Stereochemistry of the [2+2] cycloaddition of chlorosulfonyl isocyanate to chiral alkoxyallenes derived from 1,3-alkylidene-L-erythritol and -D-threitol

Article abstract of DOI:10.1002/ejoc.200400557

The [2+2] cycloaddition of chlorosulfonyl isocyanate to alkoxyallenes derived from ethylidene and benzylidene erythritols and threitols proceeds with a moderate asymmetric induction in the case of the erythritols and with a very low induction in the case of threitols. This indicates that the erythritol derivatives may exist in solution in one predominant conformation while the threitol derivatives behave as a conformational ensemble. The conformations of alkoxyallenes were studied with variety of NMR techniques as well as using ab initio calculations. The results thus obtained were in a full agreement with our predictions based on the stereoselectivity of cycloaddition. The azetidinones obtained by the cycloaddition were subjected to intramolecular alkylation at the nitrogen atom to provide the corresponding tricyclic cephams. The absolute configurations of the resultant azetidinones and cephams were assigned using NMR and CD spectroscopy. Wiley-VCH Verlag GmbH & Co, KGaA, 69451 Weinheim, Germany, 2005.

Full text of DOI:10.1002/ejoc.200400557

FULL PAPER
Stereochemistry of the [2؉2] Cycloaddition of Chlorosulfonyl Isocyanate to
Chiral Alkoxyallenes Derived from 1,3-Alkylidene-
L
-erythritol and - -threitol
D
Tong Thanh Danh,[a] Wojciech Bocian,[b] Lech Kozerski,[a,b] Patrycja Szczukiewicz,[a]
Jadwiga Frelek,[a] and Marek Chmielewski*[a]
Keywords: Alkoxyallenes / [2ϩ2] Cycloaddition / β-Lactams
The [2+2] cycloaddition of chlorosulfonyl isocyanate to alkoxy-
allenes derived from ethylidene and benzylidene erythritols
and threitols proceeds with a moderate asymmetric induction
in the case of the erythritols and with a very low induction
calculations. The results thus obtained were in a full agree-
ment with our predictions based on the stereoselectivity of
cycloaddition. The azetidinones obtained by the cycloaddi-
tion were subjected to intramolecular alkylation at the nitro-
in the case of threitols. This indicates that the erythritol deriv- gen atom to provide the corresponding tricyclic cephams.
atives may exist in solution in one predominant conformation The absolute configurations of the resultant azetidinones and
while the threitol derivatives behave as a conformational en- cephams were assigned using NMR and CD spectroscopy.
semble. The conformations of alkoxyallenes were studied
with variety of NMR techniques as well as using ab initio
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2005)
Introduction
tional character of the 1,2-O-isopropylidene-d-xylofuranose
scaffold, however, the previously presented syntheses are of
limited value.[4]
Several years ago, we demonstrated that the [2ϩ2] cyclo-
addition of chlorosulfonyl isocyanate (CSI) to the chiral al-
koxyallenes derived from 1,2-O-isopropylidene-d-xylofur-
anose in toluene solution provided β-lactams with moderate
stereoselectivity.[1] A comparison of the steady-state NOE
coefficients measured for the 3-O-allenyl-substituted fu-
ranoses in [2H8]toluene solution with conformations gener-
ated by the molecular mechanics program enabled charac-
terisation of the most favourable ground-state geometry of
the ether fragment as the s-cis conformation.[2] The NOE-
based assignment corresponded well to that obtained from
the X-ray structure analysis.[2] Based on the assumption
that the transition state of the cycloaddition resembles the
ground-state conformation of alkoxyallene, we were able to
propose a stereochemical model which plausibly rational-
ised the results of the direction and magnitude of the asym-
metric induction.[1,2]
For the present studies we decided to synthesise alkoxy-
allenes derived from the 1,3-alkylidene l-erythritol[5Ϫ7] and
d-threitol[7Ϫ9] with the intention of examining the general
applicability of the stereochemical model of [2ϩ2] cycload-
dition which we recently proposed.[2] Moreover, in the case
of the readily removable benzylidene protecting group we
could anticipate creating a route to the 3,4-disubstituted-5-
oxacephams suitable for the introduction of a carboxylic
function at C-2 and a variety of substituents at the C-7 car-
bon.[3,4]
Results and Discussion
Synthesis of Alkoxyallenes, Cycloaddition and Formation of
Cephams
The readily available 1,3-O-ethylidene- and 1,3-O-benzyl-
idene-l-erythritols (1[5]and 2,[6,7] respectively), as well as the
1,3-O-ethylidene- and 1,3-benzylidene-d-threitols (11[8]and
12[7,9]) were transformed into the corresponding alkoxy-
allenes (9, 10 and 19, 20) by a standard four-step reaction
sequence involving tritylation of the primary hydroxyl
group,[11] formation of the propargyl ether and the double
methylation-rearrangement (Scheme 1).[12]
The [2ϩ2] cycloaddition of chlorosulfonyl isocyanate to
9, 10, 19 and 20 was carried out in toluene solution in the
presence of pulverised anhydrous sodium carbonate.[13] The
chlorosulfonyl substituent was removed from the resultant
cycloadduct’s nitrogen atoms by the Red-Al reduction.[14]
Intramolecular alkylation of β-lactams derived from the
1,2-O-isopropylidene-d-xylofuranose leads to cephams hav-
ing an exo-propylidene group which enables introduction of
a variety of substituents at the carbon atom next to the β-
lactam carbonyl group.[3,4] Due to the specific multifunc-
[a]
Institute of Organic Chemistry of the Polish Academy of
Sciences,
01-224 Warsaw, Kasprzaka 44/52, Poland
Fax: (internat.) ϩ 48-22-632-37-89
E-mail: chmiel@icho.edu.pl
[b]
National Institute of Public Health,
00-725 Warsaw, Chelmska 30/34, Poland
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2005, 429Ϫ440
DOI: 10.1002/ejoc.200400557
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
429

T. T. Danh, W. Bocian, L. Kozerski, P. Szczukiewicz, J. Frelek, M. Chmielewski
FULL PAPER
Scheme 1
In the case of erythritols 9 and 10, the reactions pro-
ceeded with moderate stereoselectivity to provide the cyclo-
adducts 21, 22 and 27, 28 in the ratio of about 5:1 and 3:1,
respectively. In the case of threitols 19 and 20 the stereo-
selectivity of the formation of β-lactams 37, 38 and 39, 40
was much less satisfactory and amounted to 1:1 and 1.4:1,
respectively. In comparison with the corresponding vinyl
ethers 62[15] and 63,[16] the cycloadditions giving alkoxy-
allenes 9, 10 and 19, 20 proceeded with a significantly lower
stereoselectivity in both the erythritol and threitol series.
Detritylation of the mixture 21/22 followed by tosylation
of the resultant alcohol and subsequent intramolecular al-
kylation at the β-lactam nitrogen atom provided corre-
sponding mixtures of cephams 33/34. The mixtures 21/22,
23/24, 25/26 and 33/34, however, could not be separated
into pure, individual components. In the case of the benzyl-
idene protected series, the same reaction sequence applied
to the cycloadducts 27/28 provided, after detritylation and
tosylation, a mixture of 31/32 which could be easily sepa-
rated by chromatography. Both diastereomers 31 and 32
were independently transformed into the corresponding ce-
The cycloadditions of CSI to 47 and 46 were carried out
phams 35 and 36. In the threitol series the cycloadducts 37/ following our standard procedure[13] to afford correspond-
38 and 39/40 could be easily separated into the pure compo- ing mixtures of the β-lactams 48/49 and 54/55 in ratios 1:1
nents. However, the detritylation of the β-lactams 37, 38, and 1.65:1, respectively. Subsequent desilylation followed by
39 and 40 in the presence of an acid catalyst failed, causing the tosylation and intramolecular alkylation at the nitrogen
decomposition of the β-lactam fragment. Therefore, we de- atom led to the corresponding cephams 58/59 and 60/61
cided to prepare the TBS-protected alkoxyallenes 46 and 47 which were separated into pure components by chromatog-
following the standard silylation procedure.
raphy.
430
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 429Ϫ440

[2ϩ2] Cycloadditions of Chlorosulfonyl Isocyanate to Chiral Alkoxyallenes
FULL PAPER
36[19] and ref.[17]). The shortening of this bond indicates an
interaction of the exo double bond with the β-lactam chro-
mophoric system. Therefore, the shapes of the CD curves
differ significantly from oxacephams without the exo
double bond. The difference consists of the presence of an
additional CE occurring around 250Ϫ260 nm which can
most likely be attributed to the enone unit transition. More-
over, all CD bands are bathochromically shifted (by ap-
proximately 10 nm) in comparison with those oxacephams
without the exo double bond. However, the sign of the band
arising around 230Ϫ240 nm, attributed to the nπ* tran-
sition of the perturbed β-lactam chromophore, follows the
helicity rule.[1,17,19]
The CD data collected in Table 1 demonstrate that the
investigated oxacephams can be divided into two groups
which differ in the sign sequence of particular Cotton ef-
fects (CEs). According to this observation and the helicity
rule, compounds 33, 35, 58 and 60, with a positive sign for
the CD band around 240 nm, belong to the group with the
5a(R) configuration whilst compounds 36, 59, and 61, exhi-
biting a negative sign for this band, correspond to the 5a(S)
configuration (Figure 1). This finding is additionally cor-
roborated by the X-ray data[19] (cf. Experimental, Figure 6)
which unambiguously indicate the 5a(S) absolute configu-
ration for compound 36. The X-ray data also demonstrate
the nonplanarity of the β-lactam chromophoric system
(torsional angles of the units OϭC(7)ϪN(7a)ϪC(4a) and
The configurations of the cephams 33Ϫ36 and 58Ϫ61
were established with the help of H NMR spectroscopy
1
(NOE measurements displayed the spin-spin interaction for
the syn located protons H-4a and H-5a, whereas for those
in an anti arrangement the interactions were not found).
The assignments were further supported by CD-spec-
troscopy.
CD Spectra of Cephams and 4-Alkoxyazetidinones
The CD data of the oxacephams studied are summarised
in Table 1. In general, the absolute configurations of oxace-
phams can be assigned on the basis of the helicity rule, es-
tablished by us recently.[1,17] The rule correlates a positive
CD band around 220Ϫ240 nm with an (R) and a negative
one in the same spectral region with an (S) absolute con-
figuration at the four- and six-membered rings junction.
Our previous investigation demonstrated that the applica-
bility of the rule can be extended to oxacephams with an
exo double bond incorporated into the β-lactam ring.[1,18]
The presence of the exo double bond in the α-position to
the β-lactam carbonyl group constitutes an extension of the
chromophoric system, as evident from the C(6)ϪC(7) bond
Figure 1. CD data of the oxacephams 35 (Ϫ·Ϫ·Ϫ·) and 61
(Ϫ Ϫ Ϫ) as well as UV data for 35 (——–) were measured in aceto-
length (found to be 149.1 pm; see X-ray data of compound nitrile
Table 1. CD data of oxacephams 33, 35, 36, 58, 59, 60 and 61 measured in acetonitrile and the absolute configurations determined on
the basis of their CD data
Compound
Configuration
∆ε (λmax) [molϪ1dm3cmϪ1(nm)]
33
35
36
58
59
60
61
R
R
S
R
S
R
S
Ϫ7.24 (193.5)
Ϫ8.44 (200.0)
ϩ1.45 (218.5)
ϩ5.24 (240.0)
ϩ12.71 (241.0sh)
Ϫ5.92 (231.5)
ϩ3.19 (237.0)
Ϫ0.96 (229.0)
ϩ1.09 (242.5)
Ϫ11.90 (237.0sh)
Ϫ0.25 (288.0)
ϩ14.62 (218.0)
ϩ1.79 (208.5)
Ϫ5.10 (203.5)
ϩ1.36 (257.0)
Ϫ0.11 (280.0)
ϩ0.32 (254.0)
Ϫ0.23 (271.5)
ϩ0.27 (262.5)
Ϫ0.47 (195.0)
Ϫ6.67 (197.5)
Ϫ2.23 (197.5)
Ϫ13.87 (214.5)
Ϫ17.51 (218.0)
Eur. J. Org. Chem. 2005, 429Ϫ440
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
431

