Catalytic asymmetric epoxidation of alkenes with arabinose-derived uloses

Article abstract of DOI:10.1016/S0040-4020(03)00145-5

Four L-erythro-2-uloses were readily prepared from L-arabinose via a reaction sequence involving Fischer glycosidation, acetalization and oxidation. Bulky steric sensors at the anomeric center could enhance the stereoselectivity of the dioxirane epoxidation and one of the uloses performed with good enantioselectivity towards trans-stilbene (up to 90% ee). However, the catalysts decomposed during the epoxidation and the maximum chemical yield was only 13% under the basic conditions. Three L-threo-3-uloses could overcome the decomposition problem based on the electron withdrawing effect of the ester group(s) α to the ketone functionality. The best chemical yield was up to 93% using a ketone with two flanking ester groups. One of the improved uloses displayed moderate enantioselectivity towards trans-disubstituted and trisubstituted alkenes (40-68% ee).

Full text of DOI:10.1016/S0040-4020(03)00145-5

Tetrahedron 59 (2003) 2159–2168
Catalytic asymmetric epoxidation of alkenes
with arabinose-derived uloses
*
Tony K. M. Shing, Yiu C. Leung and Kwan W. Yeung
Department of Chemistry, The Chinese University of Hong Kong, Shatin, Hong Kong, People’s Republic of China
Received 4 October 2002; revised 19 December 2002; accepted 23 January 2003
Abstract—Four L-erythro-2-uloses were readily prepared from L-arabinose via a reaction sequence involving Fischer glycosidation,
acetalization and oxidation. Bulky steric sensors at the anomeric center could enhance the stereoselectivity of the dioxirane epoxidation and
one of the uloses performed with good enantioselectivity towards trans-stilbene (up to 90% ee). However, the catalysts decomposed during
the epoxidation and the maximum chemical yield was only 13% under the basic conditions. Three L-threo-3-uloses could overcome the
decomposition problem based on the electron withdrawing effect of the ester group(s) a to the ketone functionality. The best chemical yield
was up to 93% using a ketone with two flanking ester groups. One of the improved uloses displayed moderate enantioselectivity towards
trans-disubstituted and trisubstituted alkenes (40–68% ee). q 2003 Elsevier Science Ltd. All rights reserved.
1. Introduction
ized cis-alkenes2 with high enantioselectivity. In recent
years, chiral dioxiranes3–5 have become promising reagents
for asymmetric epoxidation. The work reported by Shi and
his co-workers are the most impressive.5 Various types of
alkenes have been epoxidized with good ee using chiral
dioxiranes generated in situ from Oxonew and D-fructose-
derived ketones.5 However, L-fructose is not commercially
available and its scarcity diminishes its attractiveness as the
enantio-complement since a short and facile preparation of
Catalytic asymmetric epoxidation of alkenes is a versatile
synthetic method to introduce chirality into organic
molecules. This protocol is useful in the synthesis of
many natural products and bioactive compounds. Catalytic
asymmetric epoxidation based on transition metal con-
taining complexes is well developed and chiral epoxides
could be obtained from allylic alcohols1 and unfunctional-
Figure 1. Catalysts and alkenes for asymmetric epoxidation.
Keywords: asymmetric epoxidation; ulose; arabinose.
*
Corresponding author. Tel.: þ852-2609-6344; fax: þ852-2603-5057; e-mail: tonyshing@cuhk.edu.hk
0040–4020/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4020(03)00145-5

2160
T. K. M. Shing et al. / Tetrahedron 59 (2003) 2159–2168
Table 1. Study of ulose 1 in catalyzing in situ epoxidation of trans-stilbene
Entry
Conditionsa
pH
mol%
Yield (%)b
ee (%)c
1
2
3
4
A
A
B
B
7
7
10
10
30
300
30
0
0
13
42
–
–
90
87
300
a
Conditions A. Reactions were carried out at 08C (bath temperature) with
substrate (0.1 mmol), ketone, Oxonew (0.5 mmol) and NaHCO3
(1.55 mmol) in CH3CN: 4£1024 M aqueous EDTA 2.5:1, v/v; Condition
B. Reactions were carried out at 08C (bath temperature) with substrate
(0.1 mmol), ketone, Oxonew (0.14 mmol) and K2CO3 (0.58 mmol) in
CH3CN: 0.05 M Na2B4O7·10H2O in 4£1024 M aqueous EDTA 2.5:1,
v/v.
Isolated yield.
Enantioselectivity was determined by 1H NMR analysis of the epoxide
products directly with Eu(hfc)3 shift reagent.
Figure 2. Design of catalyst based on steric effect and spiro transition state.
L-fructose is still a problem remaining to be solved.5d Other
uloses derived from D-arabinose, D-mannose, D-glucose
and L-sorbose were also reported by Shi,5i but they afforded
poor chemical yields (0–57% yield) on epoxidation. The
controlling factors and structural requirements for high
enantioselectivity are not well understood5p and there is
still ample room for the discovery of new ulose catalysts
for asymmetric epoxidation. Our long-term interest in the
application of carbohydrates in asymmetric synthesis has
prompted us to search for efficient ulose catalysts to induce
chirality with high ee in asymmetric epoxidation. The
abundance of arabinose in the chiral pool and its
commercial availability in large quantities for both
enantiomers have made it the first choice for our studies.
Over the years, we had described the use of arabinose-
derived alcohols as chiral auxiliaries in asymmetric Diels–
Alder reaction6,7 and in asymmetric Hosomi–Sakurai
reaction.8 In this paper, we report our results on asymmetric
epoxidation of unfunctionalized alkenes, using Oxonew as
oxidant, catalyzed by L-erythro-2-uloses 1–4 and L-threo-
3-uloses 5–7 (Fig. 1). The design of the catalyst is based on
the established spiro transition states4,5a,d,g for the dioxirane
epoxidation. The favoured transition state should result
from the minimum steric intereaction between the dioxirane
substituents and the substrate. Firstly, for uloses 1–4, the
approach of the alkenes substrate is controlled by the
isopropylidene fused ring and therefore should be on the less
hindered a-face. Following the same reasoning, the attack
b
c
of the alkenes on the dioxiranes derived from uloses 5–7
should be on the less hindered b-face. Secondly, the
favoured spatial orientation of the alkene should impose
minimum steric interaction with the steric sensor, i.e. the
aglycone in uloses 1–4 and the b-benzoate in uloses 5–7.
For an example, Figure 2 shows that the favoured transition
state for the epoxidation should be (A) according to the
reasoning presented above.
2. Results and discussion
Uloses 1–4 were readily prepared from L-arabinose via a
standard reaction sequence involving Fischer glycosidation,
acetalization, and oxidation (Scheme 1).9 The glycosides
8–10, prepared from their corresponding alcohols, were
converted into their respective acetonides 11, 13 and 14 on
treatment with 2,2-dimethoxypropane (DMP) and acetone.
Acetal 12 was prepared from the acetalization reaction with
benzophenone using a Dean and Stark apparatus and the
chemical yield was poor (up to 40% yield). The poor yield
may be caused by the steric repulsion between the benzyl
aglycone in glycoside 8 and benzophenone or by the lower
Scheme 1. Syntheses of uloses 1–4. Key: (a) ROH, AcCl (cat.), rt; (b) For 11, 13 and 14: Hþ, DMP, acetone, rt; 12: Hþ, benzophenone, benzene, reflux;
˚
(c) PDC, AcOH (cat.), 4 A MS, CH2Cl2, rt.

T. K. M. Shing et al. / Tetrahedron 59 (2003) 2159–2168
2161
Table 3. Study of uloses 1–4 in catalyzing in situ epoxidation of trans-
stilbene
Entrya
Catalyst
Yield (%)b
ee (%)c
Configd
1
2
3
4
1
2
3
4
13
10
10
8
89
6
82
90
(2)-(S,S)
(2)-(S,S)
(2)-(S,S)
(2)-(S,S)
a
All reactions were carried out at 08C (bath temperature) with substrate
(0.1 mmol), ketone (0.03 mmol), Oxonew (0.14 mmol) and K2CO3
(0.58 mmol) in CH3CN: 0.05 M Na2B4O7·10H2O in 4£1024 M aqueous
EDTA 2.5:1, v/v.
Isolated yield.
Enantioselectivity was determined by 1H NMR analysis of the epoxide
products directly with Eu(hfc)3 shift reagent.
b
c
d
The absolute configurations were determined by comparing the measured
optical rotations with the reported ones.12
the ee of the stilbene oxide is up to 90% (Table 1, entry 3).
The structure of 1 gives good stereochemical communi-
cation between the catalyst and the substrate. The
preponderant formation of the (S,S)-enantiomer may be
rationalized by the favoured transition state shown in
Scheme 2. Firstly, the approach of the trans-stilbene should
be on the a-face because the b-face is shielded by the
acetonide fused ring. Secondly, the steric repulsion between
the phenyl group and the benzyl aglycone in transition state
(D) is absent in (C).
