed structure for Au(I)-complexes.
and with Au(I) and Au(III) (Fig. 1). In this type of potentially
tridentate ligands, there are two types of nitrogen donor atoms
and one oxygen phenolate [16]. The nitrogen on the pyrrolidine
ring is expected to act as a hemilabile ligand, while the Schiffs
base nitrogen and oxygen act as the principal donors. Along
with ligand 1, a binaphthyl derivative (2) was also examined
for comparison.
The preparation of metal complexes was carried out under
dinitrogen by conventional Schlenk tube techniques. Solvents
were carefully degassed before use, and the AuCl(PPh3) was
prepared as described previously [17]. Au(I) complexes were
synthesized following Eqs. (1) and (2):
2.1.1. [(C28H34N2)O(Au(PPh3)3)]Cl2 [1Au(I)]
Color: orange brown. Yield: >85%. m.p: 100 ◦C (dec.).
ΛM > 200 ꢀ−1 cm2 mol−1. C82H78Au3Cl2N2OP3 calcd. for
(1862.2): C, 52.9; H, 4.2; N, 1.5; P, 5.0; Au, 31.7; found:
C, 52.8; H, 4.7; N, 2.0; P, 5.2; Au, 32.1%. IR (KBr, cm−1):
ν = 1629, 1602 (C=N, C=C). UV–vis (λ, nm): 424, 375,
1
337, 296, 267, 231. H RMN (CD3OD, ppm): δ = 8.24 (H,
s, CH=N); 7.80–6.96 (m, Harom); 4.04–3.45 (4H, m, CH2Bz
,
ꢀ
H5,5 ); 3.53–3.49 (1H, m, H2); 3.00–2.46 (2H, m, CH=NCH2);
ꢀ
ꢀ
2.25–1.88 (7H, m, CH3, H4,4 , H3,3 ); 1.39 (9H, s, C(CH3)3).
13C RMN (CD3OD, ppm): δ = 169.17 (C–O–Au); 158.63
(CH=N); 135.66, 133.54, 131.16, 130.41, 130.24, 128.97
(Carom); 67.30 (C2); 30.89 (C(CH3)3); 30.32 (C(CH3)3); 24.16
(CH3); 21.05 (C4). 31P NMR (CD3Cl, ppm): δ = 33.79. EM
(m/z): 1377 [(AuPPh3)3]+, 415 (L+1).
(C28H34N2)O−K+ + AuCl(PPh3)
ꢀ
ꢁ
→ (C28H34N2)O(Au(PPh3)3) [Cl]2 + KCl,
(C32H30N2)O−K+ + AuCl(PPh3)
(1)
(2)
2.1.2. [(C32H28N2)O(Au(PPh3)3]Cl [2Au(I)]
ꢀ
ꢁ
Color: orange. Yield: >90%. m.p: 170 ◦C (dec.). ΛM
=
→ (C32H28N2)O(Au(PPh3)3) Cl + KCl.
119 ꢀ−1 cm2 mol−1. C86H73Au3ClN2OP3 calcd. for (1869.79):
C, 55.2; H, 3.9; N, 1.5; P, 5.0; Au, 31.6; found: C, 54.8; H, 3.6;
N, 1.1; P, 4.6; Au, 31.9%. IR (KBr, cm−1): ν = 1615 (C=N,
C=C). UV–vis (λ, nm): 453, 413, 375, 321, 294, 266, 235. 1H
Then AuCl(PPh3) (1 mmol) in THF was added to a solution of
the phenolate ligand (1 mmol) in THF (20 ml) at 40 ◦C. The re-
sultant mixture was stirred for 2 h, cooled to room temperature,
filtered, and then concentrated under vacuum. The residue was
extracted with dichloromethane and precipitated with pentane,
washed several times, filtered, and dried to afford the respec-
tive yellow brown, air-stable solids complexes in high yields.
Properties, analytical and characteristics of the complexes are
as follows.
