Selective 4,6-O-benzylidene formation of methyl α-d-mannopyranoside using 2,6-dimethylbenzaldehyde

Article abstract of DOI:10.1016/j.carres.2014.02.024

While methyl α-d-glucopyranosides and α-d-galactopyranosides selectively form 4,6-O-benzylidenes when reacted with excess benzaldehyde in the presence of acid catalyst methyl α-d-mannopyranosides does not exhibit the same selectivity because of the cis-ar

Full text of DOI:10.1016/j.carres.2014.02.024

Accepted Manuscript
Selective 4,6-O-benzylidene formation of methyl α-D-mannopyranoside using
2,6-dimethylbenzaldehyde
Louis J. Liotta, Jennifer F. Chalmers, Jessica N. Falco Marshall, Timothy E.
Ferreira, Hannah E. Mullen, Nicholas J. Pace
PII:
S0008-6215(14)00085-8
DOI:
http://dx.doi.org/10.1016/j.carres.2014.02.024
CAR 6692
Reference:
Carbohydrate Research
To appear in:
Received Date:
Revised Date:
Accepted Date:
4 January 2014
20 February 2014
22 February 2014
Please cite this article as: Liotta, L.J., Chalmers, J.F., Falco Marshall, J.N., Ferreira, T.E., Mullen, H.E., Pace, N.J.,
Selective 4,6-O-benzylidene formation of methyl α-D-mannopyranoside using 2,6-dimethylbenzaldehyde,
Carbohydrate Research (2014), doi: http://dx.doi.org/10.1016/j.carres.2014.02.024
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Selective 4,6-O-benzylidene formation of methyl α-D-mannopyranoside using
2,6-dimethylbenzaldehyde
Louis J. Liotta*, Jennifer F. Chalmers, Jessica N. Falco Marshall, Timothy E. Ferreira, Hannah
E. Mullen, Nicholas J. Pace
Stonehill College, 320 Washington Street, Easton, MA 02357
Phone: 1-508-565-1043 Fax: 1-508-565-1469 Email: lliotta@stonehill.edu
ABSTRACT
While methyl α-D-glucopyranosides and α-D-galactopyranosides selectively form 4,6-O-
benzylidenes when reacted with excess benzaldehyde in the presence of acid catalyst methyl α-
D-mannopyranosides does not exhibit the same selectivity because of the cis-arrangement of the
C2 and C3 hydroxyl groups. The selectivity for the 4,6-O-benzylidene is restored by using 2,6-
dimethylbenzaldehyde instead of benzaldehyde. In addition the excess 2,6-
dimethylbenzaldehyde is easily recovered from the reaction by extraction with petroleum ether
and can be reused without further purification. The 2,6-dimethylbenzylidene exhibits properties
similar to the unsubstituted benzylidene with regards to chemical synthesis.
1. Introduction
As we developed a general synthesis of all the possible stereoisomers of
polyhydroxylated pyrrolidines, pyrrolizidines and indolizidines, we require the 4,6-O-
benzylidene protected methyl pyranosides of different sugars including D- and L-glucose, D-
galactose, D- and L-mannose, D- and L-allose and D-altrose.1 The 4,6-benzylidenes of both
glucose isomers and the D-galactose were directly prepared in high yields from the
1

2
Scheme 1. 4,6-O-Benzylidene formation of methyl α-D-mannopyranoside
corresponding methyl pyranosides of these sugars. Both allose isomers and the altrose are easily
prepared from the corresponding isomers of the methyl pyranoside of glucose.2,3 Preparing the
4,6-O-benzylidene of both the D- and L-isomers methyl mannopyranosides in high yield has
been particularly challenging due to the formation of both the 2,3-O-benzylidene (4S and 4R)
and the 2,3,4,6-di-O-benzylidene (5S and 5R) byproducts. Our solution to this problem has been
to prepare the 2,6-dimethylbenzylidene of the methyl mannopyranosides instead of the
unsubstituted benzylidenes (Scheme 1).
2. Results and discussion
A number of methods are reported to give reasonable yields of the desired 4,6-O-
benzylidene 2 of methyl α-D-mannopyranoside.4-14 We tried a number of the reported methods
utilizing both benzaldehyde 6 and benzaldehyde dimethyl acetal 7 as the benzylidene forming
reagent. Reacting methyl α-D-mannopyranoside 1 with a 1:1 mixture of redistilled, anhydrous
benzaldehyde 6 and 98+% formic acid for two minutes at room temperature followed by
quenching with a K2CO3 solution resulted in 4,6-O-benzylidene 2 in 32% yield and the

