Enantioselective trapping of an α-chiral carbanion of acyclic nitrile by a carbon electrophile

Article abstract of DOI:10.1039/c2cc00082b

An α-chiral nitrile carbanion generated by deprotonation of enantioenriched O-carbamoyl cyanohydrin was trapped in situ with ethyl cyanoformate to give the corresponding ester derivative in 92% yield and 90:10 er, providing the first example of trapping of an α-chiral acyclic nitrile carbanion that has been considered to be very configurationally labile. The Royal Society of Chemistry 2012.

Full text of DOI:10.1039/c2cc00082b

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 2897–2899
www.rsc.org/chemcomm
COMMUNICATION
Enantioselective trapping of an a-chiral carbanion of acyclic nitrile by a
carbon electrophilew
Michiko Sasaki,a Tomo Takegawa,a Hidaka Ikemoto,a Masatoshi Kawahata,b
Kentaro Yamaguchib and Kei Takeda*a
Received 5th January 2012, Accepted 27th January 2012
DOI: 10.1039/c2cc00082b
An a-chiral nitrile carbanion generated by deprotonation of
enantioenriched O-carbamoyl cyanohydrin was trapped in situ
with ethyl cyanoformate to give the corresponding ester derivative
in 92% yield and 90 : 10 er, providing the first example of trapping
of an a-chiral acyclic nitrile carbanion that has been considered to
be very configurationally labile.
which is explained in terms of enhanced angle strain in the
inversion transition state.9 It was also found in those studies
that the corresponding chiral aliphatic nitrile derivative
suffered extensive racemization under the same conditions.
We report the first enantioselective in situ trapping of a chiral
carbanion generated by deprotonation of an enantioenriched
acyclic a-chiral nitrile at a practically promising level.
Successful embodiment of the formidable task would rely
heavily on how to retain the configurational integrity of the
extremely labile a-nitrile carbanions during the trapping process.
The key elements in our approach are the use of N,N-dialkyl-
carbamoyloxy groups as a fixing agent for the lithium to the
carbon atom in chiral carbanions generated by deprotonation of
an enantioenriched precursor and the use of a highly reactive
electrophile. Although N,N-dialkylcarbamoyloxy groups have
become a powerful tool for generation of a chiral carbanion in
combination with (ꢀ)-sparteine10 since being introduced by
Hoppe as dipole-stabilizing and directing groups in electrophilic
substitution of allyllithium derivatives,11 they have also been
used as a fixing agent for a chiral carbanion in the absence of an
external chiral ligand.12 Thus, Hoppe and coworkers reported
the generation of chiral a-lithio derivatives of 2-alkenyl, 2-alkynyl,
and benzyl carbamates,13 which are much more configurationally
stable in comparison with the corresponding a-nitrile carbanions.
First, we chose enantioenriched O-carbamoyl cyanohydrin
derivative 1a, readily obtained from the corresponding O-acetyl
derivative, as a substrate on the basis of our previous observation,7
and we searched for a carbon electrophile that allows a rapid
in situ trapping of a carbanion generated by deprotonation of 1a.
