Stereoselective tandem epoxidation-alcoholysis/hydrolysis of glycals with molybdenum catalysts

Article abstract of DOI:10.1002/adsc.201000614

Molybdenum catalysts are efficient and selective catalysts for the tandem epoxidation/alcoholysis or epoxidation/hydrolysis of glucal and galactal derivatives. In glucal derivatives the selectivity is mainly controlled by the allylic substituent at position 3 of the glycal, obtaining in general the products derived from the initial formation of the α-epoxide (gluco) when this hydroxy group is protected, while products derived from the β-epoxide (manno) are mainly obtained when it is unprotected. In galactal derivatives the estereoselectivity is always high to give the α-epoxide (galacto) and independent of the protecting groups. Copyright

Full text of DOI:10.1002/adsc.201000614

FULL PAPERS
DOI: 10.1002/adsc.201000614
Stereoselective Tandem Epoxidation–Alcoholysis/Hydrolysis of
Glycals with Molybdenum Catalysts
Irene Marꢀn,a M. Isabel Matheu,a Yolanda Dꢀaz,a,* and Sergio Castillꢁna,*
a
Departament de Quꢀmica Analꢀtica i Quꢀmica Orgꢂnica, Universitat Rovira i Virgili, C/Marcel·lꢀ Domingo s/n, 43007
Tarragona, Spain
Fax : (+34)-977-558-446; e-mail: yolanda.diaz@urv.cat or sergio.castillon@urv.cat
Received: July 30, 2010; Revised: November 2, 2010; Published online: December 7, 2010
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/adsc.201000614.
Abstract: Molybdenum catalysts are efficient and se- products derived from the b-epoxide (manno) are
lective catalysts for the tandem epoxidation/alcoholy- mainly obtained when it is unprotected. In galactal
sis or epoxidation/hydrolysis of glucal and galactal derivatives the estereoselectivity is always high to
derivatives. In glucal derivatives the selectivity is give the a-epoxide (galacto) and independent of the
mainly controlled by the allylic substituent at posi- protecting groups.
tion 3 of the glycal, obtaining in general the products
derived from the initial formation of the a-epoxide Keywords: carbohydrates; dihydroxylation; directing
(gluco) when this hydroxy group is protected, while effect; molybdenum; oxidation; stereoselectivity
Introduction
have only recently been identified. Thus, Ru-porphy-
[19]
rin,[16] CH3ReO3 (MTO),[17,18] Ti
Glycoconjugates play an important role in many bio- turelloꢃs peroxotungstate (PW4O24
ACHTUNGTRNE(NGNU O-i-Pr)4 and Ven-
3À [19]
)
have been
logical processes such as immune response, inflamma- used in the sequential epoxidation–alcoholysis or ep-
tion, pathogen recognition, etc.[1] This fact has spurred oxidation–hydrolysis of glycals. The epoxides initially
the research for new synthetic methodologies looking obtained are opened in situ to give 2-hydroxy glyco-
for an easy and selective access to these compounds. sides or dihydroxy derivatives in the presence of the
In this way glycals are versatile building blocks for epoxidation catalysts, depending on whether the reac-
the synthesis of oligosaccharide motifs,[2] 2-deoxygly- tion is carried out in the presence of alcohols or
cosides,[3] C-glycosides,[4] nucleosides[5] and other bio-
logically important molecules.[6–8] Glycals are usually
activated by epoxidation,[9] to furnish 1,2-anhydro
sugars, which are used as glycosyl donors[2] or as pre-
cursors of more stable glycosyl donors.[10] Glycals are
also activated by dihydroxylation to give protected
sugar 1,2-diols, which have been used in the synthesis
of O-glycosides,[11] C-glycosides[12] and in intramolecu-
lar O-glycosylations.[13]
The most common epoxidation method uses dime-
thyldioxirane[9] (DMDO). However, DMDO is unsta-
ble and explosive and its use presents serious draw-
backs. Other stoichiometric reagents include m-chlo-
ACHTUNGTRENNUNG
roperbenzoic acid (mCPBA)/KF[14] and diphenyl sulf-
oxide/Tf2O.[15] Epoxides trans to the substitutent at C-
3 are obtained as the major product (Scheme 1a). The
need of atomic economy and environmental aware-
ness fuelled the development of new catalytic epoxi-
dation methods. Catalysts for epoxidation of glycals Scheme 1.
Adv. Synth. Catal. 2010, 352, 3407 – 3418
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3407

Irene Marꢀn et al.
FULL PAPERS
water, respectively (Scheme 1b). Other methods for reaction in which the active catalyst, which is presum-
synthesising 1,2-diols from glycals are based on the di- ably a peroxy-molybdenum, is formed.
hydroxylation reaction by OsO4-NMO,[11b,c] and
Thereafter, the intact active complex can react with
[20]
RuCl3/CeCl3·7H2O/NaIO4, and on the one-pot ep- the alkene, to irreversibly form the epoxide and
oxidation–hydrolytic opening using DMDO in ace- alkoxo-substituted metal complex. Substitution of the
tone (Scheme 1c).[21] Sugar 1,2-diols can also be classi- alkoxo-substituted complex by fresh hydroperoxide
cally prepared by hydrolysis of orthoesters.[22]
could re-form the active catalyst and complete the
The transition metal-catalyzed processes afore- catalytic cycle. The analogous H2O2-mediated process
mentioned mostly afford products mainly derived of involves the intermediacy of an Mo-peroxo complex.
the epoxidation trans to the C-3 substituent Since the discovery by Mimoun et al.[26] that peroxo-
(Scheme 1b). Interestingly, the dihydroxylation (epox- Mo complexes can stoichiometrically epoxidize al-
idation/hydrolysis) of unprotected glucal using MoO3 kenes under anhydrous conditions in organic solvents,
as a catalyst in water as a solvent affords mainly man- the mechanism of the oxygen transfer process has
nose, in a process that involves directed b-epoxida- been a subject of intensive debate. The long-standing
tion, probably via a hydrogen bond between the controversy about the mechanism of olefin epoxida-
neighbouring allylic hydroxy group and the catalyst.[23] tion
with
2A(OPr3)] has recently been settled with the
help of density functional methods.[27] The theoretical-
h2-diperoxomolybdenum
complexes
3À [19]
Only Venturelloꢃs peroxotungstate (PW4O24
)
pro- [MoO(O2) CHTUNGTRENNUNG
vided a similar selectivity in unprotected glycals.
However, in spite of the results previously men- ly predicted energies of the transition states and the
tioned using MoO3 and the facts that molybdenum calculated intrinsic reaction coordinates show that the
complexes are readily available and one of the most oxygen transfer takes place in a single-step reaction
active and efficient catalysts for alkene epoxidation,[24] via a direct attack of the alkene to the peroxo group
no systematic investigations of their use in glycal ep- as proposed by Sharpless,[28] thus ruling out Mimounꢃs
oxidation–alcoholysis or epoxidation–hydrolysis have metalla-2,3-dioxolane model.[26]
been carried out. In this report we show that a variety
We initially evaluated the epoxidation of unprotect-
of molybdenum complexes catalyze these processes ed glucal (1) under similar conditions to those previ-
with high conversions and selectivities.
ously explored,[23] using two molybdenum catalysts in
polar solvents such as water and methanol and using
different oxidizing reagents. The results are collected
in Table 1. The key role of the molybdenum complex
in the oxygen transfer process was first established by
Results and Discussion
The general mechanism for the molybdenum-cata- the lack of reactivity of the oxidant towards 1 in its
lyzed epoxidation reaction has been extensively stud- absence (results not displayed in Table 1).
ied and can summarily be viewed as a stepwise pro-
When the reaction was carried out in water with
cess (Scheme 2).[25] Using alkyl hydroperoxides as ter- readily available MoO3 as a catalyst precursor and
minal oxidant, the first step is a rapid and reversible H2O2 as stoichiometric oxidant, mannose (2b) was
almost exclusively obtained, in agreement with that
reported by Bilik[23] et al. (Table 1, entry 1). Note that
in no case was the epoxide obtained, as a conse-
quence of the high reactivity of the anhydro sugar in
the presence of nucleophilic solvents. The use of
methanol as solvent and working in similar conditions
allowed us to obtain exclusively methyl a-mannopyro-
side (3b) as consequence of an epoxidation–methanol-
ysis process (entries 2 and 3), although the reaction
rate was lower than in water. In order to seek for
more general oxidation conditions that would involve
protected or partially protected glycals, less polar sol-
vents and more soluble oxidizing agents should also
be explored. With this purpose in mind, oxidation re-
actions of 1 using tert-butyl hydroperoxide (TBHP) as
a terminal oxidant were explored next. Reactions
with this oxidant driven in methanol took place with
complete conversion although they required higher
temperatures than with H2O2 (entries 4 and 5). Inter-
estingly, the stereoselectivity was observed to depend
Scheme 2.
on the solvent for solubilizing TBHP. Thus, an almost
3408
asc.wiley-vch.de
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 3407 – 3418