T. T. Danh, W. Bocian, L. Kozerski, P. Szczukiewicz, J. Frelek, M. Chmielewski
FULL PAPER
Table 2. CD data of β-lactams 21, 27, 37, 39 and 40 recorded in acetonitrile and the absolute configurations determined on the basis of
their CD data
Compound
Configuration
∆ε (λmax) [molϪ1dm3cmϪ1 (nm)]
21
27
37
39
40
R
R
S
S
R
ϩ5.20(185.0)
Ϫ7.71 (194.5)
Ϫ14.21 (209.5)
Ϫ12.59 (194.5)
Ϫ28.43 (199.5)
Ϫ24.68 (196.5)
Ϫ5.67 (218.0)
ϩ0.58 (256.5)
ϩ0.31 (256.0)
Ϫ1.01 (257.0)
Ϫ1.76 (261.5)
ϩ1.59 (249.5)
Ϫ4.08 (227.0)
ϩ3.64 (223.0)
ϩ3.57 (224.5)
ϩ3.27 (222.5)
ϩ22.72 (186.0)
ϩ15.88 (186.0)
OϭC(7)ϪN(7a)ϪC(8) equal to 174° and 28°, respectively)
thus allowing the application of the helicity rule for the ster-
eochemical assignments.
Elucidation of the Geometry of the Alkoxyallenes
The moderate or low stereoselectivity of the [2ϩ2] cyclo-
additions of CSI to alkoxyallenes prompted us to investi-
gate more deeply the stereochemical model of these reac-
tions. Bearing in mind the Hammond postulate, which
states that for the exothermic reactions the transition-state
resembles the starting materials in energy and geometry, we
decided to find the ground-state conformations of alkoxy-
allenes in order to use them to explain the low stereooselec-
tivity in these reactions.
In an attempt to properly simulate the cycloaddition re-
action conditions, we recorded the nuclear Overhauser en-
hancements η (%) at the reaction temperature in toluene
(Ϫ78 °C) for alkoxyallenes 10 and 20.
However, for both compounds in [D8]toluene and
CD2Cl2, the rotation correlation time of a solute at the
studied frequency (500 MHz for 1H NMR) apparently
meets the condition ω0τc ϭ 1.12 resulting in null Over-
hauser enhancements. Therefore, the NOEs were instead ac-
quired at 303 K (vide infra). In addition, during the D
NMR experiment, it was observed that the threitol deriva-
tive 20 has strong temperature-dependent chemical shifts
whereas the erythritol derivative 10 has much less tempera-
ture-dependent chemical shifts in [D8]toluene. This indi-
cates that the former compound may exist as a confor-
mational ensemble while the latter exists mostly in one pre-
dominant conformation.
The absolute configurations of some representative mon-
ocyclic azetidinones with an exo double bond next to the β-
lactam carbonyl group were established from their circular
dichroism spectra. The relevant CD data are summarised in
Table 2. Our previous study indicates, that in the present
case, neither the helicity rule nor the β-lactam octant rule
are directly applicable for the stereochemical correlation.[1]
The helicity rule cannot be applied here due to the planarity
of the β-lactam chromophore. The presence of the exo
double bond participating in the chromophoric system is
the reason for the lack of applicability of the the β-lactam
octant rule. Based on a combined analysis of the NMR
spectroscopic, X-ray and chiroptical data for such a class of
β-lactam derivatives, we previously developed an empirical
correlation connecting a positive sign for the CE arising
around 220 nm with an (S) absolute configuration and a
negative sign for the same CD band with an (R) absolute
configuration.[1]
Application of the aforementioned regularity to the azet-
idinones 21, 27, 37, 39 and 40 led to the conclusion that
compounds 21 and 27 with a negative sign for the 220 nm
CD band possess an (R) absolute configuration whilst those
with a positive sign for the same CD band (compounds 37
and 39) possess the (S) absolute configuration. Among the
investigated monocyclic azetidinones, the only exception is
compound 40 with its two positive long wavelength CE
To aid the recognition of the stereochemistry of the stud-
bands at around 220 nm and 250 nm. The latter band most ied substrates, molecular modelling of the conformational
probably belongs to the nπ* excitation of the enone and the space was undertaken using DFT calculations. Out of a
former band to the nπ* excitation of the β-lactam chromo- large set of starting structures, shown in Figure 2, the op-
1
1
phore. However, both bands overlap with the Lb and La timised geometries were derived which represent the confor-
benzene transitions from the benzene chromophore adjac- mational families, 10AϪE, of the allenyl moiety, whereas
ent to the stereogenic centre present in the molecule. In the the dynamics of the trityloxymethyl group give rise to the
case of compound 40, and in contrast to compound 39, a three conformations, designated by 10a, 10b and 10c in each
contribution of the 1La benzene transition most likely over- family (Figure 3). These geometries, in turn, enabled back-
rides and surpasses the contribution of the β-lactam chrom- calculation of the theoretical NOEs which were then com-
ophore leading to the cumulatively positive CE around pared with the experimental values. The results are shown
220 nm, i.e. opposite to the predicted one. Despite this ap- in Table 3 for isomer 10. Since the conformational space
parent complexity in the CD spectrum, it seems reasonable of the allenyl moiety is crucial for stereoselection, Table 3
to assume, that in the case of compound 40, the sign of CE contains data illustrating the effects of H-1Ј irradiation on
around 250 nm also correctly correlates the CD data with other protons in a dioxane skeleton which maintains a rigid
the structure. Thus, the absolute configuration of azetidi- conformation. The NOEs observed upon irradiation of
none 40 can be assigned as (R).
other protons, being less informative with respect to studied
432
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org Eur. J. Org. Chem. 2005, 429Ϫ440

[2ϩ2] Cycloadditions of Chlorosulfonyl Isocyanate to Chiral Alkoxyallenes
FULL PAPER
the effect on proton H-6A. Considering the effect on
CH2Ϫ4, however, a small admixture (few %) of family 10Da
can be allowed at room temperature. The stability of the
chemical shifts in the DNMR experiment does not support
the assumption of the dynamic mobility in this case and it
can therefore be concluded that the conformation family
10A, represented by the trityloxymethyl rotamers 10a,c, is
a prevailing one in solution and also at the cycloaddition
reaction temperatures.
As mentioned above, compound 20 shows a different
temperature dependence in [D8]toluene and CD2Cl2. The
temperature dependence of 20 in both solvents is shown in
Table 4. It is apparent, that in [D8]toluene, the axial ring
protons are shifted to lower frequencies at lower tempera-
tures. Also, whereas equatorial H-5E and H-6E do not
change appreciably, H-1Ј moves to higher frequencies. A de-
tailed analysis of the trends is difficult since there are two
moieties in a molecule which may exert diverse shielding
effects upon the ring protons, i.e. the trityl and allenyl func-
tions. In principle, in [D8]toluene, the observed temperature
dependence the of chemical shifts of isomer 20 may also be
attributed to the specific solvation by the aromatic solvent.
The latter rationale, however, cannot be applied to the
methylene-d2 solution. As can be clearly seen from Table 4,
the axial ring protons do not show a temperature depen-
dence in CD2Cl2 but remarkable shifts can be observed for
H-5E and H-6E, i.e. the protons closest to the allenyl moiety.
This may serve as corroborative evidence indicating the
conformational inhomogeneity of compound 20 due to the
conformational freedom of the allenyl moiety (Figure 4,
Figure 5).
Figure 2. Low energy conformations of the allenyl fragment in
compound 10
The recognition of the conformational space deduced
from the DNMR spectroscopic results is further supported
by the NOE experimental results reported in Table 5. It is
apparent from the data shown in Table 5 that the confor-
mational space of the trityloxymethyl moiety does not exert
Figure 3. Low energy conformations of the trityloxymethyl frag-
ment of compound 10
problem, are presented in the supplementary information an effect on the NOEs of a system since these are quite
(Table 1S in the Supporting Information). uniform for the conformations 20aϪ20c (Figure 5). In con-
It is clear from Table 3 that the conformations 10a and trast, the allenyl moiety conformational space is critical for
10c from family 10A can themselves satisfactorily reproduce the NOEs observed in the experiment. It is clear that none
the experimental results whereas the data corresponding to of the conformational families can satisfactorily reproduce
other families (10B, C, D or E) cannot. The conformation the experimental results and that the conformation of the
10B can be rejected on the basis of the unreliable effect on type 20B (Figure 4) can be excluded due to unreliable en-
the H-5A proton and family 10C can be rejected based on hancement of H-6E and H-5E. On the other hand, while
Table 3. NOEs in various conformers of 10 on a given proton upon irradiation of the H-1Ј proton, CD2Cl2 solution, 303 K (sample
(5 mg/0.75 mL solvent) was degassed using the ‘‘freeze-thaw’’ technique and sealed under argon)
EXP
10Aa
10Ab
10Ac
10Ba
10Bb
10Bc
10Ca
10Cc
10Da
10Dc
10Eb
H-2A
H-6E
H-4A
H-6A
H-5A
CH2-4
Me2-9
Ϫ0.1
0.8
0.5
0.4
10.5
2.2
0.0
1.5
0.0
Ϫ0.2
14.6
0.0
4.4
0.0
Ϫ0.8
12.9
0.0
0.3
0.0
0.0
11.3
0.0
0.7
0.3
0.4
1.6
0.0
0.8
0.3
0.4
1.2
0.0
0.6
0.3
0.3
1.6
Ϫ0.2
0.6
1.2
7.2
0.7
Ϫ0.3
Ϫ0.4
2.2
10.1
0.7
0.0
0.3
0.8
0.7
0.8
0.0
0.4
0.8
0.6
0.7
Ϫ0.5
0.0
8.6
0.6
0.3
0.2, Ϫ0.1
0.4
0.8
0.6, Ϫ0.5 Ϫ0.5, 0.1 Ϫ0.1, 0.4 1.0, Ϫ0.2 0.0, 0.0 Ϫ0.1, Ϫ0.7 0.4, Ϫ0.1 Ϫ0.1, Ϫ0.4 2.3, Ϫ0.4 0.2, 1.2
1.1
1.6
1.4
1.2
1.4
0.8
0.8
0.5
2.1
2.5
1.1
1.9
1.9
2.3
1.5
1.0
1.7
1.5
2.5
2.4
2.6
2.4
Eur. J. Org. Chem. 2005, 429Ϫ440
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
433

