Diastereoselective Synthesis of Cycloalkylamines by Samarium Diiodide-Promoted Cyclizations of α-Amino Radicals Derived from α-Benzotriazolylalkenylamines

Article abstract of DOI:10.1021/jo960957d

The condensation of ω-unsaturated aldehydes with benzotriazole and secondary amines affords α-benzotriazolylalkenylamines that exist in solution as mixtures of the corresponding benzotriazol-1-yl and benzotriazol-2-yl isomers resulting from their rapid dissociation into iminium cations and the benzotriazolyl anion. The reduction of these adducts with samarium diiodide (SmI₂) takes place with formation of the benzotriazolyl anion and α-amino alkenyl radicals that undergo 5- or 6-exo-trig cyclizations leading to substituted cycloalkyl- or cycloheteroalkylamines. The presence of an electron-withdrawing substituent in the alkene subunit is required for efficient cyclizations. The formation of cyclopentylamines takes place with unusually high 1,5-cis selectivity (hex-5-enyl radical numbering), and the presence of a 2- or 4-Me substituent also imparts high 1,2- or 1,4-trans stereoinduction, respectively. The corresponding six-membered rings, however, are formed with low diastereoselectivity. Semiempirical calculations performed on model systems suggest that a stabilizing secondary orbital interaction between the amino group and the electron-deficient alkene might in part account for the enhanced cis-selectivity encountered.

Full text of DOI:10.1021/jo960957d

J . Org. Chem. 1997, 62, 1125-1135
1125
Dia ster eoselective Syn th esis of Cycloa lk yla m in es by Sa m a r iu m
Diiod id e-P r om oted Cycliza tion s of r-Am in o Ra d ica ls Der ived fr om
r-Ben zotr ia zolyla lk en yla m in es
J ose´ M. Aurrecoechea,*,† Beatriz Lo´pez,† Alvaro Ferna´ndez,† Ana Arrieta,‡ and
Fernando P. Coss´ıo‡
Departamento de Quı´mica Orga´nica, Facultad de Ciencias, Universidad del Pa´ıs Vasco, Apartado 644,
48080 Bilbao, Spain and Departamento de Qu´ımica Orga´nica, Facultad de Qu´ımica, Universidad del
Paı´s Vasco, Apartado 1072, 20080 San Sebastia´n, Spain
Received May 24, 1996X
The condensation of ω-unsaturated aldehydes with benzotriazole and secondary amines affords
R-benzotriazolylalkenylamines that exist in solution as mixtures of the corresponding benzotriazol-
1-yl and benzotriazol-2-yl isomers resulting from their rapid dissociation into iminium cations and
the benzotriazolyl anion. The reduction of these adducts with samarium diiodide (SmI2) takes
place with formation of the benzotriazolyl anion and R-amino alkenyl radicals that undergo 5- or
6-exo-trig cyclizations leading to substituted cycloalkyl- or cycloheteroalkylamines. The presence
of an electron-withdrawing substituent in the alkene subunit is required for efficient cyclizations.
The formation of cyclopentylamines takes place with unusually high 1,5-cis selectivity (hex-5-enyl
radical numbering), and the presence of a 2- or 4-Me substituent also imparts high 1,2- or 1,4-
trans stereoinduction, respectively. The corresponding six-membered rings, however, are formed
with low diastereoselectivity. Semiempirical calculations performed on model systems suggest that
a stabilizing secondary orbital interaction between the amino group and the electron-deficient alkene
might in part account for the enhanced cis-selectivity encountered.
In tr od u ction
acceptor in the addition of a radical generated on the
other functional group, as shown by Fallis6b for the
coupling between alkyl halides and hydrazones. This
behavior therefore parallels the well known ability of
imines,8 oxime ethers,9 and hydrazones6,8b,10 to trap
intramolecularly carbon radicals generated by the tin
method.11
Samarium diiodide (SmI2) has emerged in the synthetic
field as a versatile one-electron reducing agent capable
of promoting a number of useful C-C bond-forming
processes with high chemo-, regio-, and stereoselectivity.1
SmI2-promoted couplings of carbonyl groups with alkyl-,
acyl-, and aryl-halides, or with allylic-, propargylic-, and
R-benzoyloxy esters, as well as pinacol-type couplings are
all well established procedures that have led to useful
synthetic applications.1 The chemistry of the aza-
analogues of the carbonyl group is less developed. In
contrast to the high reactivity of aldehydes and ketones
toward SmI2, aliphatic imines are reduced with great
difficulty.2-4 The imine-olefin coupling, unknown with
SmI2, has only recently been reported5 with Sm/Me3SiX
acting as a Sm(II) equivalent.
Iminium ions display a much higher reactivity toward
SmI2, as the electron transfer to the cation produces a
stabilized R-amino radical.12 When this process is carried
out in the presence of CSA, the R-amino radicals can be
trapped intramolecularly by suitably positioned C-C
π-bonds to afford cyclic products.13,14 From a synthetic
perspective, the tactical combination of this kind of
reactivity with a convenient method for the convergent
In contrast to the behavior shown by imines, other aza-
derivatives of carbonyl compounds such as hydrazones6
and oxime ethers7 participate in efficient Sm(II)-promot-
ed intramolecular couplings with aldehydes and, in the
hydrazone case, also with alkyl halides.6 However, the
nitrogen functionality seems to act in these cases as the
(8) (a) Bowman, W. R.; Stephenson, P. T.; Young, A. R. Tetrahedron
Lett. 1995, 36, 5623. (b) Bowman, W. R.; Stephenson, P. T.; Terrett,
N. K.; Young, A. R. Tetrahedron 1995, 51, 7959.
(9) (a) Booth, S. E.; J enkins, P. R.; Swain, C. J .; Sweeney, J . B. J .
Chem. Soc., Perkin Trans. 1 1994, 3499. (b) Keck, G. E.; McHardy, S.
F.; Murry, J . A. J . Am. Chem. Soc. 1995, 117, 7289. (c) Marco-Contelles,
J .; Destabel, C.; Gallego, P.; Chiara, J . L.; Bernabe, M. J . Org. Chem.
1996, 61, 1354, and references cited therein.
(10) Kim, S.; Cheong, J . H.; Yoon, K. S. Tetrahedron Lett. 1995, 36,
6069.
(11) Iminium cations are, on the other hand, reduced to the
corresponding amines by the hydride reagent and are of no use in this
methodology. See ref 8b.
(12) Review on R-amino radicals: Renaud, P.; Giraud, L. Synthesis
1996, 913.
† Facultad de Ciencias.
‡ Facultad de Qu´ımica.
X Abstract published in Advance ACS Abstracts, J anuary 1, 1997.
(1) Recent reviews: (a) Molander, G. A. Chem. Rev. 1992, 92, 29.
(b) Molander, G. A. In Organic Reactions; Paquette, L. A., Ed.; J ohn
Wiley & Sons Inc: New York, 1994; Vol. 46, pp 211-367. (c) Imamoto,
T. Lanthanides in Organic Synthesis; Academic Press: London, 1994.
(d) Molander, G. A.; Harris, C. R. Chem. Rev. 1996, 96, 307.
(2) Ferna´ndez-Acebes, A. Doctoral Thesis, Universidad del Pa´ıs
Vasco, 1995.
(3) (a) Imamoto, T.; Nishimura, S. Chem. Lett. 1990, 1141. (b)
Enholm, E. J .; Forbes, D. C.; Holub, D. P. Synth. Commun. 1990, 20,
981.
(4) Souppe, J .; Danon, L.; Namy, J . L.; Kagan, H. B. J . Organomet.
Chem. 1983, 250, 227.
(5) Akane, N.; Hatano, T.; Kusui, H.; Nishiyama, Y.; Ishii, Y. J . Org.
Chem. 1994, 59, 7902.
(6) (a) Sturino, C. F.; Fallis, A. G. J . Am. Chem. Soc. 1994, 116,
7447. (b) Sturino, C. F.; Fallis, A. G. J . Org. Chem. 1994, 59, 6514.
(7) Chiara, J . L.; Marco-Contelles, J .; Khiar, N.; Gallego, P.; Desta-
bel, C.; Bernabe´, M. J . Org. Chem. 1995, 60, 6010.
(13) Martin, S. F.; Yang, C. P.; Laswell, W. L.; Ru¨eger, H. Tetrahe-
dron Lett. 1988, 29, 6685.
(14) Alternative methods for the generation and synthetic use of
R-amino radicals: Radical translocation, (a) Murakami, M.; Hayashi,
M.; Ito, Y. J . Org. Chem. 1992, 57, 793. (b) Williams, L.; Booth, S. E.;
Undheim, K. Tetrahedron 1994, 50, 13697. (c) Booth, S. E.; Benneche,
T.; Undheim, K. Tetrahedron 1995, 51, 3665. Decarbonylation of
R-amino acid chlorides, (d) Collin, J .; Namy, J . L.; J ones, G.; Kagan,
H. B. Tetrahedron Lett. 1992, 33, 2973. Photoinduced electron transfer
on amines, (e) Santamaria, J . Pure Appl. Chem. 1995, 67, 141. (f) Das,
S.; Kumar, J . S. D.; Shivaramayya, K.; George, M. V. J . Chem. Soc.,
Perkin Trans. 1 1995, 1797. (g) Khim, S. K.; Mariano, P. S. Tetrahedron
Lett. 1994, 35, 999. (h) Hoegy, S. E.; Mariano, P. S. Tetrahedron Lett.
1994, 35, 8319. (i) Yoon, U. C.; Mariano, P. S. Acc. Chem. Res. 1992,
25, 233.
S0022-3263(96)00957-7 CCC: $14.00 © 1997 American Chemical Society

1126 J . Org. Chem., Vol. 62, No. 4, 1997
Aurrecoechea et al.
Sch em e 1
Ch a r t 1
radical 3. If suitably located reactive π-bonds are present,
an intramolecular radical addition takes place with
formation of cycloalkylamines 4 (path a)19b or pyrrolidines
6 (path c).19a In the absence of such functionality radical
dimerization leads to vicinal diamines 5 (path b).19c
Following an earlier communication19b we now disclose
full details of the preparation of cyclic products 4 starting
from simple aldehydes and amines using the strategy
depicted in Scheme 1. We have also studied the depen-
dence of the stereochemistry of cyclization on the double
bond geometry and the effect of Me substituents on
different locations around the carbon chain undergoing
cyclization. Finally, the extension of this chemistry to
the synthesis of six-membered cyclic systems has been
briefly explored.
preparation of iminium cations from simple aldehyde and
amine precursors15 would constitute a very practical
entry into both nitrogen-containing heterocycles and
amino-substituted carbocycles,16 as it is generally de-
scribed in the lower portion of Scheme 1.