Stereoselective Tandem Epoxidation–Alcoholysis/Hydrolysis of Glycals
Table 1. Tandem epoxidation–hydrolysis/glycosylation of glucal with molybdenum catalysts.
Entry
Catalyst
Peroxide
Solvent
Temp. [8C]
Time [h]
Conv. [%][a,b]
Products
Ratio[a]
4:96
1[c]
2[c]
3[c]
4[d]
5[d]
6[d]
7[d]
MoO3
MoO3
MoO3
MoO3
MoO3
Mo(CO)6
Mo(CO)6
H2O2 in H2O
H2O2 in H2O
H2O2 in H2O
TBHP 5.5M in C9H20
TBHP 70 wt% in H2O
TBHP 5.5 M in C9H20
TBHP 70 wt% in H2O
H2O
r.t.
r.t.
r.t.
50
50
50
50
48
48
96
72
72
72
72
>98
50
2a/2b
3b
3b
3a/3b
3a/3b
3a/3b
3a/3b
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
>98
>98
>98
>98
>98
45:55
9:91
55:45
9:91
[a]
1
Determined by integration in the H NMR spectrum of anomeric proton signals of compounds in the final reaction crude.
Selectivity>98%.
[b]
[c]
[d]
Conditions: 1.71 mmol glycal, 1% mol catalyst, 2.5 mL H2O2 (5%) in the solvent.
Conditions: 0.36 mmol glycal, 5 mol% catalyst, 1.2 mL TBHP, 2 mL MeOH.
1:1 mixture of methyl b-glucopyranoside (3a) and
In order to assess the factors influencing the activi-
methyl a-mannopyranoside (3b) was obtained when ty of the catalyst, reactivity of the starting glycals,
TBHP was dissolved in nonane, while the stereoselec- protecting group effect and influence on the facial se-
tivity was 9:91 when TBHP in water was used (en- lectivity in product formation, a range of differently
tries 4 and 5). The anomeric configuration of the protected or partially unprotected glucals (4–10), and
gluco derivative 3a is b, while it is a for the manno galactals (11–14) were prepared using standard litera-
derivative 3b. This is a consequence of the selective ture procedures and used as starting materials for the
trans opening of the gluco and manno epoxides ini- epoxidation reaction (Figure 1).
tially formed under the reaction conditions. The use
Table 2, Table 3, Table 4 and Table 5 summarize the
of TBHP with more soluble molybdenum catalysts results obtained in the molybdenum-catalyzed epoxi-
such as Mo(CO)6 provided similars results to those dation/hydrolysis or epoxidation/alcoholysis of pro-
from MoO3 (entries 6 and 7).
tected glycals with tert-butyl hydroperoxide as an oxi-
A general trend of these epoxidation reactions with dant. Different [Mo]/TBHP catalytic systems and dif-
polar solvents is the long reaction times. This deacti- ferent solvents were explored using tri-O-acetyl-d-
vating effect has been already observed,[26,28,29] and glucal (4) as a substrate (Table 2, entries 1–7). Thus,
two reasons have been proposed, depending on the the reaction of 4 with Mo(CO)6 in MeOH gave a mix-
oxidant used. On the one hand, coordination of a ture of compounds 15a(b)/15b(a) in a 70:30 ratio as a
basic ligand (solvent molecule) to a peroxo-Mo com- consequence of a tandem epoxidation–trans epoxide
plex induces a decrease of electrophilicity of the opening process (Table 2, entry 1). The reaction
peroxo group due to donation of electron density needed 36 h for completion, in line with the results of
from the basic solvent via the metal center, thus low- Table 1. The gluco derivative 15a was now principally
ering its reactivity.[30] On the other hand, the coordi- obtained, in agreement with that reported for other
nating solvent may compete with the stoichiometric oxidation catalysts, and on the contrary to that ob-
TBHP oxidant for the substitution of the alcohol served in the reaction of unprotected glucal.[17–19,23]
ligand in the final molybdenum complex thus slowing The use of less polar solvents such as toluene did not
down the generation of the catalytic species.[25b,c]
prevent epoxide opening, probably due to the Lewis
Figure 1. Structures of glycals 4–14 used in Mo/TBHP dihydroxylation.
Adv. Synth. Catal. 2010, 352, 3407 – 3418
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3409