3
appreciable yields of the 2,3-O-benzylidene 4 (10%) and the 2,3,4,6-di-O-benzylidene 5 (25%)
(Table 1, entry 1).4 The methyl α-D-mannopyranoside 1 was next reacted with benzaldehyde 6
(12.5 equiv.) in DMF containing pTsOH (0.13 equiv.) at 50˚ C for one hour.5 Analysis of the
crude product after evaporation of the DMF indicated the mixture contained a 42% yield of 4,6-
O-benzylidene 2, a 6% yield of 2,3-O-benzylidene 4 and a 48% yield of 2,3,4,6-di-O-
benzylidene 5 (entry 2). Quenching of the reaction with a solution of K2CO3 before evaporating
the DMF resulted in a 52% yield of the 4,6-O-benzylidene, a 17% yield of the 2,3-O-benzylidene
and a 18% yield of the 2,3,4,6-di-O-benzylidene (entry 3).
Reacting the methyl α-D-mannopyranoside 1 with benzaldehyde dimethyl acetal 7 (2
equiv.) in the ionic liquid 3-butyl-1-methylimidazolium tetrafluoroborate containing pTsOH (0.1
equiv.) at 80˚ C for two hours before quenching with Na2CO3 resulted in a mixture of the 4,6-O-
benzylidene 2 (29%), the 2,3-O-benzylidene 4 (1%), and the 2,3,4,6-di-O-benzylidene 5 (38%)
(Table 1, entry 4).6 Using DMF as the solvent and pTsOH (0.10 equiv.) as the catalyst with
benzaldehyde dimethyl acetal 7 (1.01 equiv.) at 60˚ C under reduced pressure to remove
methanol for 2 hours and quenching with K2CO3 produced a mixture of the 4,6-O-benzylidene 2
(47%), the 2,3-O-benzylidene 4 (15%), and the 2,3,4,6-di-O-benzylidene 5 (24%) (entry 5).7-10
Yields of the 4,6-O-benzylidene 2 of about 50% were not as high as some of those reported in
the literature7 but similar to reports by Leino et. al in which they indicated that the 2,3-O-
benzylidene compounds were the major byproducts.8-10 Our results are also similar to the results
reported in a paper by Patroni et. al., in which they investigated a number of methods to
synthesize the desired 4,6-O-benzylidene of methyl α-D-mannopyranoside.15 When using
benzaldehyde dimethyl acetal 7 in DMF with pTsOH as the catalyst they obtained a 53:8:8:16:15
mixture of the 4,6-O-benzylidene 2, the (S)-2,3-O-benzylidenes 4S, the (R)-2,3-O-benzylidene

4
4R, the (S)-2,3,4,6-di-O-benzylidene 5S and the (R)-2,3,4,6-di-O-benzylidene 5R. In order to
improve the yield of the desired 4,6-O-benzylidene 2 we tried a number of ways to selectively
cleave the 2,3-O-benzylidene of the 2,3,4,6-di-O-benzylidene 5.16,17 The methods we
investigated were not synthetically viable.
It should be noted that two promising methods to selectively prepare the 4,6-O-
benzylidene of mannopyranosides in high yields have appeared in the literature while this work
was being completed. The one method used an acid-free organic catalyst with benzaldehyde
dimethyl acetal and reported the selective formation of the 4,6-O-benzylidene in 86% yield.18
The second method reacted the fully silylated mannopyranoside with benzaldehyde and TMS-
OTf at -78˚ C in CH2Cl2 and resulted in a 92% yield of the 4,6-benzylidene.19 We did not
attempt either of these methods.
Since both in our investigation and that of Patroni it appeared that the 4,6-O-benzylidene
formed much faster than the 2,3-O-benzylidene and that the 2,3,4,6-di-O-benzylidene product
was mostly being formed from the 4,6-O-benzylidene,15 we speculated that substituted
benzaldehydes may be more selective due to steric hindrance. Molecular models and MM2
calculations indicated that methyl substitution at the 2- and 6- positions of the benzene ring
would experience very little interference in the 4,6-O-benzylidene but would experience
interference from the C2 and C3 hydrogens when forming the R isomer of the 2,3-O-benzylidene
and from the C4 hydrogen when forming the S isomer (Figure 1). This steric interference is
more pronounced when the pyranoside ring is made more ridged by first forming the 4,6-O-
benzylidene. Thus, we would predict a small amount of the 2,3-O-benzylidene may still be
observed but the formation of the 2,3,4,6-di-O-benzylidene should be substantially reduced. The
calculated streric energy of the 2,3,4,6-di-O-benzylidene using the 2,6-dimethylbenzaldehyde

5
was calculated to be 52.9 kcal/mole for the R-isomer and 51.4 kcal/mole for the S-isomer. This
compared to 44.5 kcal/mole and 43.7 kcal/mole for the R- and S-isomers respectively when
using unsubstituted benzaldehyde. The calculated energies for the 4,6-O-benzylidenes were 30.4
kcal/mole for the unsubstituted benzaldehyde and 33.8 kcal/mole for the 2,6-
dimethylbenzaldehyde.
In practice using 2,6-dimethylbenaldehyde 9 to prepare the 4,6-O-benzylidene of methyl
α-D-mannopyranoside was much more selective, giving no dibenzylidene when the reaction was
conducted and fully quenched at 50˚ C. In addition, the reaction is more environmentally
friendly than using benzaldehyde because the unreacted 2,6-dimethylbenzaldehyde can easily be
recovered by extraction with petroleum ether and reused without any purification, giving
comparable if not better yields than the new 2,6-dimethylbenzaldehyde. A similar recovery of
benzaldehyde was not possible both because benzaldehyde was more volatile and thus
evaporated during DMF removal and because benzaldehyde is more susceptible to air oxidation
forming benzoic acid.
Initially both 2-methylbenzaldehyde 8 and the 2,6-dimethylbenzyaldehyde 9 were
investigated as to their usefulness in selectively providing the 4,6-O-benzylidene of methyl α-D-
mannopyranoside. It quickly became evident that the mono-methylated benzyaldehyde provided
more selectivity than benzaldehyde but the 2,3,4,6-di-O-benzylidene product was still formed in
an appreciable yield (Table 1, entry 6). Consequently, further experimentation was continued
only with the 2,6-dimethylbenzaldehyde 9 (entries 7-15). The use of formic acid as

