Tandem transesterification and intramolecular cycloaddition of α-methoxycarbonylnitrones with chiral acyclic allyl alcohols: Systematic studies on the factors affecting diastereofacial selectivity of the cycloaddition

Article abstract of DOI:10.1021/jo001015i

Factors affecting the stereochemical course of the intramolecular cycloaddition of intermediary α-allyloxycarbonylnitrone resulting from transesterification of α-methoxycarbonylnitrones 1a-d with chiral allyl alcohols 5 or 6 were investigated systematically. It was found that the factors of diastereofacial selection are highly dependent on the geometries of the allyl alcohols. In the cases where primary or secondary chiral (Z)-allyl alcohols are used, A(1,3)-strain arising from the chiralities in the (Z)-nitrone transition states of the intramolecular cycloaddtion is the most important factor. In contrast, in the case of the reaction using chiral nitrones 1c,d and (E)-allyl alcohols, steric interaction between the chiral N-substituent and the trans substituent of the olefin moiety in the intermediate is dominant. These aspects were applied to geometry-differentiated cycloaddition using a mixture of (E)-5 and (Z)-5. As a typical example, treatment of bulky 1b with a 1:1 mixture of (E)-5 and (Z)-5 in the presence of a catalytic amount of TiCl⁴ and MS 4A gave 7b as the predominant product among four possible products.

Full text of DOI:10.1021/jo001015i

2602
J . Org. Chem. 2001, 66, 2602-2610
Ta n d em Tr a n sester ifica tion a n d In tr a m olecu la r Cycloa d d ition of
r-Meth oxyca r bon yln itr on es w ith Ch ir a l Acyclic Allyl Alcoh ols:
System a tic Stu d ies on th e F a ctor s Affectin g Dia ster eofa cia l
Selectivity of th e Cycloa d d ition
Osamu Tamura,*,† Naka Mita,‡ Toshiyuki Okabe,‡ Tatsuya Yamaguchi,‡ Chieko Fukushima,‡
Masayuki Yamashita,# Yasuhiro Morita,‡ Nobuyoshi Morita,‡ Hiroyuki Ishibashi,† and
Masanori Sakamoto*,‡
Faculty of Pharmaceutical Sciences, Kanazawa University, Takara-machi, Kanazawa 920-0934, J apan,
Meiji Pharmaceutical University, Noshio, Kiyose, Tokyo 204-8588 J apan, and Kyoto Pharmaceutical
University, Misasagi, Yamashina, Kyoto 607-8414, J apan
tamura@dbs.p.kanazawa-u.ac.jp
Received J uly 6, 2000
Factors affecting the stereochemical course of the intramolecular cycloaddition of intermediary
R-allyloxycarbonylnitrone resulting from transesterification of R-methoxycarbonylnitrones 1a -d
with chiral allyl alcohols 5 or 6 were investigated systematically. It was found that the factors of
diastereofacial selection are highly dependent on the geometries of the allyl alcohols. In the cases
where primary or secondary chiral (Z)-allyl alcohols are used, A(1,3)-strain arising from the
chiralities in the (Z)-nitrone transition states of the intramolecular cycloaddtion is the most
important factor. In contrast, in the case of the reaction using chiral nitrones 1c,d and (E)-allyl
alcohols, steric interaction between the chiral N-substituent and the trans substituent of the olefin
moiety in the intermediate is dominant. These aspects were applied to geometry-differentiated
cycloaddition using a mixture of (E)-5 and (Z)-5. As a typical example, treatment of bulky 1b with
a 1:1 mixture of (E)-5 and (Z)-5 in the presence of a catalytic amount of TiCl4 and MS 4A gave 7b
as the predominant product among four possible products.
In tr od u ction
and stereoselectivties were achieved. For the intra-
molecular cycloadditions, however, long and multistep
synthetic routes have been required to set up two reaction
sites, nitrone groups and olefin moieties, at appropriate
positions in the molecules. As a solution for this draw-
back, we recently explored “tandem transesterification
and intramolecular 1,3-dipolar cycloaddition methodol-
ogy”.5,6 In this methodology, transesterification of equili-
brating mixtures of (E)- and (Z)-R-methoxycarbonyl-
nitrones (1)with allylalcohols 2 provides R-allyloxycarbonyl-
nitrones 3 as the intermediates, which, in turn, undergo
intramolecular cycloaddition via (Z)-nitrone transition
states A7 to give relative stereochemistry-controlled
cycloadducts 4 in one step (eq 1). It should be noted that
the tandem process acquires operational simplicity of
intermolecular cycloaddition as well as excellent regio-
1,3-Dipolar cycloaddition of nitrones with olefins leads
to spontaneous formation of carbon-carbon bonds and
carbon-oxygen bonds to give isoxazolidines, in which the
geometries of the olefins are reflected in the stereochem-
istries of the products.1 Since reductive cleavage of
nitrogen-oxygen bonds of the resulting isoxazolidines is
quite facile, nitrone cycloaddition is a useful method for
the syntheses of γ-amino alcohols.1,2 Intermolecular cyclo-
addition of a nitrone, however, gives principally four
isomers: endo and exo-isomers as well as two sets of
regioisomers.1 In the case of intramolecular cycloaddi-
tions of nitrones connected with olefin moieties by tethers
of appropriate lengths, the regiochemical problem can be
neglected and diastereoselectivity (endo- and exo-selec-
tivity) can also be dramatically improved.3 This meth-
odology has been successfully applied to the syntheses
of natural products such as alkaloids, â-lactam antibiot-
ics, and amino sugars.4 In these syntheses, high regio-
(3) For reviews see: (a) Annunziata, R.; Cinquini, M.; Cozzi, F.;
Raimondi, L. Gazz. Chim. Ital. 1989, 119, 253. (b) Padwa, A.;
Schoffstall, A. M. Intramolecular 1,3-Dipolar Cycloaddition Chemistry.
In Advances in Cycloaddition; Curran, D. P., Ed.; J AI Press: Green-
wich, 1990; Vol. 2, p 1. (c) Wade, P. A. Intramolecular 1,3-dipolar
cycloaddition. In Comprehensive Organic Synthesis; Trost, B. M.,
Fleming, I., Semmelhack, M. F., Eds.; Pergamon Press: Oxford, 1991;
Vol. 4, p 1113.
(4) For a review, see: Frederickson, M. Tetrahedron, 1997, 53, 403.
For representative examples, see: (a) Wovkulich, P. M.; Uskokovic,
M. R. J . Am. Chem. Soc. 1981, 103, 3956; Wovkulich, P. M.; Uskokovic,
Tetrahedron 1985, 41, 3455. (b) Ihara, M.; Takahashi, M.; Fukumoto,
K.; Kametani, T. J . Chem. Soc., Chem. Commun. 1988, 9. (c) Smith,
A. L.; Williams, S. F.; Holms, A. B.; Hughes, L. R.; Lidert, Z.;
Swithenbank, C. J . Am. Chem. Soc. 1988, 110, 8696. (d) J ung, M. K.;
Vu, B. T. J . Org. Chem. 1996, 61, 4427. (e) Denmark, S. E.; Martin-
borough, E. A. J . Am. Chem. Soc. 1999, 121, 3046 and references sited
therein. (f) Werner, K. M.; de los Santos, J . M.; Weinreb, S. M. J . Org.
Chem. 1999, 64, 686. (g) Williams, G. M.; Roughley, S. D.; Davies, J .
E.; Holms, A. B.; Adams, J . P. J . Am. Chem. Soc. 1999, 121, 4900.
* To whom correspondence should be sent.
† Kanazawa University.
‡ Meiji Pharmaceutical University.
#
Kyoto Pharmaceutical University.
(1) For reviews of cycloadditions of nitrones, see: (a) Confalone, P.
N.; Huie, E. M. Org. React. 1988, 36, 1. (b) Deshong, P.; Lander, S.
W., J r.; Leginus, J . M.; Dicken, C. M. In Advances in Cycloaddition;
Curran, D. P., Ed.; J AI Press: Greenwich, 1988; Vol. 1, p 87. (c) Little,
R. D. Thermal cycloadditions. In Comprehensive Organic Synthesis;
Trost, B. M., Fleming, I., Eds.; Pergamon Press: Oxford 1991; Vol. 5,
p 239. (d) Padwa, A. Intermolecular 1,3-dipolar cycloadditions. In
Comprehensive Organic Synthesis; Trost, B. M., Fleming, I., Semmel-
hack, M. F., Eds.; Pergamon Press: Oxford, 1991; Vol. 4, p 1069.
(2) Cicchi, S.; Goti, A.; Brandi, A.; Guarna, A.; De sarlo, F.
Tetrahedron Lett. 1990, 31, 3351.
10.1021/jo001015i CCC: $20.00 © 2001 American Chemical Society
Published on Web 03/27/2001

R-Methoxycarbonylnitrones
Sch em e 1
J . Org. Chem., Vol. 66, No. 8, 2001 2603
F igu r e 1. Nitrones and allyl alcohols employed.
absolute stereochemistries by using an acyclic allyl
alcohol, a chirality (or chiralities a*-c*) should be
introduced in the intramolecular system.8 To investigate
efficient factor to affect the diastereofacial selection, we
devoted our efforts to examine systematically diastereo-
facial selectivities in light of the position of chirality (a*-
c*) and geometry of olefin moiety by employing nitrones
1a -d , primary allyl alcohols (Z)- and (E)-5, and second-
ary allyl alcohols 6 (Figure 1). We have now found that
the factors of diastereofacial selection depend on the
geometry of the allyl alcohol. Thus, in the case where
chiral (Z)-5 or (Z)-6 is used, A(1,3)-strain arising from
chirality a* or b* is the important factor in the control of
the diastereofacial selectivity, while in the case where
(E)-6 is used, chirality c* controls the diastereofacial
selection, and the combined use of chiralities b* and c*
causes double asymmetric induction to exhibit high
selectivity.9
Sch em e 2
and stereoselectivity of intramolecular cycloaddition.7 In
that work,5b we also found that the reactivities of (Z)-
allyl alcohols are much higher than those of (E)-allyl
alcohols, and that this phenomenon could be extended
to “geometry-differentiated cycloaddition” in which the
reaction of 1 with a mixture of (E)- and (Z)-allyl alcohols
predominantly gives intramolecular cycloadducts from
(Z)-allyl alcohols.
