Ligand-accelerated enantioselective propargylation of aldehydes via allenylzinc reagents

Article abstract of DOI:10.1021/ol200043n

An enantioselective propargylation of aldehydes using an allenylzinc reagent generated in situ via a zinc-iodine exchange reaction is described. The enantioselectivity is controlled by addition of a catalytic amount of readily accessible and highly tunable amino alcohol ligand L13. A wide range of aldehydes can be propargylated to afford valuable and versatile homopropargyl alcohols in good to excellent yields with high levels of enantiopurity.

Full text of DOI:10.1021/ol200043n

ORGANIC
LETTERS
2011
Vol. 13, No. 8
1900–1903
Ligand-accelerated Enantioselective
Propargylation of Aldehydes via
Allenylzinc Reagents
Barry M. Trost,* Ming-Yu Ngai, and Guangbin Dong
Department of Chemistry, Stanford University, Stanford, California 94305, United
States
bmtrost@stanford.edu
Received January 6, 2011
ABSTRACT
An enantioselective propargylation of aldehydes using an allenylzinc reagent generated in situ via a zinc-iodine exchange reaction is described.
The enantioselectivity is controlled by addition of a catalytic amount of readily accessible and highly tunable amino alcohol ligand L13. A wide
range of aldehydes can be propargylated to afford valuable and versatile homopropargyl alcohols in good to excellent yields with high levels of
enantiopurity.
Chiral homopropargyl alcohols are useful synthetic
building blocks that are routinely utilized in the synthesis
of a wide variety of bioactive natural products and pharma-
ceutical compounds.1 Moreover, these compounds offer an
opportunity for dual reactivity: (1) the alkyne functional
group can undergo hydration or oxidative cleavage to
provide aldehydes or ketones, and can be reduced to provide
olefins or alkanes,2 and (2) the chiral alcohols frequently act
as directing groups for many essential reactions, such as
epoxidation, cyclopropanation, carbonyl reduction, and
hydrosilylation.3,4 In addition, the homopropargyl alcohol
itself is readily converted to synthetically useful heterocycles
such as polyhydrofurans and polyhydropyrans in a highly
chemoselective and atom-economical fashion.5
The most attractive and straightforward method for
synthesizing enantiopure homopropargylic alcohols is
the direct catalytic rather than stoichiometric asymmetric
carbonyl propargylation. Several approaches have been
established for this purpose. For example, Keck and Den-
mark have respectively reported a Lewis acid-6 and Lewis
base-catalyzed7 asymmetric carbonyl propargylation, yet
these methodsemployeda stoichiometricamount of highly
toxic tin reagents. Catatlytic asymmetric Nozaki-Hiyama
propargayltions have also been reported with respectable
levels of enantioselectivity, but the requirement of toxic
chromium source and long reaction time is still of concern
with this method.8 During the course of our studies,
(1) For selected examples: (a) Carter, R. G.; Weldon, D. J. Org. Lett.
2000, 2, 3913–3916. (b) O’Sullivan, P. T.; Buhr, W.; Fuhry, M. A. M.;
Harrison, J. R.; Davies, J. E.; Feeder, N.; Marshall, D. R.; Burton, J. W.;
Holmes, A. B. J. Am. Chem. Soc. 2004, 126, 2194–2207. (c) Trost, B. M.;
Dong, G. B. Nature 2008, 456, 485–488. (d) Liu, S. B.; Kim, J. T.; Dong,
C. G.; Kishi, Y. Org. Lett. 2009, 11, 4520–4523. (e) Francais, A.; Leyva,
A.; Etxebarria-Jardi, G.; Ley, S. V. Org. Lett. 2010, 12, 340–343.
(2) Stang, P. J.; Diederich, F. Modern Acetylene Chemistry; VCH:
New York, 1995.
(3) Hoveyda, A. H.; Evans, D. A.; Fu, G. C. Chem. Rev. 1993, 93,
1307–1370.
(4) For review of hydrosilation of alkynes, see: Trost, B. M.; Ball,
Z. T. Synthesis 2005, 853–887.
(5) (a) Quayle, P.; Rahman, S.; Ward, E. L. M.; Herbert, J. Tetra-
hedron Lett. 1994, 35, 3801–3804. (b) McDonald, F. E.; Gleason, M. M.
