Reaction of cyclic ketals with ceric ammonium nitrate in acetonitrile/water

Article abstract of DOI:10.1016/S0040-4020(01)01006-7

The possibility of performing the hydrolysis of cyclic acetals and ketals under basic or mildly alkaline conditions by ceric ammonium nitrate (CAN) in MeCN/H₂O or in 10% aq. methanol is rejected. However a role for Ce(IV) as Lewis acid is evidenced at pH 4.4. Under these conditions, which allow hydrolysis of cyclic ketals leaving glycosidic bonds unaltered, a selectivity of ketal cleavage due to steric hindrance is also observed. Similar yields are obtained when acetone/water replaces MeCN/H₂O as solvent.

Full text of DOI:10.1016/S0040-4020(01)01006-7

TETRAHEDRON
Pergamon
Tetrahedron 58 ,2002) 129±133
Reaction of cyclic ketals with ceric ammonium nitrate in
acetonitrile/water
Emiliano Manzo, Gaspare Barone, Emiliano Bedini, Alfonso Iadonisi, Lorenzo Mangoni
and Michelangelo Parrillip
Á
Dipartimento di Chimica Organica e Biologica, Universita di Napoli Federico II, Complesso Universitario Monte S. Angelo Via Cintia,
480126 Napoli, Italy
Received 28 June 2001; revised 10 September 2001; accepted 4 October 2001
AbstractÐThe possibility of performing the hydrolysis of cyclic acetals and ketals under basic or mildly alkaline conditions by ceric
ammonium nitrate ,CAN) in MeCN/H2O or in 10% aq. methanol is rejected. However a role for Ce,IV) as Lewis acid is evidenced at pH 4.4.
Under these conditions, which allow hydrolysis of cyclic ketals leaving glycosidic bonds unaltered, a selectivity of ketal cleavage due to
steric hindrance is also observed. Similar yields are obtained when acetone/water replaces MeCN/H2O as solvent. q 2002 Elsevier Science
Ltd. All rights reserved.
1. Introduction
When 1 was dissolved in a mixture of MeCN and a buffer
solution of borate/HCl ,Merck) at pH 8 and treated with
CAN ,4%) at 608C, according to a new procedure of the
same authors,2 the expected ketal cleavage did not occur. In
order to verify whether the nature of the starting material
might induce a different reaction course, the non-carbo-
hydrate substrates 2a, 2b and 3, previously used,1 were
employed as well. Since the precise procedure for K2CO3
addition was not reported, we used two distinct protocols,
obtaining slightly different results:
Two very interesting papers, recently, claimed the possi-
bility of hydrolysing cyclic acetals and ketals under basic1
or mildly alkaline2 conditions by ceric ammonium nitrate
,CAN) in an acetonitrile/water mixture. Due to the
importance of ketals and acetals as alkaline-stable protec-
tive groups in carbohydrate chemistry, the possibility of
cleaving them in basic or mildly alkaline conditions,
where the glycoside linkage is stable, would have opened
new perspectives in the protection±deprotection strategy in
carbohydrate synthesis. Therefore we have undertaken an
in-depth investigation on the reaction conditions in order to
reproduce the results reported,1,2 using as substrate 2,10:4,6-
di-O-isopropylidenesucrose 1 which contains both glyco-
side and ketal linkages.
,a) addition at 708C, under argon, of an aqueous solution
of CAN to the MeCN sample solution containing the
required amount of solid carbonate;
,b) addition at 708C, under argon, of a cloudy solution of
carbonate and CAN to the MeCN sample solution.
2. Results and discussion
When 1 was treated with CAN in a 1:2 MeCN/H2O mixture
at 708C according to the described conditions,1 both the
cyclic ketals and the glycosidic bond were cleaved very
easily, but the solution was strongly acidic ,about pH 0.2)
in contrast with the claimed slightly acidic conditions.1 In
the presence of K2CO3, 1 was completely unaltered or the
cyclic ketal opening occurred to an extent of 80% according
to the mode of the carbonate addition ,described later).
Keywords: ceric ammonium nitrate; hydrolysis; cyclic acetals and ketals;
glycoside linkages; ceric ammonium nitrate.