T. T. Danh, W. Bocian, L. Kozerski, P. Szczukiewicz, J. Frelek, M. Chmielewski
FULL PAPER
Table 4. D NMR study of 20 in [D8]toluene and CD2Cl2 solutions (in italic)[a]
1H NMR chemical shifts, δ [ppm] at various temperatures (K)
Proton
303°
253°
243°
233°
223°
213°
203°
H-1Ј
6.52, 6.52
5.35, 5.59
4.45, 4.42
3.99, 4.20
3.81, 3.88
3.68, 3.37
3.55, 3.33
3.53, 3.95
6.57, 6.55
5.28, 5.60
4.46, 4.40
3.92, 4.24
3.84, 3.93
3.73, 3.30
3.47, 3.21
3.42, 3.96
6.60
5.25
4.46
3.87
3.84
3.75
3.43
3.35
6.62
5.23
4.46
3.83
3.83
3.76
3.40
3.33
6.63, 6.56
5.22, 5.58
4.47, 4.37
3.80, 4.25
3.84 3.95
3.79, 3.25
3.38, 3.10
3.29, 3.96
6.66
5.21
4.47
3.78
3.85
3.78
3.36
3.25
6.71
5.17
4.47
3.76
3.85
3.80
3.32
3.17
H-2A
H-6E
H-4A
H-5E
CH2H-4[b]
CH2L-4[c]
H-6A
[a]
Sample was degassed using the ‘‘freeze-thaw’’ technique and sealed under argon (5 mg/0.75 mL solvent). The chemical shifts were
calibrated with respect to TMS as an internal standard. [b] High frequency proton of the CH2OTr group. [c] Low frequency proton of the
CH2OTr group.
neither conformation 20a nor 20c alone is consistent with
experimental values also, their mixture in unequal pro-
portions can be considered. This is not possible when mix-
ing the family 20B with the other two families (20A or 20C).
It can therefore be concluded that compound 20 exists as a
conformational ensemble consisting of families 20A and
20C, their proportions being highly temperature-dependent
(DNMR results, Table 4). This supports the view that under
the cycloaddition reaction conditions, the compounds may
exist in multiple conformational states which are appreci-
ably populated.
Conclusion
Figure 4. Low energy conformations of the allenyl fragment in
compound 20
The ground state conformation of the alkoxyallenes as-
signed from the experimental NOE coefficients explains
well the direction and magnitude of the asymmetric induc-
tion in the [2ϩ2] cycloaddition reactions. It has been shown
that compound 10 exists in solution under the reaction con-
ditions as one predominant conformer having an s-trans ge-
ometry for the vinyl ether fragment, whereas compound 20
exists as a conformational ensemble. Consequently, for the
former, one can propose a consistent model of the tran-
sition state based on the lowest ground-state conformation
10A and the approach of the isocyanate anti to the bulky
trityloxymethyl group, whereas for the latter (compound 20)
the geometry of the transition state could not be reliably
defined.
Figure 5. Low energy conformations of the trityloxymethyl frag-
ment in compound 20
Table 5. NOEs (%) for various conformers of 20 (Figure 4, Figure 5) on a given proton upon irradiation of the H-1Ј proton, CD2Cl2
solution, 303 K; sample (7 mg/0.75 mL solvent) was degassed using the ‘‘freeze-thaw’’ technique and sealed under argon
EXP
20Aa
20Ab
20Ac
20Ba
20Bb
20Bc
1.7
12.0
0.7
20Ca
20Cb
20Cc
H-2A
H-6E
H-4A
H-6A
H-5E
CH2-4
Me2-9
0.0
3.3
Ϫ0.9
Ϫ0.6
6.5
0.2
1.4
Ϫ1.7
Ϫ1.6
23.3
0.2
3.6
Ϫ1.7
Ϫ2.4
23.5
0.3
7.6
Ϫ1.4
Ϫ3.0
1.7
11.4
0.7
Ϫ2.3
2.0
1.8
11.7
0.8
Ϫ2.4
2.1
0.1
2.0
Ϫ0.1
Ϫ0.6
3.0
0.2
3.2
Ϫ0.1
Ϫ0.9
4.0
0.1
2.8
0.0
Ϫ0.7
2.6
Ϫ2.5
1.8
17.3
2.2
1.7, 1.8
0.0, 0.4
2.0, 0.3
0.5, 0.1
3.0, 2.5
Ϫ0.3, 0.2
2.7, 2.1
Ϫ0.5, 3.1
3.3, 3.2
3.5, Ϫ0.6
3.0, 3.4
0.0, 0.2 Ϫ0.6, 2.3
2.3, 2.9 4.9, 3.0
2.3, 0.4 Ϫ0.1, 0.1
6.2, 6.1 2.4, 3.9
434
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 429Ϫ440

[2ϩ2] Cycloadditions of Chlorosulfonyl Isocyanate to Chiral Alkoxyallenes
FULL PAPER
from 0.6 to 0.9, were taken into account when calculating the NOE
enhancements. The NOEs are accurate to Ϯ 1%.
It was shown that the circular dichroism spectroscopic
measurements of monocyclic azetidinones and cephams
could provide sound stereochemical and spectroscopic as-
signments.
The back calculations of NOEs were accomplished using the BU-
ILDUP program kindly supplied by M. P. Williamson.[21] This pro-
gram, utilising a full relaxation matrix of the spin system under
consideration, allows calculation of steady state NOE enhance-
ments using specific correlation times for individual CϪH vectors
derived from 13C-T1 measurements. The calibration of NOEs can
be accomplished by the ‘‘external relaxation’’ parameter, ρext, ap-
plied to isolated geminal protons serving as a reference within a
molecule.
Experimental Section
General Remarks: Melting points were determined on a Koefler
hot-stage apparatus. NMR spectra were recorded using Bruker Av-
ance 500 and Varian Mercury 400 instruments. IR spectra were
recorded on a PerkinϪElmer FTIR Spectrum 200 spectrometer.
UV spectra were measured on a Cary 100 spectrometer in aceto-
nitrile. CD spectra were recorded between 180 and 400 nm at room
temperature with a JASCO J-715 spectropolarimeter using aceto-
nitrile solutions. Solutions with concentrations in the range 0.8 ϫ
10Ϫ4 to 1.2 ϫ 10Ϫ3 mol·dmϪ3 were examined in cells with a path
length of 0.1 or 1 cm. Mass spectra were recorded using an AMD-
604 instrument from Inectra GmbH and HPLC-MS were recorded
with Mariner and API 356 detectors. Optical rotations were meas-
ured using a JASCO P 3010 polarimeter at 22 Ϯ 3 °C. Column
chromatography was performed with E. Merck Kiesel Gel
(230Ϫ400 mesh).
T1 Measurements: In order to properly set the timing of the steady-
state NOE experiments, 1H-T1 measurements were performed using
a Varian implemented program. The inversion recovery method
was used to measure the relaxation times with a relaxation delay
of 25 s after a π pulse application and incremented recovery time.
The data are shown in Table 6.
The 13C-T1 measurements were also performed for the isomer 10
in order to calculate the correlation time, τc, for various CϪH vec-
tors using Equation (1):
1/T1(13C) ϭ (µ0/4π)2 N γ2Hγ2C h τc/r6
(1)
CH
NOE Measurements: The low temperature spectra were recorded
on a Bruker AVANCE 500 MHz NMR spectrometer. The NOE
experiments at room temperature (303 K) were conducted on a
Varian INOVA 500 MHz NMR spectrometer on samples degassed
using the ‘‘freeze-thaw’’ technique and sealed under argon. The
chemical shifts were calibrated relative to TMS as an internal
standard. Steady-state NOEs[20] were acquired by applying a 10 s
irradiation, a 2 s acquisition and a 1 s delay between each of 64
scans. A 90 degrees pulse width was used with a spectroscopic
window of 5000 Hz and 16 k data points.
where (µ0/4π) is permeability of a vacuum; γ2Hγ2 are the mag-
C
netogyric ratios of the proton and carbon nuclei; h is Planck’s con-
stant; τc is the rotational correlation time of a given CϪH vector;
r is the length of a CϪH vector; N is the number of protons at-
tached to a given carbon. The values are given in Table 7.
Ab Initio Calculations: Starting structures were generated by MM
and pre-optimised using standard force field parameters from the
AMBER 6.0 program.[22]
The spectra were recorded in an absorption mode and phased in
Final geometries for all structures were fully optimised using the
raw using the same parameters. The integrals were read within the B3LYP density functional model and the 6Ϫ31G* basis sets using
same integral limits set for each signal in all irradiated positions,
the same as in a reference irradiation spectrum. In addition, the
integrals were corrected for the integral variation of the residual
CHDCl2 signal. Saturation factors of various multiplets, ranging
the Gaussian 98 suite of programs.[23]
Because of the large molecular sizes of both isomers, we were not
able to apply an advanced calculation protocol to allow for precise
energy calculations. This concerns the application of a large enough
1
Table 6. H-T1 (s) relaxation times of protons in 10 at 303 K in CD2Cl2 solution
T1 (s) of given H[a]
H-1Ј
H[b]
L [c]
H-2A
H-6E
H-5A
H-4A
H-6A
CH2
CH2
5.10 Ϯ 0.09
1.52 Ϯ 0.02
0.68 Ϯ 0.06
1.66 Ϯ 0.02
1.18 Ϯ 0.07
0.58 Ϯ 0.01
0.58 Ϯ 0.01
0.57 Ϯ 0.01
[a]
Sample was degassed using the ‘‘freeze-thaw’’ technique and sealed under argon. The chemical shifts were calibrated relative to TMS
[b]
[c]
as an internal standard. Methyl groups had a T1 of 1.71 s Ϯ0.02 s.
High frequency proton of the CH2OTr group. Low frequency
proton of the CH2OTr group.
Table 7. 13C-T1 (s) relaxation times of carbon atoms in 10 at 303 K in CD2Cl2 solution
Carbon atoms
C-1Ј
C-2
C-4
C-5
C-6
C-8
CH2
CH3
13C-T1 s[a]
0.76 Ϯ 0.01 0.70 Ϯ 0.01 0.58 Ϯ 0.04
0.61 Ϯ 0.01 0.34 Ϯ 0.01 14.00 Ϯ 0.01
91.10 Ϯ 0.01 81.1 Ϯ 0.1
0.33 Ϯ 0.01 2.3 Ϯ 0.01
83.10 Ϯ 0.01 7.5 Ϯ 0.1
τc ϫ 10Ϫ12 s[b] 71.1 Ϯ 0.1
78.3 Ϯ 0.1
96.9 Ϯ 0.1
Ϫ
[a]
[b]
Ca. 40 mg /0.75 mL of a solute dissolved in CDCl3, degassed using the ‘‘freeze-thaw’’ technique and sealed under argon was used.
Proton correlation time for a given carbon atom.
Eur. J. Org. Chem. 2005, 429Ϫ440
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
435

