Silyl fluoride electrophiles for the enantioselective synthesis of silylated pyrrolidine catalysts

Article abstract of DOI:10.1021/jo200991q

Chiral silylated pyrrolidine catalysts are obtained in high yield and enantioselectivity by sparteine-mediated lithiation of N-Boc-py rolidine and addition to silyl fluoride electrophiles. The activity and enantioselectivity of a new tert-butyldiphenylsilylpyrrolidine catalyst has been demonstrated for various asymmetric Michael reactions at 5 mol% catalyst loading and affords up to 99% ee for asymmetric Michael reactions with aldehydes and nitro-olefins. Acetaldehyde donors proceed with yields up to 77% and enantioselectivities up to 96% ee, avoiding common side reactions that often lower yields. Insight into the mechanism of pyrrolidine-based catalysts is provided by demonstrating ESI mass spectrometry evidence for activation of a nitro acceptor by formation of a hydrogen-bonding adduct with the catalyst amine. Analysis of reaction intermediates using mass spectrometry provides evidence that the pyrrolidine catalyst also plays a role in activating nitro-olefins through hydrogen-bonding.

Full text of DOI:10.1021/jo200991q

ARTICLE
pubs.acs.org/joc
Silyl Fluoride Electrophiles for the Enantioselective
Synthesis of Silylated Pyrrolidine Catalysts
Kaleb I. Jentzsch, Taewoo Min, Jennifer I. Etcheson, James C. Fettinger, and Annaliese K. Franz*
Department of Chemistry, University of California, One Shields Avenue, Davis, California 95616, United States
S Supporting Information
b
ABSTRACT: Chiral silylated pyrrolidine catalysts are obtained
in high yield and enantioselectivity by sparteine-mediated
lithiation of N-Boc-pyrrolidine and addition to silyl fluoride
electrophiles. The activity and enantioselectivity of a new tert-
butyldiphenylsilylpyrrolidine catalyst has been demonstrated for
various asymmetric Michael reactions at 5 mol % catalyst loading
and affords up to 99% ee for asymmetric Michael reactions with
aldehydes and nitro-olefins. Acetaldehyde donors proceed with
yields up to 77% and enantioselectivities up to 96% ee, avoiding
common side reactions that often lower yields. Insight into the
mechanism of pyrrolidine-based catalysts is provided by demonstrating ESI mass spectrometry evidence for activation of a nitro
acceptor by formation of a hydrogen-bonding adduct with the catalyst amine. Analysis of reaction intermediates using mass
spectrometry provides evidence that the pyrrolidine catalyst also plays a role in activating nitro-olefins through hydrogen-bonding.
’ INTRODUCTION
groups are most commonly utilized in protecting group strategies
for alcohols where a SiꢀO bond is formed upon substitution of a
silyl electrophile; the formation of a SiꢀC bond and the
corresponding quaternary silyl group can be readily accomplished
using the same substitution strategy.7ꢀ9
Enantioselective catalysis provides an efficient method to
access chiral molecules for synthetic building blocks, single
enantiomer pharmaceuticals, biological screens, and natural pro-
ducts. Traditionally, this field has been dominated by Lewis acid
and transition-metal-based catalysts; however, these catalysts tend
to be more toxic, expensive, and sensitive to moisture and air.
Recently, there has been an interest to develop organocatalysts
that have complementary reactivity to existing catalysts, with the
advantages of being cheaper to produce and easier to handle. The
development of new types of selective organocatalysts continues
to gain attention with the rising cost of metals and concerns for
environmental sustainability. Organocatalysts are increasingly
utilized in diverse synthetic transformations to obtain highly
functionalized complex targets.1ꢀ3 The activation mechanism
and stereocontrol elements of organocatalysts are of particular
interest because they are considered to mimic the active sites of
enzymes in biological systems.4
We are interested in the design of new catalysts incorporating
silicon functional groups to impart steric and/or electronic effects
to induce asymmetry.5 Silicon is positioned directly below carbon
on the periodic table and can be utilized in many synthetic
transformations that are similar to carbon, while also providing
effects andopportunitiesnotavailableforcarbon analogs. Because
of the electropositive nature of silicon and the reversal of bond
polarity relative to carbon, the incorporation of a silyl group may
alter the stabilization of transition states and intermediates when
organosilicon species are involved. With a covalent radius ap-
proximately 50% larger and bond lengths approximately 20%
longer than carbon,6 silicon can exert a greater steric influence
compared to related carbon analogues. Sterically demanding silyl
Herein we describe the use of silyl fluoride electrophiles for
the efficient synthesis of three enantiopure silylated pyrrolidine
catalysts including the new sterically demanding tert-butyldi-
phenylsilyl analogue. Although bulky silyl chloride electrophiles
are effective silylating agents for the protection of alcohols,10
their reactivity decreases as the steric interactions around the
silicon increase. The increased electrophilicity of a silyl fluoride
reagent11 compared to the silyl chloride reagent allows for
enantioselective reactions with sterically demanding silicon
centers. We have demonstrated that the silylated pyrrolidines
are effective catalysts for the asymmetric Michael addition of
aldehydes to nitro-olefins, providing high diastereoselectivity
and excellent enantioselectivity.12,13 Using ESI-MS, new me-
chanistic insight is provided regarding the Michael reaction
through the detection of an adduct suggesting hydrogen-bond-
ing activation of the nitroalkene by the aminocatalyst. While
examples of aminocatalysis are widespread in the literature,14ꢀ16
the full details regarding the role of weak stabilizing interactions
and steric factors that control selectivity have not been firmly
established,15,17 and thus, silyl variants may improve mechanistic
understanding and lead to new reactivity. Our research18
provides insight regarding this new class of silylated pyrrolidine
catalysts,19,20 as well as an efficient synthetic method to access
Received: May 17, 2011
Published: July 19, 2011
r
2011 American Chemical Society
7065
dx.doi.org/10.1021/jo200991q J. Org. Chem. 2011, 76, 7065–7075
|

The Journal of Organic Chemistry
ARTICLE
Table 1. Enantioselectivity of Silyl Fluoride Electrophiles for the Synthesis of Silylated Pyrrolidinesa
yield of
eeb of
yield enrichment
method A (%)
yield enrichment
method B (%)
eeb enrichment
A or B (%)
entry
Ph2RSi-X
yield of (S)-1 (%)
(S)-2 HBr (%)
(S)-2 HBr (%)
3
3
1
2
3
4
Ph2MeSiCl
Ph2MeSiF
Ph3SiF
84
91
93
90
ndc
86
85
86
46
95
92
88
nd
43
47
46
nd
76
69
68
nd
99
99
99
Ph2-t-BuSiF
a Enrichment method A: recrystallization from 99:2 CH2Cl2/MeOH. Enrichment method B: trituration with CH2Cl2/hexanes. b Determined by HPLC
analysis using chiral stationary phase after nitrogen protection with tosylate or benzoyl group; see the Supporting Information for details. c nd = not determined.
bulky silylated pyrrolidines with high yields and high enantio-
purity in two synthetic steps. This represents the first example of
silyl fluoride reagents being utilized in a substitution reaction for
enhanced enantioselectivity in the formation of SiꢀC bonds,21
which should represent a general strategy for enantioselective
synthesis of silicon-containing products for catalysis and biolo-
gical applications.22ꢀ24
We have also demonstrated that this silyl fluoride synthetic
strategy is effective to access the even bulkier triphenylsilyl
pyrrolidine (S)-2b and tert-butyldiphenylsilyl pyrrolidine (S)-
2c with high yields and excellent enantioselectivities that are only
slightly reduced by the bulkier silanes (92% and 88% ee, entries
3 and 4, respectively). Upon deprotection using ZnBr2 and
enantioenrichment with a simple trituration in CH2Cl2/hexane,
36
(S)-2aꢀc HBr are all obtained with 99% ee, with only a slight
3
decrease in yield. Using the same (ꢀ)-sparteine-mediated asym-
metric deprotonation pathway with chlorotriphenylsilane is
known to result in a nearly racemic mixture and 66% yield,19
and the tert-butyldiphenylsilyl chloride is sufficiently bulky such
that no reaction is observed. An alternate route to the triphe-
nylsilyl pyrrolidine (S)-2b, reported by Strohmann and co-
workers,20 employs diphenyldimethoxysilane followed by sub-
stitution of the silylether group with phenyllithium. While this
synthetic strategy affords excellent enantioselectivity of 99% ee
after recrystallization, the overall yield for this sequence was only
34%. Upon investigation of the triisopropylsilyl fluoride reagent,
we observed an apparent limit to this strategy, and no reaction
was observed. This outcome is attributed to the steric effect in
combination with an electronic effect resulting from the re-
placement of aryl groups with alkyl groups on the silicon.
A comparison of purification methods shows that trituration is
the most efficient method of purification while simultaneously
enriching the enantiopurity of the catalysts (Table 1). Treatment
’ RESULTS AND DISCUSSION
We have investigated silyl fluoride reagents as effective silylat-
ing agents with the goal to establish a general route for the
enantioselective formation of CꢀSi bonds and demonstrate the
efficient synthesis of silylated pyrrolidines (such as (S)-2aꢀc)
that incorporate bulky silanes with high enantioselectivity. Silyl
fluoride reagents such as diphenylmethylsilyl fluoride and tri-
phenylsilyl fluoride are commercially available, and others can be
synthesized in one step from the corresponding silyl chloride
reagent.25 To compare the rate and enantioselectivity between
silyl chloride and silyl fluoride electrophiles, we investigated the
synthesis of chiral silylpyrrolidines (S)-2a through (S)-2c using a
(ꢀ)-sparteine-mediated asymmetric deprotonation of N-Boc-
pyrrolidine26ꢀ29 with sec-BuLi, followed by addition to a silyl
halide reagent (Table 1). Initial use of a silyl chloride electrophile
for a moderately sterically hindered diphenylmethylsilyl group
afforded high yields of (S)-2a, but the observed enantioselectivity
was very low at only 46% ee after deprotection (entry 1). In
contrast, the silyl fluoride variant proceeded efficiently with an
increased enantioselectivity of 95% ee (entry 2). The dramatic
improvement in enantioselectivity is attributed to the enhanced
electrophilicity such that the silylation proceeds at a temperature
that is sufficiently low to maintain high configurational stability of
the chiral organolithium intermediate. The reduced enantios-
electivity with silyl chloride electrophiles suggests that the
addition does not proceed until the reaction begins to warm
up, which compromises the configurational stability of the
organolithium species and degrades the enantioselectivity.30,31
Although the increased reaction rates for silyl fluoride reagents
compared to silyl chloride reagents has been previously
reported,32ꢀ34 this is the first example using silyl fluoride
reagents as a strategy to enhance enantioselectivity for CꢀSi
bond formation in a substitution reaction.21,35
of catalysts (S)-2aꢀc HBr with Amberlyst-A2137 resin affords
3
the free amine catalysts (S)-2aꢀc in nearly quantitative yield
with no degradation of enantiopurity. The absolute configuration
of catalysts (S)-2aꢀc HBr was determined by X-ray crystal-
3
lographic analysis.38 It should be noted that both the (R)- and
(S)-configurations of the catalyst structures can be readily
accessed using the sparteine surrogates described by O’Brien
and co-workers.39,40
The Michael addition of propionaldehyde to β-nitrostyrene
was investigated as a model reaction to compare the catalytic
properties of silylated pyrrolidine catalysts (S)-2aꢀc (Table 2).