The condensation of aldehydes with secondary amines
in the presence of benzotriazole provides a very conve-
nient entry into reactive iminium cations 2 as a result
of the easy room temperature dissociation and intercon-
version of the corresponding adducts 1,1′ in solution
(Scheme 1).17 This property has led to numerous syn-
thetic applications where the benzotriazole anion is
displaced from 1,1′ by a variety of nucleophiles under
mild reaction conditions.18 When these adducts are
treated in solution with SmI2, a rapid reduction of the
iminium cation takes place to afford a neutral R-amino
Resu lts a n d Discu ssion
Representative aldehydes 8-10 (Chart 1) were selected
for this study. The choice of enals bearing electron-
withdrawing substituents on the alkene portion fol-
lowed literature precedence on related radical cycliza-
tions14g-i,19a,20 and some initial experiments performed
with aldehyde 7 lacking such groups (vide infra).
The route for aldehydes 8,9 is shown in Scheme 2. The
E/Z mixtures obtained in the Horner-Wadsworth-
Emmons reaction21 were either separated and each
isomer oxidized independently (8a ,b,e,9b) or carried to
the aldehyde stage and separated there (8c,d ). A slightly
modified route was followed for aldehyde 9a . Alkylation
of 2-hydroxybenzaldehyde with cinnamyl bromide or
methyl 4-bromocrotonate led uneventfully to aldehydes
10.
(15) General references for the preparation and uses of iminium ca-
tions: (a) Paukstelis, J . V.; Cook, A. G. In Enamines. Synthesis,
Structure and Reactions, 2nd ed.; Cook, A. G., Ed.; Marcel Dekker:
New York, 1988; pp 275-346. (b) Kleinman, E. F. In Comprehensive
Organic Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon Press:
New York, 1991; Vol. 2, pp 893-951. (c) Overman, L. E.; Ricca, D. J .
In Comprehensive Organic Synthesis; Trost, B. M., Fleming, I., Eds.;
Pergamon Press: New York, 1991; Vol. 2, pp 1007-1046. (d) Hiemstra,
H.; Speckamp, W. N. In Comprehensive Organic Synthesis; Trost, B.
M., Fleming, I., Eds.; Pergamon Press: New York, 1991; Vol. 2, pp
1047-1082. See also: (e) ref 14e. (f) Pandey, G. Synlett 1992, 546. (g)
Risch, N.; Arend, M. Angew. Chem., Int. Ed. Engl. 1994, 33, 2422, and
references therein.
P r ep a r a tion of Ben zotr ia zole In ter m ed ia tes. The
condensation of aldehydes 7-10 with a slight excess (5-
10%) each of benzotriazole and a secondary aliphatic
amine at room temperature, in the presence of 4 Å
(16) Some recent references on cycloalkylamine natural products:
(a) J enkins, G. N.; Turner, N. J . Chem. Soc. Rev. 1995, 24, 169. (b)
Kobayashi, Y.; Shiozaki, M. J . Org. Chem. 1995, 60, 2570. (c) Reference
7. (d) Isono, N.; Mori, M. J . Org. Chem. 1995, 60, 115. (e) Kiguchi, T.;
Tajiri, K.; Ninomiya, I.; Naito, T.; Hiramatsu, H. Tetrahedron Lett.
1995, 36, 253. (f) Trost, B. M.; Vanvranken, D. L. J . Am. Chem. Soc.
1993, 115, 444.
(17) Katritzky, A. R.; Yannakapoulou, K.; Kuzmierkiewicz, W.;
Aurrecoechea, J . M.; Palenik, G. J .; Koziol, A. E.; Szczesniak, M.;
Skarjune, R. J . Chem. Soc., Perkin Trans. 1 1987, 2673.
(19) (a) Aurrecoechea, J . M.; Ferna´ndez-Acebes, A. Synlett 1996, 39.
(b) Aurrecoechea, J . M.; Ferna´ndez-Acebes, A. Tetrahedron Lett. 1993,
34, 549. (c) Aurrecoechea, J . M.; Ferna´ndez-Acebes, A. Tetrahedron
Lett. 1992, 33, 4763.
(20) (a) Choi, J . K.; Ha, D. C.; Hart, D. J .; Lee, C. S.; Ramesh, S.;
Wu, S. J . Org. Chem. 1989, 54, 279. (b) Kano, S.; Yuasa, Y.; Asami,
K.; Shibuya, S. Chem. Lett. 1986, 735. (c) Padwa, A.; Dent, W.;
Nimmesgern, H.; Venkatramanan, M. K.; Wong, G. S. K. Chem. Ber.
1986, 119, 813. (d) Padwa, A.; Nimmesgern, H.; Wong, G. S. K. J . Org.
Chem. 1985, 50, 5620.
(18) Reviews: (a) Katritzky, A. R.; Lan, X. F.; Fan, W. Q. Synthesis
1994, 445. (b) Katritzky, A. R.; Rachwal, S.; Hitchings, G. J . Tetrahe-
dron 1991, 47, 2683.
(21) (a) Takacs, J . M.; Helle, M. A.; Seely, F. L. Tetrahedron Lett.
1986, 27, 1257. (b) Takacs, J . M.; Myoung, Y. C.; Anderson, L. G. J .
Org. Chem. 1994, 59, 6928.

Diastereoselective Synthesis of Cycloalkylamines
J . Org. Chem., Vol. 62, No. 4, 1997 1127
Sch em e 2
Sch em e 3
Sch em e 4
molecular sieves,22 led to the corresponding R-benzotria-
zolylalkenylamines 15-21 (Schemes 3, 5, 6, 8). Adducts
15-21 were usually obtained as thick oils that were
difficult to purify23 and, for preparative purposes, it was
decided to use them in a crude form. As a consequence,
the cyclization yields reported below are always given for
two steps and are referred to the starting aldehyde.
1
Nevertheless, routine inspection of the H NMR spectra
of the adducts always confirmed their formation and
allowed an assessment of their degree of purity to be
made. Thus, two sets of resonances were observed in the
region δ 5.1-5.9, corresponding to the methine NCHN
proton indicating that, as expected, adducts 15-21 exist
in solution as mixtures of 1- and 2-substituted benzotri-
azoles (e.g., 1,1′ in Scheme 1).24 Also characteristic of
these adducts are the signals due to the benzotriazole
ring protons. Thus, a doublet at δ 8.0-8.1 is typical of
H-4 (benzotriazole numbering) in the benzotriazol-1-yl
isomer whereas a multiplet at δ 7.8-8.0 was assigned
to H-4 and H-7 of the corresponding 2-substituted
isomer.17 Products other than 15-21 and excess benzo-
triazole and starting amine were either absent or in
minor amounts. Furthermore, mass recovery was always
in the range 93-100%, after adjusting for the excess
cyclized products 22 and 23 were isolated in a ratio of
1.2/1 and a low combined 27% yield. The presence of
other uncharacterized products (probably of dimeric
structure19c) complicated the separation and purification
process.26 Better results were obtained with aldehydes
that contained electron-withdrawing substituents, as
described below.
The formation of 22 and 23 is readily explained as a
result of a 5-exo-trig cyclization of radical 25 (Scheme 4)
followed by the disproportionation of the resulting cy-
clized radical 26. This explanation is also consistent with
the consumption of less than 2 equiv of SmI2 in this
reaction. Alternatively, a tertiary organosamarium in-
termediate derived from 26 could also afford the observed
products via â-hydride elimination and organometallic
disproportionation pathways;27 however, the reduction of
a tertiary alkyl radical by SmI2 appears to be a difficult
process,28 particularly in the absence of strong donor
ligands. In any case, while these results clearly establish
the radical nature of the cyclizations, they also evidence
that unactivated olefins such as that present in 15 will
not be suitable partners for a fast radical cyclization of
the R-amino radicals involved here. Similar conclusions
have been reached after attempted cyclizations20c,d of
2-azahex-5-enyl radicals generated from R-amino sulfides
by the tin method. A likely cause for these seemingly
slow cyclizations is the interaction between the nitrogen
1
starting materials present estimated by H NMR inte-
1
gration. Preparation details and full H NMR data for
15-21 are collected in Tables 4 and 5 (supporting
information).25
Initial cyclization studies were conducted on adduct 15
(Scheme 3) derived from commercial aldehyde 7. When
a 0.1 M THF solution of SmI2 was added dropwise to a
solution of 15 at 25 °C the reagent was rapidly consumed,
as indicated by the disappearance of the SmI2 charac-
teristic blue color. After the addition of approximately
1.5 equiv the blue color persisted indicating that the
starting material had been completely consumed. How-
ever, a complex reaction mixture resulted from which
(22) (a) Katritzky, A. R.; Yannakopoulou, K. Synthesis 1989, 747.
(b) Katritzky, A. R.; Long, Q. H.; Lue, P.; J ozwiak, A. Tetrahedron 1990,
46, 8153.
(23) Simple R-benzotriazolylalkylamines decompose in contact with
water and traces of acid or silica gel chromatography, reverting to the
starting aldehyde and amine: Katritzky, A. R.; Yannakopoulou, K.;
Lue, P.; Rasala, D.; Urogdi, L. J . Chem. Soc., Perkin Trans. 1 1989,
225.
(26) Cyclopentanes 22 and 23 could not be completely separated
from each other nor could they be conveniently purified. However, their
NMR (1H and 13C) and GC-MS data clearly support the proposed
structures.
(27) Molander, G. A.; Mckie, J . A. J . Org. Chem. 1993, 58, 7216.
(28) (a) Curran, D. P.; Fevig, T. L.; J asperse, C. P.; Totleben, M. J .
Synlett 1992, 943. (b) Curran, D. P.; Totleben, M. J . J . Am. Chem.
Soc. 1992, 114, 6050.
(24) Adducts 15-20 derived from aliphatic aldehydes 7-9 exist as
1/1 mixtures of the corresponding benzotriazol-1- and -2-yl isomers
whereas in adducts 21, derived from aromatic aldehydes 10, the
1-isomer markedly predominates.
(25) Further characterization by 13C NMR and MS for selected cases
is also included in the supporting information.

1128 J . Org. Chem., Vol. 62, No. 4, 1997
Aurrecoechea et al.