J. Am. Chem. Soc. 1996, 118, 6648–6659. (c) Trost, B. M.; Rhee, Y. H. J.
Am. Chem. Soc. 1999, 121, 11680–11683. (d) Consorti, C. S.; Ebeling, G.;
Dupont, J. Tetrahedron Lett. 2002, 43, 753–755. (e) Trost, B. M.; Rhee,
Y. H. J. Am. Chem. Soc. 2003, 125, 7482–7483.
(6) Keck, G. E.; Krishnamurthy, D.; Chen, X. Tetrahedron Lett.
1994, 35, 8323–8324.
(7) Denmark, S. E.; Wynn, T. J. Am. Chem. Soc. 2001, 123, 6199–
6200.
(8) (a) Bandini, M.; Cozzi, P. G.; Umani-Ronchi, A. Polyhedron
2000, 19, 537–539. (b) Bandini, M.; Cozzi, P. G.; Melchiorre, P.; Tino,
R.; Umani-Ronchi, A. Tetrahedron: Asymmetry 2001, 12, 1063–1069. (c)
Inoue, M.; Nakada, M. Org. Lett. 2004, 6, 2977–2980. (d) Usanov, D. L.;
Yamamoto, H. Angew. Chem., Int. Ed. 2010, 49, 8345–8348.
r
10.1021/ol200043n
Published on Web 03/10/2011
2011 American Chemical Society

Fandrick and co-workers reported a highly enantioselec-
tive copper-catalyzed asymmetric carbonyl silylpropargy-
lation, but a costly and elaborate ligand was employed.9
In our recent efforts toward the synthesis of bryostatin
16, we required an asymmetric carbonyl propargylation
reaction that was straightforward and facile to employ.1c
We envisioned that the addition of allenylzinc species to
aldehydes would be ideal because the allenylzinc can be
generated in situ from a propargyl iodide via zinc-iodine
exchange. The use of allenylzinc reagents would obviate
the need to isolate sensitive allenylmetal reagents and
circumvent the use of toxic allenylstannane complexes.
The availability of a wide range of propargyl halides would
also allow for the formation of a variety of allenylzinc
reagents. Importantly, the corresponding addition of al-
kylzinc reagents to aldehydes is known to be ligand-
accelerated by a variety of amino alcohol ligands. We
hypothesized that the same would hold true for additions
with allenylzinc complexes, providing a strategy for the
development of an asymmetric propargylation reaction.
Chiral amino alcohol ligands can be easily accessed from a
commercially available chiral pool, which offers great
ligand tunability for various substrates.
Our interest in chiral aminoalcohol ligands is attributed
to their low cost, easy accessibility and high modularity.
Most of these ligands can be synthesized from the com-
mercially available amino acids via protection followed by
additions of Grignard or organolithium reagents (Scheme 1).
Each of these steps allows independent modification of the
chiral space of the ligands. Therefore, the chiral amino
alcohol offers a versatile template, enabling well-defined
structure-selectivity studies. Amino alcohols L present
three structural elements that may be independently opti-
mized: (a) the backbone of the ligand, (b) the amine
protecting group, and (c) the groups appended to the
tertiary carbinol center.
Scheme 1. Chiral Amino Alcohol Ligand Synthesis
Prior reports attempting to use these zinc reagents proved
disappointing.10 Unlike the ligand-accelerated additions of
diaryl- or dialkyl-zinc to carbonyl compounds, where the
uncatalyzed reaction undergoes a disfavored four-centered
transition structure,11 the uncatalyzed addition of allenyl-
zinc species12 to aldehydes proceeds via a six-centered
transition structure to afford the racemic homopropargyl
alcohol.13 The fast background reaction has hampered the
development of highly enantioselective ligand-accelerated
carbonyl propargylation reactions using allenylzinc re-
agents. Therefore, there is a great need to develop active
amino alcohol ligands that induce high chirality transfer.
