Reaction between N-alkylhydroxylamines and chiral enoate esters: More experimental evidence for a cycloaddition-like process, a rationale based on DFT theoretical calculations, and stereoselective synthesis of new enantiopure β-amino acids

Article abstract of DOI:10.1021/jo0159082

The reactions between N-benzyl- and N-methylhydroxylamine and chiral enoate esters, derived from D-glyceraldehyde and (-)-verbenone, respectively, have been investigated. Theoretical calculations show that the most favorable mechanism involves the concert

Full text of DOI:10.1021/jo0159082

2402
J . Org. Chem. 2002, 67, 2402-2410
Articles
Rea ction betw een N-Alk ylh yd r oxyla m in es a n d Ch ir a l En oa te
Ester s: Mor e Exp er im en ta l Evid en ce for a Cycloa d d ition -lik e
P r ocess, a Ra tion a le Ba sed on DF T Th eor etica l Ca lcu la tion s, a n d
Ster eoselective Syn th esis of New En a n tiop u r e â-Am in o Acid s
Albertina G. Moglioni,†,‡ Elena Muray,† J ose´ A. Castillo,† AÄ ngel AÄ lvarez-Larena,§
Graciela Y. Moltrasio,‡ Vicenc¸ Branchadell,*,† and Rosa M. Ortun˜o*,†
Departament de Quı´mica, Universitat Auto`noma de Barcelona, 08193 Bellaterra, Barcelona, Spain,
Unitat de Cristal.lografia, Universitat Auto`noma de Barcelona, 08193 Bellaterra, Barcelona, Spain, and
Departamento de Quı´mica Orga´nica, Facultad de Farmacia y Bioqu´ımica, Universidad de Buenos Aires,
1113 Buenos Aires, Argentina
rosa.ortuno@uab.es
Received J uly 10, 2001
The reactions between N-benzyl- and N-methylhydroxylamine and chiral enoate esters, derived
from D-glyceraldehyde and (-)-verbenone, respectively, have been investigated. Theoretical
calculations show that the most favorable mechanism involves the concerted cycloaddition of the
hydroxylamine to the substrate. This result is in good agreement with the stereospecificity observed
when the trisubstituted olefins are used. The open-chain adducts have been isolated when the
processes are carried out at low temperatures and for short reaction times. These compounds evolve
to the corresponding isoxazolidinones on standing at room temperature or under acid catalysis.
The high π-facial diastereoselection has been rationalized on the basis of steric effects induced by
the dioxolane ring for D-glyceraldehyde derivatives or by the cyclobutane gem-dimethyl substitution
for esters prepared from (-)-verbenone. As an application of these reactions, new â-amino acids
have been synthesized in a highly efficient and stereocontrolled manner.
In tr od u ction
acids,9 among other interesting products. In 1976, H.
Stamm10 studied the addition of N-methylhydroxylamine
to ethyl crotonate and stated, for the first time, that the
process takes place through the 1,4-conjugate addition
or Michael-type reaction (Scheme 1, pathway a) of the
hydroxylamine to afford a â-hydroxyamino ester (P 3)
which was isolated and identified by its spectroscopic
data. These authors confirmed the role of such a com-
pound as an intermediate in the isoxazolidinone forma-
tion since it spontaneously cyclized on standing for
several days in chloroform solution. On the basis of the
acidity of N-methylhydroxylamine (pKa 4.6), the addition
was believed to occur assisted by the coordination of the
hydroxyl proton to the carbonyl.10 This coordination was
also invoked by S. W. Baldwin to explain the stereose-
lectivity observed in the reactions between (R-methyl-
benzyl)hydroxylamine and â-substituted acrylate esters,
where neither the double bond geometry nor the size of
the â-substituent seemed to have any effect on the
diastereoselectivity of the overall process.2 In that work,
it was suggested that the reaction is irreversible since
isoxazolidinones do not equilibrate under the reaction
conditions (refluxing benzene in the presence of potas-
sium carbonate) and, moreover, pure open-chain inter-
mediates lead to single isoxazolidinones under these
conditions.
Since J . E. Baldwin reported in 1984 that the addition
of N-substituted hydroxylamines to R,â-unsaturated
esters is a general procedure for the synthesis of isox-
azolidin-5-ones,1 the reaction between N-alkylhydroxy-
lamines and conjugated esters,2-4 lactones,5,6 or lactams7
has been employed in the synthesis of isoxazolidinyl
nucleoside analogues,3,4,8 carbapenems,5a and â-amino
† Departament de Qu´ımica, Universitat Auto`noma de Barcelona.
E-mail: (V.B.) vicenc@klingon.uab.es.
‡ Universidad de Buenos Aires. E-mail: (G.Y.M.) gmoltra@ffyb.uba.ar.
§ Unitat de Cristal.lografia, Universitat Auto`noma de Barcelona.
E-mail: angel.alvarez@uab.es.
(1) Baldwin, J . E.; Harwood, L. M.; Lombard, M. J . Tetrahedron
1984, 40, 4363.
(2) Baldwin, S. W.; Aube´, J . Tetrahedron Lett. 1987, 28, 179.
(3) (a) Xiang, Y.; Gi, H.-J .; Niu, D.; Schinazi, R. F.; Zhao, K. J . Org.
Chem. 1997, 62, 7430. (b) Niu, D.; Zhao, K. J . Am. Chem. Soc. 1999,
121, 2456.
(4) (a) Merino, P.; Franco, S.; Merchan, F. L.; Tejero, T. Tetrahe-
dron: Asymmetry 1998, 9, 3945. (b) Merino, P.; Franco, S.; Merchan,
F. L.; Tejero, T. J . Org. Chem. 2000, 65, 5575.
(5) (a) Maciejewski, Panfil, I.; Belzecki, C.; Chmielewski, M. Tetra-
hedron 1992, 48, 10363. (b) J urczak, M.; Chmielewski, M. Tetrahedron
Lett. 1995, 36, 135. (b) J urczak, M.; Socha, D.; Chmielewski, M.
Tetrahedron 1996, 52, 1411. (c) Frelek, J .; Panfil, I.; Gluzinski, P.;
Chmielewski, M. Tetrahedron: Asymmetry 1996, 7, 3415. (d) Panfil,
I.; Abramski, W.; Chmielewski, M. Carbohydr. Chem. 1998, 17, 1395.
(6) Pan, S.; Wang, J .; Zhao, K. J . Org. Chem. 1999, 64, 4.
(7) Langlois, N.; Calvez, O.; Radom, M.-O. Tetrahedron Lett. 1997,
38, 8037.
(8) Pan, S.; Amankulor, N. M.; Zhao, K. Tetrahedron report number
454. Tetrahedron 1998, 54, 6587.
(9) Ishikawa, T.; Nagai, K.; Kudoh, T.; Saito, S. Synlett 1998, 1291.
(10) Stamm, H.; Steudle, H. Tetrahedron Lett. 1976, 3607.