6
Table 1
Reaction Conditions for Benzylidene Formation of Methyl α-D-Mannopyranoside
Entry Mannoside Benzaldehyde Catalyst Solvent Temp. Time
Acid
Quenchc
Yield of Benzylideneb
Recovered
Aldehyde
Conc.a
(mole eq.)
6 (6.1)
6 (12.5)
6 (12.5)
7 (2.0)
7 (1.01)
8 (11.4)
9 (8.7)
9 (8.7)
9 (9.1)
9 (9.7)
9 (9.4)
9 (9.5)
9 (8.8)
9 (28.4)
9 (9.2)
(mole eq.)
4,6-
32%
2,3-
10%
2,3,4,6-
1.
1.60
HCOOH HCOOH
(16.5)
rt
2 min
1 hr
(b)
(a)
(b)
(d)
(b)
(b)
(b)
(b)
(a)
(c)
(c)
(c)
(b)
(b)
(b)
25%
48%
18%
38%
24%
17%
0%
0%
0%
2.
0.629
0.665
0.515
0.443
0.632
1.03
pTsOH
(0.13)
DMF
50˚ C
50˚ C
42%
52%
29%
47%
64%
29%
0%
6%
17%
1%
3.
pTsOH
(0.13)
pTsOH [bmim]BF4 80˚ C 2 hrs
DMF
1 hr
0%
4.
---
(0.01)
pTsOH
(0.10)
pTsOH
(0.13)
5.
DMF
DMF
60˚ C 2 hrs
15%
16%
0%
---
6.
50˚ C
1 hr
6 hrs
16 hrs
---
7.
HCOOH HCOOH
HCOOH HCOOH
rt
rt
---
8.
1.03
0%
0%
---
9.
0.644
0.648
0.667
0.658
0.672
0.336
0.681
pTsOH
(0.13)
pTsOH
(0.13)
pTsOH
(0.13)
DMF
DMF
DMF
50˚ C 2 hrs
50˚ C 1 hr
49%
60%
60%
29%
67%
(64%) (4%)
67% 5%
(66%) (4%)
47% 6%
0%
50%
5%
82%
71%
59%
73%
82%
75%
69%
10.
11.
12.
13.
14.
15.
4%
50˚ C 23 hrs
50˚ C 4 hrs
5%
4%
pTsOH
(0.13)
pTsOH
(0.13)
DMF/
sieves
DMF
9%
0%
50˚ C
50˚ C
80˚ C
1 hr
1 hr
1 hr
4%
0%
pTsOH
(0.26)
pTsOH
DMF
DMF
0%
0%
(0.13)
a mmoles of methyl α-D-mannopyranoside per milliliter of solvent, byields of individual compounds in
mixture determined by NMR isolated yield in parentheses, cacid quenching conditions: (a) none, (b) K2CO3
solution, (c) K2CO3 solid, (d) Na2CO3 solution.
both the catalyst and the solvent under various conditions displayed the expected selectivity for
the 4,6-O-benzylidene over the 2,3,4,6-di-O-benzylidenes but yields of the desired 4,6-O-
benzylidene were relatively low (entry 7). The desired 4,6-O-benzylidene also appeared to be
unstable under the reaction conditions since after 6 hours the amount of 4,6-benzylidene
produced decreased, being unobservable after 16 hours (entry 8). Consequently, conditions were
sought in which a catalytic amount of acid would be used in a solvent capable of solubilizing the

7
starting mannopyranoside and the 2,6-dimethylbenzaldehyde as opposed to having a very large
excess of acid as was the case when using formic acid as both the solvent and the catalyst.
The common benzylidene forming conditions of pTsOH as the catalyst in DMF met these
conditions and proved to be both selective and much higher yielding. Proper quenching of the
reaction before proceeding to the workup appears to be critical. Removing most of the DMF
under vacuum at 80˚ C before neutralizing the acid catalyst resulted in a large amount of the
2,3,4,6-di-O-benzylidene product (entry 9). Neutralizing the acid with dry K2CO3 before
evaporating off the DMF resulted in a 4-5% yield of the 2,3,4,6-di-O-benzylidene (entries 10-11)
while neutralizing the acid with a K2CO3 solution resulted in complete elimination of the
dibenzylidene (entries 12-15). Increasing the reaction time did not affect the yield of the 4,6-O-
benzylidene (entry 10 versus 11). Likewise a higher concentration of aldehyde starting material
did not affect the yield of the desired benzylidene (entry 13 versus 14). An attempt to drive the
reaction by conducting the it at a higher temperature (80˚ C) or by removing water with
molecular sieves actually resulted in less of the desired 4,6-O-benzylidene product and more the
2,3-O-benzylidene byproduct (entries 15 and 12).
The best results were obtained with using 8.8 mole equivalents of the 2,6-
dimethylbenzyaldehyde and 0.13 mole equivalents of p-toluenesulfonic acid in DMF at 50˚ C for
one hour and quenching with a solution of K2CO3 (entry 13). After evaporating off most of the
DMF the reaction mixture is partitioned between petroleum ether and water. The unreacted 2,6-
dimethylbenzyaldehyde is recovered in pure form simply be evaporating off the petroleum ether.
This recovered aldehyde gave identical if not better yields when used in the reaction than the
original commercial aldehyde. The methyl 4,6-O-(2,6-dimethylbenzylidene)-α-D-

