Use of methanesulfonic acid in the reductive ring-opening of O-benzylidene acetals

Article abstract of DOI:10.1016/j.carres.2006.10.009

Methanesulfonic acid was shown to be an efficient and convenient substitute for ethereal HCl in reductive 4,6-O-benzylidene acetal ring-opening reaction with sodium cyanoborohydride in THF. Normal regioselectivity was observed, the 6-O-benzyl ethers with

Full text of DOI:10.1016/j.carres.2006.10.009

Carbohydrate Research 342 (2007) 627–630
Note
Use of methanesulfonic acid in the reductive ring-opening
of O-benzylidene acetals
Alexander I. Zinin, Nelly N. Malysheva, Anna M. Shpirt,
Vladimir I. Torgov and Leonid O. Kononov*
N. D. Zelinsky Institute of Organic Chemistry of the Russian Academy of Sciences, Leninsky prosp., 47, 119991 Moscow,
Russian Federation
Received 15 June 2006; received in revised form 6 October 2006; accepted 6 October 2006
Available online 13 October 2006
Dedicated to the memory of Professor Nikolay K. Kochetkov
Abstract—Methanesulfonic acid was shown to be an efficient and convenient substitute for ethereal HCl in reductive 4,6-O-benzyl-
idene acetal ring-opening reaction with sodium cyanoborohydride in THF. Normal regioselectivity was observed, the 6-O-benzyl
ethers with free 4-OH group being the major products of the reaction.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Keywords: Benzylidene acetal; Reduction; Sodium cyanoborohydride; Methanesulfonic acid; Regioselectivity
Chemical synthesis of oligosaccharides currently relies
on the use of partially protected mono- and oligosaccha-
ride building blocks. The reductive opening of 4,6-O-
benzylidene acetals (and other benzylidene acetals with
1,3-dioxane ring) with sodium cyanoborohydride in
THF in the presence of HCl in Et2O (the so-called ‘Gar-
egg opening’)1–3 is often used for large-scale preparation
of 6-O-benzyl derivatives due to a low cost of the re-
agents. The main drawback of this procedure is associ-
ated with the necessity of preparation of soln of HCl
in Et2O by bubbling gaseous HCl (taken from a gas cyl-
inder or generated from NaCl) through cold Et2O under
anhyd conditions. This is a tedious and skill-demanding
procedure due to the necessity of ensuring strictly anhyd
conditions. Although a soln of HCl in Et2O is now com-
mercially available, the problem of keeping it anhyd
remains relevant. For this reason, other experimentally
less demanding approaches became popular. Particu-
larly worthy of note are the Et3SiH–TFA (Ref. 4), BH3Æ
Me2NH–BF3ÆOEt2 (Ref. 5) and Me3NÆBH3–AlCl3 (Ref.
6) systems. Recently, a more efficient variant of the lat-
ter method was described7 that emphasizes the necessity
of performing the reductive opening of benzylidene ace-
tals with Me3NÆBH3 in the presence of water (3:1 AlCl3–
water). Obviously, a strong acid is required3 to effect the
reductive opening of benzylidene acetals, hydrogen chlo-
ride in diethyl ether being used mainly for historical rea-
sons. We reasoned that another protic acid with an
appropriate pKa value, which is soluble in THF and
can be prepared anhyd, would be a possible alternative
for ethereal HCl in this reaction. THF is the solvent of
choice due to its ability to dissolve a wide range of com-
pounds including salts. Thus, for example, the presence
of several amide groups in the molecule of a benzylidene
derivative makes the substance so polar that it is usually
difficult to use other ether-type solvents such as diethyl
ether or 1,4-dioxane for the reaction due to poor sub-
strate solubility in these solvents. The use of THF allows
one to minimize the volume of solvent used, a feature
very useful on a large scale. A report concerning
the use of trifluoromethanesulfonic acid (TfOH) in
THF as the proton source in this reaction has been
published but no experimental details were provided.8
The major drawback of TfOH is its extremely high cost.