When catalyst (S)-2a HBr is used directly at 10 mol % without
3
adding triethylamine, we observed only low yields (20ꢀ30%)
after 5 days. Upon addition of triethylamine to generate the free
amine catalyst (S)-2a in situ, the reaction proceeds efficiently to
afford the Michael addition product in 1 h favoring the syn
7066
dx.doi.org/10.1021/jo200991q |J. Org. Chem. 2011, 76, 7065–7075

The Journal of Organic Chemistry
ARTICLE
Table 2. Catalyst Comparison for Michael Reaction
higher enantioselectivity of 88% ee using a toluene/THF
solvent system. Initially, an 88:12 diastereomeric ratio was
observed for (S)-2a HBr, but further investigations demon-
3
strated the importance of aldehyde purity on the rate and
diastereoselectivity, presumably due to the presence of trace
propionic acid when using excess aldehyde. A more optimal
93:7 diastereomeric ratio was obtained using freshly distilled
propionaldehyde with 5 mol % of propionic acid added
(Table 2, entry 5), which matches the Strohmann and Bolm
reports of a 93:7 and 95:5 diastereomeric ratio, respectively. It
was noted that the addition of catalytic propionic acid is
particularly necessary for faster reaction times of 1ꢀ5 h. When
comparing product results for the same model reaction using
catalyst (S)-2b,20 we observe a 97:3 ratio at 0 °C, compared to
the 91:9 diastereomeric ratio reported by Strohmann at 4 °C.
Thus, diastereoselectivity and reaction time are influenced by the
purity or batch of the propionaldehyde, which can account for
these variations. For consistently high diastereoselectivity and
faster reaction rates, these results indicate that the use of
freshly distilled propionaldehyde with 5 mol % of propionic acid
is optimal.
entry
catalyst
solvent
hexanes
hexanesd
drb syn/anti
eec (syn)
1
2
(S)-2a HBr
no reaction
88:12
89:11
88:12
93:7
3
(S)-2a HBr
81
82
82
78
92
93
93
98
98
99
99
3
3
(S)-2a
hexanes
4
(S)-2a HBr
9:1 hexanes/THFd
9:1 hexanes/THFd
hexanesd
3
5
(S)-2a HBre
3
6
(S)-2b HBr
95:5
3
7
(S)-2b
hexanes
94:6
8
(S)-2b HBr
9:1 hexanes/THFd
hexanesd
97:3
91:9f
92:8f
92:8f
3
9
(S)-2c HBr
3
10
11
12
(S)-2c
hexanes
X-ray structure analysis of catalysts (S)-2aꢀc HBr provides
3
(S)-2c HBr
9:1 hexanes/THFd
hexanesg
an opportunity to compare the steric volume of the silyl side
chains and analyze the factors that account for the observed
selectivity.41ꢀ44 Bond lengths between the pyrrolidine carbon
3
(R)-7
93:7
a Conversion calculated using 1H NMR spectroscopy with 3,4,5-trichlor-
opyridine as an internal standard. b Determined by 1H NMR analysis of
unpurified reaction. c Determined by HPLC analysis with OD-H chiral
stationary phase. d Addition of 5 mol % of Et3N to reaction. e Reaction run
with distilled propionaldehyde and 5 mol % of propionic acid added;
and silicon are approximately 1.91 Å for catalysts (S)-2a-c HBr,
3
which is 20% longer than a typical carbonꢀcarbon bond of 1.54
Å (Table 3). Using a simple geometric equation for the volume
of an irregular tetrahedron, an estimate for the steric volume of
the side chains on pyrrolidine was calculated (Table 3).45 Using
the steric volume alone as a predicting tool for selectivity would
catalyst (S)-2a HBr was 93% ee for this experiment. Reaction time is 3 h.
3
f When using distilled propionaldehyde with either 5 or 100 mol % of
propionic acid added, a diastereomeric ratio of 94:6ꢀ98:2 was obtained.
See the Supporting Information for details. g Reaction run under
literature conditions;12 structure of (R)-7 shown in Table 3.
suggest that catalysts (S)-2c HBr, (S)-6, and the silyl ether (R)-
3
7 should impart a similar enantio- and diastereoselectivity to the
corresponding Michael products. In fact, (S)-2c HBr and (R)-
3
7 give very similar results in terms of yield and enantioselec-
tivity (Table 2, entries 11 and 12); however, (S)-6 affords a
lower diastereomeric ratio of 78:22.12,20 These results indicate
that steric factors are not solely responsible for catalyst perfor-
mance. In addition to the steric effects, the silicon may affect
secondary weak interactions of the polar intermediates that play
a role in stereoselectivity. The electropositive nature of the
silicon atom leads to a reversal of charge distribution, and the
ability of a silyl group to stabilize a β-silyl carbocation is well-
documented.46ꢀ48 An increase in the steric volume correlates
with a higher diastereoselectivity, while the presence of a bulky
nonplanar substituent, such as a tert-butyl or trimethylsilyl ether
group, correlates with a higher enantioselectivity. Catalyst (S)-
2a containing a simple methyl group affords products with
reduced enantioselectivity as expected due to the smaller steric
volume.
addition product (5aa) with 88:12 diastereoselectivity. Optimi-
zation experiments revealed that nonpolar solvents such as
hexanes are optimal for rate and diastereoselectivity, and catalysts
(S)-2aꢀc HBr are active with 5 mol % catalyst loading without
3
reducing the reaction rate or stereoselectivity. Comparing the
purified free amine catalysts with the triethylamine-treated HBr
salts showed no significant change in the rate, selectivity, or yield
for the Michael reaction of propionaldehyde with nitrostyrene
(Table 2); however, it was noted that the purity of propionalde-
hyde can effect the diastereoselectivity (entry 5, vide infra). For
practical reasons, it was preferred to use triethylamine treatment
of (S)-2a-c HBr, which is crystalline, for easier handling since
3
the free amine structures are liquids. To increase solubility, THF
was also investigated as a 10% cosolvent mixture with hexanes,19
and comparable yields are observed. On the basis of this
preliminary comparison, pyrrolidines (S)-2b HBr and (S)-2c
The scope of Michael donors and acceptors was then
3
3
HBr were identified as the most selective catalysts, with (S)-
explored with catalyst (S)-2c HBr (Tables 4 and 5). The
3
2c HBr imparting the highest enantioselectivity (entries 9ꢀ11).
silylated catalyst generally provides high yields and excellent
enantio- and diastereocontrol for a variety of donor and
acceptor partners. Hydroxy and alkoxy aldehydes have been
previously reported for conjugate additions with nitroolefins,
generally giving high yields but only moderate to low diastereo-
selectivity.49,50 For our system, the free amine catalyst (S)-2c
was necessary for a reasonable reaction rate with the benzylox-
yacetaldehdye donor and gives comparable results of both yield
and diastereomeric ratio to literature catalysts (Table 4, entry 5).49
The more challenging branched aldehyde donors proceed with
3
Comparing results for catalyst (S)-2a in the Michael addition
reaction to those previously reported by both Strohmann20 and
Bolm19 highlights the variations in selectivity observed based
on different reaction conditions. While yields are high across all
three examples, the reaction times and selectivities for the
model reaction of propionaldehyde to nitrostyrene show some
variation based on reaction conditions. An 82% ee is reported
both here and by Strohmann and co-workers using hexanes
solvent systems, whereas Bolm and co-workers reported a
7067
dx.doi.org/10.1021/jo200991q |J. Org. Chem. 2011, 76, 7065–7075

The Journal of Organic Chemistry
ARTICLE
Table 3. Steric Comparison of Pyrrolidine Catalysts and Correlation to Selectivity Observed in the Michael Reaction of
Propionaldehyde to Nitrostyrenea
pyrrolidine CꢀX
bond length (Å)
XꢀPh bond
length (Å)
XꢀR bond
length (Å)
tetrahedral
volume (Å3)
catalyst
X (Si or C)
R =
dr syn/anti
ee (syn)
(S)-2a HBr
Si
Si
Si
C
C
CH3
Ph
1.91
1.91
1.92
1.57
1.54
1.87
1.87
1.88
1.55
1.54
1.85
1.87
1.90
1.55
1.42
14.87
28.78
19.36
21.98
20.27
93:7
97:3
92:8
78:22b
93:7
78
93
3
(S)-2b HBr
3
(S)-2c HBr
t-Bu
Ph
99
96b
3
(S)-6 HCl
3
(R)-7 HBrc
OTMS
99
3
a Product 5aa selectivity data taken from Table 2 entries 5, 8, 11, and 12. b Product 5aa selectivity data for catalyst (S)-6 reported in the literature.20
c Catalyst (R)-7 HBr was crystallized as the protonated tetrabromozincate(II) salt monohydrate, with two protonated catalyst units per ZnBr4 unit; see
3
the Supporting Information.
Table 4. Michael Donor Scope with Catalyst (S)-2c HBr
Table 5. Michael Acceptor Scope with Catalyst (S)-2c HBr
3
3
time (h) yield (%) dra (syn/anti) eeb (syn)
entry
acceptor Ar =
5
yield (%)
dra (syn/anti)
eeb (syn)
entry donor R =
5
1
2c
3c
4c
5c
Ph
5aa
5ab
5ac
5ad
5ae
87
82
88
80
96
92:8
97:3
98:2
98:2
97:3
99
95
92d
95e
92
1
2
CH3
H
5aa
5ba
5ca
5da
5ea
1
96
2
87
92:8
n/a
99
95
97
93
95
52
56
4-ClC6H4
2-FC6H4
4-MeOC6H4
75
3
n-Bu
Bnc
OBnd
93
93:7
94:6
61:39
n/a
4
5
89
2-Furyl
1
5
8
76
a Determined by H NMR analysis of unpurified reaction. b Deter-
6
7e
(CH3)2 5fa
(CH3)2 5fa
168
168
168
96
26
30
mined by HPLC analysis with chiral stationary phase. c Reactions
n/a
performed with 5 mol % of propionic acid and catalyst (S)-2c HBr
3
8
9e
i-Pr
i-Pr
5ga
5ga
no reaction
81
was 96% ee. d Determined after reductive amination to the 2,3-
disubstituted pyrrolidine followed by N-protection using 4-MeO-
benzoyl chloride; see structure 5ac0. e Determined after reduction to
the alcohol using NaBH4.