Scheme 2. Spiro transition states for the expoxidation catalyzed by ulose 1.
reactivity of the carbonyl group in benzophenone. Oxidation
of the free alcohol in acetals 11–14 with PDC gave uloses
1–4 in good yields. Uloses 1–4 had steric sensors of
different sizes that may induce different degree of enantio-
selectivity in the epoxidation of alkenes.
Experimental results (Table 1) show that pH is an important
factor for the catalytic dioxirane epoxidation.5d,f Studies
were carried out to investigate the optimum pH conditions
for the epoxidation using ulose 1 as catalyst and the results
are summarized in Table 2. The reaction was carried out at
08C with 1 mmol of substrate and 30 mol% of ulose 1. The
epoxidation reactions essentially stopped after 2 h at each
pH value studied. The pH was monitored by a pH meter and
adjusted with a buffer solution. Epoxidation only occurred
between pH 9 and 10 (Table 2, entries 3 and 4). The
chemical yield of the epoxidation was still poor (13%). It is
noteworthy that no catalyst could be recovered, hinting at
decomposition of the catalyst during reaction.
With the catalysts in hand, we initiated our investigation
using trans-stilbene as a test substrate and ulose 1 as the
catalyst under different reaction conditions. The results are
summarized in Table 1. In Table 1 (entries 1 and 2), no
stilbene oxide was observed under neutral reaction con-
ditions (conditions A). The starting material, stilbene, was
recovered but the ulose catalyst 1 could not be isolated after
column chromatography. Under the basic conditions, the
chemical yield of stilbene oxide was 42% with 3 mole
equiv. of ulose 1 (Table 1, entry 4). The yield was decreased
to 13% with 30 mol% of 1 used (Table 1, entry 3) and the
catalyst could not be recovered after work-up. This
indicated that ulose 1 decomposed during the epoxidation
but we could not isolate any decomposition products.
Although ulose 1 did not afford good chemical yields, it did
induce good enantioselectivity towards trans-stilbene and
According to our design of the catalyst shown in Figure 2,
the change in the structure of the steric sensor, i.e. the
aglycone moiety, may improve the enantioselectivity of the
catalytic epoxidation. For this reason, studies were carried
out to investigate the epoxidation catalyzed by uloses 2–4.
The reactions were carried out at 08C with 0.1 mmol of
substrate and 30 mol% of catalyst. The conditions were
maintained at pH 9–10 in aqueous CH3CN and the results
are summarized in Table 3. The uloses 2–4 with steric
sensors different from that in 1 did not improve the chemical
yield of the catalytic epoxidation. The highest chemical
yield was still only 13% (Table 3, entry 1), but the
enantioselectivity was sensitive to the size of the steric
sensor. Uloses 1 and 4 have steric sensors of similar size at
the anomeric center (a benzyl group and a (40-methyl)benzyl
group). They both induced 90% ee towards trans-stilbene
(Table 3, entries 1 and 4). When the benzyl sensor was
replaced by a less bulky methyl group as in ulose 3, the ee
decreased to 82% (Table 3, entry 3). On the other hand, the
size of the acetal group also can affect the enantioselectivity.
Table 2. pH study of ulose 1 in catalyzing in situ epoxidation of trans-
stilbene
Entrya
pH
Yield (%)b
ee (%)c
Config.d
1
2
3
4
5
6
7
8
9
10
11
12
No reaction
No reaction
12
13
No reaction
No reaction
–
–
90
89
–
–
–
(2)-(S,S)
(2)-(S,S)
–
–
–
a
All reactions were carried out at 08C (bath temperature) with substrate
(1 mmol), ketone (0.3 mmol), Oxonew (2.5 mmol) in CH3CN:buffer
(2.5:1, v/v).
Isolated yield.
Enantioselectivity was determined by 1H NMR analysis of the epoxide
products directly with Eu(hfc)3 shift reagent.
b
c
d
The absolute configurations were determined by comparing the measured
optical rotations with the reported ones.12

2162
T. K. M. Shing et al. / Tetrahedron 59 (2003) 2159–2168
demanding a-pivalate uloses were synthesized as new
catalysts. The ester group was chosen as the steric sensor
because it could overcome the problem of decomposition.
The electron withdrawing ester groups displayed poor
migratory aptitude in Baeyer–Villiger reaction.11 As a
result, the decomposition problem of the ulose during
epoxidation would be solved. L-erythro-3-uloses 5 and 6
were readily prepared from glycoside 8 as shown in Scheme
3 and dibenzoate ulose 7 was prepared from acetonide 11 as
shown in Scheme 4. Uloses 5–7 were used as catalysts for
the asymmetric epoxidation and the results are summarized
in Table 4.
Figure 3. By-products from epoxidation catalyzed by ulose 4.
The dramatic decrease of the ee (6%) for ulose catalyst 2
was the evidence (Table 3, entry 2).
Compounds 15 and 16 (Fig. 3) were isolated from the
epoxidation reaction of ulose 4 (Table 3, entry 4). It was
evident that the poor chemical yield was caused by the
decomposition of the catalyst but there was insufficient
evidence that the Baeyer–Villiger reaction10 was the major
decomposition pathway for the L-arabino-2-ulose. Most
research groups propose that Baeyer–Villiger reaction was
the main decomposition pathway, but only in two cases
were the lactone by-products successfully isolated.3f,5j They
also gave evidence that the change of migratory aptitude of
the a-carbon might inhibit the decomposition of the ketone
catalyst.
Ulose catalysts 5–7 afforded good chemical yields (76–
93%) of the epoxides in catalytic asymmetric epoxidation
(Table 4, entries 1–12). The C-4 ester steric sensor
overcame the problem of decomposition and the catalyst
could be recovered after epoxidation. The ees were poor for
ulose 7 (up to 54% ee), especially for substrate b (16% ee,
Table 4, entry 6). The keto group in ulose 7 is more
electrophilic than that in 5 and in 6. It should induce higher
reactivity to form the dioxirane and hence better chemical
yields resulted. The enantioselectivity of ulose 6 (entries 2,
5, 8, 10 and 12) was superior to that of ulose 5 (entries 1, 4,
7, 9 and 11). Changing the size of the steric sensor from
the benzoate to the more bulky pivalate group enhanced
the ee of the epoxides. The preponderant formation of the
(S,S)-enantiomer may be rationalized by the favoured
The L-erythro-2-uloses 1–4 gave poor chemical yields for
the catalytic dioxirane epoxidation. The reason was the
decomposition of the catalyst during epoxidation. The
stability of new ulose catalysts to overcome the decompo-
sition was then considered. a-Benzoate and the more steric
Scheme 3. (a) Syntheses of uloses þ5 and 6. Key: (a) tetramethoxybutane, Hþ, MeOH, reflux; (b) RCl, pyridine, DMAP(cat.); (c) 90% TFA, CH2Cl2; (d) i) H2,
˚
Pd/C, EtOH–EtOAc; (ii) DMP, H , acetone; (e) PDC, AcOH (cat.), 4 A MS, CH2Cl2, rt.
Scheme 4. Synthesis of ulose 7. Key: (a) BzCl, pyridine, DMAP(cat.); (b) 90% TFA, CH2Cl2; (c) PhCHO, Hþ, benzene, Dean–Stark trap; (d) NBS,
˚
AIBN(cat.), benzene–H2O; (e) PDC, AcOH (cat.), 4 A MS, CH2Cl2, rt.

T. K. M. Shing et al. / Tetrahedron 59 (2003) 2159–2168
2163
L-arabino-ulose catalysts had to be explored in order to
enhance the chemical yield and ee of the asymmetric
dioxirane epoxidation. Research in this direction is in
progress.
