1,3-syn-Diaxial Repulsion of Typical Protecting Groups Used in Carbohydrate Chemistry in 3-O-Substituted Derivatives of Isopropyl d -Idopyranosides

Article abstract of DOI:10.1021/acs.joc.7b01167

The strength of 1,3-syn-diaxial repulsion was evaluated for main types of protecting groups (alkyl, silyl, and acyl) usually used in carbohydrate chemistry. As molecular probes for this study, derivatives of isopropyl 2-O-benzyl-4,6-O-benzylidene-α-d-idop

Full text of DOI:10.1021/acs.joc.7b01167

Subscriber access provided by UNIV OF NEWCASTLE
Article
1,3-syn-Diaxial repulsion of typical protecting groups used in carbohydrate
chemistry in 3-O-substituted derivatives of isopropyl D-idopyranosides
Bozhena Sergeyevna Komarova, Alexey G. Gerbst, Anastasiia M. Finogenova,
Andrey S. Dmitrenok, Yury Evgenievich Tsvetkov, and Nikolay E. Nifantiev
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.7b01167 • Publication Date (Web): 27 Jul 2017
Downloaded from http://pubs.acs.org on July 28, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
The Journal of Organic Chemistry is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 40
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
1,3-syn-Diaxial repulsion of typical protecting groups used in
carbohydrate chemistry in 3-O-substituted derivatives of isopropyl D-
idopyranosides
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Bozhena S. Komarova, Alexey G. Gerbst, Anastasiia M. Finogenova, Andrey S. Dmitrenok, Yury
E. Tsvetkov, Nikolay E. Nifantiev*
Laboratory of Glycoconjugate Chemistry, N.D. Zelinsky Institute of Organic Chemistry, Russian
Academy of Sciences, Leninsky Prospect 47, 119991 Moscow, Russia
*Corresponding author. Tel, Fax: +7-499-135-87-84
nen@ioc.ac.ru
ABSTRACT: The strength of 1,3-syn-diaxial repulsion was evaluated for main types of protecting
groups (alkyl, silyl and acyl) usually used in carbohydrate chemistry. As molecular probes for this
study, derivatives of isopropyl 2-O-benzyl-4,6-O-benzylidene-α-D-idopyranoside bearing allyl,
acetyl and tert-butyldiphenylsilyl (TBDPS) protecting groups at O-3 were prepared from p-
O
4
methoxyphenyl D-galactopyranoside. The equilibrium between S2 and C1 conformations in these
compounds was investigated using 3JH,H and 3JС,H coupling constants that were determined from 1D
1H-NMR and 2D J-resolved HMBC spectra in various solvents. The analysis of the corresponding
coupling constants calculated using DFT/B3LYP/pcJ-1 approximation applied to conformations
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 40
1
2
3
4
5
6
7
8
9
optimized at DFT/B3LYP/6-311++G** level supported the investigation. Proportions of conformers
in the equilibrium revealed the highest repulsion between the 3-allyloxy group and the isopropoxy
aglycon and its dependence on the solvent polarity. Differences in the conformational behavior of 3-
O-allyl and 3-O-acetyl-α-D-idopyranoside derivatives complied with the notion that higher electron
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
density on O-3 increased 1,3-syn-diaxial repulsion. 3-O-TBDPS derivative existed mainly in C1
conformation. The attenuation of the 1,3-syn-diaxial repulsive interaction indicates that TBDPS has
stereoelectronic properties that may have significance in context of fixing unnatural pyranoside
conformation with the help of silyl groups but have been disregarded until now.
Introduction
Protecting groups have a profound effect on the result of glycosylation, a key reaction of
oligosaccharide synthesis.1 We have long been studying so-called remote anchimeric assistance, its
manifestations2-4 and possibility to adjust it for the targeted stereoselective oligosaccharide
synthesis.1 A puzzling fact found during our studies is that the anchimeric assistance by remote acyl
groups comes into effect in some cases and almost does not appear in others. For example, presence
of potentially participating groups at O-3 and O-4 of fucosyl donors5 allows for stereoselective
preparative synthesis of 1,2-cis-fucosides, while an acyl group at O-3 in glucosyl donors has almost
no influence6 on stereochemistry of glucosylations.
To find a root cause of this discrepancy, we performed analysis of putative conformational
equilibrium of oxacarbenium ions derived from a 3-O-acetylated glucosyl donor (Scheme 1a) in a
manner how it is usually done for predicting selectivity of glycosylations with substituted
pyranosydes.7-10 It can be seen that benzyloxy substituents at C-2 and C-4 in conformer B can
undergo destabilizing 1,3-syn-diaxial repulsion that might preclude the formation of bicyclic
stabilized cation C (Scheme 1a). This explains the lack of α-directing influence of potentially
participating acyl groups at O-3 of the glucoside ring. An additional argument in favor of this
ACS Paragon Plus Environment

Page 3 of 40
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
explanation is the behavior of 4,6-O-benzylidene-protected mannopyranosyl donors, in which 1,3-
syn-diaxial interaction between substituents at C-2 and C-4 cannot occur. These donors exhibit
excellent β-selectivity when they bear a benzyl group at O-3, but become completely α-selective
upon replacement of this benzyl group with a potentially participating benzoyl one.11
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The 1,3-syn-diaxial destabilizing interactions were involved for the explanation12,13 of
glycosylation selectivity with diverse pyranosyl donors with flexible conformations. In these works,
Woerpel and Marel pointed at possible development of a disfavoring 1,3-syn-diaxial interaction
3
between alkoxy substituents at C-2 and C-4 in H4 conformation of the oxacarbenium ion with the
gluco configuration. The energy of this disfavoring 1,3-syn-diaxial interaction reaches 3 kcal/mol in
1,3-dimethoxycyclohexane.14 However, there is no experimental data that would clearly show the
effect of 1,3-syn-diaxial repulsion in case of typical carbohydrate protecting groups. This work
represents an attempt to assess the impact of this repulsion between O-3 bearing conventional
protecting groups and a bulky alkoxy aglycon on the conformational equilibrium of pyranosides on
the example of frequently used acetyl, allyl and tert-butyldiphenylsilyl protections in α-D-
idopyranoside derivatives 2a-c as model compounds (Scheme 1b). The idopyranose ring is known15-
18
for its conformational mobility. α-Idopyranose has four axial oxygens at C-2, C-3, C-4 and C-1,
whose destabilizing 1,3-syn-diaxial interactions are outweighed by anomeric effect of axially
directed oxygen at C-1 and equatorial orientation of the alkoxymethylene group at C-5.
Conformational behavior of unprotected idopyranosides has been extensively investigated in context
of studies of heparin and its analogues.19 An array of protected and partially protected
idopyranosides was studied by Angyal15,16 and Paulsen.17,18
Conformational features of idopyranosides along with their NMR coupling constants have
been well established, and those data served us as a starting point in the present research. Recently,20
an efficient method was reported for the preparation of 4,6-O-benzylidene-protected D-
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 40
1
2
3
4
5
idopyranosides from corresponding 4,6-O-benzylidene-protected D-galactopyranosides. Taking
advantage of the availability of 4,6-O-benzylidene-protected D-idopyranosides, α-isopropyl idosides
6
7
8
9
2a-c along with their β-isopropyl counterparts 12a-c (Scheme 2) were chosen as suitable models for
our investigation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
SCHEME 1. (a) Putative 1,3-syn-diaxial interactions in conformer B of oxacarbenium ion
derived from glucopyranosyl donor 1. (b) idopyranoside models 2a-c used in this study
ACS Paragon Plus Environment