*
Corresponding author. Tel.: +7 495 9383610; fax: +7 495 1355328;
e-mail: kononov@ioc.ac.ru
0008-6215/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2006.10.009

628
A. I. Zinin et al. / Carbohydrate Research 342 (2007) 627–630
In addition, exceptionally high acidity of TfOH may
pose problems in some cases. The weaker trifluoroacetic
acid does not promote the reductive ring-opening of 4-
methoxybenzylidene acetals with NaBH3CN in DMF,9
but fails to promote the respective opening of benzyli-
dene acetals. Obviously, another acid, less acidic than
TfOH and more acidic than TFA, is required. We
assumed that the reasonably cheap methanesulfonic
acid (MsOH, pKa 1.92)10 could be a good candidate.
Indeed, we have found that MsOH can promote this
reaction leading regioselectively to the expected prod-
ucts with free hydroxy group at C-4, which were isolated
in high yields. The conditions of the reaction
after charring the TLC plate with acid (8:0.7 CHCl3–
MeOH, Rf 0.47) and can be separated by silica gel col-
umn chromatography (they were identified from the
NMR spectra: dH 1.50–1.65 (m, 20H, CCH2C), 3.30–
3.50 (m, 21H, CH2OCH2, CH2OH); dC 26.4 (9.5C,
CCH2C), 60.8 (1C, CH2OH), 70.5 (10C, CH2CH2-
OCH2)). Therefore, in order to minimize THF polymer-
ization and related side processes, which dramatically
decrease the yield of the target product, the reaction
should be quenched with a base (Et3N or aqueous NaH-
CO3) at least within 1–2 h after complete consumption
of the starting benzylidene acetal (at this point the reac-
tion mixture can safely be left overnight at room temper-
ature without affecting the yield of the target product).
In fact, this side process does not pose a major problem
since in most cases small amounts of THF oligomers
could be easily separated from the target 6-O-benzyl
ethers by chromatography and crystallization. When
the reaction is performed on a large scale, the additional
advantages of using MsOH rather than ethereal HCl
become evident. Thus, the addition of relatively small
volumes of MsOH (or its soln in THF) is necessary to
effect benzylidene acetal ring opening, while much larger
volumes of a soln of HCl in Et2O are needed to create
the required acidity of the medium.
A direct comparison of the two approaches (based on
the use of MsOH and HCl) was made using 4-methoxy-
phenyl glycoside 1a11 and 3-(trifluoroacetamido)propyl
glycoside 1b12 as representative examples. Both reac-
tions were complete within 1–3 h provided the required
acidity of the reaction medium was maintained and an
excess of NaBH3CN was present. The isolated yields
of 6-O-benzyl ethers 2a,b were comparable to those re-
ported.11,12 In one case, when chromatographic mobility
of the desired product 2a was similar to that of THF
oligomers and its direct purification was difficult, the
crude reaction mixture was acetylated (Ac2O/Py) and
the product separated from the by-products followed
by O-deacetylation to 2a.