Table 4. Study of uloses 5–7 in catalyzing in situ epoxidation of alkenes
Entrya
Catalyst
Substrate
Yield (%)b
ee (%)c
Config.d
1
2
3
4
5
6
7
8
5
6
7
5
6
7
5
6
5
6
5
6
a
a
a
b
b
b
c
78
80
87
77
82
93
82
84
76
77
93
89
54
67
54
27
43
16
24
40
67
68
48
59
(2)-(S,S)12
(2)-(S,S)12
(2)-(S,S)12
(2)-(S,S)13
(2)-(S,S)13
(2)-(S,S)13
(2)-(S,S)14
(2)-(S,S)14
(þ)-(S)13
(þ)-(S)13
(2)
3. Experimental
Melting points were measured in degrees Celsius and
uncorrected. IR spectra were recorded on a FT-IR
spectrophotometer as neat films on KBr plates. Optical
c
9
d
d
e
10
11
12
1
rotations were obtained at 589 nm. H NMR spectra were
e
(2)
measured at either 250 MHz or 500 MHz. 13C NMR spectra
were obtained at 62.9 MHz. 2D NMR spectra were
measured at 500 MHz. Elemental analyses were performed
at either the Shanghai Institute of Organic Chemistry, The
Chinese Academy of Sciences, China, or MEDAC Ltd,
Department of Chemistry, Brunel University, UK. Mass
spectra were obtained using EI or FAB technique. All
reactions were monitored by analytical TLC being
performed on Merck aluminum precoated plates of silica
gel 60 F254 with detection by spraying with 5% w/v
dodecamolybdophosphoric acid in ethanol and subsequent
heating. All columns were packed wet using E. Merck silica
gel 60 (230–400 mesh) as the stationary phase and eluted
using the flash chromatographic technique. THF was
distilled from sodium benzophenone ketyl under a nitrogen
atmosphere.
a
All reactions were carried out at room temperature with substrate
(0.1 mmol), ketone (0.01 mmol), Oxonew (1 mmol) and NaHCO3
(3.1 mmol) in CH3CN:4£1024 M aqueous EDTA (5:1, v/v) for 2.5 h.
Isolated yield.
b
c
Enantioselectivity was determined by 1H NMR analysis of the epoxide
products directly with Eu(hfc)3 shift reagent.
d
The absolute configurations were determined by comparing the measured
optical rotations with the reported ones.
transition state shown in Scheme 5. Firstly, the approach of
the alkene should be from the top face (b-face) because the
a-face is hindered by the acetonide fused ring. Secondly, the
steric repulsion between the phenyl group and the C-4 ester
group in transition state (F) is absent in (E). However, the
stereochemical communication between the catalyst and
substrate was still moderate (up to 68% ee for substrate d).
In summary, L-erythro-2-uloses 1–4 were easily prepared
from L-arabinose. The anomeric aglycone steric sensors of
suitable size (e.g. uloses 1 and 4) exhibited good stereo-
chemical communication towards trans-stilbene (up to 90%
ee). The chemical yields of the epoxides catalyzed by uloses
1–4 were poor (up to 13% yield) because of the
decomposition of the ulose catalysts during the reaction.
L-threo-3-Uloses 5–7 were also readily prepared as
catalysts and they could overcome decomposition based
on the electron withdrawing effect of the ester group(s) at
the a-position. The chemical yields were good (up to 93%
for ulose 7), but moderate results were obtained for
enantioselectivity (up to 68% ee for ulose 6). Thus, new
3.1. General in situ epoxidation procedure under neutral
conditions
To a stirred solution of alkene (0.1 mmol), ketone
(0.01 mmol) and n-Bu4NHSO4 (0.5 mg) in CH3CN (5 mL)
containing a buffer (1 mL, 4£1024 M aqueous EDTA) was
added dropwise, concomitantly, a solution of Oxonew
(1 mmol) in aqueous EDTA (3 mL, 4£1024 M) and a
solution of NaHCO3 (3.1 mmol) in H2O (3 mL) via two
dropping funnels. The pH of the mixture was maintained at
about 7–7.5 over a period of 2.5 h. The reaction mixture
was then poured into water (10 mL), extracted with CHCl3
(3£10 mL). The combined extracts were dried with
Scheme 5. Spiro transition states for the epoxidation catalyzed by ulose 5 or 6.

2164
T. K. M. Shing et al. / Tetrahedron 59 (2003) 2159–2168
anhydrous MgSO4, and filtered. The filtrate was concen-
trated under reduced pressure to give a residue that was
fractionated by flash column chromatography to give the
epoxide.
(2 mL, 29.4 mmol) was added dropwise to methanol
(40 mL, 0.56 mol) at 08C. After the addition, L-arabinose
(5 g, 33.3 mmol) was added slowly in portions and the
solution was maintained at 0–58C for 1 h. Then, the
resulting mixture was heated under reflux for 7 h and
cooled at 08C overnight. The crude white solid was filtered
and was recrystallized from EtOH to give the glycoside 9
(4.2 g, 77%) as white crystals: mp 170–1718C; [a]2D2¼
þ240.7 (c 0.7, DMF) {lit.,9 mp 169–1708C; [a]D20¼þ244 (c
1.7, H2O)}; 13C NMR (CD3OD) d 102.0, 70.8, 70.7, 70.3,
63.9, 55.8.
3.2. General in situ epoxidation procedure under basic
conditions
A stirred solution of alkene (0.1 mmol), ketone (0.03 mmol)
and n-Bu4NHSO4 (0.5 mg) in CH3CN (5 mL) containing
a buffer (2 mL, 0.05 M Na2B4O7·10H2O in 4£1024
M
aqueous EDTA) was cooled to 08C. A solution of Oxonew
(0.14 mmol in 1 mL 4£1024 M aqueous EDTA) and a
solution of K2CO3 (0.58 mmol in 1 mL H2O) were added
dropwise over a period of 1.5 h. The reaction mixture was
then poured into water (10 mL), extracted with CHCl3
(3£10 mL). The combined extracts were dried with
anhydrous MgSO4, and filtered. After removal of solvent
from the filtrate under reduced pressure, the residue was
fractionated by flash column chromatography to give the
epoxide.
3.3.3. 40-Methylbenzyl b-L-arabinopyranoside (10).
Acetyl chloride (2 mL, 29.4 mmol) was added dropwise to
40-methylbenzyl alcohol (30 g, 245.9 mmol) at 08C. After
the addition, L-arabinose (5 g, 33.3 mmol) was added
slowly and the solution was maintained at 0–58C for 1 h.
The resulting mixture was vigorously stirred at 658C for 5
days. Et2O (100 mL) was then added to the cooled solution.
The precipitate was filtered and was washed by Et2O
(20 mL). Recrystallization from EtOH gave the glycoside
10 (7.0 g, 83%) as white crystals: mp 178–1808C; Rf 0.74
(MeOH–CHCl3, 1:3); [a]2D0¼þ223 (c 1.0, DMF) {lit.,7 mp
178–1808C; [a]2D0¼þ221.2 (c 1.2, DMF)}; IR (MeOH)
3.3. Experimental procedure for epoxidation at different
pH
1
3210 (OH) cm21; H NMR (CD3OD) d 7.20 (4H, m), 4.89
A stirred solution of alkene (1 mmol), ketone (0.3 mmol)
and n-Bu4NHSO4 (1 mg) in CH3CN (25 mL) containing a
buffer [(i) pH 7, 10 mL 4£1024 M aqueous EDTA adjusted
(1H, d, J¼2.5 Hz), 4.68 (1H, d, J¼11.7 Hz), 4.49 (1H, d,
J¼11.7 Hz), 3.89–3.57 (5H, m), 2.33 (3H, s); 13C NMR
(d6-DMSO) d 135.0, 128.6, 127.5, 98.7, 69.1, 68.7, 68.3,
68.2, 63.2, 20.6; MS (EI) m/z 149 (Mþ2105, 6.5), 106
(100), 105 (82).
with 1.0 M NaHCO3; (ii) pH 7.5–8.5, 10 mL 4£1024
M
aqueous EDTA adjusted with 1.0 M KOH; (iii) pH
M
8.5–10.5, 10 mL 0.05 M Na2B4O7·10H2O in 4£1024
aqueous EDTA adjusted with 1.0 M KH2PO4; (iv) pH
11–12, 10 mL 0.05 M K2HPO4/0.1 M NaOH, (2:1, v/v)]
was cooled to 08C. Oxonew (0.14 mmol in 10 mL
4£1024 M aqueous EDTA) and a solution of K2CO3
(0.58 mmol in 10 mL H2O) were added dropwise over a
period of 1.5 h. The reaction mixture was then poured into
water (10 mL), extracted with CHCl3 (3£20 mL). The
combined extracts were dried with anhydrous MgSO4, and
filtered. After removal of solvent from the filtrate under
reduced pressure, the residue was fractionated by flash
column chromatography to give the epoxide.