In this type of reaction, the use of chiral cyclic second-
ary allyl alcohols enables the control of the absolute
stereochemistries of three newly formed stereogenic
centers (Scheme 1),5 whereas the use of chiral acyclic allyl
alcohols, as shown in Scheme 2, generates two possibili-
ties of the diastereofacial selections (si-si attack B or
re-re attack C) due to their flexibilities. To control the
Resu lts a n d Discu ssion
1. Th e Effect of Ch ir a lity a * in Ca ses Wh er e
P r im a r y Allyl Alcoh ols a r e Used . First, we examined
the reactions of nitrones 1a ,b with primary allyl alcohol
(Z)-5 and (E)-5 to gain insights into the effects of tether-
outer chirality (chirality a*) in the intramolecular cy-
cloaddition. As shown in Scheme 3, treatment of nitrones
1a ,b with primary allyl alcohol (Z)-5 in the presence of
catalytic amounts of titanium tetrachloride and molec-
ular sieves 4A (MS4A) induced smooth transesterification
and intramolecular cycloaddition to afford the cycload-
ducts 7a ,b as the major isomers along with small
amounts of 8a ,b independent of substituents (R) of the
nitrones 1a ,b. The stereochemistry of the cycloadduct 7a
(5) (a) Tamura, O.; Yamaguchi, T.; Noe, K.; Sakamoto, M. Tetrahe-
dron Lett. 1993, 34, 4009. (b) Tamura, O.; Okabe, T.; Yamaguchi, T.;
Gotanda, K.; Noe, K.; Sakamoto, M. Tetrahedron 1995, 51, 107. (c)
Tamura, O.; Okabe, T.; Yamaguchi, T.; Kotani, J .; Gotanda, K.;
Sakamoto, M. Tetrahedron 1995, 51, 119. For closely related tandem
acylation and intramolecular cycloaddition, see: (d) Tan, D. S.; Foley,
M. A.; Stockwell, B. R.; Shair, M. D.; Schreiber, S. L. J . Am. Chem.
Soc. 1999, 121, 9073. For tandem hemiacetalization and intramolecular
nitrone cycloaddition, see: (e) Kanemasa, S.; Tsuruoka, T.; Wada, E.
Tetrahedron Lett. 1993, 34, 87. (f) Kanemasa, S.; Tsuruoka, T.;
Yamamoto, H. Tetrahedron Lett. 1995, 36, 5019.
(6) For a reviews on tandem reactions, see: (a) Tietze, L. F.; Beifuss,
U. Angew. Chem., Int. Ed. Engl. 1993, 32, 131. See also: (b) Freder-
ickson, M.; Grigg, R.; Markandu, J .; Thornton-Pett, M.; Redpath, J .
Tetrahedron 1997, 53, 15051 and references therein. (c) Denmark S.
E.; Hurd, A. R. J . Org. Chem. 2000, 65, 2875 and references sited
therein.
(7) Owing to its operational simplicity, Schreiber and co-workers
successfully applied the tandem nitrone cycloaddition concept to
construction of a large chemical library. See ref 5d. They proposed an
(E)-nitrone transition state instead of (Z)-nitrone transition state in
Scheme 1. However, preliminary calculations for transition state A
(R ) Bn, R′ ) H) derived from 1a and cyclohex-2-en-1-ol and the
corresponding (E)-nitrone transition state showed that the (Z)-nitrone
tansition state is more stable by 9.6 kcal/mol than is the (E)-nitrone
transition state. Therefore, the stereochemical aspects in this work
are discussed by employing the (Z)-nitrone transition state model. See
Supporting Information.
(8) For reviews of diastereofacial selectivity of intramolecular cy-
cloaddition of nitrones, see: (a) Chiacchio, U.; Rescifina, A.; Romeo,
G. Targets Heterocycl. Syst. 1997, 1, 225. (b) Gothelf, K. V.; J ørgensen,
K. A. Chem. Rev. 1998, 98, 863. See also the following recent
examples: (c) Aurich, H. G.; Biesemeier, F.; Harms, K. Liebigs Ann./
Recl. 1997, 469. (d) Aurich, H.; Soeberdt, M.; Harms, K. Tetrahedron
1999, 55, 1249. (e) Marcus, J .; Brussee, J .; Van del Gen, A. Eur. J .
Org. Chem. 1998, 2513. (f) Uddin, M. J .; Kikuchi, M.; Takedatsu, K.;
Arai, K.; Fujimoto, T.; Motoyoshiya, J .; Kakehi, A.; Iriye, R.; Shirai,
H.; Yamamoto, I. Synthesis 2000, 365. (g) Broggini, G.; Folcio, F.;
Sardone, N.; Sordone, N.; Sonzogni, M.; Zecchi, G. Tetrahedron:
Asymmetry 1996, 7, 797. (h) Arnold, W.; Mohr, P. Synlett 1996, 1215.
(i) Annunziata, R.; Benaglia, M.; Clinquini, M.; Cozzi, F.; Raimondi,
L. Eur. J . Org. Chem. 1998, 1823. (j) Hems, W. P.; Tan, C.-H.; Stork,
T.; Feeder, N.; Holms, A. B. Tetrahedron Lett. 1999, 40, 1393. (k) Tan.
C.-H.; Stork, T.; Feeder, N.; Holmes, A. B. Tetrahedron Lett. 1999, 40,
1397. (l) Aurich, H. G.; Geiger, M.; Gentes, C.; Harms, K.; Ko¨ster, H.
Tetrahedron 1998, 54, 3181. (m) Shing, T. K. M.; Zhong, Y.-L.; Mak,
T. C. W.; Wang R.-J .; Xue, F. J . Org. Chem. 1998, 63, 414. (n)
Majumdar, S.; Bhattacharjya, A.; Patra, A. Tetrahedron Lett. 1997,
38, 8581. (o) Bar, N. C.; Roy, A.; Achari, B.; Mandal, S. B. J . Org. Chem.
1997, 62, 8948.
(9) Preliminary communication for a part of this work: Tamura,
O.; Yamaguchi, T.; Okabe, T.; Sakamoto, M. Synlett 1994, 620.

2604 J . Org. Chem., Vol. 66, No. 8, 2001
Tamura et al.
afford the major isomer 7 predominantly.12 The very slow
reaction of nitrone 1a with allyl alcohol (E)-5 may be
explained by taking into acount the steric hindrances in
both models F and G (Scheme 5).
Sch em e 3a
2. Th e Effect of Ch ir a lities b* a n d c* in th e Ca se
Wh er e Secon d a r y Allyl Alcoh ol (Z)-6 is Used . Next,
we investigated the tandem reaction of nitrones 1a -d
with allyl alcohol (Z)-6 to determine whether tether-inner
chirality b* affects the diastereofacial selectivity of the
cycloaddition. We also chose (Z)-6 to clarify the effects of
chiralities c* of 1c,d on the diastereofacial selectivities.
Since the diastereofacial selectivities should be reflected
directly in the relative stereochemistries between 5- and
6-positions of the rigid products 12 and 13, the methyl
group of (Z)-6 would be an excellent stereochemical
indicator for diastereofacial intramolecular cycloaddition
(Table 1). Treatment of nitrone 1a with alcohol (Z)-6 in
the presence of a catalytic amount of titanium tetrachlo-
ride and MS4A led to smooth transesterification and
intramolecular cycloaddition even at room temperature
to afford 12a accompanied by a small amount of 13a
(entry 1). The more bulky nitrone 1b also reacted with
(Z)-6 under the same conditions, giving 12b and 13b in
an excellent combined yield with high diastereofacial
selectivity (entry 2). It was also found that reactions of
chiral nitrones 1c,d with (Z)-6 proceeded smoothly to give
12c,d as the major cycloadducts in excellent yields with
high diastereofacial selectivities which were independent
of the chiralities of nitrones 1c,d (entries 3,4).
a
(a) 0.1 equiv TiCl4, MS4A, ClCH2CH2Cl, rt.
Sch em e 4a
Since the relative stereochemistries of cycloadducts 12
must be opposite to those of 13 with respect to their
methyl groups, the stereochemical assignments of 12 and
13 were readily made based on the coupling constants
a
(a) 0.1 equiv TiCl4, MS4A, ClCH2CH2Cl, 80 °C.
Sch em e 5
1
(J ab) in their H NMR spectra, as shown in Figure 2. For
example, the smaller J ab (3.3 Hz) of 12a clearly shows
that Ha and Hb have a trans-relationship, while the larger
coupling constant (J ab ) 6.6 Hz) of 13a indicates that Ha
and Hb bear a cis relationship. The stereochemistries of
12b-d and 13b-d were also assigned in the same
manner, and the stereochemistries of 12a and 12d were
further confirmed by the NOE difference spectra.
The observed selectivities may be rationalized by
considering transition state models H and I in the
intramolecular cycloaddition step, as shown in Scheme
6. In the case where (Z)-6 is used, the transition state
model H appears to be more favorable than model I, since
I has a severe A(1,3)-strain between the methyl group
and the bulky pivaloyloxymethyl group (R1). Moreover,
the R group lies too far to interact with the R1 group.
Accordingly, the reactions of the chiral nitrones 1c,d with
(Z)-6 gave 12c,d predominantly, independent of chirality
c* on the R group by way of H. This tendency indicates
that the chirality b* of (Z)-6 makes the olefin moiety react
from the si face by A(1,3)-strain.
was definitely established by transformation of 7a into
3,5-dinitrobenzoate 9 and its X-ray diffraction. On the
other hand, intramolecular cycloaddition products 10 and
11 could not be isolated from 1a and (E)-5 under similar
conditions since the cycloaddition was very slow despite
the smooth transesterification (Scheme 4).10
As depicted in Scheme 5, the observed diastereofacial
selection shown in Scheme 3 can be rationalized by
considering transition models D and E with minimum
A(1,3)-strains11 resulting from chirality a*. Since model
E has severe steric interaction between the nitrone
moiety and the bulky tert-butyldimethylsilyloxy group,
the reaction would proceed mainly by way of model D to
3. Th e Effect of Ch ir a lities b* a n d c* in th e Ca se
Wh er e th e Secon d a r y Allyl Alcoh ol (E)-6 is Used .
