Reaction of 1,2-trans-glycosyl acetates with phosphorus pentachloride: New efficient approach to 1,2-trans-glycosyl chlorides

Article abstract of DOI:10.1016/S0040-4039(02)02446-2

Reaction of phosphorus pentachloride with 1,2-trans-glycosyl esters is described. The reaction mechanism presumably involves formation of a tetrachlorophosphonium ion as one of the key reactive intermediates, which can be induced either by Lewis acids or by using acetonitrile as the reaction solvent. Two novel, efficient methods for the synthesis of the thermodynamically unstable glycosyl chlorides were developed based on this reaction.

Full text of DOI:10.1016/S0040-4039(02)02446-2

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 9577–9580
Reaction of 1,2-trans-glycosyl acetates with phosphorus
pentachloride: new efficient approach to 1,2-trans-glycosyl
chlorides
Farid M. Ibatullina,* and Stanislav I. Selivanovb
aBiophysics Division, Petersburg Nuclear Physics Institute, Gatchina, 188300, Russia
bChemical Department, Petersburg State University, St. Petersburg, Russia
Received 30 July 2002; revised 11 October 2002; accepted 25 October 2002
Abstract—Reaction of phosphorus pentachloride with 1,2-trans-glycosyl esters is described. The reaction mechanism presumably
involves formation of a tetrachlorophosphonium ion as one of the key reactive intermediates, which can be induced either by
Lewis acids or by using acetonitrile as the reaction solvent. Two novel, efficient methods for the synthesis of the thermodynam-
ically unstable glycosyl chlorides were developed based on this reaction. © 2002 Elsevier Science Ltd. All rights reserved.
Acylated 1,2-trans-glycosyl halides are important syn-
thetic intermediates for stereoselective 1,2-cis glycosyla-
tion. They have been widely employed for synthesis of
various 1,2-cis linked sugar derivatives including nitro-
phenyl glycosides,1 glycosyl azides,2 thioglycosides,1,3
glycosyl thioesters,4 and, consequently, thiooligo-
saccharides.4a,b,5
Although 1,2-trans-glycosyl chlorides can also be pre-
pared by chlorination of alkyl or aryl thioglycosides
under milder conditions, allowing the use of acid sensi-
tive protecting groups,13–15 sugar peracetates are still
more attractive precursors of acetylated glycosyl chlo-
rides due to their availability at a lower price.
In the course of our investigation of Lewis acid cata-
lyzed reactions of 1,2-trans-glycosyl esters with various
nucleophiles,16 it has been established that phosphorus
pentachloride is a convenient reagent for synthesis of
1,2-trans-glycosyl chlorides. Previously, phosphorus
pentachloride was used for b-glycosyl chloride synthesis
under quite strong reaction conditions.17 According to
the report, b-glucosyl and b-galactosyl chlorides were
prepared by passing a stream of hydrogen chloride over
5 h through the heated (90–100°C) mixture of corre-
Several methodologies have been developed to synthe-
size thermodynamically unstable glycosyl chlorides.6
Generally, they involve the reaction of 1,2-trans-glyco-
syl acetates with various chlorinating reagents in an
appropriate solvent. According to this approach b-gly-
cosyl chlorides were prepared using anhydrous tin tet-
rachloride,7 titanium tetrachloride8 or aluminum
chloride9 in non-polar solvents or with hydrogen chlo-
ride in dry ether.10 A combination of dichloromethyl
methyl ether–boron trifluoride etherate has been sug-
gested as a more convenient chlorinating reagent,11 but
the high toxicity of dichloromethyl methyl ether and its
relatively high price are obvious drawbacks of the
approach. Recently, a mixture of thionyl chloride and
acetic acid has been utilized for conversion of sugar
peracetates into the title compounds,12 though the con-
version requires considerable time (18 h) for
completion.
