An approach to 2-cyanopyrrolidines bearing a chiral auxiliary

Article abstract of DOI:10.1016/j.tetasy.2007.01.008

The reaction of 2-methyl-2-((1-phenylethyl)amino)propanenitrile with different γ-halocarbonyl compounds is investigated. The influence of different parameters such as the nature of the substrate and solvent, is discussed. The reaction is considered as a convenient route to 2-cyanopyrrolidines in the case of aliphatic γ-halocarbonyl compounds possessing a reasonably reactive carbonyl group. In many cases, the products can be obtained as single enantiomers. It is shown that cyclopropyl ketones can also react with 2-methyl-2-((1-phenylethyl)amino)propanenitrile to give 2-cyanopyrrolidines. A mechanistic scheme is proposed in order to explain the experimental facts observed.

Full text of DOI:10.1016/j.tetasy.2007.01.008

Tetrahedron: Asymmetry 18 (2007) 290–297
An approach to 2-cyanopyrrolidines bearing a chiral auxiliary
Oleksandr O. Grygorenko,a Nataliya A. Kopylova,a Pavel K. Mikhailiuk,a
Anja Meißnera and Igor V. Komarova,b,
*
aDepartment of Chemistry, Kyiv National Taras Shevchenko University, Volodymyrska Street, 64, Kyiv 01033, Ukraine
bEnamine Ltd, Alexandra Matrosova Street, 23, Kyiv 01103, Ukraine
Received 20 December 2006; accepted 9 January 2007
Available online 2 February 2007
Abstract—The reaction of 2-methyl-2-((1-phenylethyl)amino)propanenitrile with different c-halocarbonyl compounds is investigated.
The influence of different parameters such as the nature of the substrate and solvent, is discussed. The reaction is considered as a con-
venient route to 2-cyanopyrrolidines in the case of aliphatic c-halocarbonyl compounds possessing a reasonably reactive carbonyl group.
In many cases, the products can be obtained as single enantiomers. It is shown that cyclopropyl ketones can also react with 2-methyl-2-
((1-phenylethyl)amino)propanenitrile to give 2-cyanopyrrolidines. A mechanistic scheme is proposed in order to explain the experimental
facts observed.
Ó 2007 Elsevier Ltd. All rights reserved.
O
1. Introduction
CN
N
MeOH
Ph
Ph
N
+
+
2-Cyanopyrrolidines have been used as building blocks for
the synthesis of cage systems, alkaloids and their
analogues, amino acids and other biologically active
compounds.1 One of the most interesting applications of
2-cyanopyrrolidines in organic synthesis is their transfor-
mation to proline analogues.2 Therefore, general methods
for the synthesis of 2-cyanopyrrolidines are of great inter-
est. In most cases these compounds are chiral, and stereo-
selective approaches to their synthesis are often required.
Ph
N
H
CN
NC
reflux, 30 h
Ph
3a (40%)
Cl
2
3b (40%)
1
O
CN
CH3CN
N
+
1
N
+
+
N
NC
reflux, 30 h
Ph
Cl
Ph
5a (8%)
NC
6 (24%)
4
5b (8%)
Scheme 1.
Recently, we found3 that compound 1 reacted with 3-chloro-
methylcyclohexanone 2 to give diastereomeric 2-cyano-
pyrrolidines 3a and 3b in a good yield (Scheme 1).
3-Chloromethylcyclopentanone 4, which possesses a less
reactive carbonyl group compared to compound 2, gave
2-cyanopyrrolidines 5a and 5b in lower yield; the main
product of the reaction was the unexpected compound 6.
The diastereomers obtained in both reactions shown in
Scheme 1 were separated by column chromatography.
presence of the chiral auxiliary, the a-phenylethylamine
moiety, might allow a separation of diastereomers and/or
chiral induction in further transformations of the 2-cyano-
pyrrolidines. On the other hand, this chiral auxiliary can be
easily removed by hydrogenation. Therefore, we decided to
investigate the scope of the reaction of 1 with c-halo-
ketones and c-haloaldehydes. The carbonyl compounds
7–11 with a different reactivity of the carbonyl group were
selected as model substrates in addition to 2 and 4.
The reaction of a-aminonitrile 1 with c-halocarbonyl com-
pounds might have great potential as a stereoselective
approach to 2-cyanopyrrolidines and their derivatives. The
2. Results and discussion
*
Several parameters, namely, the nature of the substrate and
the solvent, temperature, reagents ratio, etc. might
Corresponding author. Tel.: +38 044 239 33 15; fax: +38 044 235 12 73;
e-mail: ik214@yahoo.com
0957-4166/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2007.01.008

O. O. Grygorenko et al. / Tetrahedron: Asymmetry 18 (2007) 290–297
291
H
O
O
O
O
Cl
O
Br
Cl
Cl
Br
CF3
H
F
7
8
9
10
11
influence the reaction of a-aminonitrile 1 with different
c-chlorocarbonyl compounds.
reaction with any of the nucleophiles present in the reac-
tion mixture, resulting in very low yields of the expected
2-cyanopyrrolidines.
