Platinum-containing hyper-cross-linked polystyrene as a modifier-free selective catalyst for L-sorbose oxidation

Article abstract of DOI:10.1021/ja0107834

Impregnation of hyper-cross-linked polystyrene (HPS) with tetrahydrofuran (THF) or methanol (ML) solutions containing platinic acid results in the formation of Pt(II) complexes within the nanocavities of HPS. Subsequent reduction of the complexes by H

Full text of DOI:10.1021/ja0107834

10502
J. Am. Chem. Soc. 2001, 123, 10502-10510
Platinum-Containing Hyper-Cross-Linked Polystyrene as a
Modifier-Free Selective Catalyst for L-Sorbose Oxidation
†
†
†
†
†
S. N. Sidorov, I. V. Volkov, V. A. Davankov, M. P. Tsyurupa, P. M. Valetsky,
,‡
‡
‡
§
§
L. M. Bronstein,* R. Karlinsey, J. W. Zwanziger, V. G. Matveeva, E. M. Sulman,
§
⊥
,⊥
N. V. Lakina, E. A. Wilder, and R. J. Spontak*
Contribution from the NesmeyanoV Institute of Organoelement Compounds, Moscow 117813, Russia,
Department of Chemistry, Indiana UniVersity, Bloomington, Indiana 47405, TVer Technical UniVersity,
TVer 170000, Russia, and Departments of Chemical Engineering and Materials Science & Engineering,
North Carolina State UniVersity, Raleigh, North Carolina 27695
ReceiVed March 26, 2001
Abstract: Impregnation of hyper-cross-linked polystyrene (HPS) with tetrahydrofuran (THF) or methanol (ML)
solutions containing platinic acid results in the formation of Pt(II) complexes within the nanocavities of HPS.
Subsequent reduction of the complexes by H2 yields stable Pt nanoparticles with a mean diameter of 1.3 nm
in THF and 1.4 nm in ML. The highest selectivity (98% at 100% conversion) measured during the catalytic
oxidation of L-sorbose in water is obtained with the HPS-Pt-THF complex prior to H2 reduction. During an
induction period of about 100 min, L-sorbose conversion is negligible while catalytic species develop in situ.
The structure of the catalyst isolated after the induction period is analyzed by X-ray diffraction, transmission
electron microscopy, and X-ray photoelectron spectroscopy. Electron micrographs reveal a broad distribution
of Pt nanoparticles, 71% of which measure less than or equal to 2.0 nm in diameter. These nanoparticles are
most likely responsible for the high catalytic activity and selectivity observed. The formation of nanoparticles
measuring up to 5.9 nm in diameter is attributed to the facilitated intercavity transport and aggregation of
smaller nanoparticles in swollen HPS. The catalytic properties of these novel Pt nanoparticles are highly robust,
remaining stable even after 15 repeated uses.
Introduction
phiphilic block copolymer micelles in selective solvents, for
instance, can be used to control the size, size distribution, shape,
and morphology of nanoparticles. The tailored synthesis of well-
defined amphiphilic block copolymers is, however, extremely
demanding due to the required purity of all the reactant species.
This practical shortcoming drives the ongoing search for more
robust and inexpensive polymeric matrices exhibiting nano-
Noble metal nanoparticles formed in various media and
stabilized by different mechanisms remain a subject of intense