Page 5 of 40
The Journal of Organic Chemistry
1
2
3
Results and discussion
4
5
6
7
8
9
Synthesis of α- and β-isopropyl idopyranosides. The synthesis of idopyranoside derivatives
2a-c and 12a-c was started from preparation of dimesylate 4. Galactose diol 321 was mesylated with
mesyl chloride in pyridine and dimesylate 4 was treated with potassium tert-butoxide in allyl
alcohol according to the recently developed method.20 As a result, D-idopyranoside 5 was obtained
in 71% yield. A small amount of the epoxy intermediate, p-methoxyphenyl 2,3-anhydro-4,6-O-
benzylidene-β-D-talopyranoside 6, was also isolated. Inversion of configurations of C-2 and C-3 was
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
confirmed by reduced vicinal coupling constants between H-1, H-2 and H-3 in the H NMR
spectrum of product 5 relatively to starting 4.
Benzylation of 5 with BnBr in DMF easily gave 7a. Attempts of PdCl2-catalyzed deallylation
of 7a in order to obtain alcohol 8 failed. To overcome this problem, allyl removal was performed
with low-valent titanium.22 More than fourfold excess of Ti(OPr-i)4/Mg relative to starting 7a was
used to prepare 8 and minimize the formation of diol 9, the main byproduct of this reaction. Target
product 8 was obtained under these conditions in 59% yield. The structures of 8 and 9 were proved
by their NMR spectra and HRMS data.
N-Phenyltrifluoroacetimidoyl group was shown to be one of the most reliable leaving groups
for performing glycosylations with 2-O-benzylated donors.2,6,23-25 Therefore, N-
phenyltrifluoroacetimidates 11a-c were used to prepare isopropyl D-idosides 2a-c and 12a-c.
These donors were synthesized from 7a, acetate 7b and silyl ether 7c; the latter two
derivatives were obtained from 8 by conventional acetylation and silylation, respectively. Anomeric
p-methoxyphenyl groups in 7a-7c were removed by treatment with CAN in a two-phase
acetonitrile–benzene–water mixture to produce hemiacetals 10a-c. Donors 11a-c were prepared by
the reaction of 10a-c with N-phenyltrifluoroacetimidoyl chloride in the presence of K2CO3. It should
be noted that imidates 11a-c proved to be much more sensitive to hydrolysis upon silica gel column
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 40
1
2
3
4
5
6
7
chromatography than the similarly protected glucosyl donor,6 and the usual addition of a drop of
Et3N to the eluent was not enough to fully prevent the product decomposition.
SCHEME 2. Synthesis of idopyranoside probes 2a-c.a
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
Conditions and reagents: (a) MsCl, Py, CH2Cl2 (93%); (b) t-BuOK, AllOH, benzene (71% for 5; 5 % for 6); (c) BnBr,
NaH, DMF (99%); (d) Ti(OPr-i)4, Mg, TMSCl, THF (59% for 8; 5% for 9); (e) Ac2O, Py (89%); (f) TBDMSCl, Py,
imidazole (79%); (g) CAN, benzene-CH3CN-H2O (90% for 10a; 90% for 10b; 57% for 10c); (h) CF3C(NPh)Cl, K2CO3,
acetone (77% for 11a; 84% for 11b, 78% for 11c); (i) i-PrOH, TfOH (5 mol %), AW-300, CH2Cl2 (40% for 2a, 46% for
12a; 28% for 2b, 60% for 12b; 42% for 2c, 54% for 12c).
To minimize the loss of the donors, the time of contact with silica gel upon chromatographic
purification should not exceed 40-60 min independently of the scale. To avoid decomposition of
especially labile silylated 11c, its purification was carried out at a temperature below –10 °C.
TfOH-catalyzed glycosylation of isopropyl alcohol with imidates 11a-c afforded mixtures of
isopropyl idosides 2a-c and 12a-c in yields of about 90% and α:β ratios ranging from 1:1.15 to
1:2.2. Anomeric configurations of 2a-c and 12a-c were deduced from their ROESY spectra and
ACS Paragon Plus Environment

Page 7 of 40
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
1JC1,H1 coupling constant values. Thus, β-anomers 12a-c were featured with the presence of
correlations between H-1 and H-5 in their ROESY spectra, while such correlations were not
observed in the spectra of α-anomers 2a-c. In addition, β-anomers 12a-c and α-anomers 2a-c had
9
1
the JC1,H1 coupling constant values of 156-157 Hz and 169-171 Hz respectively. These values are
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
characteristic for the axial and equatorial orientation of the anomeric protons.26-28
Instrumentation for NMR study. NMR spectra were recorded in CDCl3, CD2Cl2, CD3CN or
toluene-d8 solutions on Bruker spectrometer AV-600 with 0.05% TMS as reference (1H 0.0 ppm, 13C
1
0.0 ppm) at different temperatures. The resonance assignment in H and 13C NMR spectra was
performed by gradient enhanced 2D gCOSY, gNOESY, gHSQC, gJ-HMBC experiments as well as
TOCSY and ROESY experiments. Standard pulse sequences were used. A mixing time was set to
100 ms in the TOCSY experiments. A spin-lock time of 150 ms was used in the ROESY
experiments.
3
Experimental JC–H constants were measured using J-HMBC29 experiment. The experimental
error of thus measured 3JC–H was found by us to have the value of 0.5 Hz.30 The spectral widths were
1
13
about 2 ppm for H region and 40 ppm for C region and did not include resonances of aglycon
groups. The data were collected in the echo/anti-echo mode. The length of gradients was 1 ms, and
the recovery time was 100 µs. The spectra were acquired with 80–120 t1 increments and 64–256
scans per increment. 1024 points were collected during the acquisition time t2. The HMBC
preparation delay was optimized for JC–Hmin = 1.5 and 2.0 Hz. In cases when there was a necessity to
measure the 3JC,H that had values less than 2 Hz J-HMBC experiments were optimized for 1.5 Hz. In
all other cases delay in the pulse sequence was optimized for 2 Hz in order to overcome the weak
sensitivity of J-HMBC experiments and simplify developing of all needed cross correlations.
The upscaling coefficient k was 25–60. The relaxation delay was 1s. The third order low-pass
J-filter was introduced for the suppression of one bond constant (1JC–H) in the range from 125 to 180
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 40
1
2
3
4
5
Hz. The forward linear prediction to 1024 points was used in F1. The processing was performed with
p/2 shifted sine square function in both dimensions.
6
7
8
9
Computational details. All calculations were performed in gas phase using DALTON2015
software.31 6-311++G** basis set was used as supplied with the program, pcJ-1 basis set32 was
obtained from EMSL Basis Set Exchange.33 Before the calculation of the constants, starting
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
4
structures representing S2 and C1 conformations were built for each molecule. They were then
subjected to the first order geometry optimization at DFT/B3LYP/6-311++G** level, Baker
conditions were used to determine convergence. Hessian indices were zero in all the cases. Then
coupling constants were calculated for the optimized structures using DFT/B3LYP/pcJ-1
approximation. Preliminary calculations showed that the greatest contribution to all the constants of
interest was provided by the Fermi-contact term, while the summary contributions from all other
terms did not exceed 2% of each JC,H/JH,H value due to the mutual compensation of diamagnetic and
paramagnetic spin-orbit terms (Sample table of JC,H calculated using all the terms is given in
Supporting Information, Section 1, Table S1). Hence only this term was further computed which
resulted in a considerable speed-up. Cremer-Pople parameters for the optimized conformers were
computed using an on-line service.34 All the absolute energies are provided in the Supporting
Information (Section 1.3, Table S20).
Identification of conformer ratio in equilibrium in CDCl3 solution. Influence of O-3
substituents. According to the classic works,16,18 in benzylidene-protected α-idopyranosides like
2a-c, 1,3-syn-diaxial repulsion between aglyc1on and a substituent at O-3 causes distortion of the
4
O
1
normal C1 conformation towards preference of S2. Additionally, S5 conformer can also be
regarded as possible. Its formation might be expected either from 4C1 through 4H5-like conformation
O
which seems unfavorable in the case of 4,6-O-benzylidene derivatives, or from S2 conformation
1
through B2,5. DFT calculations of S5 conformers for α-compounds 2a-c showed that their energies
ACS Paragon Plus Environment

Page 9 of 40
The Journal of Organic Chemistry
1
2
3
4
O
lay in the middle between C1 and S2 except for 2c where it had the highest relative energy
4
5
(Supporting information, Section 1.3, Table S20). Hence B2,5 conformers were examined only for
6
7
O
compounds 2a and 2b and their geometry optimizations led to S2 conformers. In our opinion, this
8
9
1
means that the S5 conformer cannot be reached in the structures under study due to kinetic reasons
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
and is not considered further.
It was suggested based on the work by Lemieux35 that the proportion of S2 conformation in
O
the equilibrium of a 3-O-substituted α-idopyranoside derivative should reflect the degree of 1,3-syn-
diaxial repulsive interaction between the O-3 substituent and the aglycon. For the detection of the
conformational changes in the sugar ring, vicinal proton-proton and carbon-proton coupling
constants were measured in the hope that the two datasets would complement each other. While
3JH,H constants are routinely used in conformational analysis of pyranoside rings, JC,H coupling
3
constants are more seldom employed for this task. Two examples are inspiring: the first is the
determination of conformation of C-5–C-6 side chain in pyranosides36 and the second describes
evaluation of pyranoside conformation of unprotected idopyranosides.37
Assuming observable vicinal constants (3J) to be linear combinations of the corresponding
constants for separate conformers in the equilibrium, proportions of the conformers were to be
calculated on the basis of experimental 3J magnitudes and values of the corresponding constants for
separate conformers calculated using a quantum mechanical approach.
Geometry optimization of studied structures 2a-c and 12a-c (Scheme 2) gave the following
results. Starting 4C1 conformations for both α- (2a-c) and β-idosides (12a-c) during the optimization
showed just a slight change in the ring shape producing the same conformation very slightly
O
O
distorted towards Н5 or Н1 in case of α- and β-isomers respectively (Table 1). For convenience,
4
O
this conformation is further on denoted simply as C1. However, when starting S2 conformations
were subjected to the optimization, there was significant difference depending on the anomeric
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 40
1
2
3
4
5
6
7
8
configuration. Thus, α-idosides 2a-c all retained OS2 conformation, while their β-counterparts ended
up in a conformation almost exactly intermediate between 1S5 and B2,5 (Table 1). In our opinion, this
suggests that in the studied β-idosides neither pure skew, nor pure boat conformations seem to exist.
9
1
The H-1H coupling constants for the optimized conformations of 2a-c and 12a-c along with
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
those measured experimentally are presented in Table 2. Since the quantity of OS2 (or 1S5/B2,5 in case
of compounds 12a-c) conformer reflects the degree of 1,3-syn-diaxial repulsion, it seemed
reasonable to calculate their proportions (PS) from these data. To this end, equation (1) was used,
PS=100•(e-b)/(a-b)
(1)
that is a general solution for the proportion brought from assumption that observable coupling
constants are linear combinations of the constants calculated for the separate conformers. In this
equation e stands for the experimental magnitude of the vicinal coupling constant, and a and b are
the corresponding values computed for the skew and 4C1 conformers respectively. The proportion of
the skew conformers was calculated for each compound under investigation (Table 2).
TABLE 1. Cremer-Pople Parameters of the DFT Optimized Structures for 2a-c, 12a-c
Cremer-Pople
Starting
parameters
Resulting conformation
Compound
conformation
φ
θ
Q
4C1
OS2
4C1
OS2
4C1
OS2
330.25 11.48
326.32 86.40
335.56 11.13
325.78 87.92
330.69 10.65
324.63 86.78
0.48
Slightly distorted 4C1 (→OH5)
2a
0.73
0.48
0.74
0.48
0.72
OS2
Slightly distorted 4C1 (→OH5)
2b
2c
OS2
Slightly distorted 4C1 (→OH5)
OS2
ACS Paragon Plus Environment