We first varied the reaction conditions using ketones 2 and
4 as the model substrates. Isolation of the products by col-
umn chromatography and investigation of the reaction
mixtures by NMR and GC–MS were used to study the
reaction progress. In a previous paper,3 we reported only
the optimized conditions for both the reactions of 2 and
4; herein, we summarize all our experiments. It was found
that the solvent nature has a dramatic influence on the
reaction progress and outcome. No 2-cyanopyrrolidines 5
and 6 were formed from 4 in low-polar solvents such as
benzene or chloroform; these results are in accordance with
those reported in the literature4 for the analogous transfor-
mation. Only trace amounts of 5 and 6 were formed from 4
in methanol, whereas moderate yields of these products
were obtained in other polar solvents such as acetonitrile,
dimethylsulfoxide or 2-propanol. In the case of ketone 2,
bridged 2-cyanopyrrolidines 3 were obtained in high yields
either in methanol, acetonitrile or DMSO, although the
yield of 3 was diminished in benzene as the solvent.
Moderate yields of the corresponding 2-cyanopyrrolidines
were obtained from ketones 4 (Scheme 1) and 8 (Scheme
3 ).
Ph
CH3CN
8
+
1
N
CN
reflux, 30 h
14a,b (13%)
1 : 0.4 mixture of
diastereomers
Scheme 3.
As we have reported earlier,3 unexpected product 6 was
mainly formed in the case of 4. We suggested that cyclo-
propane 15 and 4-chlorocyclohexanone 16 could be inter-
mediates in the formation of compound 6. Therefore the
reaction of 1 with compounds 15 and 16 was performed
to test this hypothesis. In the case of 16, the yields of com-
pounds 5a, 5b and 6 (10%, 7% and 23%, respectively) were
practically identical to those for the reaction of compound
4. Cyclopropane 15 also gave cyanopyrrolidines 5a, 5b and
6 when reacted with 1 in the presence of 1.05 equiv of a-
phenylethylamine hydrochloride despite the yields of the
products were lower (2%, 2% and 13% for 5a, 5b and 6,
respectively). No reaction proceeded without a chloride
ion donor.
Increasing the reaction time and/or temperature mainly
influenced the rate of the reaction and not the product dis-
tribution and yields, both of 3 and 5 and 6.
The substrate nature appeared to be the dominating factor
for determining the reaction progress. Carbonyl group
reactivity towards nucleophiles was shown to be highly
critical for achieving good yields of 2-cyanopyrrolidines.
While the reactive c-chloroketone 2 gave 2-cyanopyrrol-
idines in a good total yield, compound 7 possessing a very
low-electrophilic carbonyl group did not give the corre-
sponding 2-aryl-2-cyanopyrrolidines at all. Even after a
prolonged reaction time, only the starting compound was
recovered.
O
O
Cl
15
16
Reaction of 1 with ketone 11 possessing a highly reactive
carbonyl group led to a complex mixture of products.
Traces of the trifluoromethyl-substituted 2-cyanopyrrol-
idines 12a and 12b were detected in the reaction mixture
by means of NMR and LC–MS. One of these diastereo-
mers was isolated although in a very low yield (less than
1%). Compound 13 was also isolated in a 5% yield (Scheme
2). We believe that high reactivity of ketone 11 led to its
These results allowed us to suggest that other cyclopro-
panes might be used as starting compounds for the synthe-
sis of 2-cyanopyrrolidines by a cascade sequence: cyclo-
propane ring opening—Strecker reaction—intramolecular
nucleophilic substitution. It indeed proved to be true, at
least for cyclopropyl ketones 17 and 18 (Scheme 4à).
Other cyclohexanone derivatives 19 and 20, which we have
tested in this reaction were obtained from (R)-carvone 21
by the method described in the literature.5 The reaction
Ph
O
CH3CN
11
1
O
+
+
N
CN
CF3
reflux, 30 h
CF3
13 (5%)
12a,b (1%)
Total yield of 14 is 63% by NMR.
à Yield of 14 is detected by NMR.
Scheme 2.

292
O. O. Grygorenko et al. / Tetrahedron: Asymmetry 18 (2007) 290–297
O
was obtained in 66% yield as 1:0.7 mixture of diastereomers
(Scheme 6).
CH3CN
reflux, 30 h
+
1 +
3a (38%)
+
3b (41%)
Ph
NH+
3 Cl
Ph
17
CH3CN
150 h
O
CH3CN
reflux, 30 h
10
+
1
N
CN
+
14a,b (15%)
+
1
NH+
3 Cl
Ph
23a,b (66%)
18
Scheme 6.
Scheme 4.
The reaction of aminonitrile 1 with aromatic aldehyde 9
also proceeded at an ambient temperature as in the case
of 10. However, a complex mixture of unidentified com-
pounds was obtained. We were only able to isolate isoindo-
lone 24 by column chromatography in a 12% yield.6
Supposedly, the unstable isoindole 25 was formed as the
main product of the reaction, which underwent oxidation
by air oxygen forming 24. We trapped isoindole 25 by a
Diels–Alder reaction with maleimide 26; compound 27
was obtained in a 40% yield (Scheme 7).
of 20 with reagent 1 proceeded much more slowly than the
reaction of 2; after 70 h of reflux, we obtained product 22
in a 39% yield (Scheme 5).
O
O
O
.
Me3SOI
py HCl
KOH, DMF
CH3CN
Cl
We believe that it was aromaticity of isoindole 25 that
led to its formation instead of the corresponding 2-
cyanopyrrolidine.
