Regio- and enantioselective hydroamination of dienes by gold(I)/menthol cooperative catalysis

Article abstract of DOI:10.1002/anie.201104076

Alcohol is key: Regio- and enantioselective hydroamination of 1,3-dienes has been achieved with the dinuclear catalyst (R)-DTBM-SEGPHOS. The rate and selectivity of the reaction are enhanced by alcohol additives like menthol, which coordinates the cationic gold(I) to generate a Br?nsted acid that can participate in catalysis. Mbs=p-methoxybenzenesulfonyl. Copyright

Full text of DOI:10.1002/anie.201104076

DOI: 10.1002/anie.201104076
Gold Catalysis
Regio- and Enantioselective Hydroamination of Dienes by Gold(I)/
Menthol Cooperative Catalysis**
Osamu Kanno, Wataru Kuriyama, Z. Jane Wang, and F. Dean Toste*
Gold(I)-mediated hydroamination reactions have emerged as
attractive methods for the formation N-containing hetero-
cycles.[1,2] While significant progress has been made in the
asymmetric hydroamination of allenes, enantioselective gold-
catalyzed hydroamination of simple alkenes and dienes has
been limited to reactions in which urea is employed as a
nucleophile.[3] Furthermore, as simple Brønsted acids are also
known to be effective catalysts for alkene hydroamination[4]
and gold(I) triflates are often employed in olefin activation,[5]
the role of gold in these reactions is unclear. In 2009, we found
that, in presence of stoichiometric amounts of base, alkyl-
gold(I) complexes could be formed by gold-promoted addi-
tion of nitrogen nucleophiles to unactivated alkenes
[Eq. (1)].[6,7] This result led us to posit that generation of an
produce appreciable amounts of pyrrolidine (Table 1,
entry 1), we were pleased to find that the combination of
(R)-DTBM-SEGPHOS(AuCl)2 (5) and AgBF4 in dichloro-
Table 1: Initial screen of catalyst, protecting group, and solvent.
Entry
1
PG
Cat. Solvent Conv. [%][a] 2/3
ee [%]
3[c]
2
1
2
3
4
5
1a Ts
1a Ts
1b Mbs
1b Mbs
1b Mbs
4
5
5
5
5
CH2Cl2
CH2Cl2
CH2Cl2
MeOH
iPrOH
trace
5
18
81
>99
–
1:0
–
–
35
2
–
–
–
84
1:0
1:0.5[b]
1:1.2[b] 11[d] 92
[a] Conversion was determined by 1H NMR spectroscopy and HPLC
analysis. [b] A very small amount of (Z)-3b was included (1–3%). [c] The
ee value corresponds to that of the E isomer. [d] 11% ee was observed for
the S enantiomer. (R)-DTBM-SEGPHOS=(R)-(+)-5,5’-bis[di(3,5-di-tert-
butyl-4-methoxyphenyl)phosphino]-4,4’-bi-1,3-benzodioxole, PG=pro-
tecting group, Ts=4-toluenesulfonyl.
acidic species was important for catalytic turnover in the
reported alkene hydroamination reactions. Moreover, Tilley
et al. proposed that the catalyst in the related platinum-
catalyzed reaction is a platinum sulfonamide complex derived
from coordination of the Lewis-acidic metal to the relatively
acidic sulfonamide nucleophile.[8] On the basis of these
reports, we envisioned that simple alcohols might serve an
analogous role in Lewis acid activated Brønsted acid cata-
lyzed[9,10] processes with gold(I) complexes playing the role of
the Lewis acid. Herein, we report the application of this
hypothesis to the development of an enantioselective hydro-
amination of 1,3-dienes.[11]
methane gave pyrrolidine 2 exclusively in 5% conversion
after 24 hours (entry 2). After examining a variety of protect-
ing groups on the nitrogen atom, we found that the Mbs-
protected (Mbs = p-methoxy benzenesulfonyl) amine gave a
small enhancement in reactivity (entry 3). Having established
reaction conditions for productive hydroamination, we envi-
sioned that addition of a potential Brønsted acid would
enhance the reactivity of the catalyst system. Accordingly,
when the solvent was switched to MeOH (entry 4), we saw
dramatic increases in the reaction rate. Additionally, both
products 2 and 3 were observed, with the major product 3
formed in 92% enantioselectivity (entry 5) when iPrOH was
employed as solvent. The pronounced rate acceleration and
change in product distribution suggested that alcohols are
important for controlling the regioselectivity and rate of
reaction.
