Synthesis of β-lactams and cyclo-β-dipeptides from β-amino acids: Experimental observations and theoretical analysis

Article abstract of DOI:10.1016/S0040-4020(00)01169-8

The cyclization of β-amino acids by means of activating agents is one of the most useful approaches for the construction of β-lactams; however, we found that when PhP(O)Cl₂ (in Et₃N) is employed as the activating agent, cyclization o

Full text of DOI:10.1016/S0040-4020(00)01169-8

TETRAHEDRON
Pergamon
Tetrahedron 57 (2001) 1883±1890
Synthesis of b-lactams and cyclo-b-dipeptides from b-amino
acids: experimental observations and theoretical analysis
a,p
Jaime Escalante, Miguel A. Gonz a lez-Tototzin, Judit Avi nÄ a, Omar Mu nÄ oz-Mu nÄ iz
a
b
b
b,p
and Eusebio Juaristi
a
Centro de Investigaciones Qu Âõ micas, Universidad Aut o noma del Estado de Morelos, Av. Universidad 1001,
6
2210 Cuernavaca, Mor., Mexico
b
Departamento de Qu Âõ mica, Centro de Investigaci o n y de Estudios Avanzados del Instituto Polit e cnico Nacional,
Apartado Postal 14-740, 07000 Mexico, D.F., Mexico
Received 4 February 2000; accepted 11 December 2000
AbstractÐThe cyclization of b-amino acids by means of activating agents is one of the most useful approaches for the construction of
b-lactams; however, we found that when PhP(O)Cl (in Et N) is employed as the activating agent, cyclization of the derived `active ester'
2
3
affords varying amounts of cyclo-b-dipeptides, depending on reactions conditions (solvent, temperature, and concentration), as well as on the
substitution pattern in the starting b-amino acid. Theoretical rationalization of the experimental results was achieved by modeling studies of
the presumed intermediates, both at semi-empirical (MNDO, AM1, and PM3) and at ab initio (HF/3-21G) levels. In the latter calculations,
simulation of solvents was accomplished by means of self-consistent reaction ®eld theory (Onsager's method). q 2001 Elsevier Science Ltd.
All rights reserved.
1
. Introduction
common methods for the construction of b-lactams via8
intramolecular condensation (pathway A in Scheme 1).
However, we have found that when the activating agent is
The chemistry of b-amino acids is of continuing interest in
modern organic chemistry. Application to b-lactam syn-
1
PhP(O)Cl in the presence of excess triethylamine, two
2
2
thesis and to the synthesis of structurally rich b-peptides
3
b-amino acids, 3-(benzylamino)propanoic acid and
3-(benzylamino)-3-methylpropanoic acid, afford cyclic
b-dipeptides (pathway B in Scheme 1).
has led to growing attention to the area. In particular,
b-dipeptides have been shown to possess interesting bio-
4
logical properties as well as greater enzyme-inhibiting
5
abilities than b-lactams. Furthermore, short oligomers
made of b-amino acids give rise to stabilized helical struc-6
tures and can exhibit a resistance to enzymatic hydrolysis.
2. Results and discussion
2
.1. b-Lactam formation
Activation of the carboxyl group on a b-amino acid by
means of an organophosphorus reagent is one of the most
7
3-(Benzylamino)-3-phenylpropanoic acid (rac-1a) was
Scheme 1.
Keywords: azetidinones; peptides and polypeptides; theoretical studies; solvent effects; amino acids and derivatives.
Corresponding authors; e-mail: jaime@servm.fc.uaem.mx; juaristi@relaq.mx
p
0040±4020/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)01169-8

1
884
J. Escalante et al. / Tetrahedron 57 (2001) 1883±1890
Table 1. Lactamization of 3-(benzylamino)-3-phenylpropanoic acid (1a) under varying reaction conditions
a
Yield (%)
Entry
Concentration (M) of rac-1a
Solvent
Temperature (8C)
Time (h)
1
2
3
4
5
6
7
0.1
Benzene
Benzene
Benzene
Benzene
MeCN
80
80
40
80
82
40
25
20
20
20
20
20
96
96
78
99
49
99
75
55
31
0.01
0.01
0.005
0.01
0.01
0.01
CH
CH
2
Cl
Cl
2
2
2
a
b-Lactam (^)-2a is the only product found in this reaction.
Table 2. b-Lactam formation with various N-substituents in the b-amino-
b-phenyl-propanoic acid substrate
achieved at higher reaction temperatures. Finally, it is
important to point out that no evidence for dipeptide for-
mation was found; that is, only starting b-amino acid rac-1a
accompanies the product in those cases where conversion
was not complete.