21
19 (83%)
20 (47%)
8
CN
CN
4
CH3CN
5
N
N6
Ph
2
Ph
+
20
+
1
The mechanistic considerations described in the litera-
ture4,7 allowed us to suggest aminonitriles, such as 28, as
the key intermediates of the reaction of 1 with c-halocar-
bonyl compounds. Scheme 8 illustrates this for the reaction
of 1 with one of the enantiomers of ketone 4. In the case of
cyclic ketones, only one diastereomer of aminonitrile 28
(namely, 28a) can undergo the cyclization, while the sec-
ond, 28b, has to be converted to 28a.
1
7
reflux, 70 h
3
22a (19%)
22b(20%)
Scheme 5.
Since the starting c-chloroketone 20 was enantiomerically
pure, we expected only one stereoisomer of product 22 to
be formed. Surprisingly, compound 22 was obtained as a
1:1 mixture of diastereomers 22a and 22b, which were suc-
cessfully separated by means of semi-preparative HPLC.
Indirect proof of this scheme is the observation of inter-
mediates of type 28.7 We also observed intermediates 30
and 31 by GS–MS analysis of the reaction mixture.
The reaction between cyclopropane derivative 19 and 1
under the conditions described above only yielded starting
compounds after 40 h of refluxing. No products 22a and
22b were detected.
In our opinion, it is critical for the success of the reaction,
that 1 provides hydrogen cyanide and a-phenylethylamine
in a low concentration. A higher concentration of these
nucleophiles might cause undesirable side reactions.3,4,7
The carbonyl group of aldehydes is far more reactive than
that of ketones, therefore, it was not surprising that the
reaction of 1 with aldehyde 10 proceeded already at an
ambient temperature. The expected 2-cyanopyrrolidine 23
The results obtained in this study prompted us to make
some additional comments and refinements to the scheme
described above. The pathway through iminium salts
mentioned in a previous work4 seems to not be dominant:
9
+
1
Ph
N
O
N
O
O
Ph
Ph
26
[O]
O
N
N
N
O
O
O
24 (12%)
25
27 (40%)
Scheme 7.

O. O. Grygorenko et al. / Tetrahedron: Asymmetry 18 (2007) 290–297
293
HO CN
+
Ph
N
Ph
HN CN
Cl
O
29
30
Ph
Cl
Cl
4
O
HN CN
28a
+
+
+
Cl
O
28b
Ph
N
H
CN
1
N
Ph
+
HO
CN
this way is impossible
for cyclic ketones
Cl
31
Ph
O
.
HCl
N
N+
NC
+
+
Cl
HO
CN
Ph
5a
32
Scheme 8.
B; -BH+Cl-
BH+Cl-; -B
BH+Cl-; -B
B; -BH+Cl-
4
15
16
ACH; -1
ACH; -1
ACH; -1
1; -ACH
1; -ACH
1; -ACH
N
Ph
B; -BH+Cl-
Ph
BH+Cl-; -B
N
N
Ph
6
5
BH+Cl-; -B
B; -BH+Cl-
Cl
Cl
34
31
33
Scheme 9. (ACH = acetone cyanohydrin).
cyclic c-chloroketones reacted with 1 as efficiently as acy-
clic ones, although the first should form highly strained
anti-Bredt intermediates such as 32. The experimental facts
also point out that additional equilibria should be
accounted for, namely, formation of the cyclopropane
derivatives, as illustrated by Scheme 9. Any of the amines
present in the reaction mixture can play the role of the base
(B) necessary for these transformations.
This mechanistic scheme also explains the epimerization
during the synthesis of 2-cyanopyrrolidines from 2-substi-
tuted cyclohexanone 20. It proceeds due to reversible
deprotonation of the starting c-chloroketone, which is
the first step in the formation of the corresponding cyclo-
propane derivative 19 (Scheme 10).
3. Conclusions
The structure of the unexpected product 6 obtained from
compound 4, observation of 15 and 33 by GS–MS
analysis of the reaction mixture and formation of
2-cyanopyrrolidines by the reaction of cyclopropanes
15, 17 and 18 with reagent 1 are all evidences for these
suggestions.
Reaction of a-aminonitrile 1 with c-halocarbonyl com-
pounds is the result of a subtle balance of the reactivity
of the substrate, solvent nature and specific properties of
1 as the source of a-phenylethylamine and hydrogen
cyanide (Fig. 1). This approach can be considered as a

294
O. O. Grygorenko et al. / Tetrahedron: Asymmetry 18 (2007) 290–297
O
4.1. (1R,4R,6S)-4-Isopropenyl-1-methylbicyclo[4.1.0]-
heptan-2-one 19
O
B; -BH+
BH+; -B
BH+; -B
B; -BH+
20
Trimethylsulfoxonium iodide (50.4 g, 0.229 mol) and 12.9 g
(0.229 mol) of potassium hydroxide were dissolved in
330 ml of DMF at 30–40 °C under vigorous stirring. The
clear solution formed was cooled to 20 °C, and the solution
of 32.8 g (0.219 mol) of (R)-carvone in 70 ml of DMF was
added. The obtained reaction mixture was stirred for 3 h
and allowed to stand overnight, then poured into water
and extracted with dichloromethane. The combined
extracts were washed with water, dried over MgSO4
and evaporated under reduced pressure to give 30 g
(0.183 mol) of 20, which was used for the next step without
further purification. For spectral and physical data see
Ref. 13.