Next, we examined the reactions of nitrones 1a -d with
(E)-6 to determine the factors affecting the diastereofacial
selectivity in the cases where (E)-allyl alcohols are used
(Table 2). In contrast to the reactions employing (Z)-6,
reactivities and diastereofacial selectivities in the reac-
tions using (E)-6 were highly dependent on the substit-
(10) When the cycloaddition is not completed, the cycloadducts can
not be isolated by chromatography since the polarities of the cycload-
ducts are between those of the (E)- and (Z)-isomers of the starting
nitrone.
(11) Evans, D. A. In Asymmetric Synthesis; Morrison, J . D., Ed.;
Academic Press: London, 1984; Vol 3, p 1. Hoffmann, R. W. Chem.
Rev. 1989, 89, 1841. Hoveyda, A. H.; Evans, D. A.; Fu, G. C. Chem.
Rev. 1993, 93, 1307.
(12) The diastereofacial selection may be also explained in terms of
dipole-dipole interaction between the silyloxy group and the oxygen
atom of the nitrone group. See ref 8i.

R-Methoxycarbonylnitrones
J . Org. Chem., Vol. 66, No. 8, 2001 2605
Ta ble 1. Ta n d em Tr a n sester ifica ton a n d In tr a m olecu la r Cycloa d d ition of A-Meth oxyca r bon yln itr on es 1a -d w ith Allyl
Alk oh ol (Z)-6a
a
b
All reactions were carried out with MS4A in 1,2-dichloroethane at room temperature. The yields were obtained after purification by
column chromatography on silica gel. c The ratios were estimated on the basis of the 1H NMR spectra of the diastereomeric mixtures.
(E)-6 exhibited a reversal of selectivity, giving rise to 14d
predominantly with moderate selectivity (entry 4).
The methyl groups of the cycloadducts 14a ,c,d and
15a ,c,d were again good indicators for the stereochemical
assignments (Figure 3). For example, the smaller J ab (4.0
Hz) in the 1H NMR spectrum of 14a clearly shows a trans
1
relationship of Ha and Hb, while the H NMR spectrum
of 15a having a cis relationship of Ha and Hb shows a
larger coupling constant (J ab ) 6.6 Hz). The stereochem-
istries of the other cycloadducts 14c,d and 15c,d were
also assigned on the basis of their J values (J ab), and were
confirmed by NOEs.
The diastereofacial selectivities of the reactions of
nitrones 1a -d with allyl alcohol (E)-6 were rather
complicated compared to those of the reactions using (Z)-
6. These selectivities may be interpreted by taking into
account (Z)-nitrone transition state models J and K
(Scheme 7). In J and K, A(1,3)-strains were released more
F igu r e 2. Selected coupling constants of 12a , 13a , and 12d
and NOEs of 12a and 12d .
than they were in H and I, the models for the reactions
of 1 with (Z)-6 (see Scheme 6). As a result, the reaction
of N-benzyl nitrone 1a having no chirality gave poor
diastereofacial selectivity. Since the R group of the
nitrones occupies a position very close to the R1 group in
J and K, bulky nitrone 1b (R ) CHPh2) showed a very
low reactivity due to the steric interaction between the
R and R1 groups. Due to the closeness between R1 and
R, chiral environment is conveyed from c* to R1 group;
therefore, the combined uses of chiralities of the nitrones
1c,d and chirality b* of (E)-6 cause double asymmetric
induction.13 Thus, high diastereofacial selectivity was
achieved by employing N-(S)-1-phenylethyl nitrone 1c
and (E)-6 to afford 15c by way of K due to making a
matched pair.13b On the other hand, N-(R)-1-phenylethyl
nitrone 1d forms a mismatched pair13b with (E)-6, and
the effect of the chirality c* of 1d overcomes that of the
chirality b* of (E)-6, affording the cycloadduct 14d in
moderate selectivity by way of J . These results clearly
showed that chirality c* of 1 strongly affects the diaste-
Sch em e 6
uents (R) of nitrones 1a -d . Thus, the reaction of N-
benzyl nitrone 1a with (E)-6 gave cycloadducts 14a and
15a in a ratio of 40:60 (entry 1). Although transesterifi-
cation of N-benzhydryl nitrone 1b with (E)-6 proceeded,
intramolecular cycloaddition was quite slow even at a
higher temperature for a longer reaction time (entry 2).10
Chiralities c* of R groups in N-1-phenylethyl nitrones
1c,d strongly affected diastereofacial selectivities in the
reactions employing (E)-6 (entries 3 and 4). Cycloaddition
of nitrone 1c having an (S)-1-phenylethyl group with
(E)-6 took place in a highly stereoselective manner to
afford 15c in high yield (entry 3). In contrast, reaction
of nitrone 1d bearing an (R)-1-phenylethyl group with
(13) (a) Heathcock, C. H. In Asymmetric Synthesis; Morrison, J . D.,
Ed.; Academic Press: London, 1984; Vol.3, p 111. (b) Masamune, S.;
Choy, W.; Petersen, J . S.; Sita, L. R. Angew. Chem., Int. Ed. Engl. 1985,
24, 1.

2606 J . Org. Chem., Vol. 66, No. 8, 2001
Tamura et al.
Ta ble 2. Ta n d em Tr a n sester ifica ton a n d In tr a m olecu la r Cycloa d d ition of A-Meth oxyca r bon yln itr on es 1a -d w ith Allyl
Alk oh ol (E)-6a
a
b
All the reactions were carried out with MS4A in 1,2-dichloroethane. The yields were obtained after purification by column
chromatography on silica gel. c The ratios were estimated on the basis of the 1H NMR spectra of the diastereomeric mixtures. Not
d
determined.10
Sch em e 8a
a
(a) 0.1 equiv TiCl4, MS4A, ClCH2CH2Cl.
F igu r e 3. Selected coupling constants of 14a , 15b, 15c, and
15d and NOEs of 15a , 15c and 15d .
could also be observed in reactions with achiral allyl
alcohols. As shown in Scheme 8, the reaction of 1c with
(Z)-cinnamyl alcohol (Z)-16 gave a ca. 1:1.8 mixture of
si-si attack product 17 and re-re attack product 18. In
contrast, use of the (E)-cinnamyl alcohol (E)-16 showed
a much higher diastereofacial selectivity. Thus, a similar
reaction of 1c with (E)-16 gave a 1:4 mixture of si-si
attack product 19 and re-re attack product 20.
Sch em e 7
As shown in Figure 4, the information on the effects
of chiralities b* and c* can be summarized by using a
(Z)-nitrone transition state model. In cases where (Z)-
allyl alcohols are used, chirality b* is the most important
factor in terms of A(1,3)-strain between R1 and R2. In
cases where (E)-allyl alcohols are used, the chirality c*
becomes the most important factor due to the closeness
between R and R1.
4. Geom etr y-Differ en tia ted In tr a m olecu la r Cy-
cloa d d ition . In our recent work, we developed a method
for preparing cycloadducts of (Z)-allyl alcohols from
geometrical mixtures of allyl alcohols, namely, geometry-
reofacial selectivities of the reaction with (E)-allyl alcohol,
in contrast to the small effects of chirality c* of 1 in the
reaction with (Z)-allyl alcohol (Z)-6. In the reactions of
1c and 1d , the (S)-1-phenylethyl group favors re face
attack of the nitrone, and the (R)-1-phenylethyl group
has a tendency of si face attack.14
The difference between the effects of chirality c* with
the use of a (Z)-allyl alcohol and the use of its (E)-isomer
(14) (a) Baldwin, S. W.; McFadyen, R. B.; Aube´, J .; Wilson, J . D.
Tetrahedron Lett. 1991, 32, 4431. (b) Aurich, H. G.; Quintero, J .-L. R.
Tetrahedron 1994, 50, 3929. See also refs 4a and 8g.

R-Methoxycarbonylnitrones
J . Org. Chem., Vol. 66, No. 8, 2001 2607
L via the equilibrium, giving mainly 7 and 8 by way of
transition states D and E (Scheme 5). Furthermore, in
the case of using 1b having a bulky R group, 7b was
obtained in high selectivity among the four possible
isomers by a combination of good (Z)-selection and high
diastereofacial selection, as depicted in Scheme 5.
Con clu sion
We have investigated diastereofacial selectivity of
intramolecular cycloaddition of chiral R-allyloxycarbonyl
nitrones by employing the tandem process. Although both
the (E)-nitrone transition state and the (Z)-nitrone
transition state can be considered, each stereochemical
outcome obtained in this work is well explained by taking
into account the (Z)-nitrone transition state (typically,
Figure 4). Further applications of the present tandem
cycloaddition are now currently under investigation.15
F igu r e 4. Major factors affecting diastereofacial selectivities
in the intramolecular cycloaddition.
Sch em e 9a
Exp er im en ta l Section
Melting points are uncorrected. 1H NMR (270 MHz) and 13C
NMR (67.8 MHz) spectra were measured in CDCl3. Mass
spectra were determined at an ionizing voltage of 70 eV.
Unless oterwise stated, all reactions were performed in flame-
dried glassware under an atmosphere of dry argon. Solutions
were evaporated under reduced pressure with a rotary evapo-
rator. The nitrones, methyl [(phenylmethyl)imino] acetate
N-oxide (1a ),5b methyl [(diphenylmethyl)imino]acetate N-oxide
(2b),5b methyl [(S)-(1-phenylethyl)imino]acetate N-oxide (2c),16
and methyl [(R)-(1-phenylethyl)-imino]acetate N-oxide (2d )16
were prepared by reported methods.
