Ruthenium-catalysed epimerisation of carbohydrate alcohols as a method to determine the equilibria for epimer interconversion in hexopyranosides

Article abstract of DOI:10.1002/ejoc.201100504

Ruthenium-catalysed epimerisation of secondary carbohydrate alcohols was used to determine the thermodynamic equilibrium between non-anomeric epimers of partially protected glucose, mannose and allose derivatives. A cyclopentadienylruthenium catalyst was used to epimerise each of two pure epimeric alcohols in two separate experiments. The epimerisation reactions were run until the same ratio of epimers was obtained from the two experiments. Secondary carbohydrate alcohols can be epimerised using a cyclopentadienyl ruthenium complex. In this way, the relative stabilities of different hexopyranoside configurations can be determined.

Full text of DOI:10.1002/ejoc.201100504

FULL PAPER
DOI: 10.1002/ejoc.201100504
Ruthenium-Catalysed Epimerisation of Carbohydrate Alcohols as a Method to
Determine the Equilibria for Epimer Interconversion in Hexopyranosides
Clinton Ramstadius,[a][‡] Annika M. Träff,[a][‡] Patrik Krumlinde,[a] Jan-E. Bäckvall,*[a] and
Ian Cumpstey*[a,b]
Keywords: Carbohydrates / Ruthenium / Hydrogen transfer / Epimerisation
Ruthenium-catalysed epimerisation of secondary carbo-
hydrate alcohols was used to determine the thermodynamic
equilibrium between non-anomeric epimers of partially pro-
tected glucose, mannose and allose derivatives. A cyclo-
pentadienylruthenium catalyst was used to epimerise each
of two pure epimeric alcohols in two separate experiments.
The epimerisation reactions were run until the same ratio of
epimers was obtained from the two experiments.
Introduction
Transition metal complexes can catalyse the redox in-
terconversion of alcohols and ketones in reactions that are
conceptually related to, but mechanistically distinct from,
the classical Meerwein–Ponndorf–Verley–Oppenauer reac-
tion,[1] with transition metal hydrides being formed as inter-
mediates.[2] An oxidation–reduction (dehydrogenation–hy-
drogenation) sequence results in the racemisation (or epi-
merisation) of alcohols. Beyond its basic importance, the
homogeneous catalytic hydrogen-transfer redox reaction
has also been coupled with other concepts in organic chem-
istry to give methods for alcohol functionalisation, as for
example with enzymatic kinetic resolution to give a dy-
namic kinetic resolution,[3] or for the formation of C–N,
C–O or C–C bonds by a sequence of oxidation–activation,
coupling and re-reduction.[4] In particular, dynamic kinetic
Figure 1. Ruthenium complexes used in the racemisation or epi-
merisation of secondary alcohols.
resolution based on the racemisation (or epimerisation) of of green chemistry.[9] They are also very rich in alcohol
secondary alcohols with ruthenium catalysts 1a–c (Fig- functionality, so the reactivity, and especially the catalytic
ure 1) has found many applications in enantioselective syn- reactivity, of carbohydrate alcohols is an important area for
thesis.[3] The catalyst 1c, activated with tBuOK, has been research. The Oppenauer oxidation is not used for carbo-
shown to have exceptional reactivity, racemising, for exam- hydrate alcohols,[10,11] and judging by earlier reports in the
ple, 1-phenylethanol with t1/2 Ͻ2 min at room temp. with literature, secondary carbohydrate alcohols appear to be
0.5 mol-% catalytic loading.[5] Hence this catalyst was cho- very unreactive towards oxidation by transition-metal com-
sen for the study described in this paper. The reaction plex-catalysed hydrogen transfer: it has been shown that the
mechanism of racemisation of secondary alcohols by cata- anomeric hemiacetal of carbohydrates may be oxidised to
lyst 1c has been studied in some detail,[6–8] and an overview the lactone by rhodium or ruthenium complexes operating
is given in Scheme 1.
by a hydrogen-transfer mechanism, without affecting un-
Carbohydrates are a cheap, renewable raw material, and protected secondary hydroxy groups, confirming their low
so are of great interest for the future from the point of view reactivity.[14–19] One paper does describe low-yielding oxi-
dation of OH-5 in a glucofuranose derivative by dehydroge-
nation with a RuH2(PPh3)4 catalyst.[19] Also, with hetero-
[a] Department of Organic Chemistry, The Arrhenius Laboratory,
Stockholm University,
geneous catalysts some reactivity has been reported.[20,21]
10691 Stockholm, Sweden
E-mail: jeb@organ.su.se
ian.cumpstey@sjc.oxon.org
The reason for this low reactivity is probably primarily
electronic in origin: The transfer of hydride from the
alcohol C atom to the metal centre is rate-determining for
electron-deficient alcohols.[28] It is known that electron-
[b] Institut de Chimie des Substances Naturelles, CNRS,
91198 Gif-sur-Yvette CEDEX, France
[‡] C. R. and A. M. T. contributed equally to this paper.