The possible effect of the substituent at nitrogen was
explored (Table 2). We can see that when R is an aryl
group, higher yields are obtained than when R is an alkyl
group; thus, b-lactamization is facilitated by incorporation
of a phenyl group in the N-substituent of the b-amino acid.
Nevertheless, even when RˆCH (CH ) , this method is
Entry
R
Yield (%)
1
2
3
4
C
6
H
5
99
95
72
62
p-C
o-CH
CH (CH
6
H
5
±C
6
H
4
3
O±C
6
H
4
3
2 3
3
2
)
3
compared favorably with the other methods for b-lacta-
9
mization of rac-1a.
b-Lactam formation in this reaction (Tables 1 and 2) may
proceed via intermediate A, as shown in Scheme 2. The
reaction is run with 2 mol equiv. of triethyl amine; there-
fore, the conjugate base of the amino acid must be the initial
species of the mechanism.
treated with 1.5 equiv. of phenylphosphonic dichloride in
the presence of excess triethylamine under varying reaction
conditions (Table 1). Benzene, especially under high dilu-
tion, gave the best yields of b-lactam formation (rac-2a) and
is therefore generally recommended as a solvent. Aceto-
nitrile (at 828C) was also an effective solvent for b-lactam
formation (Table 1); by contrast, dichloromethane afforded
low yields, even at long reaction times (Tˆ25 or 408C).
In order to explore the effect of substitution at C(2) in the
starting b-amino acid, we carried out the reaction with
(^)-N,a-dibenzyl-b-alanine, rac-4, in benzene at 0.01 M
concentration, and under re¯ux. (Scheme 3). In this case,
Some of the data presented in Table 1 give evidence of
signi®cant concentration and temperature effects. So,
comparison of entries 1 and 2 (or 4) show increased yields
of b-lactam at lower concentration. This result can be inter-
preted in terms of intramolecular lactamization, being
favored over intermolecular peptide formation. On the
other hand, examination of entries 2 and 3, as well as entries
a syn-periplanar aromatic stacking effect may facilitate
formation of the b-lactam.
10
2.2. Cyclo-b-dipeptide formation
As indicated in Scheme 1, an alternative route, for the
reaction of b-amino acids with PhP(O)Cl , affords
6
and 7, show clearly that higher yields of lactam 2a are
2
Scheme 2.

J. Escalante et al. / Tetrahedron 57 (2001) 1883±1890
1885
Scheme 3.
cyclo-b-dipeptides, presumably via cyclization of linear
b-dipeptides. In particular, b-amino acids 6a and rac-6b
provided cyclo-b-dipeptides 3a and rac-3b in moderate
yields (Table 3). These dipeptides are formed in similar
yields at high and low concentrations of the reaction
media (see entries 1±3 and 6±8 in Table 3). While best
yields of 3a and rac-3b were found in benzene, acetonitrile
and CH Cl were also effective solvents for cyclo-b-dipep-
2
2
tide formation (cf. entries 4, 5 and 9 in Table 3). Reaction
temperature seems to be important in the yield; thus, no
cyclo-b-dipeptide rac-3b is observed in the reaction of
rac-6b in CH Cl at ambient temperature (compare entries
2
2
Chart 1.
10 and 11 in Table 3).
Other researchers apparently missed the identi®cation of
cyclo-b-dipeptides in reactions, attempting the lactami-
zation of b-amino acids with activating agents. Neverthe-
less, the preparation of cyclo-b-dipeptides has been
stationary phase) established that the latter are only a
minor side product, 6% in the case of 6a and 20% in the
1
1
1
case of rac-6b. On the other hand, H NMR and infrared
spectroscopic analysis of the crude product obtained from
the reaction of 6a with PhP(O)Cl discards, by comparison
1
described in a few cases. Although, ordinary H and
2
1
13
C
NMR spectra, and even EI mass spectra, may be unsuitable
for distinction between b-lactams 2 and cyclo-b-dipeptides
2
1
4
with authentic samples, the presence of linear dipeptide 7,
linear tripeptide 8, and cyclo-b-tripeptide 9 (Chart 1).
1
3
3, characteristic infrared bands allow easy differentiation.
In particular, whereas, b-lactams 2 and 5 present carbonyl
Interestingly, the formation of cyclo-b-dipeptide rac-3b
from b-amino acid rac-6b appears to be highly diastereo-
13
2
stretch absorptions around 1730±1750 cm (see Section
1
1
selective. Indeed, H and C NMR spectra, and HPLC
4
1
), cyclo-b-dipeptides 3 exhibit CvO values close to
640 cm .