sponding b-D-peracetates with a 19 fold excess of
freshly distilled PCl5. The real role of the phosphorus
pentachloride in the process is not clear; however, it
certainly slows the reaction, because in the absence of
PCl5 the conversion proceeds faster and under much
milder conditions.10,18
Phosphorus pentachloride has also been applied for the
conversion of 1,2-trans-glycosyl acetates into 2-O-
trichloroacetyl-b-D-glycosyl chlorides by a 2.5–5 h
reflux of the reactants in carbon tetrachloride.19 Forma-
tion of 1,2-trans-glycosyl chlorides was not observed in
that case. Moreover, the proposed reaction mechanism
did not consider 1,2-trans-glycosyl chlorides, even as
possible intermediates. We, however, believe that per-
* Corresponding author. Tel.: +7-81271-46056; fax: +7-81271-32303;
e-mail: ibatullin@omrb.pnpi.spb.ru
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)02446-2

9578
F. M. Ibatullin, S. I. Seli6ano6 / Tetrahedron Letters 43 (2002) 9577–9580
Based upon these observations, we have developed a
novel approach for the conversion of sugar peracetates
into 1,2-trans-glycosyl chlorides.
acetylated 1,2-cis-glycosyl chlorides, isolated as minor
products of the reaction,19e are the result of anomeriza-
tion of initially formed 1,2-trans-glycosyl chlorides.
Typical procedures are as follows. Method A. To a
stirred suspension of a peracetylated sugar (10 mmol)
and phosphorus pentachloride (2.3 g, 11 mmol) in dry
methylene chloride (20 mL), boron trifluoride etherate
(10–20 mL) was added. After stirring for 5 min TLC
analysis showed complete disappearance of the starting
material. The reaction mixture was diluted with DCM
(20 mL) and then washed with ice-cold water, saturated
ice-cold NaHCO3 solution (2×30 mL), and again ice-
cold water, dried over anhydrous MgSO4 and evapo-
rated under reduced pressure. The residue was
co-evaporated with toluene, triturated under hexane
and crystallized (if necessary) from an appropriate sol-
vent to give the desired compound.
We have found that in the presence of Lewis acids,
phosphorus pentachloride readily reacts with sugar b-
D
-peracetates under very mild reaction conditions lead-
ing to b-glycosyl chlorides. So, treatment of a stirred
solution of 1,2-trans-glycosyl acetate in methylene chlo-
ride with a slight excess of PCl5 and a catalytic amount
of boron trifluoride etherate causes a weak exothermic
reaction, which is usually complete in a few minutes
producing the target compounds in high yields.
Boron trifluoride etherate and anhydrous aluminum
chloride were the best catalysts for the reaction,
although in case of AlCl3 the reaction has an induction
period, probably due to its slower solubility in the
reaction mixture.
Method B. To a stirred solution of a peracetylated
sugar (10 mmol) in anhydrous acetonitrile (20 mL)
powdered phosphorus pentachloride (2.5 g, 12 mmol)
was added in one portion. The reaction mixture was
stirred at room temperature for 5–15 min to comple-
tion, then diluted with DCM (40 mL) and toluene (20
mL), and worked up as described above to give the
desired product.
The reaction mechanism presumably involves initial
+
formation of a tetrachlorophosphonium ion, PCl4 ,
induced by the Lewis acid. The phosphonium ion
attacks the carbonyl oxygen of the acetoxy group at the
anomeric position, producing phosphorus oxychloride,
acetyl chloride and a 1,2-acyloxonium ion (Scheme 1).
The acyloxonium ion, in turn, reacts with PCl5 and
+
produces glycosyl chloride and a PCl4 ion which ini-
Contrary to the high reactivity of 1,2-trans-glycosyl
acetates, 1,2-cis-isomers do not react with phosphorus
pentachloride under the described conditions for steric
reasons. In this case, the possible elimination of the
acetoxy group from the anomeric position of the sugar
can not be accompanied by 1,2-acyloxonium ion forma-
tion. Moreover, phosphorus pentachloride generally
does not react with the carbonyl group of esters.22 The
only known exception is the reaction of PCl5 with alkyl
formates, leading to dichloromethyl methyl ether,23
which is often utilized for the conversion of sugar
acetates into the title compounds by treatment with a
large (up to 13 molar) excess of the reagent.11
tiates a new cycle of the transformation.