10.1021/jo0159082 CCC: $22.00 © 2002 American Chemical Society
Published on Web 03/16/2002

Reaction of N-Alkylhydroxylamines and Enoate Esters
J . Org. Chem., Vol. 67, No. 8, 2002 2403
Sch em e 1
Sch em e 2
substituent attached to the stereogenic centers already
present in the substrates. More difficult is the rational-
ization and the prediction of the diastereoselection in
flexible open-chain molecules. For instance, P. Merino
reported the influence on the π-facial diastereoselectivity
exerted by the Z/E configuration of the double bond and
by the protecting groups of the 1,2-amino alcohol subunit
in substrates derived from L-serine.4 In turn, K. Zhao
described the reaction of disubstituted chiral alkenoates,
derived from D-glyceraldehyde, with N-methylhydroxy-
lamine to give syn adducts as major isomers independent
of the Z/E geometry of the double bond.3a This result is
in agreement with the previous work by S. W. Baldwin.2
However, the factors determining the stereocontrol have
not been systematized.
In this paper, we report our results on the study of the
reaction of N-alkylhydroxylamines with chiral R,â-
unsaturated esters to rationalize its mechanism and to
establish its scope to produce optically pure isoxazolidi-
nones in a stereocontrolled manner. Thus, the reactions
between N-benzylhydroxylamine and di- or trisubstituted
(Z)- and (E)-alkenoates (Scheme 2), derived from D-
glyceraldehyde and (-)-verbenone, respectively, have
been investigated. DFT theoretical calculations have been
undertaken to ascertain whether a cycloaddition-like
process is more favorable than a two-step mechanism.
Moreover, the stereocontrol due to a dioxolane or a
cyclobutyl moiety as internal inductor of the π-facial
diastereoselection has been shown and justified both by
For trisubstituted olefins, if the addition takes place
through zwitterion P 1 (Scheme 1, path a) or through
intermediate P 2 (path b), the stereoselectivity in the
production of hydroxyester P 3 and, consequently, of
isoxazolidinone P 4 would be dependent on the stereo-
chemical outcome of the proton-transfer processes. Re-
cently, K. Zhao has postulated a concerted mechanism
through a cyclic transition state for the addition of
N-methylhydroxylamine to conjugated esters (Scheme 1,
pathway c). Accordingly, the OH proton would be in-
tramolecularly transferred to the R-carbonyl carbon,
giving a zwitterion, P 5, which tautomerizes fast to the
corresponding â-hydroxyamino ester with retention of the
Z/E configuration of the starting alkene. This hypothesis
was supported by elegant deuteration experiments and
by the observation that the Z/E relative stereochemistry
remained unaltered in isoxazolidinones when achiral
trisubstitued alkenoates were used.3b
The combination of its stereocontrolled outcome with
the use of chiral substrates or reagents makes this
reaction extremely attractive for the stereoselective
synthesis of optically pure products of biological interest,
and therefore, efforts directed to elucidate the origin of
the diastereoselection have been made. When the double
bond is located in conformationally constrained cyclic
molecules such as unsaturated lactones5,6 or lactams,7 the
stereochemistry of the major products is easily rational-
ized as the result of the anti attack with respect to the

2404 J . Org. Chem., Vol. 67, No. 8, 2002
Moglioni et al.
Sch em e 3a
a
Reagents and conditions: (a) H2, Pd(OH)2/C, H2O/EtOH, 4 atm.
experiments and by calculations. Factors influencing the
cyclization of the intermediates have also been explored.
Finally, the synthetic usefulness of the obtained chiral
isoxazolidinones has been evidenced in the preparation
of enantiopure new â-amino acids. These compounds
contain one or two new stereogenic centers created in the
addition step with determined absolute configuration.
In preliminary experiments, dichloromethane solutions
of compounds (Z)- and (E)-1 were reacted in separate
experiments with 1.2 equiv of N-benzylhydroxylamine
hydrochloride, in the presence of triethylamine. In both
cases, the yield and stereoselectivity were similar. There-
fore, mixtures of Z/E olefinic isomers were used in further
experiments. â-Hydroxyamino esters 7 (Scheme 3) were
the sole products detected when the reaction between 1
and N-benzylhydroxylamine was performed at -20 °C
overnight. These compounds cyclized smoothly on stand-
ing to give isoxazolidinones 8. For instance, a 5:3.5
mixture of 7 and 8 resulted after 1 day at room temper-
ature, conversion being complete after several days.
Attemps to accelerate the cyclization by stirring the
mixture in the presence of ZnCl2, as previously de-
scribed,4,6 failed. This result is in accordance with that
reported by Merino.4 When the reaction of 1 with N-
benzylhydroxylamine was performed at room tempera-
ture for 17 h, a mixture of isomeric syn/anti-7, contami-
nated with some amount of 8, was obtained after the
usual treatment (see the Experimental Section).15 Chro-
matography on silica gel afforded the isoxazolidinones
syn-8 and anti-8 as the only products which were fully
characterized. We concluded that acid silica gel catalyzed,
therefore, the cyclization of 7 into 8. On the contrary,
chromatography on neutral Baker silica allowed the
isolation of syn-7 and anti-7, which were characterized
by their spectroscopic data.
Resu lts a n d Discu ssion
1. Rea ction s betw een N-Ben zylh yd r oxyla m in e
a n d Alk en oa tes 1-4. Syn th esis of Ch ir a l Oxa zoli-
d in on es. Alkenoates 1 (Z- and E-isomers) (Scheme 2)
are commercially available. Alternatively, they can be
easily synthesized from D-glyceraldehyde through Wittig
condensation.11 Alkenoates 2 (Z and E isomers) were
prepared according to the procedures described in the
literature.12 Condensations of aldehydes 5 and 6, ob-
tained from (-)-verbenone,13 with the appropriate phos-
phonates afforded the new compounds (Z)- and (E)-3(4)
as shown in Scheme 2. Thus, reaction of 5 with Still’s
reagent, (CF3CH2O)2POCH2CO2CH3,14 in the presence of
sodium hydride afforded stereoselectively Z-isomers of
3 and 4, repectively, in 60% yield. Alternatively, conden-
sation of 5 with the anion derived from dimethyl (meth-
oxycarbonylmethyl)phosphonate provided E-isomers in
55-60% yield. A stereoisomeric mixture resulted from
the reactions between aldehyde 5 or 6 and methyl
(triphenylphosphoranylidene)acetate, the E-isomer being
the major product.
In these reactions, the yield of the adducts was 75-
90% and the syn/ anti ratio was about 10:1 as determined
1
by HPLC and H NMR analysis of the reaction crudes.
(11) Mann, J .; Partlett, N. K.; Thomas, A. J . Chem. Res., Synop.
1987, 369 and references therein.
(12) Ibuka, T.; Akimoto, N.; Tanaka, M.; Nishii, S.; Yamamoto, Y.
J . Org. Chem. 1989, 54, 4055.
(13) Moglioni, A. G.; Garc´ıa-Expo´sito, E.; Aguado, G. P.; Parella, T.;
Branchadell, V.; Moltrasio, G. Y.; Ortun˜o, R. M. J . Org. Chem. 2000,
65, 3934.
(15) In contrast with the results previously reported by Zhao,3a
examination of the crude resulting from the reaction of (Z)-1 with
N-methylhydroxyalmine at room temperature for 17 h, performed in
our laboratory, revealed the presence of the corresponding oxazolidi-
none (as a 10:1 syn/anti mixture) as the only product (89% yield), the
open-chain â-hydroxylamino esters not being detected by careful 1H
NMR analysis.
(14) Stille, C. W.; Gennari, C. Tetrahedron Lett. 1983, 24, 4405.