Gen er a l P r oced u r e for th e Ta n d em Tr a n sester ifica -
tion a n d In tr a m olecu la r Cycloa d d ition of th e Nitr on es
1 w ith Allyl Alcoh ols (5 or 6). To a stirred suspension of an
allyl alcohol 5 (or 6) and MS4A in 1,2-dichloroethane was
added successively titanium tetrachloride (0.1 M solution in
1,2-dichloroethane) and a solution of nitrone 1 in 1,2-dichlo-
roethane at room temperature. After stirring under the
conditions indicated in Schemes 3, 4, 8, and 9 and Tables 1
and 2, a small amount of water was added to the mixture and
the mixture was stirred for 1 h. The mixture was filtered
through a pad of Celite, and then the filtrate was diluted with
water, extracted with CH2Cl2, and dried over MgSO4. After
filtration, the filtrate was concentrated in vacuo. The residue
was purified by column chromatography on silica gel to give
cyclized products.
a
(a) 0.1 equiv TiCl4, MS4A, ClCH2CH2Cl, rt, 92 h. (b) The
mixture after completion of the reaction contained a 35:65 ratio
of (Z)-5 and (E)-5 (GLC analysis, OV-1, 0.2 mm × 25 m).
Sch em e 10
differentiated cycloaddition.5b We applied this method to
diastereofacial selective cycloaddition by employing mix-
tures of primary allyl alcohol (Z)- and (E)-5. As shown
in Scheme 9, treatment of N-benzyl nitrone 1a with an
excess amount of a 1:1 mixture of (Z)-5 and (E)-5 under
conditions similar to those in Scheme 3 afforded mainly
the cycloadduct 7a along with two diastereomers. In the
case where a bulkier N-benzhydryl nitrone 1b was used,
the reaction proceeded more (Z)-selectively to afford only
7b and 8b with high diastereofacial selectivity (7b/8b),
which was the same as that obtained in Scheme 3. After
completion of the reaction, GLC analysis revealed that
(Z)-5 selectively reacted with 1b.
(1R,4R,5R)-4-[(S)-1-(ter t-Bu tyld im eth ylsilyloxy)eth yl]-
2-p h e n ylm e t h yl-3,7-d ioxa -2-a za b icyclo[3.3.0]oct a n -8-
on e (7a ) a n d Its (1S,4S,5S)-Isom er (8a ). Following general
procedure, a diastereomeric mixture of 7a and 8a (345 mg,
81%, 7a /8a ) 89:11) was obtained from 1a (217 mg, 1.1 mmol),
(Z)-5 (356 mg, 1.6 mmol), titanium tetrachloride (0.1 M
solution in 1,2-dichloroethane, 1.70 mL, 0.17 mmol), and MS4A
(780 mg) after purification by column chromatography on silica
gel (CH2Cl2-Et2O, 30:1). Further purification by column
chromatography on silica gel (Et2O-hexane, 1:2) gave pure
7a and 8a . 7a : mp 98-99 °C (hexane-AcOEt); [R]25 +41.2
D
(c 1.53, CHCl3); IR (CHCl3) 1780 cm-1; H NMR δ 0.09 (3 H,
1
s), 0.10 (3 H, s), 0.87 (9 H, s), 1.26 (3 H, d, J ) 5.9 Hz), 3.54 (1
H, tdd, J ) 8.1, 7.1, and 3.6 Hz), 3.99 (1 H, dq, J ) 7.1 and 5.9
Hz), 4.00 (1 H, d, J ) 13.5 Hz), 4.01 (1 H, d, J ) 8.1 Hz), 4.08
(1 H, br t, J ) 7.1 Hz), 4.26 (1 H, d, J ) 13.5 Hz), 4.28 (1 H,
dd, J ) 9.9 and 8.1 Hz), 4.71 (1 H, dd, J ) 9.9 and 3.6 Hz),
7.16-7.33 (5 H, m); 13C NMR δ -4.7, -3.7, 17.8, 21.9, 25.6,
44.1, 66.6, 66.8, 82.9, 127.7, 128.4, 129.1, 136.0, 174.0; HRMS
calcd for C20H31NO4Si 377.2022, found 377.2026. Anal. Calcd
Since geometrical isomerization of the allyl alcohols
does not take place under the reaction conditions (Schemes
3, 4 and 8; Tables 1 and 2), transesterification may play
a key role of the (Z)-selectivities observed (Scheme 10).
Transesterification of 1 with (Z)- and (E)-5 takes place
to give intermediates L and M. Interconversion between
L and M may be sufficiently rapid due to facile transes-
terification of esters L and M with (E)-5 and (Z)-5,
respectively. Moreover, cycloaddition of M should be
unfavorable due to their steric interactions in the transi-
tion states F and G, as shown in Scheme 5. As a result,
the intramolecular cycloaddition mainly proceeds from
(15) For example, see: (a) Tamura, O.; Mita, N.; Kusaka, N.; Suzuki,
H.; Sakamoto, M. Tetrahedron Lett. 1997, 38, 429. (b) Tamura, O.;
Yoshida, S.; Sugita, H.; Mita, N.; Uyama, Y.; Morita, N.; Ishiguro, M.;
Kawasaki, T.; Ishibashi, H.; Sakamoto, M. Synlett 2000, 1553.
(16) Belzecki, C.; Panfil, I. J . Org. Chem. 1979, 44, 1212.

2608 J . Org. Chem., Vol. 66, No. 8, 2001
Tamura et al.
for C20H31NO4Si: C, 63.63; H, 8.28; N, 3.71. Found: C, 63.49;
55.45; H, 4.19; N, 9.19. Found: C, 54.98; H, 4.23; N, 9.11.
Cr ysta l Da ta of 9. Colorless and column single crystals of 9
were grown in hexane-AcOEt and a single crystal (0.20 mm
× 0.20 mm × 0.20 mm) was mounted on a glass fiber. All
measurements were made on a Rigaku AFC7R difractometer
with filtered Cu KR radiation (λ ) 1.54178 Å) and a rotating
anode generater at 296 K. The structure was solved by direct
method with SHELEXS-86 and refined by the full-matrix
least-squares method. All the non-hydrogen atoms were
refined anisotropically, and all hydrogen refinements were
located by calculation. The final cycles of the least-squares
were based on 1431 observed reflections (I > 3.00s I/I) and
H, 8.36; N, 3.64. 8a : mp 102-103.5 °C (hexane-Et2O); [R]25
D
-38.4 (c 1.28, CHCl3); IR (CHCl3) 1781 cm-1; 1H NMR δ -0.07
(3 H, s), 0.02 (3 H, s), 0.83 (9 H, s), 1.16 (3 H, d, J ) 6.3 Hz),
3.47 (1 H, tdd, J ) 8.3, 5.9, and 4.9 Hz), 3.89-4.03 (1 H, m),
3.98 (1 H, d, J ) 8.3 Hz), 4.03 (1 H, d, J ) 12.8 Hz), 4.20 (1 H,
dd, J ) 7.9 and 5.9 Hz), 4.27 (1 H, d, J ) 12.8 Hz), 4.29 (1 H,
dd, J ) 9.6 and 8.3 Hz), 4.46 (1 H, dd, J ) 9.6 and 4.9 Hz),
7.28-7.43 (5 H, m); HRMS calcd for C20H31NO4Si 377.2022,
found 378.2127. Anal. Calcd for C20H31NO4Si: C, 63.63; H,
8.28; N, 3.71. Found: C, 63.33; H, 8.31; N, 3.68.
(1R,4R,5R)-4-[(S)-1-(ter t-Bu tyld im eth ylsilyloxy)eth yl]-
2-d ip h en ylm et h yl-3,7-d ioxa -2-a za b icyclo[3.3.0]oct a n -8-
on e (7b) a n d Its (1S,4S,5S)-Isom er (8b). Following general
procedure, a diastereomeric mixture of 7b and 8b (359 mg,
81%, 7b/8b ) 89:11) was obtained from 1b (269 mg, 1.0 mmol),
(Z)-5 (324 mg, 1.5 mmol), titanium tetrachloride (0.1 M
solution in 1,2-dichloroethane, 1.0 mL, 0.1 mmol), and MS4A
(600 mg) after purification by column chromatography on silica
gel (CH2Cl2-Et2O, 30:1). Further purification by column
chromatography on silica gel (Et2O-hexane, 1:2) gave pure
299 variable parameters. Formula C21
H19N3O9, formula weight
) 457.40, monoclinic, space group P21 (#4), a ) 5.877(1) Å, b
) 10.147(1) Å, c ) 17.4531(7) Å, b ) 96.234(8)°, V ) 1034.6(2)
Å, Z ) 2, Dcalc ) 1.468 g/cm3, R ) 0.029 (Rw ) 0.042), GOF )
1.18.
(1S,4R,5R ,6S)-6-Met h yl-2-(p h en ylm et h yl)-4-p iva loyl-
oxym eth yl-3,7-d ioxa -2-a za bicyclo[3.3.0]octa n -8-on e (12a )
a n d It s (1R,4S,5R,6S)-Isom er (13a ) (Ta ble 1, E n t r y 1).
