Unexpectedly superior enantioselectivity for trans-stilbene cis-dihydroxylation over anchored triosmium carbonyl species in confined Al-MCM-41 channels

Article abstract of DOI:10.1039/b417787h

Superior enantioselectivity in the dihydroxylation of trans-stilbene catalysed by anchored triosmium carbonyl species without using a chiral modifier is observed inside sterically congested MCM-41 channels; this effect is more pronounced through the introduction of surface Al sites into the silicate. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b417787h

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Unexpectedly superior enantioselectivity for trans-stilbene
cis-dihydroxylation over anchored triosmium carbonyl species in
confined Al-MCM-41 channels{
Vale´rie Caps,{ Ioannis Paraskevas and Shik Chi Tsang*
Received (in Cambridge, UK) 25th November 2004, Accepted 25th January 2005
First published as an Advance Article on the web 7th February 2005
DOI: 10.1039/b417787h
ratio of 5. The binding energy of 103.6 eV for Si 2p3/2 was similar
to the one found in pure silica MCM-41 (103.6 eV). The binding
energy of 75.4 eV for Al 2p also corresponded to a highly oxidised
state of Al (binding energy for Al0 is 72.9 eV), which was expected
to be more acidic than in aluminas (74.7 eV) and aluminosilicates
(75.0 eV). Thus, this Al-MCM-41 surface contains a high
concentration of highly electron deficient, i.e. strongly Lewis
acidic, Al sites.
Superior enantioselectivity in the dihydroxylation of trans-
stilbene catalysed by anchored triosmium carbonyl species
without using a chiral modifier is observed inside sterically
congested MCM-41 channels; this effect is more pronounced
through the introduction of surface Al sites into the silicate.
Catalytic asymmetric dihydroxylation of olefins to produce 1,2-
diols is a key reaction in the fine chemicals industry. The
3-component process using a catalytic amount of soluble osmium
tetroxide together with an oxidant and a pro-chiral ligand is well-
documented.1 However, the difficulty in recovering the expensive
chiral ligand and the soluble Os species of an extremely toxic
nature prompts economic and environmental concerns with this
traditional process. Heterogenisation of this catalytic system is thus
highly desirable.2 Work has been reported on the immobilisations
of chiral ligand(s) onto a polymer,3 silica4 or SBA-15.5 It is noted
that, using this approach, the toxic OsO4 is still primarily used as a
soluble species. Systems where catalytic osmium species are
immobilised through precipitation onto nitrogen-containing poly-
mers,6 microencapsulation inside polymers7 or ion-exchange onto
charged ligands covalently bound to resin or silica,8 followed by
using soluble pro-chiral ligands, have also been reported.
Os3(CO)12 (99%) was purchased from Strem Chemicals, stored
in a dry, air-tight glove-box to avoid air/moisture contamination
and used without further purification. It was then applied to the
freshly prepared MCM-41 or Al-MCM-41 support using a
previously reported CVD technique.9 Basically, all manipulations
were carried out under an inert atmosphere or vacuum. The
support was dried by heating under vacuum for 4 h at 150 uC. A
predetermined amount of the Os3(CO)12 precursor (32 mg) was
placed in a Schlenk tube together with 200 mg of the dried
support. The tube was then evacuated, sealed and heated at 120 uC
for 40 h to allow for the reaction between the metal precursor and
the support to take place. Unattached species were subsequently
removed by heating in a vacuum at 120 uC for 1 h. After the
reactor tube was cooled to room temperature and exposed to air,
the resulting material is referred to as the ‘‘as-synthesised’’ (AS)
sample. A subsequent calcination of the AS samples to 220 uC in
air for 2 h was also carried out for comparison. Detailed
methodologies concerning testing the catalyst materials for the
reaction and product analysis can be found in the ESI.{
In this communication, chemical vapour deposition (CVD) of
Os3(CO)12 onto dried, freshly prepared MCM-41 or Al-MCM-41
mesoporous silicate was carried out; the Os loaded sample was
then tested for the dihydroxylation of trans-stilbene in liquid phase
using N-methylmorpholine N-oxide as an oxidant. We noted an
unexpectedly high enantioselectivity for this catalysed reaction,
without using a chiral ligand or oxidant.