Eur. J. Org. Chem. 2011, 4455–4459
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4455

J.-E. Bäckvall, I. Cumpstey et al.
FULL PAPER
Scheme 1. Mechanistic overview of the epimerisation/racemisation reaction as catalysed by the ruthenium complex 1c.
withdrawing substituents on the alcohol moiety disfavour is 2.5:1 (using 0.05 equiv. catalyst),[30] whereas the nickel-
this process,[28] and the polyoxygenated nature of the carbo- catalysed route gives a ratio around 1.4:1–1.7:1 (using
hydrate means that secondary alcohol functionalities are ex- 1 equiv. catalyst).[31] In these C-2 epimerisation reac-
pected to be electron-deficient and their dehydrogenation to tions,[30–32] a free hemiacetal is required, meaning that as
be slow. Steric arguments against coordination of the well as the interconversion of C-2 epimers (and possibly
alcohol to the metal are almost certainly less important: other side-reactions), the situation is complicated by (rapid)
secondary carbohydrate alcohols are sterically hindered, equilibration of the anomers at C-1.
but bulky tert-butoxide is known to be a ligand during the
activation of the ruthenium epimerisation catalyst complex
1c (Scheme 1).[7] The presence of multiple Lewis basic coor-
dination sites in carbohydrates may also play some role in
Results and Discussion
their low reactivity towards redox catalysts.
We began by examining the reactivity of a readily avail-
In this paper, we describe the ruthenium-catalysed epi- able model secondary carbohydrate alcohol, the mono-un-
merisation of secondary carbohydrate alcohols. Using this protected furanose derivative diacetone glucose 2, and its
reaction, we are able to experimentally address the position C-3 epimer 3, with ruthenium catalyst 1c (activated with
of equilibrium and hence the relative stabilities of different tBuOK). Despite the very high reactivity of this catalyst in
hexopyranosides, as well as a hexofuranose derivative. Mea- the epimerisation of secondary alcohols,[5] no reaction was
surement of the equilibrium configuration of secondary hy- seen with the carbohydrate alcohol 2 in toluene after several
droxy groups (i.e. not the anomeric position) around the hours at room temperature.
pyranose ring, and thus the relative thermodynamic stabili-
Nevertheless, heating the carbohydrate alcohol 2 with the
ties of the different pyranoside configurations, is a matter catalyst did result in some epimerisation, and after long re-
of fundamental importance, especially given the importance action times at high temperature (120 °C), with a change of
of carbohydrates, and pyranosides in particular, in nature. solvent to p-xylene, significant reaction was seen, giving a
However, it is not a straightforward task, and information 70:30 mixture of diastereomers 2 and 3, as judged by the
about non-anomeric configurational equilibria has been ob- 1H NMR spectrum of the crude product mixture. Repeating
tained only in some special cases. An enzyme, UDP-glu- the reaction starting from the C-3 epimer 3 gave the same
cose-4Ј-epimerase, catalyses the reversible interconversion mixture of diastereomers, indicating that the thermo-
of (anomerically locked) α UDP-glucose and α UDP-galac- dynamic equilibrium for the epimerisation reaction had
tose, allowing a direct measurement of the relative stabilities been reached (Table 1, entry 1). We went on to examine
of α-glucose and α-galactose derivatives.[29] The equilibrium a series of pairs of epimeric secondary mono-unprotected
constant has been estimated to be K ≈ 3 in favour of the α- pyranoside alcohols. Each of the alcohols 4–11[33] were ex-
glucose derivative. On unprotected sugars, the molybde- posed to the epimerisation catalyst in p-xylene at 120 °C
num-catalysed Bilik reaction[30] and nickel-mediated reac- (Table 1, entries 2–5).
tions[31] interconvert 2-epimeric aldoses by carbon-skeleton-
In most cases (Table 1, entries 1, 2, 4 and 5) the same
rearrangement mechanisms. The equilibrium ratio of free product composition, within experimental error, was
glucose and mannose as measured by the Bilik procedure reached from both directions (i.e., starting from each of the
4456
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 4455–4459

Ru-Catalysed Epimerisation of Carbohydrate Alcohols
Table 1. Reactions were carried out with (η5-C5Ph5)Ru(CO)2Cl (1c) (0.05 equiv.), tBuOK (0.06 equiv.) in p-xylene at 120 °C for the times
stated.