When LDA (1.1 equiv.) was added to a solution of 1a and BnBr
(5.0 equiv.) in Et2O and the reaction was allowed to proceed at
the same temperature for 5 min, benzylated product 3 was
obtained in 45% yield and 52 : 48 enantiomeric ratio (er), along
with the starting material in 42% yield and 62 : 38 er (Table 1,
entry 1). The low enantiomeric purities of the benzylated product
and of the starting material recovered indicate that the reactivity
of BnBr is not high enough to trap the carbanion before
racemization. Therefore we turned our attention to ethyl
chloroformate and ethyl cyanoformate14 as more reactive
electrophiles, which were previously shown to be highly
reactive towards a-nitrile anions.15 The reaction with ethyl
chloroformate was completed within 5 min to give ester
derivative 4a in 85% yield, but the enantiomeric ratio
Although a-nitrile carbanions have been extensively used for
carbon–carbon bond formation because of their powerful nucleo-
philic character due to the small steric demand,1 extension to an
enantioselective version using enantiopure chiral carbanions next
to a nitrile group has been considered to be challenging because the
chirality is immediately lost by the formation of an sp2-hybridized
keteniminate. Carlier reported that the calculated inversion barrier
for lithioacetonitrile was extremely low (0.45 kcal molꢀ1),2 and
X-ray analysis of lithiophenylacetonitrile indicated the geometry to
be planar.3 For carbonyl counterparts, Kawabata et al.,4 Carlier
et al.,5 and others have proposed the concept of ‘‘memory of
chirality’’ or ‘‘self-regeneration of stereocenters via stereolabile
axially chiral intermediates’’6 that allows for the enantioselective
introduction of an electrophile at an enolizable chiral center next to
carbonyl functions. However, since the method is based on transfer
of the central chirality at a chiral center to a transient axial chirality
in the enolate intermediate, it would not be applicable to a-nitrile
carbanions that probably cannot generate axially chiral inter-
mediates. Although we have also reported that an a-nitrile
carbanion formally generated via an SE20 process in allylsilicates can
be trapped by benzyl bromide7 and protic solvents8 in modest and
good enantioselectivities, respectively, the possibility of intervention
of a concerted process not involving a discrete a-nitrile carbanion
cannot be ruled out and the methods lack substrate generality.
The only exception is a chiral cyclopropyl nitrile carbanion in
which enantiospecific deuteration can be achieved in basic CH3OD,
a Department of Synthetic Organic Chemistry, Graduate School of
Biomedical Sciences, Hiroshima University, 1-2-3 Kasumi, Minami-Ku,
Hiroshima, 734-8553, Japan. E-mail: takedak@hiroshima-u.ac.jp;
Fax: +81 82-257-5184; Tel: +81 82-257-5184
b Tokushima Bunri University, 1314-1 Shido, Sanuki, Kagawa,
769-2193, Japan
w Electronic supplementary information (ESI) available: Experimental
procedures, characterization, copies of 1H and 13C NMR spectra and
chiral HPLC chromatogram for all new compounds. CCDC 860190
and 860191. For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/c2cc00082b
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 2897–2899 2897

View Article Online
Table 1 Reaction of 1a with LDA in the presence of an electrophile
Table 3 Effects of solvent and temperature
4a
1a
Yield (%) er
74 : 26 57 100 : 0
Entry Solvent T/1C Base
Yield (%) er
3a, 4a
1a
1
2
3
4
5
6
7
8
Et2O
ꢀ80 LDA
39
26
19
32
46
Entry
Electrophile
Yield (%)
er
Yield (%)
er
Toluene ꢀ80 LDA
58 : 42 69
68 : 32 77
66 : 34 66
75 : 25 44
75 : 25 64
100 : 0
100 : 0
100 : 0
100 : 0
100 : 0
100 : 0
100 : 0
1
2
3
BnBr
ClCO2Et
CNCO2Et
45
85
39
52 : 48
52 : 48
74 : 26
42
—
57
62 : 38
—
100 : 0
THF
Et2O
Et2O
Et2O
Et2O
Et2O
ꢀ80 LDA
ꢀ50 LDA
ꢀ98 LDA
ꢀ98 LTMP 28
ꢀ98 LiNEt2 10
—
—
85
93
remained low (52 : 48 er) (entry 2), suggesting the reactivity of
chloroformate to be still too low. On the other hand, the use of
ethyl cyanoformate greatly enhanced the enantiomeric ratio to
74 : 26, although the chemical yield sharply decreased to 39%
and the starting material was recovered in 57% yield without
racemization (entry 3). The results could be understood by
assuming that LDA serves not only as a base but also as a
nucleophile for cyanoformate because of its higher electro-
philicity to give some byproducts and then a substantial
amount of the starting material remained intact.
ꢀ98 t-BuLi
—
racemization of 2 and the reaction of LDA with cyanoformate at
lower temperatures. We next examined the effect of bulkiness and
basicity of a base, which possibly affect the rates of deprotonation
and of the reaction with cyanoformate. With less and more
hindered bases, LiNEt2 and LTMP, respectively, a decrease in
chemical yield was observed in both cases (entries 6 and 7). The use
of t-BuLi resulted in complete recovery of the starting material,
probably due to the steric hindrance in deprotonation (entry 8).