87:13
90f
a Determined by 1H NMR analysis of unpurified reaction. b Determined
by HPLC analysis with chiral stationary phase. c Chloroform used as
solvent. d Reaction run using the free amine catalyst (S)-2c at 10 mol %
and without addition of Et3N. e Reaction run at 20 mol % catalyst
loading. f Determined after reduction to the alcohol using NaBH4.
substrates had minimal effect on the isobutyraldehyde donor,
while the longer chain isovaleraldehyde donor became active to
afford an 81% yield (Table 4, entries 7 and 9). When investi-
gating the acceptor scope (Table 5), it was noted that con-
sistently high enantioselectivity is observed and the addition
reaction typically completes in 3ꢀ5 h. These selectivity results
are consistent with the literature reports for DFT computa-
tional studies for aminocatalysts and also observed experimen-
tal results.53ꢀ55
low yield or no reaction (Table 4, entries 6 and 8), and we
attribute this dramatically reduced catalytic activity to a slow
iminiumꢀenamine formation51 due to disfavored steric inter-
actions by the increased steric bulk of the silyl chain. While
branched substrates are known to exhibit dramatic solvent
effects in the Michael reaction (giving yields from 10 to 96%
depending only on the solvent used),52 changing solvent
systems showed no effect on the yields with our catalyst.
Increasing the catalyst loading to 20 mol % for the branched
Using catalyst (S)-2c with 20 mol % catalyst loading, the
Michael reaction with acetaldehyde affords high yields and up to
96% ee for various nitrostyrenes (Table 6). This reaction also
7068
dx.doi.org/10.1021/jo200991q |J. Org. Chem. 2011, 76, 7065–7075

The Journal of Organic Chemistry
ARTICLE
proceeds using 5 mol % catalyst loading, but longer reaction
times, up to 4 days, are required (entry 2). The Michael addition
of acetaldehyde is known to be a particular challenge due to the
formation of significant byproduct.56,57 List and co-workers have
previously reported that prolinol silyl ether catalysts, such as (R)-7,
are suitable catalysts for acetaldehyde Michael donors; how-
ever, using bulkier silyl ethers tends to give lower product
conversion with no significant increase in enantioselectivity.56
Although our earlier experiments suggested no significant ad-
vantage to using the silyl pyrrolidine as the free amine catalyst,
here with the acetaldehyde reaction we observed that triethyla-
mine conditions promote significant byproduct formation and
afford lower product conversion (entry 3). A control experiment
where the purified acetaldehyde Michael product 5bc was
combined with triethylamine and excess acetaldehyde showed
signs of degradation after 30 min, and within 48 h the Michael
product had been completely consumed. Using the free amine
silyl pyrrolidine was sufficient to overcome this problem and
provide the acetaldehyde products with acceptable yields. Com-
paring a reaction with freshly distilled acetaldehyde showed no
effect on the reaction rate or enantioselectivity; however, a
reduced yield was observed with the addition of 20 mol %
propionic acid (entry 4 vs 5).
The enamine mechanism for aldol and Michael reactions
with aminocatalysts has been extensively discussed in the litera-
ture.58,59 To complement the existing mechanistic insight, we
monitored the formation of intermediates using ESI-MS where
even low concentrations of intermediates containing the catalyst
can be directly observed due to the ionization sensitivity of the
instrument.60ꢀ62 For this purpose, a 10 μL aliquot from a
mixture of catalyst (S)-2c, nitrostyrene, and propionaldehyde
under reaction conditions was diluted into 2 mL of MeOH and
injected for MS analysis (positive mode). Immediately after
addition of the aldehyde, masses corresponding to enamine
formation 8 (m/z 350.2, calcd; m/z 350.0 observed), nitro
complexation 9 (m/z 459.2, calcd; m/z 459.0 observed), and
Michael adduct formation 10 (m/z 499.3, calcd; m/z 499.4
observed) were observed by ESI-MS (Figure 1).
Table 6. Acetaldehyde Scope for Michael Reaction with
Catalyst (S)-2c HBr
3
entry
acceptor Ar =
5
time (h)
yield (%)
eea
1
2b
3c
4
Ph
5ba
5ba
5ba
5bb
5bb
5bc
5bd
5be
3
96
3
75
75
56
77
43
76
62
63
95
nd
nd
92
93
96
92
90
Ph
Ph
4-ClC6H4
4-ClC6H4
2-FC6H4
4-MeOC6H4
2-Furyl
3
5d
3
Of particular note is the observation of m/z 459 corresponding
to the formation of adduct 9, which indicates a binding interaction
between the nitrostyrene and the silylated pyrrolidine catalyst.
The formation of this adduct, as observed by ESI-MS, suggests a
binding interaction between either the NꢀH or silicon group
from the catalyst and an oxygen on the nitro group. From the
literature, both 1H and 29Si NMR experiments provide evidence
for these types of binding interactions between nitroolefins and
6
3
7
48
48
8
a Determined by HPLC analysis on chiral stationary phase after reduc-
tion to the alcohol using NaBH4. b Reaction run with 5 mol % of catalyst.
c Reaction run with 20 mol % of catalyst (S)-2c HBr in the presence of
3
20 mol % of Et3N. d Reaction run with 20 mol % of propionic acid added.
Figure 1. ESI-MS analysis (observed m/z values) of enamine mechanism and proposed activation of nitro-olefin with catalyst (S)-2c: (A) enamine
observed upon mixing catalyst (S)-2c with propionaldehyde; (B) observation of hydrogen-bonding complex upon mixing catalyst (S)-2c with
nitrostyrene; (C) reaction mixture with 5 mol % of catalyst (S)-2c.
7069
dx.doi.org/10.1021/jo200991q |J. Org. Chem. 2011, 76, 7065–7075

The Journal of Organic Chemistry
ARTICLE
Figure 2. ESI-MS evidence of nitrostyrene binding to pyrrolidines and thioureas.
various binding donors.15,62 Hydrogen-bonding activation of
electrophiles containing nitro groups has been previously pro-
posed for thiourea62ꢀ64 catalysts, and Lewis acid activation65 of
electrophiles is well-known. In order to further understand the
observed binding of the catalyst to the nitro acceptor, we
performed a series of mass spectrometry experiments with various
pyrrolidine species and also a thiourea to compare with previous
reports (Figure 2). Donor and acceptor pairs were combined in a
1:1 ratio in methanol, allowed to sit for 10 min, and then further
diluted into methanol just prior to injection for ESI-MS analysis
(positive mode). Evidence for adduct formation was observed for
the thiourea and all NH pyrrolidines, including a binding of
catalyst (S)-2c to 4-chloronitrostyrene (11) where the with-
drawing nature of the Cl was expected to diminish the interaction.
Only the N-protected pyrrolidine (S)-1b did not show any adduct
formation by MS. This result suggests that silyl pyrrolidine
catalyst (S)-2c can activate nitro-olefin acceptors through hydro-
gen-bonding between the catalyst NꢀH and the nitro group,
rather than an interaction between the Lewis basic oxygen of the
nitro group with the electropositive silicon.66 While previous
reports from Okino et al.63 and Li et al.62 have demonstrated
hydrogen bonding between nitro groups and thiourea organoca-
talysts using 1H NMR titration studies, this is the first evidence for
pyrrolidine activation of a nitro acceptor through hydrogen-
bonding interactions supported by ESI-MS. Therefore, this
suggests evidence for a transition-state model involving two
pyrrolidine catalyst molecules: one pyrrolidine activating the
carbonyl donor through enamine formation and a second pyrro-
lidine activating the nitro-olefin acceptor through hydrogen-
bonding.
formation of a hydrogen-bonding adduct rather than the
covalent conjugate addition product.
’ CONCLUSIONS
In conclusion, we have demonstrated an efficient synthesis of
silylated pyrrolidine catalysts using the enhanced reactivity of
silyl fluoride electrophiles to improve yield and enantioselec-
tivity. While the increased reaction rates for silyl fluoride reagents
compared to silyl chloride reagents has been previously reported,
this is the first case that demonstrates the use of silyl fluoride
reagents as a strategy to enhance enantioselectivity for CꢀSi
bond formation in a substitution reaction.21 Thus, this may
provide a general strategy for enantioselective silylation. Here,
this method allows the preparation of several sterically demand-
ing silylated pyrrolidines that are difficult to synthesize or not
otherwise attainable. We have demonstrated the activity and
enantioselectivity of the new (S)-tert-butyldiphenylsilyl pyrroli-
dine catalyst (S)-2c for various asymmetric Michael reactions at 5
mol % catalyst loading and specifically highlighted the utility of
this catalyst for the more challenging reactions with acetalde-
hyde. We provide further insight into the mechanism of pyrro-
lidine-based catalysts by demonstrating the first example of ESI-
MS evidence for activation of a nitro acceptor by formation of a
hydrogen-bonding adduct with the catalyst amine. Incorporating
silicon as a functional group into existing and future catalyst
structures may be useful to enhance catalytic properties for
improved enantio- and diastereoselectivity, and lower the neces-
sary catalyst loading. Exploration of new enantioselective catalyst
structures based on the properties of silicon are currently under-
way in our laboratory.
It can also be considered that the conjugate addition of a
pyrrolidine catalyst to the nitroacceptor provides an alternate
explanation for the adduct observed at m/z 459. While this
conjugate addition may be facilitated under MS conditions,
’ EXPERIMENTAL SECTION
General Information. Commercially available reagents were
obtained from commercial sources and used without further purification
unless otherwise indicated, particularly to be noted for propionaldehyde,
which was freshly distilled for some reactions. The following abbrevia-
tions are used throughout: ethyl acetate (EtOAc), acetonitrile (MeCN),
dichloromethane (DCM), isopropanol (IPA), methanol (MeOH),
enantiomeric excess (ee), triethylamine (Et3N). All Michael addition
reactions were performed in glass vials with Teflon caps and exposure to
atmospheric conditions. All 1H and 13C spectra were recorded at
ambient temperature at 300, 400, and 600 MHz or 75, 100, and 150
MHz, respectively. 29Si NMR spectra were recorded at ambient tem-
1
experiments performed using H NMR spectroscopy do not
detect formation of the conjugate addition product. When 1H
NMR spectroscopy is used to monitor a mixture of catalyst (S)-
2a with nitrostyrene in a 1:1 mixture in deuterated chloroform,
no new signals are observed. Based on previous reports,
diagnostic signals for the formation of a conjugate addition
product would appear as two signals near 4.4 and 4.9 ppm,
corresponding to two protons R to the nitro group.67 If the
conjugate addition product is formed, then the concentration is
too low to be detected using NMR methods. The similar adduct
formation observed for both thiourea and pyrrolidine experi-
ments would also indicate a common opportunity for the
1
perature at 119 MHz. The H spectral data are reported as follows:
chemical shift in parts per million downfield from tetramethylsilane on
7070
dx.doi.org/10.1021/jo200991q |J. Org. Chem. 2011, 76, 7065–7075

The Journal of Organic Chemistry
ARTICLE
the δ scale, multiplicity (s, singlet; d, doublet; t, triplet; q, quartet; s,
septet; m, multiplet; dd, doublet of doublets, and b, broadened),
coupling constant (Hz), and integration. Carbon NMR chemical shifts
are reported in ppm from tetramethylsilane with the solvent reference
employed as the internal standard (deuterochloroform (CDCl3)) at
77.16 ppm.