The formation of the tetrachlorophosphonium ion as
an initial intermediate may occur due to the ease of
ionization of phosphorus pentachloride in the presence
of Lewis acids.20 It is also known that in polar solvents,
for example, in acetonitrile, PCl5 is almost completely
dissociated into PCl4 and PCl6 ions.20 Therefore, to
prove the proposed mechanism, the reaction was per-
formed in acetonitrile without any catalyst. Like the
Lewis-catalyzed reaction, fast disappearance of starting
peracetates and simultaneous formation of b-glycosyl
chlorides was observed by TLC analysis. The conver-
sion required less than 15 min for its full completion.
Thus, it can be concluded that the reaction of phospho-
rus pentachloride with 1,2-trans-glycosyl acetates pro-
ceeds via primary formation of tetrachlorophos-
phonium ion, induced either by Lewis acid or by the
ionizing power of the solvent.21
+
−
It has been reported that synthesis of glycosyl halides of
oligosaccharides is often accompanied by interglyco-
sidic bond cleavage, decreasing the yield of target prod-
ucts.11d,12 To evaluate the stability of interglycosidic
bonds under the described conditions, we synthesized
b-maltosyl chloride using our methods. In both cases
Scheme 1. Proposed mechanism for halogenation of sugar acetates with phosphorus pentachloride.

F. M. Ibatullin, S. I. Seli6ano6 / Tetrahedron Letters 43 (2002) 9577–9580
Table 1. Synthesis of 1,2-trans-glycosyl chlorides
9579
the compound was isolated in 92% yield. NMR spectra
and optical rotation values of the obtained samples
were essentially the same, leading to the conclusion that
cleavage of interglycosidic bonds under the described
conditions does not occur.
References
1. Apparu, M.; Blanc-Muesser, M.; Defaye, J.; Driguez, H.
Can. J. Chem. 1981, 59, 314–320.
2. (a) Gyo¨rgydea´k, Z.; Paulsen, H. Liebigs Ann. Chem.
1977, 1987–1991; (b) Takeda, T.; Sugiura, Y.; Ogihara,
Y.; Shibata, S. Can. J. Chem. 1980, 58, 2600–2603; (c)
Ogawa, T.; Nakabayashi, S.; Shibata, S. Agr. Biol.
Chem. 1983, 47, 281–286; (d) Gyo¨rgydea´k, Z.; Szila´gyi,
L. Liebigs Ann. Chem. 1987, 235–241; (e) Peto¨, C.;
Batta, G.; Gyo¨rgydea´k, Z.; Cztaricskai, F. Liebigs Ann.
Chem. 1991, 505–507.
3. (a) Blanc-Muesser, M.; Driguez, H.; Joseph, B.; Viaud,
M. C.; Rollin, P. Tetrahedron Lett. 1990, 31, 3867–3868;
(b) Blanc-Muesser, M.; Vigne, L.; Driguez, H. Tetra-
hedron Lett. 1990, 31, 3869–3870.
4. (a) Blanc-Muesser, M.; Defaye, J.; Driguez, H. Carbo-
hydr. Res. 1978, 67, 305–328; (b) Blanc-Muesser, M.;
Defaye, J.; Driguez, H.; Marchis-Mouren, G.; Seigner,
C. J. Chem. Soc., Perkin Trans. 1 1984, 1885–1889; (c)
Gadelle, A.; Defaye, J.; Pedersen, C. Carbohydr. Res.
1990, 200, 497–498.
The proposed methods have been applied for the syn-
thesis of b-chlorides of
D-glucopyranose, D-galactopy-
ranose,
D
-xylopyranose and maltose (Table 1).24 Some
of the syntheses were scaled up to 100 mmol amounts.
The crude products obtained by method A are suffi-
ciently pure and do not require additional purification.
