Enhancing selectivity in oxidation catalysis with sol-gel nanocomposites

Article abstract of DOI:10.1039/b505949f

Valuable organic compounds such as α-hydroxy acids are easily synthesised with relevant selectivity enhancement using a sol-gel hydrophobized nanostructured silica matrix doped with the organocatalyst TEMPO: A materials science based synthetic route which cannot be achieved via classical homogeneous synthesis. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b505949f

View Article Online / Journal Homepage / Table of Contents for this issue
C O M M U N I C A T I O N
Enhancing selectivity in oxidation catalysis with sol–gel
nanocomposites
Pamela Gancitano,a Rosaria Ciriminna,a Maria Luisa Testa,a Alexandra Fidalgo,b Laura M.
Ilharcob and Mario Pagliaro*b
a Istituto per lo Studio dei Materiali Nanostrutturati, CNR, via U. La Malfa 153, 90146,
Palermo, Italy
b Centro de Qu´ımica-F´ısica Molecular, Instituto Superior Te´cnico, Complexo I-Av. Rovisco Pais
1, 1049-001, Lisboa, Portugal. E-mail: mario.pagliaro@ismn.cnr.it
Received 27th April 2005, Accepted 12th May 2005
First published as an Advance Article on the web 23rd May 2005
Table 1 Dihydroxylation of terminal alkenes
Valuable organic compounds such as a-hydroxy acids are
easily synthesised with relevant selectivity enhancement
using a sol–gel hydrophobized nanostructured silica matrix
doped with the organocatalyst TEMPO: A materials science
based synthetic route which cannot be achieved via classical
homogeneous synthesis.
Substratea
Product
Time/min Diol/ketoneb (%)
Materials science, and sol-gel science and technology in
particular,1 are providing a variety of novel solutions to the
requirements of modern organic chemistry where highly selective
and stable catalysts are in urgent demand for replacing stoichio-
metric reagents in many fundamental transformations.2 Several
catalytic processes, therefore, can now be carried out over highly
porous, nanostructured silica oxides doped with one (or more)
reactive catalytic species, with easy recovery (and recycling) of
the precious catalyst at the end of the process.3
0.5
97/3
0.5
94/6
Our approach to heterogenizing catalytic conversions utilizes
doped organically modified silica4 xerogels (ORMOSILs), devel-
oped over the past few years as recyclable catalysts for a variety
of conversions,5 according to processes such as:
3
3
100/0
100/0
CH3Si(OCH3)3 + Si(OCH3)4 + H2O
→ [(CH3)SiOmHn]p + CH3OH
(1)
where in the unbalanced eqn. (1), m < 2, n is always = 0 and the
value of p approaches Avogadro’s number.6 Here, in addition
to describing a convenient catalytic strategy to synthesise
fundamental components in several important compounds with
potent biological activity such as a-hydroxy acids,7 we also
show that significant improvement in the selectivity of the
organocatalyst may be achieved by its entrapment at the inner
surface of a microporous hydrophobized sol–gel silica matrix,
exploiting the spatial restrictions imposed this entrapment.
Our two-step oxidation process involves firstly the syn-
dihydroxylation of a terminal alkene catalyzed by ruthenium,
followed by oxidation of the resulting diol with bleach mediated
by entrapped TEMPO (2,2,6,6-tetramethylpiperidine 1-oxyl
radical).8
Syn-diols 2a–d (Table 1) were thus obtained starting from
terminal alkenes by using the “flash” RuO4-catalyzed dihydrox-
ylation protocol in a biphasic solvent system (ethyl acetate :
acetonitrile : water = 3 : 3 : 1) in the presence of 7% mol RuO4
at 0 ◦C.9
A variety of olefins can be selectively oxidised in the presence
of diverse oxidisible functional groups. Thus, aliphatic alkenes
(1-hexene and 1-hexen-3-ol) were neatly converted into diols 2c
and 2d with no traces of cleavage products; while with aromatic
alkenes the dihydroxylation reaction was even faster and a very
short reaction time was crucial to minimise competitive glycol
cleavage due to RuO4-mediated electrocyclic fragmentation.9b
In the synthesis of diol 2a, for example, 4% mol of RuCl3
and 0.5 min were the optimal reaction parameters to obtain
the diol in high yield, with longer reaction times or the
a All reactions were performed on a 1 mmol scale in a solvent mixture
of ethyl acetate (3 mL)/acetonitrile (3 mL)/water (1 mL) at 0 ◦C using
4–7 mol% RuCl3 (see text) and 1.5 equiv. of NaIO4. b Calculated by
GC-MS.