Cl
Cl
-Cl-
22a
19
22b
Scheme 10.
100
80
60
40
20
0
4.2. (2S,3R,5R)-3-(Chloromethyl)-5-isopropenyl-2-methyl-
cyclohexanone 20
7
8
4
2
10 11
Pyridinium hydrochloride (15 g, 130 mmol) was added to a
solution of 10.6 g (65 mmol) of compound 20 in 70 ml of
acetonitrile. The mixture was refluxed for 36 h, then poured
into water and extracted with ether. The combined organic
extracts were dried over MgSO4 and evaporated under re-
duced pressure. The residue was purified by column chro-
matography (benzene–ethyl acetate (13:1) as an eluent) to
give 6.1 g (31 mmol, 47%) of 21, which was used for the
further transformations. To obtain an analytical sample
of 21, the product was chromatographed once more [hex-
ane–benzene (2:1) as an eluent]. [a]D = ꢀ18.8 (c 0.41,
MeOH). 1H NMR (CDCl3), d: 4.92 (s, 1H, C(CH3)@CH2),
4.75 (s, 1H, C(CH3)@CH2), 3.68 (m, 2H, CH2Cl), 2.8 (m,
1H), 2.65 (dd, 1H, J = 14.8 and 3.2 Hz), 2.51 (dd, 1H,
J = 14.8 and 6.0 Hz), 2.39 (m, 1H), 2.09 (m, 2H, CH2),
1.93 (m, 1H), 1.82 (s, 3H, C(CH3)@CH2), 1.30 (d,
3H, J = 6.8 Hz, 2-CH3). 13C NMR (CDCl3), d: 205.0
(C@O), 144.6 (C@CH2), 111.5 (C@CH2), 46.9, 45.2, 42.3,
39.7, 38.6, 29.5, 20.8, 10.5. IR: 1712 (m C@O). MS (m/z):
201 (MH+), 165 (M+ꢀCl). Anal. Calcd for C11H17ClO:
C, 65.83; H, 8.54; Cl, 17.66. Found: C, 65.52; H, 8.25;
Cl, 17.87.
Figure 1. Total yields of 2-cyanopyrrolidines obtained from different c-
halocarbonyl compounds (the substrates are given in the order of
increasing reactivity of their carbonyl group towards nucleophiles).
convenient route to 2-cyanopyrrolidines in the case of ali-
phatic c-halocarbonyl compounds. The presence of a chiral
auxiliary—a-phenylethylamine moiety—allows chromato-
graphic separation of stereoisomers in many cases; hence
the synthesis of enantiomerically pure products is possible.
4. Experimental
All air- and moisture-sensitive reactions were performed
under an argon atmosphere using a standard Schlenk tech-
nique. The solvents were purified according to a standard
procedures.8 2-Methyl-2-((1-phenylethyl)amino)propane-
nitrile 1,3 3-chloromethylcyclohexanone 2,5 3-chloro-
methylcyclopentanone 4,5 5-chloro-2-pentanone 8,5
2-(bromomethyl)benzaldehyde 9,9 4-chlorobutyralde-
hyde 10,10 1,1,1-trifluoro-5-bromo-2-pentanone 11,11 bi-
cyclo[3.1.0]hexan-2-one 15,3 4-chlorocyclohexanone 16,12
bicyclo[4.1.0]heptan-2-one 173 were prepared according to
the procedures described in the literature. All other starting
materials were purchased from Acros, Merck, Aldrich
and Fluka chemicals. Melting points are uncorrected.
Analytical TLC was performed using Polychrom SI F254
plates. Column chromatography was performed using
4.3. Reaction of 1 with c-halocarbonyl compounds
(general procedure)3
Aminonitrile 1 (1.05 equiv) was added to a solution of
1 equiv of c-halocarbonyl compound in dry acetonitrile.
The mixture obtained was refluxed or was allowed to stand
under an argon atmosphere under the conditions listed in
Table 1, then poured into excess of 10% sodium hydroxide
solution and extracted with dichloromethane. The com-
bined extracts were dried (MgSO4) and evaporated under
reduced pressure. The residue was analysed by NMR and
GS–MS and/or chromatographed.
1
Kieselgel Merck 60 (230–400 mesh) stationary phase. H
NMR, 19F NMR and 13C NMR spectra were recorded
on Varian Unity Plus 400 spectrometer at 400.1, 376.7
and 100.7 MHz, respectively. Chemical shifts are reported
in parts per million downfield from TMS (1H and 13C
NMR) or C6F6 (19F NMR) as an internal standards. IR
spectra were obtained on Nicolet Nexus 470 spectrometer.
mmax (cmꢀ1) values in IR spectra are given for the main
absorption bands. Mass spectra were recorded on Agilent
1100 LCMSD SL instrument with chemical ionization
(CI). Optical rotation values were measured on a Perkin-
Elmer 341 polarimeter.
4.4. Reaction of 1 with cyclopropyl ketones
(general procedure)
Aminonitrile 1 (1.05 equiv) and 1.05 equiv of (S)-a-phenyl-
ethylamine hydrochloride were added to a solution of
1 equiv of cyclopropyl ketone in dry acetonitrile. The

O. O. Grygorenko et al. / Tetrahedron: Asymmetry 18 (2007) 290–297
Table 1. Reaction of a-aminonitrile 1 with different substrates
295
Entry no.