Irene Marꢀn et al.
FULL PAPERS
Table 2. [Mo]-catalyzed oxidation of glucals (4–7). Influence of the solvent.[a]
Entrya)
Substrate
Catalyst
Solvent
Temp. [8C]
Time [h]
Conv [%]
Products
Ratio[b]
1
2
Mo(CO)6
MeOH
MeOH
CH2Cl2
toluene
toluene
toluene
toluene
50
50
r.t.
30
40
80
50
36
36
6
3
96
1
>98
>98
>98
>98
>98
>98
>98
15a/15b
15a/15b
16a/16b
16a/16b
16a/16b
16a/16b
16a/16b
70:30
70:30
65:35
65:35
67:33
65:35
62:38
MoO
MoO
MoO
N
3[c]
4[c]
5[c]
6[c]
7[c]
Mo(CO)6
Mo(CO)6
MoO2Cl2
2
8[c]
Mo(CO)6
MoO2ACTHUNGTRENNU(G acac)2
MoO2Cl2
toluene
toluene
toluene
80
30
50
1
3
3
>98
>98
>98
17a/17b
17a/17b
17a/17b
82:18
75:25
77:23
9[c]
10[c]
11[c]
12[c]
Mo(CO)6
MoO2ACHTNUTRGENNG(U acac)2
toluene
toluene
80
30
21
24
>98
>98
18a/18b
18a/18b
96:4
94:6
13[c,d]
14[c,d]
Mo(CO)6
MoO2A(acac)2
toluene
toluene
80
30
21
72
0
0
19a/19b
19a/19b
–
–
CTHUNGTRENNUNG
[a]
Conditions: 0.36 mmol glycal, 5 mol% catalyst, 1.2 mmol TBHP 5.5M in nonane, 2 mL solvent.
Determined by NMR by integration of the anomeric protons in the crude reaction mixture.
The reaction crude was acetylated to give the corresponding acetyl derivatives.
[Si]=TIPS.
[b]
[c]
[d]
acidity of the own catalyst (Table 2, entries 4–7). For ed to be trans, an a/b mixture was observed in each
practical reasons, the crude reaction mixture was ana- case as a result of an anomerization process. Two sig-
lyzed after acetylation of the free hydroxy groups in nals corresponded to gluco derivatives, derived from
the resulting oxidation products (Scheme 3) to afford the a-epoxide (attack from the bottom face of the
around 2:1 mixtures of 16a/16b. Thus, the diastereose- double bond), and other two signals to the manno de-
lectivity of this reaction and subsequent reactions rivatives, resulting from the b-epoxide, (attack from
1
were calculated by integration of the H NMR signals the upper face of the double bond). In all cases epoxi-
of the anomeric protons of the crude product mixture dation–hydrolysis reactions were carefully monitored
after acetylation. The spectral data obtained were as- by TLC and the reported reaction times correspond
signed by comparison with those reported in the liter- to the disappearance of starting glycal. The stereo-
ature.[31] Although the opening of epoxides is expect- chemistry at position 2 in the final acetate products
Scheme 3. Tandem epoxidation–hydrolysis of glycals with [Mo] catalysts.
3410
asc.wiley-vch.de
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 3407 – 3418

Stereoselective Tandem Epoxidation–Alcoholysis/Hydrolysis of Glycals
reflects the stereoselectivity in the initial epoxide for- in glucal 4 (and other glucal derivatives), with all its
mation.
The temperature conditions seemed to be crucial stereoselectivities should be expected, as it is the case.
when Mo(CO)6 was used as a catalyst in apolar sol- In comparison, reaction with the more hindered
substituents in a pseudo-equatorial orientation,[36] low
vents and temperatures of 808C are required for com- glucal 5 afforded better stereoselectivities (Table 2,
plete conversion of 4 in reasonable reaction times entries 8–10), obtaining gluco derivative 17a as a
(entries 5 and 6). As an Mo(0) species, Mo(CO)6 re- major product. Considering that tri-O-benzyl is con-
4
quires an induction period prior to any epoxidation, formationally more biased towards the H5 half-chair
in which the molybdenum complex is oxidized by the conformation than the tri-O-acetyl glucal itself, the in-
alkyl hydroperoxide or the oxygen peroxide to its crease in a-epoxidation might be explained by steric
highest oxidation state, the active Mo(VI)[25a,32] Most hindrance of the protecting benzyl moiety. The exis-
probably, the dissociation of the carbonyl ligands in tence of a p-stacking interaction of the aryl group of
the starting complex in apolar solvents is the limiting the benzyl group at C-3 or C-6 and the p system in
process for activation of the complex at low tempera- the glycal thus blocking the b face could not be dis-
ture.
The nature of the oxidant reagent solution in the
carded either.
Glucals 6 and 7, with bulkier protecting groups
stereoselectivity outcome of the epoxidation reaction were then tested. Glucal 6 afforded the gluco deriva-
was also explored starting from 4 and using Mo(CO)6 tive 18a with excellent stereoselectivity when
as a catalyst precursor, but there were not significant Mo(CO)6 and MoO
2ACHTUNGRTEN(NUNG acac)2 were used as catalyst
differences between the use of TBHP 5.5M in nonane (Table 2, entries 11 and 12).
(Table 2, entry 6) and the use of TBHP 70 wt% in
Unexpectedly, reaction of 7 did not occur, neither
water (98% conversion, 2 h, gluco/manno ratio= with Mo(CO)6 nor with MoO2ACHTNUTRGNE(NUG acac)2 (Table 2, en-
55:45, results not displayed in Table 2).
The reaction was then tried using another soluble ed to the significant steric demand of the bulky TIPS
tries 13 and 14). This lack of reactivity can be attribut-
4
catalyst, MoO2ACHTUNGTRENNUNG(acac)2, in different solvents. In the groups, which destabilize the H5 conformation due to
presence of MeOH, the corresponding methyl glyco- a serious gauche interaction between the protecting
sides 15a and 15b were obtained with a similar stereo- groups at positions 3 and 4 and force the glucal into
5
selectivity to that from Mo(CO)6 (entry 2). Reaction the H4 conformation, where the substituents at C-3,
in dichloromethane or toluene led to 16a/16b with C-4 and C-5 are in a pseudo-axial orientation. These
slightly lower stereoselectivities than with methanol inhibit access of the electrophilic peroxy-molybednum
(entries 3 and 4). Accordingly to the previous results species to the double bond.[19]
obtained with Mo(CO)6, the use of non-coordinating
solvents led to shorter reaction times.
Following previous experiments, we decided to
carry out the dihydroxylation of 4 and 5 with N-heter-
Other (dioxo)Mo(VI) catalyst precursors, like ocyclic ligands-containing [Mo] complexes in order to
MoO2Cl2, allowed performance of the reaction in modulate the reactivity and selectivity of the oxida-
short reaction times but with virtually the same ste- tion (Table 3).[17b,37] It is well known that this kind of
reoselectivity (Table 2, entry 7). The use of MoO3 was Mo complexes presents better solubilities in apolar
not considered with protected glycals as a conse- solvents and lower Lewis acidity, preventing the for-
quence of its insolubility problems in non-polar sol- mation of diols. In our experiment, the use of the
vents.
complex C1/TBHP[38] or the MoO3/L1/TBHP catalytic
In general, epoxidation–hydrolysis of glycal 4 pro- system did not prevent epoxide opening and only diol
ceeded with poor selectivities, probably as a conse- derivatives were obtained as the main reaction prod-
quence of steric and conformational factors. Glycals ucts. Moreover, reactions were in general slower
are well known to be conformationally flexible.[33] (Table 3 versus Table 2) in agreement with reported
Moreover, the conformational preference between epoxidation reactions catalyzed by N,N-bidentate
the normal half-chair (4H5 conformation) and the con- ligand-containing Mo complexes.[39] This deactivating
formationally inverted isomer (5H4 conformation) is effect is in line with that previously stated for the
known to be sensitive not only to configuration but effect of coordinating solvents in the decrease of elec-
also to hydroxy group protection.[34] More important- trophilicity of the Mo-peroxo or alkylperoxo-Mo
4
ly, this H5/5H4 distribution can affect the stereoselec- complex and has been also determined by theoretical
tivity of addition reaction at the glycal alkene. In fact, studies on peroxo-Mo complexes.[32,40] Higher stereo-
reported data on electrophilic addition reactions of selectivies in comparison with the unmodified catalyt-
glucals locked in a 4H5 conformation have usually ic systems could be obtained although catalytic per-
shown low diastereofacial selectivities, with a slight formance was random. Thus, in terms of stereoselec-
preference for the addition from the a face.[35] On the tivity, the MoO3/L1/TBHP catalytic system worked
basis of conformational NMR-based studies, describ- best with glycal 4 (Table 3, entry 2), whereas oxida-
4
ing the H5 half-chair as the most stable conformation
Adv. Synth. Catal. 2010, 352, 3407 – 3418
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3411