Reaction of N-Alkylhydroxylamines and Enoate Esters
J . Org. Chem., Vol. 67, No. 8, 2002 2405
F igu r e 1. Structures of compounds 9, 10, and 14 as determined by X-ray structural analysis.
Sch em e 4
F igu r e 2. Newman projection for the active conformers
related to unsaturated esters 3 and 4. The arrow marks the
preferential attack of the hydroxylamine on the double bond.
See also ref 17.
lished by X-ray analysis (Figure 1) and can be rational-
ized as the result of a preferential orientation of the
attack on the (C2-re)-face of the double bond induced by
the gem-dimethyl substitution of the cyclobutane (Figure
2). This π-facial diastereoselection has also been observed
in the cycloaddition of diazomethane to other related
cyclobutyl alkenoates and rationalized on the basis of
conformational bias for these molecules.17 By extension,
the configuration of 15 was assigned as depicted in
Scheme 4.
2. Ster eoselective Syn th esis of â-Am in o Acid s.
Isoxazolidinones 8-10, and 15, are precursos to new
enantiopure â-amino acids through hydrogenolysis of the
N-O bond concomitant to the benzyl group removal.
Thus, hydrogenation of syn-8 by using palladium hydrox-
ide on charcoal as catalyst, in 7:1 H2O/EtOH, under 4
atm of pressure, afforded product 11 in 90% yield. In a
similar manner, hydrogenation of 9 and 10 produced
almost quantitatively the R-methyl-â-amino acids 12 and
13, respectively, as hygroscopic solids (Scheme 3). Finally,
catalytic hydrogenation of 15 provided the γ-cyclobutyl-
â-amino acid 16 in 90% yield (Scheme 4).
In this way, reactions of N-benzylhydroxylamine with
several chiral enoates and subsequent N-O reduction
resulted in an efficient method to synthesize stereose-
lectively different types of â-amino acids.
3. Th eor etica l Ca lcu la tion s. a . Rea ction of Hy-
d r oxyla m in e w ith Meth yl Acr yla te. Calculations have
been done according to the procedures described in the
Computational Details of the Experimental Section. First,
we have studied the attack of hydroxylamine on methyl
acrylate (MA) as a model system. Figure 3 and Table 1
The syn configuration was assigned to the major diaste-
reoisomer according to our previous experience on the
predominant π-facial diastereoselection involved in other
cycloaddition reactions of these alkenoates,16 and in
agreement with the results reported by Zhao on the
reactions of 1 with N-methylhydroxylamine.3a
Trisubstituted olefins 2 did not react under the condi-
tions described above for the parent compound 1. Nev-
ertheless, stereospecific additions were accomplished
under more energetic conditions. Thus, (Z)-2 was made
to react with N-benzylhydroxylamine hydrochloride in
the presence of excess sodium ethoxide in boiling ethanol
for 15 h. After column chromatography on silica gel,
crystalline isoxazolidinone 9 was obtained in 59% yield,
in a single syn isomeric form, as the only defined product.
Similarly, the diastereomeric compound 10 was synthe-
sized from (E)-2 in 55% yield. The configuration of these
compounds was assigned by X-ray structural analysis
(Figure 1), showing the excellent stereocontrol in the
creation of the new stereogenic centers.
The reactions of cyclobutyl alkenoates 3 and 4 were
also explored. E isomers reacted faster than Z ones, but
both stereoisomers converged into isoxazolidinones 14
and 15, respectively (65-70% yield), by treatment with
N-benzylhydroxylamine in dichloromethane at room tem-
perature overnight (Scheme 4). Open-chain intermediates
were not detected. The configuration of 14 was estab-
(16) (a) Casas, R.; Parella, T.; Branchadell, V.; Oliva, A.; Ortun˜o,
R. M.; Guingant, A. Tetrahedron 1992, 48, 2659. (b) Mart´ın-Vila`, M.;
Hanafi, N.; J ime´mez, J . M.; AÄ lvarez-Larena, A.; Piniella, J . F.;
Branchadell, V.; Oliva, A.; Ortun˜o, R. M. J . Org. Chem. 1998, 63, 3581.
(c) Muray, E.; AÄ lvarez-Larena, A.; Piniella, J . F.; Branchadell, V.;
Ortun˜o, R. M. J . Org. Chem. 2000, 65, 388.
(17) Moglioni, A. G.; Garc´ıa-Expo´sito, E.; AÄ lvarez-Larena, A.; Bran-
chadell, V.; Moltrasio, G. Y.; Ortun˜o, R. M. Tetrahedron: Asymmetry
2000, 11, 4903.

2406 J . Org. Chem., Vol. 67, No. 8, 2002
Moglioni et al.
F igu r e 3. Optimized geometries of the stationary points
corresponding to the reaction between hydroxylamine and MA.
Selected interatomic distances are in angstroms.
Ta ble 1. Rela tive En er gies a n d Th er m od yn a m ica l
P a r a m eter s a t 298.15 K a n d 1 a tm for th e Sta tion a r y
P oin ts Cor r esp on d in g to th e Rea ction of NH2OH w ith
Meth yl Acr yla te
F igu r e 4. Optimized geometries of the Z and E isomers of
alkenoate 1. τ is the C1-C2-C3-C4 dihedral angle.
b. Rea ction s of Hyd r oxyla m in e a n d N-Meth ylh y-
d r oxyla m in e w ith Alk en oa tes (Z)-1 a n d (E)-1. Figure
4 presents the optimized geometries of (Z)-1 and (E)-1.
We have considered several conformations arising from
the rotation around the C1-C2 and C3-C4 bonds. In
both geometric isomers, an s-cis arrangement of the
carbonyl group is the preferred one. Regarding the
rotation around C3-C4, for (Z)-1 there is only one energy
minimum in which the C4-H bond is nearly eclipsed
with the C3-C2 double bond. On the other hand, two
different conformational minima have been located for
(E)-1. In the most stable one ((E)-1a ) the C3-C2 double
bond is eclipsed with the C4-O bond. The other con-
former ((E)-1b) is only 0.2 kcal mol-1 higher in energy
than (E)-1a , and it presents a C4-H eclipsed arrange-
ment, similar to the one corresponding to (Z)-1.
We have located the transition states corresponding
to the syn and anti attacks of hydroxylamine and N-
methylhydroxylamine on (Z)-1 and (E)-1. For (E)-1, we
have considered the attack of hydroxylamine on conform-
ers a and b. In all cases, the lower-energy transition-
state structures correspond to the attack on (E)-1b.
Table 2 presents the values of selected geometry
parameters of the transition states, potential energy
barriers, and activation thermodynamic parameters for
the reactions of hydroxylamine and N-methylhydroxy-
lamine with (Z)-1 and (E)-1. The structures of the
transition states for the reactions of N-methylhydroxy-
lamine are shown in Figure 5.
structa
∆Eb
∆Hb
∆Sc
∆Gb,d
TS(P 2)
P 2
TS(P 5)
P 5
17.8
6.5
11.1
1.6
18.5
8.3
10.2
3.7
-41.8
-42.1
-43.0
-40.8
31.0 (31.0)
20.9 (20.9)
23.0 (20.6)
15.9 (12.0)
a
b
d
See Figure 3. In kcal mol-1
.
c In cal K-1 mol-1
.