8
mannopyranoside 3 was extracted from the aqueous layer with dichloromethane and purified by
chromatography in 64% yield.
Scheme 2. Utility of 2,6-dimethylbenzylidene as a protecting group
The 2,6-dimethylbenzylidene exhibited chemistry identical to the unsubstituted
benzylidene when the methyl 4,6-O-(2,6-dimethylbenzylidene)-α-D-mannopyranoside 3 was
substituted for the methyl 4,6-O-benzylidene-α-D-glucopyranoside in our synthetic scheme
developed for synthesizing polyhydroxylated pyrrolizidines.1 Thus the C2 and C3 hydroxyl
groups were readily protected as benzyl ethers by treating with KOH and benzyl bromide in
toluene,20 yielding the desired fully protected mannopyranoside 10 in 98% yield (Scheme 2).
This compared to a 91% yield for the corresponding reaction for the unsubstituted 4,6-O-
benzylidene of the glucopyranoside.1 The 2,6-dimethylbenzylidene group was then readily
hydrolyzed by treating with iodine in wet methanol.20 The desired 4,6-deprotected
mannopyranoside 11 was isolated in 97% yield (Scheme 2) as compared to 100% yield for the
hydrolysis of unsubstituted benzylidene of glucopyranoside.1
One of the most useful aspects of 4,6-benzylidene protected pyranosides is that they can
be opened regioselectively under different reduction conditions to give a free hydroxyl group at
either the C4 or C6 position.8,21-29 To further investigate the utility of the 2,6-
dimethylbenzylidene protecting group, the benzyl protected 4,6-O-(2,6-dimethylbenzylidene)-α-
D-mannopyranoside 10 was subjected to two different reductive ring opening conditions, one
selective for the C4 hydroxyl (NaBH3CN, HCl)21 and one for the C6 hydroxyl (BH3•THF, TMS-

9
OTf)22 (Scheme 3). Under these two different conditions the 2,6-dimethylbenzylidene exhibited
reactivity similar to the unsubstituted benzylidene. With the NaBH3CN and HCl, the
mannopyranoside with the C4 hydroxyl group 13 was isolated in 70% yield. This compares to
reported yields when starting with the unsubstituted benzylidene of 83%. With BH3•THF, TMS-
OTf the mannopyranoside with the C6 hydroxyl group 12 was isolated in 91% yield. This
compares to a reported yield of the C6 hydroxyl compound of 85% when starting with the
unsubstituted benzylidene.
Scheme 3. Regioselective opening of 2,6-dimethylbenzylidene: (a) NaBH3CN, HCl, 0˚ C, 1.5
hrs; (b) BH3•THF, TMS-OTf, rt, 1.5 hrs.
A more selective and environmentally friendly way to prepare the 4,6-O-benzylidene of
methyl α-D-mannopyranoside in good yield was developed using 2,6-dimethylbenzaldehyde as
opposed to benzaldehyde. Notably, this method gave no dibenzylidene byproduct, which is a
major problem typically encountered when using the unsubstituted benzaldehyde. An additional
advantage of our method is that the conditions allowed for recovery and reuse of the unreacted
2,6-dimethylbenzaldehyde, which was used in excess. The 2,6-dimethylbenzylidene exhibits
similar reactivity towards hydrolysis and selective reduction as the unsubstituted benzylidene.
3. Experimental
3.1 General methods

10
Thin layer chromatography was performed using Merck silica gel 60 F254 plates. Plates were
visualized using 20% ethanolic phosphomolybdic acid. Flash column chromatography was
performed using Biotage SNAP KP-Sil columns. Crude and purified products were allowed to
dry on a vacuum line attached to a Welch 1400 DuoDeal vacuum pump overnight to ensure
complete dryness. All NMR spectra was collected using the Bruker Avance 300 MHz
Spectrometer in CDCl3 containing 0.03% TMS as in internal standard. Infrared spectroscopy
was performed on a ThermoNicolet Avatar 370 FT-IR spectrophotometer. Optical rotations were
collected using a Jasco P-1010 polarimeter. The exact mass measurements were performed
using JEOL AccuTOF DART mass spectrometer (Boston College 2609 Beacon Street Chestnut
Hill MA, USA). 2,6-Dimethylbenzyaldehyde was from Combi-Blocks Inc. Toluene was
acquired from the mBraun solvent purification system. DMF from Acros was classified as extra
dry and stored over molecular sieves. All other chemicals were from either Fisher Scientific or
Sigma-Aldrich.
3.2 Experimental Procedures
3.2.1 Methyl 4,6-O-(2,6-dimethylbenzylidene)-α-D-mannopyranoside (3) Methyl α-D-
mannopyranoside 1 (1.043 g, 5.38 mmol) and p-TsOH•H2O (0.133 g, 0.699 mmol) were dried
over night in a vacuum oven at 60˚ C. After removing from the oven, the methyl α-D-
mannopyranoside 1 was suspended in DMF (8 mL) while stirring under N2. The 2,6-
dimethylbenzaldehyde (6.38 mL, 47.6mmol) was added followed by the dried p-TsOH. The
reaction was heated to 50˚ C while stirring under N2 for 1 h. After cooling to rt, the reaction was
quenched by adding K2CO3 (0.080 mg, 0.58 mmol) as a solution in H2O (2 mL). The DMF was
mostly evaporated off under vacuum at 80˚ C. The resulting mixture was mixed with petroleum