Irene Marꢀn et al.
FULL PAPERS
Table 3. Sequential epoxidation–hydrolysis of glucals 4 and 5 catalyzed by Mo/ligand catalytic systems.
Entry[a]
Substrate
Catalyst
Temp. [8C]
Time [h]
Conv. [%][b,c]
Products
Ratio[b]
1
2
complex C1
MoO3/L1
50
80
48
24
>98
>98
16a/16b
16a/16b
55:45
76:24
3
4
complex C1
MoO3/L1
50
80
6
24
>98
50
17a/17b
17a/17b
88:12
nd
[a]
Conditions: 0.36 mmol glycal, 5 mol% catalyst, 1.2 mmol TBHP 5.5M in nonane, 2 mL solvent.
Determined by NMR by integration of the anomeric protons of acetylated compounds of the crude reaction mixture.
Selectivity>98%.
[b]
[c]
tion of glycal 5 was more efficient with complex C1 dation-hydrolysis, the diacetyl glucal 10 was reacted
(Table 3, entry 3). under the optimized reaction conditions (Table 4,
Epoxidation–hydrolysis reactions from differently entry 4) to furnish 16a/16b (gluco/manno ratio, ca.
protected glucals 2–7 showed diastereoselectivities 20:80) with the manno-configured pyranosyl acetates
mainly governed by steric factors, so that the initial as the major products. Note that the manno selectivity
epoxidation occurs preferentially by attack from the increased in comparison to the results obtained from
opposite face to that of OR group at C-3. After these conformationally-fixed glucal 8, and that it is practi-
initial results, and taking into account that the epoxi- cally reversed in comparison to the obtained from
dation–hydrolysis of unprotected glucal affords man- compound 4 (Table 2, entry 4).
nose (see Table 1) which involves electrophilic oxygen
To continue studying the effect of glycal configura-
transfer syn to the C-3 hydroxy moiety, we decided to tion in the stereoselectivity, the reaction of differently
investigate next the reaction of glucals containing a protected galactals 11–14 with [Mo]/TBHP was
free hydroxy group on that position.
Thus, 8 was treated with TBHP (Table 4, entries 1 with TBHP in the presence of MoO
and 2) in the presence of MoO2A(acac)2 and complex after acetylation, compound 21a as the only product
tested. The reaction of tri-O-acetyl-d-galactal (11)
CTHUNGTRENNUNG
CTHUNGTRENNUNG
C1, to obtain 20b (manno) as the major product, with (Table 5, entry 1). The reaction is much more stereo-
a stereoselectivity gluco:manno 30:70. The goal of selective from galactal 11 than from glucal derivative
using conformationally-restricted 8, which is locked in 4 (Table 5, entry 1 vs. Table 2, entry 4). This is in
4
a H5 on account of its isopropylidene group, is to agreement with that previously observed by other
gain insight into the parameters that govern the selec- groups,[19] and indicates that the configuration at C-4
tivity in oxygen transfer mechanism. Compared to re- plays a major role in determining the degree of ster-
sults in Table 1 or Table 2, epoxidation of a glucal 8 eoselection, with the OR group at C-4 directing the
occurs with reversed selectivity, leading preferentially attack of oxygen preferentially to the opposite face
to a mannose derivative, thus showing the implication when it is placed in an axial position.
of the allylic hydroxy group in the oxygen transfer
The reaction of tri-O-benzyl-d-galactal (12) and tri-
process. This fact was confirmed by epoxidation of O-pivaloyl-d-galactal (13) under similar conditions
the acetyl-protected derivative 9 (Table 4, entry 3). In also provided compounds 22a and 23a with an excel-
this case, the products derived from the gluco epoxide lent stereoselectivity (Table 5, entries 2 and 3). Espe-
20a were predominantly formed, as expected for an cially significant results were obtained in the reaction
epoxidation reaction dominated by steric factors.
of galactal 14, which has the 6-OH unprotected. The
With the aim of testing the influence of the confor- reaction was considerably slower and finished after
mational freedom in the stereoselectivity of the epoxi- 24 h to give 24a/24b with a stereoselectivity galacto:ta-
3412
asc.wiley-vch.de
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 3407 – 3418

Stereoselective Tandem Epoxidation–Alcoholysis/Hydrolysis of Glycals
Table 4. Mo-catalyzed sequential epoxidation–hydrolysis of partially protected glycals 8–10.[a]
Entry[a]
Substrate
Catalyst
MoO2A(acac)2
Temp. [8C]
Time [h]
Conv. [%][b,c]
Products
Ratio[b]
1
2
G
30
50
24
48
>98
50
20a/20b
20a/20b
30:70
28:72
complex C1
3
MoO
MoO
G
50
30
53
24
>98
>98
20a/20b
16a/16b
65:35
20:80
4
[a]
Conditions: 0.36 mmol glycal, 5 mol% catalyst, 1.2 mmol TBHP 5.5M in nonane, 2 mL toluene.
Determined by integration in the H NMR spectrum of the anomeric protons of acetylated compounds in the crude reac-
[b]
1
tion mixture.
Selectivity>98%.
[c]
Table 5. Mo-catalyzed sequential epoxidation–hydrolysis of differently protected galactal derivatives 11–14.
Entry[a]
1
Substrate
Catalyst
Temp. [8C]
Time [h]
4
Conv. [%][b,c]
Products
Ratio[b]
MoO
MoO
MoO
(acac)2
30
>98
21a/21b
>98:2
2
3
C
30
30
3
7
>98
>98
22a/22b
23a/23b
97:3
97:3
2ACHTUNGTNER(NUNG acac)2
4
MoO
(acac)2
30
24
>98
24a/24b
70:30
[a]
Conditions: 0.36 mmol glycal, 5 mol% catalyst, 1.2 mmol TBHP 5.5M in nonane, 2 mL solvent.
Determined by integration in the H NMR spectrum of the anomeric protons of acetylated compounds in the crude reac-
[b]
1
tion mixture.
Selectivity>98%.
[c]
lo ratio 70:30 (Table 5, entry 4). The increase in the tion by the endo face, may be indicative of a directing
percentage of isomer 24b, resulting from the epoxida- effect of the hydroxy group at C-6.
Adv. Synth. Catal. 2010, 352, 3407 – 3418
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3413