Ta ble 1. P r ep a r a tion of Disu bstitu ted Cyclop en ta n es 27 fr om Ald eh yd es 8a -d , Ben zotr ia zole, a n d Am in es
entry
aldehyde
R1
R2
(CH2)2O(CH2)2
(CH2)2O(CH2)2
(CH2)2O(CH2)2
methoda
product
yieldb
ratio c/t
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
8a
8a
8b
8a
8a
8a
8a
8a
A
B
B
A
B
B
B
B
B
B
B
A
A
B
B
A
B
B
27a
27a
27a
27b
27b
d
27d
27e
27f
27g
27h
27i
27j
27j
27j
27k
27k
27k
50
75
39
39
69
-
63
32
57
61
53
63
60
67
65
65
71
62
c
c
1:1.8
4.3:1
16:1
-
c
c
c
c
c
2:1
2:1
10:1
6.7:1
2.3:1
8.1:1
4.5:1
CH2Ph
CH2Ph
CH2Ph
CH2Ph
CH2Ph
CH2Ph
(CH2)2N(Ph)(CH2)2
CH2Ph
CH2Ph
CH2Ph
CH2Ph
CH2CO2Et
(CH2)2Ph
Ph
8a
8a
8a
(CH2)2CN
3-butenyl
CH2Ph
(CH2)2O(CH2)2
(CH2)2O(CH2)2
(CH2)2O(CH2)2
allyl
8c, 8d e
8c, 8d e
8c
8d
8c, 8d e
8c
allyl
allyl
allyl
allyl
allyl
8d
a
Method A: SmI2 added dropwise to benzotriazole adduct at 25 °C. Method B: The adduct added to SmI2 at T < -20 °C and the
reaction mixture warmed to 25 °C. Isolated yields are referred to the starting aldehyde (two steps). c A single diastereomer to the limit
b
of detection of 1H and 13C NMR. Complex mixture of products. See text. e A 65/35 8c/8d mixture was used.
d
Sch em e 5
mixture to slowly reach room temperature (method B).
It is likely that the low temperatures ensure a low
equilibrium concentration of iminium cations30 while
maintaining a relative high concentration of SmI2. This
is expected to comparatively slow down undesired inter-
molecular side-reactions and accelerate the reduction of
the cyclized radical, thus providing a forward thrust for
the overall process. Results using this latter method are
collected in Table 1 along with selected examples run
with method A for comparison. The two-step sequence
works well with aliphatic amines (both cyclic and acyclic)
but less efficiently so with the single aniline derivative
that was tried (entry 8). Nevertheless, the sequence is
tolerant of some useful functionality (entries 9, 11, 16-
18) in the amino group. Of particular value is the
preparation of mono- and dibenzylated cyclopentylamines
since the benzyl group can in principle be easily
removed14a,31 to yield the corresponding secondary and
primary amines, respectively. Although less utilized, the
allyl group (entries 16-18) can also serve this purpose.32
As seen in Table 1, both (E)-CO2Et- and (E)-CN-groups
are effective alkene-activators toward radical cyclization.
However, in the (Z)-series the presence of a CO2Et group
had a great diminishing impact in the cyclization yield
(entry 3) whereas this was largely unaffected by a CN
group (entries 15, 18).
lone pair and the unpaired electron. This confers the
radical both a strong stabilization, which is partially lost
in the transition state leading to cyclization, and a
nucleophilic character which is not matched by the
polarity of simple alkyl-substituted olefins. In line with
these ideas, both the protonation13 and the incorporation
of electron-withdrawing substituents29 on the nitrogen
atom have resulted in efficient cyclizations since the
aforementioned effects are subdued or disappear. Simi-
larly, the use of electron-deficient olefins20a has proven
useful in the reactions of R-acylamino radicals, as the
cyclized radical is also stabilized and the polarity of both
reaction partners is better matched. Our results using
neutral R-amino radicals and electron-deficient olefins
(vide infra) support these ideas.
Syn th esis of Disu bstitu ted Cyclop en ta n es. The
treatment of benzotriazole adducts 16, 17 (derived from
aldehydes 8a -d containing CO2Et- or CN-substituted
olefins) with SmI2, under the conditions described above
for 15, produced much cleaner reaction mixtures and
resulted in the highly diastereoselective formation of the
corresponding cyclopentanes 27 that were isolated in
moderate overall yields over two steps starting from
aldehydes 8 (Scheme 5, Table 1).19b Significantly, these
reactions require 2 equiv of SmI2, indicating that cycliza-
tion is followed by fast further reduction to afford a
samarium enolate (or a R-cyano organosamarium). While
this dropwise addition of SmI2 to the substrate (method
A) conveniently afforded the desired products (as reported
in our preliminary communication19b), better yields and
diastereoselectivities were realized by mixing a dilute (0.1
M) solution of the adducts 16, 17 with 0.1 M SmI2 (2.2
equiv) at temperatures below -20 °C and allowing the
Further synthetic advantage could in principle be
gained from the functionalized radical initially obtained
in the cyclization step or, alternatively, from the sa-
marium enolate resulting from its reduction by a second
equivalent of SmI2. However, no tandem cyclization33
product was observed in the reactions involving allyl-
amine derivatives (entries 16-18), indicating that further
reduction of the radical formed in the cyclization took
place faster than its alternative cyclization onto the
(30) The presence of Sm(III), a Lewis acid, might facilitate an
expectedly slow dissociation at temperatures below 25 °C. See: Katritz-
ky, A. R.; Gordeev, M. F. J . Org. Chem. 1993, 58, 4049.
(31) (a) Kocienski, P. J . Protecting Groups; Thieme: Stuttgart, 1994;
p 220. (b) Hormuth, S.; Reissig, H. U. J . Org. Chem. 1994, 59, 67.
(32) (a) Lemaire-Audoire, S.; Savignac, M.; Genet, J . P.; Bernard,
J . M. Tetrahedron Lett. 1995, 36, 1267. (b) Talukdar, S.; Banerji, A.
Synth. Commun. 1995, 25, 813. (c) Ito, H.; Taguchi, T.; Hanzawa, Y.
J . Org. Chem. 1993, 58, 774. (d) Garrohelion, F.; Merzouk, A.; Guibe,
F. J . Org. Chem. 1993, 58, 6109. (e) Laguzza, B. C.; Ganem, B.
Tetrahedron Lett. 1981, 22, 1483.
(29) (a) Udding, J . H.; Tuijp, C. J . M.; Hiemstra, H.; Speckamp, W.
N. Tetrahedron 1994, 50, 1907. (b) Esch, P. M.; Hiemstra, H.; de Boer,
R. F.; Speckamp, W. N. Tetrahedron 1992, 48, 4659.
(33) Bunce, R. A. Tetrahedron 1995, 51, 13103.

Diastereoselective Synthesis of Cycloalkylamines
J . Org. Chem., Vol. 62, No. 4, 1997 1129
Sch em e 6
Sch em e 7
readily observed at temperatures between -20 and 25
°C for most substrates, required prolonged stirring at
room temperature for the 4-methyl-substituted adducts
19 derived from aldehyde 9b; in the case of the diben-
zylamine adduct 19a this also led to some degradation.
For example, quenching the reaction of 19a after 20 min
at room temperature led to a 58% isolated yield of cyclic
products 29a ,a ′, along with ∼20% of recovered starting
aldehyde 9b, but continued stirring at room temperature
for 22 h increased the yield of 29a ,a ′ to only 60%. Of
some mechanistic significance is the isolation in the
reaction of 19a of small amounts of the bicyclic ketone
30a . This product originated as the result of a 1,5-
translocation36 of the cyclized radical 31 (Scheme 7) that
yields a stabilized benzylic R-amino radical 32. Reduc-
tion to the organosamarium14a-c intermediate 33 and
intramolecular attack on the ester carbonyl would then
lead to the observed product 30a . While the formation
of 30a confirms the radical nature of these reactions, the
small extent to which this occurs indicates that, under
the reaction conditions, reduction of the cyclized radical
to form a samarium enolate is faster than other alterna-
tive radical processes such as translocation or cyclization,
as discussed above.
Syn th esis of Six-Mem ber ed Rin gs. A brief explora-
tion into the formation of six-membered rings pointed out
a severe limitation in the use of this methodology. While
yields can still be reasonable in some cases, the diaste-
reoselectivity is significantly eroded (Scheme 8, Table 3).
Nevertheless, it is interesting that the impact of chain
elongation is less pronounced here than in the related
SmI2/CSA treatment on preformed iminium cations
where cyclization on unactivated alkenes is shut off in
going from 5-exo to the 6-exo cases.13 As noted else-
where,19c the presence of a Ph substituent in place of CO2-
Et in the adducts derived from 10a led only to the
corresponding vicinal diamine showing that the Ph group
does not provide in this case enough activation for a fast
radical cyclization and evidencing again the need for
electron-deficient olefins in these cyclizations.
Ta ble 2. P r ep a r a tion a of Tr isu bstitu ted Cyclop en ta n es
28, 29 fr om Ald eh yd es 9a ,b, Ben zotr ia zole, a n d Am in es
aldehyde
product
yieldb
diast ratio
9a
9a
9b
9b
28a
28b
29a ,a ′
29b,b′
66
57
19:1
c
6:1
4.5:1
60d (72e)
71
a
b
Prepared according to general method B. Isolated yields are
referred to the starting aldehyde (two steps). c Ratio (approxi-
mately 19:1 by 1H NMR) could not be precisely determined.
Combined yield of 29 and 30a (58:1). e Yield based on recovered
d
aldehyde 9b.
allyl group. Furthermore, attempts to trap34 the pre-
sumed samarium enolate precursor of the morpholine
derivative obtained in entry 2 were unsuccessful. Thus,
cyclopentylamine c-27a (Scheme 5, Table 1) was still the
only product obtained when benzaldehyde was added to
the reaction mixture once the total consumption of SmI2
had been observed. Either the samarium enolate is
rather unreactive or it gets protonated before benzalde-
hyde is introduced. The ethyl ester derivative derived
from 8a and N-benzylglycine (entry 6) provided a complex
mixture of products.
Syn th esis of Tr isu bstitu ted Cyclop en ta n es. The
introduction of a 2- or 4-Me35 substituent on the carbon
chain undergoing cyclization resulted in the efficient
formation of trisubstituted cyclopentanes 28, 29 (Scheme
6, Table 2). Overall yields over two steps are satisfactory,
and a high stereoselectivity is maintained. Out of four
possible diastereomers, one was always very predomi-
nantly obtained, amounting to ∼95% of the product for
cyclopentanes 28 and ∼85% for 29 (Table 2). In addition
to a somewhat decreased selectivity, the introduction of
the 4-Me substituent also caused a significant reduction
in reactivity. Thus, decoloration of the SmI2 reagent,
(34) Several examples of successful aldol-type processes with sa-
marium enolates formed in SmI2-mediated reactions have been
reported: (a) Curran, D. P.; Wolin, R. L. Synlett 1991, 317. (b) Enholm,
E. J .; J iang, S. Tetrahedron Lett. 1992, 33, 313. (c) Enholm, E. J .; J iang,
S. Heterocycles 1992, 34, 2247. (d) Enholm, E. J .; J iang, S. Tetrahedron
Lett. 1992, 33, 6069. (e) Enholm, E. J .; J iang, S. J .; Abboud, K. J . Org.
Chem. 1993, 58, 4061. (f) Enholm, E. J .; Trivellas, A. Tetrahedron Lett.
1994, 35, 1627. (g) Arime, T.; Takahashi, H.; Kobayashi, S.; Yamaguchi,
S.; Mori, N. Synth. Commun. 1995, 25, 389. (h) Enholm, E. J .; Schreier,
J . A. J . Org. Chem. 1995, 60, 1110.