Herein, we report the first systematic efforts toward the
development of a ligand-accelerated enantioselective pro-
pargylation of various aldehydes utilizing allenylzinc re-
agents, and readily accessible and highly tunable chiral
amino alcohols as putative ligands for Zn(II).
Our initial studies focused on the reaction of cinnamal-
dehyde (1a) with in situ-generated allenylzinc14 in the
presence of a catalytic amount of various amino alcohols
(Table 1). After an extensive ligand screen, we found that
the enantioselectivity of the reaction was sensititve to the
ring size of the amino alcohol ligands (entries 1-4).
Interestingly, while five-membered ring derivative L3 af-
forded the product with the highest enantiomeric ratio
(entry 3), both three- and six-membered ring systems failed
to induce any chirality (entries 1 and 4). Replacement of
the methylene group at the C-4 position with a sulfur atom
had a detrimental effect on the asymmetric induction of the
reaction (entry 5). We also found that the N-protecting
group of the aminoalcohol ligand has a great influence on
the enantioselectivity. Ligands with a free secondary amine
(L6) or a N-methyl protecting group (L7) afforded product
2a with lower er (entries 6 and 7). Highly bulky protecting
groups such as 9-anthracenylmethyl (L8) also furnished
the product with diminished selectivity (entry 8). We next
turned our attention to examining the impact of the
substituents at the carbinol carbon (entries 9-13). While
an electron-rich substituent slightly suppressed the enan-
tioselectivity of the reaction, an electron-deficient substi-
tuent gave comparable results to the phenyl system (entries
9 and 10). Using the ligand with alkyl substituents afforded
the product with low enantiomeric purity (entry 11).
Although 2-naphthyl-substituted ligand L12 afforded the
product with disappointing selectivity, 1-naphthyl-substi-
tuted ligand L13 provided homopropargyl alcohol 2a with
the highest asymmetric induction (entries 12 and 13).
With ligand L13 in hand, we explored the effect of various
solvents on the enantioselectivity (Table 2, entries 1-4).
(9) (a) Fandrick, D. R.; Fandrick, K. R.; Reeves, J. T.; Tan, Z. L.;
Tang, W. J.; Capacci, A. G.; Rodriguez, S.; Song, J. H. J.; Lee, H.; Yee,
N. K.; Senanayake, C. H. J. Am. Chem. Soc. 2010, 132, 7600–7601. For
copper-catalyzed asymmetric propargylation of ketones, see: (b) Shi,
S. L. L.; Xu, W.; Oisaki, K.; Kanai, M.; Shibasaki, M. J. Am. Chem. Soc.
2010, 132, 6638–6639.
(10) Several cases of applying amino alcohol ligated allenylzinc
species in total syntheses were reported, but they proceeded with very
low enantioselectivities, for examples, see: (a) Marino, J. P.; McClure,
M. S.; Holub, D. P.; Comasseto, J. V.; Tucci, F. C. J. Am. Chem. Soc.
2002, 124, 1664–1668. (b) Pommier, A.; Stepanenko, V.; Jarowicki, K.;
Kocienski, P. J. J. Org. Chem. 2003, 68, 4008–4013.
(11) Pu, L.; Yu, H. B. Chem. Rev. 2001, 101, 757–824 and references
cited therein.
(12) Propargylzinc and allenylzinc species are interconvertable
through a facile propargylic rearrangement. In the absence of a sub-
stiutent on the allenyl portion, allenylzinc species are thermodynami-
cally more stable than propargylzinc species. This results in selective
formation of the propargyl product. Marshall, J. A.; Gung, B. W.;
Grachan, M. L. Modern Allene Chemistry; VCH: Weinheim, Germany,
2004.
(13) Fandrick, D. R.; Fandrick, K. R.; Reeves, J. T.; Tan, Z.;
Johnson, C. S.; Lee, H.; Song, J. J.; Yee, N. K.; Senanayake, C. H.
Org. Lett. 2010, 12, 88–91.
(14) A mixture of propargyl/allenyl iodide solution (85 wt % in
toluene) was prepared from propargyl bromide solution (80 wt % in
toluene) and sodium iodide through a Finkelstein reaction protocol.