classical 0.07 equiv. RuO4 leading to ketone as the sole reaction
(fragmentation) product.
To synthesize the corresponding a-hydroxy acids, the diols
thereby obtained were further oxidised with TEMPO/NaOCl
with the idea of exploiting the known selectivity of the nitroso-
nium ion TEMPO+ for primary alcohols.10
Using 10% mol TEMPO dissolved in a 1 : 1 solution of
◦
water : acetonitrile cooled to 0 C and buffered to pH 9.1, the
reaction proceeded with good yields in the case of aliphatic diols
affording the corresponding a-hydroxy acids (79% in the case of
1,2-hexanediol). The oxidation of the aromatic diol 2a (Table 2),
on the other hand, afforded ( )atrolactic acid in 24% yield,
yielding the ketone as the main reaction product (55%).
Such glycol cleavage is due to unselective oxidation of the diol
secondary hydroxyl by TEMPO+,11 which in its turn is favoured
by the activated aromatic substrate. Indeed, when a deactived
diol such as the p-Cl derivative was used as the substrate (2b),
the cleavage was reduced and the acid : ketone ratio (65 : 20)
changed in favour of the acid.
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 3 8 9 – 2 3 9 2
2 3 8 9
©

View Article Online
Table 2 TEMPO mediated oxidation of vic diols to a-hydroxy acids
Substratea
Product
Product yieldb (%) Ketone (%)
2ac
24
60
65
55
40
20
2a
2bc
2b
2c
80
79
5
0
a The reactions were performed on a 1 mmol scale in a solution of water
(6 mL)/acetonirile (6 mL) at 0 ◦C buffered to pH 9.1 using 10 mol%
of ORMOSIL-entrapped TEMPO. b Isolated yield. c The oxidation is
carried out under homogeneous conditions.
Fig. 1 The SEM photograph (52000×) of ORMOSIL-entrapped
TEMPO (above) clearly reveals the nanostructured, spongy morphology
of the material which is completely different to the non-porous smooth
surface of polystyrene beads (below, photo courtesy of Prof. Anthony
Barrett) commonly used in organic synthesis as catalyst support.
Since the homogeneous TEMPO/bleach reaction protocol
applied to the conversion of aromatic diols gives the moderate
selectivity reported above, we turned our attention to our
newly developed sol–gel entrapped TEMPO catalyst.12 Indeed,
we have recently established that in the oxidative synthesis of
(aromatic) amino hydroxy acids from aminodiols catalysed by
TEMPO encapsulated in an hydrophobized silica matrix, the
heterogeneous oxidation over the sol–gel catalyst results in three-
fold selectivity improvement.13
In this case too, the heterogenous oxidation of diol 2a using an
entirely (100%) methyl-modified silica xerogel (SiMe100)14 leads
to an inversion of selectivity compared to the homogeneous
conversion, with an acid : ketone ratio of 60 : 40. Moreover,
diol 2b is now converted almost entirely (80 : 5) into the
corresponding a-hydroxy acid, while conversion of aliphatic
substrates such as 3c proceeds smoothly, also over encapsulated
TEMPO, affording high yields of the acid.
selectivity in comparison to the catalyst either in solution or
surface-bound to a non porous material.