Thus, we examined the effect that a range of achiral and
chiral alcohols additives had on the catalytic reaction in
CH2Cl2 (Table 2). In particular, we hypothesized that a chiral
alcohol may “match” the chiral information enforced by 5 and
lead to higher levels of selectivity. Gratifyingly, we found that
(À)-menthol provided the desired products in quantitative
conversion and a 1:9 ratio of 2b to 3b with excellent
enantioselectivity (95% ee) for the major product (entry 6).
Furthermore, the amount of (À)-menthol used could be
reduced to 2.0 equivalents without significantly impacting the
regio- or enantioselectivity (entry 7).
Our initial studies focused on the reaction of diene 1 with
catalytic amounts cationic gold(I) complexes (see the Sup-
porting Information for a complete list). While attempts to
catalyze the reaction with triphenylphosphinegold(I) did not
[*] Dr. O. Kanno, Dr. W. Kuriyama, Z. J. Wang, Prof. F. D. Toste
Department of Chemistry, University of California Berkeley
Berkeley, CA 94720 (USA)
E-mail: fdtoste@berkeley.edu
[**] This work was supported by the NIHGMS (RO1 GM073932). We
thank Takasago Int. Co. for the generous donation of phosphine
ligands and Johnson Matthey for a donation of AuCl3. We thank Dr.
Anthony Iavarone for help with mass spectroscopy. Z.J.W. thanks
the Hertz Foundation for a graduate fellowship. O.K. and W.K.
acknowledge support from the Daiichi-Sankyo Co., Ltd. and
Takasago Int. Co., respectively.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201104076.
Angew. Chem. Int. Ed. 2011, 50, 9919 –9922
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9919

Communications
Table 2: Hydroamination in the presence of alcohol additives.
While hydroamination of dienes to form the six-membered
ring remains difficult for the standard substrate (entry 5),
incorporation of aryl groups into the backbone allow the
transformation to be performed in modest to good yield
(entry 7–8). Additionally, endocyclic amides could be pre-
pared readily with modest regioselectivity and excellent
enantioselectivity for the vinyl-substituted product (entry 9).
The absolute configurations of the products were determined
by analogy from the crystal structure of the product 3k (see
the Supporting Information).
To investigate the mechanism of the hydroamination and
to determine the role of gold and menthol, we monitored the
reaction by 1H NMR spectroscopy. For this purpose, we chose
substrate 1j, as its reaction is significantly faster than any
other substrate and thus suitable for studies by point kinetics.
We examined the reaction of 1j under two reaction con-
ditions: 1) with 2.0 equivalents of (À)-menthol as an additive
and 2) with no additives (Scheme 1). As two products can be
Entry Additive
Equiv Conv. [%][a] 2/3[b] ee [%][c]
1
2
3
4
5
6
7
tBuOH
cyclohexanol
(R)-methyl lactate
10
10
10
94
97
40
46
94
1:0.6 91
1:1.5 97
1:0.3
1:1.2 90
1:1.9 94
1:9.0 95
1:6.3 95
–
(R)-1-phenethyl alcohol 10
(+)-menthol
(À)-menthol
(À)-menthol
10
10
2
>99
95
[a] Conversion was determined by 1H NMR spectroscopy and HPLC
analysis. [b] A very small amount of (Z)-3b was included (1–3%). [c] The
ee value corresponds to that of (E)-3b. See the Supporting Information
for ee values of 2b.
With these results in hand, we were able to perform the
cyclization of a variety of 1,3-dienes at ambient temperature
in good yield and enantioselectivity for the major product
(Table 3). Although the gem-dimethyl group in the backbone
of the substrate was not required for good selectivity, the
reaction was significantly slower without the substitution
(entry 2) By incorporating N-protected piperidine into the
backbone, the spiro-bicyclic product 3e could be formed in
excellent yield, regioselectivity, and with 89% ee (entry 4).
Scheme 1. Relative rate enhancement by (À)-menthol for formation of
2j and 3j under two different reaction conditions.