Since the relatively low chemical yield in the reaction with
cyanoformate is probably attributable to the competitive consump-
tion of LDA by cyanoformate that was used in large excess to
accelerate the reaction with carbanion 2, we decided to use equal
amounts of LDA and cyanoformate. The use of five equivalents
resulted in a remarkable improvement in chemical yield (Table 4,
entry 1). A decrease to three equivalents afforded almost the
same result (entry 2), although a further decrease (1.1 equiv.) of
the reagents led to a lowering of chemical yield (entry 3).
This analysis led us to use a reduced amount of cyanoformate,
which would give a better chemical yield by suppressing its
reaction with LDA. On changing its amount from five equivalents
to one equivalent, a dramatic improvement of chemical yield from
39% to 76% was achieved as predicted, but the enantiomeric
ratio was considerably lowered from 74 : 26 er to 63 : 37 er
(Table 2, entries 1–5) probably because of the decreased
concentration of cyanoformate. These results reveal that the
rates of reaction of lithiocarbanion 2 with ethyl cyanoformate
are similar in magnitude to that of racemization of a-nitrile
carbanion 2, which led us to conceive the idea that careful
manipulation of experimental parameters would provide
opportunities for successful enantioselective trapping of chiral
a-nitrile-stabilized carbanions generated by deprotonation.
Change in solvent to toluene or THF decreased both the
chemical yield and enantiomeric ratio (Table 3, entries 1–3).
However, it should be noted that the use of THF provided 4 in
moderate enantiomeric ratio (68 : 32), in sharp contrast to the
result of chirality transfer in [2,3]-Wittig rearrangement employing
a chiral allyl benzyl carbanion in which almost complete racemiza-
tion was observed in THF against Et2O.16 Lowering of the reaction
temperature improved the chemical yield and enantiomeric ratio
(entries 1, 4 and 5) probably resulting from suppression of both the
Having established that the use of equal amounts of LDA and
cyanoformate gave an excellent chemical yield and a moderate
enantiomeric ratio, we next focused our attention on improvement
in enantioselectivity. We ascribed the origin of the moderate
enantioselectivity observed under the above conditions to
deceleration of the reaction of the carbanion with cyanoformate
by the steric hindrance around the carbon atom next to the CN
group. The steric congestion also retards the deprotonation
process and, as a consequence, gives rise to the reaction of LDA
with cyanoformate that may lead to more chance for racemization
by lowering of the concentration of cyanoformate. From this
point of view, we examined reactions using substrates bearing
a carbamoyl group with different steric bulkiness. As expected,
dimethylcarbamoyl derivative 1b, the least sterically hindered,
showed the highest enantioselectivity (85 : 15 er) (Table 5, entry 2),
Table 2 Effect of the amount of ethyl cyanoformate
Table 4 Effects of the amount of LDA and cyanoformate
4a
1a
Entry
X (equiv.)
Yield (%)
er
Yield (%)
er
4a
1a
1
2
3
4
5
5.0
4.0
3.0
2.0
1.0
39
41
51
69
76
74 : 26
73 : 27
67 : 33
67 : 33
63 : 37
57
53
47
26
—
100 : 0
100 : 0
100 : 0
100 : 0
—
Entry
X
Yield (%)
er
Yield (%)
er
1
2
3
5.0
3.0
1.1
89
92
77
74 : 26
74 : 26
73 : 27
—
—
16
—
—
100 : 0
c
2898 Chem. Commun., 2012, 48, 2897–2899
This journal is The Royal Society of Chemistry 2012

View Article Online
Table 5 Effect of the alkyl group in carbamate
electrophile in up to 90 : 10 er by taking advantage of a
combination of the fixing ability of a carbamoyl group and
the high reactivity of an acylating agent. As a result, the scope of
the fixation effect of carbamates was greatly expanded to include
lithiocarbanions next to a nitrile group. The upper limit has so
far been carbanions between allyl and benzyl groups.17a To our
knowledge, this represents the first example of successful
enantioselective trapping of an a-chiral acyclic nitrile carbanion.