3.3.4. Benzyl 3,4-O-isopropylidene-b-L-arabinopyrano-
side (11). To a solution of glycoside 8 (2.8 g, 11.6 mmol) in
acetone (30 mL) were added 2,2-dimethoxypropane
(1.5 mL, 12.2 mmol) and p-TsOH (5 mg). The resulting
suspension was stirred at rt for 12 h and all the suspension
had dissolved. The reaction was quenched with NEt3 and
then concentrated in vacuo to give a crude yellow syrupy
residue. Flash column chromatography (EtOAc–hexane,
1:1) of the residue gave a white solid which was
recrystallized from EtOAc–hexane to give the acetonide
11 (2.9 g, 89%) as white crystals: mp 56–578C; Rf 0.50
(EtOAc–hexane, 1:1); [a]2D0¼þ217.8 (c 1.2, CHCl3) {lit.,9
mp 54–568C; [a]2D0¼þ220 (c 1, CHCl3)}; IR (CHCl3) 3440
3.3.1. Benzyl b-L-arabinopyranoside (8). Acetyl chloride
(7.5 mL, 109.5 mmol) was added dropwise to BnOH
(130 mL, 1.3 mol) which was cooled in an ice-water bath
with stirring. L-Arabinose (25 g, 166.5 mmol) was added
and the reaction temperature was raised to rt. The resulting
mixture was vigorously stirred at rt for 5 d. Et2O (500 mL)
was then added to precipitate the benzyl glycoside that was
filtered and the crude white solid washed with Et2O
(100 mL). Recrystallization from EtOH gave the glycoside
8 (35.0 g, 88%) as white crystals: mp 177–1788C; Rf 0.66
(MeOH–CHCl3, 5:1); [a]2D0¼þ210.5 (c 1.2, DMF) {lit.,9
mp 172–1738C; [a]2D0¼þ209 (c 0.4, H2O)}; IR (MeOH)
3280 (OH) cm21; 1H NMR (CD3OD) d 7.61–7.48 (5H, m),
5.07 (1H, d, J¼2.3 Hz), 4.92–4.72 (2H, d, J¼12.0 Hz),
4.08–3.98 (4H, m), 3.78 (1H, dd, J¼12.5, 2.5 Hz); 13C
NMR (d6-DMSO) d 138.1, 128.0, 127.3, 127.2, 98.9, 69.1,
68.7, 68.4, 68.2, 63.2; MS (EI) m/z 163 (Mþ277, 13.9), 131
(15.1), 92 (100).
(OH) cm21 13C NMR d 137.1, 128.5, 127.9, 109.2, 96.9,
;
75.9, 72.9, 69.9, 69.8, 60.1, 27.8, 25.9; MS (EI) m/z 265
(Mþ215, 5.2), 189 (10.6), 91 (100).
3.3.5. Benzyl 3,4-O-diphenylmethylidene-b-L-arabino-
pyranoside (12). To a solution of 8 (1.0 g, 4.2 mmol) in
benzene (50 mL) was added benzophenone (2.2 g,
12.3 mmol) and p-TsOH (5 mg). The resulting solution
was heated under reflux with a Dean–Stark trap for 5 d. The
solvent was then evaporated under reduced pressure, and the
residue was flash column chromatographed (hexane–Et2O,
3:1) to give a colorless syrup, 12 (0.2 g, 12%): Rf 0.58
(hexane–EtOAc, 1:1); [a]2D0¼þ78.8 (c 0.7, CHCl3); IR
1
(EtOAc) 3442 (OH) cm21; H NMR (CDCl3) d 7.53–7.27
(15H, m), 4.93 (1H, d, J¼3.0 Hz), 4.79 (1H, d, J¼11.7 Hz),
4.55 (1H, d, J¼11.7 Hz), 4.35 (1H, t, J¼6.3 Hz), 4.18–4.15
(1H, m,), 2.35 (1H, brs); 13C NMR (CDCl3) d 143.5, 137.6,
129.2, 128.8, 128.8, 128.7, 127.7, 126.8, 126.7, 97.3, 77.0,
74.0, 70.4, 70.0, 66.0, 60.4; MS (FAB) m/z 405 (MþHþ,81),
3.3.2. Methyl b-L-arabinopyranoside (9). Acetyl chloride

T. K. M. Shing et al. / Tetrahedron 59 (2003) 2159–2168
2165
327 (50), 219 (6), 183 (100), 154 (6). Anal. Calcd for
C25H24O5: C, 74.24; H, 5.98. Found: C, 73.79; H, 5.90.
3.3.11. 40-Methylbenzyl 3,4-O-isopropylidene-b-L-
erythro-pentopyranosid-2-ulose (4). Oxidation of aceto-
nide 14 (1.7 g, 5.8 mmol) in a similar manner as in the
preparation of ulose 1 gave ulose 4 (1.4 g, 83%) as a
yellowish syrup: Rf 0.5 (EtOAc–hexane, 1:1); [a]2D0¼
1H NMR (CDCl3) d 7.20 (4H, m), 4.89 (1H, s), 4.76 (1H, d,
J¼11.4 Hz), 4.69 (1H, d, J¼5.7 Hz), 4.56 (1H, d, J¼
11.4 Hz), 4.52 (1H, dd, J¼5.7 and 2.1 Hz), 4.30 (1H, dd,
J¼13.5 and 2.1 Hz), 4.10 (1H, d, J¼13.5 Hz), 2.36 (3H, s),
1.46 (3H, s), 1.39 (3H, s); 13C NMR (CDCl3) d 199.3, 138.7,
133.4, 129.8, 129.7, 128.9, 128.8, 110.9, 99.4, 78.1, 75.9,
70.4, 59.1, 27.7, 26.6, 21.7; MS (FAB) m/z 315 (MþNaþ,
9), 293 (MþHþ, 0.4), 289 (7), 154 (72), 105 (100). Anal.
Calcd for C16H20O5: C, 65.74; H, 6.90. Found: C, 66.21; H,
7.33.
3.3.6. Methyl 3,4-O-isopropylidene-b-L-arabinopyrano-
side (13). Acetonation of glycoside 9 (3.3 g, 20 mmol) in a
similar manner as in the preparation of the acetonide 11
gave the acetonide 13 (3.7 g, 90%) as a syrup: Rf 0.4
(hexane–EtOAc, 1:1); [a]2D0¼þ218.8 (c 1.2, CHCl3) {lit.,9
[a]2D0¼þ220 (c 1, CHCl3)}; MS (FAB) m/z 205 (MþHþ,
42), 173 (65), 154 (100), 137 (67).
þ119.5 (c 0.8, CHCl3); IR (CHCl3) 1634 (CvO) cm21
;
3.3.7. 40-Methylbenzyl 3,4-O-isopropylidene-b-L-arabino-
pyranoside (14). Acetonation of glycoside 10 (3.2 g,
12.6 mmol) in a similar manner as in the preparation of
acetonide 11 gave acetonide 14 (3.0 g, 81%) as white
crystals: Rf 0.50 (hexane–EtOAc, 1:1): mp 179–1808C;
[a]D22¼þ119 (c 1.0, CHCl3) {lit.,7 mp 178–1808C; [a]2D2¼
1
þ117.8 (c 1.2, CHCl3)}; IR (CHCl3) 3530 (OH) cm21; H
3.3.12. Acetal 15. A suspension of 8 (0.5 g, 2.1 mmol) in
methanol (15 mL) containing 2,2,3,3-tetramethoxybutane15
(0.45 g, 2.5 mmol), trimethyl orthoformate (0.9 mL) and
camphorsulfonic acid (5 mg) was heated under reflux for
12 h. The cooled reaction mixture was then treated with
powdered NaHCO3 (ca. 0.5 g), filtered and the filtrate
concentrated under reduced pressure. The crude residue was
flash chromatographed (hexane–EtOAc, 2:1) to afford
acetal 15 as a white solid (0.56 g, 76%). Recrystallization
from hexane–EtOAc gave white prisms: mp 140–1418C;
[a]2D3¼þ5 (c 1.5, CHCl3); Rf 0.45 (hexane–EtOAc, 1:1); IR
NMR (CDCl3) d 7.20 (4H, m), 4.92 (1H, d, J¼3.7 Hz), 4.75
(1H, d, J¼11.7 Hz), 4.51 (1H, d, J¼11.7 Hz), 4.25–4.16
(2H, m), 4.02 (1H, dd, J¼13.1, 2.4 Hz), 3.93 (1H, dd, J¼
13.1, 1.2 Hz), 3.83–3.75 (1H, m), 2.36 (3H, s), 2.20 (1H, d,
J¼7.8 Hz), 1.54 (3H, s), 1.37 (3H, s); 13C NMR (CDCl3) d
137.8, 134.0, 129.2, 128.1, 109.1, 96.8, 75.9, 72.9, 69.9,
69.6, 60, 27.8, 25.8, 21.1; MS (EI) m/z 279 (Mþ215, 1.0),
189 (3.1), 105 (100).
