The influence of boric acid on the acetylation of aldoses: 'One-pot' syntheses of penta-O-acetyl-β-D-glucofuranose and its crystalline propanoyl analogue

Article abstract of DOI:10.1039/b002841j

When glucose and boric acid are heated in acetic acid a soluble compound forms from which, with acetic anhydride and catalytic amounts of sulfuric acid, a mixture consisting of >90% of the glucofuranose per-acetates (α : β ratio 1 : 1.8) is obtained in high yield. In the absence of the sulfuric acid partial acetylation takes place and penta-O-acetyl-β-D-glucofuranose (α: β ratio 1 : 52) is obtainable in good yield by removal of boric acid and completion of the esterification by addition of acetic anhydride and pyridine. A new, crystalline glucofuranose, penta-O-propanoyl-β-D-glucofuranose, is obtained in 58% yield in a 'one-pot' procedure using boric acid. The effects of boric acid on the acid-catalysed acetylation of other aldo-hexoses and -pentoses suggest that the synthesis of furanosyl per-esters could be successful with xylose and idose as well as with glucose.

Full text of DOI:10.1039/b002841j

View Article Online / Journal Homepage / Table of Contents for this issue
The influence of boric acid on the acetylation of aldoses:
‘one-pot’ syntheses of penta-O-acetyl-ꢀ-D-glucofuranose
and its crystalline propanoyl analogue
1
Richard H. Furneaux, Phillip M. Rendle* and Ian M. Sims
Industrial Research Limited, PO Box 31-310, Lower Hutt, New Zealand.
E-mail: p.rendle@irl.cri.nz
Received (in Cambridge, UK) 10th April 2000, Accepted 11th May 2000
Published on the Web 2nd June 2000
When glucose and boric acid are heated in acetic acid a soluble compound forms from which, with acetic
anhydride and catalytic amounts of sulfuric acid, a mixture consisting of >90% of the glucofuranose per-acetates
(α:β ratio 1:1.8) is obtained in high yield. In the absence of the sulfuric acid partial acetylation takes place and
penta-O-acetyl-β--glucofuranose (α:β ratio 1:52) is obtainable in good yield by removal of boric acid and
completion of the esterification by addition of acetic anhydride and pyridine. A new, crystalline glucofuranose,
penta-O-propanoyl-β--glucofuranose, is obtained in 58% yield in a ‘one-pot’ procedure using boric acid.
The effects of boric acid on the acid-catalysed acetylation of other aldo-hexoses and -pentoses suggest that the
synthesis of furanosyl per-esters could be successful with xylose and idose as well as with glucose.
Introduction
Results and discussion
While the direct preparations, from glucose, of either penta-O-
acetyl-α- or β--glucopyranose 2 are straightforward,1 the
furanosyl isomers 3, which have value as starting materials
for making glucofuranosides and their C-, N- and S-linked
analogues, require several steps. A common method is to
constrain the glucose in the furanose form by way of its 1,2-O-
isopropylidene ketal, and in this way the furanosyl per-esters
can be obtained in 41% overall yield after five steps.2 Otherwise
they have been prepared from thiofuranosides made by kinetic-
ally controlled cyclizations of glucose dithioacetals using
mercury() catalysts (five steps from glucose, 31% yield3), and
recently by the acetolysis of octyl glucofuranosides4 which were
in turn made by the coupling of octan-1-ol with glucose in the
presence of FeCl3 (46% overall yield5). In some non-specific
acetolysis reactions the furanosyl acetates are formed together
with the pyranosyl isomers: glucose with acetic anhydride in
the presence of Montmorillonite clays gives products con-
taining 7% of the former,6 whereas the per-acetates formed
on acetolysis of methyl β--glucopyranoside in the presence
of iron() chloride comprise 48% of these furanosyl products.7
We report here that, in the presence of boric acid, glucose can
be converted directly and efficiently to its furanosyl per-esters.
Because of the polyfunctionality and isomeric complexity of
glucose and the common structural ambiguity of boric acid
[B(OH)3, B(OH)4Ϫ], many esters could be formed by their
combination, but no specific products appear to have been
isolated. Nevertheless, reaction of the sugar with the acid (1 mol
equiv.) in acetone in the presence of a strong acid catalyst gives
crystalline 1,2-isopropylidene-α--glucofuranose 3,5-borate,8
and 1H, 11B and 13C NMR9,10 and thermodynamic11 evidence is
consistent with boric acid and boronic acids [RB(OH)2] react-
ing by themselves with the sugar in the furanose form to give
1,2:3,5-bis-cyclic esters. Phenylboronic acid affords the stable,
well known 1,2:3,5-bis-ester.12 With this in mind and with the
information that the above mentioned 1,2-ketal can be used to
make 6-substituted glucoses,8 and that the product derived by
reaction of glucose with boric acid (2 mol equiv.) in acetone can
be converted in three steps to a penta-O-benzoylglucofur-
anose,13 we have investigated the influence of added boric acid
on the acetylation of -glucose and other aldoses.