11
ether (75 mL) and H2O (50 mL). After vigorous mixing the aqueous layer was removed and the
petroleum ether extracted with H2O (2 x 50 mL). The petroleum ether was dried over Na2SO4,
decanted and evaporated in vacuo to yield pure 2,6-dimethybenzaldehyde (4.830 g, 36.04 mmol,
82%). The combined aqueous layers were extracted with CH2Cl2 (3 x 75 mL). The CH2Cl2
extracts were combined, dried over Na2SO4, decanted and evaporated to an oil. The crude
product was chromatographed through silica gel first using pure CH2Cl2 followed by 40:1
CH2Cl2/CH3OH and then 20:1 CH2Cl2/CH3OH. Fractions were analyzed by TLC in 20:1
CH2Cl2/CH3OH. This procedure resulted in the isolation of pure 3 as a clear oil (Rf=0.14, 1.065
1
g, 3.44 mmol, 64%). [α]23D 70.4 (c 1.53, CH2Cl2). H-NMR (300 MHz, CDCl3): (m, 3 H, Ar-
H), 5.92 (s, 1 H, benzylidene-H), 4.67 (d, 1 H, J=1.3 Hz, C1-H), 4.27 (m, 1 H, C6-H), 3.95 (m, 1
H, C3-H), 3.91 (m, 1 H, C2-H), 3.86-3.76 (m, 3 H, C5-H, C4-H, C6-H), 3.39 (s, 3 H, OCH3),
3.00 (br, 1 H, C3-OH), 2.91 (br, 1 H, C2-OH), 2.47 (s, 6 H, Ar-CH3). 13C NMR (75 MHz,
CDCl3): 136.93 (C, Ar), 132.74 (C, Ar), 129.14 (CH, Ar), 128.95 (CH, Ar), 101.43 (CH, C1)
101.31 (CH, ArCHO2), 79.32 (CH, C4), 70.80 (CH, C2), 69.23 (CH2, C6), 68.61 (CH, C3),
62.92 (CH, C5), 55.24 (CH3, OCH3) 20.53 (CH3, Ar-CH3). IR: 3415 (m, br), 2930 (m), 1597
(w), 1468 (w), 1444 (m), 1377 (w), 1349 (w), 1276 (w), 1201 (m), 1132 (s), 1106 (s), 1087 (s),
1067 (s), 1033 (s), 997 (m), 977 (s), 962 (s), 914 (w), 862 (w), 801 (w), 774 (m), 736 (m), 684
(m), 651 (w), 628 (w) cm-1. HRMS calculated for C16H23O6 (M+ + H) 311.14946, found
311.14971.
3.2.2 Methyl 2,3-di-O-benzyl-4,6-(2,6-dimethylbenzylidene)-α-D-mannoopyranoside (10)
Methyl 4,6-O-(2,6-dimethylbenzylidene)-α-D-mannopyranoside 3 (1.05 g, 3.39 mmol), KOH
(3.35g, 59.8 mmol), and benzyl bromide (3.4 mL, 28.5 mmol) were suspended in toluene (140
mL) and the reaction was heated to reflux under nitrogen. The reaction was monitored by TLC

12
(5:1:1 hexane/EtOAc/CH2Cl2) for disappearance of 3 (Rf=0.09) and the appearance of 6
(Rf=0.64) at which point (approx. 6 h) the reaction was allowed to cool to rt. Toluene (400 mL)
was added to the reaction mixture and the mixture was washed with water (2 x 500 mL). The
toluene was dried with Na2SO4, decanted and evaporated to an oil. The crude product was
chromatographed through silica gel using hexane/EtOAc/CH2Cl2 first at an 20:1:1 ratio, then a
10:1:1 ratio. Fractions were analyzed by TLC in 10:1:1 hexane/EtOAc/CH2Cl2. This procedure
resulted in the isolation of pure 10 as a clear oil (Rf=0.42, 1.63 g, 3.33 mmol, 98%). [α]23D 9.2 (c
1
3.60, CH2Cl2). H-NMR (300 MHz, CDCl3): 7.39-6.96 (m, 13 H, Ar-H), 5.94 (s, 1 H,
benzylidene-H), 4.79 (d, 1 H, J=12.3 Hz, CHHPh), 4.72-4.67 (m, 3 H, C1-H, CH2Ph), 4.57 (d, 1
H, J=12.6 Hz, CHHPh), 4.20 (m, 2 H, C4-H, C6-H), 3.89 (dd, 1 H, J=3.2, 9.9 Hz, C3-H), 3.81-
3.78 (m, 3 H, C6-H, C5-H, C2-H), 3.29 (s, 3 H, OCH3), 2.47 (s, 6 H, Ar-CH3). 13C NMR (75
MHz, CDCl3): 138.80 (C, Ar), 138.30 (C, Ar), 137.03(C, Ar), 133.53 (C, Ar), 129.09 (CH, Ar),
128.68 (CH, Ar), 128.52 (CH, Ar), 128.35 (CH, Ar), 128.16 (CH, Ar), 127.89 (CH, Ar), 127.56
(CH, Ar), 127.51 (CH, Ar), 101.29 (CH, ArCHO2), 100.65 (CH, C1), 79.70 (CH, C4), 76.58
(CH, C2), 76.17 (CH, C3), 73.69 (CH2, OBn), 72.95 (CH2, OBn), 69.35 (CH2, C6), 64.25 (CH,
C5), 55.11 (CH3, OCH3) 20.78 (CH3, Ar-CH3). IR: 3087 (m), 3063 (s), 3029 (s), 2905 (s), 2734
(w), 1952 (w), 1597 (m), 1589 (m), 1497 (s), 1454 (s), 1373 (s), 1318 (s), 1279 (s), 1200 (s),
1170 (s), 1054 (s), 998 (s), 958 (s), 910 (s), 875 (m), 855 (m), 798 (s), 773 (s), 736 (s), 698 (s),
680 (s), 651 (w), 630 (w) cm-1. HRMS calculated for C30H35O6 (M+ + H) 491.24336, found
491.24270.
3.2.3 Methyl 2,3-di-O-benzyl-α-D-mannopyranoside (11) Methyl 2,3-di-O-benzyl-4,6-(2,6-
dimethylbenzylidene)-α-D-mannoopyranoside 10 (1.29 g, 2.63 mmol), I2 (0.51 g, 1.99 mmol)
and H2O (0.5 g, 27.8 mmol) were dissolved in methanol (190 mL) and the reaction was heated to