study due to their promising catalytic properties, which are often
1
-4
superior to those of traditional heterogeneous catalysts.
Polymers in solutions and in bulk are often employed to stabilize
such nanoparticles. Over the past few years, reports of controlled
nanoparticle formation in nanostructured polymeric media,
structures that function in both aqueous and organic media. One
such matrix is hyper-cross-linked polystyrene (HPS).1
2,13
Due
5-11
especially macromolecular surfactants, have appeared.
Am-
to its high cross-link density, which can exceed 100%, HPS
consists of nanosized rigid cavities of comparable size in the
*
To whom correspondence should be addressed.
Nesmeyanov Institute of Organoelement Compounds.
Indiana University.
Tver Technical University.
North Carolina State University.
†
2-3 nm range. It is readily produced by chemically incorporat-
‡
§
⊥
ing methylene groups between adjacent phenyl rings in dissolved
polystyrene homopolymer or gelled poly(styrene-r-divinylben-
zene) copolymer in the presence of ethylene dichloride. Good
solvent and conformational network rigidity are generally
necessary to prepare hyper-cross-linked polymers possessing
(
1) Schmid, G., Ed. Clusters and Colloids; VCH: Weinheim, Germany,
1
994.
(
2) Baiker, A.; Grunwaldt, J.-D.; M u¨ ller, C. A.; Schmid, L. Chimia 1998,
5
2, 517.
(
3) Bradley, J. S. Schr. Forschungszent. J u¨ lich, Mater. 1998, 1 (Physik
der Nanostrukturen), D6.1.
4) B o¨ nnemann, H.; Braun, G.; Brijoux, W.; Brinkmann, R.; Tilling, A.
S.; Seevogel, K.; Siepen, K. J. Organomet. Chem. 1996, 520, 143.
extremely high inner surface areas (typically in the range of
2
1
000 m /g). In the present study, we use HPS with the following
(
characteristics: (i) a formal degree of cross-linking of 200%,
2
(
5) Antonietti, M.; Wenz, E.; Bronstein, L.; Seregina, M. AdV. Mater.
(ii) an apparent inner surface area of 833 m /g, (iii) a sharp
1
995, 7, 1000.
pore size distribution maximum at about 2 nm in diameter, and
(
4.
6) Chan, Y. N. C.; Schrock, R. R.; Cohen, R. E. Chem. Mater. 1992, 4,
(iv) particle sizes ranging from 0.2 to 0.4 mm in diameter.
2
(
7) Chan, Y. N. C.; Craig, G. S. W.; Schrock R. R.; Cohen, R. E. Chem.
A unique feature of HPS is its ability to swell in a wide
variety of different solvents, even thermodynamically poor ones
e.g., water). Such versatility greatly facilitates incorporation
Mater. 1992, 4, 885.
8) Saito, R.; Okamura, S.; Ishizu, K. Polymer 1992, 33, 1099. Saito,
R.; Okamura, S.; Ishizu, K. Polymer 1993, 34, 1183, 1189.
9) Moffitt, M.; McMahon, L.; Pessel, V.; Eisenberg, A. Chem. Mater.
995, 7, 1185.
(
(
(
of various organometallic compounds into the nanostructured
1
(
10) Spatz, J. P.; Roescher, A.; M o¨ ller, M. AdV. Mater. 1996, 8, 337.
(12) Davankov, V. A.; Tsyurupa, M. P. React. Polym. 1990, 13, 27.
(13) Tsyurupa, M. P.; Davankov, V. A. J. Polym. Sci.: Polym. Chem.
Ed. 1980, 18, 1399.
(11) Bronstein, L. M.; Chernyshov, D. M.; Valetsky, P. M.; Wilder, E.
A.; Spontak, R. J. Langmuir 2000, 16, 8221.
1
0.1021/ja0107834 CCC: $20.00 © 2001 American Chemical Society
Published on Web 10/05/2001