Substrate
Reaction conditions
Eluent for the chromatography
Products (yields, %)
Temperature
Time (h)
1
2
3
4
5
6
7
8
9
7
8
Reflux
Reflux
Ambient
Ambient
Reflux
Reflux
Reflux
Reflux
Reflux
Reflux
Reflux
70
30
50
150
10
30
30
30
30
40
70
Hexane–ethyl acetate (2:1)
Hexane–ethyl acetate (2:1)
Hexane–ethyl acetate (5:1)
Hexane–ethyl acetate (5:1)
Hexane–ethyl acetate (20:1)
The same as for 43
—
14a(9), 14b(4)a
21(12)
9
10
11
15
16
17
18
19
20
23a(39), 23b(27)
12a(1), 13(5)
5a(2), 5b(2), 6(13)
5a(10), 5b(7), 6(23)
3a(39), 3b(39)
14a(11), 14b(4)b
—
The same as for 43
The same as for 23
The same as for 8
Hexane–ethyl acetate (5:1)
The same as for 19
10
11
22a(19), 22b(20)
a Total yield of 14 is 63% by NMR.
b Yield of 14 is detected by NMR.
mixture obtained was refluxed under an argon atmosphere
under the conditions listed in Table 1 and then worked up
as described above.
(dd, J = 17.5 and 8.0 Hz, 5-CH2 of 14b), 2.35 (m, 1H, 3-
CH2 of 14b), 2.29 (m, 1H, 3-CH2 of 14a), 1.96 (m, 1H, 3-
CH2 of 14a), 1.85 (m, 3H, 3-CH2 and both 4-CH2 of
14a), 1.74 (m, 2H, 4-CH2 of 14b), 1.67 (s, 3H, CH3 of
14b), 1.49 (d, 3H, J = 6.0 Hz, CHCH3 of 14a), 1.49 (d,
3H, J = 6.5 Hz, CHCH3 of 14b), 1.03 (s, 3H, CH3 of
14a). 13C NMR (CDCl3), d: 145.6 (ipso-C6H5 of 14a),
144.3 (ipso-C6H5 of 14b), 128.4, 128.3, 127.4, 127.3,
127.2, 127.1, 121.5 (CN of 14a), 120.9 (CN of 14b), 61.1
(C-2 of 14a), 61.0 (CHCH3 of 14b), 59.3 (C-2 of 14b),
58.6 (CHCH3 of 14a), 52.1 (5-CH2 of 14b), 49.1 (5-CH2
of 14a), 41.7 (3-CH2 of 14b), 40.4 (3-CH2 of 14a), 27.0
(CH3 of 14b), 25.5 (CH3 of 14a), 23.7 (CHCH3 of 14b),
20.9 (CHCH3 of 14a), 20.8 (4-CH2 of 14b), 19.9 (4-CH2
of 14a). To assign the signals in NMR spectra and to con-
4.5. 1-[(1S)-1-Phenylethyl]-2-(trifluoromethyl)pyrrolidine-2-
carbonitrile 12a
This was obtained from 1,1,1-trifluoro-5-bromo-2-penta-
1
none 11 in a 1% yield. H NMR (CDCl3), d: 7.13–7.24
(m, 5H), 4.44 (q, J = 6.8 Hz, 1H, CHCH3), 2.89 (q,
J = 7.2 Hz, 1H, 5-CH2), 2.75 (m, 1H, 5-CH2), 2.39 and
2.32 (m, 2H, 3-CH2), 1.78 (m, 2H, 4-CH2), 1.49 (d,
J = 6.8 Hz, 3H, CHCH3). 13C NMR (CDCl3), d: 143.2
(ipso-C6H5), 131.0 (q, J = 252 Hz, CF3), 128.3, 126.9,
126.5, 117.2 (CN), 68.2 (C-2), 55.5 (CHCH3), 45.3, 35.7,
23.0 (CH3), 14.7. 19F NMR (CDCl3), d: 84,6 (s, CF3). IR:
2232 (m C„N). MS (EI, m/z): 268 (M+), 105. Anal. Calcd
for C14H15F3N2: C, 62.68; H, 5.64; N, 10.44. Found: C,
62.39; H, 5.25; N, 10.42.
1
firm the cyclic structure of 11, H–1H COSY, IPT, HSQC
and HMBC experiments were carried out. IR: 2219 (m
C„N). MS (m/z): 188 (M+ꢀCN), 105. Anal. Calcd for
C14H18N2: C, 78.46; H, 8.47; N, 13.07. Found: C, 78.09;
H, 8.25; N, 12.94.
4.6. 1-[2-(Trifluoromethyl)tetrahydrofuran-2-yl]acetone 13
4.8. 3-Isopropenyl-8-methyl-6-[(1R)-1-phenylethyl]-6-azabi-
cyclo[3.2.1]octane-5-carbonitrile 22
This was obtained from 1,1,1-trifluoro-5-bromo-2-penta-
none 11 in a 5% yield. 1H NMR (CDCl3), d: 3.85 (d,
J = 7.2 Hz, 2H, OCH2), 2.88 (d, J = 14.4 Hz, 1H,
C(O)CH2), 2.41 (d, J = 14.4 Hz, 1H, C(O)CH2), 2.34 (m,
1H), 2.11 (s, 3H, CH3), 2.05 (m, 1H), 1.94 (m, 2H). 13C
NMR (CDCl3), d: 201.2 (CO), 124.6 (q, J = 281.5 Hz,
CF3), 81.9 (q, J = 27.6 Hz, CCF3), 69.2 (OCH2), 43.2,
31.1 (CH3), 28.5, 24.6. 19F NMR (CDCl3), d: 84,7 (s,
CF3). IR: 1716 (m C@O). MS (EI, m/z): 127 (M+ꢀCF3),
69 ðCF3þÞ, 43 (CH3CO+). Anal. Calcd for C8H11F3O2: C,
48.98; H, 5.65. Found: C, 48.75; H, 5.44.