Corresponding author. Tel.: 139-081674147; fax: 139-081674393;
e-mail: parrilli@unina.it
Under the ®rst conditions, ketal cleavage occurred to an
extent of 16%, in the case of 3, and only in trace amounts
for 2a and 2b while for 1 the reaction proceeded to 80% with
p
0040±4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020,01)01006-7

130
E. Manzo et al. / Tetrahedron 58 /2002) 129±133
the opening of only the cyclic ketals. In no case did the
yields change on increasing the reaction times. Under the
second conditions all compounds remained completely
unaltered. We suggest that the reported1 ketal opening of
2a, 2b and 3 in the presence of carbonate could depend on
the dissolution rate of added solid carbonate which is slower
than the rate of ketal opening, thereby carbonate is unable to
neutralise the acid present, due to the hydrolysis of the
Ce,IV) ion, before ketal cleavage has partially occurred.
This is in agreement with the fact that yields did not change
with reaction time. To support this hypothesis we measured
the time necessary to neutralise a 1:2 MeCN/H2O CAN
solution, whose concentration was identical to that utilised
in the reactions on 2a, 2b and 3, adding solid carbonate at
708C. The neutralisation took more than 1 min, a time
comparable with the rate of ketal cleavage.1 All the above
results indicated that cyclic ketal/acetal cleavage is deter-
mined mainly by protic acid environments and that under
neutral or basic conditions these processes do not occur.
This is supported by the fact that in no case the precipitation
of cerium hydroxide, which occurs under neutral condi-
tions,3 was reported. Accordingly, when we worked under
carefully checked neutral or basic conditions the cerium
hydroxide precipitated. At this stage one of authors ,E.
HNO3 at same pH, but without CAN, 1 was unaltered after
3 h.
When this work was ®nished we were acquainted with
another paper on the ability of CAN ,1.2 equiv.) to open
acetals in aqueous methanol at room temperature in the
presence of excess NaHCO3 or Na2CO3.5 Under the
described conditions we have also found that the deprotec-
tion of 2a occurred to 85% after 30min but the pH of the
solution was strongly acid and no clouding of the solution
was observed. In addition, in the presence of solid carbonate
the reaction proceeded to only 32% after 10min and this
yield did not change after 30min, suggesting that the
reaction occurred only for the time necessary to dissolve a
suf®cient amount of carbonate to neutralise the acidity due
to the Ce,IV) hydrolysis. Accordingly, when the carbonate
was added with CAN as solution, the deprotection reaction
did not proceed.
In order to evaluate the solvent effect we performed some
reactions replacing MeCN with acetone or THF. Acetone
gave similar yields and product distribution as MeCN
whereas a slower reaction occurred in THF, due to its cation
complexing ability which decreases the Lewis acidity of
Ce,IV).
Á
M.) joined Marko's group in order to compare our reaction
conditions with theirs. First of all the pH of the solution was
measured when a catalytic amount of CAN ,4%)2 in 1:1
MeCN/H2O was dissolved at 698C; this value was 1.68.
Under this set of conditions 3 was deprotected within
10min, as happened when the same pH value was obtained
with HNO3 without CAN. This showed no peculiar effect of
CAN suggesting that, under these conditions, in both cases
there is a protic acid hydrolysis. Then it was con®rmed that
adding to a solid mixture of substrate 3, K2CO3 and CAN
,2.5 equiv.) a 1:2 MeCN/H2O mixture at 708C the yield of
deprotected ketone ranged from 10to 30%, both after 1 and
24 h, according to the rapidity of carbonate dissolution. On
the other hand, if an aqueous solution of K2CO3 and CAN
,2.5 equiv.) ,in fact it is a suspension because a part of the
Ce,IV) precipitates as hydroxide) was added to a solution of
The use of acetone, which can improve the sample solu-
bility, was tested on some carbohydrate acetonides. Table
1 shows the different reaction courses with and without
Ce,IV), both performed at pH 4.4. From these data is
con®rmed the role of Lewis acid for Ce,IV) at pH 4.4 and
its different reactivity with respect to the proton. In particu-
lar, whereas the opening of terminal isopropylidenes occurs
to the same extent both with the Ce,IV) ion and the proton
,entries 2 and 4), the cleavage of 4,6 isopropylidenes is
clearly favoured by Ce,IV) ,entries 1 and 3), which is also
the only one able to cleave the 3,4-acetonide of galactose
,entry 5) and the 4,6-benzylidene acetal ,entry 6). Since in
these last cases the reaction yields were rather low even for
longer reaction times we thought that the concentration of
Ce,IV) in solution could decrease owing to its precipitation
as hydroxide. As a matter of fact the clouding of the reaction
mixture increased with time. In order to con®rm this
hypothesis we tested the effectiveness of CAN solution at
pH 4.4 with time by performing reactions with solutions,
which were kept at room temperature for few hours before
their use. Using a 3 h old CAN solution, whose pH was
unaltered, 3 reacted to only 8% after 1 h instead of 80±
85%, obtained with a freshly-prepared solution, showing a
drastic decrease reactivity in agreement with the low
concentration of Ce ion remained in solution. This observa-
tion allowed us to increase strongly the reaction yields by
adding portions of CAN solution every 2 h ,Table 1).