[a] Ratio determined by integration of signals in the 1H NMR spectrum of the crude reaction product mixture. [b] Two additional
portions of catalyst (0.05 equiv. each) were added during the reaction as thermodynamic equilibrium between the interconverting epimers
was not reached with the initial amount (0.05 equiv.) of catalyst: starting from 6 with only 0.05 equiv. catalyst, after 48 h the ratio 6/7
was 95:5; starting from 7 with only 0.05 equiv. catalyst, after 96 h the ratio 6/7 was 55:45; starting from 8 with only 0.05 equiv. catalyst,
after 48 h the ratio 8/9 was 95:5; starting from 9 with only 0.05 equiv. catalyst, after 120 h the ratio 8/9 was 62:38.
two epimers) indicating that equilibrium had been reached
(Figure 2). In one case (Table 1, entry 3), equilibrium was
not reached from at least one direction, but the dia-
stereomeric ratio of 6/7 was 93:7 and 88:12 when starting
from 6 or 7, respectively, indicating that a diastereomeric
mixture close to the equilibrium composition had been
reached. The reactions were quite clean: by-product forma-
tion was minimal. The catalyst is apparently deactivated un-
der the harsh reaction conditions and long reaction times
used; for two of the pairs of interconverting epimers (viz 6–
7 and 8–9), the same composition was not reached from
both directions with 0.05 equiv. catalyst, but the addition
of further catalyst at intervals during the reaction induced
further reaction and progress towards equilibrium (Table 1,
entries 3 and 4).
The alcohols with OH-2 free (4–7, Table 1, entries 2 and
3) equilibrate between gluco and manno configurations. En-
try 2 clearly shows that the α-manno-configured alcohol 5
has a similar stability to the α-gluco-configured alcohol 4 ing from 4; d) crude product mixture starting from 5.
1
Figure 2. Extracts of the 400 MHz H NMR spectra of a) isolated
4; b) isolated 5; c) crude product mixture from epimerisation start-
Eur. J. Org. Chem. 2011, 4455–4459
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4457

J.-E. Bäckvall, I. Cumpstey et al.
FULL PAPER
(K393 K ≈ 0.8). In fact, the equilibrium is shifted slightly in indeed this was seen in most cases, while in one case, a
favour of the manno derivative. This is a counterintuitive diastereomeric mixture close to the equilibrium composi-
result if diaxial interactions are the main consideration. tion was obtained. The reaction requires somewhat forcing
However, on the top face of a pyranoside ring, diaxial inter- conditions, but the carbohydrate derivatives appear to be
actions are lower in energy than those in a cyclohexane due stable and by-product formation was minimal. Equilibrium
to the absence of axial substituents on the ring oxygen and constants are expected to change with temperature and sol-
also due to the slightly distorted ring pucker. For the β- vent, and the protecting group pattern and the presence of
gluco–β-manno interconversion (Table 1, entry 3), equilib- between 5 and 15 mol-% of the catalytic complex may also
rium was not reached, but we can say that here the β-gluco influence the position of equilibrium to some extent. Hence
alcohol 6 is more stable than the β-manno alcohol 7 with the values obtained here cannot be treated as absolute. Nev-
the equilibrium constant in the range 7 Յ K393 K Յ 13. The ertheless, as far as we are aware, the thermodynamic equi-
difference between the behaviour of the α (4 and 5) and β librium positions of α-gluco vs. α-manno, of β-gluco vs. β-
(6 and 7) pairs is a manifestation of the Δ2 effect, which manno, or of gluco vs. allo interconversions have not been
refers to the observed greater difference in stability of the α measured experimentally before, which lends the results re-
over the β anomer for mannose than for glucose.[34–36] This ported here some significance.
has been explained as being due to destabilisation of the
β-manno configuration due to the close proximity of three
oxygen atoms (O-1, O-2 and O-5) or by an enhanced stabili-
sation of α-manno configuration due to the 180° angle be-
tween neighbouring dipoles (C1–O1 and C2–O2).