Irene Marꢀn et al.
FULL PAPERS
Figure 2. Modified Adamꢃs model for transitions-state struc-
tures of the Mo-catalyzed epoxidation of allylic alcohols.
Scheme 4.
alcohols and relating them with the regio- and diaste-
reoselective epoxidation of 1-methylgeraniol with Ti,
According to the literature, it seems clear that the they proposed for both metals, a metal-alcoholate
reaction of TBHP with an oxo-Mo complex leads to complex model.[44e] However, this proposal is in dis-
the formation of an Mo(VI) alkylperoxo complex,
ACHTUNGTRENaNGNU greement with the non-coordinative model proposed
which then transfers oxygen to the olefin.[38,41] Several by Di Furia and others.
studies previously attempted to isolate the intermedi-
Notwithstanding, since Adamꢃs proposal for the
ate Mo(VI)-tert-butyl peroxide complex without suc- Mo-catalyzed transition state structures is mainly
cess.[42] In order to gain insight in the nature of the based on the observed erythro/threo diastereoselectiv-
active catalytic species in the Mo-catalyzed epoxida- ities of chiral allylic alcohols, we believe that such a
tion of glycals, we decided to monitor the process by model can still be valid in terms of dihedral angle of
solution phase IR, also without success.
the transition state, as long as the metal-alkoxo bond
Another interesting issue, the role of the allylic hy- would be replaced with a hydrogen bond between the
droxylic group in the metal-catalyzed directed epoxi- peroxy oxygen atom and the hydroxy group in the al-
dation reactions of allylic alcohols, has been already lylic alcohol (Figure 2).
studied.[43,44] Di Furia et al. used geraniol and linalool
Taking into account all these pieces of evidence,
as probe substrates for gaining insight into the mecha- the most obvious result extracted from Table 4 is the
nism of the Mo-catalyzed epoxidation of alkenes.[44c] syn-epoxidation of partially protected glycals with the
Based on previous experiments, the same authors had allylic group unmasked. This is indeed indicative of
already proposed that the Mo-catalyzed epoxidation some kind of directing effect by the hydroxy group.
reaction of geraniol with hydrogen peroxide was con- The precedent information stated above, altogether
sistent with an intermolecular rather than intramolec- with the experimental stereoselectivity ratios obtained
ular oxygen transfer that involved the formation of a in this work, which are moderately high, tend to point
hydrogen bond between the peroxo-Mo complex and to a hydrogen-bond assistance model rather than the
the alcohol moiety in the allylic alcohol coordinative metal-alkoxo model. A qualitative con-
(Scheme 4b).[45] Further experimentation led them to formational analysis of glycal derivatives altogether
conclude that the moderate activation of Mo-cata- with the adjustement to the previous described inter-
lyzed epoxidation of allylic alcohols compared with molecular Mo-catalyzed epoxidation model via hydro-
the very large one observed for V systems might not gen-bond activation may also give some justification
be due to
a
coordinated substrate as in
V
to the degree of stereoselection of the process.
(Scheme 4a), but is more reasonable to propose the
As stated above, the optimum dihedral angle a for
involvement of a transition state effect, i.e., a hydro- Mo-catalyzed epoxidation transition state conforma-
gen-bonding assistance in the oxygen transfer proc- tion of allylic alcohols was determined to be in the
ess.[44c] Moreover, Sheldon et al. arrived at similar range of 70–908, which would fit more consistently
5
conclusions when using pinane hydroperoxide (PHP) with the glycal H4 half-chair conformation, where the
as mechanistic probe in the metal-catalyzed epoxida- allylic alcohol is in a pseudo-axial position, rather
4
tion of alkenes to distinguish between oxometal and than in the H5 conformation, in a pseudo-equatorial
peroxometal pathways.[44d] In the late 1990s, Adam position. Conformationally fixed glycals (8), though,
5
et al. proposed a transition state structure for the Ti- cannot attain the H4 conformation, and therefore the
and Mo-TBHP epoxidation of chiral allylic alcohols, resulting transition state structure will not benefit
mainly dominated by 1,3A strain, where the preferred from the same stabilization degree that should have
4
dihedral angle a for the epoxidation of the peroxy in the inverted H5 conformation (Scheme 5a).[46] On
complex lies between 708 and 908. On the basis on the other hand, conformationally mobile glycal 10 is
4
the similar diastereoselectivity ratios of Ti and Mo most likely to be preferentially in the H5 conforma-
systems of the epoxidation reactions of chiral allylic tion. Nevertheless, if epoxidation is agreed to proceed
3414
asc.wiley-vch.de
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 3407 – 3418

Stereoselective Tandem Epoxidation–Alcoholysis/Hydrolysis of Glycals
Analogous considerations might be applied to ep-
oxidation of galactal 14 unprotected at position 6. Si-
multaneous protection of hydroxy groups at positions
3 and 4 as an isopropylidene acetal is not expected to
prevent conformational equilibrium between half-
chair conformations. On one hand, epoxidation of 14
from the most stable half-chair 4H5 conformation
might be expected to proceed from the a face, based
on the steric interaction with the isopropylidene
acetal group. In such a conformation, the hydroxy-
methyl chain in a pseudo-equatorial position is in a
too distal position to establish any activating effect.
5
The H4 isomer conformation, though, may benefit
from intermolecular hydrogen assistance from the hy-
droxymethyl moiety in
(Scheme 5c).
a pseudo-axial position
Table 6 displays a comparison of the selectivities in
epoxidation/alcoholysis or epoxidation/hydrolysis of
glycals obtained in this work using molybdenum cata-
lysts, with those obtained with other catalysts, particu-
larly MTO, titanium tetraisopropoxide and Venturel-
loꢃs catalysts. Although glycosides or hydroxy deriva-
tives are obtained depending on the papers, it is as-
sumed that the gluco/manno ratio depends on the se-
lectivity of the epoxidation step.
Scheme 5. Conformational equilibrium of glucals
8 (re-
stricted, a), 10 (unrestricted, b) and galactal 14 (c), and di-
rected intermolecular epoxidation through hydrogen-bond
assistance.
Stereoselectivity in the epoxidation of tri-O-acetyl-
d-glucal (4) or tri-O-benzyl-d-glucal (5) is moderate
with all the catalysts except with TiACTHNUTRGEN(UNG O-i-Pr)4, which
under Curtin–Hammet conditions, ground state con- provides excellent selectivities (entries 2 and 3). An
formations are not determinant of the product distri- important improvement in the selectivity using TMO
bution but rather the relative energies of transition was obtained when the reaction was carried out in the
5
state structures. In this case, H4 half-chair is not the presence of nitrogen ligands. A similar effect, al-
most populated conformation but is the conformation though less significant was obtained with molybde-
that benefits from the greatest stabilization in the num catalysts. More important, however, is the effect
transition state structure, with dihedral angles in the of the protecting groups. Thus, the presence of bulky
range of what is considered an optimum value. The protecting groups such as the pivaloyl allows us to
slight increase in the manno selectivity obtained with reach very high selectivities using molybdenum cata-
glycal 10 compared to that obtained with 8 is consis- lysts (entry 4); although the presence of TIPS pre-
tent with this appreciation (Scheme 5b).
cludes the reaction. A really different behaviour
Table 6. Comparison of selectivity a/b epoxide in the tandem epoxidation-alcoholysis or hydrolysis with different catalysts.
Entry Substrate
MTO[a],[18] MTO[17] MTO[b,c]
modif.[17]
Ti
Pr)4
(O-i-
Venturelloꢃs
method[a],[19]
[Mo] (this
work)
[a],[19]
1
2
3
4
glucal unprotected
66:33
66:33
86:14
40:60
94:6
>98:2
94:6(a)
<2:98
86:14
86:14
0:100
70:30
86:14
96:4(b)
3,4,6-O-Ac
3,4,6-O-Bn
3,4,6-O-Si-t-BuMe2 (a),
3,4,6-O-Piv (b)
4,6-O-isopropylidene
4,6-O-Ac
64:36
85:15
77:23
93:7
5
6
7
8
9
66:33
75:25
28:72
20:80
>98:2
97:3
galactal OAc
95:5
>98:2 97:3
>98:2
>98:2
>98:2
>98:2
>98:2
OBn
OPiv
97:3
[a]
[b]
[c]
Reactions performed in methanol and consequently the methyl glycoside was obtained.
Tandem epoxidation/phosphorylation in ionic liquids.
Tandem epoxidation phosphorylation in the presence of nitrogen ligands (mainly pyridine).
Adv. Synth. Catal. 2010, 352, 3407 – 3418
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3415