Page 11 of 40
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
4C1
OS2
4C1
OS2
35.68
284.43 97.18
32.11 11.96
283.13 97.80
36.48 11.19
285.73 97.55
11.01
0.54
0.69
0.53
0.69
Slightly distorted 4C1 (→OH1)
1S5/B2,5
12a
Slightly distorted 4C1 (→OH1)
1S5/B2,5
12b
12c
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4C1
OS2
0.54
0.69
Slightly distorted 4C1 (→OH1)
1S5/B2,5
O
As it can be seen, for instance, from the percentage of S2 conformer in 2a, there is a
considerable dispersion in calculated proportions. Thus, calculations using data for J2,3 suggest that
O
2a exists at 73% in the S2 conformation, meanwhile, this value calculated from J4,5 is 60%, and,
finally, consideration of J6S,5 demonstrates the value of only 20%. In order to estimate which of the
values could be treated as reliable, a degree of uncertainty (or uncertainty propagation, σgen) was
calculated that arises from the implementation of equation (1) to three values each of which is
measured or calculated with its proper independent uncertainty (Supporting information. Section 2).
The uncertainty propagation calculations demonstrate that in case of compounds 2a-c only
proportions calculated using J1,2 and J2,3 values can be treated as reliable (Table 2), while
proportions calculated using other constants should be neglected as non-significant.
It can be concluded from the data for compounds 2a-c that 3-O-allylidopyranoside 2a persists
as mainly OS2 conformer with OS2 proportion 73±9%. Acetylated idosyl derivative 2b seems to have
4
O
predominantly C1 conformer with the range of S2 proportion 32±8%. Conformer ratio for idosyl
derivative 2c bearing a TBDPS protective group was found to be 26±9%. The latter result evidences
that silylated α-isopropyl idoside 2c tends to adopt the 4C1 conformation despite the fact that in this
case the bulky TBDPS substituent in axial position is supposed to have a steric repulsion with the
isopropyl group in the aglycon.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 40
1
2
3
3
TABLE 2. Experimental and calculated JH,H coupling constants (Hz) and conformer
4
5
percentages of 2a-c, 12a-c
6
7
8
9
Compound
2a
Coupling constants
Exp. values
J1,2
J2,3
J3,4
3.4
3.7
2.2
80
J4,5 J6S,5 J6R,5
5.0
9.0
2.5
2.9
1.9
60
1.5
1.1
1.6
20
2.5
3.4
2.5
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Calculated OS2
6.2 11.7
Calculated 4C1
0.8
78
1.7
73
Percentage of OS2 conformer
Uncertainty σgen
Exp. values
-
a
±17
3.0
±9
5.3
(±60)b
3.1
4.7
2.5
27
--b
--b
--b
2.3
3.5
2.4
2b
2.5
3.0
1.9
55
0.9
1.1
1.7
Calculated OS2
6.1 12.6
Calculated 4C1
0.8
42
1.8
32
Percentage of OS2 conformer
Uncertainty σgen
Exp. Values
-
-
a
a
±15
2.5
±8
4.4
±40
2.7
3.8
2.3
27
--b
--b
1.5
1.2
1.6
25
--b
2.1
3.4
2.4
2c
2.2
2.8
1.9
33
Calculated OS2
6.0 11.8
Calculated 4C1
0.8
33
1.8
26
Percentage of OS2 conformer
Uncertainty σgen
Exp. values
-
a
±16
1.4
±9
3.4
--b
--b
--b
--b
2,1
2.5
2.7
12a
2.8
0.7
2.3
2.0
2.6
2.1
1.2
1.5
1.4
Calculated 1S5/B2,5
Calculated 4C1
3.3 10.0
1.2
10
2.8
8
Percentage of 1S5/B2,5 conformer
Uncertainty σgen
Exp. Values
-
-
-
-
a
a
a
a
±50 ±18
1.5 3.2
--b
--b
--b
--b
12b
2.6
1.8
1.4
2.2
ACS Paragon Plus Environment

Page 13 of 40
The Journal of Organic Chemistry
1
2
3
4
Calculated 1S5/B2,5
Calculated 4C1
2.9 10.6
0.8
2.4
2.6
2.0
1.5
1.6
2.4
2.6
1.2
18
--b
0c
2.8
5
5
6
7
8
9
a
a
a
a
Percentage of 1S5/B2,5 conformer
Uncertainty σgen
-
-
-
-
±14
2.8
--b
--b
0c
--b
0c
--b
1.8
2.7
2.6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
d
12c
Exp. Values
-
Calculated 1S5/B2,5
Calculated 4C1
3.0 10.2
0.9
2.3
2.7
2.1
1.4
1.5
1.3
2.8
0
a
a
a
a
a
Percentage of 1S5/B2,5 conformer
Uncertainty σgen
-
-
-
-
-
--b
±15
--b
--b
--b
--b
aPercentage of conformer cannot be calculated because experimental value falls out of the range between two
corresponding constants calculated for separate conformers. bUncertainties with values more than 50% are omitted; also
the corresponding percentage values can be neglected and are given only to demonstrate their dispersion. cMeasurement
of coupling constant with 0.1 Hz exactness was not possible because of signal broadening. dCoupling constant cannot be
determined because of signals overlap.
Manifestations of TBDPS, TBDMS and TIPS bulkiness are well known. Thus, these groups
are used in carbohydrate chemistry not only as protective groups but also as substituents capable of
changing the pyranose conformation.38-40 Main driving force for the conformational flip induced by
TBDPS, TBDMS38 or TIPS is thought to be a steric repulsion between them when these groups have
a vicinal arrangement. But the example of 2c clearly showed that the ability of TBDPS to exert
steric hindrances plays a subordinate role in the case of the 1,3-syn-relashionship. The propensity of
TBDPS to be axial in 2c also correlates with observations41,42 that the O-TBDPS group in silylated
cyclohexanol is prone to adopt an axial orientation. The reason for weakening of the 1,3-syn-diaxial
interaction in silylated 2c may lie in the properties of the Si–O linkages, which are remarkable for
low basicity of oxygen in both Si-O-Si43,44 and Si-O-C45,46 structures, despite higher anionicity
relative to C–O bond. NBO analysis showed that the lone pair of oxygen is involved in a resonance-
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 40
1
2
3
4
type nO→s*Si-C interaction.43,44 This effect might diminish the electron density on O-3 in 2c and thus
lead to predominance of the 4C1 conformer.
5
6
Three β-isomers, 12a-c, were also studied in order to get an estimation of the repulsive
interaction strength between substituents at C-2 and C-4, which also had mutual 1,3-syn-diaxial
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
spatial relationship in C1 conformation in both α- (2a-c) and β- (12a-c) isopropyl glycosides. On
the other hand, both chair and skew conformations of β-idosides lack O-3–O-1 1,3-syn-diaxial
repulsion.
As can be seen from Table 2, the J3,4, J4,5 and J5,6 values fall out of the range between two
1
corresponding constants calculated for 4C1 and S5/B2,5 conformers of 12a-c. Meanwhile, the values
of J1,2 appear to be inaccurate since uncertainty propagation for proportion calculated using this
constant is more than 20%. Taking the values of J2,3, allylated β-idopyranoside 12a as well as its
acetylated and silylated counterparts 12b and 12c, all adopt mainly 4C1 conformation, likely having
O
the percentage of S2 between 0-26%, 0-19% and 0-15% respectively. Therefore, it should be
concluded that the interaction between the substituents at C-2 and C-4 does not contribute much to
the preference for OS2 conformer in 2a-c, which is thus mainly governed by the relative instability of
the 4C1 conformer due to repulsion between the substituents at C-1 and C-3. In this way, the validity
of compounds 2a-c as models for the study of 1,3-syn-diaxial repulsion is proved.
Relative energies of the conformers obtained after the geometry optimization calculations
demonstrated (Supporting information, Section 1.3, Table S20) that 2a-c and 12a-c may not exist as
O
4
pure S2 or C1, which are actually the conformers found by simple conformational analysis.17
However, no strict correlation between the calculated energies and the supposed percentage of
conformers could be established.
Influence of solvents on the conformation. The dielectric constant of solvent was shown to
influence the conformational equilibrium of pyranosides by way of modulating the 1,3-syn-diaxial
ACS Paragon Plus Environment

