Preparation and characterization of two new N-heterocyclic carbene gold(I) complexes and comparison of their catalytic activity to Au(IPr)Cl

Article abstract of DOI:10.1021/om100504z

Gold catalysts with N-heterocyclic carbene ligands have become an attractive tool for organic synthesis. Two new gold complexes (9, 11) are prepared and their structures are determined by X-ray structural analysis. Their catalytic activities have been studied with a new intramolecular cyclopropanation reaction starting from the propargyl esters tethered to a terminal alkene. Although no or low enantioselectivity was observed with 9 or 11, the insight gained from this study is the importance of the steric effects of the NHC ligand.

Full text of DOI:10.1021/om100504z

3450 Organometallics 2010, 29, 3450–3456
DOI: 10.1021/om100504z
Preparation and Characterization of Two New N-Heterocyclic
Carbene Gold(I) Complexes and Comparison of Their Catalytic
Activity to Au(IPr)Cl
Benjamin W. Gung,*,† Lauren N. Bailey,† Derek T. Craft,† Charles L. Barnes,‡ and
Kristin Kirschbaum§
†Department of Chemistry and Biochemistry, Miami University, Oxford, Ohio 45056, ‡Elmer O. Schlemper
X-ray Diffraction Center, University of Missouri;Columbia, Columbia, Missouri 65211, and Department
§
of Chemistry, University of Toledo, Toledo, Ohio 43606
Received May 23, 2010
Gold catalysts with N-heterocyclic carbene ligands have become an attractive tool for organic
synthesis. Two new gold complexes (9, 11) are prepared and their structures are determined by X-ray
structural analysis. Their catalytic activities have been studied with a new intramolecular cyclopro-
panation reaction starting from the propargyl esters tethered to a terminal alkene. Although no or
low enantioselectivity was observed with 9 or 11, the insight gained from this study is the importance
of the steric effects of the NHC ligand.
Introduction
Structurally well-defined Au(NHC)Cl complexes have been
prepared and found to be effective precatalysts comparable
to or better than Au(I)PR3Cl complexes.4-10 We recently
reported a tandem [3,3]-rearrangement/[4Cþ3C]11,12 cyclo-
addition reaction catalyzed by the previously reported Au-
(NHC)Cl/AgSbF6 complex, where NHC = 1,3-bis(diiso-
propylphenyl)imidazol-2-ylidene (IPr), I, Chart 1, developed
by Nolan and co-workers.4 In an effort to conduct the
[4Cþ3C] cycloaddition reaction in an enantioselective man-
ner, we needed chiral Au(NHC)Cl complexes. However,
currently only a few chiral Au(NHC)Cl complexes with
catalytic activity have been reported, in contrast to their
Au(I)PR3Cl counterparts,13,14 which have been successfully
used in enantioselective reactions.15-27 Unfortunately, the
Au(I) salt with a phosphine ligand did not work as well as
N-Heterocyclic carbenes (NHCs) have emerged as a class
of ligands with exceptional properties,1 mainly due to their
outstanding performance in metal-catalyzed reactions.2,3
*To whom correspondence should be addressed. E-mail: gungbw@
muohio.edu.
(1) Hahn, F. E.; Jahnke, M. C. Heterocyclic Carbenes: Synthesis and
Coordination Chemistry. Angew. Chem., Int. Ed. 2008, 47 (17), 3122–
3172.
(2) Baskakov, D.; Herrmann, W. A.; Herdtweck, E.; Hoffmann, S. D.
N-Heterocyclic Carbenes. 49. Chiral N-Heterocyclic Carbenes with
Restricted Flexibility in Asymmetric Catalysis. Organometallics 2007,
26 (3), 626–632.
(3) Cesar, V.; Bellemin-Laponnaz, S.; Gade, L. H. Chiral N-Hetero-
cyclic Carbenes As Stereodirecting Ligands in Asymmetric Catalysis.
Chem. Soc. Rev. 2004, 33 (9), 619–636.
(4) de Fremont, P.; Scott, N. M.; Stevens, E. D.; Nolan, S. P.
Synthesis and Structural Characterization of N-Heterocyclic Carbene
Gold(I) Complexes. Organometallics 2005, 24 (10), 2411–2418.
(5) de Fremont, P.; Marion, N.; Nolan, S. P. Cationic NHC-Gold(I)
Complexes: Synthesis, Isolation, and Catalytic Activity. J. Organomet.
Chem. 2009, 694 (4), 551–560.
(6) Marion, N.; Carlqvist, P.; Gealageas, R.; de Fremont, P.;
Maseras, F.; Nolan, S. P. [(NHC)Au-I]-Catalyzed Formation of Con-
jugated Enones and Enals: An Experimental and Computational Study.
Chem.;Eur. J. 2007, 13 (22), 6437–6451.
(7) Marion, N.; Gealageas, R.; Nolan, S. P. [(NHC)Au-I]-Catalyzed
Rearrangement of Allylic Acetates. Org. Lett. 2007, 9 (14), 2653–2656.