3 in MeCN at 708C no reaction occurred. As for the reaction
Á
solution was more dilute than ours: using same volumes we
2
with borate/HCl buffer we discovered that Marko's buffer
Á
reached pHˆ8, with clouding of the medium, while Marko
obtained a clear solution but at pH 2. In order to see if
Ce,IV) performs a Lewis acid function as well, we and
Marko's group together investigated the course of reactions
accomplished at equal pH but with and without CAN. To
this end the pH value of the solution was increased by
adding pyridine until the solution began to become cloudy
due to cerium hydroxide precipitation. This happens at pH
4.4. Under these conditions 3 at 688C with 3% of CAN in
1:1 MeCN/H2O was deprotected to 80±85% after 1 h and
completely after 2 h. Interestingly, at the same pH obtained
with HNO3 without CAN, Compound 3 was deprotected to
only 10% after 3 h at 688C. These results suggested that the
role of Ce,IV) as Lewis acid works but only when the proton
concentration is low. Under these conditions CAN ,3%)
showed a selectivity due to steric hindrance. Actually both
2a and its reduced 2b derivative were unaltered at pH 4.4 at
688C. Under these conditions 1 gave quantitatively sucrose,
after 1 h, showing that the ketal groups were cleaved while
the glycoside linkage was unaltered. On the other hand with
3. Conclusion
In conclusion, the possibility to achieve the ketal cleavage
with CAN in 1:2 MeCN/H2O or in 10% aq. methanol in the
presence of carbonate is rejected. However, CAN in 1:1
MeCN/H2O is able to cleave selectively sterically
unhindered cyclic ketals under mildly acidic conditions
,pH 4.4), exploiting its strong Lewis acid character. In

E. Manzo et al. / Tetrahedron 58 /2002) 129±133
131
Table 1. Yields of reactions at pH 4.4 with CAN and without CAN ,in parentheses). The yields were determined by GC-MS analysis of the acetylated reaction
mixtures. All reactions were accomplished in 1:1 acetone/water mixture at 688C
Entry
Substrate
Time ,h)
2,3-Acetonide of Manp
Man
Glc
Gal
a-d-GlcOMe
1
1
4
82 ,10)
72 ,12)
0 ,0)
25 ,1)
2
3
1
4
30,32)
55 ,55)
0,1)
3 ,4)
1
4
79 ±
85 ,10)
4
5
1
4
35 ±
61 ,50)
1
4
12 ,0
15 ,0
60,0)
5a
6
1
4
10,0)
34 ,0)
93 ,0)
61 ,0)
5a
5b
a
Adding portions of CAN ,3%) every 2 h.
Adding portions of CAN ,1%) every 2 h.
b
addition, these conditions allow hydrolysis of cyclic ketals
leaving glycosidic bonds unaltered and show a different
reactivity from that of the proton at the same pH. Work is
in progress to test the selectivity of these conditions for the
deprotection of other hydroxyl protective groups.