In paren-
theses are given the values in CH2Cl2 solution (ꢀ ) 8.93).
summarize the results obtained. We have optimized the
geometries of P 2 (Scheme 1, pathway b) and P 5 (pathway
c). The formation of these products takes place in one
step through transition states TS(P 2) and TS(P 5),
respectively. The transition vector of TS(P 2) shows that
the main component of the reaction coordinate is the
formation of the C-N bond. On the other hand, the main
component of the transition vector of TS(P 5) is the
proton transfer from the OH group of hydroxylamine to
the olefin. Table 1 shows that the formation of P 5 is
kinetically much more favorable than the formation of
P 2. Moreover, when the effect of solvation by CH2Cl2 is
taken into account in the gas-phase optimized geom-
etries, the preference for the formation of P 5 is further
increased. We have also tried to optimize the geometry
of several conformers of the zwitterionic intermediate P 1
(Scheme 1, pathway a), but all attempts failed, even when
the effect of solvation by CH2Cl2 was considered in the
geometry optimization. All these results agree with the
concerted mechanism postulated by Zhao3b (Scheme 1,
pathway c). Consequently, this is the only mechanism
that we have considered for the reactions of (Z)-1 and
(E)-1.
The values of the potential energy barriers show that
for (Z)-1 the syn attack is slightly preferred over the anti
Ta ble 2. Selected Geom etr y P a r a m eter s,a P oten tia l En er gy Ba r r ier s, a n d Activa tion Th er m od yn a m ica l P a r a m eter s a t
298.15 K a n d 1 a tm for th e Tr a n sition Sta tes Cor r esp on d in g to th e Rea ction s of R-NH-OH w ith (Z)-1 a n d (E)-1
R
TSb
C3-N
C2-H
O-H
τc
ωd
∆EqThinSpacee
∆HqThinSpacee
∆SqThinSpacef
∆GqThinSpacee
H
Z-syn
Z-anti
E-syn
E-anti
Z-syn
Z-anti
E-syn
E-anti
1.88
1.86
1.87
1.88
1.88
1.86
1.84
1.87
1.45
1.50
1.48
1.52
1.45
1.52
1.51
1.57
1.19
1.16
1.18
1.15
1.19
1.15
1.15
1.12
82.3
67.7
89.8
60.5
81.3
67.3
86.7
61.2
-1.3
-9.5
6.8
-7.3
5.4
-13.3
13.7
-12.3
11.0
11.3
9.9
11.3
9.4
9.7
7.6
9.1
10.0
10.8
9.0
10.7
8.5
9.3
7.0
9.0
-44.2
-50.7
-44.2
-47.6
-46.6
-53.0
-45.9
-49.9
23.2
25.9
22.2
24.9
22.4
25.1
20.7
23.9
Me
a
b
d
Distanes are in angstroms and dihedral angles in degrees. See Figure 5. c C5-C4-C3-C2 dihedral angle. C2-C3-N-O dihedral
angle. e In kcal mol-1 f In cal K-1 mol-1
. .

Reaction of N-Alkylhydroxylamines and Enoate Esters
J . Org. Chem., Vol. 67, No. 8, 2002 2407
F igu r e 5. Optimized geometries of the transition states of
the reactions of N-methylhydroxylamine with (Z)-1 and
(E)-1.
one. This preference is augmented when activation Gibbs
energies are considered. Regarding the reactions of (E)-
1, the potential energy barriers are also lower for the syn
attack, but the energy difference between the syn and
anti transition states is larger than those corresponding
to (Z)-1. In this case, the preference for the syn attack
also increases when activation Gibbs energies are con-
sidered.
Therefore, in all cases, the syn attack is kinetically the
most favorable one, in excellent agreement with the
selectivity experimentally observed. Moreover, the E
isomer is predicted to be slightly more reactive than the
Z isomer.
F igu r e 6. Side views of the transition states of the reactions
of N-methylhydroxylamine with (E)-1 and (E)-1. The dioxolane
group has been omitted for clarity. φC and φH correspond to
the N-C3-C2-C1 and N-C3-C2-H′ dihedral angles, re-
spectively.
(90°) corresponding to the equilibrium geometry of the
reactants. Figure 6 shows that pyramidalization involves
mainly the variation of the dihedral angle corresponding
to the group cis with respect to dioxolane. In this way,
for the Z transition states the attack of the proton takes
place from the same side as the ester group, whereas for
the E transition states the attack is produced from the
other side.
The analysis of the geometries of the transition states
shows that the C3-N bond distance only slightly changes
from one case to another, whereas more significant
differences are observed for the C2-H and H-O dis-
tances. The values of these two distances show that the
degree of proton transfer for the syn transition states is
larger than for the anti transition states. Since this
proton transfer is the main contribution to the transition
vector, the syn transition states appear later along the
reaction coordinate than the anti ones.
When the hydroxylamine molecule approaches the
olefin, steric repulsions between both fragments come
into play. One way to relieve this repulsion is the rotation
around the C3-C4 bond. The comparison between the
values of the τ torsion angle at the olefin equilibrium
geometries (Figure 4) and at the transition states (Table
2) shows that the change of this torsion angle is larger
for the anti transition states than for the syn ones, thus
showing than the anti attack is more sterically demand-
ing than the syn attack. The presence of the methyl group
in N-methylhydroxylamine does not seem to have a
relevant effect. Regarding the ω torsion angle, Table 2
shows that there are deviations from planarity. The
presence of the methyl group of N-methylhydroxylamine
leads to larger torsion angles. For (Z)-1 the torsion
around C3-N moves the hydroxylamine O atom toward
the ester group, whereas for (E)-1, the O atom is moved
away from the ester group. This different behavior of both
geometrical isomers can be related to the pyramidaliza-
tion of C2. Figure 6 presents side views of the transition
states corresponding to the reactions of N-methylhy-
droxylamine with (Z)-1 and (E)-1. The pyramidalization
can be related to the deviation of the φC and φH dihedral
angles with respect to the planar situation (angles about
Con clu d in g Rem a r k s
We have studied by means of theoretical calculations
different mechanistic pathways for the addition of hy-
droxylamine to methyl acrylate. The results obtained
point out that the most favorable mechanism involves a
concerted process through a cyclic transition state. This
result is in agreement with the stereospecificity observed
for additions of N-alkylhydroxylamines to olefinic sub-
strates.
Low temperatures favor the formation of open-chain
adducts which evolve to isoxazolidinones by heating or
during chromatography on acid silica gel. When chiral
alkenoates are used, an excellent π-facial diastereose-
lection has been observed and rationalized. Consequently,
chiral isoxazolidinones have been synthesized from trisub-
stituted olefins in a highly stereocontrolled manner.
These compounds are useful synthetic precursors to
enantiomerically pure new â-amino acids. This feature
has been illustrated herein with several examples.
Exp er im en ta l Section
Alkenoates 111 and 212 (Z and E isomers) as well as
aldehydes 5 and 613 were prepared according to the procedures
described in the literature. Flash column chromatography was
carried out on silica gel (240-400 mesh). Baker silica (40 µm)
was used for the chromatography of acid-sensitive products.