This was obtained as 1:1 mixture of (1R,3R,5S,8S)- (22a)
and (1R,3R,5S,8R)- (22b) diastereomers from ketone 19
in a 39% yield.§ A sample of compound 22 was separated
by semi-preparative HPLC to give pure diastereomers
22a and 22b (Zorbax SB-C18 semi-preparative column,
CH3CN–H2O (70:30) as an eluent, 30 °C, flow rate—
1.7 ml/min, retention times 27.96 min (22a) and 29.66 min
(22b)).
4.9. (1R,3R,5S,8S)-3-Isopropenyl-8-methyl-6-[(1R)-1-phen-
ylethyl]-6-azabicyclo[3.2.1]octane-5-carbonitrile 22a
4.7. 2-Methyl-1-((1S)-1-phenylethyl)pyrrolidine-2-carbo-
nitrile 14
1H NMR (CDCl3), d: 7.37 (d, J = 7.5 Hz, 2H, o-C6H5),
7.32 (t, J = 7.5 Hz, 2H, m-C6H5), 7.24 (t, J = 7.2 Hz, 1H,
p-C6H5), 4.76 (s, 1H, CH3C@CH2), 4.75 (s, 1H,
CH3C@CH2), 4.11 (q, J = 6.7 Hz, 1H, CH(C6H5)CH3),
3.10 (dd, J = 10.0 and 5.6 Hz, 1H, exo-7-CH2), 2.52
(m, 1H, 8-CH), 2.25 (d, J = 10.0 Hz, 1H, endo-7-CH2),
This was obtained as a mixture of (2R)- and (2S)-diastereo-
mers (14a:14b = 1:0.4) from 5-chloro-2-pentanone 8 in a
1
13% yield. H NMR (CDCl3), d: 7.22–7.39 (m, 5H, C6H5
of 14a and 5H, C6H5 of 14b), 3.94 (q, J = 6.7 Hz, 1H,
CHCH3 of 14a), 3.85 (q, J = 7.0 Hz, 1H, CHCH3 of
14b), 3.22 (m, 1H, 5-CH2 of 14a), 2.89 (m, 1H, 5-CH2 of
14b), 2.80 (dd, J = 17.5 and 8.0 Hz, 5-CH2 of 14a), 2.48
§ (R)-Isomer of 1 was used for the reaction.

296
O. O. Grygorenko et al. / Tetrahedron: Asymmetry 18 (2007) 290–297
2.13–2.33 (m, 3H, 3-CH, endo-4-CH2 and 1-CH), 1.85 (dd,
J = 13.6 and 12.4 Hz, 1H, exo-4-CH2), 1.75 (s, 3H,
CH3C@CH2), 1.66 (t, J = 13.2 Hz, 1H, endo-2-CH2), 1.61
(d, J = 6.5 Hz, 3H, CH(C6H5)CH3), 1.36 (m, 1H, exo-2-
CH2), 1.23 (d, J = 7.0 Hz, 3H, 8-CHCH3). 13C NMR
(CDCl3), d: 147.6 (C@CH2), 145.5 (ipso-C6H5), 128.6 (m-
C6H5), 127.4 (C6H5), 127.3 (C6H5), 123.5 (CN), 110.2
(C@CH2), 60.6 and 60.5 (5-C and CH(C6H5)CH3), 56.6
(7-CH2), 44.9 (8-CH), 37.3 (3-CH), 37.0 (1-CH), 31.5 (4-
CH2), 28.4 (2-CH2), 24.1 (CH(C6H5)CH3), 20.7
(CH3C@CH2), 10.3 (8-CHCH3). To assign the signals in
143.9 (ipso-C6H5 of 23a), 128.7, 128.6, 127.6, 127.4,
127.2, 121.0, 118.0 (CN of 23a), 117.8 (CN of 23b), 61.8
(CHCH3 of 23b), 61.6 (CHCH3 of 23b), 53.2 (2-CH of
23a), 52.0 (2-CH of 23b), 50.4 (5-CH2 of 23b), 48.8 (5-
CH2 of 23a), 29.8 (3-CH2 of 23b), 29.5 (3-CH2 of 23a),
23.1 (CHCH3), 23.0 (CHCH3), 21.9 (4-CH2), 21.8 (4-
CH2). IR: 2220 (m C„N). MS (m/z): 174 (M+ꢀCN), 105.
Anal. Calcd for C13H16N2: C, 77.96; H, 8.05; N, 13.99.
Found: C, 78.17; H, 7.69; N, 14.33.