,5 ml) by reduction with NaBH4 ,500 mg) under stirring
for 3 h at room temperature. The reaction was quenched
by adding acetic acid ,500 ml), the solvent was removed
by N2 stream and the residue was treated twice with 1:9
acetic acid/methanol mixture, and then ®ve times with
only methanol. Compound 2b was identi®ed by GC±MS
of its acetylated derivative. Compound 3 was prepared by
dissolving methyl nonyl ketone ,Fluka, 500 mg, 2.9 mmol)
in dry DMF ,Fluka, 1.5 ml). Ethylenglycol ,Fluka, 245 ml,
4.4 mmol) and p-toluenesulfonic acid ,Aldrich, 10mg)
were added. The reaction ran at 508C overnight, under
stirring. After addition of 0.25 M NH4HCO3, the reaction
mixture was extracted with hexane, which was dried with
anhydrous Na2SO4 and evaporated under vacuum. The
product 3 was identi®ed by the following diagnostic
GLC±MS fragments: m/z 87 ,100%), [CH3CO2C2H4]1
and m/z 199 ,17%), [C10H19CO2C2H4]1. GLC±MS spectra
were measured with a Hewlett±Packard 5890instrument.
4. Experimental
4.1. General
CAN, sucrose and 2a were purchased from Fluka, inositol
from Carlo Erba, buffer borate/hydrochloric acid from
Merck, pyridine from Romil, MeCN for HPLC from Lab-
Scan. All products were used without further puri®cation.
Compound 1 and the protected monosaccharides in Table 1
were prepared according to described procedures.4,6
Compound 2b was obtained from 2a ,300 mg) in EtOH

132
E. Manzo et al. / Tetrahedron 58 /2002) 129±133
Analysis condition: SPB-5 capillary column ,SUPELCO,
30m £0.25 mm i.d., ¯ow rate 1 ml/min, He as carrier gas),
using the following temperature programmes: for 1: 1508C
for 5 min, 150!3308C at 38C/min, 3308 for 0min; for 2a
and 2b: 808C for 2 min, 80!2408C at 88C/min, 2408 for
3 min; for 3: 1508C for 3 min, 150!2808C at 108C/min,
2808 for 10min pH-meter ,Orion) equipped with electrode
by Hamilton.
708C, in argon atmosphere, and a solution of CAN ,33 mg,
0.0602 mmol) in H2O ,100 ml) was added. The reaction
progress, measured by GC±MS, showed that 1 gave sucrose
,80%) and starting material ,20%) after 10 min. The yield
did not rise on increasing the reaction time up to 24 h.
Compound 2a or 2b ,22 mg, 0.141 mmol) and solid
K2CO3 ,199 mg, 1.421 mmol) were treated at 708C, under
argon, with MeCN ,0.3 ml) and a solution of CAN ,195 mg,
0.356 mmol) in H2O ,0.6 ml) was added. After 1 h, the
GC±MS analysis revealed only traces of deprotected
product, while the main product was starting material. The
distribution of products did not change after 24 h.
Compound 3 ,60mg, 0.280mmol) and solid K 2CO3
,398 mg, 2.843 mmol) were treated at 708C with MeCN
,0.6 ml), under argon, and a solution of CAN ,390 mg,
0.712 mmol) in H2O ,1.2 ml) was added. After usual
work-up the yield of deprotected ketal accounted for only
16% both after 1 and 24 h, the remaining product being
starting material.
4.2. Reaction of compounds 1, 2a, 2b and 3 with ceric
ammonium nitrate $2.5 equiv.) in MeCN/H2O at 708C
4.2.1. Measure of reaction progress. The progress of reac-
tion for 1 and 2b was followed by GC±MS of acetylated
derivative of crude reaction. After stopping the reaction
with pyridine, the mixture was lyophilised and, after drying,
a pyridine solution of inositol ,2 mg/ml) was added, as
internal standard. The acetylation was performed with
acetic anhydride at room temperature under stirring over-
night. After evaporation under N2 stream, the residue was
treated with H2O/CHCl3 mixture. The organic phase was
injected into GC±MS. For 2a, the reaction mixture after
quenching with pyridine was extracted with CHCl3, the
organic phase was evaporated and dissolved in a solution
of acetylated inositol in CHCl3 ,2 mg/ml). For 3, the
reaction mixture was cooled at room temperature and
0.25 M NH4CO3 solution and hexane were added. The
organic phase was dried on anhydrous Na2SO4 and evapo-
rated under vacuum. The solid was dissolved in CHCl3
solution of inositol acetate ,2 mg/ml) and analysed by
GC±MS.
4.2.4.2. Reactions with K2CO3, using cloudy solution
of salt. Compound 1 ,10 mg, 0.0237 mmol) was dissolved
in MeCN ,50 ml) at 708C, and a suspension of CAN ,33 mg,
0.0602 mmol) and K2CO3 ,34 mg, 0.243 mmol in H2O
,100 ml) was added, in argon atmosphere. After 24 h, the
GC±MS analysis showed only unaltered starting material.