(8) Alonso, I.; Trillo, B.; Lopez, F.; Montserrat, S.; Ujaque, G.;
Castedo, L.; Lledos, A.; Mascarenas, J. L. Gold-Catalyzed [4Cþ2C]
Cycloadditions of Allenedienes, including an Enantioselective Version
with New Phosphoramidite-Based Catalysts: Mechanistic Aspects of the
Divergence between [4Cþ3C] and [4Cþ2C] Pathways. J. Am. Chem.
Soc. 2009, 131, 13020.
(9) Trillo, B.; Lopez, F.; Montserrat, S.; Ujaque, G.; Castedo, L.;
Lledos, A.; Mascarenas, J. L. Gold-Catalyzed [4Cþ3C] Intramolecular
Cycloaddition of Allenedienes: Synthetic Potential and Mechanistic
Implications. Chem.;Eur. J. 2009, 15 (14), 3336–3339.
(10) Nieto-Oberhuber, C.; Lopez, S.; Echavarren, A. M. Intramole-
cular [4þ2] Cycloadditions of 1,3-Enynes or Arylalkynes with Alkenes
with Highly Reactive Cationic Phosphine Au(I) Complexes. J. Am.
Chem. Soc. 2005, 127 (17), 6178–6179.
(11) Gung, B. W.; Craft, D. T.; Bailey, L. N.; Kirschbaum, K. Gold-
Catalyzed Transannular [4þ3] Cycloaddition Reactions. Chem.;Eur.
J. 2010, 16, 639.
(12) Gung, B. W.; Bailey, L. N.; Wonser, J. Gold-Catalyzed Inter-
molecular [4C þ 3C] Cycloaddition Reactions. Tetrahedron Lett. 2010,
51, 2251.
(13) Matsumoto, Y.; Selim, K. B.; Nakanishi, H.; Yamada, K.;
Yamamoto, Y.; Tomioka, K. Chiral Carbene Approach to Gold-
Catalyzed Asymmetric Cyclization of 1,6-Enynes. Tetrahedron Lett.
2010, 51 (2), 404–406.
(14) Matsumoto, Y.; Yamada, K. I.; Tomioka, K. C-2 Sym-
metric Chiral NHC Ligand for Asymmetric Quaternary Carbon Con-
structing Copper-Catalyzed Conjugate Addition of Grignard Reagents
to 3-Substituted Cyclohexenones. J. Org. Chem. 2008, 73 (12), 4578–
4581.
(15) Munoz, M. P.; Adrio, J.; Carretero, J. C.; Echavarren, A. M.
Ligand Effects in Gold- and Platinum-Catalyzed Cyclization of Enynes:
Chiral Gold Complexes for Enantioselective Alkoxycyclization. Orga-
nometallics 2005, 24 (6), 1293–1300.
(16) Gonzalez-Arellano, C.; Corma, A.; Iglesias, M.; Sanchez, F.
Enantioselective Hydrogenation of Alkenes and Imines by a Gold
Catalyst. Chem. Commun. 2005, 27, 3451–3453.
(17) Kadzimirsz, D.; Hildebrandt, D.; Merz, K.; Dyker, G. Isoin-
doles and Dihydroisoquinolines by Gold-Catalyzed Intramolecular
Hydroamination of Alkynes. Chem. Commun. 2006, 6, 661–662.
r
pubs.acs.org/Organometallics
Published on Web 07/13/2010
2010 American Chemical Society

Article
Organometallics, Vol. 29, No. 15, 2010 3451
Chart 1. Initial Screen of Gold Complexes for Intramolecular
[4Cþ3C] Cycloaddition Reactions
Results and Discussion
Our recent success in the gold-catalyzed transannular
[4Cþ3C] cycloaddition reactions11 led us to extend the scope
of the reaction to the intramolecular version. Initially we
screened the gold complexes that have been reported in the
literature, i.e, the Au(I) complexes with each of the follow-
ing ligands: a NHC ligand (I, Au(IPr)Cl),4 a Buchwald
o-biphenylphosphine ligand (II),10 and an electron-deficient
phosphine ligand (III),10 or the Au(III) complex (IV)30
(Chart 1).
Our initial study on attempted gold-catalyzed intramole-
cular [4Cþ3C] cycloaddition reactions was more compli-
cated comparing to that on the transannular version.11 The
scope and limitation of the study have not been completed
for the intramolecular [4þ3] cycloaddition reactions. One
surprising outcome was observed for substrate diene-pro-
pargyl ester 1. No desired product was obtained with up to
20 mol % of any one of the gold complexes II-IV despite the
fact that both complexes II and IV were effective in the
transannular version of the [4þ3] cycloaddition.11 A mixture
of [4þ3] and formal [2þ2] cycloaddition products 2 and 3
was isolated, although in poor yields, when 10% of Au-
(IPr)Cl complex I and an equal amount of AgSbF6 were
employed, eq 1.