Interestingly, when the reaction of the adduct 20a
(derived from 8e and morpholine) was carried out by
dropwise addition of the SmI2 solution to the substrate
(36) Examples of related radical translocations in amine deriva-
tives: (a) Denenmark, D.; Hoffmann, P.; Winkler, T.; Waldner, A.; De
Mesmaeker, A. Synlett 1991, 621. (b) Curran, D. P.; Liu, H. T. J . Chem.
Soc., Perkin Trans. 1 1994, 1377. (c) Sato, T.; Kugo, Y.; Nakaumi, E.;
Ishibashi, H.; Ikeda, M. J . Chem. Soc., Perkin Trans. 1 1995, 1801.
See also refs 14a-c.
(35) The hex-5-enyl radical numbering will be used throughout the
discussion.

1130 J . Org. Chem., Vol. 62, No. 4, 1997
Aurrecoechea et al.
Sch em e 8
Ster eoch em istr y of Cycliza tion s. For disubstituted
cycloalkanes the use of NOE experiments did not provide
conclusive evidence about the stereochemical identity of
the cyclic products. Recourse was then made to 13C NMR
chemical shift differences between the isomers as ster-
eochemical criteria.41 The major diastereomers c-27a,i-k
and c-34a (see Schemes 5, 8 and Table 1) displayed for
carbons C2, C1′, and C2′ resonances (Table 6, supporting
information) that were significantly upfield from the
corresponding signals in the minor isomers t-27 and
t-34a ′, thus allowing the assignment of a cis-stereochem-
istry for the former.41,42 The stereochemistry of the rest
of disubstituted products was assigned by analogy with
the cases included in Table 6.
In trisubstituted cyclopentanes 28, 29 the major dia-
stereomers featured again a 1,5-cis relationship between
the two newly created stereogenic centers whereas the
third substituent adopted a trans-disposition with respect
to the other two. This stereochemical assignment relied
on the observation of NOE effects as well as on 13C NMR
chemical shift differences between the isomers (Tables 6
and 7, supporting information). The tentative assign-
ment given in this way to the minor dibenzylamine
product 29a ′ (Scheme 6) was also supported by the
observation of proton resonances for C5′-Me and H-2′ that
were shielded with respect to those found in the major
isomer 29a .
As shown by the results obtained in the formation of
disubstituted cyclopentanes (Table 1) the diastereoselec-
tivity of the 5-exo-cyclization showed a dependence on
reaction conditions, activating group present in the
alkene, and double bond geometry. Thus, the comparison
between entries 4 and 5 in Table 1 seems to indicate that
diastereoselectivity increases with decreasing reaction
temperature. This is also apparent in entries 13-18
where the selectivities obtained for the individual geo-
metric isomers at low temperatures would translate into
higher than observed selectivities for reactions run at 25
°C with the mixtures of isomers. With a single exception
(entry 5), substrates containing an (E)-CO2Et group
cyclized, under the improved conditions, with overwhelm-
ing cis-selectivity as only one diastereoisomer was found.
The selectivity dropped somewhat when the (E)-CN group
was used in place of CO2Et (entries 14, 17). However,
even in these cases the selectivity is remarkably high
when compared with the 2.6/1 cis/trans preference re-
ported for the cyclization of the 1-methylhex-5-enyl
radical.43a In the (Z)-series, we have observed a mixed
behavior; thus, while for a (Z)-CO2Et substituent the cis-
trans selectivity was reversed and greatly diminished
(entry 3), a (Z)-CN group continued to impart high, albeit
slightly lower, cis-selectivity (entries 15, 18). The pos-
sible origin of these selectivities will be discussed later
(vide infra).
Ta ble 3. P r ep a r a tion a of Six-Mem ber ed Cyclic P r od u cts
34, 35 fr om Ald eh yd es 8e, 10b, Ben zotr ia zole, a n d Am in es
aldehyde
product
yield (%)b
diast ratio
8e
8e
10b
10b
10b
34a
34b
35a
35c
35d
54
34
54
53
60
2.3:1
1:1
1.5:1c
1:1
1:1
a
According to general procedure B. b Isolated yields are referred
to the starting aldehyde (two steps). c The stereochemical identity
of the products could not be unambiguously determined.
at 25 °C (method A), only 1.5 equiv of the reducing agent
were consumed, and the bicyclic product 36 (8/1 diaster-
eomeric mixture) was obtained along with the expected
cyclohexylamine 34a (5/1 cis/trans) in a 68% combined
yield and 1/1 34a /36 ratio. The bicyclic products 36 were
not detected when the reaction was performed under the
general optimized conditions (method B). A possible
origin of bicycles 36 is an intramolecular cycloaddition37
of an enamine formed by Sm(III)-promoted â-elimination
of benzotriazole from 20a .38 The precise mechanism of
formation of 36 might actually be more complex since the
reaction appears to consume more SmI2 than expected if
the formation of cyclohexylamine 34a were the only SmI2-
consuming process. Furthermore, the treatment of the
39
same adduct 20a with SmI3 at room temperature
afforded no bicyclic product. The starting material was
rapidly consumed, but a complex mixture of uncharac-
terized products was obtained. In any event, the rapid
destruction of the adduct 20a in the presence of Sm(III)
could in part be held responsible for the only modest
yields generally obtained in the cyclizations reported here
and hints that the advantage of method B over method
A might also reside in the higher Sm(II)/Sm(III) ratio
maintained in the former case throughout the reaction.40
The stereochemical preferences observed in these reac-
tions are in agreement with Beckwith’s model43 for the
transition structure of radical cyclizations. According to
this model the chain undergoing cyclization adopts a
chairlike conformation with the substituents occupying
(37) Cook, A. G. In Enamines. Synthesis, Structure and Reactions,
2nd ed.; Cook, A. G., Ed.; Marcel Dekker: New York, 1988; p 347.
(38) Although the formation of enamines from N-[(N′,N′-dialkylami-
no)alkyl]benzotriazoles has been effected only with NaH in refluxing
THF,22b certain Sm(III)-induced â-eliminations have been reported in
reductions with SmI2. See: Angle, S. R.; Rainier, J . D. J . Org. Chem.
1992, 57, 6883.
(40) See ref 38 for related observations.
(41) Whitesell, J . K.; Minton, M. A. Stereochemical Analysis of
Alycyclic Compounds by C-13 NMR Spectroscopy; Chapman and Hall:
London, 1987.
(42) Yadav, V.; Fallis, A. G. Can. J . Chem. 1991, 69, 779.
(43) (a) Beckwith, A. L. J .; Schiesser, C. H. Tetrahedron 1985, 41,
3925. (b) Spellmeyer, D. C.; Houk, K. M. J . Org. Chem. 1987, 52, 959.
(c) Beckwith, A. L. J .; Zimmermann, J . J . Org. Chem. 1991, 56, 5791.
(d) Beckwith, A. L. J . Chem. Soc. Rev. 1993, 22, 143.
(39) Yu, Y. P.; Lin, R. H.; Zhang, Y. M. Tetrahedron Lett. 1993, 34,
4547.

Diastereoselective Synthesis of Cycloalkylamines
J . Org. Chem., Vol. 62, No. 4, 1997 1131
Ch a r t 2
F igu r e 1. Contour map of the SOMO of the aminomethyl
radical, computed at UHF/6-31G* level. Contour values in au
range from +0.30 (solid lines) to -0.22 (dotted lines), with a
step of -0.025. The black arrow emphasizes the preferencial
reactive site of this radical in nucleophilic addition processes.
preferentially pseudoequatorial positions. Detailed stud-
ies on the stereoselectivity of cyclization have been car-
ried out for simple 1-, 2-, 3-, or 4-substituted hex-5-enyl
radicals, leading to the corresponding disubstituted cy-
clopentanes. However, the corresponding studies on 1,2-,
1,3-, and 1,4-disubstituted hex-5-enyl radicals are scarce.44
While the direction of the stereochemical preference
in our cyclizations is conveniently rationalized by Beck-
with’s model, the extent to which it occurs was not
expected. Unconstrained simple alkyl-substituted acyclic
radicals display relatively moderate selectivities43a,45
except for the 4-substituted cases, where allylic strain
may contribute decisively to the higher preferences
observed.46 In comparison, the cyclizations of the 1-ami-
nohex-5-enyl radicals reported here afford exclusively cis-
disubstituted cyclopentanes in almost every case when
the alkene moiety is activated by an (E)-CO2Et group.
Furthermore, a very high degree of 1,5-cis preference is
maintained with the introduction of a 2- or 4-Me sub-
stituent, while the 4,5-selectivity is comparable to that
obtained with the corresponding parent Me-substituted
hex-5-enyl radicals, and the 2,5-selectivity is considerably
higher than in this latter case. Interestingly, a recent
paper also reports on some unusually high stereoselec-
tivities in the cyclizations of related R-amidoyl radicals
that yield disubstituted cyclopentanes.36b
invoked to explain the preference of the 1-methylhex-5-
enyl radical to cyclize preferentially through the more
sterically demanding cis-transition state.43b,48 Likewise,
a strong donor-acceptor-type interaction between the
amino group and the electron-deficient alkene might be
responsible for the enhanced cis-selectivity in the cy-
clizations of 1-amino-substituted hex-5-enyl radicals.
To test this hypothesis we have carried out some SCF-
MO computations on the transition structures (TSs)
depicted in Chart 2, using the semiempirical Hamiltonian
PM3,49 at both UHF and half-electron50 (HE) levels of
theory. The heats of formation of the chair and boat
saddle points TSCa -j and TSBa -j obtained by both
UHF/PM3 and HE/PM3 methods are collected in Table
8 (supporting information). These data predict a pref-
erential formation of cis-aminocyclopentanes regardless
of the geometry at the double bond present in the starting
olefinic moiety, in agreement with the generality of our
experimental results (vide supra). However, this prefer-
ence is only modest when the double bond is unsubsti-
tuted but markedly increases with the presence of
electron-withdrawing group (EWG) substituents on the
double bond. The reasons underlying this stereocontrol
can be understood upon inspection of the orbitals involved
in the addition. Thus, the SOMO of the aminomethyl
Although the exact nature of this high selectivity
remains to be defined, the origin might reside on the
nucleophilic and electrophilic character of the R-amino
radical and olefin trap, respectively, in a presumably
highly polarized transition state.43d,47 Attractive interac-
tions between the Me group and the π-bond have been
radical (H2NCH2 ) is a distorted pyramidalized51 π*
•
orbital between the N and C atoms (Figure 1). A direct
consequence of this π-character, as our PM3 calculations
show, is the presence of a strong stabilizing in-phase
secondary orbital interaction52 between the SOMO of the
aminoalkyl radical and the LUMO of the alkene subunit
in the cis saddle points (Figure 2). This interaction is
favored by the eclipsed conformation between the N, C1,
C5, and CR atoms present in the transition states (TS)
leading to cis-products (ω ≈ 0, Figure 2) and does not
(44) (a) Bertrand, M. P.; De Riggi, I.; Lesueur, C.; Gastaldi, S.;
Nouguier, R.; J aime, C.; Virgili, A. J . Org. Chem. 1995, 60, 6040. (b)
Feldman, K. S.; Romanelli, A. L.; Ruckle, R. E.; J ean, G. J . Org. Chem.
