A kinetic study on the reductive opening of the diphenylmethylene acetal in methyl 2,3-O-diphenylmethylene-α-L-rhamnopyranoside

Article abstract of DOI:10.1016/j.carres.2011.04.041

Reductive opening of the diphenylmethyl acetal in methyl 2,3-O-diphenylmethylene-α-L-rhamnopyranoside has been investigated by kinetic studies, and the results have been compared to those recently obtained by quantum chemical calculations. In contrast to

Full text of DOI:10.1016/j.carres.2011.04.041

Carbohydrate Research 346 (2011) 2004–2006
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Note
A kinetic study on the reductive opening of the diphenylmethylene acetal in
methyl 2,3-O-diphenylmethylene-
a
-L-rhamnopyranoside
a
b,c,
b
a
d
c
⇑
}
Dezso Szikra , Attila Mándi
, Anikó Borbás , István P. Nagy , István Komáromi , Attila Kiss-Szikszai ,
Mihály Herczeg b, Sándor Antus b,c
a Department of Physical Chemistry, University of Debrecen, H-4032 Debrecen, Hungary
b Research Group for Carbohydrates of the Hungarian Academy of Sciences, H-4032 Debrecen, Hungary
c Department of Organic Chemistry, University of Debrecen, H-4032 Debrecen, Hungary
d Thrombosis, Haemostasis and Vascular Biology Research Group of the Hungarian Academy of Sciences, University of Debrecen, H-4032 Debrecen, Hungary
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 3 May 2011
Reductive opening of the diphenylmethyl acetal in methyl 2,3-O-diphenylmethylene-a-L-rhamnopyrano-
side has been investigated by kinetic studies, and the results have been compared to those recently
obtained by quantum chemical calculations. In contrast to the previous theoretical calculations which
related only to the presumably rate limiting step of the reductive opening, the reaction system LiAlH4,
AlCl3, and the title compound consists of at least four simultaneous reactions. Nevertheless, reasonable
agreement can be found between the activation Gibbs free energy obtained from kinetic measurements
and the theoretically calculated ones in spite of the experimental errors and the approximate nature of
theoretical calculations.
Dedicated to Professor András Lipták on the
occasion of his 75th birthday
Keywords:
Diphenylmethylene acetal
Reductive opening
Chloroalane
Ó 2011 Elsevier Ltd. All rights reserved.
Kinetics
Reductive opening of cyclic acetals is very important in syn-
thetic carbohydrate chemistry, facilitating selective protection of
polyol systems. Good examples are the 4,6-O-benzylidene acetals
of hexopyranosides,1,2 which can be opened selectively under spe-
cific conditions from both sides yielding 4-O- or 6-O- ethers.3,4 Re-
agents applied for reductive opening usually consist of an
electrophile (a Lewis or a protic acid) and a hydride donor. Mix-
the latter ones were the major products. In the 4-unsubstituted case
a schematic reaction mechanism was proposed in which first a 4-O-
chloroalane derivative (2) forms, followed by the opening of the
dioxolane ring, and finally 3 is hydrolyzed by the addition of water
(Scheme 1).
From the quick H2 formation at the initial phase of the reaction
the conclusion can be drawn that 1?2 reaction step is substan-
tially faster than the 2?3 one, therefore, the 2?3 transformation
can be regarded as the rate limiting step. Recently, a theoretical
study has been presented on the presumed (2?3) rate limiting
step of the reductive opening reaction shown in Scheme 1.12 Based
on the quantum chemical calculations, a modified reaction mech-
anism was proposed for the dioxolane ring opening. It was found
that aluminium spontaneously coordinates to the O3 atom forming
a strong, almost fully developed single bond (a), which leads to
weakening of the O3–Cacetalic bond. There follows a single step
reaction which is, however, rather asynchronous. First the C–O
bond practically vanishes (b), and only then begins the C–H bond
formation (c). Zero point corrected activation energies were found
to be around 125–130 kJ/mol with DFT methods, consistently,
while HF, MP2, and ONIOM calculations resulted in somewhat
higher transition state energies.
5
tures of lithium aluminium hydride (LAH) and AlCl3 are widely
used for this purpose, since in an equilibrium reaction various
alanes can form, depending on the ratio of LAH and AlCl3, differing
in their acidity and hydride-donating ability.6 Very few kinetic
experiments7,8 and theoretical work9 can be found in the literature,
however, for the ring opening of cyclic acetal derivatives, and these
papers deal mainly with benzylidene acetals and borane reagents.