Org. Lett., Vol. 13, No. 8, 2011
1901

aldehyde. Interestingly, it is crucial to use a propargyl/
allenyl iodide solution (85 wt % in toluene). Performing the
reaction employing a neat proparyl/allenyl iodide reagent
afforded homopropargyl alcohol 2a with poorer yield and
selectivity (entry 11).16 The exact role of toluene additive
remains unclear, and is currently under investigation.
Table 1. Selected Ligand Optimization Studiesa
Table 2. Selected Reaction Optimization Experimentsa
entry
solventb
concn (M)
yield (%)c
erd
1
CH2Cl2
Toluene
THF
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.05
0.25
0.1
0.1
99
95
80
80
83
99
0
83:17
64:36
57:43
60:40
78:22
83:17
entry
ligand
conv(%)
erb
2
3
1
L1
100
100
100
100
100
100
100
100
100
100
100
100
100
50:50
44:56
2
L2c
L3
4
Hexane
e
5
CH2Cl2
3
75.5:24.5
50:50
e,f
4
L4c
L5
6
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
g
7
5
52.5:47.5
60:40
8
80
86
0j
81:19
77:23
6
L6
9
7
L7
65:35
h
j
10
11
8
L8
71:29
80
74:26
9
L9
70:30
10
11
12
13
L10
L11
L12c
L13
75.5:24.5
57:43
a See Supporting Information for detailed experimental procedures.
b Solvents were degassed by freeze-pump-thaw procedures. c Isolated
yields. d Er was determined by chiral HPLC. e CH2Cl2 was used without
degassing. f 2,6-Di-tert-butylphenol (50 mol %). g ZnMe2 (220 mol %)
was used. h Propargyl bromide (80 wt % in toluene) was used. i Only
ethyl transfer by product was obtained. j Neat propargyl/allenyl iodide
was used.
48.5:51.5
83:17
a See Supporting Information for detailed experimental procedures.
b Er was determined by chiral HPLC. c (R)-enantiomerwas used.
Under the standard reaction conditions using a mixture
of propargyl/allenyl iodide solution (85 wt % in toluene), a
range of alkenyl, aryl, heteroaryl, alkyl aldehydes were
surveyed to determine the scope of this method (Table 3).
While cinnamaldehyde afforded the product (2a) in ex-
cellent yield with 83:17 er, R-methyl cinnamaldehyde gave
the product (2b) with quantitative yield and enhanced
enantiomeric ratio (90:10) (entries 1 and 2). Aryl aldehydes
bearing different substituents proceeded smoothly to fur-
nish the corresponding products in good to excellent yields
with high levels of enantiopurity (entries 3-12). In addi-
tion, heteroaromatic aldehydes such as 2-formyl furan and
2-formyl thiophene, and aliphatic aldehydes participated
in the propargylation reaction with high yields and useful
levels of enantioselectivity (entries 13-15). The absolute
stereochemistry of 2cwas determined by comparison of the
HPLC retention time with that of the literature data,8c and
the stereochemistry of the other homopropargyl alcohols
were tentatively assigned by analogy.
We found that dichloromethane (CH2Cl2) was superior to
toluene, THF and hexane. It is noteworthy that CH2Cl2
needed to be degassed, otherwise the reaction afforded the
product with lower yield and enantioselectivity (entries 1
and 5). We believe this result is caused by the formation of
highly reactive propargyl radicals when oxygen is present.15
Indeed, performing the reaction in the presence of 2,6-di-
tert-butylphenol as a radical scavenger afforded the product
with quantitative yield and 83:17 er (entry 6). Replacing
diethylzinc with dimethylzinc failed to give any desired
product (entry 7). In additional studies, we found that
varying the reaction concentration diminished the reaction
conversion and the enantioselectivity (entries 8 and 9). The
source of propargyl/allenyl halide also has a remarkable
effect on the reaction. For example, propargyl bromide did
not give any homopropargylic alcohol product. Instead,
only the ethyl transfer byproduct was obtained (entry 10).
Presumably, this can be attributed to the slow zinc-bro-
mine exchange relative to the addition of ethyl group to the
(16) After screening a wide range of arene additives, we were able to
restore the reaction conversion and selectivity by addition of 150 mol %
naphthalene. However, the homopropargyl alcohol product was accom-
panied with 15% ethyl transfer byproduct. More studies are underway
to examine the exact role of an arene additive in this transformation.