RMN (CD3Cl, ppm): δ = 7.93–7.38 (14H + 45H, m, Harom
,
CH=N); 1.40 (3H, s, CH3); 1.29 (9H, s, C(CH3)3). 13C RMN
(CD3Cl, ppm): δ = 168.1 (C–O–Au); 152.63 (CH=N); 135.35,
134.48, 130.56 (Carom); 33.90 (C(CH3)3); 33.46 (C(CH3)3);
20.50 (CH3). 31P NMR (CD3Cl, ppm): δ = 33.79. EM (m/z):
1833 (M+–Cl), 1377 [(AuPPh3)3+], 459 [AuPPh3]+.

C. González-Arellano et al. / Journal of Catalysis 238 (2006) 497–501
499
IR spectra were recorded with a Bruker IFS 66v/S spec-
3. Results and discussion
trophotometer (range, 4000–200 cm−1) in KBr pellets. 1H
NMR, 13C NMR, 31P NMR spectra were taken with Varian
XR300 and Bruker 200 spectrometers. Chemical shifts being
referred to tetramethylsilane (internal standard) or to PO4H3
(31P).
The complexes were identified by elemental analysis, MS,
and nuclear magnetic resonance (NMR) spectroscopy. The
mass spectrum (by electro-spray mass spectrometry) shows the
ion molecular, [AuPPh3+], [(AuPPh3)3+], and peaks from the
fragments due to elimination of the phosphine ligand. The high
stability of these complexes originates from the donor capac-
ity of the imine nitrogen atom assisting in the bonding of the
second gold atom [18].
Fourier transform infrared spectra (Fig. 2a) are characteris-
tic of the binding of imine nitrogen. The 1600 cm−1 frequen-
cies can be assigned to C=C and azomethine C=N vibrations,
shifted to lower wavenumbers (relative to the free ligands) due
to N-coordination of the imine [19]. The absence of the ν(OH)
2.2. Catalytic experiments
The reaction was carried out in a 25-ml vessel at 130 ◦C
for 20–180 min. In a typical run, a mixture of aryl halide
(10 mmol), boronic acid (15 mmol), aqueous potassium car-
bonate or phosphate (20 mmol), and catalyst (0.3 mmol) in
3 ml of o-xylene was stirred for the desired time. The solution
was allowed to cool, and a 1:1 mixture of ether/water (20 ml)
was added. The organic layer was washed, separated, further
washed with another 10 ml portion of diethyl ether, dried with
anhydrous MgSO4, and filtered. The solvent and volatiles were
completely removed under vacuum to give the crude product,
which, when subjected to column chromatographic separation,
resulted in pure compounds. The reaction was followed by
gas chromatography–mass spectroscopy (GC–MS). GC analy-
ses were performed in a Hewlett-Packard 5890 II with a flame
ionization detector and/or a Hewlett-Packard G1800A with a
quadrupole mass detector using a cross-linked methylsilicone
column.
band (present in the spectra of the free ligands at ∼3430 cm−1
)
is in accordance with loss of the –OH proton. The infrared
spectra also show bands assigned to the phosphine ligand. New
bands in the 500–600 cm−1 region are ascribed to ν(M–O).
The electronic absorption spectra for all complexes (Fig. 2b)
were obtained using 10−3–10−5 M dichloromethane solutions
in the 200–800 nm range. The complexes show several band
maxima in the UV region agre∗eing with∗the assignment of
the bands as intraligand π → π , n → π , transitions in the
aromatic ring, azomethine group, and charge-transfer transi-
tion [20]. The higher molar extinction coefficient of the bands in
(a)
(b)
Fig. 2. (a) FTIR and (b) electronic absorption spectra for Au(I) complexes.