3.3.8. Benzyl 3,4-O-isopropylidene-b-L-erythro-pento-
pyranosid-2-ulose (1). Acetonide 11 (1.8 g, 6.4 mmol)
was dissolved in dry CH2Cl2 (15 mL). PDC (2.6 g, 7 mmol),
1
(CHCl3) 3469 (OH) cm21; H NMR d 7.42–7.26 (5H, m),
4.95 (1H, d, J¼3.0 Hz), 4.75 (1H, d, J¼12.3 Hz), 4.67 (1H,
d, J¼12.3 Hz), 4.19 (1H, dd, J¼10.5, 3.0 Hz), 4.14 (1H, dd,
J¼10.5, 2.7 Hz), 3.93 (1H, m), 3.82 (1H, dd, J¼12.6,
1.0 Hz), 3.70 (1H, dd, J¼12.6, 1.5 Hz), 3.26 (3H, s), 3.22
(3H, s), 1.85 (1H, brs), 1.33 (3H, s), 1.31 (3H, s); 13C NMR
d 137.6, 128.2, 127.9, 127.5, 100.2, 100.1, 96.8, 69.3, 68.0,
65.7, 65.1, 62.9, 47.9, 47.8, 17.7; MS (EI) m/z 322
(Mþ2MeOH, 5), 291 (3), 91 (100). Anal. Calcd for
C18H26O7: C, 61.00; H, 7.39. Found: C, 60.87; H, 7.43.
˚
powdered 4 A molecular sieve (2 g) and glacial acetic acid
(0.1 mL) were added slowly. The mixture was stirred at rt
for 4 h and was suction filtered through a bed of silica gel.
The filtrate was concentrated under reduced pressure to give
a crude colorless syrup. Flash column chromatography
(Et2O–hexane, 3:7) gave ulose 1 (1.6 g, 89%) as a colorless
syrup: Rf 0.50 (EtOAc–hexane, 1:1); [a]2D2¼þ236.7 (c 0.7,
CHCl3) {lit.,9 [a]D20¼þ239.6 (c 0.01, CHCl3)}; MS (FAB)
m/z 301 (MþNaþ, 54), 279 (MþHþ, 85), 263 (78), 171 (99),
101 (100).
3.3.13. Benzoate 16. To a solution of alcohol 15 (1.5 g,
4.24 mmol) in dry CH2Cl2 (20 mL) containing pyridine
(0.67 mL, 8.48 mmol) and a catalytic amount of DMAP was
added benzoyl chloride (0.6 mL, 5.09 mmol). The reaction
mixture was stirred for 24 h at rt and was then poured into
saturated NH4Cl (2 mL). The aqueous phase was extracted
with CH2Cl2 (3£10 mL) and the combined organic extracts
were washed with brine (5 mL), dried (MgSO4), and
filtered. Concentration of the filtrate followed by flash
column chromatography (hexane–Et2O, 5:1) provided 18
(1.88 g, 97%) as a syrup: [a]2D3¼þ5.3 (c 1.2, CHCl3); Rf
3.3.9. Benzyl 3,4-O-diphenylmethylidene-b-L-erythro-
pentopyranosid-2-ulose (2). Oxidation of acetonide 12
(3.8 g, 9.4 mmol) in a similar manner as in the preparation
of ulose 1 gave ulose 2 (3.0 g, 79%) as a yellowish syrup: Rf
0.60 (EtOAc–hexane, 3:7); [a]2D0¼þ60.5 (c 0.3, CHCl3); IR
1
(CHCl3) 1636 (CvO) cm21; H NMR (CDCl3) d 7.53–
7.25 (15H, m), 4.93 (1H, s), 4.79 (1H, d, J¼3 Hz), 4.76 (1H,
d, J¼2.4 Hz), 4.40 (1H, d, J¼11.7 Hz), 4.45–4.47 (1H, m),
4.24–4.25 (2H, m); 13C NMR (CDCl3) d 197.8, 129.3,
129.2, 129.0, 128.9, 128.8, 128.6, 127.2, 127.1, 127.0, 99.2,
78.2, 77.1, 70.7, 59.4; MS (FAB) m/z 403 (MþHþ, 38), 325
(31), 207 (40), 183 (100). Anal. Calcd for C25H22O5: C,
74.61; H, 5.51. Found: C, 74.09; H, 5.63.
0.65 (hexane–Et2O. 1:1); IR (CHCl3) 1690 (CvO) cm21
;
1H NMR (CDCl3) d 8.07–7.29 (10H, m), 5.30 (1H, d,
J¼1.5 Hz), 5.02 (1H, d, J¼3 Hz), 4.77 (1H, d, J¼12.6 Hz),
4.71 (1H, d, J¼12.6 Hz), 4.36 (1H, dd, J¼10.5, 3 Hz), 4.31
(1H, dd, J¼10.5, 3 Hz), 3.93 (1H, d, J¼13 Hz), 3.81 (1H,
dd, J¼13, 1.5 Hz), 3.29 (3H, s), 3.27 (3H, s), 1.34 (3H, s),
1.22 (3H, s); 13C NMR d 166.2, 137.3, 132.8, 130.3, 129.7,
128.2, 127.9, 127.5, 99.8, 96.7, 70.4, 69.5, 65.6, 64.0, 61.6,
47.8, 47.7, 17.5; MS (EI) m/z 458 (Mþ, 2), 444 (100). Anal.
Calcd for C25H30O8: C, 65.49; H, 6.59. Found: C, 65.38; H,
6.59.
3.3.10. Methyl 3,4-O-isopropylidene-b-L-erythro-pento-
pyranosid-2-ulose (3). Oxidation of acetonide 13 (1.5 g,
7.3 mmol) in a similar manner as in the preparation of
ulose 1 gave ulose 3 (1.2 g, 83%) as a yellowish syrup:
Rf 0.40 (EtOAc–hexane, 1:1); [a]2D0¼þ136.5 (c 1.2,
CHCl3) {lit.,9 [a]D20¼þ138.3 (c 0.08, CHCl3)}; MS (FAB)
m/z 255 (MþNaþ, 19), 203 (MþHþ), 187 (27), 171 (50),
154 (37).
3.3.14. Pivalate 17. To a solution of alcohol 15 (1.5 g,

2166
T. K. M. Shing et al. / Tetrahedron 59 (2003) 2159–2168
4.24 mmol) in dry CH2Cl2 (20 mL) containing pyridine
(0.67 mL, 8.48 mmol) and a catalytic amount (10 mg) of
DMAP was added pivaloyl chloride (PivCl) (0.6 mL,
4.6 mmol). The reaction mixture was stirred for 24 h at rt
and was then poured into saturated NH4Cl (2 mL). The
aqueous phase was extracted with CH2Cl2 (3£10 mL) and
the combined organic extracts were washed with brine
(5 mL), dried (MgSO4), and filtered. Concentration of the
filtrate followed by flash column chromatography (hexane–
Et2O, 9:1) provided ester 17 (1.83 g, 99%) as white crystals:
mp 117–1188C; [a]D23¼þ12 (c 0.3, CHCl3); Rf 0.75
3.3.17. 4-O-Benzoyl-1,2-O-isopropylidene-b-L-arabino-
pyranose (20). To a suspension of palladium-on-charcoal
(10 mg) in ethanol (5 mL) was stirred at rt for 1 h under H2
atmosphere (ballon). A solution of the diol 18 (340 mg,
1 mmol) in EtOAc (8 mL) was added and the resulting
mixture was stirred for 24 h under H2 atmosphere. The
mixture was then filtered through celite and the filtrate was
concentrated to give the triol that was used for next step
without purification. The triol was dissolved in acetone
(30 mL), 2,2-dimethoxypropane (0.15 mL, 1.2 mmol) and
p-TsOH (10 mg) was added. The resulting suspension was
stirred at rt for 12 h. The reaction was quenched with NEt3
and then concentrated in vacuo to give a crude yellow syrup.
Flash column chromatography (Et2O–hexane, 1:1) gave
acetonide 20 (164 mg, 56%) as a white solid: mp 138–
1398C; [a]2D3¼þ20 (c 4.5, CHCl3); Rf 0.37 (hexane–Et2O,
1:1); IR (CHCl3) 3489 (OH), 1716 (CvO) cm21; 1H NMR
d 8.03–7.41 (5H, m), 5.43 (1H, d, J¼3.3 Hz), 5.38–5.33
(1H, m), 4.44 (1H, t, J¼3.6 Hz), 4.20 (1H, t, J¼3.6 Hz),
4.02 (1H, dd, J¼12, 4.8 Hz), 3.88 (1H, dd, J¼12, 7.2 Hz),
2.41 (1H, brs), 1.58 (3H, s), 1.39 (3H, s); 13C NMR d 165.6,
133.5, 129.7, 129.3, 128.4, 111.1, 96.4, 77.8, 68.9, 67.2,
60.5, 27.7, 26.0; MS (EI) m/z 295 (Mþþ1, 8), 279 (46), 237
(100). Anal. Calcd for C15H18O6: C, 61.22; H, 6.16. Found:
C, 61.15; H, 6.12.