T. T. Danh, W. Bocian, L. Kozerski, P. Szczukiewicz, J. Frelek, M. Chmielewski
FULL PAPER
basis set and consideration of a solvent. Therefore our confidence
in the relative energy values between sets of isomers is limited.[24]
22 (minor): δ ϭ 1.32 (d, J ϭ 5.0 Hz, 3 H, CH3), 1.45, 1.95 [2 s, 6
H, (CH3)2Cϭ], 3.56 (ddd, J ϭ 1.5, 3.4, 9.4 Hz, 1 H, H-3), 3.92 (m,
1 H, H-2), 5.34 (s, 1 H, H-4Ј), 5.90 (br. s, 1 H, NH). Anal. taken
for the mixture, C31H33NO5 (449.61): calcd. C 74.53, H 6.66; found
C 73.8, H 7.1.
Compounds: 1,3-O-benzylidene-4-O-trityl-l-erythritol (4), 1,3-O-
ethylidene-2-O-propargyl-4-O-trityl-l-erythritol (5), 1,3-O-benzyl-
idene-2-O-propargyl-4-O-trityl-l-erythritol (6), 1,3-O-ethylidene-2-
O-(but-2Ј-ynyl)-4-O-trityl-l-erythritol (7), 1,3-O-benzylidene-2-O-
(but-2Ј-ynyl)-4-O-trityl-l-erythritol (8), 1,3-O-benzylidene-l-thre-
itol (12), 1,3-O-benzylidene-4-O-trityl-l-threitol (14), 1,3-O-ethyl-
idene-2-O-propargyl-4-O-trityl-l-threitol (15), 1,3-O-benzylidene-
2-O-propagyl-4-O-trityl-l-threitol (16), 1,3-O-ethylidene-2-O-(but-
2Ј-ynyl)-4-O-trityl-l-threitol (17), 1,3-O-benzylidene-2-O-(but-2Ј-
ynyl)-4-O-trityl-l-threitol (18), 1,3-O-ethylidene-4-O-tert-butyldi-
methysilyl-l-threitol (41), 1,3-O-benzylidene-2-O-(but-2Ј-ynyl)-l-
threitol (42), 1,3-O-ethylidene-2-O-propargyl-4-O-tert-butyldime-
thysilyl-l-threitol (43), 1,3-O-benzylidene-2-O-(but-2Ј-ynyl)-4-O-
tert-butyldimethysilyl-l-threitol (44) and 1,3-O-ethylidene-2-O-
(but-2Ј-ynyl)-4-O-tert-butyldimethysilyl-l-threitol (45) were ob-
tained according to standard procedures. Details of their syntheses,
spectroscopic and analytical data are provided in the Supporting
Information.[25]
(4ЈR)- and (4ЈS)-1,3-O-Ethylidene-2-O-[3Ј-(1ЈЈ-methyl-ethylidene)-
2Ј-oxoazetidin-4Ј-yl]-L-erythritol (23/24): The mixture 21/22 was de-
tritylated with 0.2% p-TsOH in MeOH at room temperature (60
min, TLC monitoring). The crude product was purified by column
chromatography using hexane/ethyl acetate, 40:60 v/v, as an eluent
to give a mixture of compounds 23/24 in a ratio of about 5:1 (78%).
IR (film): ν˜ ϭ 1745, 3306 cmϪ1. HRMS (LSIMS): calcd. for [M ϩ
Hϩ] C12H20NO5: 258.1342; found 258.1353. 1H NMR selected sig-
nals taken for the mixture (500 MHz, CDCl3, ppm): Compound 23
(major): δ ϭ 1.32 (d, J ϭ 5.0 Hz, 3 H, CH3), 1.82, 2.04 (2 s, 6 H,
(CH3)2Cϭ), 4.05(dd, J ϭ 5.3, 10.7 Hz, 1 H, H-1), 4.7 (q, J ϭ
5.0 Hz, 1 H, CH3CH), 5.55 (s, 1 H, H-4Ј), 6.5 (br. s, 1 H, NH);
compound 24 (minor): δ ϭ 1.34 (d, J ϭ 5 Hz, 3 H, CH3), 1.81,
2.05 (2 s, 6 H, (CH3)2Cϭ), 4.12 (dd, J ϭ 5.3, 10.7 Hz, 1 H, H-1),
4.71 (q, J ϭ 5.0 Hz, 1 H, CH3CH), 5.51 (s, 1 H, H-4Ј), 6.87 (br. s,
1 H, NH).
1,3-O-Ethylidene-2-O-(3Ј-methylbuta-1Ј,2Ј-dienyl)-4-O-trityl-L-
(4ЈR)- and (4ЈS)-1,3-O-Ethylidene-2-O-[3Ј-(1ЈЈ-methylethylidene)-2Ј-
oxoazetidin-4Ј-yl]-4-O-tosyl-L-erythritol (25/26): In a ratio of ca.
erythritol (9): To a solution of 7 (0.7 g, 1.58 mmol) in dry THF (7
mL) under argon at Ϫ45 °C was added BuLi (0.7 mL of 2.7 m in
heptane, 1.9 mmol). After 25 min, while the temperature was main-
tained, MeI (0.12 mL, 1.9 mmol) was added. Stirring was con-
tinued for 40 min while warming to room temperature. Sub-
sequently, diethyl ether (50 mL) and brine (50 mL) were added.
The organic layer was separated, washed with water, dried (MgSO4)
and concentrated. The residue was purified on a silica gel column
using hexane/ethyl acetate, 98:2 v/v, as an eluent to give 9 (0.43 g,
5:1, the mixture 25/26 was obtained from 23/24 by a standard tosyl-
ation procedure. The crude product was purified by column chro-
matography using hexane/ethyl acetate, 1:1 v/v, as an eluent to give
25/26 (75%). IR (film): ν˜ ϭ 1741, 3452 cmϪ1. HRMS (LSIMS):
1
calcd. for [M ϩ Hϩ] C19H26NO5S: 412.1430; found: 412.1431. H
NMR selected signals taken for the mixture (500 MHz, CDCl3,
ppm): Compound 25 (major): δ ϭ 1.23 (d, J ϭ 5 Hz, 3 H, CH3),
1.80, 2.02 (2 s, 6 H, (CH3)2Cϭ), 2.46 (s, 3 H, Ts), 3.62 (ddd, J ϭ
2.4, 4.2, 9.5 Hz, 1 H, H-3), 3.8 (m, 1 H, H-2), 4.03 (dd, J ϭ 5.3,
10.8 Hz, 1 H, H-1), 4.6 (q, J ϭ 5.0 Hz, 1 H, CH3CH), 5.55 (s, 1
H, H-4Ј), 6.67 (br. s, 1 H, NH); compound 26 (minor): δ ϭ 1.22
(d, J ϭ 5 Hz, 3 H, CH3), 1.80, 2.06, (2 s, 6 H, (CH3)2Cϭ), 4.62 (q,
J ϭ 5.0 Hz, 1 H, CH3CH), 5.49 (s, J ϭ 5.0 Hz, 1 H, H-4Ј), 6.81
(br. s, 1 H, NH).
60%). [α]2D0 ϭ Ϫ12.4 (c ϭ 0.1, CH2Cl2). IR (film): ν˜ ϭ 1959 cmϪ1
.
HRMS (LSIMS): calcd. for [M ϩ Naϩ] C30H32O4Na: 479.2198;
1
found 479.2210. H NMR (500 MHz, CDCl3, ppm): δ ϭ 1.40 (d,
J ϭ 5.0 Hz, 3H, CH3), 1.66, 1.75 [2d, J ϭ 2.0 Hz, 6H, (CH3)2Cϭ],
3.23 (dd, J ϭ 4.8, 10.3 Hz, 1 H, CH2OTr), 3.32 (dd, J ϭ 2.0,
10.3 Hz, 1 H, CH2OTr), 3.39 (dd, J ϭ 10.0, 11.0 Hz, 1 H, H-1),
3.65 (ddd, J ϭ 2.0, 4.7, 9.4 Hz, 1 H, H-3), 3.94 (m, 1 H, H-2), 4.26
(dd, J ϭ 5.0, 11.0 Hz, 1 H, H-1Ј), 4.7 (q, J ϭ 5.0 Hz, 1 H, CH3CH),
6.24 (sept, J ϭ 2.0 Hz, 1 H, HCϭ). 13C NMR (CDCl3, ppm): δ ϭ
20.9, 23.0, 23.2, 63.4, 66.9, 68.7, 79.6, 86.7, 99.4, 112.9, 117.5,
127.2, 127.6, 128.0, 128.4, 128.9, 129.2, 144.5, 188.5. C30H32O4
(456.58): calcd. C 78.92, H 7.06; found C 78.9, H 7.1.
(2R,4aR,5aR,8aS)- and (2R,4aR,5aS,8aS)-6-Isopropylidene-2-
methyl-1,3,5-trioxa-7a-azacyclobuta[b]decalin-7-one (33/34): Com-
pounds 25/26 (0.1 g, 0.24 mmol) dissolved in anhydrous CH3CN (5
mL) were treated with Bu4NBr (0.1 g, 0.3 mmol) and pulverised
K2CO3 (0.03 g, 0.5 mmol). The mixture was stirred at reflux for 2 h
(TLC). Subsequently, toluene (10 mL) was added, the mixture fil-
tered through Celite and the solvents evaporated. The crude prod-
uct was separated by chromatography on silica gel using hexane/
ethyl acetate, 7:5 v/v, as an eluent to give 33/34 in a ratio of about
4.7:1 (0.05 g, 87%). IR (film): ν˜ ϭ 1773 cmϪ1. HRMS (LSIMS):
calcd. for [M ϩ Hϩ] C12H18NO4: 240.2793; found 240.1231. 1H
NMR selected signals taken for the mixture (500 MHz, CDCl3,
ppm): Compound 33(major): δ ϭ 1.31 (d, J ϭ 5.0 Hz, 3 H,
CH3ϪC2), 1.81, 2.05 [2 s, 6 H, (CH3)2Cϭ], 4.70 (q, J ϭ 5 Hz, 1
H, H-2), 5.55 (s, 1 H, H-5a); compound 34 (minor): δ ϭ 1.33 (d,
J ϭ 5 Hz, 3 H, CH3ϪC2), 2.03, 2.17 [2 s, 6 H, (CH3)2Cϭ], 4.74
(q, J ϭ 5 Hz, 1 H, H-2), 5.40 (s, 1 H, H-5a).
(4ЈR) and (4ЈS) 1,3-O-Ethylidene-2-O-[3Ј-(1ЈЈ-methylethylidene)-2Ј-
oxoazetidin-4Ј-yl]-4-O-trityl-L-erythritol (21/22): To a suspension of
anhydrous Na2CO3 (0.14 g, 1.32 mmol) in dry toluene (2 mL) was
added chlorosulfonyl isocyanate (0.11 mL, 1.32 mmol). The mix-
ture was stirred and upon cooling to Ϫ70 °C a solution of alkoxyal-
lene 9 (0.41 g, 0.9 mmol) in dry toluene (2 mL) was added dropwise.
The mixture was stirred for 30 min, diluted with toluene (6 mL),
treated with Red-Al (1.32 mL, 1 m solution in toluene) and left for
30 min while the temperature was maintained. The cooling bath
was then removed and water (0.2 mL) was added at 0 °C. After 15
min of intensive stirring, the suspension was filtered through Celite,
the solvent evaporated and the residue purified by chromatography
on silica gel to give 21/22 (0.29 g, 66%) in a ratio of ca. 5:1. IR
(film): ν˜ ϭ 1751, 3281 cmϪ1. HRMS (LSIMS): calcd. for [M ϩ
1,3-O-Ethylidene-2-O-(3Ј-methylbuta-1Ј,2Ј-dienyl)-4-O-trityl-
D-
threitol (19): Compound 19 was obtained from 17 according to the
Naϩ] C31H33NO5Na: 522.2256; found 522.2273. 1H NMR selected procedure describe for 9 (48%). [α]2D0 ϭ Ϫ36.4 (c ϭ 0.1, CH2Cl2).
signals taken for the mixture (500 MHz, CDCl3, ppm): Compound
21 (major): δ ϭ 1.42 (d, J ϭ 5.0 Hz, 3 H, CH3), 1.46, 1.96 (2 s, 6
H, (CH3)2Cϭ), 3.58 (ddd, J ϭ 2.0, 4.2, 9.3 Hz, 1 H, H-3), 3.87 (m,
1 H, H-2), 4.06 (dd, J ϭ 5.3, 10.8 Hz, 1 H, H-1), 4.72 (q, J ϭ 5 Hz,
IR (film): ν˜ ϭ 1957 cmϪ1. HRMS (LSIMS): calcd. for [M ϩ Naϩ]
C30H32O4Na: 479.21983; found 479.21834. 1H NMR (500 MHz,
CDCl3, ppm): δ ϭ 1.32 (d, J ϭ 5.0 Hz, 3 H, CH3), 1.73, 1.80 [2 d,
J ϭ 2.0 Hz, 6 H, (CH3)2Cϭ], 3.21, 3.42 (2 m, 2 H, CH2OTr), 3.72
1 H, CH3CH), 5.28 (s, 1 H, H-4Ј), 6.05 (br. s, 1 H, NH). Compound (dd, J ϭ 1.3, 12.5 Hz, 1 H, H-1), 3.84 (m, 1 H, H-2), 3.92 (m, 1 H,
436 © 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.eurjoc.org
Eur. J. Org. Chem. 2005, 429Ϫ440