Page 15 of 40
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
repulsion.16-18,35 This may consequently have an effect on stereoselectivity of glycosylations through
the influence on the conformational equilibrium of oxacarbenium ions, when one of the putative
conformers has 1,3-syn-diaxial substituent arrangement. Although the influence of solvents on the
selectivity of glycosylations has been comprehensively investigated,47-49 we suggested that 1,3-syn-
diaxial repulsion may be one of the driving forces that define the pattern of solvent effects.
To investigate this question in-depth, conformational behavior of α-idopyranosides 2a-с was
studied in three solvents, CD2Cl2, CD3CN and toluene-d8, which are deuterated analogues of the
most frequently used solvents in glycosylation reactions.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1H NMR and J-HMBC spectra were recorded for compounds 2a,c in CD2Cl2, CD3CN and
toluene-d8, for compound 2b in CD3CN and toluene-d8 (Tol-d8). Besides, spectra of allyl idoside 2a
were recorded in a CD2Cl2-toluene-d8 mixture (1:1). The resulting set of 1H-1H coupling constants is
presented in Table 3. The computed values of constants for distinct conformers and dielectric
constants are also given. The major changes of magnitudes were observed for the J2,3 constants.
TABLE 3. 3JH,H coupling constants (Hz) of 2a-с determined in different solvents
Compound
2a
Solvent
Calculated OS2
Calculated 4C1
Tol-d8
J1,2 J2,3 J3,4 J4,5 J6S,5 J6R,5 OS2 (%)a
6.2 11.7 3.7 2.9 1.1 3.4
ε
0.8 1.7 2.2 1.9 1.6 2.5
5.3 9.5 3.4 2.7
0
2.8
78
73
72
68
65
2
5
CDCl3
5.0 9.0 3.5 2.3 1.5 2.5
b
CD2Cl2-Tol-d8, (1:1) 4.8 8.8 3.5
-
0
2.4
2.5
b
CD2Cl2
CD3CN
4.7 8.5 3.4 2.8
-
9
b
b
4.5 8.2 3.4
-
0.9
-
36
2b
Calculated OS2
Calculated 4C1
6.1 12.6 4.7 3.0 1.1 3.5
0.8 1.8 2.5 1.9 1.7 2.4
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 40
1
2
3
4
5
6
7
Tol-d8
CDCl3
3.4 5.8 3.3 2.3 1.0 2.4
3.0 5.3 3.1 2.5 0.9 2.3
37
32
43
b
b
CD3CN
3.4 6.4 3.4 2.3
-
-
8
9
2с
Calculated OS2
Calculated 4C1
Tol-d8
6.0 11.8 3.8 2.8 1.2 3.4
0.8 1.8 2.3 1.9 1.6 2.4
3.5 6.2 3.3 2.6 1.5 2.4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
44
26
22
25
CDCl3
2.5 4.4 2.7 2.2 1.5 2.1
b
CD2Cl2
2.1 4.0
-
2.4 1.6 1.8
CD3CN
2.3 4.3 2.8 2.3
0
0
aPercentage of the S2 conformation calculated by eq. 1 from the J2,3 values. Coupling constant cannot be determined
O
b
due to signals overlap.
Changes in coupling constant values for 2a measured in different solvents correlated with
solvent dielectric constants (comparison of main peak forms can be found in the Supporting
information, Page S54). It is particularly noticeable for the J2,3 values. That is, OS2/4C1 ratio depends
on solvent polarity as long as J2,3 is considered as estimation for this ratio. Decrease of the relative
O
proportion of S2 in CD2Cl2 in comparison to CDCl3 correlates with higher polarity of CD2Cl2. A
O
mixture of CD2Cl2 and toluene-d8 (1:1) generates an equilibrium in which the percentage of S2
becomes a mean of those in CD2Cl2 and in toluene-d8. The conformer ratios were also the same in
CDCl3 and in the 1:1 mixture of CD2Cl2 and toluene-d8. In the most polar solvent, CD3CN, the
percentage of the OS2 conformer demonstrated the lowest value among all the tested solvents.
These results suggest that in context of glycosylation mechanism, the largest impact from the
1,3-syn-diaxial repulsion on the conformation of the intermediate oxacarbenium cations should be
expected in toluene. In acetonitrile this influence should be the smallest. Conformational effects of
the 1,3-syn-diaxial repulsion in CD2Cl2 and CDCl3 lie in the middle between those in toluene-d8 and
ACS Paragon Plus Environment

Page 17 of 40
The Journal of Organic Chemistry
1
2
3
4
5
CD3CN, but still one would expect stronger repulsion in CDCl3 than in CD2Cl2. Mixing of solvents
of different polarity reveals an additive effect.
6
7
8
9
1H-1H coupling constants in spectra of 2b,c demonstrate the same tendencies as observed for
structure 2a, that is, they suggest that the skewed conformer is prevalent in a less polar solvent
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
(toluene-d8). However, in a more polar solvent (CD3CN) definite preference for C1 chair
conformation can only be observed for compound 2a. For structure 2c the effect is less pronounced
when comparing CDCl3 and CD3CN, and for 2b the trend is reversed. This means that the behavior
of the studied compounds in different solvents can be regulated also by other forces besides the 1,3-
syn-diaxial repulsion. For example, changes in overall dipole moments upon the conformational
transformations can be regarded as one of such factors. To study this question in detail, the same
experiments were performed for the β-idosides, 12a-c (Table S33, Section 3.4 in Supporting
information) and the results for all the compounds in the work were analyzed in comparison with the
calculated dipole moments (Table S21, Section 1.4 in Supporting information). This analysis
showed that in all the α-structures differences in dipole moments between chair and skewed
conformations were rather small. Molecule 2b was the only example where the skewed
4
conformation was even slightly more polar than C1. In the β-series all the chair conformers were
strongly more polar than the skewed ones. However, coupling constant changes in different solvents
in this series were weak and contradictory. Some deviations in H-1H coupling changes for
1
compounds 2b and 2c in CD3CN as compared to 2a could possibly be explained by the fact that in
the former two the 1,3-syn-diaxial repulsion is expectedly decreased and, especially in the polar
solvent, dipole moments started to have some effect. In summary, all this indicates that changes of
the dipoles play a minor role as compared to the 1,3-syn-diaxial repulsion in compound 2a, while in
2b and 2c in polar CD3CN their influence might be stronger. For the β-compounds only absence of
the 1,3-syn-diaxial repulsion can be clearly established.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 40
1
2
3
4
5
6
7
8
9
In case when the proposed conformer for the oxacarbenium ion has a 1,3-syn-diaxial
relationship of two substituents, the difference in stereoselectivity of glycosylations in different
solvents can be explained by the influence of solvent polarity on the strength of 1,3-syn-diaxial
interaction. Indeed, there is plenty of examples,50-52 when selectivity differs depending on solvent
polarity. For example, donors with the gluco configuration, whose plausible oxacarbenium ions may
have 1,3-syn-diaxial relationship between 2-OBn and 4-OBn, demonstrate53,54 different
stereoselectivity in CH2Cl2 and CH3CN. There is little evidence that toluene changes selectivity of
glucosylations relative to CH2Cl2. In the only found literature example demonstrating consistent
comparison of glucosylation selectivities in CH2Cl2 and toluene without ether co-solvents, a
pronounced difference between these two solvents is shown. However, the resulting selectivities are
overshadowed by the presence of modulating DMF additive.55 The idea that the influence of solvent
polarity on the oxacarbenium ion conformation and, hence, on selectivity can be mediated by 1,3-
syn-diaxial interaction correlates well with the hypothesis that the stereoselectivity is explained by
solvent-induced variations in conformational preferences of the oxacarbenium cation.56,47
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Applicability of 3JC,H couplings for study of conformational equilibria in idopyranosides.
3
Although knowledge of JH,H couplings allows for estimation of conformer ratio, in case of
idopyranosides 2a-c and 12a-c even thorough examination of JH3,H4, JH4,H5 and JH5,H6 couplings
provided little information about real torsional angles at C3-C5. In order to learn more about the
conformational shape of the studied compounds, J-resolved HMBC spectra were registered for 2a-c
and 12a-c in different solvents and vicinal carbon-proton (3JC,H) coupling constants were extracted
from them (Tables S27-S32, Section 3.3 in Supporting information).
Applicability of the resulting 3JC,H constants for the calculation of conformer proportions of α-
D-idosides 2a-c was evaluated by analysis of uncertainty propagation. Calculated percentage and
uncertainties are provided only for the case of 2a in CDCl3 (Table S26, Section 3.2 in Supporting
information) since resulting uncertainties in all the other cases were very much alike.
ACS Paragon Plus Environment