The pyranoside alcohols with OH-3 free (8–11, Table 1,
entries 4 and 5) equilibrate between gluco and allo configu-
rations. Entry 5 shows that at equilibrium, the β-allo-con-
Experimental Section
General Procedures for Epimerisation: All epimerisation reactions
were carried out under dry argon atmosphere using standard
Schlenk technique. All syringes used were purged with Ar three
times before use. All glassware was flame-dried and left to cool to
room temp. in a desiccator and then placed under Ar. Addition of
figured alcohol 11 is less stable than the β-gluco alcohol 10
with an equilibrium constant of K393 K = 2.8Ϯ0.2. In entry
4, equilibrium was reached after addition of a total of
0.15 equiv. of the catalyst, and the α-allo alcohol 9 is less
extra ruthenium complex (η5-C5Ph5)Ru(CO)2Cl (1c) was carried
out under an argon atmosphere in a glove box. The p-xylene was
dried with sodium wire. Dry THF was obtained from a VAC sol-
vent purifier. Ruthenium complex (η5-C5Ph5)Ru(CO)2Cl (1c) was
stable than the α-gluco alcohol 8 with an equilibrium con- synthesised according to a literature procedure.[8]
stant of K393 K = 8.1Ϯ0.5. Hence for both anomers the
For Carbohydrate Alcohols that were Solids: The carbohydrate
gluco configuration is more stable than the allo, but the dif-
ference in stability is greater between the α compounds (8
and 9) than the β compounds (10 and 11), which is consis-
alcohol (37 mg, 0.10 mmol) and (η5-C5Ph5)Ru(CO)2Cl (1c)
(3.2 mg, 0.005 mmol, 0.05 equiv.) were added to a flame-dried
Schlenk tube (10 mL), which was purged with Ar three times. p-
tent with a steric clash between the axial OH-3 and the Xylene (1 mL) was added by syringe and the resulting solution was
stirred under Ar at room temp. for 10 min. tBuOK (12 μL of a
0.5 m solution in THF, 0.006 mmol, 0.06 equiv.) was then added,
and the solution typically immediately changed colour from pale
yellow to orange-yellow. After 10 min, the reaction mixture was
submerged in an oil-bath at 120 °C and left to stir for 24, 48, 72
or 120 h. After this time, the reaction mixture was allowed to cool
to room temp., transferred to a round-bottomed flask (rinsing with
CH2Cl2) and concentrated in vacuo. The crude reaction product
was dried on a vacuum pump and then analysed by 1H NMR spec-
troscopy.
axial anomeric oxygen in the α-allo compound 9 that is not
present in the β-allo derivative 11.
In their experiments on the epimerisation of unprotected
methyl glycopyranosides with Raney nickel, Koch and Stu-
art,[24,25] and later Perlin,[22] observed that thermodynamic
equilibrium was not reached (i.e. the same composition of
the product mixture was not reached starting from each of
the components – indeed the same mixture was not seen
starting from any two of the components tested), but it is
noteworthy that significant α-GlcǞα-Gal and α-GlcǞα-
Man epimerisation away from the α-glucoside was seen.
This observation, along with the data from the skeletal re-
arrangement reactions[30,31] (see above) and now our result
for the α-Glc–α-Man equilibrium position, is at odds with
For Carbohydrate Alcohols that were Oils: The carbohydrate
alcohol was dissolved in the appropriate volume of p-xylene to give
a 0.1 m stock solution, and then stirred over molecular sieves (4 Å)
for 3 h. (η5-C5Ph5)Ru(CO)2Cl (1c) (3.2 mg, 0.005 mmol,
0.05 equiv.) was added to a flame-dried Schlenk tube (10 mL),
an assumption that α-gluco is obviously the most stable which was purged with Ar three times. The p-xylene solution
(1 mL) of the carbohydrate alcohol was added to this tube by sy-
ringe, and the resulting solution was stirred under Ar at room temp.
for 10 min. tBuOK (12 μL of a 0.5 m solution in THF, 0.006 mmol,
0.06 equiv.) was then added, and the solution typically immediately
changed colour from pale yellow to orange-yellow. After 10 min,
the reaction mixture was submerged in an oil-bath at 120 °C and
left to stir for 24, 48, 72 or 120 h. After this time, the reaction
mixture was allowed to cool to room temp., transferred to a round-
bottomed flask (rinsing with CH2Cl2) and concentrated in vacuo.
configuration for glycopyranosides.
Conclusions
The results described in this article represent one of the
first reports of the reaction of secondary carbohydrate
alcohols in transition-metal complex-catalysed hydrogen-
transfer redox reactions.[19] The epimerisation reaction is
The crude reaction product was dried on a vacuum pump and then
1
expected to give a thermodynamic mixture of products, and analysed by H NMR spectroscopy.