Compound 2a or 2b ,22 mg, 0.141 mmol) was dissolved in
MeCN ,0.3 ml) at 708C, under argon, and a suspension of
CAN ,195 mg, 0.356 mmol) and K2CO3 ,199 mg,
1.421 mmol) in H2O ,0.6 ml) were added. After 24 h the
GC±MS analysis showed only starting material.
Compound 3 ,60mg, 0.280mmol) was dissolved in
MeCN ,0.6 ml) at 708C under argon, and a suspension of
4.2.2. Conditions as reported.1 Compound 1 ,10mg,
0.0237 mmol) was dissolved in MeCN ,50 ml) at 708C in
argon atmosphere, and a solution of CAN ,33 mg,
0.0602 mmol) in H2O ,100 ml) was added. The pH of the
solution was 0.2. Then the mixture was stirred at 708C and
after 2 h the hydrolysis of 1 in glucose and fructose was
complete. Compound 2a ,22 mg, 0.141 mmol) was
dissolved in MeCN ,0.3 ml) at 708C under argon and a
solution of CAN ,195 mg, 0.356 mmol) in H2O ,0.6 ml)
was added. After 30min the ketal deprotection was
complete.
CAN ,390mg, 0.712 mmol) and K CO3 ,398 mg,
2
2.843 mmol) in H2O ,1.2 ml) was added. After 24 h, only
starting material was present in the crude reaction mixture.
4.3. Measurements of pH variation with temperature
and in presence of solid K2CO3
Solid CAN ,3.9 g) was dissolved in H2O ,12 ml) and MeCN
,6 ml). The pH measured at room temperature was 0.2.
When the mixture was heated to 708C its pH value
decreased to 20.76. Adding, under this conditions, solid
K2CO3 ,3.98 g) the pH was 6.2 after 1 min and 8.42 after
4 min.
For compound 2b, the same amount and conditions of 2a
were used, obtaining the same results. Compound 3 ,60mg,
0.280 mmol) was dissolved in MeCN ,0.6 ml) at 708C,
4.4. Reactions of 3 with ceric ammonium nitrate $4%) in
MeCN/H2O
under argon, and
0.712 mmol) in H2O ,1.2 ml) was added. The ketal
a
solution of CAN ,390mg,
deprotection was complete after 10min.
To a solution of 3 ,30mg, 0.140mmol) in 1:1 MeCN/H 2O
mixture ,0.9 ml), CAN ,3 mg, 5.47£1023 mmol) was added
giving a clear solution at pH 1.68. After 10min 3 was
completely deprotected. The reaction was repeated under
the same condition but the pH was achieved by adding
conc. HNO3 without using CAN. Also in this case 3 was
completely deproteced after 10min.
4.2.3. Reaction conditions without ceric ammonium
nitrate. All the above reaction were repeated under identi-
cal condition, but the pH was achieved by adding conc.
HNO3 and not by CAN hydrolysis. The same results as
described earlier were obtained.
4.2.4. Reactions with carbonate according to our proce-
dures
4.5. Reactions of 1, 2a, 2b and 3 with ceric ammonium
nitrate in buffered MeCN/H2O as decribed2
4.2.4.1. Reactions with K2CO3, using solid salt.
Compound 1 ,10 mg, 0.0237 mmol) and solid K2CO3
,34 mg, 0.243 mmol) were treated with MeCN ,50 ml) at
Compound 1 ,24 mg, 0.0569 mmol) was dissolved in
MeCN ,0.175 ml) and in a buffer solution of borate/HCl

E. Manzo et al. / Tetrahedron 58 /2002) 129±133
133
,0.175 ml) at pH 8. CAN ,1.2 mg, 2.2£1023 mmol) was
added at room temperature and the reaction mixture,
which became cloudy keeping its pH value at 8, was heated
to 608C, under stirring. After 24 h, a GC±MS analysis
revealed the presence of only unaltered starting material.
Compound 2a or 2b ,44 mg, 0.282 mmol) was dissolved
in MeCN ,0.9 ml) and in a buffer solution of borate/HCl
,0.9 ml) at pH 8. CAN ,6 mg, 1.09£1022 mmol) was
added at room temperature and the reaction mixture,
which became cloudy keeping its pH value at 8, was heated
to 608C, under stirring. After 24 h, a GC±MS analysis
revealed the presence of only unaltered starting material.