It is noted from Table 1 that all these immobilised Os3(CO)12-
derived species, with and without calcination, catalyse, in the
presence of the N-methylmorpholine N-oxide oxidant, the
oxidation of trans-stilbene to 1,2-diphenyl-1,2-ethanediol with a
high yield (.64%). Greater than 78% selectivity to the principal
MCM-41 was prepared using a standard method as previously
reported.9 Highly acidic Al-MCM-41 was prepared by adding
aluminium isopropoxide in the hydrothermal synthesis of the
mesoporous silicate.10 The transmission electron micrographs of
Table 1 Stereoselectivity in the dihydroxylation of the trans-stilbene
over Os3(CO)12 species immobilised by CVD inside mesoporous
silicates
this material showed the same hexagonal array of channels of 30–
2
40 A, as MCM-41. The BET surface area (1233 m g21) was
˚
slightly below that of MCM-41 (1344 m2 g21). The actual
incorporation of aluminium inside the silicon framework was
evidenced by FTIR. Elemental analysis gave a Si/Al molar ratio of
3 : 4, indicating a high Al content. XPS gave a surface Si/Al atomic
Selectivity (%)
Yield of
diol (%) meso (R,R) (S,S)
Catalyst
32mgOs3(CO)12/MCM-41[AS]b
32mgOs3(CO)12/Al-MCM-41[AS]b
32mgOs3(CO)12/MCM-41[220C]c
68.9
64.0
80.4
0
0
0
0
22.4
5.2
45.9
47.5
77.6
94.8
54.1
52.5
{ Electronic supplementary information (ESI) available: characterisation
of Os3(CO)12 deposited MCM-41 and related structures. See http://
www.rsc.org/suppdata/cc/b4/b417787h/
{ Present address: CNRS, Institut de Recherches sur la Catalyse, 2
Avenue Albert Einstein, 69626 Villeurbanne Cedex, France. E-mail:
valerie.caps@catalyse.cnrs.fr
32mgOs3(CO)12/Al-MCM-41[220C]b 75.9
a
Reaction conditions: substrate 1 mmol, NMO 1 mmol, 18 mg Os
catalyst in acetone (18 cm3) and water (2 cm3) at 40 uC for 96 h
(refer to ESI). Os contents determined in solids ca. 7 wt% and
3 wt% (derived from TGA).
b
c
*s.c.e.tsang@reading.ac.uk
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 1781–1783 | 1781

View Article Online
diol product with minor selectivities towards benzaldehyde,
benzoin and benzyl was achieved.9 Analysis of the diol product
suggests that this heterogeneous reaction is a diastereospecific
reaction (no meso configuration of the diol is observed), like the
homogeneous counterpart, where a concerted [3 + 2] cycloaddition
of the carbon–carbon double bond onto the single Os catalytically
active site via two oxygen atoms is proposed. When the MCM-41-
or Al-MCM-41-immobilised Os3(CO)12 CVD samples are calcined
at 220 uC (2 h), they give a nearly racemic mixture of the syn
configurations (R,R) and (S,S) despite their different Os contents
(3 wt% and 7 wt%, respectively). However, it is surprising to
observe a high enantiomeric excess (ee) towards the (S,S)
configuration of the 1,2-diphenyl-1,2-ethanediol over the as-
synthesised catalysts with no chiral ligand nor oxidant blended
in the reaction medium. Careful repeated measurements of the
chirality of products confirm reproducibility of the results (see
ESI{). To ensure that the catalysis is heterogeneous in nature, the
solid and filtrate (colourless) from the product solution were
separated from each other and repeatedly tested by adding more
batches of the same substrate contents. No activity (,1%
conversion) was found using the filtrate. Instead, the solid catalyst
gave a lower yield (41.5%) on the second test indicating a degree
of deactivation but a very similar (R,R) and (S,S) product
distribution was achieved. In addition, a soluble Os3(CO)12
(13 mmol) species without adding chiral modifier to replace the
solid catalyst was tested under identical conditions. A very poor
selectivity to the diol (56.3%) with selectivities to benzaldehyde
(40.6%), benzoin (2.6%) and benzil (0.5%), respectively were
obtained. Analysing the diol clearly showed the racemic mixture in
this homogeneous system, as expected. Thus, as noted from
Table 1, the (S,S) configuration is favoured by 55% on using the
MCM-41 solid support and by 90% with the Al-MCM-41
support. This 90% ee towards the (S,S) configuration is higher
than those reported using a homogeneous OsO4–NMO–acetone :
water system containing a chiral ligand (78%).11 It is noted that the
highest reported chirality is obtained using an Os catalytic system
with a chiral modifier in ionic liquids which can give 94–97% ee
towards the (S,S) configuration.12
Fig. 1 (a) FTIR spectrum of pure MCM-41(upper curve); (b) FTIR
spectrum of 32mgOs3(CO)12/MCM-41[AS] (lower curve).
into an Os–Os bond to yield the trinuclear anchored complex
(m–H)Os3(CO)10(m–OSi–).