2
1
Ê
analyses (mporasil 125 A column) of the crude product indi-
cate the presence of a single diastereoisomer. Although, the
relative con®guration of this product has not been securely
established, molecular modeling (see Section 2.3) does lead
to the conclusion that cyclic b-dipeptide like-3b should be
more stable than the unlike isomer, that is, (R,R)- and
(S,S)-3b are of lower energy than (R,S)-3b. Therefore, it
Importantly, the infrared spectra for the crude products in
the reactions described in Table 3 showed the presence of
2
1
cyclo-b-dipeptides 3a and rac-3b (nˆ1640 cm ) and of
1
5
the carbonyl band that corresponds to b-lactams (nˆ
2
730±1750 cm ); nevertheless, HPLC analysis (mporasil
1
1
Table 3. Cyclo-b-dipeptide formation of N-benzyl-b-alanine (6a) and (^)-N-benzyl-3-methyl-b-alanine (rac-6b) under various reaction conditions
a
Yield (%)
Entry
Starting material
Concentration (M)
Solvent
Temperature (8C)
Time (h)
1
2
3
4
5
6
7
8
9
6a
6a
6a
6a
0.1
Benzene
Benzene
Benzene
80
80
80
82
40
80
80
80
82
40
25
20
20
20
20
96
20
20
20
20
96
96
48
54
41
48
43
66
68
63
35
26
,0
0.01
0.005
0.01
0.1
CH
CH
3
CN
Cl
6a
2
2
rac-6b
rac-6b
rac-6b
rac-6b
rac-6b
rac-6b
0.1
Benzene
Benzene
Benzene
0.01
0.005
0.01
0.1
CH
CH
CH
3
2
2
CN
1
0
1
Cl
Cl
2
1
0.1
2
a
Yield after ¯ash chromatography.

1
886
J. Escalante et al. / Tetrahedron 57 (2001) 1883±1890
Figure 1. Structure and solid-state conformation of unsubstituted N-benzylated cyclo-b-dipeptide 3a.
Scheme 4.
can be argued that late, product-like transition states in the
reactions leading to b-dipeptide formation discriminate in
favor of like diastereomers.
across the ring, that allows for intramolecular electrostatic
stabilization of the boat conformation.
Scheme 4 presents a reasonable reaction mechanism for
cyclo-b-dipeptide formation. Two molecules of inter-
mediate A (see Scheme 2) react to form a b-dipeptide, B,
which undergoes intramolecular amide bond formation to
afford the corresponding cyclo-b-dipeptide.
In order to con®rm that the b-lactam is stable under the
reaction conditions, a control experiment was carried out
with equimolar amounts of b-amino acid 6a and b-lactam
rac-2a, under the established conditions [PhP(O)Cl , Et N/
2
3
benzene, 808C, 20 h]. While most of the starting b-lactam
was recovered, the yield of cyclo-b-dipeptide 3a was again
(cf. Table 3) ca. 50%. These observations demonstrate that
the b-lactam is stable under the reactions conditions, with
excess amino acid present.
Some support for the proposal that b-amino acids 6 generate
initially an open-chain b-dipeptide, which then cyclizes to
give the ®nal product, comes from the con®rmation that
b-dipeptide 7 reacts under the established conditions
[
to give cyclo-b-dipeptide 3a, albeit in low yield. (Scheme
5).
0.01 M in benzene solvent, PhP(O)Cl , Et N, 808C, 20 h]
2 3
Recrystallization of cyclo-b-dipeptide 6a, afforded a suit-
able crystal for X-ray diffraction analysis. The observed
structure and solid-state conformation is presented in Fig.
1. Salient features in this crystallographic structure are (1)
the slightly twisted boat conformation of the eight
membered ring and (2) the close proximity between the
basic amide nitrogen and the carbonyl carbon situated
2.3. Molecular modeling
1
6
With the aim of understanding the contrasting reactivity
behavior of b-amino acids 1 and 6, the former giving the
Scheme 5.

J. Escalante et al. / Tetrahedron 57 (2001) 1883±1890
1887
Chart 2.
anticipated b-lactams 2, while the latter affording mainly
cyclo-b-dipeptides 3, a series of theoretical studies were
carried out.
Figure 3. Ab initio HF/3-21G conformation of lowest energy for activated
b-dipeptide B-7. Hydrogen bonding between the amino N-H group and the
phosphoryl oxygen helps in stabilizing the `folded' conformation and
renders the phosphorus more electron de®cient.