4458
www.eurjoc.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2011, 4455–4459

Ru-Catalysed Epimerisation of Carbohydrate Alcohols
[19] G. Descotes, D. Sinou, J.-P. Praly, Carbohydr. Res. 1980, 78,
25–32.
[20] A. de Raadt, M. H. Fechter, in: Topics in Current Chemistry,
215, Glycoscience: Epimerisation, Isomerisation and Rearrange-
ment Reactions of Carbohydrates (Ed.: A. E. Stütz), Springer,
Heidelberg, Germany, 2001, pp. 327–345.
Procedure for the Addition of Extra Catalyst: The reaction mixture
was allowed to cool to room temp. Additional ruthenium complex
(η5-C5Ph5)Ru(CO)2Cl (1c) (3.2 mg, 0.005 mmol, 0.05 equiv.) was
added under an argon atmosphere in a glove box. The reaction
vessel was sealed and removed from the glove box. Additional
tBuOK (12 μL, 0.5 m in dry THF, 0.006 mmol, 0.06 equiv.) was
then added to the reaction mixture. The reaction mixture was left
to stir at room temp. for 10 min and was then submerged in an
oil-bath at 120 °C and left to stir. This procedure was repeated as
required.
[21] Extensive epimerisation of the hydroxy groups of unprotected
glycosides has been observed over Raney nickel after several
days at 100 °C,[22–25] and also under hydrogenation conditions
over copper chromite under high pressure at 180 °C (along with
reduction products).[26] It has been suggested that these reac-
tions go by dehydrogenation/re-hydrogenation mechanisms. In
the heterogeneous oxidation (dehydrogenation) of carbo-
hydrates over platinum, the primary C-6 is oxidised, and the
unprotected secondary alcohols are unaffected.[27]
.
Acknowledgments
[22] F. Balza, A. Perlin, Carbohydr. Res. 1982, 107, 270–278.
[23] S. J. Angyal, J. D. Stevens, L. Odier, Carbohydr. Res. 1987, 189,
151–157.
[24] H. J. Koch, R. S. Stuart, Carbohydr. Res. 1978, 67, 341–348.
[25] H. J. Koch, R. S. Stuart, Carbohydr. Res. 1977, 59, C1–C6.
[26] P. A. J. Gorin, Can. J. Chem. 1960, 38, 641–651.
[27] H. van Bekkum, in: Carbohydrates as Organic Raw Materials
(Ed.: F. W. Lichtenthaler), VCH, Weinheim, Germany, 1991,
pp. 289–310.
[28] B. Martín-Matute, J. B. Åberg, M. Edin, J.-E. Bäckvall, Chem.
Eur. J. 2007, 13, 6063–6072.
[29] D.-F. Fan, D. S. Feingold, Plant Physiol. 1969, 44, 599–604.
[30] M. L. Hayes, N. J. Pennings, A. S. Serianni, R. Barker, J. Am.
Chem. Soc. 1982, 104, 6764–6769.
[31] a) S. Osanai, in: Topics in Current Chemistry, 215, Glycoscience:
Epimerisation, Isomerisation and Rearrangement Reactions of
Carbohydrates (Ed.: A. E. Stütz), Springer, Heidelberg, Ger-
many, 2001, pp. 43–76; b) S. Osanai, K. Inaba, S. Yoshikawa,
Carbohydr. Res. 1991, 209, 289–295.
[32] The Lobry de Bruyn Alberda van Ekenstein reaction also in-
terconverts 2-epimeric aldoses, but this reaction operates with
a different mechanism involving deprotonation at C-2, and also
results in the formation of a large amount of the 2-ketose, see:
S. J. Angyal, in: Topics in Current Chemistry, 215, Glycoscience:
Epimerisation, Isomerisation and Rearrangement Reactions of
Carbohydrates (Ed.: A. E. Stütz), Springer, Heidelberg, 2001,
pp. 1–14.
[33] The carbohydrate starting materials used have been prepared
earlier in the literature. 3: J. D. Stevens, in: Methods in Carbo-
hydrate Chemistry (Eds.: R. L. Whistler, J. N. BeMiller), Aca-
demic Press, Orlando, FL, 1972, vol. 6, pp. 123–128; 4: A.