Glucose and boric acid were heated in acetic acid until a solu-
tion was obtained (1 h). Acetic anhydride and a catalytic
amount of sulfuric acid (conc.) were then added to effect
acetolysis of the initial product (expected to be 1) and give
mixtures of the pyranose (2) and furanose (3) pentaacetates and
the acyclic aldehydohydrate heptaacetate 4 (Scheme 1) which
Scheme 1
were quantified by 1H NMR spectroscopy. Variation in the
proportions of boric acid used caused major changes in the
ratios of products formed, as illustrated in Fig. 1. In its absence,
the pyranoses 2 represented 95% of the pentaacetates formed
(α:β ratio 1.8:1) whereas, with two or more molar equivalents
of boric acid, the products contained 93% of the furanose
isomers 3 (α:β ratio 1:1.8) (Table 1). The latter observation
is consistent with the initial formation of the furanoid borate
intermediate 1 and its transesterification without sugar-ring
DOI: 10.1039/b002841j
J. Chem. Soc., Perkin Trans. 1, 2000, 2011–2014
This journal is © The Royal Society of Chemistry 2000
2011

View Article Online
Table 1 The effect of boric acid on the proportions of per-acetates formed on acetylation of aldosesa
Per-acetates (%)b
Boric acid
Sugar
(mol equiv.)
α-Pyranose
β-Pyranose
α-Furanose
β-Furanose
Aldehydohydrate
Other
α
β
Hexosesc
-Glucose
0
2.2
0
2.2
0
2.2
0
2.2
0
2.2
61
3
80
9
25
13
60
66
34
3
34
3
17
2
53
24
15
16
9
2
33
2
37
5
15
7
4
3
60
1
12
16
39
18
8
0
1
0
3
-Mannose
-Allosed
-Galactose
-Idosed
Glucofuranose
Altrofuranose
Altrofuranose
12
1
6
25
Trace
3
Trace
6
1
1
26
33
30
54
9
Pentosese
-Ribose
0
2.2
0
2.2
0
2.2
0
12
11
12
13
62f
6f
45
42
49
55
16
2
10
11
29
19
8
43
15
39
27
29
8
5
3
2
7
4
2
6
3
Arabinofuranose
Arabinofuranose
1
3
1
-Arabinose
-Xylose
6
10
47
3
-Lyxose
62
9
11
2
Xylofuranose
Xylofuranose
1
18
2
21
2.2
8
1
a Conditions are described in the Experimental section. b From the integration of the H NMR resonances of the 1-H protons. The species listed
made up >96% of the resonances in the range δH 7.0–5.6 (except for idose with added boric acid, 86%). The values for the α- and β-pyranose esters
produced in the absence of boric acid are similar to equilibrium figures.17 c NMR assignments made using data from refs. 4 and 18. d Experiments
carried out on half the scale. e NMR assignments made using data from: 19, 20. f NMR assignment made using data from ref. 21.
With the objective of obtaining a crystalline furanosyl
pentaester for use as a more convenient glucofuranosylating
reagent, acylation of glucose was repeated in the presence of
boric acid but in propanoic acid with propanoic anhydride as
acylation agent. With minor modifications to the acetylation
procedure to accommodate changes in physical properties of
the compounds involved, a solid product was isolated directly,
from which pure penta-O-propanoyl-β--glucofuranose was
obtained by recrystallization (58%). Therefore, under the con-
ditions selected, complete acylation occurred without removal
of the boric acid. This previously unknown compound is the
first readily available, crystalline per-acylated glucofuranose. By
comparison, crystalline 1-O-acetyl-2,3,5,6-tetra-O-benzoyl-β-
Fig. 1 The effects of varying amounts of boric acid on the products of
-glucofuranose, which can be used as a β-glucofuranosylating
acid-catalysed acetylation of -glucose. (Conditions as given in the
agent, can be made in four steps from glucose in an overall yield
Experimental section.)