Data for (S)-1a. Product is a clearoil, 0.586 g, 91% yield: 1HNMR(300
MHz, CDCl3, mixture of rotomers, major peaks provided) δ 7.64ꢀ7.48
(m, 4H), 7.39ꢀ7.27 (m, 6H), 4.02ꢀ3.80 (m, 1H), 3.61ꢀ3.29 (m, 1H),
3.18ꢀ2.95 (m, 1H), 2.17ꢀ1.90 (m, 1H), 1.89ꢀ1.76 (m, 1H), 1.71ꢀ1.59
(m, 1H), 1.51ꢀ1.41 (m, 1H), 1.27 (s, 9H), 0.65 (s, 3H); 13C NMR
(75 MHz, CDCl3) δ 154.8, 154.4, 136.7, 135.5, 134.9, 134.7, 129.2, 127.6,
79.2, 78.4, 46.6, 28.6, 28.3, 25.7, 24.5, ꢀ3.5, ꢀ4.3; LRMS (ESI) m/z [M +
H]+ calcd for C22H30NO2Si 368.2, found 368.4.
Data for (S)-1b. The triphenylfluorosilane was predissolved in a
minimum of Et2O (∼1.5 mL) prior to addition to the organolithium
reaction mixture. Product is a white solid, 0.700 g, 93% yield: 1H NMR (300
MHz, CDCl3, mixture of rotomers, major peaks provided) δ 7.65ꢀ7.51 (m,
6H), 7.42ꢀ7.29 (m, 9H), 4.39ꢀ4.16 (m, 1H), 3.62ꢀ3.32 (m, 1H),
3.29ꢀ3.03 (m, 1H), 2.13ꢀ1.86 (m, 2H), 1.76ꢀ1.58 (m, 1H), 1.30ꢀ1.21
(m, 1H) 1.05 (s, 9H, and minor rotomer peak at 1.17); 13C NMR (75 MHz,
CDCl3) δ154.7, 136.3, 134.5, 133.7, 129.4, 127.7, 79.4, 78.6, 46.7, 46.1, 29.4,
28.2, 26.0, 24.5; LRMS (ESI) m/z [M + H]+ calcd for C27H32NO2Si 430.2,
found 430.5.
Unless otherwise indicated, all chiral stationary phase HPLC analyses
were performed with a Daicel CHIRALCEL OD-H column (4.6 ꢁ 250 mm,
5 μm), CHIRALPAK AD-H column (4.6 ꢁ 250 mm, 5 μm), or
CHIRALPAK AS-H column (4.6 ꢁ 250 mm, 5 μm), with corresponding
guard columns, with a flow rate of 1.0 mL/min (2-propanol/hexanes
isocratic system) using a photodiode array detector and 40 °C column
oven temperature. Compounds were analyzed by HRMS on an orbitrap
spectrometer using electrospray ionization in the positive ion mode at
>60000 resolution and using typical ESI source values. These settings
result in mass accuracies <5 ppm. Some compounds were analyzed by
LRMS in the positive ion mode on a Qtrap spectrometer. Source
parameters were 5 kV spray voltage, with a curtain plate temperature of
275 °C and sheath gas setting of 15. Samples were analyzed via flow injection
analysis by injecting 20 μL samples into a stream of 80% MeOH/20%
aqueous solution with 0.1% formic acid, flowing at 300 μL/min. Optical
rotations were obtained on a polarimeter at a wavelength of 589 nm
(sodium D line) using a 1.0 dm cell. Specific rotations are reported in
degrees per decimeter at 23 °C, and the concentrations are given in
grams per 100 mL of solvent. Solvents used for optical rotations were
MeOH (reagent grade) and CHCl3 (stabilized with 0.5ꢀ1% EtOH, and
filtered through basic alumina).
Data for (S)-1c. Product is a clear oil, 0.646 g, 90% yield. [R]22
=
D
42.3 (c 0.299, MeOH); 1H NMR (600 MHz, CDCl3) δ 7.80ꢀ7.54 (m,
4H), 7.43ꢀ7.28 (m, 6H), 4.51ꢀ4.43 (m, 1H), 3.61ꢀ3.29 (m, 1H),
2.77ꢀ2.61 (m, 1H), 2.12ꢀ1.71 (m, 2H), 1.47 (s, 9H), 1.13 (s, 9H),
1.10ꢀ1.05 (m, 1H), 0.95ꢀ0.83 (m, 1H); 13C NMR (150 MHz, CDCl3)
δ 155.5, 136.8, 134.9, 134.50, 134.48, 130.3, 129.3, 129.2, 128.0, 127.7,
127.5, 79.0, 47.0, 44.8, 28.7, 28.0, 26.7, 26.1, 24.5, 18.8; HRMS (ESI) m/z
[M + H]+ calcd for C25H36NO2Si 410.2510, found 410.2509.
General Procedure for the Synthesis of Catalysts (S)-
When indicated, the progress of reactions was monitored by analytical
thin-layer chromatography using glass plates precoated with silica gel 60
F254 and visualized with UV light. Flash chromatography was per-
formed either on silica gel 60 Å (0.035ꢀ0.070 mm) or silica gel 150 Å
grade 62 (60ꢀ200 mesh).
2aꢀc HBr. A solution of (S)-1 (1.46 mmol, 1.0 equiv) in 15 mL of
3
anhydrous DCM was placed in a 100 mL round-bottom flask with a stir
bar and septum. Zinc bromide (1.65 g, 7.31 mmol, 5 equiv) was added to
the flask, and the reaction was allowed to stir for 15 h. Deionized water
(20 mL) was added to the reaction and stirring continued for 1 h. The
aqueous and organic layers were separated, and the aqueous layer was
extracted with 3 ꢁ 15 mL of DCM. The combined organic layers were
dried over Na2SO4, filtered, and concentrated to give a tan solid. The
catalyst was purified and enantioenriched upon trituration. The crude
product was dissolved in a minimal amount of DCM (∼5ꢀ15 mL)
ignoring small amounts of solid that do not completely dissolve.
Hexanes was added (20 to 50 mL), and the mixture was swirled to
General Procedure for the Conversion of Silyl Chloride
Reagents to Silyl Fluoride Reagents. A solution of tert-butyldi-
phenylchlorosilane (5.0 g, 18.19 mmol, 1.0 equiv) in anhydrous di-
methoxyethane (120 mL) was placed in an oven-dried, Ar-purged
250 mL round-bottom flask with a stir bar. Hexafluoroammonium silicate
(6.481 g, 36.38 mmol, 2.0 equiv) was added to the flask, and a reflux
condenser with positive Ar pressure was connected to the flask. The
reaction was refluxed for 5 days, allowed to cool to room temperature,
and poured over saturated aqueous ammonium chloride, the layers were
separated, and the aqueous layer was extracted 3 ꢁ 15 mL with DCM.
The combined organic layers were dried over Na2SO4, filtered, and
concentrated to give a brown oil. The crude product was purified through
vacuum distillation (bp 87 °C at 4.5 ꢁ 10ꢀ4 Torr) to give 3.83 g (81%) of
tert-butyldiphenylfluorosilane as a light yellow oil.
mix the solvents, where the purified catalyst (S)-2 HBr precipitated out
3
as a white solid that was isolated by suction filtration, rinsing with
hexanes. The product obtained was determined to be 99% ee on the
basis of chiral stationary phase HPLC analysis after N-protection. In one
out of seven cases for (S)-2c HBr, trituration afforded the catalyst with
3
96% ee instead of the expected 99% ee.
Data for (S)-2a HBr. Product is a white solid, 0.396 g 84% yield: mp
3
163ꢀ165 °C dec; [R]22D = +1.9 (c 0.316, MeOH); 1H NMR (300 MHz,
CDCl3) δ9.02 (bs, 2H), 7.76 ꢀ 7.63 (m, 2H), 7.63 ꢀ 7.50 (m, 2H), 7.49 ꢀ
7.30 (m, 6H), 3.29 (dd, J = 9.3, 9.3 Hz, 1H), 3.09 ꢀ 2.81 (m, 2H), 2.17 ꢀ
2.00 (m, 1H), 1.93 ꢀ 1.65 (m, 3H), 0.90 (s, 3H); 13C NMR (150 MHz,
CDCl3) δ 135.3, 135.1, 132.4, 132.1, 130.4, 128.5, 128.4, 47.3, 46.8, 28.3,
25.0, ꢀ5.5; 29Si NMR (119 MHz, CDCl3) δ ꢀ8.61; HRMS (ESI) m/z
[M ꢀ Br]+ calcd for C17H22NSi 268.1516, found 268.1507. Slow re-
crystallization from DCM and hexanes afforded single crystals, and the
absolute configuration was determined to be S by X-ray analysis. The
catalyst was derivatized to the N-benzoyl structure for chiral stationary
phase HPLC analysis: 99.5% ee, see (S)-15a below.
General Procedure for the Synthesis of N-Boc-Protected
Catalysts (S)-1aꢀc. A solution ofN-Boc-pyrrolidine (0.30 g, 1.75 mmol,
1.0 equiv) in anhydrous diethyl ether (8.8 mL) was placed in an oven-dried,
argon-purged 25 mL round-bottom flask with a stir bar, septum, and positive
argon pressure. (ꢀ)-Sparteine (0.52 mL, 2.28 mmol, 1.3 equiv) was added
through the septum, and the reaction was cooled to ꢀ78 °C in a dry ice/
acetone bath. Freshly titrated sec-butyl lithium (2.28 mmol, 1.3 equiv) was
added dropwise, and the reaction was allowed to stir at ꢀ78 °C. After 5.5 h,
the fluorosilane (2.63 mmol, 1.5 equiv) was added dropwise, and the mixture
was allowed to stir at ꢀ78 °C for 15 min and then allowed to warm to room
temperature for 15 min. The reaction was quenched by the addition of
deionized water (20 mL), the aqueous and organic layers were separated,
and the aqueous layer was extracted with 3 ꢁ 15 mL of EtOAc. The
combined organic layers were washed with 100 mL of 5% aqueous
phosphoric acid and then with brine, dried over Na2SO4, filtered, and then
concentrated to give a light yellow oil. The crude product was purified on a
flash silica gel column (100% hexanes to 90:10 hexanes/EtOAc) to give (S)-
1aꢀc as a clear oil.
Data for (S)-2b HBr. Product is a white solid, 0.496 g, 74% yield:
3
1
mp = 277ꢀ279 °C dec; [R]22 = +16.7 (c 0.318, MeOH); H NMR
D
(300 MHz, CDCl3) δ 9.53 (bs, 2H), 7.73 ꢀ 7.63 (m, 5H), 7.50 ꢀ 7.37
(m, 10H), 3.73 (dd, J = 9.0, 9.0 Hz, 1H), 3.02 (ddd, J = 15.4, 5.7, 5.7 Hz,
1H), 2.88 ꢀ 2.66 (m, 1H), 2.44 ꢀ 2.21 (m, 1H), 1.95 ꢀ 1.67 (m, 3H);
13C NMR (150 MHz, CD3OD) δ 137.1, 132.0, 131.7, 129.7, 48.7, 48.3,
29.5, 26.2; HRMS (ESI) m/z [M ꢀ Br]+ calcd for C22H24NSi 330.1672,
7071
dx.doi.org/10.1021/jo200991q |J. Org. Chem. 2011, 76, 7065–7075

The Journal of Organic Chemistry
ARTICLE
found 330.1676. Slow recrystallization from DCM and hexanes afforded
single crystals, and the absolute configuration was determined to be (S)
by X-ray analysis. The catalyst was derivatized to the N-benzoyl structure
for chiral stationary-phase HPLC analysis: 99.6% ee, see (S)-15b below.