Following general procedure, a diastereomeric mixture of 12a
and 13a (40.8 mg, 100%, 12a /13a ) 95:5) was obtained from
1a (22.7 mg, 0.12 mmol), (Z)-6 (32.6 mg, 0.18 mmol), titanium
tetrachloride (0.1 M solution in 1,2-dichloroethane, 120 µL, 12
µmol), and MS4A (220 mg) after purification by column
chromatography on silica gel (CH2Cl2-Et2O, 20:1). Further
purification by column chromatography on silica gel (AcOEt-
hexane, 1:3) gave pure 12a (38.8 mg) and 13a (2.0 mg). 12a :
mp 131-132 °C (AcOEt-hexane); [R]25D +4.0 (c, 1.14, CHCl3);
7b and 8b. 7b: mp 180-182 °C (hexane-AcOEt); [R]25 +9.4
D
(c 1.17, CHCl3); IR (CHCl3) 1782 cm-1; H NMR δ 0.05 (3 H,
1
s), 0.06 (3 H, s), 0.86 (9 H, s), 1.06 (3 H, d, J ) 5.9 Hz), 3.50 (1
H, dtd, J ) 8.6, 7.9, and 3.3 Hz), 3.89 (1 H, dq, J ) 7.9 and 5.9
Hz), 4.10 (1 H, d, J ) 13.5 Hz), 4.11 (1 H, d, J ) 8.6 Hz), 4.21
(1 H, dd, J ) 9.9 and 7.9 Hz), 4.66 (1 H, dd, J ) 9.9 and 3.3
Hz), 5.05 (1 H, br s), 7.15-7.58 (10 H, m); 13C NMR δ -4.9,
-3.4, 17.8, 21.9, 25.7, 43. 4, 65.9, 66.3, 66.7, 72.4, 82.8, 127.2,
127.6, 127.8, 127.9, 128.2, 128.8, 140.5, 140.9, 174.0; HRMS
calcd for C26H35NO4Si 453.2336, found 453.2343. Anal. Calcd
for C26H35NO4Si: C, 68.84; H, 7.78; N, 3.09. Found: C, 68.76;
1
IR (CHCl3) 1784, 1732 cm-1; H NMR δ 1.20 (9 H, s), 1.42 (3
H, d, J ) 6.4 Hz, spin saturation at δ ) 3.18 f 4% NOE; at δ
) 4.73 f 7% NOE), 3.18 (1 H, td, J ) 7.8 and 3.3 Hz, spin
saturation at δ ) 1.42 f 11% NOE; at δ ) 4.11 f 4% NOE;
δ ) 4.51 f 12% NOE), 4.04 (1 H, d, J ) 13.2 Hz), 4.11 (1 H,
d, J ) 7.8 Hz, spin saturation at δ ) 3.18 f 13% NOE), 4.23
(1 H, d, J ) 13.2 Hz), 4.26 (1 H, dd, J ) 11.7 and 6.3 Hz), 4.31
(1 H, dd, J ) 11.7 and 5.9 Hz), 4.51 (1 H, br q, J ) 6.4 Hz,
spin saturation at δ ) 3.18 f 10% NOE), 4.73 (1 H, qd, J )
6.4 and 3.3 Hz, spin saturation at δ ) 1.42 f 15% NOE), 7.26-
7.43 (5 H, m); 13C NMR δ: 23.2, 27.5, 39.2, 52.1, 61.5, 67.8,
74.8, 76.6, 128.2, 128.9, 129.5, 136.2, 173.3, 178.5; HRMS calcd
for C19H25NO5 347.1733, found 347.1736. Anal. Calcd for
H, 7.81; N, 3.08. 8b: mp 228-130 °C (hexane-AcOEt); [R]25
D
-32.9 (c 1.25, CHCl3); IR (CHCl3) 1782 cm-1; 1H NMR δ -0.40
(3 H, s), -0.13 (3 H, s), 0.63 (9 H, S), 1.07 (3 H, d, J ) 6.3 Hz),
3.33 (1 H, tdd, J ) 8.6, 5.9, and 4.3 Hz), 8.82 (1 H, dq, J ) 7.9
and 6.3 Hz), 3.98 (1 H, d, J ) 8.6 Hz), 4.12 (1 H, dd, J ) 9.9
and 8.6 Hz), 4.18 (1 H, dd, J ) 7.9 and 5.9 Hz), 4.32 (1 H, dd,
J ) 9.9 and 4.3 Hz), 5.09 (1 H, s), 7.06-7.50 (10 H, m); 13C
NMR δ -5.1, -4.8, 18.2, 21.2, 25.9, 44.4, 66.0, 67.0, 73.0, 84.0,
127.5, 127.9, 128.1, 128.3, 128.7, 129.1, 141.0, 141.7, 174.4;
HRMS calcd for C26H35NO4Si 453.2336, found 453.2343. Anal.
Calcd for C26H35NO4Si: C, 68.84; H, 7.78; N, 3.09. Found: C,
68.63; H, 7.80; N, 3.09.
C
19H25NO5: C, 65.69; H, 7.25; N, 4.03. Found: C, 65.74; H,
7.30; N, 3.96. 13a : mp 130-130.5 °C (AcOEt-hexane); [R]25
D
+10.1 (c 1.19, CHCl3); IR (CHCl3) 1784, 1726 cm-1; 1H NMR δ
1.19 (9 H, s), 1.59 (3 H, d, J ) 6.6 Hz), 3.47 (1 H, td, J ) 7.6
and 6.6 Hz), 4.05 (1 H, d, J ) 13.2 Hz), 4.13 (1 H, d, J ) 7.6
Hz), 4.22 (1 H, d, J ) 13.2 Hz), 4.22-4.62 (3 H, m), 4.72 (1 H,
quin, J ) 6.6 Hz), 7.22-7.47 (5 H, m); 13C NMR δ 17.1, 27.0,
38.7, 48.7, 61.6, 69.3, 75.7, 77.2, 127.8, 128.4, 129.2 135.6,
173.0, 178.1; HRMS calcd for C19H25NO5 347.1733, found
347.1738. Anal. Calcd for C19H25NO5: C, 65.69; H, 7.25; N,
4.03. Found: C, 65.45; H, 7.37; N, 3.95.
(1R,4R,5R)-4-[(S)-1-(3,5-Dinitrobenzoyl)oxyethyl]-2-phen-
ylm eth yl-3,7-d ioxa -2-a za bicyclo[3.3.0]octa n -8-on e (9). To
a stirred solution of 7a (200 mg, 0.53 mmol) in MeOH (10 mL)
was added dropwise acetyl chloride (3 mL) at 0 °C. After
stirring for 30 min under the same conditions, the mixture
was concentrated in vacuo. The residue was diluted with a
saturated solution of NaHCO3, and the mixture was extracted
with AcOEt. The organic phase was washed with a saturated
solution of NaCl and dried over MgSO4. After filtration, the
filtrate was concentrated in vacuo to give crude (1R,4R,5R)-
4-[(S)-1-hydroxyethyl]-2-phenylmethyl-3,7-dioxa-2-azabicyclo-
[3.3.0] octan-8-one (131 mg). The crude product was dissolved
in THF (10 mL) and then Et3N (0.40 mL, 2.75 mmol) and 3,5-
dinitrobenzoyl chloride (605 mg, 2.62 mmol) was added to the
solution at room temperature. After stirring for 30 min, a
saturated solution of NaHCO3 (1 mL) was added to the
mixture, and the mixture was vigorously stirred for 30 min.
The mixture was diluted with water and extracted with CH2-
Cl2. The organic phase was washed with a saturated solution
of NaCl and dried over MgSO4. After filtration, the filtrate was
concentrated in vacuo to give the residue, which was purified
by column chromatography on silica gel (AcOEt-hexane) to
(1S,4R,5R,6S)-6-Meth yl-2-(d ip h en ylm eth yl)-4-p iva loyl-
oxym eth yl-3,7-d ioxa -2-a za bicyclo[3.3.0]octa n -8-on e (12b)
a n d It s (1R,4S,5S,6S)-Isom er (13b) (Ta ble 1, E n t r y 2).
Following general procedure, a diastereomeric mixture of 12b
and 13b (47.3 mg, 100%, 12b/13b ) 95:5) was obtained from
1b (30.1 mg, 0.17 mmol), (Z)-6 (31.7 mg, 0.17 mmol), titanium
tetrachloride (0.1 M solution in 1,2-dichloroethane, 110 µL, 11
µmol), and MS4A (200 mg) after purification by column
chromatography on silica gel (CH2Cl2-Et2O, 20:1). Further
purification by preparative TLC on silica gel (Et2O-hexane,
1:1) gave pure 12b (44.9 mg) and 13b (2.4 mg). 12b: mp 225-
227 °C (AcOEt-hexane); [R]25 -11.1 (c 1.02, CHCl3); IR
D
1
(CHCl3) 1784, 1732 cm-1; H NMR δ 1.10 (9 H, s), 1.35 (3 H,
afford 9 (172 mg, 71%): mp 154 °C (hexane-AcOEt). [R]25
d, J ) 6.6 Hz), 3.17 (1 H, td, J ) 8.1 and 2.3 Hz), 4.19 (1 H,
dd, J ) 11.7 and 6.6 Hz), 4.24 (1 H, d, J ) 8.1 Hz), 4.26 (1 H,
dd, J ) 11.7 and 5.6 Hz), 4.54 (1 H, br q, J ) 6.4 Hz), 4.68 (1
H, qd, J ) 6.6 and 2.3 Hz), 5.01 (1 H, br s), 7.17-7.56 (10 H,
m); 13C NMR δ 22.9, 26.9, 38.7, 50.7, 61.1, 66.3, 72.5, 73.8,
76.0, 127.4, 127.5, 127.7, 127.9, 128.4, 128.9, 140.4, 140.6,
173.1, 177.9; HRMS calcd for C25H29NO5 423.2046, found
423.2045. Anal. Calcd for C25H29NO3: C, 70.90; H, 6.90; N,
3.31. Found: C, 71.02; H, 6.99; N, 3.24. 13b: mp 188-189 °C
D
+55.9 (c 0.501, CHCl3); IR (CHCl3) 1792, 1736 cm-1; 1H NMR
δ 1.45 (3 H, d, J ) 6.6 Hz), 3.60 (1 H, qd, J ) 6.6 and 3.0 Hz),
4.00 (1 H, d, J ) 13.0 Hz), 4.05 (1 H, d, J ) 6.9 Hz), 4.21 (1 H,
d, J ) 13.0 Hz), 4.29 (1 H, dd, J ) 10.0 and 6.9 Hz), 4.44 (1 H,
t, J ) 6.9 Hz), 4.49 (1 H, dd, J ) 10.0 and 3.0 Hz), 5.40 (1 H,
quin, J ) 6.9 Hz), 7.15-7.45 (5 H, m), 9.06 (2 H, br d, J ) 2.0
Hz), 9.17 (1 H, br t, J ) 2.0 Hz); HRMS calcd for C21H19N3O9:
457.1122, found 457.1124. Anal. Calcd for C21H19N3O9: C,

R-Methoxycarbonylnitrones
J . Org. Chem., Vol. 66, No. 8, 2001 2609
(CHCl3-hexane); [R]25 +12.6 (c 0.870, CHCl3); IR (CHCl3):
(1S,4S,5R,6S)-6-Met h yl-2-(p h en ylm et h yl)-4-p iva loyl-
oxym eth yl-3,7-d ioxa -2-a za bicyclo[3.3.0]octa n -8-on e (14a )
a n d It s (1R,4R,5S,6S)-Isom er (15a ) (Ta ble 2, E n t r y 1).