˚
(NaBH3CN, MsOH, MS 4 A, THF, pH 2–3, ꢀ20 ꢁC)
were similar to the traditional ones3 and can be expected
to be compatible with a variety of protective groups (see
˚
Scheme 1). Molecular sieves of any type (3 or 4 A, beads
or powder) were successfully used in the reaction pro-
moted by MsOH. As in the Garegg opening, it is impor-
tant to control the acidity of the medium. The best yields
were achieved when the acidity of the reaction mixture
was near pH 2.5–3 (measured with a high resolution
indicator pH paper (Acilit, E. Merck)). Very low acidity
of the reaction medium (pH > 3) does not allow the
reaction to proceed, while very high acidity (pH 2 or
lower) would lead to the formation of by-products and
partial acetal cleavage if traces of moisture were present
in the reaction mixture. The actual amount of acid re-
quired varies with the amount and type of molecular
sieves used. We have also found that although neat
MsOH could be added directly to the reaction mixture,
in most cases it is more convenient to use a freshly pre-
pared 0.5 M soln of MsOH in THF and add it in 1 equiv
portions until gas evolution ceases, then add 0.5 equiv
more to ensure the correct pH. On a large scale, most
of the required MsOH can be added neat, and then a
2 M soln of MsOH in THF is used for fine-tuning the
acidity of the reaction medium. Although MsOH does
catalyze ring-opening polymerization of THF, this side
process does not prevent from using THF as the solvent
for the reductive opening of benzylidene acetals. The
amount of THF oligomers was especially large when
the reaction mixture was left overnight at room temper-
ature. In this case, the THF oligomers are detectable
The structures of products 2a,b were inferred from
1
their H and 13C NMR spectra, which were identical
to those reported.11,12 The location of the benzyl group
at O-6 in 2a,b was deduced from the low-field position of
C-6 (d ꢀ 70 ppm—note the incorrect assignment of 13C
OBn
Ph
O
a
O
O
O
HO
R1O
OR4
OR4
R1O
NR2R3
NR2R3
1a—b
2a—b
a: R1 = Bn, R2 = R3 = Pht, R4 = 4-MeOC6H4
b: R1 = Ac, R2 = H, R3 = Ac, R4 = (CH2)3NHTFA
˚
Scheme 1. Reagents and conditions: (a) NaBH3CN, CH3SO3H, MS 4 A, THF, 20 ꢁC, 1–3 h.

A. I. Zinin et al. / Carbohydrate Research 342 (2007) 627–630
629
NMR spectrum in Ref. 11 and its absence in Ref. 12) in
comparison with that in derivatives with unprotected 6-
OH group (d 60–62 ppm). The 1H NMR spectrum of 2b
contained showed a signal for H-3 at low field (dH
5.03 ppm) and a signal for H-4 at high field (dH
3.68 ppm), which corroborates the acetylation of O-3
and the absence of acetyl group at O-4.
In conclusion, the suggested procedure for reductive
4,6-O-benzylidene acetal ring opening in the presence
of NaBH3CN, which makes use of MsOH instead of
HCl, is a convenient alternative for the existing methods
and is especially useful on a large scale since it avoids the
need for preparation of large volumes of ethereal HCl.
Due to its experimental simplicity and low cost of
reagents, this procedure may find wide applications in
academic research and industry.
(0.6 mL, 9.25 mmol) were added (pH 2.5–3) and after
1 h the reaction was complete. Triethylamine (6 mL,
43.17 mmol) and water (8 mL) were added and the reac-
tion mixture was filtered through a Celite pad. The sol-
ids were washed with MeOH (100 mL) and the filtrate
concentrated. The residue was dissolved in CHCl3
(200 mL), washed with water (3 · 100 mL), dried with
Na2SO4 and concentrated. The residue (5.4 g) was
applied on a silica gel column and eluted with
1:9 ! 1:4 EtOAc–hexane to give a fraction (3.59 g),
which contained 2a (1:4 EtOAc–benzene, Rf 0.56; 1:1
EtOAc–hexane, Rf 0.26) contaminated with THF oligo-
mers. This fraction was dissolved in anhyd Py (8 mL)
and treated with Ac2O (8 mL) for 18 h at room temper-
ature. Then MeOH (10 mL) was added and after 1 h the
mixture was concentrated. The residue was dissolved in
CHCl3 (200 mL), washed with water (2 · 100 mL), dried
with Na2SO4 and concentrated to a residue which was
purified by crystallization from Et2O–hexanes to give
the pure 4-O-acetylated derivative (2.60 g, 75%, 1:4
EtOAc–benzene Rf 0.75, 1:1 EtOAc–hexane Rf 0.56).