However, careful study of CO2 evolution during temperature
programmed decomposition (TPDe) of the anchored species in a
flowing stream of helium suggested more than one type of surface
species.9,14 In particular, the high temperature shoulder peak at
400 uC becomes more prominent in those samples with lower
initial Os3(CO)12 loadings (¡3 wt%, see ESI{) or when using the
Al-MCM-41 support with more reactive acidic sites, as presently
shown in Fig. 2. It is believed that, in these cases, the triangular
clusters are likely to align themselves "parallel" to the MCM-41
channel surface, with possibly more than one anchor point per
FTIR characterisations of the as-synthesised MCM-41-
immobilised Os3(CO)12 compared with MCM-41 show a strong
and characteristic absorption band near 1382 cm21 (Figs. 1a and
b). This band is attributed to Os–H–Os vibrations.9 More than 8
absorption peaks including 2124, 2116, 2076, 2067, 2026, 2017,
1995 and 1987 cm21 vibrations are found in the enlarged CO-
stretching frequency region (Fig. 1b). These absorption peaks
outnumber those of pure Os3(CO)12 (2069, 2037, 2016 and
2005 cm21 of D3h), which implies a decrease in the symmetry
when the molecule is adsorbed on the MCM-41 surface or a
simultaneous presence of more than one surface species or very
likely a result of both. Immobilisation of Os3(CO)12 via CVD onto
various amorphous inorganic oxides including silica was widely
reported in the eighties.13 Detailed characterisation, including the
uses of EXAFS and FTIR, showed that (m–H)Os3(CO)10(m–OSi–)
species, with the triangular osmium framework anchored upright
on the silica surface, can be synthesised at high surface coverage.13
Many of our observed FTIR vibrations indeed match the reported
values, indicating that the volatile Os3(CO)12 is likely to react with
the MCM-41 silicate in a similar way as with those silicas,
probably through oxidative addition of surface silanol groups
Fig. 2 Temperature programmed decomposition (CO2 evolution mon-
itored by on-line mass spectrometry) of anchored Os3(CO)12 on MCM-41
and Al-MCM-41 (as-synthesised samples) obtained by ramping the
samples at 10 K min21 in He (10 mL min21).
1782 | Chem. Commun., 2005, 1781–1783
This journal is ß The Royal Society of Chemistry 2005

View Article Online
trinuclear cluster from the surface oxygen (eliminating more CO
from the complex). It is noted that TGA-DTA (not shown) indeed
indicates similar Os content (7 wt%) for both the Al-MCM-41 and
the MCM-41 supports but the signal of the CO2 evolution (Fig. 2)
is clearly reduced in this acidic catalyst, which suggests the presence
of less carbonyl species available from the Al-MCM-41 support. A
similar surface species of a tri-Os carbonyl cluster with a bidentate
alkenes in liquid phase. The precise nature of the surface chiral
species is yet to be disclosed.
We are grateful to D. Law for help with SCIENTA ESCA300
X-ray photoelectron spectrometry at the RUSTI Laboratory
(UK), A. Munday for elemental analysis and Drs G. Blond, C.-M.
Yang and E. Toukoniitty for helpful discussions. Further
characterisations of the anchored Os samples can be found in
the Electronic Supplementary Information (ESI{).
oxygen donor from
a silica surface has been previously
suggested.13 On the other hand, after calcination of the as-
synthesised samples at 220 uC, the numbers of CO stretching
bands in both MCM-41/Al-MCM-41 samples have been
substantially decreased as compared to the free Os3(CO)12
cluster. Three principal FTIR CO-stretching vibrations are
now observed at 2130, 2040 and 2000 cm21, in agreement with
the previous observations that heat treatment of silica anchored
(m–H)Os3(CO)10(m–OSi–) species can produce mononuclear sur-
face species Os(CO)x (x 5 2 or 3) with C2v or C3v symmetry
through the cluster degradation.9,13
Vale´rie Caps,{ Ioannis Paraskevas and Shik Chi Tsang*
The Surface and Catalysis Research Centre, School of Chemistry,
University of Reading, Whiteknights, Reading, UK RG6 6AD.
E-mail: s.c.e.tsang@reading.ac.uk; Fax: +44 118 931 6632;
Tel: +44 118 931 6346
Notes and references
1 H. C. Kolb, M. S. VanNieuwenhze and K. B. Sharpless, Chem. Rev.,
1994, 94, 2483; L. Ahrgren and L. Sutin, Org. Process Res. Dev., 1997, 1,
425.