Melting points were determined on a hot stage and are
uncorrected. Standard 1H NMR and 13C NMR spectra were
recorded at 250 and 62.5 MHz, respectively. Chemical shifts

2408 J . Org. Chem., Vol. 67, No. 8, 2002
Moglioni et al.
in NMR spectra are given on the δ scale. Electron impact MS
and HRMS spectra were recorded at 70 eV.
performed at -20 °C overnight, and after solvent removal, the
residue was chromatographed on Baker silica (3:1 hexane/ethyl
acetate). These compounds led, on standing at room temper-
ature, to the corresponding isoxazolidinones 8.
Com p u ta tion a l Deta ils. All calculations have been done
using density functional (DFT) methods within the generalized
gradient approximation (GGA). Molecular geometries have
been fully optimized using Becke’s18 functional for exchange
and the correlation functional due to Perdew and Wang19
(BPW91). Molecular geometries have been fully optimized at
this level of calculation using the standard 6-31G(d) basis set.20
Harmonic vibrational frequencies have been calculated for all
structures to characterize them as energy minima (all fre-
quencies are real) or transition states (one and only one
imaginary frequency). These calculations have been done with
the Gaussian-98 program.21 Single-point calculations have
been done for the previously optimized geometries using an
uncontracted Slater-type orbital (STO) triple-ú basis set
supplemented with a set of d polarization functions for C, N,
and O, and with a set of p functions for H (TZP). These
calculations have been done using the ADF program.22 The
reported energy barriers have been calculated with the TZP
basis set, whereas zero-point and thermal corrections to the
energy and entropies have been calculated from frequencies
computed with the 6-31G(d) basis set. For the reaction between
hydroxylamine and methyl acrylate, the effect of solvation by
CH2Cl2 (ꢀ ) 8.93) has been taken into account at the BPW91/
6-31G(d) level of calculation using the conductor-like screening
model23 implemented in the Gaussian-98 program.
Isomeric products 7 were characterized by their 1H NMR
(CDCl3) spectroscopic data as follows: syn-7, 1.36 (s, 3H), 1.40
(s, 3H), 2.47 (dd, J ) 15.5 Hz, J ′ ) 4.8 Hz, 1H), 2.76 (dd, J )
15.5 Hz, J ′ ) 4.8 Hz, 1H), 3.52 (m, 1H), 3.68 (s, 3H), 3.79 (m,
1H), 3.97 (s, 2H), 3.99 (dd, J ) 7.7 Hz, J ′ ) 5.7 Hz, 1H), 4,44
(m, 1H), 5.00 (br s, 1H), 7.30 (br s, 5 H); anti-7, 1.32 (s, 3H),
1.37 (s, 3H), 2.63 (dd, J ) 15.6 Hz, J ′ ) 5.4 Hz, 1H), 2.89 (dd,
J ) 15.6 Hz, J ′ ) 7.9 Hz, 1H), 3.37 (m, 1H), 3.69 (s, 2H), 3.90
(dd, J ) 8.6 Hz, J ′ ) 5.5 Hz, 1H), 4.10 (dd, J ) 8.6 Hz, J ′ )
6.2 Hz, 1H), 4.27 (m, 1H), 4.75 (br s, 1H), 7.30 (br s, 5H).
Da ta for (3R,4′S)-2-N-Ben zyl-3-(2′,2′-d im eth yl-1′,3′-d i-
oxola n -4′-yl)-1,2-isoxa zolid in -5-on e, syn -8: crystals; mp
71-74 °C; [R]D +152.3 (c 1.08, CHCl3); UV λmax 234 nm, ꢀ 710
1
cm-1 M-1; IR (KBr) 1778, 1602, 1265 cm-1; H NMR (CDCl3)
1.33 (s, 3H), 1.41 (s, 3H) 2.57 (dd, J ) 17.2 Hz, J ′ ) 8.9 Hz,
1H), 2.65 (dd, J ) 17.2 Hz, J ′ ) 8.1 Hz, 1H), 3.51 (m, 1H),
3.69 (dd, J ) 8.5 Hz, J ′ ) 6.5 Hz, 1H), 4.01 (dd, J ) 8.5 Hz, J ′
) 6.5 Hz, 1H), 4.13 (d, J ) 14.0 Hz, 1H), 4.18 (m, 1H), 4.31 (d,
J ) 14.0 Hz, 1H), 7.34 (m, 5H); 13C NMR (CDCl3) 173.3, 134.9,
129.5, 128.6, 128.0, 110.3, 76.6, 66.0, 65.6, 63.2, 32.3, 26.4, 25.1.
Anal. Calcd for C15H19NO4: C, 64.97; H, 6.91; N, 5.05. Found:
C, 65.00, H, 6.68; N, 5.00.
Da ta for (3S,4′S)-2-N-Ben zyl-3-(2′,2′-d im eth yl-1′,3′-d i-
oxola n -4′-yl)-1,2-isoxa zolid in -5-on e, a n ti-8: crystals; mp
88-90 °C; [R]D -67.3 (c 1.07, CHCl3); UV λmax 232 nm, ꢀ 804
cm-1 M-1; IR (KBr) 1773, 1254 cm-1; 1H NMR (CDCl3) 1.29 (s,
3H), 1.35 (s, 3H), 2.69 (dd, J ) 18.0 Hz, J ′ ) 5.0 Hz, 1H), 2.77
(dd, J ) 18.0 Hz, J ′ ) 6.9 Hz, 1H), 3.42 (m, 1H), 3.53 (dd, J )
8.4 Hz, J ′ ) 5.0 Hz, 1H), 4.00-4.16 (complex absorption, 2H),
4.11 (d, J ) 13.2 Hz, 1H), 4.24 (d, J ) 13.2 Hz, 1H), 7.34 (m,
5H); 13C NMR (CDCl3) 175.5, 134.2, 129.6, 128.8, 128.5, 109.9,
Rea ction of N-Ben zylh yd r oxyla m in e w ith 1: Isoxa zo-
lid in on es 8 th r ou gh Hyd r oxya m in o Ester s 7. N-Benzyl-
hydroxylamine hydrochloride (132 mg, 0.8 mmol) and dry
triethylamine (120 µL, 0.9 mmol) were successively added to
a solution of a Z/E mixture of alkenoates 1 (129 mg, 0.7 mmol)
in anhydrous dichloromethane (4.5 mL). The resulting mixture
was stirred at room temperature overnight, under a nitrogen
atmosphere. Then, water was added (14 mL), the layers were
separated, and the aqueous one was extracted with dichlo-
romethane (3 × 10 mL). The combined organic phases were
dried (MgSO4), and the solvent was removed at reduced
pressure. The residue was chromatographed on silica gel
(dichloromethane) to afford 134 mg (70% yield) of a 10:1
75.5, 67.0, 64.1, 63.2, 30.3, 26.5, 24.8. Anal. Calcd for C15H19
-
NO4: C, 64.97; H, 6.91; N, 5.05. Found: C, 65.30, H, 7.17; N,
4.76.
Rea ction of N-Ben zylh yd r oxyla m in e w ith Alk en oa tes
(Z)-2 a n d (E)-2: Isoxa zolid in on es 9 a n d 10. A mixture of
(Z)-2 (229 mg, 1.1 mmol), N-benzylhydroxylamine hydrochlo-
ride (851 mg, 5.3 mmol), and sodium ethoxide (363 mg, 5.3
mmol) in absolute ethanol (7 mL) was heated to reflux
overnight. Then the mixture was cooled to room temperature,
and the solvent was removed at reduced pressure. The residue
was dissolved in dichloromethane (10 mL) and washed with
water (10 mL). The organic phase was dried (MgSO4), and the
solvent was evaporated. The residue was chromatographed on
silica gel (3:1 hexane/ethyl acetate) to afford isoxazolidinone
9 (181 mg, 60% yield). In a similar manner, isoxazolidinone
10 was obtained from (E)-2 in 52% yield.