Au(I) with a NHC ligand for the tandem 3,3-rearrangement/
[4þ3] cycloaddition reaction.12 Among the Au(I)PR3Cl
complexes used in our study of the transannular [4Cþ3C]
cycloaddition reaction, only the Au(I) catalyst (II, Chart 1)
with a Buchwald ligand worked well.11,10 However even
this catalyst was ineffective for the corresponding inter-
or intramolecular versions of tandem 3,3-rearrangement/
[4Cþ3C] cycloaddition reactions.12 On the other hand, Au-
(IPr)Cl complex I proved to be effective as a precatalyst in
the intermolecular [4Cþ3C] cycloaddition reactions.12
Therefore an effective chiral Au(NHC)Cl complex would
be the most promising chiral Au(I) precatalyst for enantio-
selective inter- and intramolecular [4Cþ3C] cycloaddition
reactions.
After a survey of the literature, we found that, despite an
upsurge of studies on chiral NHC ligands,3,28,29 currently
only a few gold catalysts with a chiral NHC ligand are known
to be catalytically active.13,15 In this report, we disclose the
synthesis, characterization, catalytic reactivity, and thermal
stability of two new chiral Au(NHC)Cl complexes. Although
almost no enantioselectivity was observed in the cyclopro-
panation reactions studied using the two new chiral Au-
(NHC) complexes, important insight into the relationship
between structure and reactivity of Au(NHC) complexes has
been obtained.
In order to understand the possible reaction pathways and
to improve the reaction selectivity and yield, we decided to
first simplify the reaction by excluding the [4Cþ3C] cycload-
dition pathway with the starting material 4a-c. Instead of
the diene moiety in propargyl ester 1, propargyl esters 4a-c
contain a terminal alkene; hence they cannot undergo a [4þ3]
cycloaddition reaction. They were prepared and studied under
the catalysis of Au(IPr)Cl complex I, Scheme 1. Surprisingly, a
smooth cyclopropanation reaction was observed. In the pre-
sence of complex I and an equal amount of AgSbF6, the
propargyl esters 4a-c were converted into bicyclic enol
esters 5a-c in high yield at room temperature. The cyclo-
propanation products (5a-c) were obtained as mixtures of
cis/trans double-bond isomers of the enol acetate, which
were inseparable by column chromatography. To facilitate
the structure identification, the products from the gold-cat-
alyzed cyclopropanation reactions were converted to their
corresponding ketones (6a-c), which allowed unambiguous
identification by NMR spectroscopy.
(18) Zhang, Z.; Widenhoefer, R. A. Gold(I)-Catalyzed Intramolecu-
lar Enantioselective Hydroalkoxylation of Allenes. Angew. Chem., Int.
Ed. 2007, 46 (1þ2), 283–285.
(19) LaLonde, R. L.; Sherry, B. D.; Kang, E. J.; Toste, F. D. Gold(I)-
Catalyzed Enantioselective Intramolecular Hydroamination of Allenes.
J. Am. Chem. Soc. 2007, 129 (9), 2452–2453.
(20) Melhado, A. D.; Luparia, M.; Toste, F. D. Au(I)-Catalyzed
Enantioselective 1,3-Dipolar Cycloadditions of Munchnones with Elec-
tron-Deficient Alkenes. J. Am. Chem. Soc. 2007, 129 (42), 12638–12639.
(21) Liu, C.; Widenhoefer, R. A. Gold(I)-Catalyzed Intramolecular
Enantioselective Hydroarylation of Allenes with Indoles. Org. Lett.
2007, 9 (10), 1935–1938.
(22) LaLonde, R. L.; Sherry, B. D.; Kang, E. J.; Toste, F. D. Gold(I)-
Catalyzed Enantioselective Intramolecular Hydroamination of Allenes.
J. Am. Chem. Soc. 2007, 129 (9), 2452–2453.
(23) Hashmi, A. S. K.; Schaefer, S.; Bats, J. W.; Frey, W.; Rominger,
F. Gold Catalysis and Chiral Sulfoxides: Enantioselective Synthesis of
Dihydroisoindol-4-ols. Eur. J. Org. Chem. 2008, 29, 4891–4899.
(24) Kleinbeck, F.; Toste, F. D. Gold(I)-Catalyzed Enantioselective
Ring Expansion of Allenylcyclopropanols. J. Am. Chem. Soc. 2009, 131
(26), 9178.
(25) Uemura, M.; Watson, L. D. G.; Katsukawa, M.; Toste, F. D.
Gold(I)-Catalyzed Enantioselective Synthesis of Benzopyrans via Re-
arrangement of Allylic Oxonium Intermediates. J. Am. Chem. Soc.
2009, 131 (10), 3464.
(26) Gonzalez, A. Z.; Toste, F. D., Gold(I)-Catalyzed Enantioselec-
tive [4þ2]-Cycloaddition of Allene-dienes. Org. Lett. 12, (1), 200-203.