Initial geometries for presumed intermediates A-1a, A-6a
and A-6b (Chart 2) were calculated by means of semi-
empirical methods (MNDO, AM1 and PM3), accessible in
1
7
the gaussian 94 package of programs.
The structures of lowest energy selected from the semi-
empirical calculations were then optimized at ab initio
1
8
19
HF/3-21G level, as available in gaussian 98. Further-
more, the role of the solvent was taken into account by
means of the general self-consistent reaction ®eld (SCRF)
model proposed for quantum chemical computations on
2
0
solvated molecules. In this model, the solvent is repre-
sented by an in®nite dielectric continuum, characterized
by its dielectric relative permittivity e, in which a cavity
is created and the solute is placed in it. The charge distri-
bution of the solute polarizes the continuum, which in turn
2
1
creates an electrostatic ®eld inside the cavity.
Figure 4. Ab initio HF/3-21G conformations of minimum energy for (a)
like-cyclo-b-dipeptide l-3b, and (b) unlike-cyclo-b-dipeptide u-3b. The like
diastereoisomer is calculated to be 3.6 kcal/mol more stable.
Fig. 2 shows the ab initio HF/3-21G calculated most stable
2
2
structures for A-1a, A-6a and A-6b in a solvent of eˆ2.3.
Most interestingly, unsubstituted b-amino acid phosphoryl-
ated intermediate A-6a presents an `extended' confor-
mation, with a considerably large distance between the
As it was indicated above, substrate 6a affords cyclo-b-
dipeptide 3a in good yield (Table 3) and this experimental
®nding, i. e. the lack of b-lactam formation, can be under-
stood in view of the preferred `extended' conformation of
the activated derivative A-1a (Fig. 2). In order to examine
the propensity of open-chain phosphorylated b-dipeptide B-
7 for cyclization, we next calculated the conformation of
minimum energy for this intermediate (Fig. 3). Pleasingly,
B-7 is estimated to adopt a `folded' conformation with a
Ê
nitrogen and the activated carbonyl, rˆ3.77 A, which
should be unfavorable for lactam formation, as experi-
mentally found. By contrast, C(3)-substituted A-1a and
A-6b present `folded' conformations, with short N´´´CvO
Ê
distances (rˆ2.71 and 2.61 A, respectively), which should
facilitate b-lactamization. This interpretation is in agree-
ment with experimental observation in the case of substrate
Ê
relatively short N´´´CvO distance, rˆ3.72 A, that should
1a (Table 1), but seems to be at odds with 6b as the starting
b-amino acid.
favor cyclization to give 3a, as experimentally observed.
Most interestingly, this folded conformation is stabilized
Figure 2. Ab initio HF/3-21G conformations of minimum energy, in a solvent of eˆ2.3, for activated b-amino acids A-1a, A-6a and A-6b.

1
888
J. Escalante et al. / Tetrahedron 57 (2001) 1883±1890
Table 4. Infrared carbonyl frequencies for representative b-lactams and
and the phosphoryl oxygen, and this renders the phosphorus
more electron de®cient, further activating dipeptide derivate
B-7 towards cyclization.
21
cyclo-b-dipeptides prepared in this work (nCvO, in cm
)
a
Compound
Experimental
Calculated
Lactam 2a
Lactam 4
Cyclo-b-dipeptide 3a
Cyclo-b-dipeptide 3b
1747
1740
1639
1641
1751
1743
1639
1639
Finally, both semi-empirical and ab initio calculations indi-
cate that the like-diastereoisomer of disubstituted cyclo-b-
dipeptide 3b is signi®cantly more stable than the unlike
isomer, because the former places both methyl substituents
pseudo-equatorial, whereas one of the methyl groups in
u-3b adopts a pseudo-axial orientation in the eight-
membered ring.
a
21
Ab initio HF/3-21G level (scaling factor to 0.9085), gaussian 98.
by hydrogen-bonding between the N-H and the phosphoryl
oxygen (Fig. 3). This hydrogen bonding might render the
phosphorus more electron de®cient, further activating
2
3
dipeptide B-7 towards cyclization.