1992, 57, 100.
(45) Beckwith, A. L. J .; Easton, C. J .; Lawrence, T.; Serelis, A. K.
Aust. J . Chem. 1983, 36, 545.
(46) RajanBabu, T. V. Acc. Chem. Res. 1991, 24, 139.
(47) The importance of a favorable polarization in the transition
state on the rates of radical cyclization and radical hydrogen abstrac-
tion processes has been stressed: (a) J ohnson, C. C.; Horner, J . H.;
Tronche, C.; Newcomb, M. J . Am. Chem. Soc. 1995, 117, 1684. (b)
Newcomb, M.; Horner, J . H.; Filipkowski, M. A.; Ha, C.; Park, S.-U. J .
Am. Chem. Soc. 1995, 117, 3674. (c) Horner, J . H.; Martinez, F. N.;
Musa, O. M.; Newcomb, M.; Shahin, H. E. J . Am. Chem. Soc. 1995,
117, 11124.
(48) Giese, B. Radicals in Organic Synthesis: Formation of Carbon-
Carbon Bonds; Pergamon: Oxford, 1986; p 149.
(49) Stewart, J . J . P. J . Comput. Chem. 1989, 10, 209.
(50) (a) Longuett-Higgins, H. C.; Pople, J . A. Proc. Phys. Soc. A68
1955, 591. (b) Dewar, M. J . S.; Trinajstic, N. J . Chem. Soc. A 1971,
1220.
(51) (a) Armstrong, D. A.; Rauk, A.; Yu, D. K. J . Am. Chem. Soc.
1993, 115, 666. (b) Schubert, S.; Renaud, P.; Carrupt, P. A.; Schenk,
K. Helv. Chim. Acta 1993, 76, 2473.

1132 J . Org. Chem., Vol. 62, No. 4, 1997
Aurrecoechea et al.
affords R-amino radicals that undergo 5- or 6-exo cycliza-
tions leading to substituted cycloalkyl- and cyclohet-
eroalkylamines. A simple and convenient synthetic
strategy combined with reasonable overall yields and
high diastereoselectivities assess to the effectiveness of
the overall synthetic scheme for the convergent synthesis
of polysubstituted cyclopentylamines. The formation of
the corresponding six-membered rings, however, will be
severely limited due to lack of diastereoselectivity.
Exp er im en ta l Section
Unless otherwise stated, reactions were conducted under
an atmosphere of dry Ar. THF and benzene were freshly
distilled from sodium/benzophenone. THF was deoxygenated
prior to use. Dichloromethane, DMSO, and DMF were dis-
tilled from CaH2. Organic extracts were dried over Na2SO4,
filtered, and evaporated under reduced pressure. Flash
column chromatography53 was performed on silica gel (230-
400 mesh). HPLC purifications were carried out with a
LiChrosorb Si60 (7 µm, 25 × 2.5 cm) column. Melting points
are uncorrected. 1H and 13C NMR spectra were obtained in
CDCl3 at 250 MHz and 62.9 MHz, respectively. Mass spectra
were obtained at 70 eV. GC-MS analysis were performed at
70-280 °C (20 °C/min) with a stationary phase of methylphen-
ylsilicone (0.25 µm, 30 m × 0.25 mm).
F igu r e 2. Schematic diagram showing the interaction be-
tween the SOMO of an aminomethyl radical and the LUMO
of an alkene. (A) Primary and secondary interactions in a TS
leading to cis-aminocyclopentanes. (B) Primary interaction in
a TS leading to trans-aminocyclopentanes.
Com p u ta tion a l Meth od s. All TS calculations were per-
formed using the semiempirical SCF-MO PM349 method. TSs
were optimized using Bartels54a and eigenvector following54b
methods. All stationary points were refined by minimization
of the gradient norm of the energy at least below 0.01 kcal/
Å-deg and characterized by harmonic vibrational frequency
analysis.55 The TSs showed only one negative eigenvalue in
their diagonalized Hessians, corresponding to motion along the
reaction coordinate. The relative preference found with this
semiempirical approach was qualitatively reproduced for
TSCa and TSCb using ab initio calculations56,57 at the UHF/
6-31G* level.
exist in those corresponding to the formation of trans-
cyclopentanes (ω ≈ -120, Figure 2). The lowering of
energy provided by the EWG substituents to the alkene
LUMO would be expected to further enhance the SOMO-
LUMO interactions thus leading to a predicted greater
cis-selectivity in these cases. It should be noted that the
secondary interaction, in principle, depends neither on
the geometry of the double bond nor on the conformation
of the five-membered ring being formed. Thus, for
sterically nondemanding substituents such as a cyano
group, the cis-product should be the major stereoisomer
formed from both (E)- and (Z)-geometries. The more
sterically demanding (and conformationally more mobile)
ethoxycarbonyl group might give rise, in the TS associ-
ated to a (Z)-CdC geometry, to steric congestion with the
amino group substituents (not included in the calcula-
tions). This could account for the lower selectivity
obtained in the cyclization of the morpholine adduct
derived from aldehyde 8b (Table 1, entry 3).
Finally, the higher stereoselectivity observed when the
olefin activating group is (E)-CO2Et, rather than (E)- or
(Z)-CN, might indicate that the rate of reduction of the
cyclized radical is also important for the stereochemical
outcome of the reaction. Thus, a very fast reduction
would lead to kinetic products in an essentially irrevers-
ible cyclization, and this is expected to be the case for
the CO2Et activating group where the oxophilicity of
trivalent samarium would facilitate the formation of the
corresponding samarium enolate. A slower reduction of
the CN-substituted cyclized radical might allow for some
reversibility in the cyclization step therefore leading to
a lower stereoselectivity.
P r ep a r a t ion of Ben zot r ia zole In t er m ed ia t es 15-21.
Gen er a l P r oced u r e. A mixture of benzotriazole (0.45 g, 3.54
mmol), a secondary amine (3.54 mmol), the appropriate
aldehyde (3.50 mmol), and 4 Å molecular sieves (1 g) in dry
benzene (15 mL) was stirred at room temperature for 12 h.
The mixture was diluted with CH2Cl2 (20 mL) and filtered
through Celite. The solvent was eliminated under reduced
pressure and the resulting thick oil maintained under vacuum
(0.1 mmHg) for 2 h and used without further manipulation.
When N-benzylaniline was used, molecular sieves were omit-
ted, and the mixture was refluxed in a Dean-Stark until the
theoretical amount of water had separated. Evaporation of
the solvent afforded the oily product that was processed as
described above. Details of the preparation of adducts 15-
21 and full 1H NMR data are collected in Tables 4, 5
(supporting information). 13C NMR and HRMS data for
selected cases are also given in the supporting information.
Rea ction s of Ben zotr ia zole Ad d u cts 15-21 w ith Sm I2.
1b
Gen er a l P r oced u r e. Meth od A. A 0.1 M solution of SmI2
was added dropwise to a stirred solution of the benzotriazole
adduct (0.38 mmol) in THF (40 mL) at 25 °C at such a rate as
to allow the SmI2 characteristic blue color to disappear before
(53) Still, W. C.; Kahn, M.; Mitra, A. J . Org. Chem. 1978, 43, 2923.
(54) (a) Bartels, R. H. Report CNA-44; University of Texas; Center
for Numerical Analysis: Austin TX, 1972. (b) Baker, J . J . Comput.
Chem. 1986, 7, 385.
(55) McIver, J . W.; Komornicki, A. K. J . Am. Chem. Soc. 1972, 94,
2625.
Con clu sion s
We have shown that the reduction with SmI2 of
iminium ions formed in situ by dissociation of R-benzo-
triazolyl amines derived from ω-unsaturated aldehydes
(56) Gaussian 94, Revision B.2, Frisch, M. J .; Trucks, G. W.;
Schlegel, H. B.; Gill, P. M. W.; J ohnson, B. G.; Robb, M. A.; Cheeseman,
J . R.; Keith, T.; Petersson, G. A.; Montgomery, J . A.; Raghavachari,
K.; Al-Laham, M. A.; Zakrzewski, V. G.; Ortiz, J . V.; Foresman, J . B.;
Peng, C. Y.; Ayala, P. Y.; Chen, W.; Wong, M. W.; Andres, J . L.;
Replogle, E. S.; Gomperts, R.; Martin, R. L.; Fox, D. J .; Binkley, J . S.;
Defrees, D. J .; Baker, J .; Stewart, J . P.; Head-Gordon, M.; Gonzalez,
C.; Pople, J . A. Gaussian, Inc., Pittsburgh PA, 1995.
(52) See reference 44b for related unfavorable secondary orbital
interactions between the SOMO of an alkoxy-substituted radical and
the HOMO of an electron-rich olefin that have been similarly used to
explain a trans-preference in radical cyclizations. We thank the Editor
for calling this work to our attention.
(57) Work is now in progress to extend these ab initio calculations
to the rest of the TSs included in Chart 2 and to related systems.

Diastereoselective Synthesis of Cycloalkylamines
J . Org. Chem., Vol. 62, No. 4, 1997 1133
the next drop was added. Addition continued until the blue
color persisted. The mixture was then stirred for 15 min and
treated with saturated K2CO3 (10 mL), followed by H2O (10
mL). The aqueous phase was extracted with EtOAc (3 × 100
mL). The crude product was purified as indicated below for
the individual cases. Meth od B. A solution of the benzotri-
azole adduct (0.59 mmol) in THF (6 mL) was added to a THF
(15 mL) solution of SmI2 (1.3 mmol) that had been cooled to
below -20 °C. The mixture was stirred for 30 min and then
allowed to reach room temperature slowly. It was then
proceeded as in method A. Details of method used, yields and
stereoselectivities are collected in Tables 1-3. The following
compounds were prepared:
) 6.9 Hz, 3H), 1.4-1.8 (m, 4H), 1.9-2.2 (m, 3H), 2.15 (dd, J )
15.4, 10.5 Hz, 1H, overlapped with m at 1.9-2.2), 2.9-3.0 (m,
1H), 3.9-4.0 (m, 2H), 4.36 (q, J ) 8.3 Hz, 1H), 4.51 and 4.60
(AB q, J ) 17.9 Hz, 2H), 6.7 (m, 2H), 7.1-7.3 (m, 8H); 13C
NMR δ 14.1, 22.8, 30.0, 31.5, 35.9, 37.9, 51.8, 60.2, 61.1, 113.2,
116.8, 126.1, 126.5, 128.4, 129.0, 139.8, 149.3, 173.1. Anal.