4.12. 2-((1R)-1-Phenylethyl)isoindolin-1-one 24
1
NMR spectra, H–1H COSY, HSQC, HMBC and NOE
experiments were carried out. IR: 2234 (m C„N). MS
(m/z): 295(MH+), 191. Anal. Calcd for C20H26N2: C,
81.59; H, 8.90; N, 9.51. Found: C, 81.94; H, 8.62; N,
9.20. The results of combustion elemental analysis were
obtained for the mixture of diastereomers 22a and 22b.
This was obtained from 2-(bromomethyl)benzaldehyde 9 in
a 12% yield.§ Mp = 140–143 °C (lit.6 142–144 °C). For
spectral and physical data see Ref. 6.
4.13. (3aR,4S,9R,9aS)-2-(4-Methoxyphenyl)-10-[(1R)-1-
phenylethyl]-3a,4,9,9a-tetrahydro-1H-4,9-epiminobenzo[f ]-
isoindole-1,3-dione 27
4.10. (1R,3R,5S,8R)-3-Isopropenyl-8-methyl-6-[(1R)-1-
phenylethyl]-6-azabicyclo[3.2.1]octane-5-carbonitrile 22b
Compound 1§ (180 mg) and 0.20 g of N-(p-methoxyphen-
1H NMR (CDCl3), d: 7.29–7.38 (m, 4H, o-C6H5 and
m-C6H5), 7.25 (t, J = 7.2 Hz, 1H, p-C6H5), 4.78 (s, 1H,
CH3C@CH2), 4.70 (s, 1H, CH3C@CH2), 4.18 (q,
J = 6.8 Hz, 1H, CH(C6H5)CH3), 3.10 (dd, J = 9.6 and
6.0 Hz, 1H, exo-7-CH2), 2.46 (dd, J = 13.5 and 5.0 Hz,
1H, endo-4-CH2), 2.26 (d, J = 10 Hz, 1H, endo-7-CH2),
2.22 (m, 1H, endo-2-CH2), 2.04 (br s, 1H, 1-CH), 1.98 (q,
J = 6.8 Hz, 1H, 8-CH), 1.77 (t, J = 13.0 Hz, 1H, exo-4-
CH2), 1.68 (s, 3H, CH3C@CH2), 1.64 (m, 1H, 3-CH),
1.61 (d, J = 6.5 Hz, 3H, CH(C6H5)CH3), 1.47 (t,
J = 12.8 Hz, 1H, exo-2-CH2), 1.30 (d, J = 7.0 Hz, 3H,
8-CHCH3). 13C NMR (CDCl3), d: 147.3 (C@CH2), 145.6
(ipso-C6H5), 128.6 (m-C6H5), 127.2 (C6H5), 127.1 (C6H5),
122.5 (CN), 110.2 (C@CH2), 62.3 (5-C), 59.4
(CH(C6H5)CH3), 53.9 (7-CH2), 50.0 (8-CH), 39.7 (1-CH),
39.3 (4-CH2), 37.9 (3-CH), 36.9 (2-CH2), 23.9
(CH(C6H5)CH3), 20.6 (CH3C@CH2), 16.4 (8-CHCH3).
yl)maleimide were added to
a solution of 122 mg
(0.61 mmol) of 2-(bromomethyl)benzaldehyde 9 in 10 ml
of dry acetonitrile. The reaction mixture was allowed to
stand in an argon atmosphere for 72 h. The solvent was re-
moved by evaporation, and the residue was chromato-
graphed with hexane–ethyl acetate (1:1) as an eluent to
give 105 mg of compound 27 (0.25 mmol, 40%).
1
Mp = 87–88 °C. [a]D = +88.5 (c 0.87, MeOH). H NMR
(CD3OD), d: 7.24 (m, 9H), 6.70 (d, 2H, J = 8.0 Hz,
C6H4OCH3), 6.22 (d, 2H, J = 8.5 Hz, C6H4OCH3), 4.74
(br m, 1H, 4-CH or 9-CH), 4.40 (br m, 1H, 9-CH or
4-CH), 3.85 (dd, J = 10.4 and 5.2 Hz, 1H, 3a-CH or 9a-
CH), 3.77 (br m, 1H, 9a-CH or 3a-CH), 3.63 (s, 3H,
OCH3), 3.16 (br m, 1H, CHCH3), 1.32 (d, J = 6.5 Hz,
CHCH3) Signals at 1H NMR spectrum are broadened
and in some cases doubled probably due to slow inversion
at N-10.14 They are sharpened upon heating the sample to
50 °C. 13C NMR (CD3OD), d: 176.0 (C@O), 175.9 (C@O),
159.6 (40-C6H4OCH3), 143.9 (ipso-C6H5), 141.0, 140.9,
128.4, 127.8 (C6H4OCH3), 127.52, 127.48, 127.3, 126.8,
124.0, 123.0 (br s), 122.9 (br s), 113.6 (10-C6H4OMe),
66.1 (C-4 or C-9), 65.6 (C-9 or C-4), 56.4 (CHCH3), 54.5
(OCH3), 46.89 and 46.88 (C-3a and C-9a), 21.3 (CHCH3).
To assign the signals in NMR spectra, 1H–1H COSY,
HSQC and HMBC experiments were carried out. IR
(KBr): 1774 (mas C@O), 1713 (ms C@O). MS (m/z): 425
(MH+), 222, 118, 105. Anal. Calcd for C27H24N2O3: C,
76.40; H, 5.70; N, 6.60. Found: C, 76.02; H, 5.98; N, 6.24.