Page 19 of 40
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
3
Large uncertainties preclude determination of the conformer percentage from JC,H couplings.
The limitation comes from everywhere: experimental uncertainties increase, same as does the
uncertainty of calculations due to the smaller range of constants; another difficulty is the small
difference between putative values of coupling constants of separate conformers. As it was found
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
(Supporting information Section 2) uncertainty of JC,H coupling constants derived from computer
modelling is ≤ 2.1 Hz, measurement uncertainty is 0.5 Hz and maximum range between couplings
of putative values of coupling constants of separate conformers is 3.9 Hz (Supporting information
Table S28, JC5,H3). These values give uncertainty propagation minimum ±43%.
Nevertheless, vicinal and geminal JC,H coupling constants are being increasingly introduced
into the field of analysis of pyranoside conformations. Serianni tested conformity of experimental
4
3
JC,H (and JC,C) of D-idosides with percentage of C1 conformer calculated from JH,H by building
plots of the percentages of 4C1 against some JC,H values.37
Changes in 3JC,H couplings in α-D-idosides 2a-c and β-D-idosides 12a-c correlate well with the
3
tendencies demonstrated by JH,H couplings manifested in different solvents. In particular, decrease
O
of JC1,H5, JC3,H1, JC1,H3, JC4,H2 and JC5,H3 is consistent with enlargement of proportion of S2
conformer.
Two plots (Figures S2 and S3, Section 3.5, Supporting information) of the percentages of 4C1
3
against JC3,H1 and JC2,H4 values were built throughout compounds 2a-c to see if behavior of JC,H
coupling constants is consistent for different compounds. Analysis of the plots reveals that changes
of JC1,H3 and JC2,H4 values for 2b and c conform one trend. However values JC3,H1 and JC2,H4 of 2a
fall out.
Influence of temperature on the conformer ratio. In an attempt to freeze out separate
conformers and measure coupling constants for each of them, 1D 1H NMR spectra of 2a in CD2Cl2
solution were recorded at temperatures as low as –80°C. But even at that temperature, only
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 40
1
2
3
4
5
6
7
8
9
decreased values of some inter-proton constants (J1,2, J2,3 and J3,4) were observed (Table 4). This
might mean that with the lowering of the temperature, the proportion of OS2 conformer goes down. If
this percentage is calculated using J2,3 constants, the quantity of the OS2 conformation changes from
68% at RT to 48% at –80 °C. This lack of success coincides with Paulsen’s results,17 when he also
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
failed to acquire spectra containing signals of separate conformers of idose even at –100 °C and
observed only decrease of coupling constants.
TABLE 4. 3JH,H coupling constants (Hz) of 2a determined in CD2Cl2 at different temperatures
Temperature/
J1,2
4.8
3.2
J2,3
8.5
7.0
J3,4
3.3
2.6
OS2 (%)a
Coupling constants
273 K
68
48
193 K
aPercentage of OS2 conformation calculated by eq. 1 from values of J2,3.
This observation also means that the impact of 1,3-syn-diaxial repulsion on the conformational
equilibrium might decrease at lower temperatures. Therefore, in case when this interaction is
important for the equilibrium of oxacarbenium ions formed during glycosylation, the ratio of
anomeric products would depend on the temperature at which the glycosylation was carried out.
Indeed, sometimes temperature influences the stereochemical result of glycosylations and this
influence seems to be differently directed for different glycosyl donors.57-59,6 Probably, the complex
pattern of the temperature dependence can be partly explained by the increasing intervention of 1,3-
syn-diaxial repulsion into the conformational equilibrium with the rise of the temperature.
Conclusions
The substituents at C-1 and C-3 in isopropyl 2-O-benzyl-4,6-O-benzylidene-α-D-idopyranosides
experience repulsive 1,3-syn-diaxial interaction to different extent depending on the type of the
protecting group at O-3. Among the three studied O-3 substituents (allyl, acetyl, and TBDPS),
which represent the principal types of protecting groups in carbohydrate chemistry, the allyl
ACS Paragon Plus Environment

Page 21 of 40
The Journal of Organic Chemistry
1
2
3
derivative displays the largest repulsive force. The destabilizing O-1/O-3 interaction caused a shift
4
5
6
7
8
9
in the conformational equilibrium C1↔OS2 towards S2. The percentage of the latter conformer
reaches 73±9%, which shows that the anomeric effect and the strain resulting from ring puckering
are outweighed by 1,3-syn-diaxial repulsion. In accordance with the assumption that electron density
on oxygen atoms determines the strength of this repulsion, the 3-O-acetyl derivative adopts
predominantly 4C1 conformation with the relative proportion of OS2 dropping down to 32±8%. In the
case of the 3-O-TBDPS derivative, the weight of OS2 conformer is only 26±9%, despite presumably
unfavorable spatial interactions between TBDPS and the isopropyl group. The observable decrease
of syn-1,3-diaxial interaction in 2c is in accordance with the known feature of the Si-O linkage
consisting in lower basicity of oxygen in comparison to C-O bond. The conformer ratios for 2b and
2c show that TBDPS diminishes electron density on oxygen more effectively than the acetyl group.
The repulsive interaction between the substituent at C-3 and 1-isopropoxy groups in the α-
idosides depends on polarity of the solvent. Particularly, in the case of the 3-allyloxy group the
4
O
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
lower content of the S2 conformer in the equilibrium correlates with the higher solvent dielectric
constant in accordance with our hypothesis of 1,3-syn-diaxial repulsion. In case of 3-acetoxy and 3-
silyloxy substituents, due to the expected decrease of this repulsion, other factors influencing the
conformational state, such as overall dipole moments, seem to play some role in polar solvents.
Assuming that stereoselectivity of glycosylation is regulated by the conformational equilibrium of
oxacarbenium ions, this opens a way to understanding how the solvent dielectric constant affects
stereoselectivity through destabilization of conformers with a 1,3-syn-diaxial relationship between
two alkoxy substituents. This is, for example, the case of donors with the gluco configuration
bearing simultaneously two alkyl protecting groups at O-2 and O-4.
Analysis of experimentally measured vicinal proton-proton couplings in combination with
DFT/B3LYP/pcJ-1 calculations of constants for separate previously optimized conformers
represents a reliable tool for determination of conformer ratio with 15% precision in case of
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 40
1
2
3
4
5
6
7
pyranosides. The use of carbon-proton vicinal couplings is limited due to adverse factors: the greater
uncertainties of coupling measurement from J-HMBC spectra and insufficient prediction accuracy
of couplings for separate conformers by computer calculations.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Experimental section
General experimental methods. Molecular sieves MS AW-300 for glycosylation reactions
were activated prior to application at 180 °C in vacuum of an oil pump for 2 h. Dichloromethane
was successively distilled from diethanolamine, P2O5, and CaH2 under Ar. Pyridine was dried by
distillation from P2O5. Analytical TLC was performed on Silica Gel 60 F254 aluminium sheets
(Merck), and visualization was accomplished using UV light or by charring at 150 °C with 10%
(v/v) H3PO4 in isopropyl alcohol. Column chromatography was performed on Silica Gel 60, 40–63
µm (Merck). Low-temperature chromatography of 11c was carried out using a jacketed column
1
13
connected to a Lauda Eco RE 1050 S cooling thermostat. Signal assignment in H and C NMR
spectra was made using COSY and 1H-13C HSQC techniques.
p-Methoxyphenyl 2,3-di-O-mesyl-4,6-O-benzylidene-β-D-galactopyranoside (4). Mesyl
chloride (19.4 mL, 0.25 mol) was added to a chilled solution of diol 3 (28.3 g, 0.075 mol) in a
mixture of dry CH2Cl2 (340 mL) and pyridine (200 mL). Cooling was removed and the reaction
mixture was stirred at rt for 5 days. After completion of the reaction (TLC control), the solvents
were evaporated and residual pyridine was removed by addition and evaporation of toluene (3×25
mL). The residue was purified by chromatography on silica gel in CH2Cl2-CH3OH (200:1→175:1)
24
to afford 4 (38.1 g, 95%): white crystals, mp 227–229 °C, Rf = 0.20 (toluene–MTBE, 2.5:1); [α]D
–9.6 (c 1, CHCl3); 1H NMR (600 MHz, CDCl3): δH 7.57–7.53 (2H, m, Ph), 7.43–7.36 (3H, m, Ph),
7.05 (2H, d, J 9.0 Hz, OCH4OCH3), 6.86 (2H, d, J 9.0 Hz, OCH4OCH3), 5.61 (1H, s, CHPh), 5.14
(1H, dd, J2,1 7.9 Hz, J2,3 10.0 Hz, H-2), 5.02 (1H, d, J1,2 7.9 Hz, H-1); 4.86 (1H, dd, J3,2 9.9 Hz, J3,4
ACS Paragon Plus Environment