Comparing to the 4,6-O-benzylidene acetals, partial hydrogenoly-
sis of 2,3-O-diphenylmethylene acetals requires milder conditions.10
Thus reductive opening of methyl 2,3-O-diphenylmethylene-a-L-
rhamnopyranoside derivatives was studied by Hajkó et al. applying
LAH and AlCl3 in a 1:1 ratio (i.e., chloroalane) in Et2O/CH2Cl2 1:1.11
While at the 4-unsubstituted derivative (1) only the 2-O-diphenyl-
methyl product was formed, at the 4-OMe or the 4-deoxy cases the
reaction resulted in a mixture of 2-O- and 3-O-benzyl ethers, where
In order to further support the reaction mechanism proposed
earlier,12 we have performed kinetic measurements to determine
the activation Gibbs free energy of the reaction experimentally.
⇑
Corresponding author. Tel.: +36 52 512 900/22257; fax: +36 52 512 900/22342.
E-mail addresses: mandia@delfin.klte.hu, atterl@yahoo.de (A. Mándi).
Methyl 2,3-O-diphenylmethylene-a-L
-rhamnopyranoside (1)13,14
0008-6215/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2011.04.041

D. Szikra et al. / Carbohydrate Research 346 (2011) 2004–2006
2005
OMe
OMe
AlH2Cl
OMe
H2O
OMe
O
O
O
O
HO
HO
O
O
:
:
Al
O
O
HO
a
O
O
O
O
O
b
Cl
Al
H
H2
Ph
Ph
Cl
Ph Ph
Ph
Ph
Ph Ph
c
2
3
4
1
Scheme 1. Previous assumption for the mechanism of the reductive opening of 1.11
was prepared from methyl
a-L-rhamnopyranoside (5) and freshly
OMe
OMe
OMe
prepared benzophenone dimethyl acetal15 in the presence of
LiAlH4
O
O
O
AlCl3
HO
O
LiH3Al
O
ClHAl
( )10-camphorsulfonic acid in MeCN (Scheme 2). Methyl 2-O-
O
O
O
O
O
O
diphenylmethyl-a-L-rhamnopyranoside (4) was synthesized
Ph
Ph
Ph
according to the literature13 with LAH/AlCl3 1:1 in Et2O/CH2Cl2
1:1 to use as a standard for the HPLC measurements.
Ph
Ph
1a
Ph
1
2
For each (1?4) reaction in the kinetic experiments we used
Scheme 3. Formation of alkoxy-aluminium hydride 1a and the chloroalane
derivative 2 in two steps caused by the subsequent addition of LAH and AlCl3.
methyl
2,3-O-diphenylmethylene-
a-
L-rhamnopyranoside
(1),
2 equiv LAH and 2 equiv AlCl3, and the reactions were carried out
at 10, 15, 20, and 25 °C. LAH was added to the sugar solution,
and after 1 min the reaction was started with the addition of AlCl3
(in order to exclude the impact of the first fast step, i.e., the LAH + -
free OH reaction) (Scheme 3). Samples taken at certain times were
analyzed by HPLC.
Based on the previous assumptions11 and the theoretical re-
sults12 the rate determining step of the reaction should follow first
order kinetics. However, our kinetic data contain considerable
experimental errors, which can mainly arise from two factors.
The first one is one of the side reactions, namely the hydrolysis
of the acetal ring, since it seems to be very sensitive to the LAH/
AlCl3 ratio which is also affected by a very small amount of water
may be present in the system. The second one is the ratio of Et2O
and CH2Cl2. Even a slight change of this ratio surprisingly seems
to alter the rate of LAH–AlCl3 induced reductive opening reactions
in a significant manner.16 Further errors can arise i.a. from the rel-
atively circuitous simple preparation process for the HPLC mea-
surements. As a consequence of these facts the order of the
reaction could not be determined with high certainty. Neverthe-
less, the experimental data do not contradict the first order
kinetics.
Figure 1. Rate constant determination for entry 1 at 10 °C (see Table S1).
Accordingly, rate constants for the consumption of 1 for form-
ing 4 were determined supposing first order kinetics. An example
can be seen in Figure 1. Primary kinetic data are listed in
Table S1 in the Supplementary data.
D
H, DS, and DG values have been determined from these rate
constants using the Eyring–Polányi equation (Fig. 2).17–19 Activa-
tion Gibbs free energy was found to be 94.09 kJ/mol.