This research was partially supported by a Grant-in-Aid for
Scientific Research (B) 22390001 (KT), a Grant-in-Aid for
Challenging Exploratory Research 2365900700 (KT) and a
Grant-in-Aid for Young Scientists (B) 22790011 (MS) from
the Ministry of Education, Culture, Sports, Science and
Technology (MEXT). We thank the Natural Science Center
for Basic Research and Development (N-BARD), Hiroshima
University, for the use of the facilities.
4a–e
1a–e
Yield
(%)
Yield
(%)
Entry
1
NR2
Solvent
Et2O
er
er
1
2
1a NiPr2
1b NMe2
92
74 : 26
85 : 15
—
—
—
—
THF : Et2O 88
(2 : 1)
3
4
5
1c NEt2
1d N(CH2)4 Et2O
1e N(CH2)5 Et2O
Et2O
89
72
85
80 : 20
76 : 24 23
80 : 20
—
—
100 : 0
—
—
Table 6 Reactions of 1b and 10b with an electrophile
Notes and references
1 (a) S. Arseniyadis, K. S. Kyler and D. S. Watt, Org. React., 1984,
31, 1; (b) F. F. Fleming and B. C. Shook, Tetrahedron, 2002, 58, 1.
2 P. R. Carlier, Chirality, 2003, 15, 340.
3 W. Zarges, M. Marsch, K. Harms and G. Boche, Angew. Chem.,
Int. Ed. Engl., 1989, 28, 1392.
4 T. Kawabata, K. Yahiro and K. Fuji, J. Am. Chem. Soc., 1991,
113, 9694.
Yield
Product (%)
Entry R Electrophile
El
er
5 P. R. Carlier, H. Zhao, J. DeGuzman and P. C.-H. Lam, J. Am.
Chem. Soc., 2003, 125, 11482.
1
2
3
4
5
6
7
H
NCCO2Et
Br NCCO2Et
CO2Et
CO2Et
Bn
CO2Et
COPh
COPh
4b
40b
3b
4b
5b
50b
92
81
89
89
90
93
84
90 : 10
88 : 12
50 : 50
53 : 47
84 : 16
89 : 11
81 : 19
6 (a) C. Wolf, Dynamic Stereochemistry of Chiral Compounds: Principles
and Applications, Royal Society of Chemistry, Cambridge, UK, 2008,
pp. 282; (b) P. R. Carlier, D. C. Hsu and S. A. Bryson, in Stereo-
chemical Aspects of Organolithium Compounds, Topics in Stereo-
chemistry, ed. R. E. Gawley and J. Siegel, Wiley, New York, 2010,
vol. 26, ch. 2; (c) D. Seebach, A. R. Sting and M. Hoffmann,
Angew. Chem., Int. Ed. Engl., 1996, 35, 2708.
7 M. Sasaki, E. Kawanishi, Y. Shirakawa, M. Kawahata, H. Masu,
K. Yamaguchi and K. Takeda, Eur. J. Org. Chem., 2008, 3061.
8 M. Sasaki, Y. Shirakawa, M. Kawahata, K. Yamaguchi and
K. Takeda, Chem.–Eur. J., 2009, 15, 3363.
H
H
H
BnBr
ClCO2Et
PhCOCl
Br PhCOCl
H
CH3CH2CH2- COCH2CH2CH3 6b
COCla
(CH3)2CHCOCl COCH(CH3)2
(CH3)3CCOCl COC(CH3)3
8
9
H
H
7b
8b
85
87
74 : 26
61 : 39
a
Five equivalents of LDA and butyryl chloride were used, respectively.
9 (a) H. M. Walborsky and J. M. Motes, J. Am. Chem. Soc., 1970,
92, 2445; For a magnesiated cyclopropyl nitrile, see: (b) P. R. Carlier
and Y. Zhang, Org. Lett., 2007, 9, 1319.
in which a mixed solvent system (THF–Et2O = 2:1) was used
because of the low solubility of 1b in Et2O.