[2ϩ2] Cycloadditions of Chlorosulfonyl Isocyanate to Chiral Alkoxyallenes
FULL PAPER
H-3), 4.29 (dd, J ϭ 1.2, 12.5 Hz, 1 H, H-1Ј), 4.74 (q, J ϭ 5.0 Hz, (2S,4aR,5aR,8aR)-
and
(2S,4aR,5aS,8aR)-6-Isopropylidene-2-
1 H, CH3CH), 6.54 (sept, J ϭ 2.0 Hz, 1 H, H-1Ј).
methyl-1,3,5-trioxa-7a-azacyclobuta[b]decalin-7-one (58 and 59):
The mixture 58/59 was obtained from 52/53 in a ratio of about 1:1
according to the procedure describe for 33/34 (84%). The mixture
was separated by chromatography using hexane/ethyl acetate, 1:1
v/v, as an eluent. Compound 58: [α]2D0 ϭ 7.6 (c ϭ 0.5, CH2Cl2). IR
(film): ν˜ ϭ 1756 cmϪ1. HRMS (ESI): calcd. for [M ϩ Naϩ]
C12H17NO4Na: 262.1050; found 262.1063. 1H NMR (500 MHz,
CDCl3, ppm): δ ϭ 1.40 (d, J ϭ 5 Hz, 3 H, CH3C-2), 1.87, 2.04 [2
s, 6 H, (CH3)2Cϭ], 3.26 (dd, J ϭ 1.8, 14.3 Hz, 1 H, H-8), 3.50 (m,
1 H, H-4a), 3.84 (dd, J ϭ 1.7, 12.5 Hz, 1 H, H-4), 4.02 (m, 1 H,
H-8a), 4.11 (dd, J ϭ 1.5, 12.5 Hz, 1 H, H-4Ј), 4.15 (dd, J ϭ 7.5,
14.3 Hz, 1 H, H-8Ј), 4.73 (q, J ϭ 5 Hz, 1 H, H-2), 5.73 (s, 1 H, H-
5a). Compound 59:[α]2D0 ϭ Ϫ3.4 (c ϭ 0.2, CH2Cl2). IR (film): ν˜ ϭ
1747 cmϪ1. HRMS (ESI): calcd. for [M ϩ Naϩ] C12H17NO4Na:
262.1050; found 262.1074. 1H NMR (500 MHz, CDCl3, ppm): δ ϭ
1.35 (d, J ϭ 5.0 Hz, 3 H, CH3ϪC2), 1.84, 2.03 [2 s, 6 H, (CH3)2Cϭ
], 3.30 (dd, J ϭ 3.9, 14.7 Hz, 1 H, H-8), 3.50 (m, 1 H, H-4a), 3.62
(m, 1 H, H-8a), 3.90 (dd, J ϭ 1.7, 12.5 Hz, 1 H, H-4), 3.95 (d, J ϭ
14.7 Hz, 1 H, H-8Ј), 4.14 (dd, J ϭ 1.7, 12.5 Hz, 1 H, H-4Ј), 4.74
(q, J ϭ 5.0 Hz, 1 H, H-2), 5.28 (s, 1 H, H-5a).
(4ЈR)- and (4ЈS)-1,3-O-Ethylidene-2-O-[3Ј-(1ЈЈ-methylethylidene)-2Ј-
oxoazetidin-4Ј-yl]-4-O-trityl-D-theritol (37 and 38): The mixture 37/
38 in a ratio of about 1:1 was obtained from compound 19 accord-
ing to the procedure describe for 21/22 (34%). The mixture was
separated into pure components by chromatography using hexane/
ethyl acetate, 7:3 v/v, as an eluent. Compound 37: [α]2D0 ϭ Ϫ18.8
(c ϭ 0.2, CH2Cl2). IR (film): ν˜ ϭ 1746 cmϪ1. HRMS (ESI): calcd.
for [M ϩ Naϩ] C31H33NO5Na: 522.2251; found 522.2277. 1H
NMR (500 MHz, CDCl3, ppm): δ ϭ 1.32 (d, J ϭ 5.0 Hz, 3 H,
CH3), 1.58, 1.96 [2 s, 6 H, (CH3)2Cϭ], 3.20 (dd, J ϭ 6.7; 9.9 Hz, 1
H, CHAHBOTr), 3.51 (dd, J ϭ 6.0, 9.6 Hz, 1 H, CHAHBOTr), 3.59
(m, 1 H, H-2), 3.82(dd, J ϭ 1.4, 12.5 Hz, 1 H, H-1), 3.89 (m, 1 H,
H-4), 4.22 (dd, J ϭ 1.4, 12.5 Hz, 1 H, H-1Ј), 4.77 (q, J ϭ 5.0 Hz,
1 H, CH3CH), 5.62 (s, 1 H, H-4Ј), 6.24 (br. s, 1 H, NH). Compound
38: [α]2D0 ϭ Ϫ9.5 (c ϭ 0.2, CH2Cl2). IR (film): ν˜ ϭ 1742 cmϪ1
.
HRMS (LSIMS): calcd. for [M ϩ Naϩ] C31H33NO5Na: 522.22408;
1
found 522.22564. H NMR (500 MHz, CDCl3, ppm): δ ϭ 1.36 (d,
J ϭ 5.0 Hz, 3 H, CH3), 1.76, 2.02 [2 s, 6 H, (CH3)2Cϭ], 3.18 (dd,
J ϭ 7.4, 9.3 Hz, 1 H, CHAHBOTr), 5.43 (dd, J ϭ 5.6, 9.2 Hz, 1 H,
CHAHBOTr), 3.76 (m, 1 H, H-2), 3.83 (dd, J ϭ 1.5, 12.6 Hz, 1 H,
H-1), 3.95 (m, 1 H, H-4), 3.99 (dd, J ϭ 1.5, 12.6 Hz, 1 H, H-1Ј),
4.78 (q, J ϭ 5.0 Hz, 1 H, CH3CH), 5.56 (s, 1 H, H-4Ј), 5.96 (br. s,
1 H, NH).
1,3-O-Benzylidene-2-O-(3Ј-methylbuta-1Ј2Ј-dienyl)-4-O-trityl-L-
erythritol (10): To a solution of 8 (2.6 g, 5.1 mmol) in dry THF (25
mL) under argon at Ϫ45 °C was added BuLi (2.7 m in heptane, 2.3
mL, 6.2 mmol). After 25 min, while the temperature was main-
tained, MeI (0.39 mL, 6.2 mmol) was added. Stirring was con-
1,3-O-Ethylidene-2-O-(3Ј-methylbuta-1Ј2Ј-dienyl)-4-O-tert-butyl- tinued for 40 min while warming to room temperature. Sub-
dimethysilyl-
D
-threitol (47): Compound 47 was obtained from 45
sequently, diethyl ether (50 mL) and brine (50 mL) were added.
The organic layer was separated, washed with water, dried (MgSO4)
according to the procedure describe for 9 (45%). [α]2D0 ϭ Ϫ12.0 (c ϭ
0.2, CH2Cl2). IR (film z CH2Cl2): ν˜ ϭ 1955 cmϪ1. HRMS (ESI): and concentrated. The residue was purified on a silica gel column,
calcd. for [M ϩ Naϩ] C17H32O4NaSi: 351.1962; found 351.1992. using hexane/ethyl acetate, 97:3 v/v, as an eluent to give the product
1H NMR (200 MHz, CDCl3, ppm): δ ϭ 0.13 0.14 [s, 3 H, (1.9 g, 72%). [α]2D0 ϭ Ϫ16.44 (c ϭ 1.2, CH2Cl2). IR (film): ν˜ ϭ 1958
(CH3)2Si], 1.03 (s, 9 H, tBuSi), 1.42 (d, J ϭ 5.0 Hz, 3 H, CH3), cmϪ1. HRMS (ESI): calcd. for [M ϩ Na]ϩ C35H34O4Na: 541.2349;
1.58, 1.65 (2 d, J ϭ 2.0 Hz, 6 H, (CH3)2Cϭ), 3.34 (dd, J ϭ 1.6, found 541.2352. 1H NMR (200 MHz, CDCl3, ppm): δ ϭ 1.78 1.87
12.4 Hz, 1 H, H-1), 3.62 (m, 1 H, H-2), 3.77 (m, 1 H, H-3), 3.92 [2 d, J ϭ 2.1 Hz, 6 H, (CH3)2Cϭ], 3.43 (dd, J ϭ 4.3, 10.3 Hz, 1 H,
(dd, J ϭ 5.6, 9.6 Hz, 1 H, CHAHBOSi), 4.08 (dd, J ϭ 7.6, 9.6 Hz,
1 H, CHAHBOSi), 4.38 (dd, J ϭ 1.3, 12.4 Hz, 1 H, H-1Ј), 4.56 (q,
J ϭ 5.0 Hz, 1 H, CH3CH).
CHAHBOTr), 3.54 (dd, J ϭ 1.9, 10.3, Hz, 1 H, CHAHBOTr), 3.73
(t, J ϭ 10.6 Hz, 1 H, H-1a), 4.00 (ddd, J ϭ 1.9, 4.3, 9.5 Hz, 1 H,
H-3), 4.30 (m, 1 H, H-2), 4.53 (dd, J ϭ 5.2, 10.6 Hz, 1 H, H-1b),
5.68 (s, 1 H, PhCH), 6.42 (sept, J ϭ 2.1 Hz, 1 H, H-1Ј). 13C NMR
(CDCl3, ppm): δ ϭ 22.6, 22.8, 62.9, 66.5, 68.8, 79.7, 86.3, 100.8,
112.8, 117.2, 126.1, 126.8, 127.7, 128.2, 128.3, 128. 8, 137.8, 144.1,
188.1. C35H34O4 (518.66): calcd. C 81.05, H 6.61; found C 81.0,
H 6.7.
(4ЈR)- and (4ЈS)-1,3-O-Ethylidene-2-O-[3Ј-(1ЈЈ-methylethylidene)-2Ј-
oxoazetidin-4Ј-yl]-4-O-tert-butyldimethysilyl-D-theritol (48/49): The
mixture 48/49 was obtained from compound 47 in a ratio of ca.
1:1, according to the procedure describe for 21/22 (30%). IR (film):
ν˜ ϭ 1748 cmϪ1. HRMS (ESI): calcd. for [M ϩ Naϩ] C18H33NO5N-
(4ЈR)- and (4ЈS)-1,3-O-Benzylidene-2-O-[3ЈЈ-(1ЈЈ-methylethylidene)-
2Ј-oxoazetidin-4Ј-yl]-4-O-trityl-L-erythritol (27/28): To a suspension
1
aSi: 394.2020; found 394.2037. H NMR (500 MHz, CDCl3, ppm,
taken for the mixture of both stereomers, selected signals): δ ϭ
1.91, 2.07 and 1.89, 2.08 [4s, 12 H, 2 ϫ (CH3)2Cϭ],5.74 and 5.68
(2 s, 2 H, H-4Ј of both isomers), 6.73 and 6.55 (2br. s, 2 H, NH of
both isomers).
of anhydrous Na2CO3 (0.28 g) in dry toluene (4 mL) was added
CSI (0.23 mL, 2.6 mmol). The mixture was stirred and upon cool-
ing to Ϫ70 °C a solution of 10 (1.2 g, 2.2 mmol) in dry toluene (4
mL) was added dropwise. The reaction mixture was stirred for 10
min, diluted with toluene (10 mL), treated with Red-Al (1 m solu-
(4ЈR)- and (4ЈS)-1,3-O-Ethylidene-2-O-[3Ј-(1ЈЈ-methylethylidene)-2Ј-
oxoazetidin-4Ј-yl]-4-O-tosyl-
pounds 48/49 (0.15 g, 0.4 mmol) was dissolved in THF (10 mL)
and TBAF·3H2O (0.13 g, 0.4 mmol) was added. The mixture was added at 0 °C. After 15 min of intensive stirring, the suspension
L-threitol (52/53): A mixture of com- tion in toluene, 2.6 mL) and left for 30 min while maintaining the
temperature. The cooling bath was removed and water (0.2 mL) as
stirred for 15 min (TLC) and the solvent was then evaporated and
the crude mixture of 50/51 tosylated by a standard procedure to
give 52/53 in a ratio of ca 1:1 (0.11 g, 67%). IR (film): ν˜ ϭ 1755
cmϪ1. HRMS (ESI): calcd. for [M ϩ Naϩ] C19H25NO7NaS:
434.1244; found 434.1269. 1H NMR (200 MHz, CDCl3, ppm,
was filtered through Celite, the solvent evaporated and the residue
purified by chromatography on silica gel to give 27/28 in a ratio ca.
3:1. IR (film): ν˜ ϭ 1749, 3281 cmϪ1. HRMS (ESI): calcd. for [M
ϩ Na]ϩ C36N35NO5Na: 584.2407; found 584.2410. C36H35NO5
(561.68): calcd. C 76.98; H 6.28; N 2.49. Found: 77.1; H 6.4; N
taken for the mixture of both stereoisomers, selected signals): δ ϭ 2.4. 1H NMR (500 MHz, CDCl3, ppm, taken for the mixture of
1.30 (d, J ϭ 5.1 Hz, 3 H, 2CH3), 1.82, 2.04 [2 s, 6 H, (CH3)2Cϭ of
one isomer], 1.86, 2.04 [2 s, 6 H, (CH3)2Cϭ of second isomer], 5.67,
5.75 (2 s, 2 H, H-4Ј of both isomers), 6.58, 6.53 (2 br. s, 2 H, 2NH
of both isomers).
both stereomers, selected signals) major isomer: δ ϭ 1.52, 2.04 [2
s, 6 H, (CH3)2Cϭ], 5.62 (s, 1 H, H-4Ј), 6.62 (br. s, 1 H, NH); minor
isomer: δ ϭ 1.73, 2.03 [2 s, 6 H, (CH3)2Cϭ], 5.38 (s, 1 H, H-4Ј),
6.20 (br. s, 1 H, NH).
Eur. J. Org. Chem. 2005, 429Ϫ440
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
437