Aldehyde (3.74 mmol, 10 equiv) was then added, and the reaction was
allowed to stir until complete as judged by TLC (100% CHCl3) and then
quenched by the addition of 1.0 N HCl (5 mL). The organic layer was
separated, andtheaqueous layer extracted with 3 ꢁ 10 mLof EtOAc. The
combined organic layers were dried with Na2SO4, filtered, and concen-
trated to give a colored oil. The crude product was purified on a flash
silica-gel column (hexanes/DCM). The purity of propionaldehyde was
shown to affect the reaction rate and diastereoselectivity, although the
enantioselectivity was not affected. See the Supporting Information for a
full table comparing selectivity for reactions performed with undistilled vs
distilled propionaldehyde and the effect of adding 5ꢀ100 mol %
propionic acid.
(2S,3R)-2-Methyl-4-nitro-3-phenylbutanal (5aa). Prepared from
0.0558 g of trans-β-nitrostyrene and propanal according to general
procedure A to afford 0.0674 g of 5aa in 87% yield after 1 h. HPLC
analysis: Chiralcel OD-H column, n-hexane/i-PrOH = 90:10, 1.0 mL/min,
40 °C, tR (syn, major) = 15.9 min; tR (syn, minor) = 21.7 min; tR (anti,
major) = 18.9 min; tR (anti, minor) = 25.4 min. 99.4% ee. dr = 92:8 based
on 1H NMR analysis of unpurified reaction. Spectral data are consistent
with literature values.68
Data for (S)-2c HBr. Product is a white solid, 0.432 g, 76% yield:
3
1
mp = 187ꢀ189 °C dec; [R]23 = +28.5 (c 0.302, MeOH); H NMR
D
(300 MHz, CDCl3) δ 10.56 (bs, 2H), 7.78 (dd, J = 6.4, 2.9 Hz, 2H), 7.69
(dd, J = 7.4, 1.6 Hz, 2H), 7.51 ꢀ 7.38 (m, 6H), 3.61 (dd, J = 9.1, 9.1 Hz,
1H), 3.24 (ddd, J = 11.1, 7.1, 4.1 Hz, 1H), 2.81 ꢀ 2.59 (m, 1H), 2.34 ꢀ
2.15 (m, 1H), 1.90 ꢀ 1.58 (m, 3H), 1.17 (s, 9H); 13C NMR (150 MHz,
CDCl3) δ 137.2, 136.7, 130.6, 130.42, 130.38, 129.2, 128.6, 128.4, 48.0,
44.5, 29.2, 28.3, 25.0, 18.3; 29Si NMR (119 MHz, CDCl3) δ ꢀ3.44;
HRMS (ESI) m/z [M ꢀ Br]+ calcd for C20H28NSi 310.1985, found
310.1991. Slow recrystallization from DCM and hexanes afforded single
crystals, and the absolute configuration was determined to be S by X-ray
analysis. The catalyst was derivatized to the N-tosylated structure for
chiral stationary-phase HPLC analysis: 99.7% ee, see (S)-15c below.
General Procedure for the Generation of Free Amine
Catalysts (S)-2aꢀc. To a solution of catalyst (S)-2 HBr (0.861
3
mmol) in DCM (10 mL) in a 25 mL round-bottom flask was added
Amberlyst-A21 basic resin (2.5 g). The mixture was allowed to stir for
1 h and then filtered to remove the Amberlyst. The resulting solution
was concentrated in vacuo, redissolved in a minimum of DCM (∼1 mL),
and then filtered through a silica gel plug using 90:10 DCM/MeOH to
remove any remaining Amberlyst residue. The solution was then
concentrated in vacuo to give (S)-2aꢀc as a lightly colored oil that
was used without further purification.
(2S,3R)-2-Butyl-4-nitro-3-phenylbutanal (5ca). Prepared from
0.0558 g of trans-β-nitrostyrene and hexanal according to general
procedure A to afford 0.0867 g of 5ca in 93% yield after 2 h. HPLC
analysis: Chiralcel OD-H column, n-hexane/i-PrOH = 95:5, 1.0 mL/min,
40 °C, tR (syn, major) = 14.6 min; tR (syn, minor) = 18.8 min; tR (anti,
major) = 15.8 min; tR (anti, minor) = 25.6 min. 96.8% ee. dr = 93:7
1
based on H NMR analysis of unpurified reaction. Spectral data are
consistent with literature values.69
Data for (S)-2a. Prepared from 0.300 g of (S)-2a HBr according to
3
the general procedure to afford 0.228 g of (S)-2a as a lightly colored oil in
(2S,3R)-4-Nitro-2-(1-phenylmethyl)-3-phenylbutanal (5da). Pre-
pared from 0.0670 g of trans-β-nitrostyrene and 0.0502 g hydrocinna-
maldehyde, modifying general procedure A by using one equiv of
aldehyde and 1.2 equiv of nitrostyrene with CHCl3 as solvent to afford
0.0943 g of 5da in 89% yield after 5 h. HPLC analysis: Chiralcel OD-H
column, n-hexane/i-PrOH = 95:5, 1.0 mL/min, 40 °C, tR (syn, major) =
41.5min; tR (syn, minor) = 45.8 min; tR (anti, major) = 59.1 min; tR (anti,
1
99% yield: [R]23 = +0.7 (c 0.924, CHCl3); H NMR (400 MHz,
D
CDCl3) δ = 7.70ꢀ7.51 (m, 4H), 7.48ꢀ7.23 (m, 6H), 2.97 (dddd, J =
11.6, 11.6, 5.8, 5.8 Hz, 1H), 2.85ꢀ2.67 (m, 2H), 2.60 (bs, 1H), 1.93
(ddd, J = 11.4, 7.4, 4.6 Hz, 1H), 1.69 (ddd, J = 13.2, 8.5, 5.0 Hz, 1H),
1.63ꢀ1.45 (m, 2H), 0.62 (s, 3H); 13C NMR (100 MHz, CDCl3) δ
135.4, 135.2, 135.0, 134.9, 129.43, 129.42, 127.90, 127.87, 48.7, 47.2,
28.6, 26.4, ꢀ6.1; 29Si NMR (119 MHz, CDCl3) δ ꢀ8.3; LRMS (ESI)
m/z [M + H]+ calcd for C17H22NSi 268.2, found 268.5.
1
minor) = 68.2 min. 92.6% ee. dr =94:6 based on H NMR analysis of
unpurified reaction. Spectral data are consistent with literature values.70
(2S,3R)-2-(Benzyloxy)-4-nitro-3-phenylbutanal (5ea). Prepared
from 0.0558 g of trans-β-nitrostyrene and 0.0842 g benzyloxyacetaldehyde,
modifying general procedure A by omitting Et3N, using 0.0116 g of the free
amine catalyst (S)-2c (10 mol %), and using 1.5 equiv of aldehyde to afford
0.0851 g of 5ea in 76% yield after 8 h. HPLC-analysis: Chiralpak AD-H
column, n-hexane/i-PrOH = 97:3, 0.5 mL/min, 40 °C, tR (syn, major) =
60.6 min; tR (syn, minor) = 65.1 min; tR (anti, major) = 87.6 min; tR (anti,
Data for (S)-2b. Prepared from 0.353 g of (S)-2b HBr according to
3
the general procedure to afford 0.278 g of (S)-2b as a white solid in 98%
yield: mp = 105ꢀ107 °C; [R]22D = +7.7 (c 0.662, CHCl3); 1H NMR
(300 MHz, CDCl3) δ 7.63ꢀ7.57 (m, 5H), 7.45ꢀ7.30 (m, 10H), 3.13
(dd, J = 10.1, 7.7 Hz, 1H), 2.97 (ddd, J = 11.7, 7.3, 4.7 Hz, 1H), 2.77
(ddd, J = 10.7, 7.4, 7.4 Hz, 1H), 2.07 (ddd, J = 15.4, 9.6, 5.8 Hz, 1H),
1.81ꢀ1.60 (m, 2H), 1.60ꢀ1.46 (m, 1H), NꢀH peak not observed; 13C
NMR (75 MHz, CDCl3) δ 136.2, 133.8, 129.8, 128.1, 49.2, 46.6, 29.2,
26.8; 29Si NMR (119 MHz, CDCl3) δ ꢀ11.9. LRMS (ESI) m/z [M +
H]+ calcd for C22H24NSi 329.2, found 329.4.
1
minor) = 73.0 min. 95.2% ee. dr =61:39 based on H NMR analysis of
unpurified reaction. Spectral data are consistent with literature values.49
(3R)-2,2-Dimethyl-4-nitro-3-phenylbutanal (5fa). Prepared from
0.0558 g of trans-β-nitrostyrene and isobutyraldehyde, modifying gen-
eral procedure A by using 20 mol % catalyst loading, to afford 0.0248 g of
5fa in 30% yield after 7 days. HPLC analysis: Chiralpak AS-H column,
n-hexane/i-PrOH = 95:5, 1.0 mL/min, 40 °C, tR (major) = 6.6 min;
tR (minor) = 7.2 min. 55.6% ee. Spectral data are consistent with
literature values.52
Data for (S)-2c. Prepared from 0.336 g of (S)-2c HBr according to
3
the general procedure to afford 0.264 g of (S)-2c as a lightly colored
semisolid in 99% isolated yield: [R]23D = +35.8 (c 0.884, CHCl3); 1H
NMR (600 MHz, CDCl3) δ 7.75ꢀ7.60 (m, 4H), 7.45ꢀ7.27 (m, 6H),
3.03 (dd, J = 10.6, 7.6 Hz, 1H), 2.88 (ddd, J = 12.2, 7.2, 5.1 Hz, 1H), 2.73
(ddd, J = 10.4, 7.5 Hz, 1H), 2.02 (dddd, J = 16.0, 12.1, 7.8, 4.6 Hz, 1H),
1.69 (dddd, J = 20.1, 15.5, 7.5, 4.5 Hz, 1H), 1.56 (dddd, J = 10.6, 10.6,
9.2, 9.2 Hz, 1H), 1.43 (dddd, J = 19.5, 16.7, 7.4, 4.1, 1H), 1.12 (s, 9H),
NꢀH peak not observed; 13C NMR (75 MHz, CDCl3) δ 136.6, 136.5,
133.6, 133.2, 129.40, 129.35, 127.8, 127.7, 48.8, 45.1, 29.4, 28.5, 26.6,
18.5; 29Si NMR (119 MHz, CDCl3) δ ꢀ3.5. HRMS (ESI) m/z [M +
H]+ calcd for C20H28NSi 310.1985, found 310.1981.