(15) For reviews on dialkylzinc in radical reactions, see: (a) Bazin, S.;
Feray, L.; Bertrand, M. P. Chimia 2006, 60, 260–265. (b) Akindele, T.;
Yamada, K. I.; Tomioka, K. Acc. Chem. Res. 2009, 42, 345–355.
1902
Org. Lett., Vol. 13, No. 8, 2011

In summary, we have developed a tin-free catalytic
asymmetric propargylation of a wide range of aldehydes
employing allenylzinc reagents and readily available chiral
amino alcohol ligands. The simple nature of the reaction
protocol and the ready availability of the amino alcohol
ligand make this new asymmetric propargylation an at-
tractive alternative to currently existing methods. In addi-
tion, the low toxicity and low cost of the allenylzinc
reagents are of particular interest from an environmental
and an economic point of view. Further efforts will be
devoted to studying reaction mechanism, assessing syn-
thetic applications, and expansion of the substrate scope.
Table 3. Substrate Scope of the Ligand-Accelerated Asym-
metric Proparylation of Aldehydes 1a-1oa
entry
substrate
PhCHdCH-
yield (%)c
er (S/R)d
1
1a
1b
1c
1d
1e
1f
99
99
99
98
93
90
85
98
80
80
95
94
80
97
80
83:17
90:10
93:7
2
PhCHdC(Me)-
Ph-
3
4
p-MeO-C6H4-
p-Me-C6H4-
m-Me-C6H4-
o-Me-C6H4-
p-F-C6H4-
p-Cl-C6H4-
p-Br-C6H4-
2-Napthyl
4-MeO-1-napthyl
Furyl
96:4
5
94:6
6
92:8
7
1g
1h
1i
94.5:5.5
93:7
8
9
87.5:12.5
90.5:9.5
92:8
10
11
12
13
14
15
1j
1k
1l
96:4
1m
1n
1o
90:10
85:15
80:20
Thiophenyl
c-Hexyl
a See Supporting Information for detailed experimental procedures.
b
c
MS (4 A) was added to give better reproducibility. Isolated yields. d Er
˚
was determined by chiral HPLC.
While a detailed mechanism for the overall catalytic
process is yet to be determined, a plausible reaction mech-
anism and stereochemical model is depicted in Figure 1.17
Entry into the catalytic cycle via deprotonation of the
amino-alcohol ligand with diethylzinc species affords ac-
tive catalyst I.18 Coordination of the allenylzinc species
(generated in situ via the zinc-iodine exchange process) to
catalyst I furnishes dinuclear zinc complex II, which then
binds to the aldehyde to give intermediate III. The alde-
hyde approaches so as to avoid a steric interaction with the
benzyl group on the ligand and exposes its si-face to the
allenyl group (IIIa vs IIIb). Propargylation of the aldehyde
via a six-centered transition structure delivers propargyl
alkoxide IV with the stereochemistry in agreement with the
experimental data. Dissociation of zinc alkoxide V regen-
erates active catalyst I to complete the catalytic cycle.
Figure 1. (a) Proposed reaction mechanism and (b) stereoche-
mical model for the ligand-accelerated propargylation of
aldehydes.
Acknowledgment. We thank the National Science
Foundation for their generous support of our programs.
M.-Y.N. thanks the Croucher Foundation for a postdoc-
toral fellowship. G.D. thanks Stanford for a graduate
fellowship.
(17) The proposed dinuclear-zinc mechanism is based on the well-
developed aminoalcohol catalyzed asymmetric dialkyl- and diarylzinc
addition to aldehydes, see ref 11. One of the referees suggested a
mononuclear-zinc mechanism, which cannot be ruled out at this stage.
(18) The reaction does not show nonlinear effects (see Supporting
Information), so a single Zn-complex is depicted in the proposed
reaction mechanism.
Supporting Information Available. Experimental pro-
cedures and characterization data for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
Org. Lett., Vol. 13, No. 8, 2011
1903