Francais, D. Urban, J.-M. Beau, Angew. Chem. Int. Ed. 2007,
46, 8662–8665; 5: S. C. Ennis, I. Cumpstey, A. J. Fairbanks,
T. D. Butters, M. Mackeen, M. R. Wormald, Tetrahedron 2002,
58, 9403–9412; 6 and 7: J. G. M. van der Ven, J. C. H. M. Wijk-
mans, J. P. Kamerling, J. F. G. Vliegenthart, Carbohydr. Res.
1994, 253, 121–140; 8: D. E. Ward, B. F. Kaller, J. Org. Chem.
1994, 59, 4230–4238; 9: C.-W. T. Chang, B. Elchert, Tetrahe-
dron Lett. 2001, 42, 7019–7024; 10: A. Roeen, J. I. Padron, J. T.
Vazquez, J. Org. Chem. 2003, 68, 4615–4630; 11: C. Grandjean,
G. Lukacs, J. Carbohydr. Chem. 1996, 15, 831–856.
[34] Y. Wang, H.-S. Cheon, Y. Kishi, Chem. Asian J. 2008, 3, 319–
326.
Belén Martín Matute carried out the preliminary experiments on
this project. Financial support from the Swedish Research Council
and INTENANT is gratefully acknowledged.
[1]
[2]
[3]
C. F. de Graauw, J. A. Peters, H. van Bekkum, J. Huskens, Syn-
thesis 1994, 1007–1017.
O. Pàmies, J.-E. Bäckvall, Chem. Eur. J. 2001, 7, 5052–5058,
and references cited therein.
a) O. Pàmies, J.-E. Bäckvall, Chem. Rev. 2003, 103, 3247–3262;
b) B. Martín-Matute, J.-E. Bäckvall, Curr. Opin. Chem. Biol.
2007, 11, 226–232; c) J. H. Lee, K. Han, M.-J. Kim, J. Park,
Eur. J. Org. Chem. 2010, 999–1015.
[4]
[5]
G. E. Dobereiner, R. H. Crabtree, Chem. Rev. 2010, 110, 681–
703.
a) B. Martín-Matute, M. Edin, K. Bogar, J.-E. Bäckvall, An-
gew. Chem. Int. Ed. 2004, 43, 6535–6539; b) G. Csjernyik, K.
Bogár, J.-E. Bäckvall, Tetrahedron Lett. 2004, 45, 6799–6802.
M. C. Warner, O. Verho, J.-E. Bäckvall, J. Am. Chem. Soc.
2011, 133, 2820–2823.
J. B. Åberg, J. Nyhlén, B. Martín-Matute, T. Privalov, J.-E.
Bäckvall, J. Am. Chem. Soc. 2009, 131, 9500–9501.
B. Martín-Matute, M. Edin, K. Bogar, F. B. Kaynak, J.-E.
Bäckvall, J. Am. Chem. Soc. 2005, 127, 8817–8825.
F. W. Lichtenthaler, S. Mondel, Pure Appl. Chem. 1997, 69,
1853–1866.
[6]
[7]
[8]
[9]
[10] M. L. Sinnott, Carbohydrate chemistry and biochemistry: struc-
ture and mechanism, RSC, London, 2007, pp. 587–588.
[11] Aldose to ketose isomerisations by hydrogen transfer mediated
by samarium or magnesium compounds have been re-
ported.[12,13]
.
[12] M. Adinolfi, A. Iadonisi, L. Mangoni, Tetrahedron Lett. 1996,
37, 5987–5988.
[13] G. Casiraghi, G. Rassu, M. Cornia, F. Nicotra, J. Chem. Soc.,
Chem. Commun. 1989, 1242–1243.
[14] M. Saburi, Y. Ishii, N. Kaji, T. Aoi, I. Sasaki, S. Yoshikawa, Y.
Uchida, Chem. Lett. 1989, 563–566.
[15] I. Isaac, I. Stasik, D. Beaupère, R. Uzan, Tetrahedron Lett.
1995, 36, 383–386.
[16] M. Bierenstiel, M. Schlaf, Eur. J. Org. Chem. 2004, 1474–1481.
[17] I. Isaac, G. Aizel, I. Stasik, A. Wadouachi, D. Beaupère, Syn-
lett 1998, 475–476.
[35] R. E. Reeves, J. Am. Chem. Soc. 1950, 72, 1499–1506.
[36] A. Neuberger, A. P. Fletcher, Carbohydr. Res. 1971, 17, 79–88.
Received: April 8, 2011
[18] G. Descotes, J.-P. Praly, D. Sinou, J. Mol. Catal. 1979, 6, 421–
430.
Published Online: June 8, 2011
Eur. J. Org. Chem. 2011, 4455–4459
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4459