In this experiment, however, the reaction steps from the full
reaction cannot be separated, that is, it includes the reaction of alk-
oxy-aluminium hydride (1a) with AlCl3 (Scheme 3), the multiple
equilibrium reaction between the excess of LAH and AlCl3, the pre-
sumable rate limiting 2?3 reaction itself and the hydrolysis of the
acetal ring resulting in 5. In other words, not only the formation of
2, but also the side reactions are included in the reaction rate cal-
culated from the experiments. Nevertheless, the ꢀ100 kJ/mol
seems to be the upper limit of the activation free energy of the rate
determining step.
Figure 2. Temperature dependence of the rate constants.
S = À99.32 J/mol K, G = 94.09 kJ/mol at 298.15 K.
DH = 64.47 kJ/mol,
D
D
Comparing the theoretically12 and the experimentally deter-
mined activation parameters, it should be noted, that by calcula-
tions the activation energies corrected by zero point vibrational
energy (ZPVE) were calculated while some kind of overall activa-
tion free energy was obtained from the experiments.
Accepting the calculated ‘overall’ experimental activation free
energy as an activation (Gibbs) free energy for the reaction we
modeled (even though this value is suffering from experimental er-
rors) it is still substantially lower than any of the previously com-
puted zero point ZPVE corrected activation energies.12 It should be
H3CO OCH3
Ph Ph
OMe
OMe
O
HO
O
HO
O
O
HO
Acetonitrile
H+
OH
Ph
Ph
5
1
Scheme 2. Preparation of 1 with transacetalization.

2006
D. Szikra et al. / Carbohydrate Research 346 (2011) 2004–2006
noted, that comparison of the calculated ZPVE-corrected energy
differences (as activation energy) to the activation free energy is
theoretically incorrect. However, at gas phase model, since the
reaction mechanism was predicted to be an intermolecular bond
rearrangement neither substantial entropy effect nor substantial
volume changes are expected and the calculated ZPVE-corrected
energy differences should be close to the theoretical free energy
differences as it was pointed out in our previous paper.12 It is true
even though the harmonic approximation of the low frequency
vibrations can cause large uncertainty in the calculated free en-
ergy, which can be an additional source of differences between
the theoretical and experimental values.
Nevertheless, one of the main reasons of the discrepancy men-
tioned above probably resides in the fact that no solvent molecules
were included in the theoretical calculations. In solvent phase the
polar solvent molecules with non-bonding electron pairs can sol-
vate more effectively the transition state than the reactant due to
the increased hard character of aluminium at transition state
geometry compared to the reactant geometry. (In the latter case
the hydride anion is partially left the aluminium.) This more preva-
lent solvation effect results in both decreased activation energy
and loss of entropy (due to the more ordered solute–solvent inter-
action) at transition state compared to the hypothetical gas phase
reaction.
1.2. Kinetic details
For each (1?4) reaction in the kinetic experiments we used
200 mg methyl 2,3-O-diphenylmethylene-a-L-rhamnopyranoside
(1), 44 mg (2 equiv) LAH, 156 mg (2 equiv) AlCl3, 20 cm3 Et2O
and 20 cm3 CH2Cl2. Reactions were carried out in a double-walled
reaction vessel at 10, 15, 20, and 25 °C under Ar atmosphere. Before
starting the reaction 2 cm3 sample was taken from the sugar solu-
tion. LAH was added and after 1 min the reaction was started with
the addition of AlCl3 (in order to exclude the impact of the first fast
step, i.e., the LAH + free OH reaction). Further samples were taken
at 2, 5, 10, 20, 30, 40, and 60 min. Since the high volatility of the
solvents used would result in a poor reproducible sample volume,
they were corrected by the mass of each sample. The 2–2 cm3 sam-
ples taken at certain times were quenched with a mixture of MeOH
and water, and the reaction solvents were evaporated by the aid of
Ar. Volumes were filled up to 10 cm3 with MeOH–water 80:20, fil-
tered with 0.45 lm syringe filter, and analyzed by HPLC.
Acknowledgments
This work was financially supported by the Hungarian Scientific
Research Fund (OTKA K-62802) and the TÁMOP 4.2.1/B-09/1/
KONV-2010-0007 project. The project is co-financed by the
European Union and the European Social Fund.
Another important reason can be that the previously calculated
rate limiting reaction is strongly asynchronous, since the C–O bond
breaking precedes the C–H bond formation and the transition state
corresponds dominantly to a hydride anion transfer. For such sys-
tems the contribution of tunneling effect is usually non
negligible.20
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.carres.2011.04.041.