(2S,3R)-2-(1-Methylethyl)-4-nitro-3-phenylbutanal (5ga). Prepared
from 0.0558 g of trans-β-nitrostyrene and isovaleraldehyde, modifying
general procedure A by using 20 mol % of catalyst (0.0292 g) and 0.0076
g of Et3N (20 mol %) to afford 0.0525 g of 5ga in 81% yield after 4 days.
HPLC analysis: Chiralcel OD-H column, n-hexane/i-PrOH = 95:5,
1.0 mL/min, 40 °C, tR (syn, major) = 12.8 min; tR (syn, minor) = 13.8
min. 91.9% ee. Due to a slight overlap in the separation, the reported
enantiomeric excess of 90% ee was calculated based on an average of two
different HPLC methods as described in the Supporting Information.
dr = 83:17 based on 1H NMR analysis of the unpurified reaction.
Spectral data are consistent with literature values.68
General Procedure A for the Asymmetric Michael Addi-
tion of Aldehydes to Nitroolefins. Triethylamine (0.0187 mmol,
0.05 equiv) was added to a stirred solution of catalyst (S)-2c HBr
3
(0.0073 g, 0.0187 mmol, 0.05 equiv) and nitroolefin (0.374 mmol, 1.0
equiv) in 9:1 hexane/THF (0.4 mL) at 0 °Cand allowed to stir for 5 min.
7072
dx.doi.org/10.1021/jo200991q |J. Org. Chem. 2011, 76, 7065–7075

The Journal of Organic Chemistry
ARTICLE
(2S,3R)-3-(4-Chlorophenyl)-2-methyl-4-nitrobutanal (5ab). Pre-
pared from 0.0687 g of trans-4-chloro-β-nitrostyrene and propanal
according to general procedure A to afford 0.0795 g of 5ab in 88% yield
after 30 min. HPLC analysis: Chiralpak AD-H column, n-hexane/i-
PrOH = 95:5, 0.5 mL/min, 30 °C, tR (syn, major) = 33.5 min; tR (syn,
minor) =25.3 min;tR (anti, major) = 29.2 min; tR (anti, minor) = 31.1 min.
97.9% ee. dr = 88:12 based on 1H NMR analysis of unpurified reaction.
Spectral data are consistent with literature values.71
(2S,3R)-3-(2-Fluorophenyl)-2-methyl-4-nitrobutanal (5ac). Pre-
pared from 0.0625 g of trans-2-fluoro-β-nitrostyrene and propanal
according to general procedure A to afford 0.0791 g of 5ac in 94%
yield after 30 min. Enantiomeric excess was determined after cycliza-
tion and N-protection to structure 5ac0 using the procedure below.
97.1% ee. dr = 90:10 based on 1H NMR analysis of unpurified
reaction. Spectral data are consistent with literature values.19
3.74 mmol, 10 equiv) was added to a stirred solution of catalyst (S)-2c
(0.0232 g, 0.0748 mmol, 0.2 equiv) and nitroolefin (0.374 mmol, 1.0
equiv) in 9:1 hexane/THF (0.4 mL) at 0 °C and then allowed to warm to
room temperature. The reaction was allowed to stir for 3ꢀ48 h until
complete as judged by TLC (100% CHCl3) and then quenched by the
addition of 1.0 N HCl (5 mL). The organic layer was separated and the
aqueous layer extracted with 3 ꢁ 10 mL of EtOAc. The combined
organic layers were dried with Na2SO4, filtered, and concentrated to give
a colored oil. The crude product was purified on a flash silica gel column
(hexanes/DCM).
(3R)-4-Nitro-3-phenylbutanal (5ba). Prepared from 0.0558 g of
trans-β-nitrostyrene and acetaldehyde according to general procedure
B to afford 0.0542 g of 5ba in 75% yield after 3 h. Product 5ba was
reduced to the corresponding alcohol using NaBH4 for HPLC analysis:
Chiralcel OD-H column, n-hexane/i-PrOH = 95:5, 1.0 mL/min, 40 °C,
tR (major) = 28.9 min; tR (minor) = 35.6 min. 94.8% ee. Spectral data are
consistent with literature values.13
(3R,4S)-3-(2-Fluorophenyl)-1-(4-methoxybenzoyl)-4-methylpyrroli-
dine (5ac0). In a 50 mL flask was dissolved 0.070 g of nitroaldehyde 5ac in
MeOH (10 mL), and 20 wt % Pd(OH)2 on carbon (0.044 g, 0.062 mmol,
0.2 equiv) was added. The reaction flask was purged with hydrogen, a
hydrogen balloon was put in place, and the reaction was allowed to stir,
refilling the hydrogen balloon as needed, for 48 h. The reaction was
filtered over Celite to remove Pd(OH)2 then concentrated in vacuo to
give a lightly colored oil. The crude pyrrolidine was redissolved in DCM
(10 mL), then Et3N (0.11 mL, 0.777 mmol, 2.5 equiv) was added, and
finally 4-methoxybenzoyl chloride (0.063 mL, 0.466 mmol, 1.5 equiv) was
added and the mixture allowed to stir for 30 min. The crude product was
purified on a flash silica gel column (80:20 hexanes/EtOAc to 50:50
hexanes/EtOAc) to give 0.0818 g (84%) of 5ac0 as a clear oil. HPLC-
analysis: Chiralpak AD-H column, n-hexane/i-PrOH = 90:10, 0.7 mL/min,
33 °C, tR (syn, major) = 22.4 min; tR (syn, minor) = 26.9 min; tR (anti,
major) = 24.3 min; tR (anti, minor) = 37.7 min. 97.1% ee. dr = 90:10
(3R)-3-(4-Chlorophenyl)-4-nitrobutanal (5bb). Prepared from
0.0687 g of trans-4-chloro-β-nitrostyrene and acetaldehyde according to
general procedure B to afford 0.0656 g of 5bb in 77% yield after 3 h.
Product 5bb was reduced to the corresponding alcohol using NaBH4 for
HPLC analysis: Chiralcel OD-H column, gradient: n-hexane/i-PrOH =
98:2, 0.7 mL/min, 30 °C for 60 min, then n-hexane/i-PrOH = 95:5,
1.0 mL/min, 30 °C for 30 min, tR (major) = 61.6 min; tR (minor) = 63.2
min. 92.1% ee. Spectral data are consistent with literature values.13
(3R)-3-(2-Fluorophenyl)-4-nitrobutanal (5bc). Prepared from 0.0625 g
of trans-2-fluoro-β-nitrostyrene and acetaldehyde according to general
procedure B to afford 0.0600 g of 5bc in 76% yield after 3 h. Product
5bc was reduced to the corresponding alcohol using NaBH4 for HPLC-
analysis: Chiralcel OD-H column, n-hexane/i-PrOH = 96:4, 0.8 mL/min,
40 °C, tR (major) = 35.4 min; tR (minor) = 39.0 min. 96.4% ee: 1H NMR
(300 MHz, CDCl3) δ 9.72 (s, 1H), 7.35ꢀ7.20 (m, 2H), 7.17ꢀ7.02 (m,
2H), 4.72 (dd, J = 7.3, 1.1 Hz, 2H), 4.27 (p, J = 7.2 Hz, 1H), 3.03 (d, J =
7.1 Hz, 2H); 13C NMR (75 MHz, CDCl3) δ 198.8, 160.9 (d, J1CF = 246.2
1
1
based on H NMR of unpurified reaction for structure 5ac: H NMR
(600 MHz, CDCl3, reported as a 1:1 mixture of rotomers) δ 7.57ꢀ7.52
(m, 2H, 1:1 mixture of rotomers), 7.29ꢀ7.22 (m, 2H), 7.12 (q, J =
7.8 Hz, 1H), 7.07ꢀ7.00 (m, 1H, 1:1 mixture of rotomers), 6.93ꢀ6.87
(m, 2H), 4.08ꢀ4.01 (m, 1H), 3.86ꢀ3.74 (m, 4H, 1:1 mixture of
rotomers), 3.60 (t, J = 10.2 Hz, 1H), 3.39ꢀ3.14, (m, 2H, 1:1 mixture
of rotomers), 2.60ꢀ2.32 (m, 1H, 1:1 mixture of rotomers), 1.06ꢀ1.95
(m, 3H, 1:1 mixture of rotomers); 13C NMR (150 MHz, CDCl3, reported
as a 1:1 mixture of rotomers) δ 169.4, 161.1, 161.6 (J1CF = 255 Hz), 132.1,
129.4, 128.8 (J2CF = 15.0 Hz), 128.7(J3CF =8.6 Hz), 128.5(J3CF =8.0 Hz)
128.3 (J4CF = 4.4 Hz), 128.1 (J4CF = 4.2 Hz), 128.0, (J2CF = 23.7 Hz),
127.8 (J2CF = 24.0 Hz), 126.7 (J3CF = 14.0 Hz), 125.9 (J3CF = 14.0 Hz),
124.5 (J4CF = 3.3 Hz), 115.8 (J3CF = 10.4 Hz), 115.7 (J3CF = 10.4 Hz),
113.6, 56.0, 55.7, 55.4, 52.3, 45.7, 43.5, 40.4, 37.7, 15.8, 15.4; LRMS (ESI)
m/z [M + H]+ calcd for C19H21FNO2 314.2, found 314.2.
Hz), 130.04 (d, J4CF = 1.7 Hz), 129.95 (d, J3CF = 2.2 Hz), 125.1 (d, J2
=
=
13.2 Hz), 124.9 (d, J3CF = 3.5 Hz), 116.3 (d, J2CF = 21.8 Hz), 77.8 (d, J4CF
CF
CF
2.7 Hz), 45.2 (d, J3 = 2.0 Hz), 33.5 (s, 1H); 19F NMR (282 MHz,
CDCl3) δ ꢀ117.1 (m).
(3R)-3-(4-Methoxyphenyl)-4-nitrobutanal (5bd). Prepared from
0.0670 g of trans-4-methoxy-β-nitrostyrene and acetaldehyde according
to general procedure B to afford 0.0518 g of 5bd in 62% yield after 48 h.
Product 5bd was reduced to the corresponding alcohol using NaBH4 for
HPLC analysis: Chiralcel OD-H column, n-hexane/i-PrOH = 95:5,
1.0 mL/min, 40 °C, tR (major) = 39.2 min; tR (minor) = 43.1 min.
91.9% ee. Spectral data are consistent with literature values.13
(3S)-3-(2-Furyl)-4-nitrobutanal (5be). Prepared from 0.520 g of (E)-
2-(2-nitroethenyl)furan and acetaldehyde according to general proce-
dure B to afford 0.0432 g of 5be in 63% yield after 48 h. Product 5be was
reduced to the corresponding alcohol using NaBH4 for HPLC analysis:
Chiralcel OD-H column, n-hexane/i-PrOH = 96:4, 0.8 mL/min, 40 °C,
tR (major) = 31.8 min; tR (minor) = 34.0 min. 90.0% ee. Spectral data are
consistent with literature values.13
(2S,3R)-3-(4-Methoxyphenyl)-2-methyl-4-nitrobutanal (5ad). Pre-
pared from 0.0670 g of trans-4-methoxy-β-nitrostyrene and propanal
according to general procedure A to afford 0.0772 g of 5ad in 87% yield
after 30 min. Due to the difficulty in separating the stereoisomers of this
substrate, the reported enantiomeric excess is an average of two different
methods of calculation as described in the Supporting Information.