(27) Teller, H.; Flugge, S.; Goddard, R.; Furstner, A., Enantioselec-
tive Gold Catalysis: Opportunities Provided by Monodentate Phos-
phoramidite Ligands with an Acyclic TADDOL Backbone. Angew.
Chem., Int. Ed. 49, (11), 1949-1953.
(28) Herrmann, W. A.; Goossen, L. J.; Kocher, C.; Artus, G. R. J.
Chiral Heterocyclic Carbenes in Asymmetric Homogeneous Catalysis.
Angew. Chem. Int. Ed. Engl. 1996, 35 (23-24), 2805–2807.
(29) Perry, M. C.; Burgess, K. Chiral N-Heterocyclic Carbene-Tran-
sition Metal Complexes in Asymmetric Catalysis. Tetrahedron: Asym-
metry 2003, 14 (8), 951–961.
Several gold-catalyzed tandem [3,3]-rearrangement/cyclo-
propanation reactions have been reported by the groups of
Furstner,31,32 Gagosz,33 and Toste.34 However, there is a
marked difference in the present reaction substrates and the
(30) Hashmi, A. S. K.; Weyrauch, J. P.; Rudolph, M.; Kurpejovic, E.
Gold Catalysis: The Benefits of N and N,O Ligands. Angew. Chem., Int.
Ed. 2004, 43 (47), 6545–6547.
(31) Furstner, A.; Hannen, P. Carene Terpenoids by Gold-Catalyzed
Cycloisomerization Reactions. Chem. Commun. 2004, 22, 2546–2547.
(32) Mamane, V.; Gress, T.; Krause, H.; Furstner, A. Platinum- and
Gold-Catalyzed Cycloisomerization Reactions of Hydroxylated
Enynes. J. Am. Chem. Soc. 2004, 126 (28), 8654–8655.

3452 Organometallics, Vol. 29, No. 15, 2010
Scheme 1. Intramolecular Cyclopropanation Catalyzed by Au(I)IPr Complex I
Gung et al.
Scheme 2. Preparation of Au(I) Complexes 9 and 11
Table 1. Selected Structural Parameters for the New Gold
Complexes 9-Cl, 11-Cl, 9-OAc, and 11-OAc
products. Previously reported cyclopropanation substrates
have the acetate group located in between the alkyne and the
alkene bonds and the products contain an endocyclic enol
acetate. In the current propargyl esters 4a-c, the ester group
is located on the farther side of the triple bond away from the
alkene. As a result, the products 5a-c have an exocyclic enol
acetate. A recent report of PtCl2-catalyzed cyclopropanation
reactions provided similar reaction products,35 but the start-
ing materials were propargyl ethers and the products con-
tained an enol ether moiety rather than an enol ester.
Au-C(1)
distance (A)
δ
C Au-C(1)
νCdO
complex
9-Cl
11-Cl
9-OAc
11-OAc
Au(IPr)Cl (I)a
(ppm)
(cm-1
)
1.998
2.003
na
na
1.942
191.0
178.4
185.5
177.3
175.1
na
na
1626.7
1627.1
na
a From ref 4.
High yields of 5a-c were obtained with catalyst I regard-
less of the characteristics of the substituent R at the pro-
pargylic position. On the basis of the study of Furstner,36 this
gold-catalyzed cyclopropanation should involve one of two
possible pathways: (1) an initial gold-activated 1,2-acetate
migration and the formation of a gold carbenoid intermedi-
ate followed by cyclopropanation with the terminal alkene,
or (2) an initial enyne cyclization with the terminal alkene
attacking the gold-activated alkyne, leading to cyclopropa-
nation and a gold carbenoid intermediate, followed by 1,2-
migration of the acetate and deauration.36 More discussion
on possible reaction mechanisms will be presented later.
With the results of the cyclopropanation reaction cata-
lyzed by Au(IPr)Cl (I) in hand, our attention turned to the
preparation of chiral Au(NHC) complexes. Among the
widespread interest in the development of better NHC lig-
ands, one of the efforts has been to examine the σ-donor and
π-acceptor properties of the ligand.37 We were interested in
comparing the catalytic activity of the Au(I) complexes, one
with a saturated NHC ligand (9) and another one with an ex-
tended conjugation system, the benzimidazolium ligand (11).
Preparation of Chiral Au(I) Complexes. The preparation of
the chrial Au(I) complexes 9 and 11 is shown in Scheme 2.
The preparation of the imidazolium salt 8 and the benzimi-
dazolium salt 10 follows the general protocol of Roland and
Alexakis,38 and the conversion from 8 and 10 to the Au(I)Cl
complexes 9-Cl and 11-Cl was carried out using the method
of Wang and Lin.39 The most widely accepted methodology
for the determination of a ligand’s electronic properties is the
IR spectroscopic analysis of the CO-stretching frequencies of
(L)Ni(CO)3, (L)Rh(CO)2Cl, (L)Ir(CO)2Cl, and related com-
plexes.40-43 The 13C chemical shifts of the carbene carbon
have been compared throughsubtraction of the corresponding
(33) Buzas, A.; Gagosz, F. Gold(I) Catalyzed Isomerization of
5-En-2-yn-1-yl Acetates: An Efficient Access to Acetoxy Bicyclo-
[3.1.0]hexenes and 2-Cycloalken-1-ones. J. Am. Chem. Soc. 2006, 128
(39), 12614–12615.