This material was dried under diminished pressure and
dissolved in anhyd THF (45 mL) under argon. To the
resulting soln, a 1 N soln of magnesium methoxide in
MeOH (45 mL) was added and the mixture was kept
at room temperature for 2.5 h. The reaction mixture
was neutralized with AcOH and concentrated. The
residue was dissolved in CHCl3 (50 mL), concentrated
to give the residue (2.42 g), which was redissolved in
benzene and purified by silica gel chromatography
(1:49 ! 1:9 EtOAc–benzene) to give pure compound
1. Experimental
1.1. General methods
The reactions were performed in argon atmosphere
using commercial reagents (Aldrich and Fluka) and
anhyd (where appropriate) solvents purified according
to standard procedures. Column chromatography was
performed on Silica Gel 60 (40–63 lm, E. Merck). Thin-
layer chromatography was carried out on plates with
Silica Gel 60 on glass or on aluminum foil (E. Merck).
The spots of compounds containing carbohydrates were
visualized with 1:10 85% H3PO4–96% EtOH with subse-
25
1
quent heating (150 ꢁC). The H and 13C NMR spectra
2a (2.24 g, 69%). ½aꢁD +56.5 (c 1.0, CHCl3); 13C NMR
were recorded on a Bruker AC-200 instrument (200.13
(CDCl3): d 55.2 (OMe); 55.5 (C-2); 70.3 (C-6); 73.7
(OCH2Ph); 73.7, 73.9 (C-4, C-5); 74.4 (OCH2Ph); 78.5
(C-3); 97.5 (C-1); 114.3, 118.6, 123.3, 127.4, 127.7,
127.9, 128.1, 128.4, 133.8, (arom. CH); 131.5, 137.6,
138.0, 150.7, 155.3 (arom. C); 167.7 (CO).
1
and 50.32 MHz, respectively). H NMR chemical shifts
are referred to the residual signal of CHCl3 (dH 7.27);
13C NMR chemical shifts, to the CDCl3 signal (dC
77.0). Signal assignments in 13C NMR spectra was based
on DEPT-135 experiments. Optical rotation was
measured on a JASCO DIP-360 polarimeter at 20–25 ꢁC.
1.3. 3-(Trifluoroacetamido)propyl 2-acetamido-3-O-acet-
yl-6-O-benzyl-2-deoxy-b-D-glucopyranoside (2b)
1.2. 4-Methoxyphenyl 3,6-di-O-benzyl-2-deoxy-2-phthal-
imido-b-D-glucopyranoside (2a)
To a mixture of solid 4,6-O-benzylidene derivative 1b12
(6.705 g, 13.29 mmol) and solid NaBH3CN (7.00 g,
111.39 mmol), anhyd THF (200 mL) was added under
argon followed by freshly activated powdered molecular
To a mixture of solid 4,6-O-benzylidene derivative 1a11
(3.25 g, 5.47 mmol) and solid NaBH3CN (1.38 g,
21.96 mmol) under argon, anhyd THF (80 mL) was
added followed by freshly activated powdered molecular
˚
sieves 4 A (9.6 g). The resulting suspension was magnet-
ically stirred at room temperature for 30 min, and a 2 M
soln of MsOH in THF was then added dropwise (using a
syringe) at such a rate that the gas evolution was steady
and vigorous stirring was possible (the reaction mixture
gradually becomes thick). The addition of MsOH soln
was continued until pH 2.5–3 was reached (1 h, 80 mL
of 2 M MsOH soln was added), and TLC showed the
presence of the product in the reaction mixture (8:0.7
CHCl3–MeOH, Rf 0.33 (1b), 0.20 (2b)) (if the product
is not detected at this point, an additional amount of
˚
sieves 4 A (4 g). The resulting suspension was stirred
(mechanical stirrer) at room temperature for 30 min,
and a soln of MsOH (1.42 mL, 21.88 mmol) in THF
(10 mL) was then added dropwise within 10 min. The
reaction mixture became thick and gas evolution began.