This idea is analogous to Fisher’s “key-in-the-lock” mecha-
nism invoked to explain the selectivity of enzymatic reactions
and is emerging as a general concept16 in the theory of
heterogeneous catalysis.
In the present case, where the selectivity for primary over
secondary hydroxyls is due to steric hindrance associated with
the preferential attack of primary alcohols by TEMPO+,10a
encapsulation of the nitroxyl radical moiety in the restricted
˚
environment provided by the cages (15 A in diameter) further
favours the conversion of less hindered primary hydroxyls.
Furthermore, contrary to what happens with catalyst tether-
ing at a material’s external pore surface, sol–gel encapsulation
ensures effective site isolation and protection. As a consequence,
the nitroxyl radical moieties are protected from intermolecular
quenching which is known to cause activity degradation in
catalysts obtained by supporting TEMPO over commercial
silica.17 In addition, they can be recycled in all seven of the
consecutive reaction cycles in which they are used, showing a
unique activity enhancement upon use which is typical of doped
ORMOSILS employed in liquid-phase catalysis.14
Organic modification of the silica cages is also required, with
the 100% methyl-modified ORMOSIL being significantly more
selective than unmodified SiO2-supported TEMPO.
Indeed, a closer look at the catalyst lamellar shaped cage
surface (by DRIFT spectroscopy, Fig. 2)18 reveals, through the
lower relative intensities of the mO–H and mSi–O bands, that the
cages are depleted of silanol groups. Besides, adsorbed water
is residual, as indicated by the extremely low intensity of the
dH–O–H band. These observations point to a low degree of
hydrophilicity in the modified ORMOSIL network. The inner
As in the case of the oxidation of alcohols14 and
aminohydroxy-alcohols,13 the ORMOSIL catalysts can also be
recycled repeatedly in the oxidation of diols, as consecutive
reaction tests show prolonged intermediate stability (which truly
measures a catalyst’s stability), due to the protecting action of
the sol–gel ORMOSIL cages towards the entrapped nitroxyl
radicals (see below).
These results provide a clear example of enhancement in
selectivity upon entrapment of an organocatalyst in the inner
porosity of a composite material, which is highly relevant to
modern organic chemistry in which solid-phase synthesis is
being widely applied (Fig. 1).15
The effect of the spatial confinement imposed by the narrow
silica pores is to restrict the possible orientations that a reactant
can assume on approaching the catalytic center. Moreover, the
intra-cage silanol groups, which hinder free tumbling of the
substrate molecule, dictate a specific orientational approach to
the catalyst through the hydrogen bonds between the substrate
(the hydroxylated diol) and the pore/cage surface. As a result of
these effects, sol–gel entrapped catalysts generally show higher
2 3 9 0
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 3 8 9 – 2 3 9 2

View Article Online
offering a practical alternative to the use of the cyanohydrins
currently employed in industrial processes.8
Acknowledgements
This work is dedicated with deep affection to Dr Fabrizio
Lo Celso. Thanks to University of Valencia’s Professor Jose
Sepulveda for helpful discussions and to Dr Donatella Capitani
(CNR-IMC, Rome) for the NMR analyses.
Notes and references
1 For an advanced account, see: D. Avnir, L. C. Klein, D. Levy, U.
Schubert and A. B. Wojcik, Organo-silica Sol–Gel Materials, in The
Chemistry of Organosilicon Compounds 2, ed. Z. Rappoport and
Y. Apeloig, Wiley, London, 1998; ch. 40, pp. 2317–2362.
2 While rendering the resulting conversion atom-efficient. For a recent
discussion of this central topic of modern synthetic chemistry, see
D. J. Cole-Hamilton, Science, 2003, 299, 1702.
3 The wide scope and importance of sol–gel doped catalysts to modern
organic synthesis was recently reviewed in: R. Ciriminna and M.
Pagliaro, Curr. Org. Chem., 2004, 18, 1851.