13
reflux. The reaction was monitored by TLC (10:1:1 hexane/EtOAc/CH2Cl2) for the
disappearance of 6 (Rf = 0.42). When the reaction was complete (approximately 3.5 h), it was
cooled to rt. Ten percent Na2S2O3 was added (32 mL), and the solution was evaporated to an oil
which was subsequently dissolved in water (225 mL) and extracted with CH2Cl2 (2 x 225 mL).
The organic layers were combined, dried with Na2SO4, decanted and evaporated to dryness to
result in pure 11 as a clear oil (0.953 g, 2.55 mmol, 97% yield). [α]23D -9.48 (c 1.69, CH2Cl2).
1H-NMR (300 MHz, CDCl3): 7.36-7.23 (m, 10 H, Ar-H), 4.73 (d, 1 H, J=1.5 Hz, C1-H), 4.65
(s, 2 H, CH2Ph), 4.58 (d, 1 H, J=11.8 Hz, CHHPh), 4.47 (d, 1 H, J=11.8 Hz, CHHPh), 4.02 (td, 1
H, J=2.4, 9.6 Hz, C4-H), 4.06-3.99 (m, 2 H, C6-CH2), 3.77 (dd, 1 H, J=1.6, 2.9 Hz, C2-H), 3.69
(dd, 1 H, J=3.0, 9.5 Hz, C3-H), 3.58 (dt, 1 H, J=4.4, 9.2 Hz, C5-H), 3.31 (s, 3 H, OCH3), 2.81
(br, 1 H, C4-OH), 2.49 (br, 1 H, C6-OH). 13C NMR (75 MHz, CDCl3): 138.10 (C, Ar),
138.07(C, Ar), 128.53 (CH, Ar), 128.42 (CH, Ar), 127.91 (CH, Ar), 127.86 (CH, Ar), 127.79
(CH, Ar), 127.75 (CH, Ar), 99.37 (CH, C1), 79.73 (CH, C3), 73.90 (CH, C2), 72.84 (CH2, OBn),
72.24 (CH, C5), 71.78 (CH2, OBn), 67.27 (CH, C4), 62.80 (CH2, C6), 54.90 (CH3, OCH3). IR:
3439 (s,br), 3088 (m), 3063 (m), 3030 (m), 2913 (s), 2834 (m), 1605 (w), 1586 (w), 1497 (s),
1454 (s), 1367 (s), 1323 (m), 1251 (m), 1198 (s), 1053 (s), 968 (s), 910 (m), 804 (m), 737 (s),
699 (s) cm-1. HRMS calculated for C21H30NO6 (M+ + NH4) 392.20731, found 392.20853.
3.2.5 Methyl 2,3-di-O-benzyl-4-O-(2,6-dimethylbenzyl)-α-D-mannopyranoside (12) Methyl
2,3-di-O-benzyl-4,6-(2,6-dimethylbenzylidene)-α-D-mannoopyranoside 10 (0.134 g, 0.274
mmol) was dissolved in CH2Cl2 (3 mL) followed by the addition of a 1 M solution of BH3 in
THF (0.55 mL, 0.55 mmol). The resulting solution was stirred at room temperature under N2
while TMS-OTf (0.005 mL, 0.026 mmol) was added. The reaction was monitored by TLC
(10:3:1 hexane/EtOAc/CH2Cl2) for the disappearance of 10 (Rf = 0.57). When the reaction was