Calcd for C22H27NO2: C, 78.30; H, 8.06; N, 4.15. Found: C,
78.30; H, 8.04; N, 4.15.
Eth yl cis-[2-[N-Ben zyl-N-(2-cya n oeth yl)a m in o]cyclo-
pen tyl]eth an oate (c-27f). Flash chromatography (14% EtOAc
in hexanes) afforded the titled compound as a colorless oil. The
analytical sample was obtained after HPLC purification (10
mL/min, 14% EtOAc in hexanes, tR ) 26 min): 1H NMR δ 1.26
(t, J ) 7.1 Hz, 3H), 1.5-1.9 (m, 6H), 2.12 (dd, J ) 15.5, 10.2
Hz, 1H), 2.30 (t, J ) 7.1 Hz, 2H), 2.5-2.6 (m, 1H), 2.72 (dd, J
) 15.5, 3.8 Hz, 1H), 2.8-2.9 (m, 2H), 2.9-3.1 (m, 1H), 3.59
and 3.75 (AB q, J ) 14.2 Hz, 2H), 4.13 (q, J ) 7.1 Hz, 2H),
7.2-7.4 (m, 5H); 13C NMR δ 14.1, 14.2, 20.7, 27.5, 29.2, 33.1,
38.2, 47.4, 56.3, 60.2, 65.2, 119.0, 127.3, 128.4, 128.5, 138.5,
173.9. Anal. Calcd for C19H26N2O2: C, 72.58; H, 8.33; N, 8.91.
Found: C, 72.50; H, 8.40; N, 8.94.
Eth yl cis-[2-(4-P h en ylp ip er a zin -1-yl)cyclop en tyl]eth a -
n oa te (c-27g). Flash chromatography (10% EtOAc in hex-
anes) afforded the titled compound as microcrystals: mp 39-
41°C; 1H NMR δ 1.26 (t, J ) 7.3 Hz, 3H), 1.4-1.9 (m, 6H),
2.01 (dd, J ) 16.2, 11.3 Hz, 1H), 2.4-2.7 (m, 7H), 3.17 (t, J )
5.0 Hz, 4H), 4.13 (q, J ) 7.3 Hz, 2H), 6.8-6.9 (m, 3H), 7.2-
7.3 (m, 2H); 13C NMR δ 14.2, 20.2, 27.0, 28.9, 33.0, 37.2, 49.0,
52.5, 60.1, 69.1, 115.8, 119.5, 129.0, 151.3, 174.2. Anal. Calcd
for C19H28N2O2: C, 72.12; H, 8.92; N, 8.85. Found: C, 72.01;
H, 8.89; N, 9.08.
Eth yl cis-[2-[N-Ben zyl-N-(3-bu ten yl)a m in o]cyclop en -
tyl]eth a n oa te (c-27h ). Flash chromatography (5% EtOAc in
hexanes) afforded the titled compound as a colorless oil. The
analytical sample was obtained after HPLC purification (10
mL/min, 4% EtOAc in hexanes, tR ) 18 min); 1H NMR δ 1.26
(t, J ) 7.2 Hz, 3H), 1.5-1.8 (m, 6H), 2.08 (dd, J ) 15.6, 10.8
Hz, 1H), 2.1-2.2 (m, 2H), 2.5-2.7 (m, 3H), 2.75 (dd, J ) 15.6,
3.5 Hz, 1H), 2.9-3.1 (m, 1H), 3.61 and 3.72 (AB q, J ) 14.3
Hz, 2H), 4.13 (q, J ) 7.2 Hz, 2H), 4.9-5.1 (m, 2H), 5.6-5.7
(m, 1H), 7.2-7.4 (m, 5H); 13C NMR δ 14.2, 20.8, 27.6, 28.9,
29.1, 33.2, 38.2, 50.8, 55.9, 60.1, 65.3, 115.4, 126.6, 128.1, 128.4,
136.9, 140.2, 174.3. Anal. Calcd for C20H29NO2: C, 76.15; H,
9.27; N, 4.44. Found: C, 75.98; H, 9.27; N, 4.67.
cis- a n d tr a n s-[2-(N,N-Diben zyla m in o)cyclop en tyl]eth -
a n en itr ile (c-27i, t-27i). Colorless oil after flash chromatog-
raphy (20% EtOAc in hexanes). Data for the mixture of
isomers: 1H NMR δ 1.4-2.1 (m, 6H), 2.22 (dd, J ) 16.8, 10.8
Hz, 1H), 2.4-2.6 (m, 1H), 2.74 (dd, J ) 16.8, 3.9 Hz, 1H), 3.0-
3.1 (m, 1H, cis-isomer), 3.40 (d, J ) 13.7 Hz, 2H, trans-isomer),
3.60 and 3.70 (AB q, J ) 14.3 Hz, 4H, cis-isomer), 3.80 (d, J )
13.7 Hz, 2H, trans-isomer), 7.25-7.35 (m, 10H); 13C NMR δ
17.1, 20.6, 20.9, 21.7, 22.0, 27.3, 28.8, 29.2, 38.5, 38.9, 54.8,
56.4, 65.3, 65.7, 119.2, 120.5, 127.0, 128.3, 128.5, 128.6, 138.9,
139.7. Anal. Calcd for C21H24N: C, 82.85; H, 7.95; N, 9.20.
Found: C, 82.50; H, 7.90; N, 9.42.
cis- a n d tr a n s-(2-Mor p h olin ocyclop en tyl)eth a n en itr ile
(c-27j, t-27j). Flash chromatography (40% EtOAc in hexanes)
afforded in order of elution c-27j and t-27j as yellowish oils.
Data for c-27j: 1H NMR δ 1.2-1.5 (m, 1H), 1.6-1.9 (m, 5H),
2.04 (dd, J ) 16.4, 9.3 Hz, 1H), 2.4-2.5 (m, 7H), 3.68 (t, J )
4.6 Hz, 4H); 13C NMR δ 16.5, 19.9, 26.3, 28.5, 37.4, 53.0, 66.8,
68.9, 120.5; HRMS calcd for C11H18N2O 194.1419, found
194.1422. Data for t-27j: 1H NMR δ 1.4-1.8 (m, 5H), 1.8-
2.0 (m, 1H), 2.0-2.2 (m, 1H), 2.3-2.5 (m, 7H), 3.67 (s, 4H);
13C NMR δ 21.6, 22.4, 25.0, 29.9, 37.2, 50.4, 67.1, 71.3, 119.1.
Anal. Calcd for C11H18N2O: C, 68.01; H, 9.34; N, 14.42. Found
(mixture of isomers): C, 67.85; H, 9.01; N, 14.17.
1-Isop r op yl-3-m eth yl-2-m or p h olin ocyclop en ta n e (22)
a n d 3-Met h yl-1-(1-m et h ylet h en yl)-2-m or p h olin ocyclo-
p en ta n e (23). General method A was applied to adduct 15.
Flash chromatography (10% EtOAc in hexanes) afforded in
order of elution cyclopentanes 22 (contaminated with some 23,
15%) and 23 (contaminated with some 22, 12%), both as
yellowish oils. Data for 22: 1H NMR δ 0.84 (d, J ) 6.5 Hz,
3H), 0.92 (d, J ) 6.5 Hz, 3H), 0.99 (d, J ) 6.8 Hz, 3H), 1.0-
1.8 (m, 6H), 1.9-2.0 (m, 1H), 3.19 (t, J ) 6.5 Hz, 1H), 2.4-2.6
(m, 4H), 3.66 (t, J ) 4.6 Hz, 4H); 13C NMR δ 18.7, 21.2, 22.1,
25.2, 31.0, 33.1, 34.1, 47.5, 50.6, 67.6, 77.2; GC-MS: tR ) 12.6
min; LRMS (EI) m/z 211 (M+); HRMS calcd for C13H24NO (M+
- 1) 210.18579, found 210.18516. Data for 23: 1H NMR δ
1.03 (d, J ) 6.7 Hz, 3H), 1.2-1.8, 1.75 (s, 3H, overlapped with
m at 1.2-1.8), 1.99 (m, 1H), 2.38 (t, J ) 7.5 Hz, 1H), 2.5-2.7
(m, 4H), 3.66 (t, J ) 4.5 Hz, 4H), 4.71 (s, 1H), 4.79 (s, 1H); 13
C
NMR δ 20.2, 20.8, 30.2, 32.4, 35.6, 49.2, 50.8, 67.5, 76.4, 110.4,
148.8; GC-MS: tR ) 12.7 min; LRMS (EI) m/z 209 (M+); HRMS
calcd for C13H23NO 209.17796, found 209.17671.
Eth yl cis- a n d tr a n s-(2-Mor p h olin ocyclop en tyl)eth a -
n oa te (27a ). Flash chromatography (20% EtOAc in hexanes,
then EtOAc) afforded in order of elution (where appropriate;
see Table 1) c-27a and t-27a as yellowish oils. Data for
c-27a : 1H NMR δ 1.25 (t, J ) 7.1 Hz, 3H), 1.3-1.8 (m, 6H),
1.97 (dd, J ) 16.2, 11.4 Hz, 1H), 2.3-2.6 (m, 7H), 3.67 (t, J )
4.7 Hz, 4H), 4.10 (q, J ) 7.1 Hz, 2H); 13C NMR δ 14.3, 20.1,
26.7, 28.8, 32.9, 37.0, 53.0, 60.1, 67.0, 69.4, 174.2. Anal. Calcd
for C13H23NO3: C, 64.70; H, 9.61; N, 5.80. Found: C, 64.30;
H, 9.56; N, 5.60. Data for t-27a : 1H NMR δ 1.25 (t, J ) 7.1
Hz, 3H), 1.2-1.3 (m, 2H, overlapped with t at 1.25), 1.5-2.5
(m, 12H), 3.68 (t, J ) 4.0 Hz, 4H), 4.12 (q, J ) 7.1 Hz, 2H);
13C NMR δ 14.2, 22.9, 25.8, 30.8, 37.5, 39.9, 50.7, 60.2, 67.3,
72.2, 174.2; HRMS calcd for C13H23NO3 241.167793, found
241.167007.