. Experimental25
4
The relative energy between like and unlike cyclo-b-dipep-
tides 3b was estimated. In apparent agreement with experi-
ment (see Section 2.2), the like diastereoisomer [(R,R)-3b
and (S,S)-3b] is calculated to be ca. 3.6 kcal/mol more stable
than the unlike [(R,S)-3b] isomer. Examination of Fig. 4
offers a ready explanation for the above observation, since
the methyl substituents in like-3b take both pseudo-equa-
torial positions in the eight-membered ring, whereas one of
the methyls necessarily exists as pseudo-axial in the boat
4.1. General procedure for the cyclization of N-sub-
stituted b-amino acids
In a ¯ask provided with magnetic stirrer and condenser was
placed the N-substituted b-amino acid (prepared by the
26
reductive amination method developed by Simpkins ),
the appropriate solvent, 2 mol equiv. of triethylamine and
.5 mol equiv. of phenylphosphonic dichloride. The reac-
1
tion mixture was re¯uxed at the chosen temperature and
time (see Tables 1±3). Concentration in a rotatory evapo-
rator gave the crude product which was puri®ed on a silica
gel column (hexane/AcOEt, 10:0!0:10) to get the product
as a colorless oil, or white crystals (in the case of 3a).
2
4
conformation of unlike-3b.
Finally, it was also gratifying to con®rm that ab initio calcu-
lations at the HF/3-21G level do reproduce the experimental
infrared carbonyl bands for typical b-lactams and cyclo-b-
dipeptides described in this work (Table 4).
4
9
2
1
4
.1.1. (^)-N-Benzyl-4-phenyl-2-azetidinone (rac-2a).
1
3
9% Yield. H NMR (CDCl , 270 MHz) d 2.85 (dd, Jˆ
3
3
.2 Hz, J ˆ14.5 Hz, 1H), 3.32 (dd, Jˆ5.1 Hz, J
ˆ
gem
gem
1
2
3
. Conclusions
4.5 Hz, 1H), 4.38 (dd, Jˆ2.5 Hz, Jˆ5.1 Hz, 1H), 3.74,
0
13
.79 (AB, JˆJ ˆ14.5 Hz, 2H), 7.33 (m, 10H). C NMR
Phenylphosphonic dichloride, PhP(O)Cl , activates (^)-
2
(CDCl , 67.9 MHz) d 44.8, 47.0, 53.6, 126.6, 127.8,
3
(
benzylamino)-3-phenylpropanoic acids, rac-1a±d, towards
1
28.6, 128.8, 129.1, 135.6, 138.0, 167.3. MS, m/z 237
1
21
b-lactamization. 2-Substituted b-amino acid rac-4 is also
converted to b-lactam rac-5 in 75% yield, in the presence
of PhP(O)Cl . In contrast, 3-(benzylamino)-3-methylpropa-
(M ). IR (n cm ) 1747. Pure compound was confronted
max
with H NMR previously reported.
1
27
2
noic acid 6a and (^)-3-(benzylamino)-3-methylpropanoic
acid rac-6b afford cyclo-b-dipeptides 3a and rac-3b,
respectively, as the main products.
4.1.2. (^)-N-p-Phenylbenzyl-4-phenyl-2-azetidinone
(rac-2b). 95% Yield. H NMR (CDCl , 270 MHz) d 2.89
1
3
3
3
(dd, Jˆ2.1 Hz, J ˆ14.7 Hz, 1H), 3.36 (dd, Jˆ5.2 Hz,
gem
0
J
ˆ14.6 Hz, 1H), 3.82, 4.83 (AB, JˆJ ˆ15.0 Hz, 2H),
gem
1
2
X-Ray diffraction analysis of cyclo-b-dipeptides 3a shows a
slightly twisted boat conformation for this interesting
analog of the well-known diketopiperazines, commonly
generated during peptide synthesis from a-amino acids.
Semi-empirical (AM1, MNDO, PM3) and ab initio (HF/
4.45 (dd, Jˆ2.2 Hz, Jˆ5.2 Hz, 1H), 7.2±7.58 (m, 14H).
C NMR (CDCl , 67.9 MHz) d 44.5, 47.0, 53.7, 126.7,
3
1
3
127.1, 127.5, 128.6, 128.9, 129.1, 129.1, 134.6, 138.0,
1
140.7, 140.7, 167.3. MS, m/z 313 (M ). IR (n
21
cm )
max
1
1753. HRMS (FAB): calcd for [M1H] C H NO:
314.1545; found: 314.1552.
2
2
20
3-21G) calculations reproduce with great accuracy the
solid-state conformation of cyclic b-dipeptide 3a.