Finally, we were pleased to find that lowering the reaction
temperature to ꢀ114 1C provided further improvement in
chemical yield and enantioselectivity, affording 4b in 92%
yield and 90 : 10 er (Table 6, entry 1). The absolute configu-
ration of the major enantiomer was determined on the basis of
X-ray crystallographic analysis with anomalous dispersion of
a compound derived from the corresponding p-bromophenyl
derivative 10b (entry 2, see ESIw), indicating that the substitu-
tion reaction occurs with inversion of the configuration, which
is consistent with expectations from previous findings.17
The optimized conditions for ethyl cyanoformate were applied
to the reactions with BnBr, ethyl chloroformate, and some acid
chlorides. Whereas the former two reagents led to the formation of
racemic or almost racemic products (entries 3 and 4), reactions
with an acid chloride provided acylated derivatives in enantiomeric
ratios depending on the reactivity of acid chlorides (entries 5–9).18
The fact that the enantiomeric ratios increase as an acid
chloride is more reactive or less bulky suggests again that
the rate of racemization of the lithiocarbanion is similar in
magnitude to that of reaction with the electrophiles.
10 (a) D. Hoppe and T. Hense, Angew. Chem., Int. Ed. Engl., 1997,
36, 2282; (b) J.-C. Kizirian, in Stereochemical Aspects of Organolithium
Compounds, Topics in Stereochemistry, ed. R. E. Gawley and
J. Siegel, Wiley, New York, 2010, vol. 26, ch. 6.
11 D. Hoppe, Angew. Chem., Int. Ed. Engl., 1984, 23, 932.
12 D. Hoppe, F. Marr and M. Bruggemann, in Organolithiums in
¨
Enantioselective Synthesis, ed. D. M. Hodgson, Springer, New York,
2003, pp. 61.
13 (a) D. Hoppe and T. Kramer, Angew. Chem., Int. Ed. Engl., 1986,
¨
25, 160; (b) S. Dreller, M. Drybusch and D. Hoppe, Synlett, 1991,
397; (c) D. Hoppe, A. Carstens and T. Kramer, Angew. Chem., Int.
¨
Ed. Engl., 1990, 29, 1424; (d) J. Clayden, Organolithiums: Selectivity for
Synthesis, Pergamon, Oxford, 2002; (e) A. Basu and S. Thayumanavan,
Angew. Chem., Int. Ed., 2002, 41, 716.
14 L. M. Mander and S. P. Sethi, Tetrahedron Lett., 1983, 24, 5425.
15 (a) K. Tanaka and K. Takeda, Tetrahedron Lett., 2004, 45, 7859; (b) K.
Tanaka and K. Takeda, Tetrahedron Lett., 2005, 46, 6429; (c) X.
Linghu, D. A. Nicewicz and J. S. Johnson, Org. Lett., 2002, 4, 2957.
16 (a) M. Sasaki, M. Higashi, H. Masu, K. Yamaguchi and K. Takeda,
Org. Lett., 2005, 7, 5913; (b) M. Sasaki, H. Ikemoto, M. Kawahata,
K. Yamaguchi and K. Takeda, Chem.–Eur. J., 2009, 15, 4663; (c) H.
Ikemoto, M. Sasaki and K. Takeda, Eur. J. Org. Chem., 2010, 6643.
17 (a) H. Ikemoto, M. Sasaki, M. Kawahata, K. Yamaguchi and
K. Takeda, Eur. J. Org. Chem., 2011, 6553; (b) A. Carstens and
D. Hoppe, Tetrahedron, 1994, 50, 6097.
18 The absolute configurations of the major enantiomers of acylated
products 5b–8b were assigned by analogy to 50b whose absolute
configuration was confirmed by X-ray crystallographic analysis
with anomalous dispersion.
In conclusion, we have demonstrated that a chiral a-nitrile
carbanion generated by deprotonation of enantioenriched
O-carbamoylcyanohydrin is able to be trapped by a carbon
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 2897–2899 2899