Page 23 of 40
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
3.7 Hz, H-3), 4.56 (1H, d, J4,3 3.7 Hz, H-4), 4.37 (1H, br dd, J6A,6B 12.7, H-6A), 4.10 (1H, dd, J6B,5
1.5 Hz, J6B,6A 12.6 Hz, H-6B), 3.79 (3H, s, OCH3), 3.60 (1H, br m, H-5), 3.20, 3.18 (6H, 2s, SCH3);
13C NMR (150 MHz, CDCl3): δC 129.2, 129.0, 126.3 (Ph), 119.2, 114.8 (MeOC6H4O), 101.1
(PhCH), 100.0 (C-1), 76.6 (C-3), 76.2 (C-2), 74.8 (C-4), 68.4 (C-6), 66.3 (C-5), 55.7 (OCH3), 39.4,
39.2 (SCH3). Anal. Calcd (%) for C22H26S2O11: C, 49.80; H, 4.94. Found: C, 49.85; H, 4.83.
p-Methoxyphenyl 3-O-allyl-4,6-O-benzylidene-β-D-idopyranoside (5). A solution of
potassium tert-butoxide (16.2 g, 0.14 mol) in allyl alcohol (56.0 mL) was added to a solution of the
dimesylate 4 (7.0 g, 0.013 mmol) in a mixture of allyl alcohol (60.0 mL), dry benzene (45.0 mL)
and dioxane (47.0 mL). After 4 days, the reaction mixture was evaporated to dryness, redissolved in
ethyl acetate (230 mL), and washed with saturated sodium chloride solution (230 mL). The aqueous
layer was extracted back with EtOAc (3×30 mL). The combined organic layers were dried with
sodium sulfate, filtered, and evaporated to dryness. The resulting solid was purified by column
chromatography on silica gel using toluene–MTBE (50:1→25:1) to give the idopyranoside 5 (3.9 g,
71%) as a white foam and anhydro-D-talopyranoside 6 (215 mg, 5 %) as a yellowish crystal solid.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
24
1
5: Rf = 0.20 (toluene–EtOAc, 20:1); [α]D –74.9 (c 1, CHCl3); H NMR (600 MHz, CDCl3):
δH 7.50–7.46 (2H, m, Ph), 7.38–7.31 (3H, m, Ph), 7.06 (2H, d, J 9.1 Hz, OCH4OCH3) , 6.82 (2H, d,
J 9.1 Hz, OCH4OCH3), 5.88 (1H, m, CH2=CHCH2O), 5.51 (1H, s, CHPh), 5.29 (1H, m,
CH2=CHCH2OA), 5.27 (1H, br s, H-1), 5.22 (1H, m, CH2=CHCH2OB), 4.40 (1H, dd, J6A,5 1.33 Hz,
12.6 Hz, H6A), 4.15 (2H, m, CH2=CHCH2), 4.12 (1H, dd, J6B,5 1.8 Hz, J6B,6A 12.5 Hz, H-6B), 4.08
(1H, m, H-4), 3.96–3.91 (2H, m, H-2, H-3), 3.89 (1H, m, H-5), 3.79 (3H, s, OCH3), 3.43 (1H, d,
JOH,H2 11.5 Hz, OH-2); 13С NMR (150 MHz, CDCl3): δC 155.1, 151.5 (ipsoMeOC4H6), 137.2
(ipsoPh), 133.7 (CH2=CHCH2O), 129.2, 128.3, 126.0 (Ph), 118.2 (MeOC4H6), 118.0
(CH2=CHCH2), 114.4 (MeOC4H6), 101.5 (PhCH), 98.5 (C-1), 76.2 (C-3), 73.4 (C-4), 71.5
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 40
1
2
3
4
5
(CH2=CHCH2), 69.5 (C-6), 67.4 (C-2), 66.8 (C-5), 55.6 (OCH3); Anal. Calcd (%) for C23H26O7: C,
66.65; H, 6.32. Found: C, 66.44; H, 6.31;
6
7
8
9
6: Rf = 0.37 (toluene–EtOAc, 20:1); mp 149–152 °C, [α]D24 –135 (c 1, CHCl3); 1H NMR (600
MHz, CDCl3): δH 7.56–7.51 (2H, m, Ph), 7.40–7.32 (3H, m, Ph), 7.12 (2H, d, J 9.1 Hz,
OC6H4OCH3), 6.80 (2H, d, J 9.1 Hz, OC6H4OCH3), 5.59 (1H, s, PhCH), 5.40 (1H, s, H-1), 4.35
(1H, m, H-4), 4.33 (1H, br d, J6A,6B 12.7 Hz, H-6A), 4.07 (1H, dd, J6B,5 2.2 Hz, J6B,6A 12.7 Hz, H-6B,
H-6B), 3.76 (3H, s, OCH3), 3.52–3.47 (2H, m, H-3, H-2), 3.41 (1H, m, H-5); 13С NMR (150 MHz,
CDCl3): δC 129.2, 128.3, 126.4 (Ph), 118.8, 114.5 (MeOC4H6), 101.3 (PhCH), 96.7 (C-1), 70.2 (C-
3), 69.6 (C-6), 62.5 (C-5), 55.6 (OCH3), 53.5 (C-2), 51.7 (C-3). The NMR data of 6 were in
accordance with those of similar anhydrotalopyranosides.20,60 HRMS (ESI-TOF) m/z [M + K]+
Calcd for C20H20O6K 395.0891; Found 395.0891.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
p-Methoxyphenyl
3-O-allyl-2-O-benzyl-4,6-O-benzylidene-β-D-idopyranoside
(7a).
Sodium hydride (60% suspension, 827 mg, 20.7 mmol) and benzyl bromide (3 mL, 25 mmol) were
added to a solution of idopyranoside 5 (3.51 g, 8.5 mmol) in dry DMF (41 mL) at +4 °C, and then
cooling was removed. After 15 min of stirring at rt, TLC showed full conversion of the starting
material. The excess of NaH was quenched by addition of MeOH (3 mL) upon cooling. The reaction
mixture was diluted with heptane (250 mL) and washed with water (300 mL). The aqueous phase
was extracted back with heptane (3×50 mL), and the combined heptane solutions were extracted
with acetonitrile (3×50 mL). The combined acetonitrile extracts were taken to dryness, and the
residue was subjected to silica gel chromatography in toluene–MTBE (70:1 → 25:1) to afford 7a
20
1
(4.26 g, 99%) as a foam: Rf = 0.20 (PE–EtOAc, 5:1); [α]D –46.0 (c 1, CHCl3); H NMR (600
MHz, CDCl3): δH 7.54–7.44 (4H, m, Ph), 7.35–7.25 (6H, m, Ph), 7.06 (2H, d, J 9.1 Hz,
MeOC6H4O), 6.83 (2H, d, J 9.1 Hz, MeOC6H4O), 5.85 (1H, m, CH2=CHCH2O), 5.50 (1H, s,
PhCH), 5.30 (1H, d, J1,2 1.6 Hz, H-1), 5.24 (1H, m, CH2=CHCH2OA), 5.17 (1H, m,
ACS Paragon Plus Environment