Average ee = 96.1%. dr =90:10 based on 1H NMR analysis of unpurified
reaction. Spectral data are consistent with literature values.72
(2S,3S)-3-(Furan-2-yl)-2-methyl-4-nitrobutanal (5ae). Prepared
from 0.0520 g of (E)-2-(2-nitroethenyl)furan and propanal according
to general procedure A to afford 0.0568 g of 5ae in 77% yield after 30
min. HPLC analysis: Chiralpak AS-H column, n-hexane/i-PrOH = 98:2,
0.8 mL/min, 30 °C, tR (syn, major) = 18.7 min; tR (syn, minor) = 20.8
min; tR (anti, major) = 22.1 min; tR (anti, minor) = 24.9 min. 96.5% ee.
dr = 87:13 based on 1H NMR of unpurified reaction. Spectral data are
consistent with literature values.12
General Procedure C for the N-Protection of 2-Silylpyrro-
lidine Catalysts to (S)-15aꢀc for HPLC Analysis. To a mixture of
(S)-2 HBr (0.049 mmol, 1.05 equiv) in DCM (2.5 mL) were added
3
triethylamine (0.016 mL, 0.116 mmol, 2.5 equiv) and then the acid
chloride of the corresponding protecting group (0.046 mmol, 1.0 equiv).
The reaction wasallowedto stir at room temperature for1 h and then was
directly purified using preparatory TLC to prepare an HPLC standard.
Data for (S)-15a. Prepared according to general procedure C using
(S)-2a HBr with benzoyl chloride for HPLC analysis. Purified by
3
preparative TLC (80:20 hexanes/EtOAc). HPLC analysis: Chiralpak
General Procedure B for the Asymmetric Michael Addi-
tion of Acetaldehyde to Nitroolefins. Acetaldehyde (0.16 mL
AD-H column, n-hexane/i-PrOH = 95:5, 1.0 mL/min, 40 °C, tR (major)
1
= 14.6 min; tR (minor) = 10.8 min. 99.5% ee. H NMR (300 MHz,
7073
dx.doi.org/10.1021/jo200991q |J. Org. Chem. 2011, 76, 7065–7075

The Journal of Organic Chemistry
ARTICLE
CDCl3) δ 7.76ꢀ7.69 (m, 2H), 7.66ꢀ7.59 (m, 2H), 7.46ꢀ7.16 (m,
11H), 4.30 (dd, J = 8.7, 8.7 Hz, 1H), 3.34 (ddd, J = 10.5, 6.8, 3.9 Hz, 1H),
3.02 (ddd, J = 9.6, 6.9, 6.9 Hz, 1H), 2.21ꢀ2.04 (m, 1H), 2.00ꢀ1.80 (m,
1H), 1.78ꢀ1.55 (m, 2H), 0.78 (s, 3H); 13C NMR (75 MHz, CDCl3) δ
169.5, 137.3, 136.5, 135.3, 135.0, 129.8, 129.5, 129.3, 128.2, 127.9, 127.2,
50.8, 47.1, 28.6, 26.9, ꢀ2.9; LRMS (ESI) m/z [M + H]+ calcd for
C24H26NOSi 372.2, found 372.2.
2 mL of methanol and directly injected for ESIꢀMS analysis in
positive mode.
’ ASSOCIATED CONTENT
S
Supporting Information. Copies of NMR spectra,
b
HPLC chromatograms, computational details, and X-ray crystal
Data for (S)-15b. Prepared according to general procedure C using
structures for (S)-2a HBr, (S)-2b HBr, (S)-2c HBr, and (R)-
3
3
3
(S)-2b HBr with 4-methoxybenzoyl chloride for HPLC analysis. Pur-
3
7 HBr (as the protonated tetrabromozincate (II) salt mono-
3
ified by preparative TLC (90:10 DCM/EtOAc). HPLC analysis: Chir-
hydrate). This material is available free of charge via the Internet
at http://pubs.acs.org.
alpak AD-H column, n-hexane/i-PrOH = 95:5, 1.0 mL/min, 40 °C, tR
1
(major) = 41.2 min; tR (minor) = 23.4 min. 99.6% ee. H NMR (300
MHz, CDCl3) δ 7.76ꢀ7.65 (m, 5H), 7.49ꢀ7.26 (m, 10H), 7.05 (d, J =
8.5 Hz, 2H), 6.74 (d, J = 8.7 Hz, 2H), 4.68 (dd, J = 8.7, 8.7 Hz, 1H), 3.76
(s, 3H), 3.50ꢀ3.35 (m, 1H), 3.09 (ddd, J = 18.2, 8.4, 8.4 Hz, 1H), 2.20
(dddd, J = 8.6, 5.9, 5.7, 5.7 Hz, 1H), 2.06ꢀ1.89 (m, 1H), 1.77ꢀ1.61 (m,
2H); 13C NMR (75 MHz, CDCl3) δ 169.2, 160.7, 136.4, 134.5, 129.4,
129.3, 129.2, 127.8, 113.2, 55.3, 51.0, 45.9, 29.5, 27.0; HRMS (ESI) m/z
[M + H]+ calcd for C30H30NO2Si 464.2046, found 464.2041.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: akfranz@ucdavis.edu.
’ ACKNOWLEDGMENT
Data for (S)-15c. Prepared according to general procedure C using
This research is supported by funds from the University of
California, Davis, NSF (CHE-0847358), and the Pittsburgh
Supercomputer Center. A.K.F. acknowledges 3M Corp. for a
Nontenured Faculty Award. K.I.J. and T.M. are recipients of the
UC Davis Bradford Borge Chemistry Fellowship. Special thanks
to Dr. James Fettinger for solving X-ray structures and Prof. Dean
Tantillo and Dr. Mike Lodewyk for helpful discussion and
assistance with computational studies.
(S)-2c HBr with p-toluenesulfonyl chloride for HPLC analysis. Purified
3
by preparative TLC (60:40 DCM/hexane). HPLC analysis: Chiralpak
AD-H column, n-heptane/i-PrOH = 99:1, 0.70 mL/min, 25 °C, tR
(major) = 51.0 min; tR (minor) = 36.7 min. 99.7% ee. 1H NMR (300
MHz, CDCl3) δ 7.81 (dd, J = 7.7, 1.6 Hz, 2H), 7.77 (d, J = 8.3 Hz, 2H),
7.57 (dd, J = 7.7, 1.6 Hz, 2H), 7.46ꢀ7.27 (m, 8H), 4.47 (dd, J = 10.0, 3.3
Hz, 1H), 3.34 (ddd, J = 12.8, 8.7, 3.9 Hz, 1H), 2.52ꢀ2.35 (m, 1H), 2.42
(s, 3H), 1.70ꢀ1.57 (m, 1H), 1.56ꢀ1.43 (m, 1H), 1.24 (s, 9H),
0.96ꢀ0.67 (m, 2H); 13C NMR (75 MHz, CDCl3) δ 143.6, 137.0,
136.8, 136.0, 133.9, 133.2, 129.9, 129.8, 129.4, 127.9, 127.8, 127.6, 49.2,
48.0, 28.1, 27.3, 23.6, 21.7, 19.1; HRMS (ESI) m/z [M + H]+ calcd for
C27H34NO2SSi 464.2074, found 464.2071.
’ REFERENCES
(1) MacMillan, D. W. C. Nature 2008, 455, 304.
(2) Rueping, M.; Merino, E.; Sugiono, E. Adv. Synth. Catal. 2008,
350, 2127.
(3) Enders, D.; H€uttl, M. R. M.; Grondal, C.; Raabe, G. Nature 2006,
441, 861.
(4) Berkessel, A.; Gr€oger, H. Asymmetric Organocatalysis - From
Biomimetic Concepts to Applications in Asymmetric Synthesis; Wiley-
VCH Verlag GmbH & Co. KGaA: Weinheim, 2005.
(5) Xu, L.-W.; Li, L.; Shi, Z.-H. Adv. Synth. Catal. 2010, 352, 243.
(6) Brook, M. A. Silicon in Organic, Organometallic, and Polymer
Chemistry; John Wiley & Sons: Hamilton, Ontario, 2000.
(7) Lipshutz, B. H.; Hungate, R. W. J. Org. Chem. 1981, 46, 1410.
(8) Block, E.; Aslam, M. J. Am. Chem. Soc. 1985, 107, 6729.
(9) Terao, J.; Kambe, N. Chem. Rec. 2007, 7, 57.
(10) Bartoszewicz, A.; Kalek, M.; Nilsson, J.; Hiresova, R.; Stawinski,
J. Synlett 2008, 2008, 37.
(11) Eaborn, C. J. Chem. Soc. 1952, 2840.
(12) Hayashi, Y.; Gotoh, H.; Hayashi, T.; Shoji, M. Angew. Chem.,
Int. Ed. 2005, 44, 4212.
(13) Gotoh, H.; Ishikawa, H.; Hayashi, Y. Org. Lett. 2007, 9, 5307.
(14) List, B. Synlett 2001, 2001, 1675.
(15) Shi, Z.-H.; Sheng, H.; Yang, K.-F.; Jiang, J.-X.; Lai, G.-Q.; Lu, Y.;
Xu, L.-W. Eur. J. Org. Chem. 2011, 66.
(16) Zheng, Z.; Perkins, B. L.; Ni, B. J. Am. Chem. Soc. 2009, 132, 50.
(17) Patil, M. P.; Sharma, A. K.; Sunoj, R. B. J. Org. Chem. 2010, 75, 7310.
(18) Min, T.; Jentzsch, K.; Franz, A. K.; Etcheson, J. Abstracts of
Papers, 239th ACS National Meeting, San Francisco, CA, Mar 21ꢀ25,
2010; American Chemical Society: Washington, DC, 2010; ORGN-201.
(19) Husmann, R.; J€orres, M.; Raabe, G.; Bolm, C. Chem.—Eur. J.
2010, 16, 12549.
(20) Bauer, J. O.; Stiller, J.; Marquꢀes-Lꢀopez, E.; Strohfeldt, K.;
Christmann, M.; Strohmann, C. Chem.—Eur. J. 2010, 16, 12553.
(21) An example of a catalytic asymmetric hydrosilylation has been
reported where higher enantioselectivity is observed using a silyl fluoride
substrate compared to a silyl chloride substrate; see: Hatanaka, Y.; Goda,
K.-I.; Yamashita, F.; Hiyama, T. Tetrahedron Lett. 1994, 35, 7981.