500
C. González-Arellano et al. / Journal of Catalysis 238 (2006) 497–501
Table 1
Influence of ligands on Au(I) reactivity for Suzuki reactions: Ph–I + Ar–
B(OH) → Ph–Ar + Ar–Ar
2
1Au(I)
2Au(I)
a
a
Ar
Conversion (%) Selectivity (%) Conversion (%) Selectivity (%)
Ph
80 (2)
41 (3)
30 (3)
–
84 (2)
48 (2)
53 (3)
49 (3)
–
3-BrPh
4-BrPh
50
70
92
85
86
88
86
b
4-MeOPh 44 (3)
3-HCOPh 40 (2)
4-HCOPh 25 (1.5)
61 (1.5)
42 (1.5)
100
100
100
Ratio catalyst: IPh = 1:30, IPh (1 mmol), aryl boronic (1.5 mmol), K PO
3
4
(2 mmol).
a
Cross-coupling/conversion.
b
Reaction with BrPh yields only homocoupling product. Reaction with
K CO as base gives 18% cross-copling with selectivity 100%. In parenthe-
2
3
sis, reaction time in hours.
the 260–350 nm range may be due to the ∗coincidence of charge
transfer, d → π∗, and intraligand π → π , n → π∗ transitions.
The bands in the 400–500 nm region correspond to d–d transi-
tions and MLCT bands.
The diamagnetic gold complexes presented here were char-
1
acterized by H, 13C, and 31P NMR spectroscopy (data given
in the Catalyst preparation section). All assignments are based
on several correlations in the two-dimensional spectra. In all
cases, the spectra show the simultaneous occurrence of two set
of signals, which are assigned on the one hand to the substituted
benzaldimine entity and on the other hand to the aliphatic (1) or
aromatic (for 2) part of the ligand. In the 1H NMR spectra, all
of the resonances are high field shifted compared with the un-
coordinated ligand and are in agreement with metallation of the
ligand with coordination of the metal atom via the imine nitro-
gen atom. The most noticeable shifts involve HC=N, C=N–
Fig. 3. Suzuki reaction results for Au(I) complexes.
4-MeOPh, 4-MePh, 3- and 4-HCOPh) and aryl halides (BrPh,
IPh) was studied. Au(I) complexes with phenyl bromide are in-
effective, yielding the homocoupling compound as a principal
product. On the other hand, gold (I) complex with ligand 1,
which is less electron-rich than 2, gives lower activity toward
the Suzuki reaction (Table 1). The effects of electron-donating
and electron-withdrawing substituents in the boronic acid on
reactivity are also given in Table 1. In general, moderate to ex-
cellent yields and good selectivity toward the cross-coupling
product are found with Au(I) complexes (Fig. 3). The best re-
activity is found with 3- or 4-formylphenylboronic acid as the
substrate. It is noteworthy that when 1,2Au(III) complexes are
used as catalysts for several of the reactions, only homocou-
pling of boronic acids is observed (Fig. 3) [9].
For comparison purposes, 2Pd(II) and Pd(PPh3)4 complexes
were prepared and tested under the same reaction conditions
(Table 2). Results (Tables 1 and 2) show that activity and se-
lectivity of 2Au(I) and Pd complexes are similar, although the
Pd(PPh3)4 is systematically more active. Nevertheless, it is in-
teresting to see that Au(I), which has the same d10 electronic
structure as Pd0, is able to catalyze the cross-coupling reaction
with high selectivity. On the other hand, changing the charge of
the gold atom [Au(III)] makes it possible to selectively catalyze
the homocoupling reaction.
ꢀ
ꢀ
CH2N–, H2 , and H5 . Deprotonation of the –OH group was
confirmed by the absence of OH resonances in the 1H spectra.
The resonance corresponding to the HC=N proton appears as
a singlet (δ = 8.5–7.2 ppm) shifted to high field due to the coor-
dination of the imine group to the metal atom through the lone
1
pair of the nitrogen atom [21]. The H NMR spectrum shows
the signals of the PPh3 protons as a multiplet at δ = 7–8 ppm.
The 13C NMR spectra show the signals assigned to the C=N
carbon high field shifted and C1 at δ ≈ 162 downfield shifted,
confirming that metallation occurred. 31P NMR spectra show
a signal at δ ≈ 34 ppm. Structures of complexes 1Au(I) and
2Au(I) that agree with the spectroscopic characterization are
presented in Fig. 1. Linear structures in which the gold is trinu-
clear can be observed.