1
(hexane–Et2O, 1:1); IR (CHCl3) 1729 (CvO) cm21; H
NMR (CDCl3) d 7.42–7.28 (5H, m), 5.15 (1H, m), 4.96
(1H, d, J¼2.7 Hz), 4.74 (1H, d, J¼12.6 Hz), 4.69 (1H, d,
J¼12.6 Hz), 4.19 (1H, dd, J¼10.8, 3 Hz), 4.15 (1H, dd,
J¼10.8, 3 Hz), 3.85 (1H, d, J¼12.3 Hz), 3.58 (1H, dd,
J¼12.9, 1.8 Hz), 3.24 (3H, s), 3.20 (3H, s), 1.30 (3H, s),
1.23 (9H, s), 1.19 (3H, s); 13C NMR d 177.8, 137.4, 128.2,
127.9, 127.5, 99.8, 99.6, 96.8, 69.5, 68.8, 65.5, 64.1, 61.8,
47.9, 47.5, 38.9, 27.0, 17.6, 17.5; MS (EI) m/z 406 (Mþ–
MeOH, 100), 348 (13). Anal. Calcd for C23H34O8: C, 63.00;
H, 7.81. Found: C, 63.21; H, 7.85.
3.3.15. Benzyl 4-O-benzoyl-b-L-arabinopyranoside (18).
To a solution of benzoate 16 (720 mg, 1.57 mmol) in
CH2Cl2 (10 mL) was added 90% aqueous TFA (0.5 mL) at
rt. The mixture was stirred for 24 h at rt, and was then
poured into saturated NaHCO3 (5 mL). The aqueous layer
was extracted with CH2Cl2 (2£10 mL). The combined
organic extracts were washed with brine (10 mL), dried over
MgSO4, and filtered. Concentration of the filtrate followed
by flash chromatography (hexane–EtOAc, 1:1) gave the
diol 18 (464 mg, 86%) as a syrup: [a]2D3¼þ222 (c 0.5,
CHCl3); Rf 0.3 (hexane–EtOAc, 1:1); IR (CHCl3) 3395
(OH), 1723 (CvO) cm21; 1H NMR d 8.07–7.33 (10H, m),
5.42 (1H, s), 5.10 (1H, d, J¼3.6 Hz), 4.79 (1H, d, J¼
11.7 Hz), 4.57 (1H, d, J¼11.7 Hz), 4.09 (1H, dd, J¼9.6,
3.3 Hz), 4.01–3.96 (2H, m), 3.88 (1H, dd, J¼12.9, 1.8 Hz),
1.94 (2H, brs); 13C NMR d 166.5, 136.7, 133.3,
129.7, 129.5, 128.5, 128.4, 128.1, 97.7, 72.0, 69.9, 69.8,
68.9, 60.9; MS (EI) m/z 344 (Mþ, 1), 253 (3), 91 (100).
Anal. Calcd for C19H20O6: C, 66.27; H, 5.85. Found: C,
66.42; H, 5.99.
3.3.18. 1,2-O-Isopropylidene-4-O-pivaloyl-b-L-arabino-
pyranose (21). To a suspension of palladium-on-charcoal
(10 mg) in ethanol (5 mL) was stirred at rt for 1 h under H2
atmosphere. A solution of diol 19 (240 mg, 0.74 mmol) in
EtOAc (8 mL) was added and the resulting mixture was
stirred for 24 h under H2 atmosphere. The mixture was then
filtered through celite and the filtrate was concentrated to
give the triol that was used for next step without
purification. The triol was dissolved in acetone (30 mL),
2,2-dimethoxypropane (0.13 mL, 1 mmol) and p-TsOH
(10 mg) was added. The resulting suspension was stirred
at rt for 12 h. The reaction was quenched with NEt3 and then
concentrated in vacuo to give a crude yellow syrup. Flash
column chromatography (Et2O–hexane, 1:1) gave the
acetonide 21 (116 mg, 57%) as a white solid: mp 52–
538C; [a]2D3¼þ0.1 (c 6.1, CHCl3); Rf 0.32 (hexane–Et2O,
1:1); IR (CHCl3) 3496 (OH), 1723 (CvO) cm21; 1H NMR
d 5.37 (1H, d, J¼3 Hz), 5.09 (1H, m), 4.29 (1H, t, J¼
3.6 Hz), 4.14 (1H, t, J¼3.6 Hz), 3.92 (1H, dd, J¼11.7,
5.1 Hz), 3.70 (1H, dd, J¼12, 6.9 Hz), 2.21 (1H, brs), 1.55
(3H, s), 1.34 (3H, s), 1.22 (9H, s); 13C NMR d 177.5, 111.0,
96.5, 77.8, 68.1, 67.3, 60.7, 38.9, 27.6, 27.0, 26.0; MS (EI)
m/z 274 (Mþ, 7), 259 (32), 217 (100). Anal. Calcd for
C13H22O6: C, 56.92; H, 8.08. Found: C, 56.76; H, 8.44.
3.3.16. Benzyl 4-O-pivaloyl-b-L-arabinopyranoside (19).
To a solution of pivalate 17 (60 mg, 0.13 mmol) in CH2Cl2
(10 mL) was added 90% aqueous TFA (0.5 mL) at rt. The
mixture was stirred for 24 h, and was then poured into
saturated NaHCO3 (5 mL). The aqueous layer was extracted
with CH2Cl2 (2£10 mL). The combined organic extracts
were washed with brine (10 mL), dried over MgSO4, and
filtered. Concentration of filtrate followed by flash chroma-
tography (hexane–EtOAc, 2:1) gave the diol 19 (34 mg,
81%) as an oil: [a]D23¼þ170 (c 1.3, CHCl3); Rf 0.44
(hexane–EtOAc, 1:1); IR (CHCl3) 3415 (OH), 1729
3.3.19. 4-O-Benzoyl-1,2-O-isopropylidene-b-L-threo-
pentopyranos-3-ulose (5). To a solution of the alcohol 20
(104 mg, 0.37 mmol) was dissolved in dry CH2Cl2 (10 mL)
˚
were added PDC (167 mg, 0.44 mmol) and powdered 4 A
1
(CvO) cm21; H NMR d 7.34–7.29 (5H, m), 5.07 (1H,
molecular sieve (170 mg). The mixture was stirred at rt for
4 h and was then suction filtered through a pad of silica gel.
The filtrate was concentrated under reduced pressure to give
a crude syrup that was flash column chromatographed
(Et2O–hexane, 1:2) to give ulose 5 (94 mg, 92%) as a white
solid: mp 100–1018C; [a]2D3¼þ18 (c 0.4, CHCl3); Rf 0.4
m), 4.97 (1H, d, J¼3.6 Hz), 4.71 (1H, d, J¼11.7 Hz), 4.50
(1H, d, J¼11.7 Hz), 3.97 (1H, dd, J¼9.9, 3.3 Hz), 3.81 (2H,
m), 3.65 (1H, dd, J¼12.9, 1.8 Hz), 2.47 (2H, brs); 13C NMR
d 178.4, 136.9, 128.4, 127.9, 97.9, 71.1, 69.7, 69.6, 68.7,
60.9, 39.0, 27.0; MS (EI) m/z 324 (Mþ, 4), 233 (23), 217
(100). Anal. Calcd for C17H24O6: C, 62.95; H, 7.47. Found:
C, 62.76; H, 7.51.
1
(hexane–Et2O, 1:1); IR (CHCl3) 1730 (CvO) cm21; H
NMR d 8.11–7.44 (5H, m), 5.88 (1H, dd, J¼6.3, 3.3 Hz),

T. K. M. Shing et al. / Tetrahedron 59 (2003) 2159–2168
2167
5.71 (1H, d, J¼4.2 Hz), 4.81 (1H, dd, J¼13.2, 6.3 Hz), 4.45
(1H, d, J¼4.2 Hz), 3.96 (1H, dd, J¼13.2, 3.3 Hz), 1.74
(3H, s), 1.41 (3H, s); 13C NMR d 197.6, 165.4, 133.6,
130.0, 128.9, 128.5, 113.4, 101.4, 78.6, 73.2, 64.5, 26.0,
25.5; MS (EI) m/z 292 (Mþ, 59), 234 (43), 105 (100). Anal.
Calcd for C15H16O6: C, 61.64; H, 5.52. Found: C, 61.66; H,
5.52.
129.5, 128.2, 127.6, 127.4, 96.0, 72.1, 69.5, 69.4, 67.8, 62.4;
MS (EI) m/z 345 (Mþþ1, 100), 288 (80). Anal. Calcd for
C19H20O6: C, 66.27; H, 5.85. Found: C, 66.29; H, 5.86.