Eth yl cis- a n d tr a n s-[2-(N,N-Diben zyla m in o)cyclop en -
tyl]eth a n oa te (27b). Colorless oil after flash chromatography
(15% EtOAc in hexanes). Data for the diastereomeric mix-
ture: 1H NMR δ 1.27 (t, J ) 7.1 Hz, 3H), 1.4-1.8 (m, 6H),
2.20 (dd, J ) 15.5, 10.7 Hz, 1H), 2.6-2.7 (m, 1H), 2.88 (dd, J
) 15.5, 3.8 Hz, 1H), 3.1 (m, 1H), 3.38 (d, J ) 13.7 Hz, 2H,
trans-isomer), 3.56 and 3.71 (AB q J ) 14.4 Hz, 4H, cis-isomer),
3.84 (d, J ) 13.7 Hz, 2H, trans-isomer), 4.10 (q, J ) 7.1 Hz,
2H), 7.2-7.4 (m, 10H); 13C NMR δ 14.2, 21.3, 21.6, 22.1, 27.4,
29.4, 30.0, 33.7, 38.4, 54.7, 56.1, 60.1, 65.5, 126.6, 128.1, 128.6,
128.7, 139.5, 140.3, 174.2. Anal. Calcd for C23H29NO2: C,
78.59; H, 8.32; N, 3.98. Found: C, 78.74; H, 8.30; N, 4.28.
Eth yl cis-[2-[N-Ben zyl-N-(2-p h en yleth yl)a m in o]cyclo-
pen tyl]eth an oate (c-27d). Flash chromatography (6% EtOAc
in hexanes) afforded the titled compound as a yellowish oil
that crystallized on standing (microcrystals): mp 38-40 °C;
1H NMR δ 1.29 (t, J ) 7.1 Hz, 3H), 1.3-1.9 (m, 6H), 2.12 (dd,
J ) 15.0, 10.8 Hz, 1H), 2.5-2.8 (m, 6H), 3.0-3.2 (m, 1H), 3.73
and 3.85 (AB q, J ) 14.1 Hz, 2H), 4.17 (q, J ) 7.1 Hz, 2H),
7.0-7.4 (m, 10H); 13C NMR δ 14.3, 20.8, 27.7, 29.1, 30.7, 33.2,
38.3, 53.4, 56.1, 60.1, 65.3, 125.7, 126.7, 128.2, 128.3, 128.5,
128.6, 140.0, 140.7, 174.3. Anal. Calcd for C24H31NO2: C,
78.86; H, 8.55; N, 3.83. Found: C, 79.05; H, 8.41; N, 3.81.
E t h yl cis-[2-(N-Ben zyl-N-p h en yla m in o)cyclop en t yl]-
eth a n oa te (c-27e). Flash chromatography (6% EtOAc in
hexanes) afforded the titled compound as a colorless oil that
crystallized on standing (microcrystals). The analytical sample
was obtained after HPLC purification (12 mL/min, 6% EtOAc
cis- a n d tr a n s-[2-(N,N-Dia llyla m in o)cyclop en tyl]eth -
an en itr ile (c-27k, t-27k). Flash chromatography (20% EtOAc
in hexanes) afforded in order of elution c-27k and t-27k as
yellowish oils. Data for c-27k : 1H NMR δ 1.4-1.6 (m, 1H),
1.6-1.9 (m, 5H,), 2.05 (dd, J ) 16.9, 10.6 Hz, 1H), 2.3-2.4
(m, 1H), 2.56 (dd, J ) 16.9, 3.9 Hz, 1H), 2.9-3.0 (m, 1H), 3.14
(d, J ) 6.3 Hz, 4H), 5.1-5.2 (m, 4H), 5.7-5.9 (m, 2H); 13C NMR
δ 16.7, 20.5, 27.1, 28.5, 38.5, 54.0, 64.5, 117.5, 120.7, 134.5.
1
in hexanes, tR ) 19 min): mp 56-58°C; H NMR δ 1.17 (t, J

1134 J . Org. Chem., Vol. 62, No. 4, 1997
Aurrecoechea et al.
Data for t-27k : 1H NMR δ 1.2-1.3 (m, 7H), 2.32 (dd, J ) 16.8,
9.1 Hz, 1H), 2.58 (dd, J ) 16.8, 4.1 Hz, 1H), 2.8-3.0 (m, 3H),
3.1-3.2 (m, 2H), 5.1-5.2 (m, 4H), 5.7-5.9 (m, 2H); 13C NMR
δ 20.7, 21.9, 22.9, 29.0, 38.5, 53.5, 66.4, 116.7, 119.4, 136.8.
Anal. Calcd for C13H20N2: C, 76.42; H, 9.87; N, 13.71. Found
(mixture of isomers): C, 76.05; H, 9.71; N, 13.39.
Eth yl [(1R*,2R*,3S*)-2-(N,N-Diben zyla m in o)-3-m eth yl-
cyclop en tyl]eth a n oa te (28a ). Flash chromatography (4%
EtOAc in hexanes) afforded the titled compound (95/5 dia-
stereomeric mixture) as a colorless oil. Data for the major
(1R*,2R*,3S*)-isomer: 1H NMR δ 1.11 (d, J ) 6.6 Hz, 3H),
1.31 (t, J ) 7.0 Hz, 3H, overlapped with m at 1.1-1.5), 1.1-
1.5 (m, 7H), 1.8-2.0 (m, 2H), 2.3-2.7 (m, 3H), 2.8-2.9 (m, 2H),
3.6 (d, J ) 14.0 Hz, 2H), 4.0-4.2 (m, 2H), 7.2-7.4 (m, 10H);
13C NMR δ 14.2, 22.2, 31.3, 32.5, 32.9, 35.6, 39.1, 55.7, 60.0,
C14H25NO2: C, 65.85; H, 9.87; N, 5.48. Found: C, 65.47; H,
9.83; N, 5.61.
E t h yl cis- a n d tr a n s-(2-Mor p h olin ocycloh exyl)et h a -
n oa te (c-34a , t-34a ). Flash chromatography (15% EtOAc in
hexanes) afforded in order of elution t-34a (16%) and c-34a
(38%) as yellowish oils. Data for t-34a : 1H NMR δ 1.0-1.3
(m, 3H), 1.26 (t, J ) 7.1 Hz, 3H, overlapped with m at 1.0-
1.3), 1.6-2.0 (m, 8H), 2.3-2.4 (m, 2H), 2.6-2.7 (m, 3H), 3.6-
3.7 (m, 4H), 4.1 (m, 2H); 13C NMR δ 14.2, 23.9, 25.9, 33.1, 36.7,
39.6, 48.9, 59.9, 67.5, 68.8, 173.9; HRMS calcd for C14H25NO3
255.18344, found 255.18383. Data for c-34a : 1H NMR δ 1.0-
1.4 (m, 4H), 1.25 (t, J ) 7.1 Hz, 3H, overlapped with m at
1.0-1.4), 1.6-1.8 (m, 4H), 2.0-2.7 (m, 8H), 3.67 (t, J ) 4.7
Hz, 4H), 4.11 (q, J ) 7.1 Hz, 2H); 13C NMR δ 14.2, 20.0, 24.1,
25.4, 29.0, 31.3, 31.7, 50.6, 60.1, 65.0, 67.3, 174.2; HRMS calcd
for C14H25NO3 255.18344, found 255.18347.
6-(E t h oxyca r b on yl)-7-m or p h olin ob icyclo[3.2.0]h ep -
ta n e (36). General procedure A was used. Flash chromatog-
raphy (15% EtOAc in hexanes) afforded in order of elution
t-34a (6%), 36a (4%), c-34a (29%), and 36a ′ (30%) as thick
colorless oils. Data for the less polar isomer of 36: 1H NMR
δ 1.23 (t, J ) 7.1 Hz, 3H), 1.4-1.9 (m, 6H), 2.2 (m, 4H), 2.37
(t, J ) 7.1 Hz, 1H), 2.6-2.9 (m, 2H), 2.90 (t, J ) 8.0 Hz, 1H),
3.66 (t, J ) 4.7 Hz, 4H), 4.10 (q, J ) 7.1 Hz, 2H); 13C NMR δ
14.2, 25.3, 26.3, 32.1, 37.9, 38.3, 45.4, 50.3, 59.9, 60.3, 66.7,
174.7; HRMS calcd for C14H23NO3 253.16779, found 253.16829.
Data for the more polar isomer of 36: 1H NMR δ 1.24 (t, J )
7.1 Hz, 3H), 1.3-1.7 (m, 6H), 2.2-2.4 (m, 4H), 2.5 (m, 1H),
2.6 (m, 1H), 2.9-3.0 (m, 2H), 3.68 (t, J ) 4.6 Hz, 4H), 4.0-4.2
(m, 2H); 13C NMR δ 14.4, 26.2, 28.1, 32.0, 35.8, 40.5, 42.7, 50.2,
60.2, 64.4, 66.6, 172.5; HRMS calcd for C14H23NO3 253.16779,
found 253.16814. Anal. Calcd for C14H23NO3: C, 66.37; H,
9.15; N, 5.53. Found: C, 66.74; H, 8.88; N, 5.40.
67.8, 126.7, 128.1, 128.3, 140.0, 173.7; HRMS calcd for C24H31
NO2 365.23548, found 365.23541.
-
Eth yl [(1R*,2R*,3S*)-3-Meth yl-2-m or p h olin ocyclop en -
tyl]eth a n oa te (28b). The crude product was purified by flash
chromatography (35% EtOAc in hexanes) and HPLC (12 mL/
min, 30% EtOAc in hexanes, tR ) 26 min) to yield the titled
compound (∼95/5 diastereomeric mixture) as a colorless oil:
1H NMR δ 1.01 (d, J ) 6.0 Hz, 3H), 1.21 (t, J ) 7.1 Hz, 3H,
overlapped with m at 1.1-1.2), 1.1-1.2 (m, 1H), 1.3-1.4 (m,
1H), 1.6-1.8 (m, 1H), 1.9-2.1 (m, 3H), 2.23 (t, J ) 6.5 Hz,
1H), 2.3-2.6 (m, 6H), 3.60 (t, J ) 4.5 Hz, 4H), 4.07 (q, J ) 7.1
Hz, 2H); 13C NMR δ 14.2, 23.0, 29.1, 32.2, 32.7, 33.8, 39.4, 52.7,
59.9, 67.1, 75.5, 174.0; HRMS calcd for C14H25NO3 255.18344,
found 255.18335.