of 16%.15
The effects of boric acid on the acid-catalysed acetylation of
rearrangement. Boric acid also results in the formation of 1%
of the acyclic heptaacetate 4, this compound having also been
observed in the products of acetolysis of methyl β--gluco-
pyranoside in the presence of iron() chloride.7 The identity
of the penta-O-acetyl--glucofuranoses produced via ester 1
was confirmed by NMR spectroscopy2,4 and by their con-
version to the known phenyl β--glucofuranoside via its
tetraacetate.14
When glucose and boric acid were heated in acetic acid–
acetic anhydride without the addition of sulfuric acid, acetyl-
ation of the sugar (or more precisely acetolysis of the proposed
borate intermediate) still occurred, and this procedure afforded
a means of preparing β--glucofuranose pentaacetate in high
yield (α:β ratio 1:52, by GLC-MS analysis). However, a
two-step process was required to achieve this, the first step
apparently locking the sugar in the β-furanosyl acetate form by
preferential acetolysis at C-1 of the borate 1 with C–O bond
fission and inversion of configuration, and the second, a
standard acetylation carried out with acetic anhydride in the
presence of pyridine after removal of boric acid as the volatile
trimethyl borate, completing the esterification. The two
anomers could not be separated by chromatography (flash silica
column).
other aldo-hexoses and -pentoses are recorded in Table 1 and
suggest that this approach to furanosyl derivatives may be
synthetically useful with xylose and idose as well as with
glucose. In the case of arabinose and galactose, boric acid
suppresses slightly the formation of the furanose forms and
may be of use for enhancing the yields of their pyranose
per-esters. In some cases partial epimerization occurred at C-2 to
give some of the isomeric sugars as their acetylated furanoses,
and notably in the cases of the structurally related mannose and
lyxose, epimerization was enhanced appreciably when boric
acid was present. Such isomerization under acetolysis con-
ditions, which is a known characteristic of aldofuranose esters
having cis-acetoxy groups at C-2 and C-3, has been summarized
by Sowa.16 When pure α--mannofuranose pentaacetate was
subjected to the boric acid–acetic acid–acetic anhydride–
sulfuric acid acetolysing conditions, no epimerization was
observed, although there was 22% conversion to the β-anomer.
This suggests that the appreciable inversion at C-2 that occurs
with mannose itself (Table 1) does so at the borate ester stage as
indicated in Scheme 2. The 2,3-borate ester function therefore
apparently performs the role played by 2,3-acetoxonium ions
during normal direct acetolyses of aldoses containing cis-2,3-
diols as proposed by Sowa.16
2012
J. Chem. Soc., Perkin Trans. 1, 2000, 2011–2014

View Article Online
EtOH); [α]D22 Ϫ30.8 (c 5.0 in CHCl3) (Found: C, 54.8; H, 7.0.
Calc. for C21H32O11: C, 54.8; H, 7.0%); δH 6.14 (1H, s, 1-H), 5.42
(1H, d, J 4.8, 3-H), 5.29 (1H, ddd, J 9.4, 2.5 and 5.2, 5-H), 5.10
(1H, s, 2-H), 4.63 (1H, dd, J 12.3 and 2.5, 6-Ha), 4.55 (1H, dd,
J 9.4 and 4.7, 4-H), 4.10 (1H, dd, J 12.3 and 5.2, 6-Hb), 2.40
(2H, q, J 7.6, CH2), 2.38 (2H, q, J 7.6, CH2), 2.37 (2H, q, J 7.6,
CH2), 2.34 (2H, q, J 7.6, CH2), 2.26 (2H, q, J 7.6, CH2), 1.17
(6H, t, J 7.6, 2 × CH3), 1.13 (6H, t, J 7.6, 2 × CH3), 1.09 (3H, t,
J 7.6, CH3); δC 174.3, 173.4, 173.0, 172.9, 172.9 (5 × C=O), 99.2
(1-C), 80.2 (4-C), 79.9 (2-C), 73.2 (3-C), 68.6 (5-C), 63.3 (6-C),
28.0, 27.7, 27.7, 27.7, 27.6 (5 × CH2), 9.4, 9.2, 9.2, 9.1, 9.1
(5 × CH3); m/z (FAB) 459.18664 (Mϩ Ϫ H, C21H31O11 requires
m/z 459.18636), 387 (Mϩ Ϫ C3H5O2, 100%), 183 (24), 137 (25),
57 (52).
Scheme 2
In summary, boric acid has a profound effect on the acetyl-
ation and acetolysis of aldoses. By its use, crystalline penta-
O-propanoyl-β--glucofuranose can be made in a simple,
inexpensive, ‘one-pot’ procedure. The suitability of this com-
pound as a glycosylating agent is under investigation.