Irene Marꢀn et al.
FULL PAPERS
among the compared catalysts is observed in the ep- Experimental Section
oxidation of unprotected glucal, TiACTHNUTRGNEN(UG O-i-Pr)4 catalyst
afforded very low selectivity, while Venturelloꢃs and General Experimental Methods
molybdenum catalysts provided almost complete ste-
All chemicals used were reagent grade and used as supplied.
Glucals 1a and 1b and galactals 11a and 11b were commer-
cially available. Glucal 9 was obtained through a conven-
tional procedure for acetylation. All other glycals were syn-
thesized according to literature procedures: 6,[47] 7,[48] 8,[49]
10,[49] 13,[48] 14.[50] HPLC grade dichloromethane (CH2Cl2),
tetrahydrofuran (THF), dimethylformamide (DMF) and di-
ethyl ether were dried using a solvent purification system
(Pure SOLV system-4ꢅ). The other solvents were purified
using standard procedures.[51] 1H and 13C NMR spectra were
recorded on a Varianꢅ Mercury 400 and Varianꢅ 400-MR,
(both at 400 MHz and 100 MHz, respectively) spectrometer
in CDCl3 as solvent, with chemical shifts (d) referenced to
reoselectivity in the manno epoxide (entry 1). This
effect is also observed with molybdenum catalysts in
partially protected derivatives. Thus, compounds de-
rived from the manno epoxide are mainly formed
when 3-OH is unprotected (entries 5 and 6). Concern-
ing galactal derivatives all the studied catalysts pro-
vide excellent selectivities (entries 7–9).
Conclusions
1
internal standards CDCl3 (7.27 ppm H, 77.23 ppm 13C) or
Highly stereoselective one-pot epoxidation–hydrolysis
or epoxidation–alcoholysis of glucal and galactal de-
rivatives was achieved using H2O2 or TBHP and mo-
lybdenum catalysts. The stereoselectivity in the epoxi-
dation-hydrolysis of glucal derivatives depends on the
presence or absence of the protecting groups. In the
absence of protecting groups the manno epoxide is
preferently obtained. Thus, mannose and methyl man-
nopyranoside can be easily and stereoselectively ob-
tained from d-glucal using molybdenum catalysts and
performing the reaction in water or methanol, respec-
tively.
However, when the reaction is carried out in a pro-
tected glucal the trans epoxide is mainly obtained.
The better stereoselectivities in this last case were ob-
tained when pivaloyl groups were present. Modified
molybdenum catalysts allowed increases, in some
cases, of the stereoselectivity of the process.
Me4Si as an internal reference (0.00 ppm), unless otherwise
specified. 2D correlation spectra (TOCSY, gCOSY, NOESY,
gHSQC, gHMBC) were visualized using the VNMR pro-
gram (Varianꢅ). ESI-MS were run on an Agilentꢅ 1100
Series LC/MSD instrument. Optical rotations were mea-
sured at room temperature in a Perkin–Elmerꢅ 241 MC ap-
paratus with 10 cm cells. Elemental analysis (C, H, N, S) was
performed on a Carlo Erbaꢅ EA 1108 Analyser in the
Servei de Recursos Cientifics (SRCiT-URV). Analytical thin
layer chromatography (TLC) was performed on Merckꢅ
silica gel 60 F254 glass or aluminium plates. Compounds were
visualized by UV (254 nm) irradiation or dipping the plate
in a suitable developing solution. Flash column chromatog-
raphy was carried out using forced flow or by gravity of the
indicated solvent on Flukaꢅ or Merckꢅ silica gel 60 (230–400
mesh). Radial chromatography was performed on 1, 2, or
4 mm plates of Kieselgel 60 PF254 silica gel, depending on
the amount of product.
In 4,6-diprotected glucal derivatives, the hydroxy
group at position 3 directs the stereoselectivity. Thus,
the gluco derivative is obtained in fully protected glu-
cals (anti attack), while the manno derivative is
mainly obtained (syn attack) when 3-OH is unprotect-
ed.
A similar behaviour to that observed for protected
glucals was stated for the protected galactal deriva-
tives, although in this case the stereoselectivity is
always good independently of the protecting groups
present, and the galacto derivative is almost exclusive-
ly obtained in all cases.
General Procedure for Catalytic Dihydroxylation
using Mo(VI)/TBHP Systems in Water – Acetylation
To a solution of glycal (1.71 mmol) in 5% aqueous or meth-
anolic hydrogen peroxide (2.5 mL), the corresponding [Mo]
catalyst (1.0 mol%) was added. After 48 h at room tempera-
ture, undissolved oxide was removed, and the excess of hy-
drogen peroxide was decomposed by treatment for 24 h at
228C with 5% palladized charcoal (5–10 mg). The filtered
mixture was then evaporated under vacuum. The resulting
1
residue was analyzed by H NMR.[24] The ratio of isomers
was determined by integration of anomeric protons of the
corresponding acetyl derivatives.
In conclusion, molybdenum salts and complexes ef-
ficiently catalyze the epoxidation of glycals and the
subsequent epoxide opening to afford glycosides or
hydroxy derivatives depending on whether the solvent
is an alcohol, water or other organic solvent. The ste-
reoselectivities obtained are similar to those of the
best catalysts reported. A directing effect was ob-
served in unprotected or partially protected glycals
that seems to be consistent with a hydrogen-bond as-
sistance model.
General Procedure for Catalytic Dihydroxylation
using Mo(VI)/TBHP Systems in Organic Solvents –
Acetylation
To a solution of glycal (0.36 mmol) in dry toluene (2 mL)
the corresponding [Mo] catalyst (5.0 mol%) and TBHP
(5.5M in nonane, 0.4 mL, 1.2 mmol) were added. The mix-
ture was heated between 30–808C over a 1–96 h period. The
reaction was monitored by TLC. When the reaction was
completed the solution was concentrated under vacuum.
The resulting residue was dissolved in pyridine (2 mL) and
acetic anhydride (1 mL) was added. The resulting solution
3416
asc.wiley-vch.de
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 3407 – 3418