Fig. 2 The DRIFT spectrum of ORMOSIL-entrapped TEMPO clearly
reveals that its pores are depleted of silanol groups.
4 Functionalized silica gels are commercial reagents with a rapidly
growing market. Doped silica gels offer several technical advantages
over well known organic resins being solvent-independent (rigid
porous structure and no swelling) and with a high density of
functional groups (small volume of gel required). Silica works in
all solvents (organic and aqueous), is easy to weigh and handle,
mechanically stable and suitable for scale-up. Sigma-Aldrich, for
instance, commercializes the silica gels produced by SiliCycle Inc.
See also the URL: www.silicycle.com.
cage surfaces not only contain hydrophobic groups but are also
deficient in hydrophilic ones; thus, in the conversion of vic-
diols, the hydrogen-bonding interaction between the hydrophilic
hydroxyl groups in the substrate and the silanol groups at
the cage’s surface is greatly diminished, while the access to
encapsulated TEMPO is spatially restricted by the narrow cages.
Conducting a thorough DRIFT spectral analysis of the
catalysts used in the present study, we have recently shown that
the presence of the co-precursor MTMS affects the structure
and the hydrophilicity/lipophilicity balance (HLB) of the sol–
gel catalyst.19 However, its content does not appreciably affect
the catalyst’s hydrophilicity. In particular, modification with
25% MTMS imparts a significant decrease in the catalyst’s
hydrophilicity without major structural changes; higher mod-
ifier content does not further influence hydrophilicity, but is
responsible for a gradual lipophilicity increase and for striking
structural changes.
The organically modified silica structure is now made of
larger, less strained six-member rings [(SiO)6] (and not by
four-member units, (SiO)4) which are able to accommodate
the unreactive methyl groups better, with the alkyl organic
groups concentrated at the cage surface, diminishing the number
of silanols at the surface and, as a consequence, the intra-
cage hydrogen bonds which limit the freedom of the dopant
molecule.18,6
5 (a) S. Campestrini, M. Carraro, U. Tonellato, M. Pagliaro and
R. Ciriminna, Tetrahedron Lett., 2004, 45, 7283; (b) R. Ciriminna
and M. Pagliaro, Chem. Eur. J., 2003, 9, 5067; (c) R. Ciriminna,
S. Campestrini and M. Pagliaro, Adv. Synth. Catal., 2003, 345,
1261; (d) R. Ciriminna, J. Blum, D. Avnir and M. Pagliaro, Chem.
Commun., 2000, 1441.
6 D. Avnir, Acc. Chem. Res., 1995, 28, 328.
7 (a) K. C. Nicolau, W. M. Dai and R. K. Kuy, Angew. Chem., Int. Ed.
Engl., 1994, 33, 15; (b) S. Shinagawa, T. Kanamaru, S. Harada, M.
Asai and H. Ookazaki, J. Med. Chem., 1987, 30, 1458.
8 The industrial synthesis starts from cyanohydrins. Catalytic routes,
more limited in scope, have also been developed based on supported
metal catalysts: (a) M. Rossi, L. Prati in Green Chemistry: Challenging
perspectives, ed. P. Tundo, P. T. Anastas, Oxford University Press,
Oxford, 2000, p. 183; or enzymes: (b) C. H. Wong and J. R. Matos,
J. Org. Chem., 1985, 50, 1992 and since 2002 the US company Diversa
has commercialized the enzyme nitrilase for the biocatalytic synthesis
of a-hydroxy acids.
9 (a) T. K. M. Shing, E. K. W. Tam, V. W.-F. Tai, I. H. F. Chung and
Q. Jiang, Chem. Eur. J., 1996, 2, 50; the method has been recently
improved with a significant (one seventh) reduction in the amount of
catalytic Ru; (b) B. Plietker and M. Niggemann, Org. Lett., 2003, 5,
3353.