1
To assign the signals in the NMR spectra, H–1H COSY,
HSQC, HMBC and NOE experiments were carried out.
IR: 2234 (m C„N). MS (m/z): 295 (MH+), 191. Anal.
Calcd for C20H26N2: C, 81.59; H, 8.90; N, 9.51. Found:
C, 81.94; H, 8.62; N, 9.20. The results of combustion ele-
mental analysis were obtained for the mixture of diastereo-
mers 22a and 22b.
4.11. 1-((1R)-1-Phenylethyl)pyrrolidine-2-carbonitrile 23
This was obtained from 4-chlorobutyraldehyde 10 in 66%
yield as 1:0.7 mixture of (2R)- and (2S)-diastereomers.§ 1H
NMR (CDCl3), d: 7.2–7.4 (m, 5H, C6H5 of 23a and
m, 5H, C6H5 of 23b), 4.15 (d, J = 6.5 Hz, 1H, 2-CH of
23b), 3.65 (q, J = 6.7 Hz, 1H, CHCH3 of 23a), 3.59 (q,
J = 6.3 Hz, 1H, CHCH3 of 23b), 3.48 (d, J = 6.5 Hz, 1H,
2-CH), 3.27 (m, 1H, 5-CH2 of 23a), 2.69 (td, J = 9.5 and
3.8 Hz, 1H, 5-CH2 of 23b), 2.55 (q, J = 8.3 Hz, 1H, 5-
CH2 of 23a), 2.30 (q, J = 9.0 Hz, 5-CH2 of 23b), 2.22 (m,
1H, 3-CH2 of 23b), 2.16 (m, 1H, 3-CH2 of 23b), 2.00 (m,
4H, 3-CH2 and 4-CH2 of 23a), 1.88 (m, 1H, 4-CH2 of
23b), 1.80 (m, 1H, 4-CH2 of 23b), 1.44 (d, J = 7.0 Hz,
3H, CHCH3 of 23a), 1.43 (d, J = 7.5 Hz, 3H, CHCH3 of
23b). 13C NMR (CDCl3), d: 144.5 (ipso-C6H5 of 23b),
Acknowledgements
The authors are thankful to Dr. Zoya V. Voitenko for
helpful discussion and to Olga V. Manoylenko for HPLC
separations.
References
1. (a) Gauthier, G.; Royer, J.; Husson, H.-P. Eur. J. Org. Chem.
2002, 1484–1489; (b) Oba, M.; Koguchi, S.; Nishiyama, K.

O. O. Grygorenko et al. / Tetrahedron: Asymmetry 18 (2007) 290–297
297
Tetrahedron 2002, 58, 9359–9363; (c) Fleming, F. F.; Funk,
L.; Altundas, R.; Tu, Y. J. Org. Chem. 2001, 66, 6502–6504.
2. (a) Huang, P.-Q.; Huang, H.-Y. Synth. Commun. 2004, 34,
1377–1382; (b) Sunilkumar, G.; Nagamani, D.; Argade, N.
P.; Ganesh, K. N. Synthesis 2003, 2304–2306; (c) Gorichko,
M. V.; Grygorenko, O. O.; Komarov, I. V. Tetrahedron Lett.
2002, 43, 9411–9412.
3. Grygorenko, O. O.; Artamonov, O. S.; Palamarchuk, G. V.;
Zubatyuk, R. I.; Shishkin, O. V.; Komarov, I. V. Tetra-
hedron: Asymmetry 2006, 17, 252–258.
8. Armarego, W. L. F.; Chai, C. L. L. Purification of Laboratory
Chemicals; Elsevier: Oxford, 2003; pp 1–609.
9. Zhang, X.-X.; Lippard, S. J. J. Org. Chem. 2000, 65, 5298–
5305.
10. Witiak, D. T.; Tomita, K.; Patch, R. J.; Enna, S. J. J. Med.
Chem. 1981, 24, 788–794.
11. Doucet-Personeni, C.; Bentley, P. D.; Fletcher, R. J.;
Kinkaid, A.; Kryger, G.; Pirard, B.; Taylor, A.; Taylor,
R.; Taylor, J.; Viner, R.; Silman, I.; Sussman, J. L.;
Greenblatt, H. M.; Lewis, T. J. Med. Chem. 2001, 44,
3203–3215.
4. Oka, M.; Matsumoto, Y.; Unno, R. Heterocycles 1997, 45,
1447–1450.
5. di Beelo, N.; Pellacani, L.; Tardella, P. A. Synthesis 1978,
227–228.
12. Grewe, R.; Lorenzen, W.; Vining, L. Chem. Ber. 1954, 87,
793–802.
13. de Faria, M. L.; de Magalhaes, R. A.; Silva, F. C.; Matias,
˜
6. Grigg, R.; Zhang, L.; Collard, S.; Keep, A. Tetrahedron Lett.
2003, 44, 6979–6982.
L. G. de O.; Ceschi, M. A.; Brocksom, U.; Brocksom, T. J.
Tetrahedron: Asymmetry 2000, 11, 4093–4103.
7. Rammeloo, T.; Stevens, C. V.; de Kimpe, N. J. Org. Chem.
2002, 67, 6509–6513.
14. Kovtunenko, V. A.; Voitenko, Z. V. Russ. Chem. Rev. 1994,
63, 997–1018, and references cited therein.