(34) Watson, L. D. G.; Ritter, S.; Toste, F. D. Asymmetric Synthesis
of Medium-Sized Rings by Intramolecular Au(I)-Catalyzed Cyclopro-
panation. J. Am. Chem. Soc. 2009, 131 (6), 2056.
(35) Ye, L.; Chen, Q.; Zhang, J. C.; Michelet, V. PtCl2-Catalyzed
Cycloisomerization of 1,6-Enynes for the Synthesis of Substituted
Bicyclo[3.1.0]hexanes. J. Org. Chem. 2009, 74 (24), 9550–9553.
(36) Furstner, A.; Hannen, P. Platinum- and Gold-Catalyzed Re-
arrangement Reactions of Propargyl Acetates: Total Syntheses of (-)-
alpha-Cubebene, (-)-Cubebol, Sesquicarene and Related Terpenes.
Chem.;Eur. J. 2006, 12 (11), 3006–3019.
(38) Winn, C. L.; Guillen, F.; Pytkowicz, J.; Roland, S.; Mangeney,
P.; Alexakis, A. Enantioselective Copper Catalysed 1,4-Conjugate
Addition Reactions Using Chiral N-Heterocyclic Carbenes. J. Organo-
met. Chem. 2005, 690 (24-25), 5672–5695.
(39) Wang, H. M. J.; Lin, I. J. B. Facile synthesis of Silver(I)-Carbene
Complexes. Useful Carbene Transfer Agents. Organometallics 1998, 17
(5), 972–975.
(40) Gusev, D. G. Electronic and Steric Parameters of 76 N-Hetero-
cyclic Carbenes in Ni(CO)(3)(NHC). Organometallics 2009, 28 (22),
6458–6461.
€
(37) Alcarazo, M.; Stork, T.; Anoop, A.; Thiel, W.; Furstner, A.
Steering the Surprisingly Modular π-Acceptor Properties of N-Hetero-
cyclic Carbenes: Implications for Gold Catalysis. Angew. Chem., Int.
Ed. 2010, 49 (14), 2542–2546.
(41) Tolman, C. A. Steric Effects of Phosphorus Ligands in Organo-
metallic Chemistry and Homogeneous Catalysis. Chem. Rev. 1977, 77
(3), 313–348.

Article
Organometallics, Vol. 29, No. 15, 2010 3453
Figure 1. ORTEP diagram of chiral NHC-AuCl complexes: 9-Cl and 11-Cl. Hydrogen atoms in complex 9-Cl have been omitted for
clarity.
Table 2. Catalytic Activities of Au(I) Complexes 9, 11, and I in the Cyclopropanation Reactions of 4a-ca
Z/E-
isomeric
ratio
substrate
LAu(I)Cl
R
time (h)
yield (%)b
4a
4b
4c
4a
4b
4c
4a
4b
4c
9
9
9
11
11
11
I
I
I
Ph
48
2
2
2
3
1.5
2
1
(65% 12a)c
97
95
(56% 12b)c
96
94
90
97
94
NA
2:1
1:1
2:1
2:1
1:1
1:0
3.6:1
2.4:1
n-propyl
isopropyl
Ph
n-propyl
isopropyl
Ph
n-propyl
isopropyl
1
a Low or near-zero optical rotations were observed for all reaction products. b All yields refer to product 5 except where indicated otherwise. c Product
12 was produced only with the specific combination of the gold complex and substrate 4a.
imidazol-2-ylidene carbene by Nolan and co-workers to
study the donor ability of the NHC ligands bound to gold-
(I).4 In an attempt to identify the difference in the chemical
environment surrounding the gold atom in complexes 9 and
11, we hoped to compare the carbonyl stretching frequencies
of complexes 9-OAc and 11-OAc. Their preparation is also
shown in Scheme 2 following the report of Baker and co-
workers.44 Our wish to compare the CdO stretching fre-
quencies of 9-OAc and 11-OAc would turn out to be a bit
consequently its influence diminished. A very small differ-
ence in CdO stretching frequencies was observed (Table 1).
To characterize the gold complexes and to examine the
stereochemical environment of the gold atom, crystals
suitable for X-ray structure analysis were grown of com-
plexes 9-Cl and 11-Cl. The ORTEP diagrams of the struc-
tures for complexes 9-Cl and 11-Cl are shown in Figure 1.