This is an essential advantage, because of the low
stability of these compounds. Method B also affords
pure enough crude products, although some were
recrystallized to give satisfactory elemental analyses. In
each case, only the b-anomer of glycosyl chlorides was
formed during halogenation, according to chemical-
shift data. The glycosyl chlorides obtained have been
successfully utilized in the synthesis of various 1,2-cis-
glycosyl azides, 1,2-cis-thioglycosides and 1,2-cis-
thiooligosaccharides.
5. (a) Blanc-Muesser, M.; Defaye, J.; Driguez, H. Tetra-
hedron Lett. 1976, 47, 4307–4310; (b) Blanc-Muesser,
M.; Driguez, H. J. Chem. Soc., Perkin Trans. 1 1988,
3345–3351; (c) Cottaz, S.; Driguez, H. Synthesis 1989,
755–758.
6. For review see: Haynes, L. J.; Newth, F. H. Adv. Car-
bohydr. Chem. 1955, 10, 207–256.
7. (a) Lemieux, R. U.; Brice, C. Can. J. Chem. 1952, 30,
295–310; (b) Lemieux, R. U.; Brice, C. Can. J. Chem.
1955, 33, 109–119.
Conclusion
The reactions described here of 1,2-trans-glycosyl
acetates with phosphorus pentachloride provides two
novel efficient methods for the synthesis of 1,2-trans-
glycosyl chlorides with yields comparable to those from
the best available methods, but superior in terms of
simplicity, mild reaction conditions, rapidity, and
availability of reagents.
8. (a) Pacsu, E. Ber. 1928, 61, 1508–1513; (b) Dick, W. E.;
Weisleder, D. Carbohydr. Res. 1976, 46, 173–182.
9. (a) Korytnyk, W.; Mills, J. A. J. Chem. Soc. 1959,
636–649; (b) Holland, C. V.; Horton, D.; Jewell, J. S. J.
Org. Chem. 1967, 32, 1818–1821.
Acknowledgements
10. (a) Freudenberg, K.; Ivers, O. Ber. 1922, 55B, 929–941;
(b) Kissman, H. M.; Baker, B. R. J. Am. Chem. Soc.
1957, 79, 5534–5540.
Authors are grateful to Prof. A. G. Shavva for fruitful
discussion and to Dr. S. P. Shevtsov and Dr. A. E.
Aleshin for assistance in the manuscript preparation.

9580
F. M. Ibatullin, S. I. Seli6ano6 / Tetrahedron Letters 43 (2002) 9577–9580
11. (a) Farkas, J.; Szabo´, I. F.; Bogna´r, R.; Anderle, D.
Carbohydr. Res. 1976, 48, 136–138; (b) Exoffier, G.;
Paillet, M.; Vignon, M. Carbohydr. Res. 1984/1985, 135,
C10–C11; (c) Kartha, K. P. R.; Jennings, H. Tetrahedron
Lett. 1990, 31, 2537–2540; (d) Kova´c, P.; Edgar, K. J. J.
Org. Chem. 1992, 57, 2455–2467.
12. Chittenden, G. J. F. Carbohydr. Res. 1992, 242, 297–301.
13. (a) Wolfrom, M. L.; Groebke, W. J. Org. Chem. 1963,
28, 2986–2988; (b) Wolfrom, M. L.; Garg, H. G.; Hor-
ton, D. J. Org. Chem. 1963, 28, 2989–2991; (c) Horton,
D.; Wolfrom, M. L.; Garg, H. G. J. Org. Chem. 1963, 28,
2992–2995.
14. (a) Kartha, K. P. R.; Field, R. A. Tetrahedron Lett. 1997,
38, 8233–8236; (b) Kartha, K. P. R.; Cura, P.; Aloui, M.;
Readman, S. K.; Rutherford, T. J.; Field, R. A. Tetra-
hedron: Asymmetry 2000, 11, 581–593.
15. Sugiyama, S.; Diakur, J. M. Org. Lett. 2000, 2, 2713–
2715.
