Synthesis of 1,2-cis- and 1,2-trans-glycosides of 2-acetamido-4,6-O- benzylidene-2-deoxy-d-glucopyranose by anomeric O-alkylation

Article abstract of DOI:10.1016/j.carres.2011.01.016

The reaction of a partially protected 1-hydroxy derivative of N-acetyl-d-glucosamine with benzyl bromide under conditions of anomeric O-alkylation was studied. It was found that the stereoselectivity of the reaction depended on the nature of the alkali me

Full text of DOI:10.1016/j.carres.2011.01.016

Carbohydrate Research 346 (2011) 685–688
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Note
Synthesis of 1,2-cis- and 1,2-trans-glycosides of 2-acetamido-4,6-O-
benzylidene-2-deoxy-D-glucopyranose by anomeric O-alkylation
⇑
Sergey S. Pertel , Oksana A. Gorkunenko, Elena S. Kakayan, Vasily Ja. Chirva
Department of Organic and Biological Chemistry, V.I. Vernadsky Taurida National University, Vernadsky ave. 4, Simferopol, Crimea 95007, Ukraine
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 1 August 2010
Received in revised form 11 January 2011
Accepted 16 January 2011
Available online 23 January 2011
The reaction of a partially protected 1-hydroxy derivative of N-acetyl-D-glucosamine with benzyl bro-
mide under conditions of anomeric O-alkylation was studied. It was found that the stereoselectivity of
the reaction depended on the nature of the alkali metal cation constituent of a transient ion pair. The sub-
stitution of the Li+ cation for K+ or complexation with a crown ether allowed the steric outcome to be
shifted from b- to a-selectivity.
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
Anomeric O-alkylation
Glycosaminide synthesis
Stereoselectivity
Ion pairs
The Curtin–Hammett principle
Hammond’s postulate
The anomeric O-alkylation method1–4 offers a successful alter-
native to traditional methods of glycoside synthesis, which are
based on electrophilic glycosyl donors. A highly nucleophilic ano-
meric alkoxide anion is formed as a result of the interaction of
the corresponding lactol with a base. This anion represents a nucle-
ophilic glycosyl donor and can be alkylated in situ to produce the
corresponding glycoside. The feasibility of the use of unprotected
or partially protected sugar derivatives is an advantage of this
method. The stereochemical results of anomeric O-alkylation were
explained3 using the concept of kinetic and thermodynamic
anomeric effects. Because of the higher reactivity of an equatorial
b-oxide, equatorial glycopyranosides are predominantly formed
under kinetic control (kinetic anomeric effect). On the other hand,
if the alkylation reaction occurs faster than the isomerization of the
anomeric oxides, axial glycopyranosides are formed predomi-
nantly, as the axial oxides usually prevail in the reaction mixture
because of the thermodynamic anomeric effect.5 Furthermore,
complexation also affects the stereoselectivity of the glycoside
synthesis in some cases where one or another anomeric alkoxide
is stabilized due to the chelate effect.6–9 Unfortunately, chelate
complexation apparently does not affect the ratio of anomeric oxi-
des in gluco- and galactopyranose series,10 which prevents using
this effect for changing the stereochemical results of anomeric
alkylation.
atives could be ensured by changes in the nature of the alkali metal
cation that serves as a counterion for the anomeric alkoxide anion.
It was reported11 that the anomeric O-alkylation of the 2-deoxy
sugar derivative in THF in the presence of KHMDS resulted in a
highly stereoselective formation of the corresponding a-glycoside.
At the same time, when NaH in DMF was used as a base, the mix-
ture of anomeric glycosides with the ratio 1:1 was formed. It seems
that such considerable change of stereoselectivity could be con-
nected not only with changing of solvent, temperature, and nature
of base, but also with replacement of alkali metal cation. As shown
by Beau et al.,12 the b-stereoselectivity of anomeric alkylation of
sodium alkoxides was considerably improved when LiBr was
added to the reaction mixture, which was aimed at promoting
the solubility of the starting unprotected sugar derivative in
DMF. Because NaBr, in contrast to LiBr, is insoluble in DMF, it is
conceivable that lithium alkoxide was formed in the reaction mix-
ture as a result of the exchange reaction, which could influence the
stereoselectivity of anomeric alkylation.