Following general procedure, a diastereomeric mixture of 14a
and 15a (75.0 mg, 71%, 14a /15a ) 40:60) was obtained from
1a (59.1 mg, 0.31 mmol), (E)-6 (85.5 mg, 0.46 mmol), titanium
tetrachloride (0.1 M solution in 1,2-dichloroethane, 300 µL, 30
µmol), and MS4A (410 mg) after purification by column
chromatography on silica gel (CH2Cl2-Et2O, 30:1). Further
purification by column chromatography on silica gel (AcOEt-
hexane, 1:3) gave pure 14a (30.0 mg) and 15a (45.0 mg). 14a :
D
1790, 1725 cm-1; 1H NMR δ 1.05 (9 H, s), 1.56 (3 H, d, J ) 5.6
Hz), 3.46 (1 H, td, J ) 7.6 and 5.6 Hz), 4.17 (1 H, dd, J ) 11.6
and 7.6 Hz), 4.27 (1 H, d, J ) 7.6 Hz), 4.51 (1 H, td, J ) 7.6
and 3.6 Hz), 4.60 (1 H, dd, J ) 11.6 and 3.6, Hz), 4.63 (1 H,
quin, J ) 5.6 Hz), 5.01 (1 H, br s), 7.17-7.55 (10 H, m); 13C
NMR δ 17.2, 26.9, 38.6, 48.3, 61.7, 68.6, 72.4, 35.4, 77.3, 127.4,
127.7, 127.8, 127.9, 128.4, 128.9, 140.5, 140.5, 173.1, 178.1;
HRMS calcd for C25H29NO5 423.2046, found: 423.2044.
(1S,4R,5R,6S)-6-Meth yl-2-[(S)-1-ph en yleth yl]-4-pivaloyl-
oxym eth yl-3,7-d ioxa -2-a za bicyclo[3.3.0]octa n -8-on e (12c)
a n d It s (1R,4S,5S,6S)-Isom er (13c) (Ta b le 1, E n t r y 3).
Following general procedure, a diastereomeric mixture of 12c
and 13c (41.6 mg, 100%, 12c/13c ) 92:8) was obtained from
1c (23.8 mg, 0.16 mmol), (Z)-6 (32.0 mg, 0.17 mmol), titanium
tetrachloride (0.1 M solution in 1,2-dichloroethane, 115 µL,
11.5 µmol), and MS4A (220 mg) after purification by column
chromatography on silica gel (CH2Cl2-Et2O, 30:1). Further
purification by column chromatography on silica gel (AcOEt-
hexane, 1:4) gave pure 12c (38.3 mg) and 13c (3.3 mg). 12c:
mp 99-99.5 °C (Et2O-hexane); [R]25 -25.1 (c 1.10, CHCl3);
D
IR (CHCl3) 1774, 1730 cm-1; H NMR δ 1.19 (9 H, s), 1.43 (3
1
H, d, J ) 6.6 Hz), 2.90 (1 H, dt, J ) 8.2 and 4.0 Hz), 3.88 (1 H,
d, J ) 8.2 Hz), 4.07-4.26 (3 H, m), 4.17 (1 H, d, J ) 14.2 Hz),
4.30 (1 H, d, J ) 14.2 Hz), 4.69 (1 H, qd, J ) 6.6 and 4.0 Hz,),
7.25-7.41 (5 H, m); 13C NMR δ 22.0, 27.1, 38.7, 54.5, 63.6,
79.9, 81.0, 127.5, 128.2, 129.0, 136.1, 172.0, 178.0; HRMS calcd
for C19H25NO5 347.1732, found 347.1733. Anal. Calcd for
C19H25NO5: C, 65.69; H, 7.25; N, 4.03. Found: C, 65.45; H,
7.20; N, 4.02. 15a : mp 80-81 °C (Et2O-hexane); [R]25D -56.3
mp 153-153.5 °C (AcOEt-hexane); [R]25 -41.1 (c 1.05,
D
1
(c 1.12, CHCl3); IR (CHCl3) 1772, 1726 cm-1; H NMR δ 1.19
CHCl3); IR (CHCl3) 1782, 1732 cm-1; 1H NMR δ 1.14 (9 H, s),
1.39 (3 H, d, J ) 6.6 Hz), 1.50 (3 H, d, J ) 6.6 Hz), 3.06 (1 H,
ddd, J ) 9.9, 8.3, and 4.0 Hz), 4.12 (1 H, d, J ) 8.3 Hz), 4.15-
4.31 (4 H, m), 4.68 (1 H, qd, J ) 6.6 and 4.0 Hz), 7.28-7.42 (5
H, m); 13C NMR δ 22.0, 22.9, 27.5, 39.2, 53.2, 61.6, 74.7, 76.5,
(9 H, s), 1.49 (3 H, d, J ) 6.6 Hz, spin saturation at δ ) 4.74
f 8% NOE), 3.18 (1 H, ddd, J ) 7.9, 6.6, and 5.3 Hz), 4.08 (1
H, d, J ) 7.9 Hz, spin saturation at δ ) 3.18 f 12% NOE),
4.12 (1 H, dd, J ) 11.6 and 5.3 Hz), 4.22 (1 H, d, J ) 13.9 Hz),
4.25 (1 H, dd, J ) 11.6 and 6.6 Hz), 4.32 (1 H, d, J ) 13.9 Hz),
4.44 (1 H, br q, J ) 5.3 Hz), 4.74 (1 H, quin, J ) 6.6 Hz, spin
saturation at δ ) 3.18 f 14% NOE), 7.27-7.43 (5 H, m); 13C
NMR d: 17.1, 27.1, 38.7, 49.7, 64.3, 74.8, 76.5, 127.5, 128.3,
128.9 136.3, 172.6 178.2; HRMS calcd for C19H25NO5 347.1733,
found 347.1729. Anal. Calcd for C19H25NO5: C, 65.69; H, 7.25;
N, 4.03. Found: C, 65.80; H, 7.31; N, 4.00.
128.1, 128.2, 128.8, 142.0, 173.8, 188.2; HRMS calcd for C20H27
-
NO5 361.1889, found 361.1885. Anal. Calcd for C20H27NO5: C,
66.46; H, 7.53; N, 3.88. Found: C, 66.30; H, 7.66; N, 3.79.
13c: mp 159-159.5 °C (AcOEt-hexane); [R]20D -39.7 (c 0.920,
CHCl3); IR (CHCl3) 1790, 1727 cm-1; 1H NMR δ 1.22 (9 H, s),
1.50 (3 H, d, J ) 6.6 Hz), 1.57 (3 H, d, J ) 6.9 Hz), 3.38 (1 H,
br td, J ) 7.6 and 5.6 Hz), 3.86 (1 H, br q, J ) 6.6 Hz), 4.14 (1
H, d, J ) 7.9 Hz), 4.25 (1 H, dd, J ) 11.6 and 7.6 Hz), 4.51 (1
H, td, J ) 7.6 and 3.3 Hz), 4.60 (1 H, quin, J ) 6.9 Hz), 4.67
(1 H, dd, J ) 11.6 and 3.3 Hz), 7.30-7.44 (5 H, m); 13C NMR
δ 17.5, 21.9, 27.5, 39.2, 48.2, 53.2, 62.3, 64.2, 69.6, 75.8, 128.1,
128.6, 129.4, 141.9, 173.8, 178.6; HRMS calcd for C20H27NO5
361.1889, found 361.1893. Anal. Calcd for C20H27NO5: C,
66.46; H, 7.53; N, 3.88. Found: C, 66.07; H, 7.63; N, 3.78.
(1S,4S,5R,6S)-6-Meth yl-2-[(S)-1-ph en yleth yl]-4-pivaloyl-
oxym eth yl-3,7-d ioxa -2-a za bicyclo[3.3.0]octa n -8-on e (14c)
a n d It s (1R,4R,5S,6S)-Isom er (15c) (Ta ble 2, E n t r y 3).
Following general procedure, a diastereomeric mixture of 14c
and 15c (90.7 mg, 100%, 14c/15c ) 8:92) was obtained from
1c (51.9 mg, 0.25 mmol), (E)-6 (70.0 mg, 0.38 mmol), titanium
tetrachloride (0.1 M solution in 1,2-dichloroethane, 750 µL, 75
µmol), and MS4A (440 mg) after purification by column
chromatography on silica gel (CH2Cl2-Et2O, 30:1). Further
purification by column chromatography on silica gel (AcOEt-
hexane, 1:6) gave pure 14c (7.3 mg) and 15c (83.4 mg). 14c:
(1S,4R,5R,6S)-6-Meth yl-2-[(R)-1-ph en yleth yl]-4-pivaloyl-
oxym eth yl-3,7-d ioxa -2-a za bicyclo[3.3.0]octa n -8-on e (12d )
a n d It s (1R,4S,5S,6S)-Isom er (13d ) (Ta b le 1, E n t r y 4).