1
mixture of isomeric syn/anti-8 as determined by H NMR and
HPLC (Backerbond column; 3:2 hexane/ethyl acetate; UV
detection, λ ) 233 nm). Chromatography on silica gel by using
3:1 hexane/ethyl acetate as eluent afforded fractions of pure
syn- and anti-8, which were fully characterized. Hydroxyamino
esters syn- and anti-7 could be isolated when the reaction was
(18) Becke, A. D. Phys. Rev. A 1988, 38, 3098.
(19) (a) Wang, Y.; Perdew, J . P. Phys. Rev. B 1991, 44, 13298. (b)
Perdew, J . P.; Chevary, J . A.; Vosko, S. H.; J ackson, K. A.; Pederson,
M. R.; Singh, D. J .; Fiolhais, C. Phys. Rev. B 1992, 46, 6671.
(20) Hehre, W. J .; Radom, L.; Schleyer, P. v. R.; Pople, J . A. Ab Initio
Molecular Orbital Theory; Wiley: New York, 1986.
Da ta for (3R,4R,4′S)-2-N-Ben zyl-3-(2′,2′-d im eth yl-1′,3′-
d ioxola n -4′-yl)-4-m eth yl-1,2-isoxa zolid in -5-on e, 9: crys-
tals; mp 128-130 °C (dichloromethane/pentane); [R]D +132.7
(c 1.01, CHCl3); UV λmax 232 nm, ꢀ 682 cm-1 M-1; IR (KBr)
(21) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery, J . A.,
J r.; Stratmann, R. E.; Burant, J . C.; Dapprich, S.; Millam, J . M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J .;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J .; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J . B.; Cioslowski, J .; Ortiz, J . V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J .; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; J ohnson, B. G.; Chen,
W.; Wong, M. W.; Andres, J . L.; Head-Gordon, M.; Replogle, E. S.;
Pople, J . A., Gaussian, Inc., Pittsburgh, PA, 1998.
(22) (a) ADF 1999: Baerends, E. J .; Be´rces, A.; Bo, C.; Boerrigter,
P. M.; Cavallo, L.; Deng, L.; Dickson, R. M.; Ellis, D. E.; Fan, L.;
Fischer, T. H.; Fonseca Guerra, C.; van Gisbergen, S. J . A.; Groeneveld,
J . A.; Gritsenko, O. V.; Harris, F. E.; van den Hoek, P.; J acobsen, H.;
van Kessel, G.; Kootstra, F.; van Lenthe, E.; Osinga, V. P.; Philipsen,
P. T. H.; Post, D.; Pye, C. C.; Ravenek, W.; Ros, P.; Schipper, P. R. T.;
Schreckenbach, G.; Snijders, J . G.; Sola`, M.; Swerhone, D.; te Velde,
G.; Vernooijs, P.; Versluis, L.; Visser, O.; van Wezenbeek, E.; Wie-
senekker, G.; Wolff, S. K.; Woo, T. K.; Ziegler, T., Scientific Computing
& Modelling NV, Amsterdam, The Netherlands, 1999. http://www.sc-
m.com. (b) Fonseca Guerra, C.; Snijders, J . G.; te Velde, G.; Baerends,
E. J . Theor. Chem. Acc. 1998, 99, 391.
1
1778, 1250 cm-1; H NMR (CDCl3) 1.30 (d, J ) 7.3 Hz, 3H);
1.32 (s, 3H), 1.43 (s, 3H), 2.92 (m, 1H), 3.33 (dd, J ) 8.2 Hz, J ′
) 5.0 Hz, 1H), 3.69 (dd, J ) 8.0 Hz, J ′ ) 7.0 Hz, 1H), 3.94 (dd,
J ) 8.0 Hz, J ′ ) 6.7 Hz, 1H), 4.18 (m, 1H), 4.20 (d, J ) 13.5
Hz, 1H), 4.34 (d, J ) 13.5 Hz, 1H), 7.30 (m, 5H); 13C NMR
(CDCl3) 177.4, 135.0, 129.3, 128.4, 127.7, 109.9, 76.5, 66.0, 65.8,
62.7, 36.7, 26.1, 25.5, 10.2. Anal. Calcd for C16H21NO4: C,
65.96; H, 7.26; N, 4.81. Found: C, 65.93; H, 7.18; N, 4.83.
Da ta for (3R,4S,4′S)-2-N-Ben zyl-3-(2′,2′-d im eth yl-1′,3′-
d ioxola n -4′-yl)-4-m eth yl-1,2-isoxa zolid in -5-on e, 10: crys-
tals; mp 79-81 (dichloromethane/pentane); [R]D +158.0 (c 1.03,
CHCl3); UV λmax 232 nm, ꢀ 734 cm-1 M-1; IR (KBr) 1767, 1652,
1252 cm-1; 1H NMR (CDCl3) 1.25 (d, J ) 7.0 Hz, 3H), 1.36 (s,
3H), 1.43 (s, 3H), 2.77 (m, 1H), 3.17 (dd, J ) 10.9 Hz, J ′ ) 7.0
Hz, 1H), 3.78 (m, 1H), 4.04 (d, J ) 14.3 Hz, 1H), 4.06 (dd, J )
8.2 Hz, J ′ ) 6.1 Hz, 1H), 4.23 (m, 1H), 4.57 (d, J ) 14.3 Hz,
1H). 7.30 (m, 5H); 13C NMR (CDCl3) 175.3, 135.3, 129.4, 128.4,
127.9, 109.8, 76.6, 73.2, 65.9, 63.3, 39.9, 26.3, 25.3, 13.9. Anal.
Calcd for C16H21NO4: C, 65.96; H, 7.26; N, 4.81. Found: C,
65.81; H, 7.26; N, 4.82.
(23) (a) Klamt, A.; Schu¨u¨rmann, G. J . Chem. Soc., Perkin Trans. 2
1993, 799. (b) Barone, V.; Cossi, M. J . Phys. Chem. A 1998, 102, 1995.

Reaction of N-Alkylhydroxylamines and Enoate Esters
J . Org. Chem., Vol. 67, No. 8, 2002 2409
Gen er a l P r oced u r es for th e Syn th esis of Alk en oa tes
3 a n d 4 fr om Ald eh yd es 5 a n d 6, Resp ectively. Standard
protocols are described. Method A: A solution of the aldehyde
(2.8 mmol) in methanol or THF (10 mL) was added, at 0 °C,
to methyl (triphenylphosphoranylidene)acetate (1.2 g, 3.36
mmol) under a nitrogen atmosphere. The mixture was then
stirred at room temperature for 15 h. After evaporation of the
solvent, the residue was extracted with boiling hexane (50 mL),
filtered, concentrated, and purified by column chromatography.
Method B: To a stirred suspension of sodium hydride (100 mg,
2.5 mmol, 60% in mineral oil) in dry THF (2 mL) was slowly
added bis(2,2,2-trifluorethyl) (methoxycarbonylmethyl)phos-
phonate (0.52 mL, 2.5 mmol) under a nitrogen atmosphere.