To test this assumption, we synthesized benzylidene derivative
113,14 (Scheme 1) by the reaction of N-acetyl-
D-glucosamine with
benzaldehyde dimethyl acetal in DMSO in the presence of a cata-
lytic amount of pyridinium perchlorate. The solution of benzyli-
dene acetal 1 in DMF was then treated with lithium (or sodium,
or potassium) t-butoxide, and the alkoxides obtained were in situ
1-O-benzylated at ꢀ30 °C with benzyl bromide. The selective
1-O-alkylation of 1 is possible in these conditions because of higher
acidity of the hemiacetal hydroxyl group compared to the other
sugar hydroxyls that allows the selective deprotonation of the gly-
coside hydroxyl.15
We have presumed that the control of anomeric alkylation ste-
reoselectivity in the case of 2-amino-2-deoxyglucopyranose deriv-
⇑
Corresponding author. Tel.: +380 652 608 379; fax: +380 652 637 589.
E-mail addresses: orgchem@crimea.edu, sergepertel@yahoo.com (S.S. Pertel).
0008-6215/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2011.01.016

686
S. S. Pertel et al. / Carbohydrate Research 346 (2011) 685–688
OH
Ph
O
O
O
HO
HO
O
a
b
HO
OH
OH
HNAc
HNAc
1
Ph
Ph
c
Ph
+
O
O
O
O
O
O
O
O
O
OBzl
HO
HO
HO
OM
HNAc
HNAc
HNAc
OBzl
2
3
4
M = Li, Na, K
Scheme 1. Reagents: (a) PhCH(OCH3)2, PyꢁHClO4, DMSO; (b) (CH3)3COM, DMF; (c) BzlBr, DMF.
Table 1
Anomeric alkylation of 1 with benzyl bromide in DMF in the presence of alkali metal t-butoxides
Entry
Base
Volume of DMF per 100 mg of 1 (mL)
Yielda (%)
Anomeric ratiob
(a/b)
1
2
3
4
5
6
7
8
(CH3)3COLi
(CH3)3CONa
(CH3)3COK
(CH3)3COK + 18K6
(CH3)3COLi
(CH3)3CONa
(CH3)3COK
2
2
2
2
1
1
1
1
74
69
73
70
74
72
72
75
1:3.3
1:2.5
4.1:1
13.0:1
1:3.0
1:3.6
2.0:1
10.0:1
(CH3)3COK + 18K6
a
Isolated yield after column chromatography.
Anomeric ratio is based on the isolated products.
b
Table 2
Optical rotationa of mixture of anomeric alcoholates 2 in DMF at ꢀ30 °C
Anomeric mixtureb
Base
Volume of DMF per 100 mg of 1 (mL)
ꢀ30
Entry
½Mꢂ
546
1
2
3
4
5
6
A
B
A
B
A
B
A
B
A
B
A
B
(CH3)3COLi
(CH3)3COLi
(CH3)3CONa
(CH3)3CONa
(CH3)3COK
(CH3)3COK
(CH3)3COLi
(CH3)3COLi
(CH3)3CONa
(CH3)3CONa
(CH3)3COK
(CH3)3COK
2
2
2
2
2
2
1
1
1
1
1
1
ꢀ132
ꢀ132
ꢀ213
ꢀ213
ꢀ147
ꢀ147
ꢀ132
ꢀ132
ꢀ213
ꢀ213
ꢀ147
ꢀ147
7c
8c
9
10
11
12
a
Unless otherwise indicated, the optical rotation was measured first at 10 min after reagent mixing and dissolution of base and then after 30, 90 and 240 min. In all cases
the value of optical rotation did not change during the period of measurements.
b
Two mixtures of anomers of 1 were used: mixture A with anomeric ratio 2:3 (a/b) and mixture B with anomeric ratio 3:1 (a/b) (see Section 1).