Following general procedure, a diastereomeric mixture of 12d
and 13d (35.6 mg, 100%, 12d /13d ) 93:7) was obtained from
1d (20.4 mg, 98 µmol), (Z)-6 (32.0 mg, 0.17 mmol), titanium
tetrachloride (0.1 M solution in 1,2-dichloroethane, 100 µL, 10
µmol), and MS4A (210 mg) after purification by column
chromatography on silica gel (CH2Cl2-Et2O, 20:1). Further
purification by column chromatography on silica gel (AcOEt-
hexane, 1:4) gave pure 12d (33.1 mg) and 13d (2.5 mg). 12d :
mp 90-92 °C (Et2O-hexane); [R]25 -30.8 (c 1.17, CHCl3); IR
D
(CHCl3) 1771, 1728 cm-1; H NMR δ 1.18 (9 H, s), 1.34 (3 H,
1
d, J ) 6.3 Hz), 1.52 (3 H, d, J ) 6.6 Hz), 2.70 (1 H, ddd, J )
7.9, 4.0, and 3.3 Hz), 3.77 (1 H, d, J ) 7.9 Hz), 4.04 (1 H, dd,
J ) 10.9 and 4.0 Hz), 4.19 (1 H, dt, J ) 5.9 and 4.0 Hz), 4.27
(1 H, dd, J ) 10.9 and 5.9 Hz), 4.36 (1 H, br), 4.63 (1 H, qd, J
) 6.6 and 3.3 Hz), 7.24-7.46 (5 H, m); 13C NMR δ 21.5, 22.1,
27.1, 38.7, 53.8, 63.5, 79.8, 80.3, 127.6 128.2, 128.4, 140.5,
172.4, 178.0; HRMS calcd for C20H27NO5 361.1889, found
361.1890. Anal. Calcd for C20H27NO5: C, 66.46; H, 7.53; N,
3.88. Found: C, 66.26; H, 7.55; N, 3.79. 15c: mp 101-101.5
mp 119-119.5 °C (AcOEt-hexane); [R]25 +47.1 (c 0.850,
D
CHCl3); IR (CHCl3) 1790, 1732 cm-1; 1H NMR δ 1.22 (9 H, s),
1.34 (3 H, d, J ) 6.6 Hz), 1.49 (3 H, d, J ) 6.3 Hz), 3.09 (1 H,
td, J ) 7.8 and 2.3 Hz, spin saturation at δ ) 1.34 f 9% NOE;
at δ ) 4.51 f 3% NOE), 3.90 (1 H, q, J ) 6.3 Hz), 4.10 (1 H,
d, J ) 7.8, Hz, spin saturation at δ ) 3.09 f 10% NOE), 4.30
(1 H, dd, J ) 11.7 and 6.3 Hz), 4.35 (1 H, dd, J ) 11.7 and 5.6
Hz), 4.51 (1 H, br q, J ) 6.4 Hz), 4.68 (1 H, qd, J ) 6.6 and 2.3
Hz, spin saturation at δ ) 1.34 f 14% NOE), 7.26-7.45 (5 H,
m, Ar-H); 13C NMR δ 21.7, 22.9, 27.0, 38.8, 50.2, 61.2, 64.0,
66.3, 73.8, 75.9, 127.6, 128.1 128.9, 141.5, 173.3, 177.9; HRMS
calcd for C20H27NO5 361.1889, found 361.1882. Anal. Calcd for
°C (Et2O-hexane); [R]25 -75.0 (c 0.800, CHCl3); IR (CHCl3)
D
1782, 1728 cm-1; 1H NMR δ 1.18 (9 H, s), 1.47 (3 H, d, J ) 6.3
Hz), 1.51 (3 H, d, J ) 6.3 Hz, spin saturation at δ ) 4.69 f
8% NOE), 3.15 (1 H, dt, J ) 6.3 and 5.6 Hz, spin saturation
at δ ) 4.69 f 12% NOE), 4.10-4.31 (3 H, m), 4.13 (1 H, br
dd, J ) 10.9 and 5.6 Hz), 4.42 (1 H, br q, J ) 5.6 Hz), 4.69 (1
H, quin, J ) 6.3 Hz, spin saturation at δ ) 1.51 f 5% NOE;
at δ ) 3.15 f 30% NOE), 7.22-7.39 (5 H, m); 13C NMR δ:
16.8, 21.7, 27.1, 38.7, 49.3, 64.9, 74.2, 126.9, 127.5, 128.6, 142.8,
172.8, 178.1; HRMS calcd for C20H27NO5 361.1889, found
361.1890. Anal. Calcd for C20H27NO5: C, 66.46; H, 7.53; N,
3.88. Found: C, 66.34; H, 7.47; N, 3.84.
C
20H27NO5: C, 66.46; H, 7.53; N, 3.88. Found: C, 66.29; H,
7.64; N, 3.83. 13d : mp 148-149 °C (AcOEt-hexane); [R]25
D
+50.6 (c 1.24, CHCl3); IR (CHCl3): 1782, 1728 cm-1; 1H NMR
δ 1.13 (9 H, s), 1.52 (3 H, d, J ) 6.6 Hz), 1.55 (3 H, d, J ) 6.6
Hz), 3.36 (1 H, dt, J ) 7.6 and 6.6 Hz), 4.11 (1 H, d, J ) 7.6
Hz), 4.11-4.35 (3 H, m), 4.48 (1 H, dd, J ) 11.2 and 3.3 Hz),
4.69 (1 H, quin, J ) 6.6 Hz), 7.25-7.45 (5 H, m); 13C NMR δ
17.4, 21.7, 26.9, 38.6, 49.9, 61.8, 67.2, 75.6, 77.3, 77.8, 127.5,
127.7, 128.3, 141.5, 173.3, 178.1; HRMS calcd for C20H27NO5
361.1889, found 361.1893. Anal. Calcd for C20H27NO5: C,
66.46; H, 7.53; N, 3.88. Found: C, 66.40; H, 7.62; N, 3.81.
(1S,4S,5R,6S)-6-Meth yl-2-[(R)-1-ph en yleth yl]-4-pivaloyl-
oxym eth yl-3,7-d ioxa -2-a za bicyclo[3.3.0]octa n -8-on e (14d )
a n d It s (1R,4R,5S,6S)-Isom er (15d ) (Ta ble 2, E n t r y 4).
Following general procedure, a diastereomeric mixture of 14d
and 15d (81.9 mg, 91%, 14d /15d ) 63:37) was obtained from
1d (51.9 mg, 0.25 mmol), (E)-6 (70.0 mg, 0.38 mmol), titanium
tetrachloride (0.1 M solution in 1,2-dichloroethane, 750 µL, 75
µmol), and MS4A (450 mg) after purification by column

2610 J . Org. Chem., Vol. 66, No. 8, 2001
Tamura et al.
1
chromatography on silica gel (CH2Cl2-Et2O, 30:1). Further
purification by column chromatography on silica gel (AcOEt-
hexane, 1:2) gave pure 14d (51.6 mg) and 15d (30.3 mg). 14d :
(Et2O-hexane); IR (CHCl3) 1786 cm-1; H NMR δ 1.71 (3 H,
d, J ) 6.6 Hz), 3.36 (1 H, br m), 4.27 (1 H, d, J ) 8.5 Hz), 4.36
(1 H, q, J ) 6.6 Hz), 4.45 (1 H, dd, J ) 10.5 and 2.6 Hz), 4.49
(1 H, dd, J ) 10.5 and 5.5 Hz), 4.97 (1 H, d, J ) 6.9 Hz), 7.37-
7.57 (10 H, m); 13C NMR δ 19.8, 50.2, 64.7, 68.5, 84.7, 126.4,
127.7, 128.0, 128.4, 128.5, 128.7, 141.5, 173.8; HRMS calcd
for C19H19NO3 309.1365, found: 309.1362. Anal. Calcd for
mp 80-81 °C (Et2O-hexane); [R]25 +56.6 (c 1.18, CHCl3); IR
D
(CHCl3) 1771, 1728 cm-1; H NMR δ 0.93 (9 H, s), 1.32 (3 H,
1
d, J ) 6.6 Hz), 1.44 (3 H, d, J ) 6.6 Hz), 2.81 (1 H, ddd, J )
5.6, 4.4, and 1.7 Hz), 4.03-4.22 (5 H, m), 4.54 (1 H, qd, J )
6.6 and 1.7 Hz), 7.15-7.32 (5 H, m); 13C NMR δ 21.0, 22.4,
27.0, 38.7, 52.3, 63.7, 64.2, 78.9, 81.9, 127.1, 127.5, 128.5, 142.4,
172.6, 177.8; HRMS calcd for C20H27NO5 361.1889, found
361.1898. Anal. Calcd for C20H27NO5: C, 66.46; H, 7.53; N,
3.88. Found: C, 66.43; H, 7.62; N, 3.83. 15d : mp 101 °C
(Et2O-hexane); [R]25D -33.9 (c 1.29, CHCl3); IR (CHCl3) 1772,
1726 cm-1; 1H NMR δ 1.21 (9 H, s), 1.46 (3 H, d, J ) 6.6 Hz),
1.47 (3 H, d, J ) 6.3 Hz, spin saturation at δ ) 4.61 f 6%
NOE), 2.94 (1 H, ddd, J ) 7.9, 6.3, and 4.8 Hz, spin saturation
at δ ) 4.61 f 7% NOE), 4.05 (1 H, dd, J ) 11.1 and 4.8 Hz),
4.2-4.6 (3 H, m), 4.61 (1 H, quin, J ) 6.3 Hz, spin saturation
at δ ) 1.43 f 16% NOE; at δ ) 2.94 f 13% NOE), 7.26-7.49
(5 H, m); 13C NMR δ 16.8, 21.8, 27.1, 38.7, 50.0, 64.0, 64.1,
64.8, 74.27, 74.34, 127.6, 128.1, 128.3, 141.8, 172.8, 178.5;
HRMS calcd for C20H27NO5 361.1889, found 361.1884. Anal.
Calcd for C20H27NO5: C, 66.46; H, 7.53; N, 3.88. Found: C,
66.80; H, 7.65; N, 3.86.
C
19H19NO3: C, 73.77; H, 6.19; N, 4.53. Found: C, 73.57; H,
6.25; N, 4.45.