After 1 h of vigorous stirring (pH 2.5–3), only a small
amount of the starting acetal 1a remained unreacted
(1:4 EtOAc–benzene, Rf 0.74 (1a), 0.56 (2a)). Additional
amounts of NaBH3CN (0.7 g, 11.14 mmol) and MsOH

630
A. I. Zinin et al. / Carbohydrate Research 342 (2007) 627–630
2 M MsOH soln (5 mL) should be added). Vigorous
stirring was continued at room temperature (20–
25 ꢁC), and after 3 h no starting material was present
in the reaction mixture (if the reaction was incomplete,
more NaBH3CN (1 equiv, 1.00 g, 13.23 mmol) should
be added accompanied by the amount of MsOH soln
necessary to attain pH 2.5–3). The reaction mixture
was cooled in an ice water bath and satd aq NaHCO3
(100 mL) was added (at this point the reaction mixture
can be left overnight at room temperature without
affecting the yield of the target product 2b) and the sol-
ids were filtered off. The filtrate was extracted with
CHCl3 (3 · 120 mL), the organic extract was washed
with satd aq NaHCO3 (3 · 50 mL), dried by filtration
through 1:1 mixture Na2SO4–Celite (solids were washed
with CHCl3) and the filtrate was concentrated to give a
foam (7.3 g), which was dissolved in CHCl3 (35 mL) and
AcOEt (4 mL) and applied on a silica gel column and
then eluted with CHCl3–MeOH mixtures (gradient
0!10% MeOH) to give pure 2b as a white foam
(6.128 g, 91%). Crystallization from AcOEt–Et2O–
hexanes afforded 3.12 g (46%) of 2b as white crystals,
Acknowledgement
This work was supported by the Russian Foundation
for Basic Research (Project No. 06-03-33035).
References
1. Garegg, P. J.; Hultberg, J.; Wallin, S. Carbohydr. Res.
1982, 108, 97–107.
2. Garegg, P. J. Pure Appl. Chem. 1984, 56, 845–858.
3. Garegg, P. J. In Preparative Carbohydrate Chemistry;
Hanessian, S., Ed.; Marcel Dekker: New York, 1997; pp
53–67.
4. DeNinno, M. P.; Etienne, J. B.; Duplantier, K. C.
Tetrahedron Lett. 1995, 36, 669–672.
5. Oikawa, M.; Liu, W.-C.; Nakai, Y.; Koshida, S.; Fukase,
K.; Kusumoto, S. Synlett 1996, 1179–1180.
6. Ek, M.; Garegg, P. J.; Hultberg, J.; Oscarson, S. J.
Carbohydr. Chem. 1983, 2, 305–312.
7. Sherman, A. A.; Mironov, Y. V.; Yudina, O. N.;
Nifantiev, N. E. Carbohydr. Res. 2003, 338, 697–703.
8. Pohl, N. L.; Kiessling, L. L. Tetrahedron Lett. 1997, 38,
6985–6988.
9. Johansson, R.; Samuelsson, B. J. Chem. Soc., Perkin
Trans. 1 1984, 2371–2374.
10. Covington, A. K.; Thompson, R. J. Solutions Chem. 1974,
3, 603–617.
11. Lu, Y.-P.; Su, B.; Li, Q.; Li, H.; Cai, M.-S.; Li, Z.-J. Acta
Chim. Sinica 2002, 60, 360–366.
22
mp 132–133 ꢁC, ½aꢁD ꢂ70.4 (c 1.0, CHCl3), lit.12
25
½aꢁ546 ꢂ93 (c 0.6, CHCl3); 13C NMR (CDCl3): d 20.9
(CH3CO2); 23.1 (CH3CON); 28.4 (CH2CH2CH2); 37.2
(CH2N); 53.8 (C-2); 66.7 (OCH2CH2); 69.9 (C-6); 70.2
(C-4); 73.6 (OCH2Ph); 74.3 (C-5); 75.4 (C-3); 101.2 (C-
1); 127.7, 127.9, 128.4 (Ph); 137.5 (arom. C); 171.0,
172.0 (CO). Anal. Calcd for C22H29F3N2O8: C, 52.15;
H, 5.77; N, 5.53. Found: C, 51.97; H, 5.82; N, 5.62.
12. Bovin, N. V.; Zemlyanukhina, T. V.; Chagiashvili, Ts. N.;
Khorlin, A. Ya. Khim. Prirodn. Soed. 1988, 777–
785 (Chem. Nat. Compd. (Engl. Transl.) 1988, 24, 659–
665).