Because of slow dissolution of lithium t-butoxide the first optical rotation was measured at 60 min after reagent mixing and then after 90 and 240 min.
c
It turned out that the anomeric ratios of the reaction products
depended on the nature of the alkali metal (Table 1). The highest
b-selectivity was observed with the lithium alkoxide for which
the proper application of Curtin–Hammett principle envisages an
evidence of equilibrium establishing among reactants, we had
been attempting to compare previously the rate of anomerisation
of anomeric oxides and the rate of their alkylation. Monitoring of
the reaction by TLC had been showing that the completion of the
alkylation process required from 17 h (in the presence of
(CH3)3COK), up to 83 h (in the presence of (CH3)3COLi) (see Section
1). For estimation of rate of glycosyl oxide anomerisation we had
been measuring optical rotation of two samples of benzylidene
derivative 1 with different anomeric ratios in the presence of alkali
metal t-butoxide in DMF at ꢀ30 °C (Table 2).
the
t-butoxide the stereoselectivity inverted and the benzyl
side 4 was formed predominantly ( /b = 4.1:1, Table 1, entry 3).
a
/b ratio was 1:3.3 (Table 1, entry 1), whereas with potassium
a-glyco-
a
An intermediate value of the anomeric ratio was observed in the
case of sodium t-butoxide. Alkylation of potassium alkoxide of
derivative 1 in the presence of an equivalent amount of 18-
crown-6 capable of binding effectively the potassium ions resulted
in a significant change in the anomeric ratio in favor of a-glycoside
4 (Table 1, cf. entries 3 and 4).
In our opinion, the results obtained can be rationalized using
Hammond’s postulate and the Curtin–Hammett principle, which
allow predicting the ratio of the reaction products in those cases
where the reactants form an equilibrium mixture. In as much as
As the data of Table 2 suggest, at least at 10 min after reagent
mixing, the optical rotation of mixture of anomeric alcoholates of
an alkali metal was constant and did not depend on the composi-
tion of initial equilibrium mixture. These data confirm significant
exceeding of the rate of anomerisation over the rate of alkylation.

S. S. Pertel et al. / Carbohydrate Research 346 (2011) 685–688
687
B
A
=/
T1
G
α
=/
=/
=/
=/
T2
T2
α
β
G
- Gα<0
=/
G - G >0
=/
T1
β
=/
β
β
α
β-
oxide
α-oxide
β-oxide
α-oxide
β-glycoside
β-glycoside
α-glycoside
α-glycoside
Reaction coordinate
Figure 1. Possible changes in the relative energies of the transition states upon changes in the anomeric oxide reactivities. A – b-selectivity, B – a-selectivity.
Hence, there was enough time for an equilibrium to be established
among the reactants; therefore, the Curtin–Hammett principle can
be used in this case for forecast of products ratios.
to an increase of its reactivity, the change of the counterion from
Li+ to Na+ and from Na+ to K+ should lead to augmentation of the
alkoxide reactivity. Complexation of a cation by a crown ether
should contribute to an additional increase in the alkoxide reactiv-
ity as such complexation should induce an additional decrease of
ion pair tightness.
In accordance with the Curtin–Hammett principle, the product
ratio is not determined by the equilibrium constant between the
reactants, but rather by the difference in the free energies of the
corresponding transition states. In the cases where the equatorial
(b-) glycoside prevails in the mixture of the anomeric alkylation
products, the difference between the energy of the transition state
in the reaction of a reactive b-oxide with an electrophile and that of
Interestingly, the stereoselectivity of 1-O-alkylation of deriva-
tive 1 depends on concentration of the reactants in the reaction
mixture; increasing the concentration, as a rule, contributes to b-
selectivity (Table 1, cf. entries 1–4 and 5–8). Apparently, this is re-
lated to change of dissociation degree of an ion pair with change of
concentration. In more dilute solutions the dissociation degree
grows and ion pair becomes less tight and more reactive. It also
should be noted, that, in more concentrated solutions, alkylation
in the presence of sodium t-butoxide was more b-selective than
in the presence of lithium t-butoxide (Table 1, cf. entries 5 and
6). Probably, this is related to association of lithium alcoholates
in more concentrated solutions,17 which can influence both the
energy of anomeric oxides and the energy of corresponding transi-
tion states and, as a result, lead to change of difference between the
energies of the transition states.
a more stable axial
the contrary, in the case of a-selective anomeric alkylation, the en-
a
-oxide, G–b ꢀ G–a , must be negative (Fig. 1A). On
ergy of the transition state derived from b-oxide must be larger
than that for the
be positive (Fig. 1B).
a
-anomer, so that the difference G–b ꢀ Ga– would
Such a change in the relative energy of transition states, which
provides the shift from b- to a-selectivity, probably occurs with an
increase in the reactivities of anomeric oxides. Indeed, the increase
in the reactivity of reactant leads to the formation of an earlier
transition state that is closer in energy to the reactant. This, accord-
ing to Hammond’s postulate, implies that the transition state and
the reactant come close together in structure (early, reactant-like
transition state). The difference between the energies of the transi-
tion states, which are close in structure to the corresponding ano-
meric oxides, should be close to the difference between the
energies of these oxides.