The observed key parameters for the two gold complexes
are compiled in Table 1 along with the previously reported
data for complex I. The observed 13C NMR chemical shift
difference between complexes 9 and complexes 11 are
consistent with previous observations of Au(NHC) com-
plexes.4 The differences in Au-C bond distances between
9-Cl and 11-Cl (1.998 vs 2.003 A) are too small to be
significant. However, the differences in Au-C bond dis-
tances between the two new complexes and complex I
appear to be real (∼2.00 vs 1.942 A). The differences
in the CdO stretching frequencies between 9 and 11 are
also too small to be significant. However, fortuitously, they
appear to parallel the stabilities of the corresponding
cationic Au(I) catalyst. The most stable Au(NHC) complex
is complex I, which has the shortest bond distance. Later we
naıve because the ester CO groups are not bonded directly to
¨
the Au atom (but through an oxygen atom spacer), and
(42) Wurtz, S.; Glorius, F. Surveying Sterically Demanding N-Het-
erocyclic Carbene Ligands with Restricted Flexibility for Palladium-
catalyzed Cross-Coupling Reactions. Acc. Chem. Res. 2008, 41 (11),
1523–1533.
(43) Furstner, A.; Alcarazo, M.; Krause, H.; Lehmann, C. W. Effec-
tive Modulation of the Donor Properties of N-Heterocyclic Carbene
Ligands by “Through-Space” Communication within a Planar Chiral
Scaffold. J. Am. Chem. Soc. 2007, 129 (42), 12676.
(44) Baker, M. V.; Barnard, P. J.; Brayshaw, S. K.; Hickey, J. L.;
Skelton, B. W.; White, A. H. Synthetic, Structural and Spectroscopic
Studies of (Pseudo)halo(1,3-di-tert-butylimidazol-2-ylidine) Gold Com-
plexes. Dalton Trans. 2005, 1, 37–43.

3454 Organometallics, Vol. 29, No. 15, 2010
Gung et al.
Scheme 3. Assignment of the Relative Configuration of the Diastereomeric Formal [2þ2] Cycloaddition Product 12b and the
Corresponding Alcohols 14a and 14b
will show that complex 11, which has the longest Au-C
bond length, is the least stable.
Catalytic Activity of Gold Complexes 9 and 11.
are depicted in Scheme 4. Both gold-stabilized carbocations
and gold carbenoids have been depicted for intermediates
involving Au(I)(NHC) catalysts. The dual characteristics of
gold complexes have been discussed extensively.36,45-48 In
the presence of catalyst Au(IPr) (I) the formation of the
cyclopropanation products 5a-c involves most likely an
initial enyne cyclization to form the cyclopropane carbo-
cation intermediate B, which is resonance-stabilized with
the gold carbenoid B0 (Scheme 4). A 1,2-acetate migration
and deauration from B would give the cyclopropanation
product 5.
The isolation of product 12a indicates a 1,3-migration of
the acetate group in the presence of gold complex 9. Accord-
ing to a computational study of Cavallo and co-workers,48,49
both 1,2- and 1,3-migration of the acetate group are low-
barrier processes and gold-coordinated allenyl ester is 8 kcal/
mol more stable than gold-coordinated propargyl ester.
Therefore, it is plausible that 1,3-acetate migration competes
with enyne cyclization initially. With substrate 4a, where
R = Ph, no cyclopropanation product 5a was observed in
the presence of Au(I) catalyst 9 or 11. It is conceivable that a
phenyl group stabilizes allene 12a by conjugation, while a
propyl or an isopropyl group may destabilize the allene
intermediate by making it more prone to electrophilic co-
ordination by the same Au(I) catalyst.
Interestingly, the reactions of the substrate 4a (R=Ph)
provided different products (12a and 12b) when Au(I) com-
plexes 9-Cl and 11-Cl were employed (Table 2). This was
especially surprising when considering that a uniformly
high yields of the cyclopropanation products 5a-c were
obtained when the catalyst Au(IPr) (I) was employed. In
the presence of complex 9-Cl/AgSbF6, substrate 4a was
converted to allene 12a and no further change was ob-
served even after prolonged stirring at room temperature
or heating to reflux in CH2Cl2. On the other hand, with
substrates 4b and 4c, the “normal” cyclopropanation pro-
ducts 5b and 5c were obtained in high yield, respectively,
as a mixture of double-bond isomers in favor of the
Z-isomer.
In the presence of complex 11-Cl/AgSbF6, substrate 4a
gave yet another product (12b). The formal [2þ2] cycloaddi-
tion reaction product 12b was isolated in 56% yield as a 2:1
mixture of double-bond isomers in favor of the Z-isomer
from 4a in the presence of complex 11-Cl/AgSbF6, while
substrates 4b and 4c produced the “normal” cyclopropanes
5b and 5c in high yield, respectively. Unfortunately no
significant optical activity was observed for reactions
employing either catalyst system 9-Cl/AgSbF6 or 11-Cl/
AgSbF6.
With substrate 4a, complex 9 appears to have lost its
catalytic activity once 12a was produced. More evidence
supporting this hypothesis is presented in the next section.