Page 25 of 40
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
CH2=CHCH2OB), 4.96 (1H, d, JBnA,BnB 12.1, PhCH2A), 4.78 (1H, d, JBnB,BnA 12.1, PhCH2B), 4.34
(1H, dd, J6A,5 1.5 Hz, J6A,6B 12.6 Hz, H-6A), 4.07 (1H, dd, J6B,5 2.1 Hz, J6B,6A 12.6 Hz, H-6B), 4.05
(2H, m, CH2=CHCH2), 3.95 (1H, m, H-4), 3.92 (1H, m, H-3), 3.76 (1H, m, H-5), 3.75–3.73 (4H, m,
OCH3, H-2); 13C NMR (150 MHz, CDCl3): δC 154.9, 151.7 (ipsoMeOC4H6), 138.6, 138.0 (ipsoPh),
134.0 (CH2=CHCH2O), 128.8, 128.1, 128.0, 127.7, 127.3, 126.7 (Ph), 118.3 (MeOC4H6), 117.6
(CH2=CHCH2), 114.3 (MeOC4H6), 101.3 (PhCH), 98.8 (C-1), 76.5 (C-3), 73.9 (C-2), 73.22
(PhCH2), 73.19 (C-4), 71.3 (CH2=CHCH2), 69.5 (C-6), 66.7 (C-5), 55.6 (OCH3); Anal. Calcd (%)
for C30H32O7: C, 71.41; H, 6.39. Found: C, 71.32; H, 6.50.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
p-Methoxyphenyl 2-O-benzyl-4,6-O-benzylidene-β-D-idopyranoside (8). Magnesium
turnings were merged into a 5% v/v solution of AcOH in EtOH until gas started to evolve, then
washed consecutively with EtOH and Et2O and dried for 15 min. Thus activated Mg (370 mg) was
added to a solution of 3-O-allyl-idopyranoside 7a (462 mg, 0.92 mmol) in dry THF (16 mL).
Ti(OPr-i)4 (1.3 mL, 4.23 mmol), TMSCl (470 µL, 3.70 mmol) and a drop of 1,2-dibromoethane
were added consecutively to this stirred heterogeneous mixture. The reaction mixture started
immediately to darken. After 6 h, when TLC showed full conversion of the starting compound 7a,
the reaction was quenched by sequential addition of Celite (3.45 g), NaF (10.1 g) and aqueous
saturated NaHCO3 (7 mL). The reaction mixture was diluted with Et2O (50 mL) and stirred for
several hours until its color became off-white. Then the solids were filtered off through a pad of
Celite, washed with Et2O (3×30 mL), and the filtrate was evaporated. The solid residue was purified
on a silica gel column (toluene–MTBE, 10:1→3:1) to give product 8 (250 mg, 59%) as a crystal
solid. Byproduct 9 (5%) was also isolated as white crystals.
8: Rf = 0.20 (toluene–MTBE, 5:1); mp 137–139 °C, [α]D24 –46.1 (c 1, CHCl3); 1H NMR (600
MHz, CDCl3): δH 7.48–7.38 (4H, m, Ph), 7.31-7.21 (7H, m, Ph + residual toluene), 7.02 (2H, d, J
9.3 Hz, OC6H4OCH3), 6.80 (2H, d, J 9.3 Hz, OC6H4OCH3), 5.46 (1H, s, PhCH), 5.38 (1H, d, J1,2 1.8
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 40
1
2
3
4
5
Hz, H-1), 4.90 (1H, d, JBnA,BnB 12.1 Hz, PhCH2A), 4.72 (1H, d, JBnB,BnA 12.1 Hz, PhCH2B), 4.31
(1H, dd, J6A,5 1.5 Hz, J6A,6B 12.3 Hz, H-6A), 4.30 (1H, dd, J3,2 4.7 Hz, J3,4 2.1 Hz, H-3), 4.02 (1H,
6
7
dd, J6B,5 2.1 Hz, J6B,6A 12.6 Hz, H-6B), 3.93 (1H, t, J4,3 2.1 Hz, H-4), 3.80 (1H, dt, J5,4 2.0 Hz, J5,6A
=
8
9
13
J5,6B 1.8 Hz, H-5), 3.76 (3H, s, OCH3), 3.67 (1H, dd, J2,1 1.8 Hz, J2,3 4.7 Hz, H-2); C NMR (150
MHz, CDCl3): δC 128.9, 128.2, 128.0, 127.8, 127.5, 126.7 (Ph), 118.6, 114.4 (OC4H6OCH3), 101.3
(PhCH), 98.0 (C-1), 76.3 (C-2), 75.7 (C-4), 73.0 (CH2Ph), 69.9 (C-3), 69.5 (C-6), 66.3 (C-5), 55.6
(OCH3); HRMS (ESI-TOF) m/z [M+NH4]+ Calcd for C27H32NO7 482.2173; Found 482.2173;
9: Rf = 0.23 (toluene–MTBE, 3:1); mp 166–168 °C, [α]D22 –108.6 (c 1, CHCl3); 1H NMR (300
MHz, CDCl3): 7.52–7.46 (2H, m, Ph), 7.38–7.33 (5H, m, Ph), 7.07 (2H, d, J 9.1 Hz, OC6H4OCH3),
6.82 (2H, d, J 9.1 Hz, OC6H4OCH3), 5.52 (1H, s, CHPh), 5.34 (1H, d, J1,2 1.0 Hz, H-1), 4.43 (1H,
dd, J6A,5 1.6 Hz, J6A,6B 12.6 Hz, H-6A), 4.30 (1H, dt, H-3), 4.11 (1H, dd, J6B,5 1.9 Hz, J6A,6B 12.6 Hz,
H-6B), 4.02 (1H, m, H-4), 3.91 (1H, dt, H-5), 3.85 (1H, ddt, J2,1 1.0 Hz, J2,3 3.3 Hz, J2,OH2 11.7 Hz,
H-2), 3.78 (3H, s, OCH3), 3.38 (1H, d, JOH2,2 11.7 Hz, OH2), 2.05 (1H, d, JOH3,3 4.5 Hz, OH3); 13C
NMR (150 MHz, CDCl3): δC 129.4, 128.4, 126.2 (Ph), 118.5, 114.6 (MeOC4H6), 101.7 (PhCH),
98.3 (C-1), 75.3 (C-4), 70.0 (C-3), 69.8 (C-2), 69.7 (C-6), 66.6 (C-5), 55.8 (OCH3); Anal. Calcd (%)
for C20H22O7: C, 64.16; H, 5.92. Found: C, 64.25; H, 6.05; HRMS (ESI-TOF) m/z [M+NH4]+ for
C20H26NO7 392.1704; Found 392.1697; [M+Na]+ Calcd for C20H22O7Na 397.1258; Found 397.1256.
p-Methoxyphenyl 3-O-acetyl-2-O-benzyl-4,6-O-benzylidene-β-D-idopyranoside (7b).
Acetic anhydride (4 mL) was added to a solution of 8 (280 mg, 0.60 mmol) in pyridine (4 mL).
When TLC showed full conversion of the starting material (2 h), the solvents were removed and
traces of pyridine were coevaporated with toluene (3×5 ml). Silica gel chromatography of the
resulting solid (12:1, toluene–MTBE) afforded acetate 7b (267 mg, 89%) as white crystals.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
22
1
7b: Rf = 0.20 (toluene–MTBE, 12:1); mp 167–169 °C, [α]D –36.0 (c 1, CHCl3); H NMR
(600 MHz, CDCl3): δH 7.52–7.43 (4H, m, Ph), 7.35–7.22 (6H, m, Ph), 7.03 (2H, d, J 9.2 Hz,
ACS Paragon Plus Environment