Identical procedure was employed for compound 3
,30mg, 0.140mmol). This was dissolved in MeCN
,0.45 ml) and in the buffer solution ,0.45 ml) at pH 8.
CAN ,3 mg, 5.47£1023 mmol) was added as above, giving
a turbid solution. After 24 h, only starting material was
present in the crude reaction mixture.
under stirring. The reaction was quenched with pyridine
and the mixture was diluted with water and extracted with
dichloromethane.
4.7. Reaction at pH 4.4 with and without ceric
ammonium nitrate $3%)
At beginning two solutions were prepared: the ®rst, A was
obtained dissolving CAN ,105 mg, 0.192 mmol) in water
,19 ml) and adding pyridine up to the beginning of precipi-
tation of cerium hydroxide. Under these conditions the pH
solution is 4.4. The second, B was obtained adding conc.
HNO3 to water until to achieve a solution at pH 4.4. Solu-
tions of 1 ,70mg, 0.166 mmol), or 2a ,26 mg, 0.167 mmol),
or 2b ,26 mg, 0.167 mmol) or 3 ,26 mg, 0.167 mmol) in 1:1
MeCN/A ,1 ml) were prepared and held with stirring at
688C. After usual work-up, 1 gave after 1 h quantitatively
sucrose, which was unaltered until 4 h. 2a and 2b were
unreacted after 4 h. 3 was deprotected for 80±85% after
1 h and completely after 2 h. Performing the above reactions
under same conditions but using solution B, that is without
CAN, 1, 2a and 2b gave, after 3 h, starting material, while 3
was deprotected only for 10%.
4.6. Reaction of 2a with ceric ammonium nitrate in 10%
aq. methanol5
Compound 2a ,16 mg, 0.1 mmol) was dissolved in 10% aq.
methanol ,0.2 ml). A solution of CAN ,66 mg, 0.12 mmol)
in 10% aq. methanol ,0.8 ml) was then added dropwise
under stirring. The reaction was quenched with pyridine
and the mixture was diluted with water and extracted with
dichloromethane.
An analogous procedure, except for the use of acetone in
place of acetonitrile, was followed in the deprotection of
saccharidic substrates on Table 1.
References
4.6.1. Reaction of 2a with ceric ammonium nitrate and
solid sodium carbonate in 10% aq. methanol.5 Compound
2a ,16 mg, 0.1 mmol) and sodium carbonate ,127 mg,
1.2 mmol) were suspended in 10% aq. methanol ,0.2 ml).
A solution of CAN ,66 mg, 0.12 mmol) in 10% aq. metha-
nol ,0.8 ml) was then added dropwise under stirring. The
reaction was quenched with pyridine and the mixture was
diluted with water and extracted with dichloromethane.
1. Ates, A.; Gautier, A.; Bern, L.; Plancher, J.-M.; Quesnel, Y.;
Â
Marko, I. Tetrahedron Lett. 1999, 40, 1799±1802.
Â
2. Marko, I.; Ates, A.; Gautier, A.; Bern, L.; Plancher, J.-M.;
Quesnel, Y.; Vanherck, J.-C. Angew. Chem., Int. Ed. Engl.
1999, 38, 3207±3209.
3. Kajimura, A.; Sumaoka, J.; Komiyama, M. Carbohydr. Res.
1998, 309, 345±351.
4. Manzo, E.; Barone, G.; Parrilli, M. Synlett 2000, 887±889.
5. Nair, V.; Nair, G. L.; Balagopal, L.; Rajan, R. Ind. J. Chem.
1999, 38B, 1234±1236.
4.6.2. Reaction of 2a with a solution of ceric ammonium
nitrate and sodium carbonate in 10% aq. methanol.5
Compound 2a ,16 mg, 0.1 mmol) was dissolved in 10%
aq. methanol ,0.2 ml). A suspension of CAN ,66 mg,
0.12 mmol) and sodium carbonate ,127 mg, 1.2 mmol) in
10% aq. methanol ,0.8 ml) was then added dropwise
6. Hanessian, S. Preparative Carbohydrate Chemistry; Marcel
Dekker: New York, 1997.