4
.1.3. (^)-N-o-Methoxybenzyl-4-phenyl-2-azetidinone
1
Ab initio calculations at the HF/3-21G level help in under-
standing the contrasting reactivity behavior of activated
(rac-2c). 72% Yield. H NMR (CDCl , 270 MHz) d 2.78
3
3 3
(dd, Jˆ1.5 Hz, J ˆ14.6 Hz, 1H), 3.28 (dd, Jˆ5.2 Hz,
gem
(
phosphorylated) b-amino acid derivatives A-1a and
J
ˆ14.6 Hz, 1H), 3.62 (s, 3H), 3.96, 4.67 (AB, Jˆ
gem
0
7.1±7.32 (m, 9H). C NMR (CDCl , 67.9 MHz) d 39.9,
1
2
A-6a. The former is estimated to adopt a `folded' confor-
mation that should facilitate b-lactam formation, whereas
the latter prefers an `extended' conformation that apparently
inhibits b-lactamization. Indeed, modeling of the activated
open-chain b-dipeptide B-7 provides a `folded' confor-
mation that should favor cyclization to give cyclo-b-dipep-
tide 3a, as experimentally observed. Most interestingly, the
J ˆ14.6 Hz, 2H), 4.39 (dd, Jˆ2.2 Hz, Jˆ5.2 Hz, 1H),
1
3
3
47.0, 54.3, 55.1, 110.3, 120.6, 123.8, 126.4, 128.2, 128.8,
129.3, 130.5, 138.8, 157.5, 167.4. MS, m/z 267 (M ). IR
1
2
(nmax cm ) 1753. HRMS (FAB): calcd for [M1H]
1
1
C H NO : 268.1338; found: 268.1345.
1
7
18
2
`folded' conformation of activated dipeptide B-7 is stabi-
lized by hydrogen-bonding between the amino N-H group
4.1.4. (^)-N-n-Pentyl-4-phenyl-2-azetidinone (rac-2d).
1
62% Yield. H NMR (CDCl , 270 MHz) d 0.79 (t, Jˆ
3

J. Escalante et al. / Tetrahedron 57 (2001) 1883±1890
1889
6
2
.8 Hz, 3H), 1.20 (m, 4H), 1.39 (q, Jˆ7.1 Hz, 2H), 2.76 (m,
Lukacs, G.; Springer: Berlin, 1993; Vol. 2, pp 621. For the
preparation of b-lactams from b-amino acids: (e) Huang, H.;
Iwasawa, N.; Mukaiyama, T. Chem. Lett. 1984, 1465.
(f) Kumieda, T.; Nagamatsu, T.; Kiguchi, T.; Hirobe, M.
Tetrahedron Lett. 1988, 29, 2203. (g) Tanner, D.; Somfai, P.
Tetrahedron 1988, 44, 613. (h) Mayachi, N.; Shibasaki, M.
J. Org. Chem. 1990, 55, 1975. (i) Murayama, T.; Kobayashi,
T.; Miura, T. Tetrahedron Lett. 1995, 36, 3703.
1
2
H), 3.32 (m, 2H), 5.0 (dd, Jˆ2.2 Hz, Jˆ4.9 Hz, 1H), 7.27
(
m, 5H). MS, m/z (relative intensity) 217 (85.4), 178 (26),
2
1
146 (33), 122 (58.5), 107 (100). IR (n
HRMS (FAB): calcd for [M1H] C H NO: 218.1545;
found: 218.1553.
cm ) 1728.
max
1
1
4
20
4
.1.5. (^)-N-3-Dibenzyl-2-azetidinone (rac-5). 735%
1
Yield. H NMR (CDCl , 399.7 MHz) d 3.0 (dd, Jˆ
3. (a) Seebach, D.; Matthews, J. L. Chem. Commun. 1997, 2015
(and references cited therein). (b) Gellman, S. H. Acc. Chem.
Res. 1998, 31, 173 (and references cited therein).
(c) Gademann, K.; Hintermann, T.; Schreiber, J. V. Curr.
Med. Chem. 1999, 6, 905.
3
3
5
1
7
4
1
1
2
.0 Hz, Jgemˆ14.5 Hz, 2H), 3.1 (dd, Jˆ5.1 Hz, J
1
.2 (m, 10H). C NMR (CDCl , 100.5 MHz) d 34.2,
ˆ
gem
4.3 Hz, 2H), 3.5 (m, 1H), 4.2±5.0 (d, Jˆ15.0 Hz, 2H),
3
3
3.9, 45.7, 50.7, 126.6, 127.6, 127.9, 128.6, 128.7, 129.1,
1
35.5, 138.0, 169.7. MS, m/z 251 (M ). IR (n
21
cm )
max
4. Palomo, C.; Aizpurua, J. M.; Cuevas, C. J. Chem. Soc., Chem
Commun. 1994, 1957.
1
740. HRMS (FAB): calcd for [M1H] C H NO:
52.1388; found: 252.1390.