Page 27 of 40
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
MeOC6H4O), 6.82 (2H, d, J 9.2 Hz, MeOC6H4O), 5.52 (1H, s, CHPh), 5.38 (1H, br t, J3,2=J3,4 2.7
Hz, H-3), 5.20 (1H, d, J1,2 1.5 Hz, H-1), 4.92 (1H, d, JBnA,BnB 12.0 Hz, BnA), 4.87 (1H, d, JBnB,BnA
12.0 Hz, BnB), 4.40 (1H, br d, J6A,6B 12.5 Hz, H-6A), 4.11 (1H, dd, J6B,5 2.1 Hz, J6B,6A 12.5 Hz, H-
6B), 3.93 (1H, br d, J4,3=J4,5 1.8 Hz, H-4), 3.78 (4H, m, H-5, OC6H4OCH3), 3.70 (1H, br d, J2,3 3.2
Hz, H-2), 2.1 (3H, s, CH3CO); 13C NMR (150 MHz, CD3Cl) δC 168.9 (CH3CO), 155.1, 151.7
(ipsoOC6H4OCH3), 138.2, 137.7 (ipsoPh (PhCH, Bn), 128.9, 128.1, 128.0, 127.8, 127.3, 126.7 (Ph
(Bn, CHPh)), 118.3, 114.4 (OC6H4OCH3), 101.4 (PhCH), 98.9 (C-1), 72.9 (PhCH2), 72.8 (C-2),
72.0 (C-4), 70.1 (C-3), 69.3 (C-6), 67.0 (C-5), 55.6 (OC6H4OCH3). HRMS (ESI-TOF) m/z
[M+NH4]+ Calcd for C29H34NO8 524.2279; Found 524.2272; [M+Na]+ Calcd for C29H30O8Na
529.1833, found 529.1828.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3-O-Allyl-2-O-benzyl-4,6-O-benzylidene-D-idopyranose (10a). Water (7.4 mL) was added
to a solution of idopyranoside 7a (232 mg, 0.46 mmol) in a mixture of CH3CN (29.4 mL) and
benzene (2.4 mL). The resulting biphasic mixture was cooled to –9 °C. At this temperature, CAN
(1.26 g, 2.3 mmol) was added to the vigorously stirred mixture, and 45 seconds later the reaction
was quenched by adding saturated aqueous NaHCO3. The mixture was diluted with EtOAc (150
mL) and washed with saturated aqueous NaHCO3. The aqueous phase was extracted back with
EtOAc (3×30 mL) and the solvent was evaporated from the combined organic phases. The solid
residue was chromatographed in toluene–MTBE (15:1 → 4:1) to give 10a (164 mg, 90%) as an α,β-
1
mixture in a ratio of 1:1.3; yellowish foam: Rf = 0.20 (toluene–MTBE, 5:1); H NMR (600 MHz,
CDCl3): δH 7.53-7.47 (2H, m, Ph), 7.41-7.22 (8H, m, Ph), 5.94-5.82 (1H, m, CH2=CHCH2O), 5.52
α
(1H, s, CHPh), 5.36 (0.41H, br d, J1,OH1 9.0 Hz, H-1 ), 5.32-5.20 (2H, m, CH2=CHCH2O), 5.01
β
β
(0.57H, br d, J1b,OH1b 12.5 Hz, H-1 ), 4.75 (0.56H, d, JBnA,BnB 11.3 Hz, PhCH2A ), 4.67 (0.83H, br s,
α
β
α,β
CH2Ph ), 4.56 (0.56H, d, JBnB,BnA 11.3 Hz, PhCH2B ), 4.39–4.31 (1H, m, H-6A ), 4.27 (0.42H, br d,
,
α
α
β
α
β
JOH1,H1 9.3 Hz, OH1 ), 4.21-4.00 (4H, m, CH2=CHCH2, H-6B , H-4 , OH1 ), 3.98 (0.42H, br m, H-
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 40
1
β
β
β
α
α
α
2
3
5 ), 3.95-3.88 (1.56H, m, H-3 , H-4 , H-3 ), 3.73 (0.56H, br m, H-5 ), 3.55 (0.42H, br m, H-2 ), 3.50
4
5
(0.56H, br m, H-2 ); 13C NMR (150 MHz, CDCl3): δC 137.9 (ipsoPh), 133.9, 133.4
β
6
7
8
(CH2=CHCH2O), 128.9, 128.3, 128.2, 128.1, 127.9, 127.7, 127.6, 126.5 (Ph), 118.5, 117.9
9
α,β
α,β
α
β
α
β
(CH2=CHCH2 ), 101.3, 101.1 (PhCH ), 94.1 (C-1 ), 92.2 (C-1 ), 74.9 (C-3 ), 74.1 (C-2 ), 73.6 (C-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
or
β
α
α
β
β
α
β
α
3 ), 73.2 (C-2 ), 72.9 (C-4 ), 72.4 (PhCH2 ), 72.2 (C-4 ), 72.1 (CH2=CHCH2 ), 72.0 (PhCH2 ), 71.5
or
α
β
α,β
α
β
(CH2=CHCH2 ), 69.91, 69.88 (C-6 ), 66.4 (C-5 ), 59.4 (C-5 ); Anal. Calcd (%) for C23H26O6: C,
69.33; H, 6.58. Found: C, 69.54; H, 6.77; HRMS (ESI-TOF) m/z [M+NH4]+ Calcd for C23H30NO6
416.2075; Found 416.2068.
3-O-Allyl-2-O-benzyl-4,6-O-benzylidene-α-D-idopyranosyl N-phenyltrifluoroacetimidate
(11a). N-Phenyltrifluoroacetimidoyl chloride (47 µL, 0.18 mmol) and K2CO3 (50 mg, 0.36 mmol)
were added to a solution of 10a (55 mg, 0.14 mmol) in acetone (4.5 mL). The mixture was
intensively stirred until TLC showed disappearance of the starting material (20 h) and then filtered
through a Celite pad. The filtrate was concentrated under reduced pressure and the residue was
purified by column chromatography (PE–EtOAc, 10:1 → 7:1 + 1% (v/v) of Et3N). Chromatographic
purification should not last more than 1 h to minimize hydrolysis of the product 11a on silica gel.
Thus, donor 11a (61 mg, 77%) was obtained as a white foam: Rf = 0.20 (PE–EtOAc, 7:1); 1H NMR
(600 MHz, (CDCl3, 323 K)): δH 7.52, 7.41, 7.36–7.24 (12H, m, Ph (Bn, Ph, NPh)), 7.09 (1H, t,
NPh), 6.85 (2H, d, NPh), 6.11 (1H, br s, H-1), 5.88 (1H, m, CH2=CHCH2O), 5.51 (1H, s, PhCH),
5.27 (1H, m, CH2=CHCH2OA), 5.21 (1H, m, CH2=CHCH2OB), 4.86 (1H, d, JBnA.BnB 12.0 Hz,
PhCH2A), 4.74 (1H, d, JBnB,BnA 12.0Hz, PhCH2B), 4.34 (1H, br d, JH-6A,H-6B 12.8 Hz, H-6A), 4.14-
4.07 (2H, m, CH2=CHCH2), 4.03 (1H, br d, JH-6B,H-6A 12.7 Hz, H-6B), 3.97 (1H, m, H-4), 3.93 (1H,
m, H-3), 3.77 (1H, m, H-2), 3.65 (1H, br m, H-5); 13C NMR (150 MHz, CDCl3): δC 134.0
(CH2=CHCH2O), 129.0, 128.6, 128.3, 128.1, 127.7, 127.6 (Ph (Bn, CHPh, NPh)), 124.0, 119.4
(NPh), 117.7 (CH2=CHCH2), 101.4 (PhCH), 95.0 (C-1), 76.6 (C-3), 73.6 (C-4), 73.2 (C-2), 73.1
ACS Paragon Plus Environment

Page 29 of 40
The Journal of Organic Chemistry
1
2
3
4
5
(PhCH2), 71.4 (CH2=CHCH2), 69.3 (C-6), 67.3 (C5); Anal. Calcd (%) for C31H30F3NO6: C, 65.37;
H, 5.31; N, 2.46. Found: C, 65.42; H, 5.36; N, 2.37.
6
Isopropyl 3-O-allyl-2-O-benzyl-4,6-O-benzylidene-α- and –β-D-idopyranosides (2a and
12a). Donor 11a (59.5 mg, 0.11 mmol) was dried by evaporation of its solution in dry toluene and
then dissolved in CH2Cl2 (0.7 mL, freshly distilled over CaH2) under Ar atmosphere. Powdered
molecular sieves AW-300 (39 mg) were added to the solution of 11a at +10 °C, the resulting
mixture was stirred at this temperature for 1 h, and then i-PrOH (32 µL, 0.42 mmol) was added. The
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
reaction mixture was cooled to –35 °C and a 3% v/v solution of TfOH in CH2Cl2 (15 µL, 5.5 µmol)
was added. One min later, the reaction was quenched by addition of Et3N (4 µL), diluted with
CH2Cl2 (5 mL), and filtered through a Celite pad. The filtrate was diluted with CH2Cl2 (25 mL) and
washed with saturated aqueous NaHCO3. The aqueous phase was extracted back with CH2Cl2 (3×5
mL) and combined organic extracts were taken to dryness. The residue was purified by silica gel
chromatography (PE–EtOAc, 7:1→4:1 + 1% v/v of Et3N) to provide 2a (18.4 mg, 40%) and 12a (21
mg, 46%), both as amorphous solid.
2a: Rf = 0.33 (PE–EtOAc, 5:1); [α]D24 +64.5 (c 1, CHCl3); 1H NMR (600 MHz, (CDCl3)): δH
7.52–7.49 (2H, m, Ph), 7.38–7.25 (8H, m, Ph), 5.94 (1H, m, CH2=CHCH2O), 5.54 (1H, s, PhCH),
5.30 (1H, m, CH2=CHCH2OA), 5.17 (1H, m, CH2=CHCH2OB), 5.01 (1H, d, J1,2 5.1 Hz, H-1), 4.76
(2H, s, PhCH2), 4.26 (1H, dd, J6A,5 1.3 Hz, J6A,6B 12.8 Hz, H-6A), 4.23 (2H, m, CH2=CHCH2), 4.14
(1H, dd, J6B,5 2.7 Hz, H-6B), 4.13 (1H, m, H-4), 4.00 (1H, h, JCH,CH3 6.1Hz, CH(CH3)2), 3.80 (1H,
m, H-5), 3.76 (1H, dd, J3,2 9.0 Hz, J3,4 3.4 Hz, H-3), 3.54 (1H, dd, J2,1 5.1 Hz, J2,3 9.0 Hz, H-2), 1.22
(3H, d, JCH3A,CH 6.2 Hz, CH3A), 1.19 (3H, d, JCH3B,CH 6.2 Hz, CH3B); 13C NMR (150 MHz, CDCl3):
δC 138.5, 138.0 (ipsoPh (PhCH, Bn), 134.9 (CH2=CHCH2O), 128.8, 128.2, 128.1, 127.8, 127.5,
126.2 (Ph (Bn, CHPh)), 116.8 (CH2=CHCH2), 100.3 (C-1), 100.1 (PhCH), 80.3 (C-3), 78.9 (C-4),
78.8 (C-2), 73.7 (PhCH2), 72.0 (CH2=CHCH2), 69.8 (CH(CH3)2), 69.3 (C-6), 62.1 (C-5), 23.7
ACS Paragon Plus Environment