16. Ibatullin, F. M.; Shabalin, K. A. Carbohydr. Lett. 2000,
3, 427–429.
17. Espinosa, F. G. Acta Salmanticensia. Ser. Cienc. 1958, 2,
53; Chem Abstr. 1959, 53, 16973h.
18. (a) Fox, J. J.; Goodman, I. J. Am. Chem. Soc. 1951, 73,
3256–3260; (b) Austin, P. W.; Hardy, F. E.; Buchanan, J.
G.; Baddiley, J. J. Chem. Soc. 1964, 2128–2137.
19. (a) Brigl, P. Z. Physiol. Chem. 1921, 116, 1; (b) Brigl, P.;
Mistele, P. Z. Physiol. Chem. 1923, 126, 120–129; (c)
Lemieux, R. U.; Huber, G. Can. J. Chem. 1953, 31,
1040–1047; (d) Lemieux, R. U.; Howard, J. Methods
Carbohydr. Chem. 1963, 2, 400–402; (e) Saito, S.; Ichi-
nose, K.; Sasaki, Y.; Sumita, S. Chem. Pharm. Bull. 1992,
40, 3261–3268.
(300 MHz, CDCl3): l 5.78 (m, 1H, H–1), 5.05–4.97 (m,
2H, H–2, H–3), 4.86 (m, 1H, J4,5a=3.0 Hz, J4,5b=3.7 Hz,
H–4), 4.36 (dd, 1H, J5a,5b=12.9 Hz, H–5a), 3.76 (dd, 1H,
H–5a), 2.10, 2.099, 2.096 (3 × s, 9H, 3 × Ac); 13C NMR
(75 MHz, CDCl3): l 169.7, 169.2, 169.0 (3 × CO-Ac),
88.5 (C-1), 70.1, 67.3, 66.8 (C-2, C-3, C-4), 61.5 (C-5),
20.75, 20.7, 20.6 (3 × CH3-Ac).
2,3,4,6-tetra-O-Acetyl-b-D
-glucopyranosyl chloride, 1H
NMR (500 MHz, CDCl3): l 5.28 (d, 1H, J1,2=8.2 Hz,
H–1), 5.20–5.12 (m, 3H, J4,5=9.7 Hz, H–2, H–3, H–4),
4.24 (dd, 1H, J6a,6b=12.5 Hz, H–6a), 4.15 (dd, 1H, H–6b),
3.80 (ddd, 1H, J5,6a=4.7 Hz, J5,6b=2.3 Hz, H–5), 2.08,
2.06, 2.01, 1.99 (4s, 4 × 3H, 4 × Ac); 13C NMR (125
MHz, CDCl3): l 170.5, 170.0, 169.2, 169.0 (4 × CO-Ac),
87.5 (C-1), 75.5 (C-5), 73.3 (C-3), 72.7 (C-2), 67.5 (C-4),
61.5 (C-6), 20.6, 20.5, 20.4 (4 × CH3-Ac).
2,3,4,6-tetra-O-Acetyl-b-D
-galactopyranosyl chloride, 1H
NMR (300 MHz, CDCl3): l 5.41 (dd, 1H, J4,5=1 Hz,
H–4), 5.36 (dd, 1H, J2,3=10.1 Hz, H–2), 5.24 (d, 1H,
J1,2=8.8 Hz, H–1), 4.99 (dd, 1H, J3,4=3.4 Hz, H–3), 4.14
(d, 2H, H–6a,b), 4.01 (td, 1H, J5,6a=J5,6b=6.4 Hz, H–5),
2.16, 2.07, 2.04, 1.97 (4s, 4 × 3H, 4 × Ac); 13C NMR (75
MHz, CDCl3): l 170.3, 170.0, 169.8, 169.1 (4 × CO-Ac),
88.1 (C-1), 74.5, 70.8, 70.6, 66.7 (C-2, C-3, C-4, C-5), 61.2
(C-6), 20.6, 20.5, 20.4 (4 × CH3-Ac).