10 (a) A. E. J. de Nooy, A. Besemer and H. van Bekkum, Tetrahedron,
1995, 51, 8023; (b) TEMPO-mediated oxidations of alcohols are now
commercial. The chemical manufacturer DSM, for instance, uses an
oxidation process analogous to that described in R. Ben-Daniel, P.
Alsters and R. Neumann, J. Org. Chem., 2001, 66, 8650; (c) For
a thorough discussion on the oxidation chemistry of TEMPO, see
A. E. J. de Nooy, A. C. Besemer and H. van Bekkum, Synthesis,
1996, 1153.
These findings offer a long awaited solution to the “alkyl
effect” for which 800–1000% activity enhancements are com-
monly observed for catalytic species sol–gel entrapped in
heavily alkylated ORMOSIL matrices.20 This also explains why,
counter to intuition, the xerogels with the highest limitations
to diffusion imposed by the narrow pore network, still posses
the highest reactivity. Indeed, a dramatic increase in the activity
of ORMOSIL-entrapped catalysts is observed exactly when the
transition from four- to six-member rings takes place.
Interestingly, these findings were somewhat anticipated by
molecular orbital calculations that clearly predicted how, sol–
gel amorphous silica being made of siloxane clusters, large
silica clusters with six-membered rings should have “very weak
hydrogen bonding between the two Si–O–H groups” and large
structural flexibility.21
Finally, also of relevance to this report is the fact that the
TEMPO moiety has been entrapped by starting from cheap
and readily available 4-oxo-TEMPO rather than the expensive
radical TEMPO itself, making the use of ORMOSIL-entrapped
TEMPO attractive from an economical viewpoint.22
11 That is, the oxidation of the secondary hydroxyl is also mediated by
TEMPO+, and due to the action of OCl−, as shown in: A. Besemer,
PhD Thesis, Delft University of Technology, 1994.
12 General oxidation procedure: NaIO4 (321 mg, 1.5 mmol) was dis-
solved under stirring in 1 mL of H2O. Next, the solution was cooled
to 0 ◦C and a 0.1 M solution of RuCl3 (400 lL, 0.04 mmol) was added
and the mixture stirred until the color turned bright yellow. Ethyl
acetate (3 mL) and acetonitrile (3 mL) were added, followed by the
olefin (1 mmol). The slurry was stirred until all starting material was
consumed (0.5 to 3 min, depending on the substrate). The reaction
was then quenched adding 5 mL of saturated Na2S2O3 solution, and
the phases were separated, extracting the aqueous layer with ethyl
acetate (3 × 15 mL). After the combined organic layer was dried
over Na2SO4 and the solvent evaporated under reduced pressure, the
crude product was purified by column chromatography on silica gel
using a DCM/ethyl acetate as mobile phase. A solution of the diol
(1 mmol) in water (6 mL)/acetonitrile (6 mL) was then placed in a
In conclusion, the results of the present study further demon-
strate the benefits of anchoring the TEMPO catalyst inside
nanocages of porous silica for which the surface HLB can be
tailored to the requirements of an organic catalytic synthesis,
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 3 8 9 – 2 3 9 2
2 3 9 1

View Article Online
thermostated reaction vessel, with 135 mg of ORMOSIL-supported
TEMPO (0.1 mmol of TEMPO, according to microanalysis). After
the reaction mixture was cooled to 0 ◦C, an aqueous solutionof
NaOCl (2.5 mL, ca. 13% w/w) was added, buffered to a pH of 9.1 by
further addition of 5% NaHCO3 (2 mL). The reaction mixture was
vigorously shaken until all substrate was consumed (TLC of reaction
mixture samples). After filtration, the filtrate was acidified adding
10 mL of a 1 M tartaric acid solution, and the resulting solution
extracted with ethyl acetate (3 × 15 mL). The combined organic
phases were dried over Na2SO4, the solvent was removed in vacuo,
and the crude product was purified by column chromatography
on silica gel using a CH2Cl2–ethyl acetate mixture as eluent.