Eth yl [(1S*,2R*,5R*)-2-(N,N-Diben zyla m in o)-5-m eth yl-
cyclop en tyl]eth a n oa te (29a ) a n d Its (1S*,2R*,5S*)-isom er
(29a ′). Following General method B, after the reaction
mixture reached room temperature it was stirred at that
temperature further 22 h. The normal workup afforded, after
flash chromatography (1% EtOAc in hexanes), the mixture of
29a , 29a ′, and 30a in a ratio (1H NMR) of 84.3:14.0:1.7. The
mixture was separated by HPLC (8 mL/min, 5% EtOAc in
hexanes) to yield in order of elution 29a ′, 29a , and 30a as
Eth yl cis- an d tr a n s-[2-(N,N-Diben zylam in o)cycloh exyl]-
eth a n oa te (34b). Flash chromatography (15% EtOAc in
hexanes) afforded the titled compound as a colorless oil. Data
for the diastereomeric mixture: 1H NMR δ 0.8-2.2 (m, 13H),
2.40 (dd, J ) 14.8, 10.7 Hz, 1H, one isomer), 2.69 (d, J ) 8.6
Hz, 2H, one isomer), 2.87 (d, J ) 14.8 Hz, 1H, one isomer),
3.35 (d, J ) 13.7 Hz, 2H, one isomer), 3.63 (d, J ) 14.6 Hz,
2H), 3.83 (d, J ) 14.6 Hz, 2H), 7.2-7.4 (m, 10H); 13C NMR δ
14.3, 20.1, 23.3, 25.8, 25.9, 26.0, 28.9, 31.8, 32.6, 34.1, 37.3,
38.5, 53.4, 55.1, 60.0, 60.2, 60.7, 62.3, 126.5, 126.7, 128.1, 128.1,
1
colorless oils. Data for 29a ′: tR 26 min; H NMR δ 0.93 (d, J
) 6.9 Hz, 3H), 1.26 (t, J ) 7.1 Hz, 3H), 1.3-1.4 (m, 2H), 1.7-
1.8 (m, 1H), 2.10 (dd, J ) 16.8, 9.8 Hz, 1H, overlapped with m
at 2.0-2.2), 2.0-2.2 (m, 1H), 2.7-2.8 (m, 1H), 2.94 (dd, J )
16.8, 2.2 Hz, 1H), 2.9-3.0 (m, 1H), 3.52 and 3.76 (AB q, J )
14.4 Hz, 4H), 4.1 (m, 2H), 7.2-7.3 (m, 10H); 13C NMR δ 14.2,
16.0, 27.9, 28.4, 29.0, 35.3, 42.0, 55.4, 60.2, 66.7, 126.6, 128.0,
128.9, 139.2, 174.7; HRMS calcd for C24H31NO2 365.23548,
found 365.23563. Data for 29a : tR 28 min; 1H NMR δ 0.99 (d,
J ) 6.2 Hz, 3H), 1.0-1.2 (m, 1H), 1.27 (t, J ) 7.2 Hz, 3H),
1.5-2.0 (m, 4H), 2.1-2.3 (m, 2H), 2.84 (dd, J ) 14.6, 4.8 Hz,
1H), 3.26 (dt, J ) 8.1, 7.6 Hz, 1H), 3.53 and 3.61 (AB q, J )
14.3 Hz, 4H), 4.0-4.2 (m, 2H), 7.2-7.5 (m, 10H); 13C NMR δ
14.2, 21.6, 26.4, 32.1, 34.2, 38.5, 46.2, 56.2, 60.1, 63.2, 126.7,
128.1, 128.6, 139.7, 174.0; HRMS calcd for C24H31NO2
365.23548, found 365.23513. Data for 30a : tR 33 min; 1H NMR
δ 0.97 (d, J ) 6.5 Hz, 3H), 1.1-1.2 (m, 1H), 1.4-1.6 (m, 2H),
1.7-1.8 (m, 2H), 2.19 (dd, J ) 12.6, 2.5 Hz, 1H, overlapped
with m at 2.0-2.2), 2.0-2.2 (m, 1H), 3.08 (dd, J ) 12.6, 7.7
Hz, 1H), 3.23 (d, J ) 13.7 Hz, 1H), 3.4-3.6 (m, 1H), 3.67 (d, J
) 13.7 Hz, 1H), 4.20 (s, 1H), 7.2-7.4 (m, 8H), 7.53 (dd, J )
8.1, 1.4 Hz, 2H); 13C NMR δ 19.8, 24.3, 32.6, 39.5, 40.5, 43.2,
54.2, 57.5, 72.9, 127.0, 128.1, 128.3, 128.4, 128.4, 128.8, 137.5,
138.9, 211.3; HRMS calcd for C22H25NO 319.19361, found
319.19324.
128.4, 128.8, 129.1, 140.3, 140.7, 174.3; HRMS calcd for C24H31
NO2 365.23548, found 365.23595.
-
cis- a n d tr a n s-3-[(Met h oxyca r b on yl)m et h yl]-4-m or -
p h olin och r om a n (35a ). Flash chromatography (25% EtOAc
in hexanes) afforded the separated isomers as yellowish oils.
Data for the less polar isomer: mp 62-64°C; 1H NMR δ 2.1-
2.7 (m, 7H), 3.20 (bs, 1H), 3.5-3.7 (m, 4H), 3.67 (s, 3H), 3.9-
4.2 (m, 1H), 4.35 (dd, J ) 10.9, 2.4 Hz, 1H), 6.8-6.9 (m, 2H),
7.1-7.2 (m, 2H); 13C NMR δ 31.2, 34.2, 51.2, 51.6, 62.7, 66.2,
67.5, 116.8, 118.9, 119.9, 129.0, 132.5, 154.7, 172.8. Data for
the more polar isomer: 1H NMR δ 2.3-2.7 (m, 6H), 2.76 (q, J
) 7.9 Hz, 1H), 3.6-3.7 (m, 5H), 3.71 (s, 3H), 3.94 (t, J ) 11.0
Hz, 1H), 4.1-4.2 (m, 1H), 6.7-7.2 (m, 4H); 13C NMR δ 32.4,
35.6, 51.7, 54.2, 60.1, 66.1, 67.6, 116.8, 119.8, 119.9, 129.1,
130.9, 154.5, 172.8. Anal. Calcd for C16H21NO4: C, 65.96; H,
7.26; N, 4.81. Found (diastereomeric mixture): C, 65.60; H,
7.12; N, 5.03.
cis- a n d tr a n s-4-(4-P h en ylp ip er a zin -1-yl)-3-[(m eth oxy-
ca r bon yl)m et h yl]ch r om a n (35c). Flash chromatography
(20% EtOAc in hexanes) afforded the diastereomeric mixture
of 35c as an oil. The isomers were separated by HPLC (12
mL/min, 27% EtOAc in hexanes). Data for the less polar
Eth yl [(1S*,2R*,5R*)-5-Meth yl-2-m or p h olin ocyclop en -
tyl]eth a n oa te (29b). Following General method B, after the
reaction mixture reached room temperature it was stirred at
that temperature further 17 h. The normal workup afforded,
after flash chromatography (30% EtOAc in hexanes), the
diastereomeric mixture of 29b and 29b′ as a colorless oil from
where the major isomer 29b could be separated by HPLC (12
mL/min, 30% EtOAc in hexanes, tR ) 23 min): 1H NMR δ 0.98
(d, J ) 6.9 Hz, 3H), 1.1-1.3 (m, 1H, overlapped with t at 1.23),
1.23 (t, J ) 7.1 Hz, 3H), 1.3-1.4 (m, 1H), 1.4-2.0 (m, 4H),
2.0-2.2 (m, 1H), 2.3-2.7 (m, 6H), 3.64 (t, J ) 4.7 Hz, 4H),
4.10 (q, J ) 7.1 Hz, 2H); 13C NMR δ 14.2, 22.9, 27.2, 30.3,
34.3, 37.5, 44.2, 52.8, 60.0, 66.9, 67.2, 174.1. Anal. Calcd for
1
isomer: tR 11.8 min; H NMR δ 2.2-2.5 (m, 2H), 2.5-2.7 (m,
3H), 2.8-2.9 (m, 2H), 3.15 (t, J ) 4.9 Hz, 4H), 3.30 (bs, 1H),
3.69 (s, 3H), 4.06 (dd, J ) 11.0, 2.0 Hz, 1H), 4.37 (dd, J )
11.0, 1.5 Hz, 1H), 6.8-7.0 (m, 5H), 7.1-7.3 (m, 4H). Data for
1
the more polar isomer: tR 13.2 min; H NMR δ 2.37 (dd, J )
15.7, 5.5 Hz, 1H), 2.5-2.8 (m, 6H), 3.13 (t, J ) 4.8 Hz, 4H),
3.71 (s, 3H), 3.83 (d, J ) 3.6 Hz, 1H), 3.99 (t, J ) 11.1 Hz,
1H), 4.15 (dd, J ) 11.1, 3.9 Hz, 1H), 6.8-7.0 (m, 5H), 7.1-7.3
(m, 4H); 13C NMR (diastereomeric mixture) δ 31.5, 32.4, 34.4,
35.6, 49.8, 50.6, 51.6, 53.6, 59.5, 62.4, 66.0, 66.4, 116.0, 116.7,
116.8, 119.4, 119.6, 119.7, 120.0, 120.1, 129.0, 130.8, 132.4,

Diastereoselective Synthesis of Cycloalkylamines
J . Org. Chem., Vol. 62, No. 4, 1997 1135
151.3, 154.4, 154.7, 172.8. Anal. Calcd for C22H26N2O3: C,
72.11; H, 7.15; N, 7.64. Found (diastereomeric mixture): C,
71.63; H, 7.23; N, 7.64.
versidad del Pa´ıs Vasco (UPV 170.310-EC201/94) is
gratefully acknowledged. We also thank the Departa-
mento de Educacio´n, Universidades e Investigacio´n
(Gobierno Vasco, Spain) for a Fellowship (to A.F-A.).
cis- a n d tr a n s-4-[N-Ben zyl-N-(2-cya n oeth yl)a m in o]-3-
[(m eth oxyca r bon yl)m eth yl]ch r om a n (35d ). Flash chro-
matography (32% EtOAc in hexanes) afforded the diastereo-
meric mixture of 35d as an oil: 1H NMR δ 2.2-2.3 (m, 7H),
3.67 and 3.68 (s, 3H), 3.6-4.3 (m, 5H), 6.9-7.5 (m, 9H); 13C
NMR δ 17.5, 18.3, 32.0, 32.6, 34.7, 34.9, 47.2, 49.5, 51.7, 56.2,
56.9, 58.5, 59.1, 66.1, 67.3, 117.1, 118.8, 120.2, 120.6, 121.0,
127.5, 127.6, 128.5, 128.6, 128.8, 129.0, 129.2, 129.6, 130.8,
138.6, 139.1, 154.8, 155.4, 172.7; GC-MS, tR ) 15.4, 16.1 min;
HRMS calcd for C22H24N2O3 364.17869, found 364.17883.
Su p p or tin g In for m a tion Ava ila ble: Preparation and
characterization data for 8-10; 13C NMR and HRMS data for
15, 16b, and 17b; Tables 4, 5, 6-8; copies of 1H and 13C NMR
spectra of compounds 8b-d , 9, 22, 23, t-27a , 28a , 28b, 29a ,
29a ′, 30a , c-34a , t-34a , 34b, 35d (44 pages). This material is
contained in libraries on microfiche, immediately follows this
article in the microfilm version of the journal, and can be
ordered from the ACS; see any current masthead page for
ordering information.
Ack n ow led gm en t. Financial support by the Direc-
cio´n General de Investigacio´n Cient´ıfica y Te´cnica
(DGICYT PB89-0412 and PB92-0449) and by the Uni-
J O960957D