14
complete (approximately 1.5 h) it was quenched by the addition of Et3N (0.150 mL, 1.08 mmol)
followed by the careful addition of CH3OH until H2 was no longer evolved. The solution was
concentrated under vacuum. To assist in evaporating the excess Et3N, the resulting residue was
three times dissolved in CH3OH (5 mL) and evaporated under vacuum. The crude product was
chromatographed through silica gel first using 10:3:1 hexane/EtOAc/CH2Cl2 and then 10:4:1
hexane/EtOAc/CH2Cl2. Fractions were analyzed by TLC in 10:3:1 hexane/EtOAc/CH2Cl2. This
procedure resulted in the isolation of pure 12 as a clear oil (Rf=0.13, 0.123 g, 0.263 mmol, 91%).
1
[α]23.5D 14.2 (c 4.32, CH2Cl2). H-NMR (300 MHz, CDCl3): 7.39-7.24 (m, 10 H, Ar-H), 7.10-
6.97 (m, 3 H, (CH3)2Ar-H), 5.00 (d, 1 H, J=10.3 Hz, CHHPh), 4.74-4.58 (m, 6 H, C1-H, CHHPh,
CH2Ph, CH2Ph), 3.98-3.92 (m, 2 H, C3-H, C4-H), 3.82 (t, 1 H, J=2.4 Hz, C2-H), 3.81-3.70 (m, 2
H, C6-H, C6-H), 3.56 (m, 1 H, C5-H), 3.30 (s, 3 H, OCH3), 2.34 (s, 6 H, Ar-CH3), 1.94 (s, 1 H,
C6-OH). 13C NMR (75 MHz, CDCl3): 138.43 (C, Ar), 138.27 (C, Ar), 138.01(C, Ar), 134.52 (C,
Ar), 128.34 (CH, Ar), 128.33 (CH, Ar), 128.21 (CH, Ar), 128.10 (CH, Ar), 127.74 (CH, Ar),
127.69 (CH, Ar), 127.49 (CH, Ar), 127.39 (CH, Ar), 99.17 (CH, C1), 80.69 (CH, C3), 74.22
(CH, C4), 74.19 (CH, C2), 72.86 (CH2, OBn), 71.94 (CH, C5), 71.54 (CH2, OBn), 69.17 (CH2,
OBn), 62.56 (CH2, C6), 54.79 (CH3, OCH3) 19.72 (CH3, Ar-CH3). IR: 3484 (m,br), 3064 (w),
3030 (m), 2914 (s), 1589 (w), 1497 (m), 1469 (m), 1454 (s), 1396 (m), 1363 (s), 1321 (m), 1267
(m), 1197 (m), 1072 (s), 970 (s), 907 (w), 846 (w), 827 (w), 803 (w), 773 (m), 737 (s), 698 (s),
600 (m), 485 (m), 471 (w), 462 (m), 426 (w), 417 (w), 408 (w), 401 (w) cm-1. HRMS calculated
for C32H40NO6 (M+ + NH4) 510.28556, found 510.28447.
3.2.5 Methyl 2,3-di-O-benzyl-6-O-(2,6-dimethylbenzyl)-α-D-mannopyranoside (13) Methyl
2,3-di-O-benzyl-4,6-(2,6-dimethylbenzylidene)-α-D-mannoopyranoside 10 (0.107 g, .219
mmol) was dissolved in a 1.0 M solution of NaBH3CN in THF. A few grains of methyl orange

15
were added to the solution. The yellow solution was cooled on ice while stirring under N2 for 15
min. A saturated solution of HCl in ether was slowly added to the reaction until the color of the
solution remained pink for five minutes (about 9 mL). The mixture was stirred at 0˚ C for 10
min and then diluted with an equal volume of CH2Cl2. After filtering through Celite, the solution
was washed sequentially with saturated NaHCO3, brine, and water. The organic layer was dried
over anhydrous Na2SO4, decanted, and evaporated under reduced pressure to a clear oil. The
crude product was chromatographed through silica gel first using 10:2:1 hexane/EtOAc/CH2Cl2
and then 10:3:1 hexane/EtOAc/CH2Cl2. Fractions were analyzed by TLC in 10:3:1
hexane/EtOAc/CH2Cl2. This procedure resulted in the isolation of pure 13 as a clear oil
1
(Rf=0.26, 0.0753 g, 0.153 mmol, 70%). [α]23.5D -4.8 (c 2.51, CH2Cl2). H-NMR (300 MHz,
CDCl3): 7.32-7.21 (m, 10 H, Ar-H), 7.09-6.97 (m, 3 H, Ar-H), 4.74 (d, 1 H, J=1.5 Hz, C1-H),
4.70-4.48 (m, 6 H, CH2Ph, CH2Ph, CH2Ph), 4.20 (td, 1 H, J=9.39, 1.53 Hz, C4-H), 3.82-3.74 m,
3 H, C6-H, C6-H, C2-H), 3.73-3.65 (m, 2 H, C3-H, C5-H), 3.32 (s, 3 H, OCH3), 2.59 (d, 1 H,
J=1.8 Hz, C4-OH), 2.38 (s, 6 H, Ar-CH3). 13C NMR (75 MHz, CDCl3): 138.35 (C, Ar), 138.29
(C, Ar), 138.04(C, Ar), 134.17 (C, Ar), 128.49 (CH, Ar), 128.37 (CH, Ar), 128.16 (CH, Ar),
127.98 (CH, Ar), 127.90 (CH, Ar), 127.74 (CH, Ar), 127.69 (CH, Ar), 99.27 (CH, C1), 79.69
(CH, C3), 74.04 (CH, C2), 72.75 (CH2, OBn), 71.92 (CH2, OBn), 71.47 (CH, C5), 70.82 (CH2,
C6), 68.31 (CH, C4), 68.03 (CH2, OBn), 54.91 (CH3, OCH3) 19.77 (CH3, Ar-CH3). IR: 3466
(m,br), 3063 (w), 3029 (m), 2910 (s), 1496 (m), 1468 (m), 1454 (m), 1362 (m), 1324 (m), 1284
(m), 1197 (m), 1058 (s), 967 (m), 908 (w), 804 (w), 772 (m), 737 (s), 698 (s), 598 (m), 500 (m),
488 (m), 479 (m), 466 (s), 455 (m), 450 (m), 429 (w), 418 (m), 401 (m) cm-1. HRMS calculated
for C32H40NO6 (M+ + NH4) 510.28556, found 510.28659.