T. T. Danh, W. Bocian, L. Kozerski, P. Szczukiewicz, J. Frelek, M. Chmielewski
FULL PAPER
(2R,4aR,5aS,8aS)-6-Isopropylidene-2-phenyl-1,3,5-trioxa-7a-
azacyclobuta[b]decalin-7-one (36): Compound 36 was obtained
(4ЈR)- and (4ЈS)-1,3-O-Benzylidene-2-O-[3Ј-(1ЈЈ-methylethylidene)-
2Ј-oxoazetidin-4Ј-yl]-L-erythritol (29/30): The mixture 27/28 was de-
from 32 according to the procedure describe for 35 (82%). [α]2D0
ϭ
tritylated with 0.2% p-TsOH in MeOH at room temperature (30
min, TLC monitoring). The crude product was purified by column
chromatography using hexane/ethyl acetate, 2:3 v/v, as an eluent to
give a mixture of compounds 29/30 in a ratio ca. 3:1 (62%). Oil.
IR (film): ν˜ ϭ 1745, 3306 cmϪ1. HRMS (ESI): calcd. for [M ϩ
Na]ϩ C17H21NO5Na: 342.1312; found 342.1334. C17H21NO5:
calcd. C 63.94, H 6.63, N 4.39; found C 64.0, H 6.7, N 4.3. 1H
NMR (400 MHz, CDCl3, ppm, taken for the mixture of both ster-
eoisomers, selected signals) major isomer 29: δ ϭ 1.65, 2.06 [2 s, 6
H, (CH3)2Cϭ], 5.53 (s, 1 H, H-4Ј), 6.78 (br. s, 1 H, NH); minor
isomer 30: δ ϭ 1.83, 2.03 [2 s, 6 H, (CH3)2Cϭ], 5.58 (s, 1 H, H-4Ј),
7.10 (br. s, 1 H, NH).
Ϫ51.6 (c ϭ 0.7, CH2Cl2). IR (film): ν˜ ϭ 1758 cmϪ1. HRMS (ESI):
calcd. for [M ϩ Na]ϩ C17H19NO4Na: 324.1206; found 324.1219.
1H NMR (500 MHz, C6D6, ppm): δ ϭ 1.50 1.96 [s, 3 H, (CH3)2Cϭ
], 2.83 (dd, J ϭ 8.0, 12.6 Hz, 1 H, H-8), 3.25 (m, 1 H, H-8a), 3.33
(m, 1 H, H-4a), 3.50 (t, 10.3 Hz, 1 H, H-4), 4.16 (m, 2 H, H-4ЈH-
8), 4.91 (s, 1 H, H-5a), 5.22 (s, 1 H, H-2). 13C NMR (CDCl3, ppm):
δ ϭ 20.0, 21.1, 42.0, 68.7, 70.3, 73.9, 82.7, 101.8, 126.1, 128.4,
129.3, 133.3, 136.8, 139.6, 165.1. C17H19NO4 (301.35): calcd. C
67.76, H 6.36, N 4.65; found C 67.9, H 6.4, N 4.6.
(4ЈR)- and (4ЈS)-1,3-O-Benzylidene-2-O-[3Ј-(1ЈЈ-methylethylidene)-
2Ј-oxoazetidin-4Ј-yl]-4-O-tosyl-L-erythritol (31 and 32): The mixture
31/32 in a ratio of about 3:1 was obtained from 29/30 by a standard
tosylation procedure. The crude product mixture was separated by
column chromatography using hexane/ethyl acetate, 1:1 v/v, as an
eluent to give 31 (58%) and 32 (20%). Compound 32 (minor):
[α]2D0 ϭ 32.48 (c ϭ 0.8, CH2Cl2). IR (film): ν˜ ϭ 1756, 3275. HRMS
(ESI): calcd. for [M ϩ Na]ϩ C24H27NO7NaS: 469.1400; found
1
496.1414. H NMR (500 MHz, CDCl3, ppm): δ ϭ 1.83, 2.05 [2 s,
6 H, (CH3)2Cϭ], 2.41 (s, 3 H, Ts), 3.66 (t, J ϭ 10.4 Hz, 1 H, H-1),
3.86 (ddd, J ϭ 2.4, 4.1, 9.5 Hz, 1 H, H-3), 3.98 (m, 1 H, H-2), 4.20
(dd, J ϭ 5.3, 10.8 Hz, 1 H, CHAHBOTr), 4.28 (dd, J ϭ 2.4, 10.8 Hz,
1 H, CHAHBOTr), 4.36 (dd, J ϭ 4.2, 10.4 Hz, 1 H, H-1Ј), 5.41 (s,
1 H, PhCH), 5.60 (s, 1 H, H-4Ј), 6.68 (br. s, 1 H, NH). 13C NMR
(CDCl3, ppm): δ ϭ 19.7, 20.6, 21.5, 63.8, 68.0, 69.6, 77.4, 82.5,
100.9, 126.0, 127.9, 128.1, 129.0, 129.7, 132.3, 134.3, 136.8, 141.2,
144.9, 164.6. C24H27NO7S (473.55): calcd. C 60.87, H 5.75, N 2.96,
S 6.77; found C 61.0, H 5.8, N 3.0, S 6.6. Compound 31 (major):
[α]2D0 ϭ Ϫ50.10 (c ϭ 0.2, CH2Cl2). IR (film): ν˜ ϭ 1756, 3281.
HRMS (ESI): calcd. for [M ϩ Na]ϩ C24H27NO7NaS: 496.1400;
found 496.1428. 1H NMR (500 MHz, CDCl3, ppm): δ ϭ 1.75, 1.99
[2 s, 6 H, (CH3)2Cϭ], 2.32 (s, 3 H, Ts), 3.56 (m, 1 H, H-3), 3.75
(m, 2 H, H-2, H-1Ј); 4.18 (m, 1 H, CHAHBOTr), 4.27 (dd, J ϭ 3.5,
11.0 Hz, 1 H, CHAHBOTr), 4.37 (dd, J ϭ 4.4, 10.8 Hz, 1 H, H-1Ј),
Figure 6. ORTEP diagram of compound 36 with the crystallo-
graphic numbering scheme[19]
1,3-O-Benzylidene-2-O-(3Ј-methylbuta-1Ј2Ј-dienyl)-4-O-trityl-D-
threitol (20): Compound 20 was obtained from 18 according to the
procedure describe for 10 (54%). [α]2D0 ϭ Ϫ45.54 (c ϭ 2.0, CH2Cl2).
IR (film): ν˜ ϭ 1955 cmϪ1. HRMS (ESI): calcd. for [M ϩ Na]ϩ
C35H34O4Na: 541.2349; found 541.2378. 1H NMR (500 MHz,
CDCl3, ppm): δ ϭ 1.79 1.86 [d, J ϭ 2.1 Hz, 6 H, (CH3)2Cϭ], 5.62
(s, 1 H, PhCH), 6.60 (hept, J ϭ 2.2 Hz, 1 H, HCϭ). 13C NMR
(CDCl3, ppm): δ ϭ 22. 7, 22.9, 61.8, 67.5, 68.1, 77.8, 86.5, 101.1,
112.3, 117.2, 126.1, 126.9, 127.7, 128.7, 137.9, 143.9, 188.8.
5.34 (s, 1 H, H-4Ј), 5.45 (s, 1 H, PhCH), 6.84 (br. s, 1 H, NH). 13
C
(4ЈR)- and (4ЈS)-1,3-O-Benzylidene-2-O-[3Ј-(1ЈЈ-methylethylidene)-
2Ј-oxoazetidin-4Ј-yl]-4-O-trityl-D-theritol (39/40): The mixture 39/
NMR (CDCl3, ppm): δ ϭ 19. 8, 20.6, 21.6, 65.4, 68.2, 69.7, 77.2,
82.9, 100.9, 126.1, 128.0, 128.1, 129.1, 129. 8, 132.4, 133. 5, 136.8,
141.4, 145.0, 164.4. C24H27NO7S (473.55): calcd. C 60.87, H 5.75,
N 2.96, S 6.77; found C 60.7, H 5.7, N 2.9, S 6.8.
40 in a ratio of about 1.4:1, respectively, was obtained from com-
pound 20 according to the procedure describe for 27/28 (63%). The
mixture was separated into pure components using hexane/ethyl
acetate, 1:1 v/v as an eluent. Compound 39 (major): [α]2D0 ϭ Ϫ54.2
(c ϭ 0.6, CH2Cl2). IR (film): ν˜ ϭ 1749, 3290 cmϪ1. HRMS (ESI):
(2R,4aR,5aR,8aS)-6-Isopropylidene-2-phenyl-1,3,5-trioxa-7a-aza-
cyclobuta[b]decalin-7-one (35): Compound 31 (1.6 g, 3.4 mmol) was
dissolved in anhydrous CH3CN (35 mL) and treated with Bu4NBr
(1.3 g, 4.0 mmol) and pulverised K2CO3 (4.6 g, 33 mmol). The mix-
ture was stirred at reflux for 2 h (TLC). Toluene (50 mL) was sub-
sequently added, the mixture filtered through Celite and the sol-
vents evaporated. The crude product mixture was separated by
chromatography on silica gel using hexane/ethyl acetate, 1:1 v/v, as
an eluent to give 35 (0.9 g, 85%). [α]2D0 240.6 (c ϭ 0.5, CH2Cl2). IR
1
calcd. for [M ϩ Na]ϩ C36H35O5Na: 584.2047; found 584.2437. H
NMR (500 MHz, CDCl3, ppm): δ ϭ 1.46, 1.88 [2 s, 6 H, (CH3)2Cϭ
], 3.15 (dd, J ϭ 5.8, 9.8 Hz, 1 H, CHAHBOTr), 3.51 (dd, J ϭ 6.4,
9.8 Hz, 1 H, CHAHBOTr), 3.56 (m, 1 H, H-2), 3.91 (dd, J ϭ 1.4,
12.7 Hz, 1 H, H-1), 4.04 (m, 1 H, H-3); 4.31 (dd, J ϭ 1.3, 12.7 Hz,
1 H, H-1Ј), 5.52 (s, 1 H, PhCH), 5.58 (s, 1 H, H-4Ј), 6.39 (br. s, 1
H, NH). 13C NMR (CDCl3, ppm): δ ϭ 19. 7, 20.5, 63.5, 68.2,
(film): ν˜ ϭ 1759 cmϪ1. HRMS (ESI): calcd. for [M ϩ Na]ϩ 70.2, 78.7, 82.0, 86.9, 101.2, 126.0, 127.2, 127.9, 128.2, 128.3, 128.6,
C17H19NO4Na: 324.1206; found 324.1227. 1H NMR (500 MHz, 128.9, 133.8, 137.7, 140.5, 143.7, 164.5. Compound 40 (minor): [α]
ϭ Ϫ12.2 (c ϭ 0.6,CH2Cl2). IR (film): ν˜ ϭ 1747, 3294 cmϪ1
.
20
C6D6, ppm): δ ϭ 1.44, 1.87 [2 s, 6 H, (CH3)2Cϭ], 3.29 (dd, J ϭ
D
7.8; 12.7 Hz, 1 H, H-8), 3.41 (t, J ϭ 10.2 Hz, 1 H, H-4), 3.46 (m, HRMS (ESI): calcd. for [M ϩ Na]ϩ C36H35O5Na: 584.2047; found
1
1 H, H-8a), 3.63 (m, 1 H, H-4a), 3.91 (dd, J ϭ 8.3, 12.7 Hz, 1 H, 584.2433. H NMR (500 MHz, CDCl3, ppm): δ ϭ 1.65, 1.91 [2 s,
H-8Ј), 4.05 (dd, J ϭ 5.1, 10.6 Hz; 1 H, H-4Ј), 5.21 (s, 1 H, H-2),
6 H, (CH3)2Cϭ], 3.12 (dd, J ϭ 6.7, 9.5 Hz, 1 H, CHAHBOTr), 3.43
(dd, J ϭ 5.9, 9.5 Hz, 1 H, CHAHBOTr), 3.70 (m, 1 H, H-2), 3.94
5.24 (s, 1 H, H-5a). 13C NMR (CDCl3, ppm): δ ϭ 20.1, 20.7, 42.2,
63.9, 69.1, 75.1, 82.4, 101.8, 126.06, 128.2, 129.1, 134.3, 136.8, (dd, J ϭ 1.4, 12.6 Hz, 1 H, H-1), 4.06 (m, 2 H, H-4, H-1Ј), 5.46 (s,
140.5, 167.2. C17H19NO4 (301.35): calcd. C 67.76, H 6.36, N 4.65;
found C 67.8, H 6.4, N 4.7.
1 H, H-4Ј), 5.52 (s, 1 H, PhCH), 5.94 (br. s, 1 H, NH). 13C NMR
(CDCl3, ppm): δ ϭ 19.8, 20.9, 63.0, 65.2, 70.5, 78.3, 81.7, 86.9,
438
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 429Ϫ440