In conclusion, we have shown that the change in nature of an
ion pair composed of an anomeric alkoxide anion and an alkali me-
tal cation may serve as an additional effective tool for controlling
anomeric O-alkylation stereoselectivity.
Because in the equatorial b-oxide there is stronger repulsion of
the lone electron pairs of the ring oxygen compared to axial
1. Experimental
a
-oxide (two gauche interactions compared to one gauche interac-
tion in the -oxide), the b-oxide is destabilized and its reactivity is
higher than that of the
-anomer (kinetic anomeric effect).1,15 If
the axial -oxide is more stable than the equatorial b-oxide be-
1.1. General methods
a
a
Reagents of reagent grade were purchased from standard ven-
dors (Aldrich and Fluka) and used without additional purification.
t-Butyl alcohol was distilled over metallic sodium. DMF was dried
using azeotropic distillation with benzene with subsequent frac-
tional vacuum distillation. Tetrahydrofuran was dried by boiling
under reflux over metallic sodium. Thin-layer chromatography
was performed on silica gel STH-1A-coated aluminum foil (Sorbpo-
limer, Russian Federation). Visualization of spots of carbohydrate
derivatives was effected by exposure of TLC plates to chlorosul-
fonic acid vapor for 5 min at room temperature followed by heat-
ing to ꢃ200 °C. Column chromatography was carried out on
Silica Gel 60 (Fluka 220–448 mesh). 1H NMR spectra were recorded
on a Varian Mercury 400 spectrometer (400.49 MHz). The chemical
shifts are referred to the signal of TMS (dH 0.0). Optical rotation was
measured with a Carl–Zeiss Polamat-S polarimeter. Melting points
were determined in capillaries and were uncorrected.
a
cause of the anomeric effect, the difference between the energies
of the transition states G–b ꢀ Ga– must increase with approximation
of the energies of the transition states and the energies of corre-
sponding oxides and the proportion of the
tion mixture must also increase.
a-glycoside in the reac-
The change in reactivities of anomeric alkoxides can occur with
the change of the alkali metal cation used as a counterion. It is well
known that the complexation activity of alkali metal cations appre-
ciably differs in as much as their size, hardness, and Lewis acidity.
The greater the Lewis acidity of a cation, the more its complexation
activity, and the more tightness of the ion pair that is built of an
anomeric alkoxide anion and an alkali metal cation. As a rule, the
lithium cation is the most powerful Lewis acid among alkali metals
cations and the potassium cation is less powerful than its sodium
counterpart.16 Because a decrease of ion pair tightness corresponds

688
S. S. Pertel et al. / Carbohydrate Research 346 (2011) 685–688
1.2. General procedure for the synthesis of benzyl 2-acetamido-
4,6-O-benzylidene-2-deoxy-b- -glucopyranoside (3) and benzyl
2-acetamido-4,6-O-benzylidene-2-deoxy- -glucopyranoside
(4) by anomeric alkylation in the presence of alkali metal tert-
butoxides
determined by 1H NMR in DMSO-d6) and with optical rotation
20
D
½
a
ꢂ
ꢀ3 (c 1.4, DMF), was obtained by crystallization of 1 from a
546
a-D
DMSO–H2O mixture.14 Mixture B, with anomeric ratio 3:1 (
a/b)
(as determined by 1H NMR in DMSO-d6) and with optical rotation
½
20
aꢂ
+23 (c 1.4, DMF) was obtained by chromatography of 1 on a
546
column with silica gel using CHCl3?CHCl3–EtOH, 100:8 (v/v)
An alkali metal (Na, K or Li, 1.68 mmol) was added to t-butyl
alcohol (6 mL) and the mixture was heated under reflux until the
metal dissolved completely. Tetrahydrofuran (3 mL) was added