The Au(I) complexes were studied for the Suzuki cross-
coupling of Br–Ph and I–Ph with a series of aryl boronic acids.
The standard reaction conditions were applied to a series of
catalysts; the conversions are tabulated in Table 1. In our exper-
iments, we chose K3PO4 as mild base, because with K2CO3,
longer reaction times are needed to obtain reasonable conver-
sions.
The general utility of the reaction conditions with a vari-
ety of aryl boronic acid substrates (aryl=Ph, 3- and 4-BrPh,

C. González-Arellano et al. / Journal of Catalysis 238 (2006) 497–501
501
Table 2
(c) J. Guzman, B.C. Gates, J. Am. Chem. Soc. 126 (2004) 2672;
Suzuki reactions with the Pd complexes as catalysts Ph–I + Ar–B(OH)
→
(d) A. Luengnaruemitchai, D.T.K. Thoa, S. Osuwan, E. Gulari, Int. J. Hy-
drogen Energy 30 (2005) 981.
2
Ph–Ar + Ar–Ar
[5] J. Hua, Q. Zheng, Y. Zheng, K. Wei, X. Lin, Catal. Lett. 102 (2005) 99.
[6] (a) A. Corma, A. Abad, P. Concepción, H. García, Angew. Chem. Int.
Ed. 44 (2005) 4066;
(b) F. Porta, L. Prati, M. Rossi, G. Scari, J. Catal. 211 (2002) 464.
[7] (a) S. Carrettin, J. Guzmán, A. Corma, Angew. Chem. Int. Ed. 44 (2005)
2242;
Pd complexes
Pd(PPh )
Pd(II)
3 4
a
a
Ar
Conversion (%) Selectivity (%) Conversion (%) Selectivity (%)
Ph
3-BrPh
4-BrPh
4-MeOPh 66 (1.5)
3-HCOPh 78 (1.5)
4-HCOPh 69 (1.5)
100 (1)
66 (3)
71 (1.5)
100
88
96
75
100
100
90 (1.5)
–
–
48 (0.5)
68 (3)
39 (3)
89
12
100
100
–
–
(b) A.S.K. Hashmi, L. Schwarz, J.-H. Choi, T.M. Frost, Angew. Chem.
Int. Ed. 39 (2000) 2285.
[8] (a) A.S.K. Hashmi, J.P. Weyrauch, M. Rudolph, E. Kurpejovic, Angew.
Chem. Int. Ed. 43 (2004) 6545;
(b) A.S.K. Hashmi, Gold Bull. 37 (2004) 51, and references therein.
[9] (a) Y. Ito, M. Sawamura, T. Hayashi, J. Am. Chem. Soc. 108 (1986) 6405;
(b) C. González-Arellano, A. Corma, M. Iglesias, F. Sánchez, Chem. Com-
mun. (2005) 3451.
a
Ratio catalyst:IPh = 1:30, IPh (1 mmol), aryl boronic (1.5 mmol), K PO
3
4
(2 mmol). In parenthesis, reaction time in hours
4. Conclusion
[10] C. González-Arellano, A. Corma, M. Iglesias, F. Sánchez, Chem. Com-
mun. (2005) 1990.
Two new Au(I)-Schiff base complexes have been synthe-
sized and characterized. These are efficient catalysts for Suzuki
coupling reactions of iodobenzene and several aryl boronic
acids to give nonsymmetrical biaryls. Their activity and selec-
tivity is similar to the corresponding Pd complexes, indicating
that Au(I) with the same d10 configuration as Pd(0) can catalyze
reactions typically catalyzed by palladium. On the other hand,
the same Shiff base complexes of Au(III) catalyze arylboronic
homocoupling with 100% selectivity.
[11] (a) J. Hassan, M. Sévignon, Ch. Gozzi, E. Schulz, M. Lemaire, Chem.