The gold-catalyzed reaction appears to favor the production
of the cyclopropanation product 5a with the more stable
complex Au(IPr) (I). In order to test this hypothesis, allene
(45) Echavarren, A. M. Carbene or Cation? Nat. Chem. 2009, 1 (6),
431–433.
(46) Benitez, D.; Shapiro, N. D.; Tkatchouk, E.; Wang, Y. M.;
Goddard, W. A.; Toste, F. D. A Bonding Model for Gold(I) Carbene
Complexes. Nat. Chem. 2009, 1 (6), 482–486.
The geometric isomers and the structure of the interesting
[2þ2] cycloaddition product 12b were assigned by comparing
the chemical shift difference of the vinyl protons and by
converting to the bicyclic ketone 13 with ozonolysis, Scheme 3.
Furthermore, removal of the acetate group from product 12b
allowed the separation of the alcohols 14a and 14b and the
determination of the double-bond configuration by a two-
dimensional NOESY spectrum. The major isomer 14a was
determined to have the Z-double bond on the basis of the
NOE effect and the unchanged chemical shift of the vinyl
proton from its precursor.
(47) Amijs, C. H. M.; Lopez-Carrillo, V.; Raducan, M.; Perez-Galan,
P.; Ferrer, C.; Echavarren, A. M. Gold(I)-Catalyzed Intermolecular
Addition of Carbon Nucleophiles to 1,5- And 1,6-Enynes. J. Org. Chem.
2008, 73 (19), 7721–7730.
(48) Marion, N.; Lemiere, G.; Correa, A.; Costabile, C.; Ramon,
R. S.; Moreau, X.; de Fremont, P.; Dahmane, R.; Hours, A.; Lesage, D.;
Tabet, J. C.; Goddard, J. P.; Gandon, V.; Cavallo, L.; Fensterbank, L.;
Malacria, M.; Nolan, S. P. Gold- and Platinum-Catalyzed Cycloisome-
rization of Enynyl Esters versus Allenenyl Esters: An Experimental and
Theoretical Study. Chem.;Eur. J. 2009, 15 (13), 3243–3260.
(49) Correa, A.; Marion, N.; Fensterbank, L.; Malacria, M.; Nolan,
S. P.; Cavallo, L. Golden Carousel in Catalysis: The Cationic Gold/
Propargylic Ester Cycle. Angew. Chem., Int. Ed. 2008, 47 (4), 718–721.
Possible Pathways Leading to the Observed Products.
Plausible pathways for the conversion of the propargyl esters
4a-c in the presence of gold catalysts to different products

Article
Organometallics, Vol. 29, No. 15, 2010 3455
Scheme 4. Possible Pathways Leading to Observed Products
12a was allowed to stir with complex I under the same con-
ditions, eq 2. Cyclopropanation product 5a was indeed pro-
duced with only 2.5% of the catalyst at room temperature.
intermediates in gold-catalyzed cycloisomerization reactions
have been compared to “nonclassical” carbocations.45,52 The
delocalization of the electrons in the reactive intermediate
can be depicted in three renditions (A, B, and C, Scheme 4).
The proposed pathway involves the cyclobutane tertiary
carbocation rendition C for the production of 12b.52 A 1,3-
acetate migration/deauration from C would produce pro-
duct 12b. The question is why only gold catalyst 11 favors the
formation of 12b. A recent study of the π-acceptor property
of NHCs in gold catalysis provides a reasonable expla-
nation.37 An electron-rich gold complex tends to stabilize
an adjacent carbocation and therefore favor the cyclopro-
panation pathway via resonance contributors B and B0; in
contrast, a more electron-deficient gold complex might favor
the rendition C, in which the cation center is not directly
bound to gold, and therefore produces the alkylidenecyclo-
butane 12b by a 1,3-acetate migration/deauration process.
The proposed pathways in Scheme 4 suggest different
catalyst stability for the three Au(I) complexes employed.
In order to find corroborative evidence, variable-tempera-
ture NMR experiments were performed on the three Au(I)
complexes (I, 9, and 11). Each Au(NHC)Cl complex was
treated with an equal molar amount of AgSbF6 in CDCl3
and stirred at room temperature for 15 min, eq 4.
The formation of the formal [2þ2] cycloaddition product
alkylidenecyclobutane 12b with a terminal alkene, to the best
of our knowledge, is unprecedented. Previously reported
gold-catalyzed intramolecular allene-ene [2þ2] cycloaddi-
tion reactions involve either trisubstituted or phenyl-substi-
tuted alkenes.50 A stable tertiary or benzylic carbocation is
usually proposed as the intermediate in the initial nucleo-
philic attack on the gold-activated allene moiety by the
electron-rich double bond.50,51 For substrate 4a, this me-
chanism requires the initial formation of 12a, which would
be activated by the same Au(I) catalyst and attacked by the
terminal alkene. The problem with this pathway is that it
invokes a primary carbocation in order to form the cyclo-
butane, eq 3. Since this is unlikely to happen,
we propose the pathway depicted in Scheme 4 to explain the
isolation of the [2þ2] cycloaddition product 12b. Reactive
The resulting suspension was filtered, the filtrate was
1
loaded into an NMR tube, and the H NMR spectra were
recorded at 30 min intervals from room temperature to
55 °C. Previously Nolan and co-workers have reported rapid
decomposition and generation of colloidal gold upon at-
tempted preparation of the cationic gold(I) complex in dry
dichloromethane or dry chloroform with silver salts.5 We
(50) Luzung, M. R.; Mauleon, P.; Toste, F. D. Gold(I)-Catalyzed
[2þ2]-Cycloaddition of Allenenes. J. Am. Chem. Soc. 2007, 129 (41),
12402–12403.