[2ϩ2] Cycloadditions of Chlorosulfonyl Isocyanate to Chiral Alkoxyallenes
FULL PAPER
101.2, 126.1, 127.2, 128.0, 128.2, 128.3, 128.6, 128.9, 134.3, 137.7,
141.0, 143.6, 164.4.
Acknowledgments
The Authors wish to thank Dr. Zofia Urbanczyk-Lipkowska for
1,3-O-Benzylidene-2-O-(3Ј-methylbuta-1Ј2Ј-dienyl)-4-O-tert-
the crystal structure refinement of compound 36.
butyldimethysilyl-D-threitol (46): Compound 46 was obtained from
44 according to the procedure describe for 10 (60%). [α]2D0 ϭ Ϫ23.4
(c ϭ 0.5, CH2Cl2). IR (film): ν˜ ϭ 1955 cmϪ1. HRMS (ESI): calcd.
for [M ϩ Na]ϩ C22H34O4NaSi: 413.2119; found 413.2139. 1H
NMR (200 MHz, CDCl3, ppm): δ ϭ 0.16 [s, 6 H, (CH3)2Si], 0.94
(s, 9 H, tBu), 1.80, 1.83 [d, J ϭ 2.1 Hz, 3 H, (CH3)2Cϭ], 5.59 (s, 1
H, PhCH), 6.59 (hept, J ϭ 2.1 Hz, 1 H, H-1Ј). 13C NMR (CDCl3,
ppm): δ ϭ Ϫ5.6, Ϫ5.2, 18.4, 22.5, 22.7, 25.9, 67.3, 67.4, 68.1, 78.3,
101.1, 112.5, 116.9, 126.6, 127.9, 128.1, 128.9, 137.3, 188.5.
[1]
´
R. Łysek, B. Furman, Z. Kałuz˙a, J. Frelek, K. Suwinska, Z.
´
Urbanczyk-Lipkowska, M. Chmielewski, Tetrahedron: Asym-
metry 2000, 11, 3131Ϫ3150.
[2]
[3]
´
R. Łysek, P. Krajewski, Z. Urbanczyk-Lipkowska, B. Furman,
Z. Kałuz˙a, L. Kozerski, M. Chmielewski, J. Chem. Soc., Perkin
Trans. 2 2000, 61Ϫ67.
[3a] J. D. Buynak, M. N. Rao, H. Pajouhesh, R. Y. Chandrasek-
aran, K. Finn, P. de Meester, S. C. Chu, J. Org. Chem. 1985,
[3b]
50, 4245Ϫ4252.
J. D. Buynak, H. Pajouhesh, D. L. Lively,
Y. Ramalakshimi, J. Chem. Soc., Chem. Commun. 1984,
948Ϫ949. [3c] J. D. Buynak, M. N. Rao, R. Y. Chandrasekaran,
E. Haley, P. de Meester, S. C. Chu, Tetrahedron Lett. 1985, 26,
5001Ϫ5004. [3d] J. D. Buynak, J. Mathew, R. M. Rao, J. Chem.
Soc., Chem. Commun. 1986, 941Ϫ942. [3e] J. D. Buynak, J. Ma-
thew, M. N. Rao, E. Haley, C. George, U. Siriwardane, J.
Chem. Soc., Chem. Commun. 1987, 735Ϫ737. [3f] J. D. Buynak,
M. N. Rao, J. Org. Chem. 1986, 51, 1571Ϫ1574.
(4ЈR)- and (4ЈS)-1,3-O-Benzylidene-2-O-[3Ј-(1ЈЈ-methylethylidene)-
2Ј-oxoazetidin-4Ј-yl]-4-O-tert-butyldimethysilyl-D-theritol (54/55):
The mixture 54/55 in a ratio of about 1.65:1 was obtained from
compound 46 according to the procedure describe for 21/22 (58%).
HRMS (ESI): calcd. for [M ϩ Na]ϩ C23H35NO5NaSi: 456.2177;
1
found 456.2188. H NMR (500 MHz, CDCl3, ppm, taken for the
mixture of both stereomers, selected signals) major isomer 54: δ ϭ
1.89, 2.06 (2 s, 6 H, (CH3)2Cϭ), 5.59 (s, 1 H, PhCH), 5.74 (s, 1 H,
H-4Ј), 6.72 (br. s, 1 H, NH); minor isomer 55: δ ϭ 1.90, 2.05 [2 s,
6 H, (CH3)2Cϭ], 5.58 (s, 1 H, PhCH), 5.77 (s, 1 H, H-4Ј), 6.53 (br.
s, 1 H, NH).
[4]
´
R. Łysek, Z. Urbanczyk-Lipkowska, M. Chmielewski, Tetra-
hedron 2001, 57, 1301Ϫ1309.
[5] [5a]
R. Barker, D. L. MacDonald, J. Am. Chem. Soc. 1960, 82,
[5b]
2301Ϫ2303.
S. A. Barker, A. B. Foster, A. H. Haines, J.
Lehmann, J. M. Webber, G. Zweifel, J. Chem. Soc. 1963,
4161Ϫ4167.
D. L. MacDonald, H. O. L. Fischer, C. E. Ballou, J. Am. Chem.
Soc. 1956, 78, 3720Ϫ3722.
A. B. Foster, A. H. Haines, J. Homer, J. Lehmann, L. F.
Thomas, J. Chem. Soc. 1961, 5005Ϫ5011.
D. D. Ball, J. K. N. Jones, J. Chem. Soc. 1958: 605Ϫ607.
J. J. Patroni, R. V. Stick, B. W. Skelton, A. H. White, Aust. J.
Chem. 1988, 41, 91Ϫ102.
A. Kazimierski, Z. Kaluza, M. Chmielewski, ARCIVOC 2004,
iii, 213Ϫ225.
(4ЈR)- and (4ЈS)-1,3-O-Benzylidene-2-O-[3Ј-(1ЈЈ-methylethylidene)-
2Ј-oxoazetidin-4Ј-yl]-4-O-tosyl-D-threitol (56/57): A mixture of com-
[6]
[7]
pounds 54/55 (0.15 g, 0.35 mmol) was dissolved in THF (10 mL)
and TBAF·3H2O (0.11 g, 0.35 mmol) was added. The mixture was
stirred for 15 min, the solvent was evaporated and the crude prod-
uct tosylated by a standard procedure to give 56/57 in a ratio of
about 1.4:1 (0.1 g, 60%). HRMS (ESI): calcd. for [M ϩ Na]ϩ
C24H27NO7NaS: 496.1400; found 496.1425. 1H NMR (500 MHz,
CDCl3, ppm, taken for the mixture of both stereomers, selected
signals) major isomer 56: δ ϭ 1.86, 2.07 (2 s, 6 H, (CH3)2Cϭ), 2.46
(s, 3 H, Ts), 5.55 (s, 1 H, PhCH), 5.68 (s, 1 H, H-4Ј), 6.83 (br. s, 1
H, NH); minor isomer 57: δ ϭ 1.90, 2.07 [2 s, 6 H, (CH3)2Cϭ],
2.46 (s, 3 H, Ts), 5.54 (s, 1 H, PhCH), 5.77 (s, 1 H, H-4Ј), 6.76 (br.
s, 1 H, NH).
[8]
[9]
[10]
[11]
M. Carda, P. Casabo, F. Gonzalez, S. Rodriguez, L. R. Dom-
ingo, J. A. Marco, Tetrahedron: Asymmetry 1997, 8, 559Ϫ577.
[12] [12a]
H. J. Reich, M. J. Kelly, R. E. Olson, R. C. Holtan, Tetra-
hedron 1983, 39, 949Ϫ.960. [12b] P. Rochet, J.-M. Vatele, J. Gore,
Synthesis 1999, 795Ϫ799.
Z. Kałuz˙a, W. Abramski, C. Bełz˙ecki, J. Grodner, D. Mostow-
[13]
[14]
´
icz, R. Urbanski, M. Chmielewski, Synlett 1994, 539Ϫ541.
(2S,4aR,5aR,8aR)- and (2S,4aR,5aS,8aR)-6-Isopropylidene-2-phe-
nyl-1,3,5-trioxa-7a-azacyclobuta[b]decalin-7-one (60 and 61): Com-
pounds 60 and 61 were obtained from 56/57 according to the pro-
cedure describe for 33 and 34. The mixture was separated into pure
components by chromatography using hexane/ethyl acetate, 1:1 v/
v as an eluent. Compound 60 (36%, minor): [α]2D0 ϭ Ϫ12.06 (c ϭ
0.2, CH2Cl2). IR (film): ν˜ ϭ 1758 cmϪ1. HRMS (ESI): calcd. for
[M ϩ Na]ϩ C17H19NO4Na: 324.1206; found 324.1224. 1H NMR
(500 MHz, CDCl3, ppm): δ ϭ 1.91, 2.09 [s, 3 H, (CH3)2Cϭ], 3.42
(dd, J ϭ 1.7, 14.2 Hz, 1 H, H-8), 3.67 (m, 1 H, H-4a), 4.10 (dd,
J ϭ 1.8, 12.5 Hz, 1 H, H-4), 4.23 (dd, J ϭ 7.3, 14.2 Hz, 1 H, H-
8Ј), 4.29 (m, 1 H, H-8a), 4.34 (dd, J ϭ 1.6, 12.5 Hz, 1 H, H-4Ј),
5.59 (s, 1 H, H-2), 5.81 (s, 1 H, H-5a). 13C NMR (CDCl3, ppm):
δ ϭ 20.2, 23.0, 29.7, 42.8, 63.8, 70.3, 71.5, 81.3, 101.1, 126.1, 128.3,
129.2, 134.9, 137.4, 140.3, 167.4. Compound 61 (58%, major):
E. Hungerbühler, M. Bioloz, J. Kalwoda, M. Langa, P. Schnei-
der, G. Sedelmeier in New Aspects of Organic Chemistry I. New
York, (Eds.: Z. Yoshida, T. Shiba, Y. Oshiro), VCH; 1989,
419Ϫ451.
[15]
[16]
[17]
[18]
´
K. Borsuk, K. Suwinska, M. Chmielewski, Tetrahedron: Asym-
metry 2001, 12, 979Ϫ981.
K. Borsuk, B. Grzeszczyk, P. Szczukiewicz, B. Przykorska, J.
Frelek, M. Chmielewski, Chirality 2004, 16, 414Ϫ421.
J. Frelek, R. Łysek, K. Borsuk, J. Jagodzinski, B. Furman, A.
Klimek, M. Chmielewski, Enantiomer 2002, 7, 107Ϫ114.
R. Łysek, K. Borsuk, M. Chmielewski, Z. Kałuz˙a, Z.
´
Urbanczyk-Lipkowska, A. Klimek, J. Frelek, J. Org. Chem.
2002, 67, 1472Ϫ1479.
[19]
Crystallographic data for 36 have been deposited with the
Cambridge Crystallographic Data Center. CCDC-247983
contains the supplementary crystallographic data for
this paper. These data can be obtained free of charge
from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data—request/cif.
[α]2D0 ϭ Ϫ147.3 (c ϭ 0.2, CH2Cl2). IR (film): ν˜ ϭ 1758 cmϪ1
.
HRMS (ESI): calcd. for [M ϩ Na]ϩ C17H19NO4Na: 324.1206;
found 324.1220. 1H NMR (500 MHz, CDCl3, ppm): δ ϭ 1.87, 2.07
[2 s, 6 H, (CH3)2Cϭ], 3.42 (dd, J ϭ 3.9, 14.8 Hz, 1 H, H-8), 3.60
(m, 1 H, H-4a), 3.89 (m, 1 H, H-8a), 4.12 (d, 14.8 Hz, 1 H, H-8Ј),
4.17 (dd, J ϭ 1.8, 12.4 Hz, 1 H, H-4), 4.33 (dd, J ϭ 1.6, 12.4 Hz,
1 H, H-4Ј), 5.37 (s, 1 H, H-5a), 5.60 (s, 1 H, H-2). 13C NMR
(CDCl3, ppm): δ ϭ 20.0, 21.4, 31.1, 42.7, 67.6, 69.0, 70.2, 81.2,
101.1, 126.2, 128.2, 128.3, 129.0, 134.5, 137.4, 138.6, 165.4.
[20]
D. Neuhaus, M. P. Williamson, The Nuclear Overhauser Effect
in Structural and Conformational Analysis, VCH Publishers,
New York, 1989, chapter 5, pp 129Ϫ189.
[21] [21a] M. P. Williamson, Magn. Reson. Chem. 1987, 25, 356Ϫ361.
[21b]
L. Kozerski, R. Kawecki, P. Krajewski, P. Gluzinski, K.
Pupek, P. E. Hansen, M. P. Williamson, J. Org. Chem. 1995,
60, 3533Ϫ3538. [21c] L. Kozerski, P. Krajewski, K. Pupek, P. G.
Eur. J. Org. Chem. 2005, 429Ϫ440
www.eurjoc.org
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
439