The mixture was cooled to -78 °C, and a solution of aldehyde
5 (200 mg, 1.01 mmol) in THF (2 mL) was added. The resulting
mixture was then stirred at room temperature for 10 h. The
reaction mixture was treated with a saturated solution of
ammonium chloride and then extracted with dichloromethane
(3 × 30 mL). The combined organic extracts were dried
(MgSO4), and the solvent was removed at reduced pressure.
The crude product was purified by chromatography (3:2
hexane/ether) to afford 150 mg (60%) of (Z)-3. Method C: To
a suspension of sodium hydride (100 mg, 2.5 mmol, 60% in
mineral oil) at -78 °C was added, slowly and with stirring,
trimethyl phosphonacetate (455 mg, 2.5 mmol) in dry THF (6
mL) under a nitrogen atmosphere. After 30 min a solution of
aldehyde 5 (400 mg, 2.02 mmol) was added. The mixture was
then stirred at room temperature for 20 h. The reaction
mixture was treated with a saturated solution of ammonium
chloride and then extracted with dichloromethane (3 × 30 mL).
The organic layer was dried (MgSO4), and the solvent was
removed at reduced pressure. The crude product was purified
by chromatography (3:2 hexane/ether) to afford 260 mg (50%)
of (E)-3.
Then, water was added (40 mL), the layers were separated,
and the aqueous one was extracted with dichloromethane (3
× 30 mL). The combined organic phases were dried (MgSO4),
and the solvent was removed at reduced pressure. The residue
was chromatographed on silica gel (2:1 hexane/AcOEt) to
afford pure compounds.
Da ta for (3S,1′S,3′R)-2-N-Ben zyl-3-[2′,2′-d im eth yl-3′-(2-
m eth yl-1,3-d ioxola n -2-yl)cyclobu tyl]-1,2-isoxa zolid in -5-
on e, 14: yield 580 mg (87%); crystals; mp 42-43 °C (hexane/
AcOEt); [R]D -69 (c 2.0, MeOH); IR (film) 2930, 1782,1638
1
cm-1; H NMR (acetone-d6) 1.07 (s, 3H), 1.16 (s, 3H), 1.19 (s,
3H), 1.65 (m, 1H), 2.00 (m, 2 H), 2.20 (m, 2H), 2.42 (dd, J )
17.2 Hz, J ′ ) 8.2 Hz, 1H), 2.97 (dd, J ) 17.2 Hz, J ′ ) 7.3 Hz,
1H), 3.75-3.95 (m, 4H), 4.07 (d, J ) 14.0 Hz, 1H), 4.18 (d, J
) 14.0 Hz, 1H), 7.30-7.36 (complex absorption, 5H); 13C NMR
(acetone-d6) 17.63, 23.96, 24.17, 31.80, 35.44, 41.78, 46.10,
50.60, 63.77, 64.21, 65.97, 66.92, 110.05, 128.27, 129.03,
129.87, 137.71, 175.16. Anal. Calcd for CHNO: C, 69.54; H,
7.88; N, 4.05. Found: C, 70.04; H, 7.89; N, 3.96.
Da t a for (3S,1′S,3′R)-2-N-Ben zyl-3-(2′,2′-d im et h yl-3′-
m e t h oxyca r b on ylcyclob u t yl)-1,2-isoxa zolid in -5-on e ,
15: yield 311 mg (50%); crystals; mp 78-80 °C (MeOH/H2-
CCl2/hexane); [R]D -66.01 (c 2.0, H2CCl2); IR (film) 2954, 1780,-
1
1733 cm-1; H NMR (HCCl3) 0.91 (s, 3H), 1.22 (s, 3H), 2.00-
2.20 (m, 3 H), 2.40 (dd, J ) 17.1 Hz, J ′ ) 8.5 Hz, 1H), 2.70
(dd, J ) 17.0 Hz, J ′ ) 7.2 Hz, 1H), 3.64 (s, 3H), 3.35 (m, 1H),
4.02 (d, J ) 14.0, 1H), 4.10 (d, J ) 14.0, 1H), 7.32 (complex
absorption, 5H); 13C NMR (CHCl3) 17.78, 24.01, 30.37, 35.06,
42.36, 44.85, 45.84, 63.31, 65.97, 127.99, 128.58, 129.06,
135.20, 172.59, 174.10; HRMS m/z calcd for
C18H23NO4
317.1627, found 317.1632 (M+).
Gen er a l P r oced u r e for th e Hyd r ogen olysis of Isox-
a zolid in on es syn -8, 9, 10, a n d 15: â-Am in o Acid s 11, 12,
13, a n d 16. A standard reaction is described as follows for
the reduction of syn-8 to afford 11. A mixture of isoxazolidinone
syn-8 (111 mg, 0.4 mmol) and 20% Pd(OH)2/C (28 mg) in
ethanol (10 mL) was hydrogenated at room temperature under
4 atm of pressure. The reaction mixture was filtered through
Celite, and the solvent was removed at reduced pressure. The
residue was dissolved in 15:2 water/ethanol and filtered
through a reversed-phase C18-cartridge eluting with water.
After solvent removal, amino acid 11 (68 mg, 90% yield) was
obtained. Following the same procedure, amino acids 12, 13,
and 16 were prepared in nearly quantitative yield.
Da ta for Meth yl (1′R,3′R)-3-[2′,2′-Dim eth yl-3′-(2-m eth -
yl-1,3-d ioxola n -2-yl)cyclobu tyl]-(Z)-2-p r op en oa te, (Z)-3:
oil; [R]D -74.91 (c 2.75, MeOH); IR (film) 2953, 1725,1650 cm-1
;
1H NMR (acetone-d6) 1.06 (s, 3H), 1.13 (s, 3H), 1.17 (s, 3H),
1.80 (complex absorption, 1H), 2.00 (m, 2H), 2.22 (dd, J ) 11.0
Hz, J ′ ) 8.0 Hz, 1H), 3.64 (s, 3H), 3.70-4.00 (m, 4H), 5.75
(dd, J ) 11.7 Hz, J ′ ) 1.1 Hz, 1H), 6.22 (dd, J ) 11.7 Hz, J ′ )
10.2 Hz, 1H); 13C NMR (acetone-d6) 17.57, 22.76, 24.77, 30.82,
39.94, 43.72, 49.69, 49.89, 63.08, 64.81, 109.04, 119.36, 150.15,
165.65. Anal. Calcd for C14H22O4: C, 66.12; H, 8.72. Found:
C, 66.04; H, 8.82.
Da ta for (3R,4′S)-3-Am in o-3-(2′,2′-d im eth yl-1′,3′-d iox-
ola n -4′-yl)p r op a n oic Acid , 11: highly hygroscopic solid
unsuitable for mp and microanalysis determinations; [R]D
+40.2 (c 2.01, methanol); IR (KBr) 3500-3200 (br), 1730, 1295
Da ta for Meth yl (1′R,3′R)-3-[2′,2′-Dim eth yl-3′-(2-m eth -
yl-1,3-d ioxola n -2-yl)cyclobu tyl]-(E)-2-p r op en oa te, (E)-3:
oil; [R] -15.60 (c 2.1, MeOH); IR (film) 2954, 1724, 1638 cm-1
;
1H NMR (acetone-d6) 0.99 (s, 3H), 1.15 (s, 3H), 1.17 (s, 3H),
1.90 (complex absorption, 1H), 2.05 (m, 1H), 2.25 (dd, J ) 9.5
Hz, J ′ ) 9.2 Hz, 1H), 2.55 (m, 1H), 3.66 (s, 3H), 3.80-4.00 (m,
4H), 5.75 (dd, J ) 15.6 Hz, J ′ ) 1.3 Hz, 1H), 6.89 (dd, J )
15.6 Hz, J ′ ) 7.4 Hz, 1H); 13C NMR (acetone-d6) 18.68, 23.96,
31.29, 44.33, 45.44, 50.63, 51.41, 64.27, 66.02, 110.04, 121.81,
149.84, 166.91. Anal. Calcd for C14H22O4: C, 66.12; H, 8.72.