Table 3: Substrate scope of diene hydroamination.
formed from 1j, the overall rate constant for the reaction is
equal to the sum of the rate constants for the formation of
each product and the ratio of rate constants for each product,
k3/k2, is directly proportional to the concentration of 2j and 3j
at all times. Interestingly, we found that (À)-menthol had
almost no effect on the rate of formation of 2j, but increased
the rate of formation of 3j by 36-fold.[12] To ensure that the
rate acceleration observed was not due to interconversion of
the two products to a thermodynamic distribution, we isolated
an authentic sample of 2j and resubjected it to 3 mol% 5,
6 mol% AgBF4, and 2.0 equivalents of (À)-menthol over-
night at room temperature. Indeed, no equilibration to 3j was
observed after 18 hours.
Entry
1
n
R
t [h]
Yield [%][a]
ee
(2/3[b])
[%][c]
1[d]
2[d]
3[d]
1b
1c
1d
1
1
1
Me
H
Ph
24
144
24
95 (1:6.3)
91 (1:6.1)
93 (1:3.5)
95
85
84
4[d]
5[d]
1e
1 f
1
2
24
–
99 (1:5.0)
89
–
Me
n.r.
Entry
Substrate
t [h]
Yield [%][a]
ee
(2/(Z)-3/(E)-3)
[%][c]
In addition, when 20% proton sponge is added to the
reaction of 1j with 5 and (À)-menthol, none of the vinyl
product 3j is formed [Eq. (2)] and the rate of overall reaction
6
7
1g
72
72
80 (1:0.5:7.5)
97
97
1h
1i
77 (1:3.9:8.2)
42 (1:1.2:2.0)
8
168
98
9
1j
6
99 (1:0.03:1.5)
94
91
is significantly reduced. However, if a related but acid-stable
substrate (1b) is treated with 10 mol% TfOH in the absence
of gold(I), the allylic product 3b is formed exclusively
[Eq. (3); Tf = trifluorosulfonyl]. Finally, mass spectrometry
of a solution of 5 and (À)-menthol show peaks at m/z = 1727
(singly charged ion) and 864 (doubly charged ion), which
10
1k
24
67 (1:<0.1:7.5)
[a] Based on the combined yield of isolated 2 and 3. [b] A very small
amount of (Z)-3 was included (1–3%). [c] The ee value corresponds to
that of (E)-3. See the Supporting Information for ee values of 2b–k.
[d] The protecting group (PG) is Mbs. n.r. =no reaction.
9920 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 9919 –9922

Experimental Section
General procedure for gold(I)-catalyzed hydroamination: A mixture
of AgBF4 (0.58 mg, 3.0 mmol) and the phosphine gold(I) chloride
complex (1.5 mmol) was suspended in 400 mL of solvent in a sealed
vial, and sonicated or stirred magnetically for 15 min at room
temperature. The resulting suspension was filtered through a glass-
microfiber plug directly into a solution of substrate (0.05 mmol) in
100 mL of same solvent. The mixture was stirred for 24 h at room
temperature. The reaction was quenched by addition of one drop of
triethylamine and the conversion was determined by 1H NMR
spectroscopy or HPLC analysis of the crude reaction mixture after
filtration though a short silica gel column. Reported yields reflect
yields of mixtures of olefin regioisomers that were isolated after
chromatography on silica gel.
correspond to a 1:1 complex of (R)-[DTBM-SEGPHO-
S(Au)2]2+ and (À)-menthol (see the Supporting Information).
These experiments lead us to believe that 2j and 3j are
being formed predominantly through two different mecha-
nisms: pathway 1 involves traditional coordination, nucleo-
philic addition, and proto-demetalation by gold similar to
what has been previously observed for allenes;[13] pathway 2
proceeds through an Brønsted acid catalyzed pathway in
which 5 acts as a Lewis acid by binding to menthol, thus
increasing its Brønsted acidity (Scheme 2). In pathway 1,
coordination of the gold catalyst to the diene enables
intramolecular addition, and proto-demetalation[14] frees the
gold catalyst. In pathway 2, gold coordinates to menthol and
generates a Bronsted acidic species that then allows diene
protonation and cyclization of the nucleophile. When a base
such as a proton sponge is added, catalysis through the acid-
mediated pathway 2 is inhibited; however, pathway 1 is still
operative, albeit at a diminished rate. Exclusive formation of
3j by Brønsted acid catalysis with TfOH is also consistent
with our mechanistic proposal.