1
7
18
5. Samy, R.; Kim, H. K.; Brady, M.; Toogood, P. L. J. Org.
Chem. 1999, 64, 2711.
4
3
.1.6. Perhydro-1,5-(N,N-dibenzyl)-diazocine-2,6-dione
6
. Seebach, D.; Gademann, K.; Ernst, M.; Hoyer, D. Angew.
Chem., Int. Ed. Engl. 1999, 38, 1223.
7. Veda, M.; Mori, H. Bull. Chem. Soc. Jpn 1992, 65, 1636.
1
a. 49% Yield. Colorless crystals, mpˆ161±1628C. H
NMR (CDCl , 300 MHz) d 2.95 (t, Jˆ6.9 Hz, 2H), 3.54
3
1
3
(
t, Jˆ6.9 Hz, 2H) 4.61 (s, 2H), 7.31 (m, 5H). C NMR
8
. Palomo, C.; Aizpurua, J. M.; Urchegui, R.; Iturburu, M.;
Ochoa, A.; Cuevas, C. J. Org. Chem. 1991, 56, 2244.
(
CDCl , 75.4 MHz) d 36.7, 42.4, 49.3, 128.1, 128.7,
3
1
21
129.1, 137.2, 170.6. MS, m/z 322 (M ). IR (nmax cm )
1639 and 1742. HRMS (FAB): calcd for [M1H]
9. (a) Kim, S.; Chang, S. B.; Lee, P. H. Tetrahedron Lett. 1987,
8, 2735. (b) Kim, S.; Lee, P. H.; Lee, T. A. J. Chem. Soc.,
Chem. Commun. 1988, 1242.
1
2
C H N O : 322.1681; found: 322.1700. X-Ray crystallo-
2
graphic structure in Fig. 1.
0
22
2
2
2
8
1
0. Aromatic p±p interactions facilitating cyclizations; e.g.
1
2a
12b
Diels±Alder reactions
and Claisen Rearrangements
4.1.7. (^)-Perhydro-1,5-(N,N-dibenzyl)-4,8-dimethyl-
diazocine-2,6-dione (rac-3b). 68% Yield. H NMR
have been postulated before: (a) Siegel, C.; Thornton, E. R.
J. Am. Chem. Soc. 1989, 111, 5722. (b) Kallmerten, J.; Gould,
T. J. J. Org. Chem. 1986, 51, 1152.
2
9
1
(
CDCl , 270 MHz) d 1.17 (d, Jˆ5.9 Hz, 3H), 2.50 (dd,
3
3
3
Jˆ2.2 Hz, J ˆ14.3 Hz, 1H), 3.03 (dd, Jˆ4.9 Hz,
gem
11. Nagao, Y.; Kumagai, T.; Tamai, S.; Matsunaga, H.; Abe, T.;
Inoue, Y. Heterocycles 1996, 42, 849.
12. (a) Rothe, M.; Timler, R. Chem. Ber. 1962, 95, 783.
J
ˆ14.5 Hz, 1H), 3.54 (m, 1H), 4.06, 4.55 (AB, Jˆ
gem
0
7.9 MHz) d 18.6, 44.2, 44.4, 47.1, 127.7, 128.3, 128.8,
13
J ˆ15.2 Hz, 2H), 7.24 (m, 5H). C NMR (CDCl ,
3
6
1
1
3
(
b) Aversa, M. C.; Bonaccorsi, P.; Giannetto, P.; Beagley,
1
36.1, 166.9. MS, m/z 350 (M ). IR (n
21
cm ) 1641 and
max
B.; Leigh, D. A.; Pritchard, R.; Truscello, A. M. J. Heterocycl.
Chem. 1992, 29, 317. (c) Sutton, P. W.; Bradley, A.; Farr a s, J.;
Romea, P.; Urp Âõ , F.; Vilarrasa, J. Tetrahedron 2000, 56, 7947.
3. Gordon, A. J.; Ford, R. A. The Chemist's Companion; Wiley:
New York, 1972; pp 196, 200.
1
747. HRMS (FAB): calcd for [M1H] C H N O :
51.2073; found: 351.2066.
2
2
27
2
2
1
1
1
Acknowledgements
4. Full experimental and spectroscopic data for 7±9 will be
reported separately.
5. For a de®nition of the like/unlike stereochemical descriptors:
We are indebted to the editor, Professor S. F. Martin, and the
reviewers for many important observations. The ®nancial
support from Conacyt-M e xico, via grants 3559P-E9607
and 33023-E2000, is gratefully acknowledged.