to the solution and the solvents were evaporated at 45 °C with a
stream of dry argon to a pasty residue. Tetrahydrofuran was added
to, and distilled from, the residue three times. To the synthesized
alkali metal t-butoxide cooled to ꢀ30 °C, a solution of 400 mg
(1.29 mmol) of benzylidene derivative 1,13,14 obtained after disso-
lution of 1 in hot DMF (the ratios between 1 and DMF are indicated
in Table 1) followed by cooling to the room temperature, was
added. The mixture was cooled to ꢀ30 °C and benzyl bromide
(0.200 mL, 1.68 mmol) was added. After 5 h the reaction mixture
was warmed to ꢀ20 °C and kept for completion of the reaction
(approximately during 12 h in the case of potassium alkoxide
and approximately during 78 h in the case of lithium one). To study
the effect of crown ether on the anomeric alkylation stereochemis-
try, we added 18-crown-6 (440 mg, 1.68 mmol) to the reaction
mixture before addition of benzyl bromide. The progress of the
conversion was monitored by TLC (20:1 CHCl3–EtOH). After com-
pletion of alkylation, the reaction mixture was diluted with water
(50 mL) and the suspension obtained was stirred for 20 min. After
12 h the precipitate was filtered off, dried, and chromatographed
on a column with silica gel using CHCl3?CHCl3–EtOH, 100:4 (v/
v) solvent systems followed by crystallization from a dioxane–pro-
pan-2-ol mixture. For the yields of anomeric benzyl glycosides 3
and 4 and anomeric ratios, see Table 1.
solvent systems.
An alkali metal (Na, K or Li, 0.84 mmol) was added to t-butyl
alcohol (3 mL) and the mixture was heated under reflux until the
metal dissolved completely. Tetrahydrofuran (3 mL) was added
to the solution and the solvents were evaporated at 45 °C with a
stream of dry argon to a pasty residue. Tetrahydrofuran was added
to, and distilled from the residue three times. To the synthesized
alkali metal t-butoxide cooled to ꢀ30 °C, a solution of 200 mg
(0.645 mmol) of benzylidene derivative 1, obtained after dissolu-
tion of anomeric mixture A or B in hot DMF (the ratios between
anomeric mixtures and DMF are indicated in Table 2) followed
by cooling to the room temperature, was added slowly under dry
argon. The mixture was shaken intensively at ꢀ30 °C until the pre-
cipitate dissolved completely. Data on optical rotation of obtained
solutions at ꢀ30 °C are shown in Table 2. Unless otherwise indi-
cated (see Table 2), the optical rotation values were measured at
first time at 10 min after reagents mixing and dissolution of base
and then after 30, 90, and 240 min. In all cases the value of optical
rotation was not changing during all the period of measurements.
Acknowledgments
Authors are grateful to Professor L.V. Backinowsky for discus-
sions and help during preparation of this manuscript and Dr. L.O.
Kononov for help in measuring NMR spectra. This work was funded
by Ministry of Education of Ukraine; registration number of financ-
ing 0109U001355.
1.2.1. Benzyl 2-acetamido-4,6-O-benzylidene-2-deoxy-b-D-
glucopyranoside 3
References
18
546
Mp 269–270 °C, lit.18 270–271 °C; ½
a
ꢂ
ꢀ105.0 (c 1.0, Py), lit.18
a 2D5
ꢂ
ꢀ89.0 (c 0.8, Py); 1H NMR (DMSO-d6): d 7.87 (br d, 1H, JNH,2
1. Schmidt, R. R. Angew. Chem., Int. Ed. Engl. 1986, 25, 212–235.
2. Schmidt, R. R. Pure Appl. Chem. 1989, 61, 1257–1270.
3. Schmidt, R. R. In Modern Methods in Carbohydrate Chemistry; Khan, S. S. H.,
O’Neill, R. A., Eds.; Harwood, 1996; pp 20–54.
4. Tamura, J. In Carbohydrates in Chemistry and Biology; Ernst, B., Hart, G. W., Sinaÿ,
P., Eds.; Wiley-VCH, 2000; pp 179–193.
5. Lubineau, A.; Escher, S.; Alais, J.; Bonnaffe, D. Tetrahedron Lett. 1997, 38, 4087–
4090.