Full experimental details including preparation of the diene
substrates, determination of enantioselectivity and HPLC data,
kinetic studies, and mass spectrometry data are provided in the
Supporting Information.
Received: June 14, 2011
Revised: June 29, 2011
Published online: September 7, 2011
Keywords: enantioselectivity · gold · heterogeneous catalysis ·
.
hydroamination · synthetic methods
[1] For recent general reviews, see: a) N. Krause, N. Winter, Chem.
Rev. 2011, 111, 1994; b) A. Fꢀrstner, Chem. Soc. Rev. 2009, 38,
3208; c) N. D. Shapiro, F. D. Toste, Synlett 2010, 675; d) A. S. K.
Hashmi, Angew. Chem. 2010, 122, 5360; Angew. Chem. Int. Ed.
2010, 49, 5232. For select examples of alkyne hydroaminination,
see: e) E. Mizushima, T. Hayashi, M. Tanaka, Org. Lett. 2003, 5,
3349; f) D. J. Gorin, N. R. Davis, F. D. Toste, J. Am. Chem. Soc.
2005, 127, 11260; g) J. Kang, H. Kim, J. Lee, S. Shin, Org. Lett.
2006, 8, 3537; h) A. S. K. Hashmi, M. Rudolph, S. Schymura, J.
Visus, W. Frey, Eur. J. Org. Chem. 2006, 4905; i) N. T. Patil, R. D.
Kavthe, V. S. Raut, V. S. Shinde, B. Sridhar, J. Org. Chem. 2010,
75, 1277. For select examples of allene hydroamination with
gold, see j) N. Krause, N. Morita, Org. Lett. 2004, 6, 4121; k) N.
Nishina, Y. Yamamoto, Angew. Chem. 2006, 118, 3392; Angew.
Chem. Int. Ed. 2006, 45, 3314; l) Z. Zhang, C. Liu, R. E. Kinder,
X. Han, H. Qian, R. A. Widenhoefer, J. Am. Chem. Soc. 2006,
128, 9066; m) R. L. LaLonde, B. D. Sherry, E. J. Kang, F. D.
Toste, J. Am. Chem. Soc. 2007, 129, 2452.
Scheme 2. Two mechanisms, pathway 1 and pathway 2, are proposed
to account for formation of 2 and 3, respectively.
[2] For reviews of enantioselective hydroamination, see: a) K. C.
Hultzsch, Org. Biomol. Chem. 2005, 3, 1819; b) K. C. Hultzsch,
Adv. Synth. Catal. 2005, 347, 367; c) R. A. Widenhoefer, X. Han,
Eur. J. Org. Chem. 2006, 4555.
[3] a) C. F. Bender, R. A. Widenhoefer, Org. Lett. 2006, 8, 5303;
b) Z. Zhang, S. D. Lee, R. A. Widenhoefer, J. Am. Chem. Soc.
2009, 131, 5372; c) P. Mukherjee, R. A. Widenhoefer, Org. Lett.
2010, 12, 1184.
[4] a) B. Schlummer, J. F. Hartwig, Org. Lett. 2002, 4, 1471; b) D. C.
Rosenfeld, S. Shekhar, A. Takemiya, M. Utsunomiya, J. F.
Hartwig, Org. Lett. 2006, 8, 4179; c) Z. Li, J. Zhang, C. Brouwer,
C.-G. Yang, N. W. Reich, C. He, Org. Lett. 2006, 8, 4175.
[5] a) C. Brouwer, C. He, Angew. Chem. 2006, 118, 1776; Angew.
Chem. Int. Ed. 2006, 45, 1744; b) J. Zhang, C.-G. Yang, C. He, J.
Am. Chem. Soc. 2006, 128, 1798; c) C. F. Bender, R. A.
Widenhoefer, Chem. Commun. 2006, 4143; d) X. Han, R. A.
Widenhoefer, Angew. Chem. 2006, 118, 1779; Angew. Chem. Int.
Ed. 2006, 45, 1747; e) R. E. Kinder, Z. Zhang, R. A. Widenhoe-
fer, Org. Lett. 2008, 10, 3157. For a related example of Brønsted
acid catalyzed hydroarylation in the presence of PtII, see M. A.