General catalyst preparation: methyltrimethoxysilane (MTMS), 3-
aminopropyl-trimethoxysilane (APTMS), tetramethyl-orthosilicate
(TMOS), 4-oxo-2,2,6,6-tetramethyl-1-piperidinoloxy free radical (4-
oxo TEMPO), and methanol were purchased from Sigma-Aldrich
and used without further purification. Ultra pure water (Millipore
Type I quality) was used in all the preparations. The chemical
entrapment of the TEMPO moiety in the sol–gel silica hybrid
materials was performed in two steps: reductive ammination of 4-
oxo-TEMPO with APTMS, followed by sol–gel polycondensation
of methyltrimethoxysilane (MTMS) and TMOS in the presence of
the TEMPO funcionalized alkoxide thereby obtained. The chosen
molar ratio was Si : MeOH:H2O = 1 : 2 : 3 for all of the catalyst
preparations. A typical catalytic ORMOSIL doped with TEMPO,
such as Si75Me, was obtained by dissolving APTMS (4 mmol) in
MeOH (2.6 mL), bringing the pH to 7 with HCl (conc.), and then
adding 4-oxo-TEMPO (1 mmol) along with NaBH3CN (0.5 mmol).
After 48 h under fast stirring at room temperature, the NaBH3CN
excess was destroyed by adding HCl (49 mmol) and a precursor
mixture made of TMOS (1.750 mL), MTMS (4.370 mL), MeOH
(2.730 mL) and H2O (5.8 mL) was added under fast stirring. The sol
gelled slowly and the alcogel obtained was sealed, left to age at room
temperature for 24 h and eventually dried at 60 ◦C for five days. The
resulting powder xerogel was washed three times with DCM under
reflux and dried again at 60 ◦C. A typical catalytic load was 0.3 mmol
TEMPO g−1
.
13 M. L. Testa, R. Ciriminna, C. Hajji, E. Zaballos-Garcia, M. Ciclosi,
J. Sepu´lveda-Arques and M. Pagliaro, Adv. Synth. Catal., 2004, 346,
655.
14 The first use of these ORMOSILS in the oxidation of alcohols is
reported in: R. Ciriminna, C. Bolm, T. Fey and M. Pagliaro, Adv.
Synth. Catal., 2002, 344, 159.
15 P. Seneci, Solid-Phase Synthesis and Combinatorial Technologies,
John Wiley and Sons, New York, 2000.
16 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.
17 T. Fey, H. Fischer, S. Bachmann, K. Albert and C. Bolm, J. Org.
Chem., 2001, 66, 8154.
18 A. Fidalgo and L. M. Ilharco, Chem. Eur. J., 2004, 10, 315.
19 R. Criminna, L. M. Ilharco, A. Fidalgo, S. Campestrini, M. Pagliaro,
Soft Matter, (submitted).
20 For instance, also in the case of catalytic processes mediated by
entrapped precious metals, see 5b, or by enzymes see: M. T. Reetz,
P. Tielmann, W. Wiesenhofer, W. Konen and A. Zonta, Adv. Synth.
Catal., 2003, 345, 717.
21 S. Hayakawa and L. L. Herch, J. Non-Cryst. Solids, 1998, 242, 131.
22 4-oxo-TEMPO is easily obtained by triacetonamine (TAA), a
polymer stabilizer marketed in thousands of tons yearly by large
companies such as Degussa at 3 $ kg−1; whereas the current market
price of TEMPO is 80–100 $ kg−1 due to the expensive reduction step
(with NaBH4) of the carbonyl group in 4-oxo-TEMPO. In order to
be adopted by industry, the price of TEMPO will have to be around
20 $ kg−1: a requirement clearly met by sol–gel entrapment of 4-oxo-
TEMPO. W. Kesber, personal communication, (prices as of March
23, 2005); M. Pagliaro, D. Avnir, J. Blum, G. Deganello, US Pat.
6797773 B1, 2004 .
2 3 9 2
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 2 3 8 9 – 2 3 9 2