2,3,6-tri-O-Acetyl-(2,3,4,6-tetra-O-acetyl-a-D-glucopyran-
osyl)-b-D-glucopyranosyl chloride (2,3,6,2%,3%,4%,6%-hepta-
O-acetyl-b-maltosyl chloride), 1H NMR (500 MHz,
CDCl3): l 5.39 (d, 1H, J1%,2%=4.0 Hz, H–1%), 5.36 (d, 1H,
J
1,2=8.0 Hz, H–1), 5.33 (dd, 1H, J3%,4%=9.8 Hz, H–3%),
5.19 (dd, 1H, J3,4=8.6 Hz, H–3), 5.03 (dd, 1H, J4%,5%
=
10.0 Hz, H–4%), 4.98 (dd, 1H, J2,3%=8.3 Hz, H–2), 4.83
(dd, 1H, J2%,3%=10.5 Hz, H–2%), 4.49 (dd, 1H, J6a,6b=12.4
Hz, H–6a), 4.22 (dd, 1H, J6%a,6%b=12.4 Hz, H–6%a), 4.21
(dd, 1H, H–6b), 4.12 (dd, 1H, J4,5=9.7 Hz, H–4), 4.04
(dd, 1H, H–6%b), 3.94 (ddd, 1H, J5%,6%a=3.9 Hz, J5%,6%b=2.3
Hz, H–5%), 3.81 (ddd, 1H, J5,6a=2.7 Hz, J5,6b=4.6 Hz,
H–5), 2.14, 2.08, 2.05, 2.02, 2.01, 2.00, 1.98 (7s, 7 × 3H,
7 × Ac); 13C NMR (125 MHz, CDCl3): l 170.43, 170.4,
170.3, 170.0, 169.9, 169.3, 169.2 (7 × CO-Ac), 95.8 (C-1%),
87.2 (C-1), 75.5, 75.3, 74.2, 72.2, 70.0, 69.2, 68.6, 68.0
(C-2, C-2%, C-3, C-3%, C-4, C-4%, C-5, C-5%), 62.5, 61.5
(C-6, C-6%), 20.77, 20.71, 20.6, 20.5, 20.49, 20.48, 20.46
(7 × CH3-Ac).
20. (a) Van Wazer, J. R. Phosphorus and its Compounds; New
York: Wiley, 1958; Vol. 1; (b) Corbridge, D. E. C.
Phosphorus. An Outline of its Chemistry Biochemistry and
Technology; 2nd ed.; Elsevier: Amsterdam-Oxford-New
York, 1980.
+
21. PCl4 ion has been postulated as a reactive intermediate
in some PCl5 involving reactions: (a) Kirsanov, A. V.;
Molosnova, V. P. Zh. Obshch. Khim. 1958, 28, 30–35;
Chem. Abstr. 1958, 52, 12760b; (b) Newman, M. S.;
Wood, L. L., Jr. J. Am. Chem. Soc. 1959, 81, 4300–4303.
22. (a) Fialkov, Y. A.; Buryanov, Y. B. Docl. Akad. Nauk
SSSR 1953, 92, 585–588; Chem. Abstr. 1954, 48, 5708c;
(b) Weygand-Hilgetag, Organish-Chemische Experimen-
tierkunst, 3rd ed., Johann Ambrosius Barth Verlag:
Leipzig, 1964.
The same compound was prepared using method B.
Although the crude product, isolated in 92% yield, was
sufficiently pure, it was recrystallized from ether–hexane
to give the sample with satisfactory elemental analysis
(yield 83%).
23. Fischer, H.; Schwarz, A Ann. 1934, 512, 239–249.
24. All synthesized compounds gave satisfactory elemental
1
analyses and were identified by optical rotations, H, 13C
25. The NMR data is presented using the convention fol-
lowed in Carbohydrate Research, (see Instructions to
Authors).
NMR, EI–MS spectroscopy. Selected spectral data:25
2,3,4-tri-O-acetyl-b-D
-xylopyranosyl chloride, 1H NMR