Phenyl ꢀ-D-glucofuranoside
Experimental
Pentaacetates 3, made as described in the acetolysis section
below and without further purification, were converted via the
tetraacetyl glycoside to the title compound as previously
described14 (30%), mp 78.5–80.5 ЊC (from butan-2-one–
petroleum spirit) (lit.,14 79–80 ЊC), [α]D20 Ϫ143.2 (c 2.0 in H2O)
{lit.,14 [α]D20 Ϫ142 (c 2.0 in H2O)} (Found: C, 56.25; H, 6.3. Calc.
for C12H16O6: C, 56.25; H, 6.3%); δH (d6-DMSO) 7.29 (2H, t,
J 7.9, 3Ј- and 5Ј-H), 6.99–6.95 (3H, m, 2Ј-, 4Ј- and 6Ј-H), 5.53
(1H, d, J 3.8, OH), 5.36 (1H, s, 1-H), 4.93 (1H, d, J 4.1, OH),
4.55 (1H, d, J 5.4, OH), 4.36 (1H, t, J 5.7, OH), 4.11 (1H, d,
J 3.8, 2-H), 4.06–4.00 (2H, m, 3- and 4-H), 3.77 (1H, dddd,
J 8.2, 6.2, 5.4 and 3.0, 5-H), 3.55 (1H, ddd, J 11.2, 5.7 and 3.0,
6-Ha), 3.36 (1H, dd, J 11.2 and 6.2, 6-Hb); δC (d6-DMSO) 157.2
(1Ј-C), 129.8 (3Ј- and 5Ј-C), 121.7 (4Ј-C), 116.4 (2Ј- and 6Ј-C),
107.3 (1-C), 82.9 (4-C), 81.3 (2-C), 75.3 (3-C), 69.8 (5-C), 64.0
(6-C); m/z (FAB) 256.09448 (Mϩ, C12H16O6 requires M,
256.09469).
General methods
All non-specialized starting materials were commercially avail-
able research-grade chemicals and were used without further
purification. Analytical TLC was carried out on pre-coated 0.25
mm thick Merck 60 F254 silica gel plates. Visualization was by
absorption of UV light, or by thermal development after spray-
ing with ammonium molybdate and cerium() sulfate in dil.
sulfuric acid. Mps were measured with a Reichert hotstage
microscope and are uncorrected. Optical rotations were
measured on a Perkin-Elmer 241 polarimeter, and [α]D-values
are given in 10Ϫ1 deg cm2 gϪ1. Microanalyses were carried out
at the Campbell Microanalytical Laboratory, University of
1
Otago. H and 13C NMR spectra were recorded at 300 MHz
and 75 MHz respectively in CDCl3 solution (unless otherwise
indicated) with TMS as internal standard. J-Values are given
in Hz. The assignments of peaks were made with assistance
of DEPT and COSY (1H,1H and H,13C) experiments. Mass
1
Acetolysis of aldoses in the absence and presence of boric acid
spectra were recorded by Mr John M. Allen, Horticultural
and Food Research Institute of New Zealand in positive mode
at 15 keV with caesium as the ionizing agent in an NBA–MeOH
or glycerol–MeOH matrix. Petroleum spirit refers to the frac-
tion of distillation range 60–80 ЊC.
The sugar (0.55 mmol), boric acid (0 or 1.21 mmol) and acetic
acid (2 cm3) were stirred at 50 ЊC for 1 h. Acetic anhydride
(2 cm3) and sulfuric acid (2 mm3) were added and heating was
continued for a further 3 h. Ice (20 g) was added and the
mixture was stirred for 1 h and then extracted with CHCl3
(3 × 20 cm3). The extracts were washed successively with
NaHCO3 (saturated aq.) and water, and dried (MgSO4), and the
1,2,3,5,6-Penta-O-acetyl-ꢀ-D-glucofuranose 3ꢀ
-Glucose (5.0 g, 27.8 mmol), boric acid (3.8 g, 60.7 mmol) and
acetic acid (100 cm3) were stirred at 50 ЊC for 1 h by which time
all the sugar had dissolved. Acetic anhydride (100 cm3) was
added and the mixture was heated at 50 ЊC for 16 h. The boric
acid was removed as trimethyl borate by the addition of meth-
anol (20 cm3) and in vacuo concentration of the subsequent
mixture to 100 cm3 and then the addition of methanol (10 cm3)
and concentration in vacuo to 50 cm3 (×2). Acetic anhydride
(100 cm3) and pyridine (100 cm3) were added and the solution
was stirred at 20 ЊC for 2 h. Following the addition of ice
(200 g), the solution was stirred for 1 h and then extracted with
CHCl3 (3 × 150 cm3). The combined extracts were washed suc-
cessively with water, HCl (2 M) and water, dried (MgSO4), and
concentrated in vacuo to give the pentaacetates 3 (10.0 g, 92%)
1
solvent was removed to give a syrup that was analysed by H
NMR spectroscopy (see Table 1).