3.3.23. Benzyl 2-O-benzoyl-3,4-O-benzylidene-b-L-ara-
binopyranoside (24). To a solution of the diol 23 (2.0 g,
5.8 mmol) in benzene (50 mL) was added benzylaldehyde
(0.7 g, 6.6 mmol) and p-TsOH (10 mg). The resulting
solution was heated under reflux with a Dean–Stark trap
for 12 h and was then poured into saturated NaHCO3
(5 mL). The aqueous layer was extracted with CH2Cl2
(2£10 mL). The combined organic extracts were washed
with brine (10 mL), dried over MgSO4, and filtered. The
filtrate was concentrated under reduced pressure, and the
residue was purified through flash column chromatography
(hexane–Et2O, 10:1) to give the major benzylidene isomer
24 (2.0 g, 80%) as a colorless syrup (1.3 g,): Rf 0.33
(hexane–Et2O, 3:1); [a]2D3¼þ142 (c 0.7, CHCl3); IR
3.3.20. 1,2-O-Isopropylidene-4-O-pivaloyl-b-L-threo-
pentopyranos-3-ulose (6). Oxidation of the acetonide 21
(100 mg, 0.36 mmol) in a similar manner as in the
preparation of ulose 5 gave ulose 6 (82 mg, 84%) as a
white solid: mp 61–638C; [a]2D3¼214 (c 0.8, CHCl3); Rf
0.34 (hexane–Et2O, 1:1); IR (CHCl3) 1729 (CvO) cm21
;
1H NMR d 5.69 (1H, d, J¼4.2 Hz), 5.59 (1H, dd, J¼6.3,
3.3 Hz), 4.67 (1H, dd, J¼12.9, 6.3 Hz), 4.37 (1H, d, J¼
4.2 Hz), 3.79 (1H, dd, J¼12.9, 3.3 Hz), 1.69 (3H, s), 1.38
(3H, s), 1.26 (9H, s); 13C NMR d 197.6, 177.3, 113.2, 101.3,
78.5, 72.3, 64.3, 38.7, 27.1, 26.0, 25.7; MS (EI) m/z 272
(Mþ, 100), 256 (78). Anal. Calcd for C13H20O6: C, 57.34; H,
7.40. Found: C, 57.34; H, 7.40.
1
(CHCl3) 1629 (CvO) cm–1; H NMR (CDCl3) d 8.12–
7.22 (15H, m), 6.27 (1H, s), 5.33 (1H, dd, J¼8.4, 3.6 Hz),
5.22 (1H, d, J¼3.6 Hz), 4.87 (1H, dd, J¼8.1, 5.1 Hz), 4.77
(1H, d, J¼12.3 Hz), 4.55 (1H, d, J¼12.3 Hz), 4.33 (1H, dd,
J¼5.1, 1.8 Hz), 4.43 (1H, d, J¼13.2 Hz), 4.07 (1H, dd, J¼
13.2, 2.4 Hz); 13C NMR (CDCl3) d 166.0, 138.7, 136.9,
133.7, 133.2, 130.1, 129.9, 129.5, 129.1, 128.3, 127.8,
127.5, 126.1, 102.8, 95.3, 74.3, 73.5, 69.9, 69.6, 58.8; MS
(EI) m/z 433 (MþHþ, 100), 432 (59), 325 (15). Anal. Calcd
for C26H24O6: C, 72.21; H, 5.59. Found: C, 72.04; H, 5.67.
3.3.21. Benzyl 2-O-benzoyl-3,4-O-isopropylidene-b-L-
arabinopyranoside (22). To a solution of acetonide 11
(1.9 g, 6.78 mmol) in dry CH2Cl2 (20 mL) containing
pyridine (1 mL), and a catalytic amount (10 mg) of
DMAP was added benzoyl chloride (0.9 mL, 8.0 mmol).
The reaction mixture was stirred for 24 h at rt and was then
poured into saturated NH4Cl (2 mL). The aqueous phase
was extracted with CH2Cl2 (3£10 mL) and the combined
organic extracts were washed with brine (5 mL), dried
(MgSO4), and filtered. Concentration of the filtrate followed
by flash column chromatography (hexane–Et2O, 3:1)
provided benzoate 22 (2.52 g, 97%) as a syrup: [a]2D3¼
þ171 (c 1.1, CHCl3); Rf 0.31 (hexane–Et2O, 3:1); IR
3.3.24. Benzyl 3,4-di-O-benzoyl-b-L-threo-pentopyranosid-
3-ulose (7). A solution of the benzylidene 24 (1.4 g,
3.2 mmol) in benzene (50 mL) containing barium carbonate
(5.0 g), NBS (0.95 g, 5.3 mmol), water (0.25 g, 13.9 mmol),
and AIBN (10 mg) was heated under reflux for 1 h. The
cooled mixture was then poured into saturated NaHCO3
(5 mL). The aqueous layer was extracted with Et2O
(2£10 mL). The combined organic extracts were washed
with brine (10 mL), dried over MgSO4, and filtered. The
filtrate was concentrated under reduced pressure to give the
crude dibenzoate 25 that was used for next step without
purification. To a solution of the dibenzoate 25 (1.1 g,
2.45 mmol) in dry CH2Cl2 (10 mL) were added slowly PDC
1
(CHCl3) 1716 (CvO) cm21; H NMR (CDCl3) d 8.11–
7.18 (10H, m), 5.18 (1H, dd, J¼8.1, 3.3 Hz), 5.12 (1H, d,
J¼3.6 Hz), 4.75 (1H, d, J¼12.3 Hz), 4.57 (1H, dd, J¼8.1,
5.7 Hz), 4.51 (1H, d, J¼12.3 Hz), 4.33 (1H, dt, J¼5.7,
1.5 Hz), 4.09 (2H, d, J¼1.8 Hz), 1.59 (3H, s), 1.38 (3H, s);
13C NMR d 165.9, 137.1, 133.1, 129.9, 129.7, 128.3,
128.2, 127.7, 127.4, 109.4, 95.4, 73.6, 73.0, 72.6, 69.5, 58.8,
28.0, 26.3; MS (EI) m/z 277 (Mþ2C7H7O, 100). Anal.
Calcd for C22H24O6: C, 68.74; H, 6.29. Found: C, 69.07; H,
6.28.
˚
(1.1 g, 2.94 mmol) and powdered 4 A molecular sieve (1 g).
The mixture was stirred at rt for 4 h and was then suction
filtered through a pad of silica gel. The filtrate was con-
centrated under reduced pressure to give a crude syrup.
Flash column chromatography (Et2O–hexane, 1:4) gave
ulose 7 (0.98 g, 69% from 24) as a syrup: [a]2D3¼þ181 (c
0.5, CHCl3); Rf 0.51 (Et2O–hexane, 1:1); IR (CHCl3) 1729,
3.3.22. Benzyl 2-O-benzoyl-b-L-arabinopyranoside (23).
To a solution of benzoate 22 (1.30 g, 3.38 mmol) in CH2Cl2
(10 mL) was added 90% aqueous TFA (0.5 mL) at rt. The
mixture was stirred for 24 h rt and was then poured into
saturated NaHCO3 (5 mL). The aqueous layer was extracted
with CH2Cl2 (2£10 mL). The combined organic extracts
were washed with brine (10 mL), dried over MgSO4, and
filtered. Concentration of the filtrate followed by flash
chromatography (hexane–EtOAc, 2:1) gave the diol 23
(1.16 g, quant. yield) as white crystals: mp 120–1228C;
[a]2D3¼þ186 (c 0.3, CHCl3); Rf 0.27 (hexane–EtOAc, 1:1);
1
1709 (CvO) cm 21; H NMR (CDCl3) d 8.08–7.24 (15H,
m), 6.05 (1H, d, J¼4 Hz), 5.48 (1H, d, J¼4 Hz), 5.41 (1H, d,
J¼1.2 Hz), 4.80 (1H, d, J¼12.3 Hz), 4.61 (1H, d, J¼
12.3 Hz), 4.30 (1H, dd, J¼13.2, 1.2 Hz), 4.22 (1H, dd,
J¼13.2, 1.2 Hz); 13C NMR (CDCl3) d 193.4, 165.0, 136.2,
133.7, 133.5, 129.9, 129.8, 128.7, 128.6, 128.5, 128.4,
128.0, 127.6, 99.1, 76.8, 74.7, 69.5, 63.4; MS (EI) m/z 339
(Mþ2C7H7O, 20), 338 (100). Anal. Calcd for C26H22O7: C,
69.95; H, 4.97. Found: C, 69.89; H, 5.23.