Stereoselective Tandem Epoxidation–Alcoholysis/Hydrolysis of Glycals
was stirred at room temperature for 2–3 h. The reaction
mixture was quenched by adding MeOH. The solution was
extracted with CH2Cl2 and washed successively with brine,
saturated aqueous NaHCO3 and water. The solution was
dried over MgSO4 and then the solvent was removed under
reduced pressure. The resulting residue was analyzed by
1H NMR. The ratio of isomers was determined by integra-
tion of the anomeric protons of the corresponding acetylat-
ed products. When the solvent is methanol, the acetylation
was not needed.
[7] For use in a novel class of glycosylations based in a
[4+2]cycloaddition see: a) G. Capozzi, A. Dios, R. W.
Frank, A. Geer, C. H. Marzabadi, S. Menichetti, C.
Nativi, M. Tamarez, M. Angew. Chem. 1996, 108, 805–
807; Angew. Chem. Int. Ed. Engl. 1996, 35, 777–779;
b) R. W. Franck, C. H. Marzabadi, J. Org. Chem. 1998,
63, 2197–2208.
[8] For the synthesis of thionucleosides from thioglycals
see: K. Haraguchi, A. Nishikawa, E. Sasakura, H.
Tanaka, K. T. Nakamura, T. Miyasaka, Tetrahedron
Lett. 1998, 39, 3713–3716.
[9] R. L. Halcomb, S. J. Danishefsky, J. Am. Chem. Soc.
1989, 111, 6661–6666.
[10] D. M. Gordon, S. J. Danishefsky, Carbohydr. Res. 1990,
206, 361–366.
Acknowledgements
[11] a) L. Shi, Y.-J. Kim, D. Y. Gin, J. Am. Chem. Soc. 2001,
123, 6939–6940; b) W. J. Sanders, L. L. Kiessling, Tetra-
hedron Lett. 1994, 35, 7335–7338; c) A. B. Charette,
J. F. Marcoux, B. Cote, Tetrahedron Lett. 1991, 32,
7215–7218.
Financial support of DGI (CTQ2008-01569-BQU), Consol-
ider Ingenio 2010 (CSD2006-0003) (Ministerio de Ciencia y
Tecnologꢀa, Spain) and technical assistance from the Servei
de Recursos Cientifics (URV) are gratefully acknowledged.
I.M. thanks MICINN for a fellowship.
[12] a) M. Carpintero, I. Nieto, A. Fernandez-Mayoralas, J.
Org. Chem. 2001, 66, 1768–1774; b) T. Vidal, A. Hau-
drechy, Y. Langlois, Tetrahedron Lett. 1999, 40, 5677–
5680; c) S.-C. Hung, C.-H.; Wong, Angew. Chem. 1996,
108, 2829–2832; Angew. Chem. Int. Ed. Engl. 1996, 35,
2671–2674.
[13] a) M. Bols, J. Chem. Soc. Chem. Commun. 1992, 913–
914; b) G. Stork, G. Kim, J. Am. Chem. Soc. 1992, 114,
1087–1088; c) F. Barresi, O. Hindsgaul, J. Am. Chem.
Soc. 1991, 113, 9376–9377.
References
[1] a) A. Varki, Glycobiology 1993, 3, 97–130; b) Y. C.
Lee, R. T. Lee, Acc. Chem. Res. 1995, 28, 321–327;
c) R. A. Dwek, Chem. Rev. 1996, 96, 683–720; d) S.
Hakomori, Adv. Cancer Res. 1989, 52, 257–331;
e) P. M. Rudd, T. Elliott, P. Cresswell, I. A. Wilson,
R. A. Dwek, Science 2001, 291, 2370–2376.
[2] a) S. J. Danishefsky, M. T. Bilodeau, Angew. Chem.
1996, 108, 1482–1522; Angew. Chem. Int. Ed. Engl.
1996, 35, 1380–1419; b) J. Y. Roberge, X. Beebe, S. J.
Danishefsky, J. Am. Chem. Soc. 1998, 120, 3915–3927;
c) F. E. McDonald, H. Y. H. Zhu, J. Am. Chem. Soc.
1998, 120, 4246–4247; d) J. Thiem, M. Gerken, J. Org.
Chem. 1985, 50, 954–958.
[3] a) D. Hou, T. L. Lowary, Carbohydr. Res. 2009, 344,
1911–1940; b) A. Kirschning, M. Jesberger, K.-U.
Schçning, Synthesis 2001, 507–540; c) A. Veyriꢆres, in:
Carbohydrates in Chemistry and Biology, (Eds.: B.
Ernst, G. W. Hart, P. Sinaꢇ), Wiley, Weinheim 2000,
Part I, Vol. I, pp 367–405; d) H. Marzabadi, R. W.
Franck, Tetrahedron 2000, 56, 8385–8417; e) J. C.
Castro-Palomino, R. R. Schmidt, Synlett 1998, 501–
503; f) K. Toshima, K. Tatsuta, Chem. Rev. 1993, 93,
1503–1531.
[14] G. Bellucci, G. Catelani, C. Chiappe, F. D’Andrea, Tet-
rahedron Lett. 1994, 35, 8433–8436.
[15] E. Honda, D. Y. Gin, J. Am. Chem. Soc. 2002, 124,
7343–7352.
[16] a) C.-J. Liu, W.-Y. Yu, S.-G. Li, C.-M. Che, J. Org.
Chem. 1998, 63, 7364–7369; b) X.-Q Yu, J.-S. Huang,
W.-Y Yu, C.-M Che, J. Am. Chem. Soc. 2000, 122,
5337–5342.
[17] a) G. Soldaini, F. Cardona, A. Goti, A. Tetrahedron
Lett. 2003, 44, 5589–5592; b) G. Soldaini, F. Cardona,
A. Goti, A. Org. Lett. 2005, 7, 725–728; c) A. Goti, F.
Cardona, G. Soldaini, C. Crestini, C. Fiani, R. Saladino,
Adv. Synth. Catal. 2006, 348, 476–486.
[18] E. C. Boyd, R. V. H. Jones, P. Quayle, A. J. Waring,
Green Chem. 2003, 5, 679–681.
[19] P. Levecque, D. W. Gammon, H. H. Kinfe, P. Jacobs, D.
De Vos, B. Sels, Adv. Synth. Catal. 2008, 350, 1557–
1568.
[20] P. Tiwari, A. K. Misra, J. Org. Chem. 2006, 71, 2911–
2913.
[4] a) S. N. Thorn, T. Gallagher, Synlett 1996, 856; b) S. Ho-
sokawa, B. Kirschbaum, M. Isobe, Tetrahedron Lett.
1998, 39, 1917–1920.
[21] S. Rani, Y. D. Vankar, Tetrahedron Lett. 2003, 44, 907–
909.
[5] a) R. Robles, C. Rodrꢀguez, I. Izquierdo, M. T. Plaza,
A. Mota, Tetrahedron: Asymmetry 1997, 8, 2959–2965;
b) Y. Dꢀaz, A. El-Laghdach, S. Castillꢁn, Tetrahedron
1997, 53, 10921–10938; c) Y. Dꢀaz, A. El-Laghdach,
M. I. Matheu, S. Castillꢁn, J. Org. Chem. 1997, 62,
1501–1505; d) Q. Chao, J. Zhang, L. Pickering, T. S.
Jahnke, V. Nair, Tetrahedron 1998, 54, 3113–3124; e) F.
Bravo, M. Kassou, Y. Dꢀaz, S. Castillꢁn, Tetrahedron
Lett. 2001, 42, 83–97.
[22] a) F. W. Lichtenthaler, T. Schneider-Adams, J. Org.
Chem. 1994, 59, 6728–6734; b) W. Broder, H. Kunz,
Carbohydr. Res. 1993, 249, 221–241; c) R. R. Schmidt,
F. Effenberger, Carbohydr. Res. 1987, 171, 59–79; d) E.
Wu, Q. Wu, Carbohydr. Res. 1993, 250, 327–333;
e) R. U. Lemieux, A. R. Morgan, Can. J. Chem. 1965,
43, 2198–2204.
[23] V. Bilik, S. Kucar, Carbohydr. Res. 1970, 13, 311–313.
[24] K. B. Sharpless, T. R. Verhoeven, Aldrichimica Acta
1979¸ 12, 63–74.
[6] For use in cyclopropanation and ring expansion see:
C. V. Ramana, R. Murali, M. Nagarajan, J. Org. Chem.
1997, 62, 7694–7703.
Adv. Synth. Catal. 2010, 352, 3407 – 3418
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3417