Acknowledgements
We wish to thank Dr John D. Stevens, University of New South
Wales, Australia for kindly providing the α-mannofuranose
pentaacetate and the 1H NMR data for many of the acetylated
sugars, Professor Arnold E. Stütz, Technische Universität
Graz, Austria for a gift of -idose and Professor Robin J. Ferrier
for help with the preparation of this manuscript.
References
as a pale yellow syrup (90% pure, α:β ratio 1:52); H and 13C
NMR data were identical to literature data (refs. 4 and 2,
respectively).
1
1 M. L. Wolfrom and A. Thompson, in Methods in Carbohydrate
Chemistry, ed. R. L. Whistler and M. L. Wolfrom, Academic Press,
New York, 1963, vol. 2, p. 212.
2 L. V. Backinowsky, S. A. Nepogod’ev, A. S. Shashkov and N. K.
Kochetkov, Carbohydr. Res., 1985, 138, 41.
3 J. D. McChesney and R. Buchman, Heterocycles, 1976, 4, 1065.
4 V. Ferrières, M. Gelin, R. Boulch, L. Toupet and D. Plusquellec,
Carbohydr. Res., 1998, 314, 79.
5 V. Ferrières, J.-N. Bertho and D. Plusquellec, Carbohydr. Res., 1998,
311, 25.
6 P. M. Bhaskar and D. Loganathan, Tetrahedron Lett., 1998, 39,
2215.
1,2,3,5,6-Penta-O-propanoyl-ꢀ-D-glucofuranose
-Glucose (5.0 g, 27.8 mmol), boric acid (3.5 g, 56.9 mmol) and
propanoic acid (75 cm3) were stirred at 70 ЊC for 1 h. Propanoic
anhydride (75 cm3) was added slowly so that precipitation was
avoided and the resulting mixture was heated at 70 ЊC for 48 h.
Ice (100 g) and water (750 cm3) were added and the mixture was
stirred vigorously for 1 h. The precipitate was collected, and
recrystallized (×2) from EtOH (20 cm3) to give the title com-
pound (7.4 g, 58%) as white prisms, mp 74.5–75.5 ЊC (from
7 D. R. McPhail, J. R. Lee and B. Fraser-Reid, J. Am. Chem. Soc.,
1992, 114, 1905.
8 L. v. Vargha, Ber. Dtsch. Chem. Ges., Teil B, 1933, 66, 704.
J. Chem. Soc., Perkin Trans. 1, 2000, 2011–2014
2013

View Article Online
9 J. C. Norrild and H. Eggert, J. Am. Chem. Soc., 1995, 117, 1479.
10 M. Makkee, A. P. G. Kieboom and H. van Bekkum, Recl. Trav.
Chim. Pays-Bas, 1985, 104, 230.
11 R. Aruga, J. Chem. Soc., Dalton Trans., 1988, 2971.
12 E. J. Bourne, E. M. Lees and H. Weigel, J. Chem. Soc., 1965, 3798.
13 P. Brigl and H. Grüner, Ber. Dtsch. Chem. Ges., Teil. B, 1933, 66,
1977.
16 W. Sowa, Can. J. Chem., 1971, 49, 3292.
17 R. U. Lemieux, in Molecular Rearrangements, ed. P. de Mayo,
Interscience, New York, 1964, pp. 709–769.
18 J. D. Stevens, University of New South Wales, Sydney, personal
communication.
19 P. L. Durette and D. Horton, J. Org. Chem., 1971, 36, 2658.
20 B. L. Kam, J.-L. Barascut and J.-L. Imbach, Carbohydr. Res., 1979,
69, 135.
14 P. Jerkeman and B. Lindberg, Acta Chem. Scand., 1963, 17, 1709.
15 R. J. Ferrier and S. R. Haines, Carbohydr. Res., 1984, 127, 157.
21 J.-P. Utille and D. Gagnaire, Carbohydr. Res., 1982, 106, 43.
2014
J. Chem. Soc., Perkin Trans. 1, 2000, 2011–2014