1
IR (CHCl3) 3435 (OH), 1629 (CvO) cm21; H NMR d
8.07–7.23 (10H, m), 5.25 (1H, dd, J¼9.9, 3.6 Hz), 5.18
(1H, d, J¼3.6 Hz), 4.77 (1H, d, J¼12.3 Hz), 4.55 (1H, d, J¼
12.3 Hz), 4.28 (1H, dd, J¼9.9, 3.6 Hz), 4.09 (1H, m), 4.01
(1H, dd, J¼12.9, 1.2 Hz), 3.84 (1H, dd, J¼12.6, 1.8 Hz),
2.18 (2H, brs); 13C NMR d 166.8, 137.2, 133.2, 129.8,
3.3.25. Cinnamyl acetate epoxide.16 1H NMR d 7.27–7.16
(5H, m), 4.39 (1H, dd, J¼12.3, 3.3 Hz), 4.00 (1H, dd, J¼
12.3, 5.7 Hz), 3.71 (1H, d, J¼1.8 Hz), 3.19–3.15 (1H, m),
2.02 (3H, s); 13C NMR d 170.3, 136.0, 128.2, 128.1, 125.4,

2168
T. K. M. Shing et al. / Tetrahedron 59 (2003) 2159–2168
63.9, 58.9, 56.0, 20.4; MS (FAB) m/z 193 (Mþþ1, 62), 133
J. H.; Cheung, K. K. J. Am. Chem. Soc. 1996, 118, 491.
(b) Yang, D.; Wang, X. C.; Wong, M.-K.; Yip, Y. C.; Tang,
M.-W. J. Am. Chem. Soc. 1996, 118, 11311. (c) Yang, D.;
Wong, M. K.; Yip, Y. C.; Wang, X. C.; Tang, M. W.; Zheng,
J. H.; Cheung, K. K. J. Am. Chem. Soc. 1998, 120, 5943.
(d) Yang, D.; Yip, Y. C.; Chen, J.; Cheung, K. K. J. Am. Chem.
Soc. 1998, 120, 7659.
(100).
Acknowledgements
This work was supported by the CUHK Direct Grant
(account no. 2060231).
5. (a) Tu, Y.; Wang, Z. X.; Shi, Y. J. Am. Chem. Soc. 1996, 118,
9806. (b) Wang, Z. X.; Tu, Y.; Frohn, M.; Shi, Y. J. Org.
Chem. 1997, 62, 2328. (c) Wang, Z. X.; Shi, Y. J. Org. Chem.
1997, 62, 8622. (d) Wang, Z. X.; Tu, Y.; Frohn, M.; Zhang,
J. R.; Shi, Y. J. Am. Chem. Soc. 1997, 119, 11224. (e) Frohn,
M.; Dalkiewicz, M.; Tu, Y.; Wang, Z. X.; Shi, Y. J. Org.
Chem. 1998, 63, 2948. (f) Wang, Z. X.; Shi, Y. J. Org. Chem.
1998, 63, 3099. (g) Cao, G. A.; Wang, Z. X.; Tu, Y.; Yu, H.;
Shi, Y. Tetrahedron Lett. 1998, 39, 4425. (h) Zhu, Y.; Tu, Y.;
Shi, Y. Tetrahedron Lett. 1998, 39, 7819. (i) Tu, Y.; Wang,
Z. X.; Frohn, M.; He, M.; Yu, H.; Tang, Y.; Shi, Y. J. Org.
Chem. 1998, 63, 8475. (j) Wang, Z. X.; Miller, S. M.;
Anderson, O. P.; Shi, Y. J. Org. Chem. 1999, 64, 6443.
(k) Wang, Z. X.; Cao, G. A.; Shi, Y. J. Org. Chem. 1999, 64,
7646. (l) Warren, J. D.; Shi, Y. J. Org. Chem. 1999, 64, 7675.
(m) Frohn, M.; Zhou, X.; Zhang, J.-R.; Tang, Y.; Shi, Y. J. Am.
Chem. Soc. 1999, 121, 7718. (n) Shu, L.; Shi, Y. Tetrahedron
Lett. 1999, 40, 8721. (o) Tian, H.; She, X.; Shu, L.; Yu, H.;
Shi, Y. J. Am. Chem. Soc. 2000, 122, 11551. (p) Tian, H.; She,
X.; Shi, Y. Org. Lett. 2001, 3, 715. (q) Tian, H.; She, X.; Xu, J.;
Shi, Y. Org. Lett. 2001, 3, 1929. (r) Zhu, Y.; Shu, L.; Tu, Y.;
Shi, Y. J. Org. Chem. 2001, 66, 1818.
References
1. (a) Katsuki, T.; Sharpless, K. B. J. Am. Chem. Soc. 1980, 102,
5974. (b) Rossiter, B. E.; Katsuki, T.; Sharpless, K. B. J. Am.
Chem. Soc. 1981, 103, 464. (c) Johnson, R. A.; Sharpless, K. B.
In Catalytic Asymmetric Synthesis; Ojima, I., Ed.; VCH: New
York, 1993; Chapter 4.1. (d) Katsuki, T.; Martin, V. S. Org.
React. 1996, 48, 1.
2. (a) Jacobsen, E. N.; Zhang, W.; Muci, A. R.; Ecker, J. R.;
Deng, L. J. Am. Chem. Soc. 1991, 113, 7063. (b) Lee, N. H.;
Muci, A. R.; Jacobsen, E. N. Tetrahedron Lett. 1991, 32, 5055.
(c) Deng, L.; Jacobsen, E. N. J. Org. Chem. 1992, 57, 4320.
(d) Palucki, M.; McCormick, G. J.; Jacobsen, E. N.
Tetrahedron Lett. 1995, 36, 5457. (e) Katsuki, T. Coord.
Chem. Rev. 1995, 140, 189. (f) Mukaiyama, T. Aldrichim. Acta
1996, 29, 59.
3. (a) Curci, R.; Fiorentino, M.; Serio, M. R. J. Chem. Soc.,
Chem. Commun. 1984, 155. (b) Curci, R.; D’Accolti, L.;
Fiorentino, M.; Rosa, A. Tetrahedron Lett. 1995, 36, 5831.
(c) Brown, D. S.; Marples, B. A.; Smith, P.; Walton, L.
Tetrahedron 1995, 51, 3587. (d) Song, C. E.; Kim, Y. H.; Lee,
K. C.; Lee, S. G.; Jin, B. W. Tetrahedron: Asymmetry 1997, 8,
2921. (e) Adam, W.; Zhao, C.-G. Tetrahedron: Asymmetry
1997, 8, 3995. (f) Denmark, S. E.; Wu, Z.; Crudden, C. M.;
Matsuhashi, H. J. Org. Chem. 1997, 62, 8288. (g) Bergbreiter,
D. E. Chemtracts: Org. Chem. 1997, 10, 661. (h) Armstrong,
A.; Hayter, B. R. Chem. Commun. 1998, 621. (i) Dakin, L. A.;
Panek, J. S. Chemtracts: Org. Chem. 1998, 11, 531. (j) Adam,
W.; Saha-Moller, C. R.; Zhao, C.-G. Tetrahedron: Asymmetry
1999, 10, 2749. (k) Carnell, A. J.; Johnstone, R. A. W.; Parsy,
C. C.; Sanderson, W. R. Tetrahedron Lett. 1999, 40, 8029.
(l) Armstrong, A.; Hayter, B. R. Tetrahedron 1999, 55, 11119.
(m) Armstrong, A.; Hayter, B. R.; Moss, W. O.; Reeves, J. R.;
Wailes, J. S. Tetrahedron: Asymmetry 2000, 11, 2057.
(n) Solladie-Cavallo, A.; Bouerat, L. Org. Lett. 2000, 2, 3531.
4. (a) Yang, D.; Yip, Y. C.; Tang, M. W.; Wong, M. K.; Zheng,
6. Shing, T. K. M.; Lloyd-Williams, P. J. Chem. Soc., Chem.
Commun. 1987, 423.
7. Shing, T. K. M.; Chow, H.-F.; Chung, I. H. F. Tetrahedron
Lett. 1996, 37, 3713.
8. Shing, T. K. M.; Li, L.-H. J. Org. Chem. 1997, 62, 1230.
9. Bennett, M.; Gill, G. B.; Pattenden, G.; Shuker, A. J.;
Stapleton, A. J. Chem. Soc., Perkin Trans. 1 1991, 929.
10. Montgomery, R. E. J. Am. Chem. Soc. 1974, 96, 7820.
11. (a) Chida, N.; Ogawa, S. J. Chem. Soc., Chem. Commun. 1997,
807. (b) Chida, N.; Tobe, T.; Ogawa, S. Tetrahedron Lett.
1994, 35, 7249. (c) Krow, G. R. Org. React. 1993, 43, 251.
12. Chang, H.-T.; Sharpless, K. B. J. Org. Chem. 1996, 61, 6456.
13. Brandes, B. D.; Jacobsen, E. N. J. Org. Chem. 1994, 59, 4378.
14. Witkop, B.; Foltz, C. M. J. Am. Chem. Soc. 1957, 79, 197.
15. Montchamp, J. L.; Tian, F.; Hart, M. E.; Frost, J. W. J. Org.
Chem. 1996, 26, 3897.
16. Ilankumaran, P.; Verkade, J. G. J. Org. Chem. 1999, 64, 3086.