In conclusion, reaction kinetics studies on the reductive opening
of methyl 2,3-O-diphenylmethylene-a-L-rhamnopyranoside can be
References
satisfactorily featured by a first order kinetics model which is in a
good accordance with the theoretical reaction mechanism pub-
lished recently.12 The moderate discrepancy between the pub-
lished theoretical and the experimental activation Gibbs free
energy as well as entropy values presented here can be rational-
ized based on the assumption of direct interaction between the sol-
ute and solvent molecules (with non-bonding electron pairs) at the
transition state.
1. Lipták, A.; Jodál, I.; Nánási, P. Carbohydr. Res. 1975, 44, 1–11.
2. Garegg, P. J.; Hultberg, H. Carbohydr. Res. 1981, 93, C10–C11.
3. Stick, R. V. Carbohydrates: The Sweet Molecules of Life; Academic Press: London,
2001. p 53.
4. Protecting Group Manipulations in Carbohydrate Synthesis; Lipták, A., Borbás, A.,
Bajza, I., Eds.Comprehensive Glycoscience I–IV; Kamerling, H., Boons, G. J., Lee,
Y., Suzuki, A., Taniguchi, N., Voragen, A. G. J., Eds.; Elsevier: Amsterdam, 2007;
Vol. 1, p 237.
5. Ashby, E. C.; Prather, J. J. Am. Chem. Soc. 1966, 88, 729–733.
6. Ashby, E. C.; Cooke, B. J. Am. Chem. Soc. 1968, 90, 1625–1630.
7. Johnsson, R.; Cukalevski, R.; Dragen, F.; Ivanisevic, D.; Johansson, I.; Petersson,
L.; Wettergren, E. E.; Yam, K. B.; Yang, B.; Ellervik, U. Carbohydr. Res. 2008, 343,
2997–3000.
1. Experimental
8. Johnsson, R.; Ohlin, M.; Ellervik, U. J. Org. Chem. 2010, 75, 8003–8011.
9. Johnsson, R.; Olsson, D.; Ellervik, U. J. Org. Chem. 2008, 73, 5226–5232.
10. Borbás, A.; Lipták, A. Carbohydr. Res. 1993, 241, 99–116.
11. Hajkó, J.; Borbás, A.; Szabovik, G.; Kajtár-Peredy, M.; Lipták, A. J. Carbohydr.
Chem. 1997, 16, 1123–1144.
12. Mándi, A.; Komáromi, I.; Borbás, A.; Szikra, D.; Nagy, I. P.; Lipták, A.; Antus, S.
Tetrahedron Lett. 2011, 52, 1256–1259.
13. Borbás, A.; Hajkó, J.; Kajtár-Peredy, M.; Lipták, A. J. Carbohydr. Chem. 1993, 12,
191–200.
14. Crich, D.; Li, M. J. Org. Chem. 2008, 73, 7003–7010.
15. Di Donna, L.; Napoli, A.; Siciliano, C.; Sindora, G. Tetrahedron Lett. 1999, 40,
1013–1014.
16. Herczeg, M.; Borbás, A., unpublished work.
17. Eyring, H. J. Chem. Phys. 1935, 3, 107–115.
1.1. General methods
Optical rotations were measured at room temperature with a
Perkin–Elmer 241 automatic polarimeter. Melting points were
determined on a Kofler hot-stage apparatus and are uncorrected.
TLC was performed on Kieselgel 60 F254 (Merck) with detection
by 5% sulfuric acid in ethanol. Column chromatography was per-
formed on Silica Gel 60 (Merck 0.062–0.200 mm). The organic
solutions were dried over MgSO4 and concentrated in vacuum.
The 1H (400.13 and 500.13 MHz) and 13C NMR (100.61 and
125.76 MHz) spectra were recorded with Bruker DRX-400 and
Bruker DRX-500 spectrometers for solutions in CDCl3. Internal ref-
erences: TMS (0.00 ppm for 1H), CDCl3 (77.00 ppm for 13C). HPLC
18. Evans, M. G.; Polányi, M. Trans. Faraday Soc. 1935, 31, 875–894.
19. Lente, G.; Fábián, I.; Poë, A. J. New J. Chem. 2005, 29, 759–760.
20. Kuhn, H.; Försterling, H. D. Principles of Physical Chemistry; John Wiley & Sons:
Chicester, 2000. pp 687–689.
separation conditions: Waters Symmetry C18 column 3.5 lm
4.6 Â 150 mm, MeOH–water 80:20 eluent, flow rate: 0.6 ml/min,
detection: 220 nm.