(51) Zhang, L. M. Tandem Au-catalyzed 3,3-Rearrangement-[2þ2]
Cycloadditions of Propargylic Esters: Expeditious Access to Highly
Functionalized 2,3-Indoline-Fused Cyclobutanes. J. Am. Chem. Soc.
2005, 127 (48), 16804–16805.
(52) Furstner, A.; Morency, L. On the Nature of the Reactive Inter-
mediates in Gold-Catalyzed Cycloisomerization Reactions. Angew.
Chem., Int. Ed. 2008, 47 (27), 5030–5033.
1
were able to record H NMR spectra of the cationic Au(I)
complexes from rt to 55 °C during a five-hour time period.
The CDCl3 we used was purchased and used as received. It
1
contains about 0.1% water, as indicated by its H NMR

3456 Organometallics, Vol. 29, No. 15, 2010
Gung et al.
Figure 2. Catalyst stability study: ([) catalyst I, (9) catalyst 9,
(2) catalyst 11.
Figure 4. 1H NMR spectra of complex 9þ. Complex 9-Cl was
stirred with 1 equiv of AgSbF6 for 15 min at rt in CDCl3. After
filtering off AgCl, the filtrate was loaded into an NMR tube and
the spectra were recorded at 30 min intervals with increasing
temperatures. A broadened quartet at 5.6 ppm is observed
initially and decreases in intensity as the temperature rises. A
second quartet at 5.4 ppm grows from about 5% initially to
predominant at 55 °C. At the end of the experiment, gold
mirroring was observed for complex 9þ.
catalyst had already decomposed when the first spectrum
was recorded (not shown). Thus the variable-temperature
NMR experiments carried out with the three gold complexes
support the hypothesis that the different results observed
with the Au(I) catalysts (I, 9, and 11) are mainly due to the
stability difference in the catalyst.
Conclusions
Figure 3. 1H NMR spectra of complex I (Au(IPr)þ). Au(IPr)Cl
was stirred with 1 equiv of AgSbF6 for 15 min at rt in CDCl3.
After filtering off AgCl, the filtrate was loaded into an NMR
tube and the spectra were recorded at 30 min intervals with
increasing temperatures.
Two new chiral gold complexes have been prepared and
their structures determined by X-ray structure analysis.
These complexes become catalytically active toward the
cyclopropanation reactions of propargyl esters 4a-c upon
treatment with a silver salt. However, their thermal stability
is inferior compared to the commercially available complex
Au(IPr)Cl (I). Future design of other chiral NHC ligands for
preparing gold catalysts should take into account the ther-
mal stability of the resulting gold complexes. Work along
this line is underway in our laboratories.
spectrum. It is possible the trace amount of water may have
served to stabilize the cationic gold complex in solution. The
cationic catalyst I showed no signs of change from room
temperature to 55 °C and from the first to the last spectrum
taken during the 5 h of time, Figures 2 and 3. The signal for
one pair of the isopropyl groups is broadened, but the
integrity of the complex remains intact and no gold mirror-
ing was observed.
For catalysts 9 and 11, it was convenient to monitor the
proton on the stereogenic methine carbon during the NMR
study. From the very first NMR spectrum, two sets of signals
for the methine proton were recorded, Figures 2 and 4. For
catalyst 9, one slightly broadened quartet and one clear
quartet were observed at 5.6 and 5.4 ppm, respectively. The
broadened quartet was assigned to the cationic catalyst
because it gradually disappears and the clear quartet gradu-
ally grows as the temperature rises. At the conclusion of the
experiments for catalyst 9, a gold mirror was deposited onto
the bottom of the NMR tube. For catalyst 11, 50% of the
Acknowledgment. B.W.G. thanks the CFR of Miami
University for a summer stipend. L.N.B. and D.T.C.
thank the Department of Chemistry & Biochemistry,
Miami University, for a Teaching Award and a Disserta-
tion Scholarship, respectively.
Supporting Information Available: Text and figures giving full
experimental details for the preparation of all new compounds
described in this paper, including copies of H and 13C NMR
spectra, and crystallographic data and refinement parameters of
compounds 9-Cl and 11-Cl. This material is available free of
charge via the Internet at http://pubs.acs.org.
1