From the above characterisation, it may not be possible to work
out the precise origin of the exciting enantioselectivity observed in
the as-synthesised mesoporous samples. However, it is apparently
clear that the mononuclear Os species anchored on the MCM-41
surface give a racemic mixture with no enantioselectivity in trans-
stilbene oxidation. On the other hand, the trinuclear Os species
anchored onto MCM-41, and in particular onto Al-MCM-41
(which accommodates more trinuclear Os species), give a very high
enantioselectivity for the same reaction. It has recently been
reported that placing chiral catalysts in confined porous structures,
such as mesoporous silicates, could significantly boost the ee
values.15 Thus, it is thought that our impressive ee in the case of
Al-MCM-41 is a result of the introduction of surface Al sites into
MCM-41 for immobilisation of more trinuclear species, hence
generating a more sterically crowded environment at higher
coverage. Perhaps the most intriguing difference between our
results and the literature is that no chiral ligand/modifier or
oxidant is used in our case. Thus, we believe that upon
immobilisation on the MCM-41 or related surfaces, a favourable
chiral orientation of the triosmium complex precursor must have
taken place, resulting in a new surface chiral catalytic species.
In summary, we have made an unprecedented observation that
heterogenisation of Os3(CO)12 on the internal surface of Al-MCM-
41 using simple chemical vapour deposition gives a superior
enantioselectivity in the dihydroxylation of trans-stilbene without
adding a chiral modifier. The present enantiomeric excess observed
reaches 90%, rendering these materials potential heterogeneous
catalysts akin to enzymes for the asymmetric dihydroxylation of
2 D. E. De Vos, B. F. Sels and P. A. Jacobs, Adv. Catal., 2002, 46, 1;
A. Severeyns, D. E. De Vos and P. A. Jacobs, Top. Catal., 2002, 19, 125.
3 D. Pini, A. Petri, A. Nardi, C. Rosini and P. Salvadori, Tetrahedron
Lett., 1991, 32, 5175; P. Salvatori, D. Pini and A. Petri, J. Am. Chem.
Soc., 1997, 119, 6929; C. Bolm and A. Gerlach, Eur. J. Org. Chem.,
1998, 21.
4 C. Bolm, A. Maischak and A. Gerlach, Chem. Commun., 1997, 2353.
5 I. Motorina and C. M. Crudden, Org. Lett., 2001, 3, 2325.
6 W. A. Herrmann, R. M. Kratzer, J. Blu¨mel, H. B. Friedrich,
R. W. Fischer, D. C. Apperley, J. Mink and O. Berkesi, J. Mol.
Catal. A: Chem., 1997, 120, 197.
7 S. Kobayashi, T. Ishida and R. Akiyama, Org. Lett., 2001, 3, 2649.
8 B. M. Choudary, N. S. Chowdari, M. L. Kantam and K. V. Raghavan,
J. Am. Chem. Soc., 2001, 123, 9220; B. M. Choudary, N. S. Chowdari,
K. Jyothi and M. L. Kantam, J. Am. Chem. Soc., 2002, 124, 5341.
9 V. Caps, I. Paraskevas and S. C. Tsang, Appl. Catal., A, 2003, 252, 37.
10 A. Matsumoto, H. Chen, K. Tsutsumi, M. Gru¨n and K. Unger,
Microporous Mesoporous Mater., 1999, 32, 55.
11 H.-L. Kwong, C. Sorato, Y. Ogino, H. Chen and K. B. Sharpless,
Tetrahedron Lett., 1990, 31, 2999.
12 C. E. Song, Chem. Commun., 2004, 1033–1043.
13 R. Psaro, C. Dossi, A. Fiso and R. Ugo, Res. Chem. Intermed., 1991, 15,
31; J. Evans, Spectrochim. Acta, 1987, 43, 1511.
14 V. Caps and S. C. Tsang, Appl. Catal., A, 2003, 248, 19.
15 B. F. G. Johnson, S. A. Raynor, D. S. Shephard, T. Mashmeyer,
J. M. Thomas, G. Sankar, S. Bromley, R. Oldroyd, L. Gladdenc and
M. D. Mantlec, Chem. Commun., 1999, 1167; J. M. Thomas,
T. Maschmeyer, B. F. G. Johnson and D. S. Shephard, J. Mol.
Catal. A: Chem., 1999, 141, 139; S. A. Raynor, J. M. Thomas, R. Raja,
B. F. G. Johnson, R. G. Bell and M. D. Mantlec, Chem. Commun.,
2000, 1925; R. Raja, J. M. Thomas, M. D. Jones, B. F. G. Johnson and
D. E. W. Vaughan, J. Am. Chem. Soc., 2003, 125, 14982; M. D. Jones,
R. Raja, J. M. Thomas, B. F. G. Johnson, D. W. Lewis, J. Rouzaud
and K. D. M. Harris, Angew. Chem., Int. Ed., 2003, 42, 4326.
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 1781–1783 | 1783