(
a) Seebach, D.; Prelog, V. Angew. Chem., Int. Ed. Engl. 1982,
1, 654. (b) Juaristi, E. In Introduction to Stereochemistry and
2
Conformational Analysis; Wiley: New York, 1991; pp 52±54.
6. Moore, J. A.; Anet, F. A. L. In Comprehensive Heterocyclic
Chemistry; Katritzky, A. R., Rees, C. W., Eds.; Pergamon:
Oxford, 1984; Vol. 7, pp 783.
1
References
1
. For reviews: (a) Drey, C. N. C. In Chemistry and Biochemistry
of the Amino Acids; Barret, G. C., Ed.; Chapman and Hall:
London, 1985; pp 25. (b) Juaristi, E.; Quintana, D.; Escalante,
J. Aldrichim. Acta 1994, 27, 3. (c) Cole, D. C. Tetrahedron
17. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T.;
Petersson, G. A.; Montgomery, J. A.; Aghavachari, K.;
Al-Laham, M. A.; Zakrzewski, V. G.; Ortiz, J. V.; Foresman,
J. B.; Cioslowski, J.; Stefanov, B. B.; Nanayakkara, A.;
Challacombe, M.; 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 94 (Revision D.4); Gaussian: Pittsburgh, PA, 1995.
18. Hehre, W. J.; Schleyer, P. v. R.; Pople, J. A. Ab Initio
Molecular Orbital Theory; Wiley: New York, 1986.
1
994, 50, 9517. (d) Cardillo, G.; Tomasini, C. Chem. Soc. Rev.
996, 23, 117. (e) In Enantioselective Synthesis of b-Amino
1
Acids; Juaristi, E., Ed.; Wiley-VCH: New York, 1997.
f) Juaristi, E.; L o pez-Ruiz, H. Curr. Med. Chem. 1999, 6, 983.
(
2
. For reviews: (a) The Chemistry of b-Lactams; Page, M. I., Ed.;
Chapman and Hall: London, 1992. (b) Craig, W. A.; Ebert,
S. C. Antimicrob. Agents Chemother. 1992, 36, 2577. (c) In
The Organic Chemistry of b-Lactams; Georg, G. I., Ed.;
Verlag Chemie: New York, 1993. (d) Southgate, R.; Branch,
C.; Coulton, S.; Hunt, E. In Recent Progress in the Chemical
Synthesis of Antibiotics and Related Microbial Products;
19. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.;
Montgomery, J. A.; Stratmann, R. E., Jr.; Burant, J. C.;

1
890
J. Escalante et al. / Tetrahedron 57 (2001) 1883±1890
Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.;
23. Smith III, A. B.; Ducry, L.; Corbett, R. M.; Hirschmann, R.
Org. Lett. 2000, 2, 3887.
Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.;
Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford,
S.; Ochterski, J.; Peterson, G. A.; Ayala, P. Y.; Cui, Q.; Ortiz,
J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox,
D. J.; Keith, D. J.; Al-Lahman, M. A.; Peng, C. Y.;
Nanyakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M.
W.; Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.;
Gonzalez, C.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A.
gaussian 98 (Revision A.7); Gaussian, Pittsburgh, PA, 1998.
24. Conformational Analysis of Medium-sized Heterocycles;
Glass, R., Ed.; VCH: New York, 1988.
25. For general experimental procedures, see Juaristi, E.;
Balderas, M.; Ramõ
Ârez-Quir o s, Y. Tetrahedron: Asymmetry
1
998, 9, 3881.
2
2
2
2
6. Simpkins, N. S.; Cain, C. M.; Cousins, R. P. C.; Coumbarides,
G. Tetrahedron 1990, 46, 523.
7. Barrow, K. D.; Spotswood, T. M. Tetrahedron Lett. 1965, 37,
3
325.
8. Crystallographic data are deposited at Cambridge Crystallo-
graphic Data Center (CDDC 152942).
9. Relative con®guration of rac-3b not yet securely established
2
2
0. Rivail, J. L.; Rinaldi, D. Chem. Phys. 1976, 18, 233.
1. (a) Onsager, L. J. Am. Chem. Soc. 1938, 60, 1486. (b) Wong,
M. W.; Frisch, M.; Wiberg, K. B. J. Am. Chem. Soc. 1991,
(
see text).
1
13, 4776. (c) Miertus, S.; Tomassi, J. Chem. Phys. 1982, 65,
39.
2
2
2. Optimized structures of A-1a, A-6a, and A-6b proved rela-
tively insensitive to changes in solvent polarity.