Th e Ta n d em Tr a n sester ifica tion a n d In tr a m olecu la r
Cycloa d d ition of th e Nitr on e 1a w ith a Geom etr ica l
Mixtu r e of Allyl Alcoh ols (Z)-5 a n d (E)-5. Following the
general procedure, a 76:11:13 mixture of three diastereomers
7a , 8a , 10a , or 11a (52.4 mg, 89%) was obtained from 1a (30.0
mg, 0.16 mmol), 5 [50.4 mg, 0.23 mmol for each (Z)-5 and (E)-
5], 0.1 M titanium tetrachloride in 1,2-dichloroethane (155 µL,
15 µmol), and MS4A (300 mg) after column chromatography
on silica gel with CH2Cl2-Et2O (30:1). The daistereomeric
distribution of the products were determined by comparing
1
doublets due to methyl groups in the H NMR spectrum with
those of 7a and 8a (see Supporting Information). 1H NMR
(CDCl3, 270 MHz) δ -0.07 (3 H × 76/100, s), 0.02 (3 H × 11/
100 + 13/100, s), 0.05 (3 H × 13/100, s), 0.09 (3 H × 76/100,
s), 0.10 (3 H × 76/100, s), 1.11 (3 H × 13/100, d, J ) 5.9 Hz),
1.16 (3 H × 11/100, d, J ) 6.3 Hz), 1.26 (3 H × 13/100, d, J )
5.9 Hz), 3.36 (1 H × 13/100), 3.47 (1 H × 11/100, tdd, J ) 8.3,
5.9, and 4.9 Hz), 3.54 (1 H × 76/100, tdd, J ) 8.1, 7.1, and 3.6
Hz), 3.91-4.12 (3 H, m), 4.16-4.37 (2 H, m), 4.45 (1 H × 13/
100, dd, J ) 9.6 and 7.9 Hz), 4.46 (1 H × 11/100, dd, J ) 9.6
and 4.9 Hz), 4.71 (1 H × 76/100, dd, J ) 9.9 and 3.6 Hz), 7.24-
7.46 (5 H, m).
(1S,4R,5S)-4-P h en yl-2-[(S)-1-p h en yleth yl]-3,7-d ioxa -2-
a za -bicyclo[3.3.0]octa n -8-on e (17) a n d Its (1R,4S,5R)-
Isom er (18). Following general procedure, a diastereomeric
mixture of 17 and 18 (28.1 mg, 91%, 17/18 ) 36:64) was
obtained from 1c (20.7 mg, 0.10 mmol), (Z)-16 (20.1 mg, 0.15
mmol), titanium tetrachloride (0.1 M solution in 1,2-dichloro-
ethane, 100 µL, 10 µmol), and MS4A (300 mg) after purification
by column chromatography on silica gel (CH2Cl2-Et2O, 40:1).
Further purification by column chromatography on silica gel
(AcOEt-hexane, 1:3) gave pure 17 (10.1 mg) and 18 (18.0 mg).
Th e Ta n d em Tr a n sester ifica tion a n d In tr a m olecu la r
Cycloa d d ition of th e Nitr on e 1b With a Geom etr ica l
Mixtu r e of Allyl Alcoh ols (Z)-5 a n d (E)-5. Following the
general procedure, an 89:11 mixture of two diastereomers 7b
and 8b (45.0 mg, 87%) was obtained from 1b (30.0 mg, 0.11
mmol), 5 [36.0 mg, 0.17 mmol for each (Z)-5 and (E)-5], 0.1 M
titanium tetrachloride in 1,2-dichloroethane (111 µL, 10 µmol),
and MS4A (300 mg) after column chromatography on silica
gel with CH2Cl2-Et2O (40:1). The daistereomeric distribution
of the products were determined by comparing doublets due
to methyl groups in the 1H NMR spectrum with those of 7b
and 8b (see, Supporting Information). 1H NMR (CDCl3, 270
MHz) δ -0.40 (3 H × 11/100, s), -0.13 (3 H × 11/100, s), 0.05
(3 H × 89/100, s), 0.06 (3 H × 89/100, s), 0.63 (9 H × 11/100,
s), 0.86 (9 H × 89/100, s), 1.06 (3 H × 89/100, d, J ) 5.9 Hz),
1.07 (3 H, d × 11/100, J ) 6.3 Hz), 3.33 (1 H, tdd × 11/100, J
) 8.6, 5.9, and 4.3 Hz), 3.50 (1 H × 89/100, dtd, J ) 8.6, 7.9,
and 3.3 Hz), 3.89 (1 H × 89/100, dq, J ) 7.9, 5.9 Hz), 4.10 (1
H × 89/100, d, J ) 13.5 Hz), 4.11 (1 H × 89/100, d, J ) 8.6
Hz), 4.21 (1 H × 89/100, dd, J ) 9.9 and 7.9 Hz), 3.80-4.20 (3
H × 11/100, m), 4.32 (1 H × 11/100, dd, J ) 9.9 and 4.3 Hz),
4.66 (1 H × 89/100, dd, J ) 9.9 and 3.3 Hz), 5.05 (1 H × 89/
100, br s), 5.09 (1 H × 11/100, s), 7.06-7.58 (10 H, m).
17: mp 234-236 °C (AcOEt-hexane); [R]25 -155.7 (c 0.730,
D
CHCl3); IR (CHCl3) 1790 cm-1; H NMR δ 1.47 (3 H, d, J )
1
6.3 Hz), 3.51 (1 H, br qd, J ) 7.9 and 1.3 Hz), 3.60 (1 H, dd, J
) 10.2 and 1.3 Hz), 3.90 (1 H, dd, J ) 10.2 and 7.9 Hz), 4.00
(1 H, q, J ) 6.3 Hz), 4.07 (1 H, d, J ) 7.9 Hz), 5.46 (1 H, d, J
) 7.9 Hz), 7.17-7.41 (10 H, m); 13C NMR δ 21.8, 45.4, 63.7,
66.5, 66.6, 81.3, 126.8, 127.5, 128.0, 128.6, 128.8, 128.9, 134.7,
141.8, 174.4; HRMS calcd for C19H19NO3: 309.1365, found
309.1366. Anal. Calcd for C19H19NO3: C, 73.77; H, 6.19; N,
4.53. Found: C, 73.54; H, 6.22; N, 4.46. 18: mp 203-205 °C
(AcOEt-hexane); [R]25 +75.0 (c 0.640, CHCl3); IR (CHCl3)
D
1779 cm-1; 1H NMR δ 1.48 (3 H, d, J ) 6.3 Hz), 3.47 (1 H, qd,
J ) 7.3 and 2.6 Hz), 3.66 (1 H, dd, J ) 10.2 and 2.6 Hz), 3.94
(1 H, dd, J ) 10.2 and 7.3 Hz), 4.15 (1 H, d, J ) 7.3 Hz), 4.25
(1 H, br), 5.21 (1 H, d, J ) 7.3 Hz), 7.11-7.44 (10 H, m); 13C
NMR δ 22.1, 47.2, 65.3, 66.8, 81.5, 126.6, 127.6, 127.8, 128.4,
128.5, 128.7, 135.1, 141.8, 174.0; HRMS calcd for C19H19NO3:
309.1365, found 309.1362. Anal. Calcd for C19H19NO3: C,
73.77; H, 6.19; N, 4.53. Found: C, 73.47; H, 6.24; N, 4.49.
(1S,4S,5S)-4-P h en yl-2-[(S)-1-p h en yleth yl]-3,7-d ioxa -2-
a za -bicyclo[3.3.0]octa n -8-on e (19) a n d Its (1R,4S,5R)-
Isom er (20). Following general procedure, a diastereomeric
mixture of 19 and 20 (70.3 mg, 94%, 19/20 ) 20:80) was
obtained from 1c (50.0 mg, 0.24 mmol), (E)-16 (48.6 mg, 0.36
mmol), titanium tetrachloride (0.1 M solution in 1,2-dichloro-
ethane, 240 µL, 24 µmol), and MS4A (400 mg) after purification
by column chromatography on silica gel (CH2Cl2-Et2O, 40:1).
Further purification by column chromatography on silica gel
(AcOEt-hexane, 1:3) gave pure 19 (14.1 mg) and 20 (56.2 mg).
Su p p or tin g In for m a tion Ava ila ble: Further discussion
for (Z)-nitrone transition state vs (E)-nitrone transition state
containing computational results (Cartesian coodinates and
total energies) of (Z)- and (E)-nitrone transition states of
cycloaddtion of 3 (R ) Bn, R′ ) H) derived from 1a and
cyclohex-2-en-1-ol; crystallographic data of compound 9 (CIF);
experimental procedure for (Z)-5, (E)-5, (Z)-6, and (E)-6; 1H
NMR of 13b, a mixture of 7a , 8a , 10a , or 11a obtained by
reaction of 1a with a 1:1 mixture of (Z)-5 and (E)-5, a mixture
of 7b and 8b obtained by reaction of 1b with a 1:1 mixture of
(Z)-5 and (E)-5, (Z)-5, (4S,Z)-4-tert-butyldimethylsilyloxy-1-
pivaloyloxy-2-pentene, (Z)-6, (E)-5, (4S,E)-4-tert-butyldimeth-
ylsilyloxy-1-pivaloyloxy-2-pentene, and (E)-6. This material is
available free of charge via Internet at http://pubs.acs.org.
19: mp 128-130 °C (AcOEt-hexane); IR (CHCl3) 1779 cm-1
;
1H NMR δ 1.56 (3 H, d, J ) 6.9 Hz), 3.06 (1 H, dtd, J ) 9.2,
6.9, and 3.3 Hz), 3.81 (1 H, d, J ) 9.2 Hz), 4.21-4.33 (3 H, m),
4.75 (1 H, d, J ) 6.9 Hz), 7.17-7.42 (10 H, m); 13C NMR δ
21.1, 51.8, 64.4, 69.2, 84.8, 126.5, 127.7, 128.1, 128.5, 128.7,
129.0, 140.0, 173.8; HRMS calcd for C19H19NO3 309.1365,
found: 309.1365. Anal. Calcd for C19H19NO3: C, 73.77; H, 6.19;
N, 4.53. Found: C, 73.59; H, 6.27; N, 4.43. 20: mp 92-94 °C
J O001015I