Irene Marꢀn et al.
FULL PAPERS
[25] a) G. R. Howe, R. R. Hiatt, J. Org. Chem. 1971, 36,
2493–2497; b) A. O. Chong, K. B. Sharpless, J. Org.
Chem. 1977, 42, 1587–1590; c) T. N. Baker, G. J. Mains,
M. N. Sheng, J. K. Jajacek, J. Org. Chem. 1973, 38,
1145–1148; d) W. J. Thiel, J. Mol. Catal. A 1997, 117,
449–454.
[26] H. Mimoun, S. de Roch, L. Sajus, Tetrahedron 1970, 26,
37–50.
[27] D. V. Deubel, J. Sundermeyer, G. Frenking, J. Am.
Chem. Soc. 2000, 122, 10101–10108.
[41] a) F. E. Kꢈhn, M. Groarke, ꢉ. Bencze, E. Herdtweck,
A. Prazeres, A. M. Santos, M. J. Calhorda, C. C.
Rom¼o, I. S. GonÅalves, A. D. Lopes, M. Pillinger,
Chem. Eur. J. 2002, 8, 2370–2383; b) A. Al-Ajlouni,
A. A. Valente, C. D. Nunes, M. Pillinger, A. M. Santos,
J. Zhao, C. C. Rom¼o, I. S. GonÅalves, F. E. Kꢈhn, Eur.
J. Inorg. Chem. 2005, 1716–1723; c) A. M. Al-Ajlouni,
A. Gꢈnyar, M.-D. Zhou, P. N. W. Baxter, F. E. Kꢈhn,
Eur. J. Inorg. Chem. 2009, 1019–1026.
[42] a) P. Chaumette, L. Mimoun, L. Saussine, J. Fischer, A.
Mitschler, J. Organomet. Chem. 1983, 250, 291-310;
b) M. Groarke, I. S. GonÅalves, W. A. Herrmann, F. E.
Kꢈhn, J. Organomet. Chem. 2002, 649, 108–112; c) see
also Refs.[25d,38]
[28] K. B. Sharpless, J. M. Towsend, D. R. Williams, J. Am.
Chem. Soc. 1972, 94, 295–296.
[29] a) G. Amato, A. Arcoria, F. P. Ballisteri, G. A. Toma-
selli, O. Bortolini, V. Conte, F. Di Furia, G. Modena, G.
Valle, J. Mol. Catal 1986, 37, 165–175; b) H. Mimoun,
Angew. Chem. 1982, 94, 750–766; Angew. Chem. Int.
Ed. Engl. 1982, 21, 734–750.
[43] A. H. Hoveyda, D. A. Evans, G. C. Fu, Chem. Rev.
1993, 93, 1307–1370.
[44] a) M. N. Sheng, J. G. Zajacek, J. Org. Chem. 1970, 35,
1839–1843; b) K. B. Sharpless, R. C. Michaelson, J.
Org. Chem. 1973, 37, 6136–6137; c) A. Arcoria, F. P.
Ballistreri, G. A. Tomaselli, F. Di Furia, G. Modena, J.
Org. Chem. 1986, 51, 2374–2376; d) H. E. B. Lempers,
I. Ripollꢆs, A. Garcia, R. E. Sheldon, J. Org. Chem.
1998, 63, 1408–1413; e) W. Adam, T. With, Acc. Chem.
Res. 1999, 32, 703–710; f) W. Adam, C. M. Mitchell, R.
Paredes, A. K. Smerz, L. A. Veloza, Liebigs Ann. Re-
cueil 1997, 1365–1369.
[30] D. V. Deubel, J. Sundermeyer, G. Frenking, Eur. J. Org.
Chem. 2001, 1819–1827.
[31] For triacetylated and tribenzylated gluco derivatives,
see: Ref.[20] For triacetylated and tribenzylated manno
derivaties, see: a) J.-R. Ella-Menye, X. Nie, G. Wang,
Carbohydr. Res. 2008, 343, 1743–1753; b) S. Valerio, A.
Pastore, M. Adinolfi, A. Iadonisi, J. Org. Chem. 2008,
73, 4496–4503.
[32] G. R. Haas, J. W. Kolis, Organometallics 1998, 17,
4454–4460, and references cited therein.
[33] M. Rico, M. Santoro, Org. Mag. Resonance 1976, 8,
49–55.
[34] a) J. Thiem, P. Ossowski, J. Carbohydr. Chem. 1984, 3,
287–313; b) W. R. Roush, D. P. Sebesta, C. E. Bennett,
Tetrahedron 1997, 53, 8825–8836.
[35] G. Grewal, N. Kaila, R. W. Franck, J. Org. Chem. 1992,
57, 2084–2092.
[36] A. A. Chalmers, R. H. Hall, J. Chem. Soc. Perkin Trans.
2 1974, 728–732.
[37] P. Ferreira, W.-M. Xue, E. Bencze, E. Herdtweck, F.
[45] O. Bortolini, F. Di Furia, G. Modena, J. Mol Catal.
1983, 19, 319–329.
[46] Another experimental evidence that underscores the
different mode of action between V and Mo is that the
reaction glycal 8 with VOACHTNUTRGNENG(U acac)2 and TBHP did not led
to even traces of the epoxidation–hydrolysis product.
[47] Y. Matsushita, K. Sugamoto, Y. Kita, T. Matsui, Tetra-
hedron Lett. 1997, 38, 8709–8712.
[48] K. H. Dçtz, F. Otto, M. J. Nieger, J. Organomet. Chem.
2001, 621, 77–88.
[49] E. W. Holla, Angew. Chem. 1989, 101, 222–223;
Angew. Chem. Int. Ed. Engl. 1989, 28, 220–221.
[50] T. Toyokuni, S. Cai, B. Dean, Synthesis 1992, 1236–
1238.
[51] D. D. Perrin, W. L. F. Armarego, Purification of Labo-
ratory Chemicals, 3rd edn., Pergamon Press, Oxford,
1989.
Kꢈhn, Inorg. Chem. 2001, 40, 5834–5841.
[38] J. M. Mithcell, N. S. Finney, J. Am. Chem. Soc. 2001,
123, 862–869.
[39] P. Carreiro, C. Monteiro, G. Yong-en, A. J. Burke, A. I.
Rodrigues, J. Mol. Catal. A 2006, 260, 295–298.
[40] C. Di Valentin, P. Gisdakis, I. V. Yudanov, N. Rçsch, J.
Org. Chem. 2000, 65, 2996–3004.
3418
asc.wiley-vch.de
ꢄ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 3407 – 3418