6. Schmidt, R. R.; Reichrath, M. Angew. Chem. 1979, 18, 466–467.
7. Schmidt, R. R.; Reichrath, M.; Moering, U. Tetrahedron Lett. 1980, 21, 3561–
3564.
8. Schmidt, R. R.; Moering, U.; Reichrath, M. Chem. Ber. 1982, 115, 39–49.
9. Tamura, J.; Schmidt, R. R. J. Carbohydr. Chem. 1995, 14, 895–911.
10. Schmidt, R. R.; Reichrath, M.; Moering, U. J. Carbohydr. Chem. 1984, 3, 67–84.
11. Morris, W. J.; Shair, M. D. Org. Lett. 2009, 11, 9–12.
12. Vauzeilles, B.; Dause, B.; Palmier, S.; Beau, J.-M. Tetrahedron Lett. 2001, 42,
7567–7570.
13. Roth, W.; Pigman, W. J. Am. Chem. Soc. 1960, 82, 4608–4611.
14. Pertel, S. S.; Gunchak, S. A.; Kakayan, E. S.; Chirva, V. Ya.; Vidal, S. Carbohydrate
Chemistry: Proven Synthetic Methods; Kovac, P., Ed.; CRC Press Taylor &
Francis Group (in press).
½
8.0 Hz, NH), 7.25–7.50 (m, 10H, Ph), 5.62 (s, 1H, PhCH), 5.31 (br
d, 1H, JOH,3 4.5 Hz, OH), 4.78 (d, 1H, PhCHb), 4.58 (d, 1H, J1,2
7.5 Hz, H-1), 4.53 (d, 1H, JCHa,CHb 12.5 Hz, PhCHa), 4.24 (dd, 1H,
J6b,5 5.0 Hz, H-6b), 3.77 (t, 1H, J6a,6b 10.0 Hz, J6a,5 10.0 Hz, H-6a),
3.54–3.70 (m, 2H, H-2, H-3), 3.46 (dd, 1H, J4,3 9.0 Hz, H-4), 3.37
(ddd, 1H, J5,4 9.0 Hz, H-5), 1.82 (s, 3H, CH3CO).
1.2.2. Benzyl 2-acetamido-4,6-O-benzylidene-2-deoxy-a-D-
glucopyranoside 4
18
546
Mp 263–264 °C, lit.18 263–264 °C; ½
a
ꢂ
+111.0 (c 1.0, Py), lit.18
½
a 2D6 +120.0 (c 1.0, Py); 1H NMR (DMSO-d6): d 7.96 (br d, 1H, JNH,2
ꢂ
7.5 Hz, NH), 7.25–7.55 (m, 10H, Ph), 5.61 (s, 1H, PhCH), 5.16 (br
d, 1H, JOH,3 5.0 Hz, OH), 4.80 (d, 1H, J1,2 3.0 Hz, H-1), 4.70 (d, 1H,
PhCHb), 4.49 (d, 1H, JCHa,CHb 12.5 Hz, PhCHa), 4.14 (ddd, 1H, J6b,6a
10.0 Hz, J6b,5 5.0 Hz, H-6b), 3.86 (ddd, 1H, J2,3 8.0 Hz, H-2), 3.72
(m, 3H, H-3, H-5, H-6a), 3.52 (ddd, 1H, J4,3 9.0 Hz, J4,5 9.0 Hz, J4,6b
9.0 Hz, H-4), 1.85 (s, 3H, CH3CO).
15. Gomez, A. M.; Lopez, J. C. In Glycoscience: Chemistry and Chemical Biology;
Fraser-Raid, B. O., Tatsuta, K., Thiem, J., Eds.; Springer: Berlin, 2001; Vol. 1, pp
467–500.
16. Houghton, R. P. Metal Complexes in Organic Chemistry; Cambridge University
Press: Cambridge, 1979. p 308.
17. Turova, N. Ya.; Novoselova, A. V. Russ. Chem. Rev. 1965, 34, 161–185.
18. Dictionary of Carbohydrates; Collins, P. M., Ed.; Chapman & Hall: London, 1998;
p 937.
1.3. General procedure for determination of optical rotations of
mixtures of anomeric alcoholates (2) in DMF solutions at ꢀ30 °C
Two mixtures of anomers of compound 1, different by composi-
tion, were prepared. Mixture A, with anomeric ratio 2:3 (a/b) (as