Found: C, 65.93; H, 8.63.
1
cm-1; H NMR (methanol-d4) 1.44 (s, 3H), 2.38 (dd, J ) 16.9
Hz, J ′ ) 8.7 Hz, 1H), 2.49 (dd, J ) 16.9 Hz, J ′ ) 4.5 Hz, 1H),
3.46 (m, 1H), 3.88 (dd, J ) 11.7 Hz, J ′ ) 9.0 Hz, 1H), 4.22 (m,
2H), 4.99 (br s, 2H); 13C NMR (methanol-d4) 176.5, 111.5, 76.7,
67.4, 53.5, 36.1, 26.8, 25.3; MS m/z (rel intens) 190 (M+ + 1,
1), 174 (6), 114 (10), 88 (100), 70 (35), 43 (44).
Da ta for (2R,3R,4′S)-3-Am in o-3-(2′,2′-d im eth yl-1′,3′-d i-
oxola n -4′-yl)-2-m eth ylp r op a n oic Acid , 12: deliquescent
solid unsuitable for mp and microanalysis determinations; [R]D
Meth yl (1′R,3′R)-3-(2′,2′-Dim eth yl-3′-m eth oxyca r bon -
ylcyclobu t yl)-(E)-2-p r op en oa t e, 4. This compound was
obtained by method A and purified by chromatography (3:1
CH2Cl2/AcOEt): yield 342 mg (54%); oil; [R]D -9.82 (c 1.12,
1
+9.2 (c 1.95, methanol); IR (KBr) 3421, 1726, 1268 cm-1; H
NMR (methanol-d4) 1.28 (d, J ) 7.5 Hz, 3H), 1.45 (s, 3H), 1.52
(s, 3H), 2.50 (m, 1H), 3.46 (dd, J ) 7.9 Hz, J ′ ) 4.5 Hz, 1H),
3.83 (m, 1H), 4.34 (m, 2H), 5.02 (br s, 2H); 13C NMR (methanol-
d4) 180.8, 111.2, 75.7, 67.8, 57.1, 41.7, 26.8, 25.6, 12.8; MS m/z
(rel intens) 204 (M+ + 1, 6), 188 (8), 102 (100), 84 (62), 72 (31),
43 (26).
CHCl3); IR (film) 2954, 1792, 1734, 1651 cm-1 1H NMR
;
(CDCl3) 0.87 (s, 3H), 1.21 (s, 3H), 2.05 (m, 1H), 2.25 (m, 1H),
2.55-2.75 (m, 2H), 3.63 (s, 3H), 3.69 (s, 3H), 5.75 (dd, J )
15.6 Hz, J ′ ) 1.3 Hz, 1H), 6.87 (dd, J ) 15.6 Hz, J ′ ) 7.2 Hz,
1H); 13C NMR (CDCl3) 18.53, 23.09, 29.82, 44.35, 44.90, 45.80,
51.24, 51.41, 121.62, 147.95, 166.73, 172.78. Anal. Calcd for
(C12H18O4)2‚H2O: C, 61.25; H, 8.14. Found: C, 61.22; H, 8.07.
Rea ction of N-Ben zylh yd r oxyla m in e w ith 3 a n d 4:
Isoxa zolid in on es 14 a n d 15. N-Benzylhydroxylamine hy-
drochloride (400 mg, 2.42 mmol) and dry triethylamine (0.5
mL, 3.75 mmol) were successively added to a solution of a Z/E
mixture of alkenoates 3 or 4 (1.97 mmol) in anhydrous
dichloromethane (15 mL). The resulting mixture was stirred
at room temperature for 60 h, under a nitrogen atmosphere.
Da ta for (2S,3R,4′S)-3-Am in o-3-(2′,2′-d im eth yl-1′,3′-d i-
oxola n -4′-yl)-2-m eth ylp r op a n oic Acid , 13: deliquescent
solid unsuitable for mp and microanalysis determinations; [R]D
-11.8 (c 1.87, methanol); IR (KBr) 3430, 1721, 1276 cm-1; 1H
NMR (methanol-d4) 1.37 (d, J ) 7.5 Hz, 3H), 1.44 (s, 3H), 1.53
(s, 3H), 2.45 (m, 1H), 3.24 (dd, J ) 7.9 Hz, J ′ ) 4.5 Hz, 1H),
3.90 (m, 1H), 4.34 (m, 2H), 5.09 (br s, 2H); 13C NMR (methanol-
d4) 180.6, 111.3, 75.75, 67.77, 58.34, 41.32, 26.78, 25.33, 16.00;
MS m/z (rel intens) 204 (M+ + 1, 1), 188 (6), 102 (100), 84 (61),
91 (35), 72 (28), 43 (34).

2410 J . Org. Chem., Vol. 67, No. 8, 2002
Moglioni et al.
Da t a for (3S,1′S,3′R)-3-Am in o-3-[2′,2′-d im et h yl-3′-(2-
m eth yl-1,3-d ioxola n -2-yl)cyclobu tyl]p r op a n oic Acid , 16:
yield 95%; crystals; mp 85-87 °C (EtOH); [R]D -26.09 (c 1.15,
This work has been financially supported by DGESIC
and AECI (Spain) and Fundacio´n Antorchas (Argentina)
through Projects PB97-0214, Q2812001B, and A-1362211-
70, respectively. Computer time from the Centre de
Supercomputacio´ de Catalunya is gratefully acknowl-
edged.
1
MeOH); IR (film) 3543, 2952,1706 cm-1; H NMR (methanol-
d4) 1.20 (s, 3H), 1.28 (s, 3H), 1.30 (s, 3H), 1.80-2.10 (m, 3H),
2.10-2.30 (m, 2H), 2.49 (dd, J ) 16.9 Hz, J ′ ) 3.0 Hz, 1H),
3.33 (dd, J ) 9.8 Hz, J ′ ) 2.9 Hz, 1H), 3.85-4.10 (m, 4H); 13
C
NMR (methanol-d4) 17.09, 23.09, 23.98, 32.06, 38.84, 41.94,
45.37, 50.21, 52.40, 64.67, 66.43, 110.64, 177.63; HRMS m/z
calcd for C13H23NO4 257.1627, found 257.1631 (M+).
Su p p or tin g In for m a tion Ava ila ble: Total energies and
geometries of energy minima and transition states. This
material is available free of charge via the Internet at
http://pubs.acs.org.
Ack n ow led gm en t. E.M. thanks the Ministerio de
Educacio´n, Spain, for a predoctoral fellowship, and
A.G.M. thanks the CONICET, Argentina, for a grant.
J O0159082