Rev. 102 (2002) 1359;
(b) A. Suzuki, J. Organomet. Chem. 576 (1999) 147;
(c) N. Miyaura, A. Suzuki, Chem. Rev. 95 (1995) 2457;
(d) A. Suzuki, in: F. Diederich, P.T. Stang (Eds.), Metal-Catalysed Cross-
Coupling Reactions, Wiley–VCH, Weinheim, 1998, pp. 49–97.
[12] (a) D.W. Old, J.P. Wolfe, S.L. Buchwald, J. Am. Chem. Soc. 120 (1998)
9722;
(b) J.P. Wolfe, S.L. Buchwald, Angew. Chem. Int. Ed. 38 (1999) 2413;
(c) J.P. Wolfe, R.A. Singer, B.H. Yang, S.L. Buchwald, J. Am. Chem.
Soc. 121 (1999) 9550;
(d) J. Yin, S.L. Buchwald, J. Am. Chem. Soc. 122 (2000) 12051;
(e) J. Yin, M.P. Rainka, X.-X. Zhang, S.L. Buchwald, J. Am. Chem.
Soc. 124 (2002) 1162;
Acknowledgments
(f) T.E. Barder, S.D. Walker, J.R. Martinelly, S.L. Buchwald, J. Am.
Chem. Soc. 127 (2005) 4685.
The authors thank Ministerio de Educación y Ciencia
(project MAT2003-07945-C02-01 and 02) and the Auricat EU
Network (HPRN-CT-2002-00174) for financial support.
[13] (a) V.P.W. Bohm, C.W.K. Gstottmayr, T. Weskamp, W.A. Herrmann,
J. Organomet. Chem. 595 (2000) 186;
(b) O. Navarro, R.A. Kelly III, S.P. Nolan, J. Am. Chem. Soc. 125 (2003)
16194.
References
[14] C. Zhang, J. Huang, M.L. Trudell, S.P. Nolan, J. Org. Chem. 64 (1999)
3804.
[1] (a) G.J. Hutchings, Catal. Today 100 (2005) 55;
(b) G.J. Hutchings, Gold Bull. 37 (2004) 3.
[15] G.A. Grasa, A.C. Hilier, S.P. Nolan, Org. Lett. 3 (2001) 1077.
[16] C. González-Arellano, E. Gutiérrez-Puebla, M. Iglesias, F. Sánchez, Eur.
J. Inorg. Chem. (2004) 1955.
[2] (a) A. Corma, H. García, Chem. Rev. 103 (2003) 4037;
(b) Ch. Wei, Ch.-J. Li, J. Am. Chem. Soc. 125 (2003) 9584;
(c) X. Yao, Ch.-J. Li, J. Am. Chem. Soc. 126 (2004) 6884;
(d) G. Dyker, Angew. Chem. Int. Ed. 39 (2000) 4237, and references
therein.
[17] M.I. Bruce, B.K. Nicholson, O.B. Shawkataly, Inorg. Synth. 26 (1989)
325.
[18] A. Kolb, P. Bissinger, H. Schmidbaur, Inorg. Chem. 32 (1993) 5132.
[19] (a) H. Onoue, I. Moritani, J. Organomet. Chem. 43 (1972) 431;
(b) H. Onoue, K. Minami, K. Nakagawa, Bull. Chem. Soc. Jpn. 43 (1970)
3480.
[3] (a) M. Haruta, T. Kobayashi, H. Sano, N. Yamada, Chem. Lett. (1987)
405;
(b) M. Haruta, N. Yamada, T. Kobayashi, S. Iijima, J. Catal. 115 (1989)
301.
[20] T.N. Sorrell, W.E. Allen, P.S. White, Inorg. Chem. 34 (1995) 952.
[21] Y. Ustynyuk, V.A. Chertov, J.V. Barinov, J. Organomet. Chem. 29 (1971)
C53.
[4] (a) B.C. Gates, J. Guzman, Angew. Chem. Int. Ed. 42 (2003) 690;
(b) S. Carrettin, P. Concepción, A. Corma, J.M. López-Nieto, V.F. Puntes,
Angew. Chem. Int. Ed. 43 (2004) 2538;