T. T. Danh, W. Bocian, L. Kozerski, P. Szczukiewicz, J. Frelek, M. Chmielewski
FULL PAPER
Blackwell, M. P. Williamson, J. Chem. Soc., Perkin Trans. 2
1997, 1811Ϫ1818.
D. A. Pearlman, D. A. Case, J. C. Caldwell, G. L. Seibel, U. C.
Singh, P. Weiner, P. A. Kollman, AMBER Program Package;
1991, UCSF; San Francisco, CA.
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M.
A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgom-
ery, Jr., R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Mil-
lam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J.
Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Po-
melli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P.
Y. Ayala, Q. Cui, K. Morokuma, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz, A.
G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I.
Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith,
M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, C. Gonzalez,
M. Challacombe, P. M. W. Gill, B. G. Johnson, W. Chen, M.
W. Wong, J. L. Andres, M. Head-Gordon, E. S. Replogle, J. A.
Pople, Gaussian 98, Gaussian, Inc., Pittsburgh PA, 1998.
J. B. Foresman, A. Frisch, Exploring Chemistry with Electronic
Structure Methods, Gaussian Inc. PA, ed. 1996, chapter 2Ϫ3,
pp. 13Ϫ59.
Supporting Information including preparative procedures,
spectroscopic and analytical data for compounds 3Ϫ8, 12,
14Ϫ18 and 41Ϫ45 as well selected NOEs are available (see also
the footnote on the first page of this article) or from the au-
thors.
[22]
[23]
[24]
[25]
Received August 5, 2004
440
© 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2005, 429Ϫ440