16
Acknowledgment
This material is based upon work partially supported by the National Science Foundation under
Grant No. 0610509. We would like to thank Marek Domin of the Mass Spectrometry Center at
Boston College for running high resolution mass spectroscopy on our samples.
References
1. Sletten, E. M., Liotta, L. J. J. Org. Chem. 2006, 71 1335-1343.
2. Qin, Huiping, Q.; Grindley. T. B. Carbohydr. Chem. 1996, 15, 95-108
3. Chang, C. T.; Yu, H.; Elchert, B. Tetrahedron Lett. 2001, 42, 7019-7023.
4. Kanie, O.; Takeda, T.; Hada, N.; Ogihara, Y. J. Carbohydr. Chem. 1991, 10, 561-581.
5. Bhattacharyya, T.; Basu, S. Indian J. Chem. 1996, 35B, 397-398.
6. Zhang, J.; Ragauskas, A. J. Carbohydr. Res. 2005, 340, 2812-2815.
7. Jalsa, N. K.; Singh, G. Tetrahedron: Asymmetry 2009, 20, 867-874.
8. Ekholm, F. S.; Arda, A.; Eklund, P.; Andre, S.; Gabius, H.-J.; Jimenez-Barbero, J.; Leino, R.
Chem. -- Eur. J. 2012, 18, 14392-14405.
9. Polakova, M.; Roslund, M. U.; Ekholm, F. S.; Saloranta, T.; Leino, R. Eur. J. Org. Chem.
2009, 870-888.
10. Ekholm, F. S.; Polakova, M.; Pawlowicz, A. J.; Leino, R. Synthesis 2009, 567-576.
11. Kumar, V.; Gauniyal, H. M.; Shaw, A. K. Tetrahedron: Asymmetry 2007, 18, 2069-2078.
12. Sau, A.; Misra, A. K. J. Carbohydr. Chem. 2011, 30, 41-46.
13. Mukhopadhyay, B.; Russell, D. A.; Field, R. A. Carbohydr. Res. 2005, 340, 1075-1080.
14. Tatina, M.; Yousuf, S. K.; Mukherjee, D. Org. Biomol. Chem. 2012, 10, 5357-5360.
15. Patroni, J. J.; Stick, R. V.; Skelton, B. W.; White, A. H. Aust. J. Chem. 1988, 41, 91-102.
16. Anwer, M. L.; Spatola, A. F. Synthesis 1980, 929-932.

17
17. Lipták, A.; Imre, J.; Harangi, J.; Nánási, P. Tetrahedron 1982, 38, 3721-3727.
18. Geng, Y.; Faidallah, H. M.; Albar, H. A.; Mhkalid, I. A.; Schmidt, R. R. Eur. J. Org. Chem.
2013, 7035-7040.
19. Patil, P. S.; Lee, C.-C.; Huang, Y.-W.; Zulueta, M. M. L.; Hung, S.-C. Org. Biomol. Chem.
2013, 11, 2605-2612.
20. Dechaux, E.; Savy, P.; Bouyain, S.; Monneret, C.; Florent, J-C. J. Carbohydr. Chem. 2000,
19, 485-501.
21. Maity, S. K.; Patra, A.; Ghosh, R. Tetrahedron 2010, 66, 2809-2814.
22. Kalikanda, J.; Li, Z. Carbohydr. Res. 2011, 346, 2380-2383.
23. Debenham, S. D.; Toone, E. J. Tetrahedron: Asymmetry 2000, 11, 385-387.
24. Tani, S.; Sawadi, S.; Kojima, M.; Akai, S.; Sato, K. Tetrahedron Lett. 2007, 48, 3103-3104.
25. Tanaka, N.; Ogawa, I.; Yoshigase, S.; Nokami, J. Carbohdr. Res. 2008, 343, 2675-2679.
26. Mikami, T.; Asano, H.; Mitsunobu, O. Chem. Lett. 1987, 2033-2036.
27. Shie, C.-R.; Tzeng, Z.-H.; Kulkarni, S. S.; Uang, B.-J.; Hsu, C.-Y.; Hung, S.-C. Angew.
Chem. Int. Ed. 2005, 44, 1665 –1668.
28. Panchadhayee, R.; Misra, A. K. Synlett 2010, 1193-1196.
29. Zhang, Y.-J.; Dayoub, W.; Chen, G.-R.; Lemaire, M. Eur. J. Org. Chem. 2012, 1960-1966.

18
Figure 1: Steric hindrance in 2,3,4,6-di-O-benzylidene of methyl α-D-mannopyranoside
resulting from methyl groups on the benzene.

19
Graphical abstract

20
Highlights
• A method to protect the C4 and C6 hydroxyls of α-mannopyranoside is provided.
• The use of 2,6-dimethylbenzaldehyde in benzylidene formation is investigated.
• Formation of benzylidenes using 2,6-dimethylbenzaldehyde is more selective.
• 2,6-dimethylbenzaldehyde is compared to benzaldehyde for benzylidene
formation.