In conclusion, we report the first examples of Lewis acid
activated Brønsted acidity in gold(I) catalysis. This novel
catalyst system was applied to the formation of pyrrolidine
and piperidine adducts through a phosphinegold(I)/menthol-
catalyzed enantioselective diene hydroamination. The regio-
selectivity and enantioselectivity of the reaction can be
controlled by the addition of menthol as a cocatalyst, which
we believe acts as a Brønsted acid when coordinated to the
gold catalyst and results in the formation of vinyl-substituted
products (3). Thus, the 1,3-diene scaffold has access to both
mechanistic pathways involved for gold(I)-catalyzed hydro-
amination, depending upon the presence of a potential
Brønsted acid, such as an alcohol. Detailed kinetic studies
of this reaction and application of Lewis acid activated
Brønsted acidity by gold(I) to other reactions are ongoing and
will be reported in due course.[15]
Angew. Chem. Int. Ed. 2011, 50, 9919 –9922
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9921

Communications
Bowring, R. G. Bergman, T. D. Tilley, Organometallics 2011, 30,
1295.
[11] The racemic hydroamination of 1,3-dienes has previously be
investigated by Brouwer and He, see reference [5a] and a) X.
Giner, C. Nꢁjera, Org. Lett. 2008, 10, 2919. For examples of
palladium-catalyzed diene hydroamination, see b) O. Lçber, M.
Kawatsura, J. F. Hartwig, J. Am. Chem. Soc. 2001, 123, 4366;
c) A. M. Johns, M. Utsunomiya, C. D. Incarvito, J. F. Hartwig, J.
Am. Chem. Soc. 2006, 128, 1828.
[6] R. L. LaLonde, W. E. Brenzovich, D. Benitez, E. Tkatchouk,
K. K. Kelly, W. A. Goddard, F. D. Toste, Chem. Sci. 2010, 1, 226.
[7] The same phenomena have been observed for palladium-
catalyzed hydroamination: B. M. Cochran, F. E. Michael, J.
Am. Chem. Soc. 2008, 130, 2786.
[8] J. L. McBee, A. T. Bell, T. D. Tilley, J. Am. Chem. Soc. 2008, 130,
16562.
[12] As determined as an average of two sets of independent runs.
[13] a) X. Zeng, M. Soleilhavoup, G. Bertrand, Org Lett. 2009, 11,
3166; b) R. L. Lalonde, J. Z. Wang, M. Mba, A. D. Lackner, F. D.
Toste, Angew. Chem. 2010, 122, 608; Angew. Chem. Int. Ed. 2010,
49, 598; c) Z. J. Wang, D. Benitez, E. Tkatchouk, W. A. God-
dard, F. D. Toste, J. Am. Chem. Soc. 2010, 132, 13064; d) A. N.
Duncan, R. A. Widenhoefer, J. Am. Chem. Soc. 2010, 132, 419.
[14] As proto-demetalation from alkylgold complexes are known to
be much slower than in the analogous reaction with vinylgold
complexes (see Ref. [6]), we do not preclude menthol from
participating in pathway 1 in the proto-demetalation step.
[15] C. H. Cheon, O. Kanno, O. F. D. Toste, J. Am. Chem. Soc. 2011,
133, 13248.
[9] For pioneering studies on the use of Lewis acid assisted Brønsted
acid catalysis to the enantioselective additions to alkenes, see:
a) K. Ishihara, S. Nakamura, H. Yamamoto, J. Am. Chem. Soc.
1999, 121, 4906; b) S. Nakamura, K. Ishihara, H. Yamamoto, J.
Am. Chem. Soc. 2000, 122, 8131; c) K. Ishihara, H. Ishibashi, H.
Yamamoto, J. Am. Chem. Soc. 2001, 123, 1505; d) K. Ishihara, H.
Ishibashi, H. Yamamoto, J. Am. Chem. Soc. 2002, 124, 3647. For
an excellent review see: e) H. Yamamoto, K. Futatsugi, Angew.
Chem. 2005, 117, 1958; Angew. Chem. Int. Ed. 2005, 44, 1924.
[10] For an alternative Brønsted acid catalyzed enantioselective
hydroamination of dienes, see: N. D. Shapiro, V. Rauniyar, G. L.
Hamilton, J. Wu, F. D. Toste, Nature 2011, 470, 245.
9922 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 9919 –9922