ESIꢀMass Spectrometry Detection of Intermediates and
Hydrogen-Bond Adducts. Reaction mixtures and samples were
prepared and analyzed using ESI-MS in the positive ion mode on a Qtrap
spectrometer. Source parameters were 5 kV spray voltage, with a curtain
plate temperature of 275 °C and sheath gas setting of 15. Samples were
analyzed via flow injection analysis by injecting 20 μL samples into a stream
of 80:20 MeOH/aqueous 0.1% formic acid, flowing at 300 μL/min. All raw
ESIꢀMS data files were analyzed using Applied Biosystems Analyst
Software v.1.4.2, build 1228. Spectra were processed for publication figures
using Systat Software Inc. SigmaPlot software v.11.0, build 11.0.0.77.
Sample Preparation for ESIꢀMS Detection of Reaction
Intermediates in Figure 1. Spectrum A. Catalyst (S)-2c (0.005 g,
0.016 mmol, 0.05 equiv) was combined with propionaldehyde (0.048 g,
0.323 mmol, 1.0 equiv) in 0.3 mL of 9:1 hexane/THF and the mixture
allowed to stir for 10 min. A 10 μL aliquot of this mixture was diluted
into 2 mL of methanol and directly injected for ESIꢀMS analysis in
positive mode.
Spectrum B. Catalyst (S)-2c (0.005 g, 0.016 mmol, 0.05 equiv) was
combined with trans-β-nitrostyrene (0.048 g, 0.323 mmol, 1.0 equiv) in
0.3 mL of 9:1 hexane/THF and allowed to stir for 1 h. A 10 μL aliquot of
this mixture was diluted into 2 mL of methanol and directly injected for
ESIꢀMS analysis in positive mode.
Spectrum C. After analysis of the reaction mixture described above to
generate graph B, propionaldehyde (0.24 mL, 3.230 mmol, 10.0 equiv)
was added. Immediately after addition of the aldehyde, a 10 μL aliquot of
the reaction mixture was diluted into 2 mL of methanol and directly
injected for ESIꢀMS analysis in positive mode.
Sample Preparation for ESIꢀMS Detection of Nitrostyr-
ene Binding to Amines (Figure 2). A mixture of trans-β-
nitrostyrene (0.010 g, 0.070 mmol, 1.0 equiv) and hydrogen bond
donor (0.070 mmol, 1.0 equiv) in 0.1 mL of methanol was allowed
to stir for 10 min. A 10 μL aliquot of this mixture was diluted into
7074
dx.doi.org/10.1021/jo200991q |J. Org. Chem. 2011, 76, 7065–7075

The Journal of Organic Chemistry
ARTICLE
(22) Sieburth, S. M.; Chen, C.-A. Eur. J. Org. Chem. 2006, 311.
(23) Tacke, R.; Schmid, T.; Penka, M.; Burschka, C.; Bains, W.;
Warneck, J. Organometallics 2004, 23, 4915.
(24) Mortensen, M.; Husmann, R.; Veri, E.; Bolm, C. Chem. Soc. Rev.
2009, 38, 1002.
(54) Wang, J.; Li, H.; Lou, B.; Zu, L.; Guo, H.; Wang, W. Chem.—
Eur. J. 2006, 12, 4321.
(55) Patil, M. P.; Sunoj, R. B. Chem.—Eur. J. 2008, 14, 10472.
(56) García-García, P.; Ladꢀep^eche, A.; Halder, R.; List, B. Angew.
Chem., Int. Ed. 2008, 47, 4719.
(25) For this study, tert-butyldiphenylfluorosilane and triisopropyl-
fluorosilane were synthesized on gram-scale from the corresponding
commercially available silyl chlorides according to the method described
by Damrauer et al.; see: Damrauer, R.; Simon, R. A.; Kanner, B.
Organometallics 1988, 7, 1161.
(57) Hayashi, Y.; Itoh, T.; Ohkubo, M.; Ishikawa, H. Angew. Chem.,
Int. Ed. 2008, 47, 4722.
(58) Mukherjee, S.; Yang, J. W.; Hoffmann, S.; List, B. Chem. Rev.
2007, 107, 5471.
(59) Patora-Komisarska, K.; Benohoud, M.; Ishikawa, H.; Seebach,
D.; Hayashi, Y. Helv. Chim. Acta 2011, 94, 719.
(60) Wu, Z.-J.; Luo, S.-W.; Xie, J.-W.; Xu, X.-Y.; Fang, D.-M.; Zhang,
G.-L. J. Am. Soc. Mass. Spectrom. 2007, 18, 2074.
(61) Marquez, C.; Metzger, J. O. Chem. Commun. 2006, 1539.
(62) Li, B.-L.; Wang, Y.-F.; Luo, S.-P.; Zhong, A.-G.; Li, Z.-B.; Du,
X.-H.; Xu, D.-Q. Eur. J. Org. Chem. 2010, 656.
(63) Okino, T.; Hoashi, Y.; Furukawa, T.; Xu, X.; Takemoto, Y.
J. Am. Chem. Soc. 2005, 127, 119.
(26) Beak, P.; Basu, A.; Gallagher, D. J.; Park, Y. S.; Thayumanavan,
S. Acc. Chem. Res. 1996, 29, 552.
(27) Beak, P.; Kerrick, S. T.; Wu, S.; Chu, J. J. Am. Chem. Soc. 1994,
116, 3231.
(28) Kerrick, S. T.; Beak, P. J. Am. Chem. Soc. 1991, 113, 9708.
(29) Wiberg, K. B.; Bailey, W. F. J. Am. Chem. Soc. 2001, 123, 8231.
(30) Weisenburger, G. A.; Faibish, N. C.; Pippel, D. J.; Beak, P. J. Am.
Chem. Soc. 1999, 121, 9522.
(31) Basu, A.; Thayumanavan, S. Angew. Chem., Int. Ed. 2002,
41, 716.
(64) Zhang, Z.-H.; Dong, X.-Q.; Chen, D.; Wang, C.-J. Chem.—Eur.
J. 2008, 14, 8780.
(32) Tenenbaum, J. M.; Woerpel, K. A. Org. Lett. 2003, 5, 4325.
(33) Jones, P. R.; Cheng, A. H. B.; Albanesi, T. E. Organometallics
1984, 3, 78.
(34) Hatanaka, Y.; Hiyama, T. J. Org. Chem. 1989, 54, 268.
(35) In a recent example of asymmetric deprotonation of N-Boc
piperidine, O’Brien and coworkers overcome a slow rate of silylation and
increase enantioselectivity by utilizing a tin-lithium exchange process
and subsequent trapping with a silyl chloride; see: Stead, D.; Carbone,
G.; O’Brien, P.; Campos, K. R.; Coldham, I.; Sanderson, A. J. Am. Chem.
Soc. 2010, 132, 7260.
(65) Miyashita, M.; Yanami, T.; Kumazawa, T.; Yoshikoshi, A. J. Am.
Chem. Soc. 1984, 106, 2149.
(66) Boarino, L.; Geobaldo, F.; Borini, S.; Rossi, A. M.; Rivolo, P.;
Rocchia, M.; Garrone, E.; Amato, G. Phys. Rev. B 2001, 64, 205308.
(67) Yadav, J. S.; Reddy, A. R.; Rao, Y. G.; Narsaiah, A. V.; Reddy,
B. V. S. Synthesis 2007, 2007, 3447.
(68) Betancort, J. M.; Barbas, C. F. Org. Lett. 2001, 3, 3737.
(69) Wang, W.; Wang, J.; Li, H. Angew. Chem., Int. Ed. 2005,
44, 1369.
(70) Kotrusz, P.; Toma, S.; Schmalz, H.-G.; Adler, A. Eur. J. Org.
(36) Kaul, R.; Brouillette, Y.; Sajjadi, Z.; Hansford, K. A.; Lubell,
Chem. 2004, 1577.
W. D. J. Org. Chem. 2004, 69, 6131.
(37) Srinivasan, N.; Yurek-George, A.; Ganesan, A. Mol. Diversity
2005, 9, 291.
(71) Luo, R.-S.; Weng, J.; Ai, H.-B.; Lu, G.; Chan, A. S. C. Adv. Synth.
Catal. 2009, 351, 2449.
(72) Andrey, O.; Alexakis, A.; Tomassini, A.; Bernardinelli, G. Adv.
Synth. Catal. 2004, 346, 1147.
(38) The structures of the silylated pyrrolidines (S)-2aꢀc HBr
3
were confirmed using HRMS and X-ray crystallography. An X-ray
structure of the silyl ether catalyst (R)-7 was also obtained (as the
tetrabromozincate) for steric volume comparison. CCDC nos.
822232ꢀ822235. See the Supporting Information for more details.
(39) Dearden, M. J.; Firkin, C. R.; Hermet, J.-P. R.; O’Brien, P. J. Am.
Chem. Soc. 2002, 124, 11870.
(40) Stead, D.; O’Brien, P.; Sanderson, A. Org. Lett. 2008, 10, 1409.
(41) Cartledge, F. K. Organometallics 1983, 2, 425.
(42) Shimizu, N.;Takesue,N.; Yamamoto,A.; Tsutsumi, T.;Yasuhara,
S.; Tsuno, Y. Chem. Lett. 1992, 1263.
(43) Golubev, O.; Panov, D.; Ploom, A.; Tuulmets, A.; Nguyen, B. T.
J. Organomet. Chem. 2007, 692, 3700.
(44) Ploom, A.; Tuulmets, A.; J€arv, J. Phosphorus, Sulfur, Silicon Relat.
Elem. 2010, 185, 2503.
(45) See the Supporting Information for details of volume
calculation.
(46) Mayr, H.; Pock, R. Tetrahedron 1986, 42, 4211.
(47) Lambert, J. B.; Zhao, Y.; Emblidge, R. W.; Salvador, L. A.; Liu,
X.; So, J.-H.; Chelius, E. C. Acc. Chem. Res. 1998, 32, 183.
(48) Brook, M. A.; Neuy, A. J. Org. Chem. 1990, 55, 3609.
ꢁ
(49) Hut'ka, M.; Polꢀaꢁckovꢀa, V.; Marꢀak, J.; Kaniansky, D.; Sebesta, R.;
ꢁ
Toma, S. Eur. J. Org. Chem. 2010, 6430.
(50) Andrey, O.; Alexakis, A.; Bernardinelli, G. Org. Lett. 2003,
5, 2559.
(51) Austin, J. F.; MacMillan, D. W. C. J. Am. Chem. Soc. 2002,
124, 1172.
(52) Mase, N.; Thayumanavan, R.; Tanaka, F.; Barbas, C. F. Org.
Lett. 2004, 6, 2527.
(53) Transition-state calculations performed using Gaussian with
the B3LYP/6-31G(d) level of theory indicates that the syn-addition is
preferred over anti-addition for all three silyl pyrrolidine catalysts in the
gas phase; see the Supporting Information and: Zhao, J.-Q.; Gan, L.-H.
Eur. J. Org. Chem. 2009, 2009, 2661.
7075
dx.doi.org/10.1021/jo200991q |J. Org. Chem. 2011, 76, 7065–7075