L-Sorbose Oxidation
J. Am. Chem. Soc., Vol. 123, No. 43, 2001 10503
Scheme 1
Scheme 2
exhibits both high catalytic activity and selectivity in the
oxidation of L-sorbose.
Results and Discussion
Characteristics of HPS-Pt Nanocomposites. Incorporation
of Pt complexes into the nanostructured matrix of HPS is
achieved by swelling HPS particles in the presence of THF or
ML solutions containing H2PtCl6. During solvent removal under
vacuum at ambient temperature, the color of the HPS powder
changes for samples prepared in THF (HPS-Pt-THF) and ML
HPS matrix. A preceding study14 addressing the formation of
Co nanoparticles in HPS has demonstrated that the HPS cavities
effectively restrict metal nanoparticle growth, thereby providing
precise control over particle size and shape. On the basis of
these results, we have surmised that this polymer matrix could
also be used to control the formation of noble metal nanopar-
(HPS-Pt-ML) solutions. This color change is inferred to be
indicative of a chemical transformation in the functionality of
2
1,22
the Pt compound. As reported earlier,
impregnation of
1
1
ticles prepared upon reduction of precursor metal complexes,
H2PtCl6 into carbon and pregraphitized carbon black in the
presence of water promotes the reduction of Pt(IV) to Pt(II),
along with the simultaneous oxidation of the carbon, as depicted
in Scheme 2. The detailed mechanism by which Pt(IV) is
reduced and carbon is oxidized is discussed elsewhere.22 In the
present work, we presume that HPS can be envisaged as
pregraphitized carbon containing both arene cycles and CH
groups, and should therefore induce the same chemical response
elicited by the reaction shown in Scheme 2.
thus expediting the development of new nanocomposite systems
with interesting catalytic properties. Moreover, since HPS swells
in most solvents, access of reactant species to catalytic sites in
a self-supporting substrate would be provided in virtually any
reaction medium.
In the present work, we examine the incorporation of
H2PtCl6 in HPS and the subsequent formation and catalytic
property development of Pt nanoparticles. An example of a
catalytic reaction of fundamental, as well as technological,
interest is the oxidation of L-sorbose to 2-keto-L-gulonic acid.
Several strategies based on chemical, electrochemical, biotech-
nological, and catalytic methods are currently available for the
oxidation of this ketose, which serves as an intermediate in
Vitamin C production. While the catalytic route promises to be
the most viable, it has been thwarted by complications. Due to
acetonation, carbohydrate functional groups are initially pro-
tected from oxidation, in which case the reaction selectivity
increases. The initial functional groups must, however, be
According to XPS data collected from both HPS-Pt-THF and
HPS-Pt-ML samples, the binding energy of Pt 4f7/2 in each
material is virtually identical at 73.6 ( 0.1 eV, which is
measurably lower than that of Pt(IV). For comparison, the
tabulated Pt 4f7/2 binding energy ranges for K2Pt(IV)Cl6,
K2Pt(II)Cl4, and Pt(0) are 74.1-74.3, 72.8-73.4, and 71.0-
71.3 eV, respectively. On the basis of these data, it is reasonable
to conclude that H2PtCl6 undergoes a chemical transformation
within the HPS matrix, resulting in the reduction of Pt(IV) to
Pt(II). Deconvolution of the XPS spectra presented in Figure 1
reveals that almost all the Pt(IV) is reduced to Pt(II) in the HPS-
Pt-THF sample (Figure 1a), whereas a residual fraction of Pt-
recovered after oxidation, resulting in tremendous loss of end-
product.1
5,16
Direct catalytic oxidation of L-sorbose to 2-keto-
17-20
L-gulonic acid by the reaction
displayed in Scheme 1 avoids
(IV) remains in the HPS-Pt-ML sample (Figure 1b).
acetonation altogether. This reaction has been performed in the
presence of O2 over a Pt or Pd catalyst deposited on activated
carbon or aluminum oxide. In this case, the reaction must be
conducted in neutral or low-alkali media. Selectivity may be
increased through the addition of modifying (promoting) agents
such as phosphine and aminophospine complexes, as well as
aromatic and cycloaliphatic amines, at optimal loading levels.
The resultant selectivity unfortunately decreases from 95% at
Compositional data acquired from elemental analysis of HPS
and its Pt derivatives are presented in Table 1. Inconsistency in
the overall composition makeup of HPS (the compositions sum
to only 98.72 wt %) is attributed to the presence of a small
amount of oxygen-containing groups, which account for about
13
1
.3 wt % oxygen. Solid-state C NMR spectra (Figure 2) have
been collected from these materials under magic angle spinning
13
(
MAS) conditions with cross polarization. The C MAS
30% conversion to 40% at 100% conversion. Thus, the two
spectrum of metal-free HPS (Figure 2a) primarily shows signals
arising from di- and trisubstituted arenes (at 125-140 ppm),
as well as from CH and CH2 groups (at 42 ppm). The
corresponding FTIR spectrum of metal-free HPS displays
several prominent signals that are inconsistent with the idealized
model in which HPS consists of only CH and CH2 groups and
1,2,3-substituted phenyl rings. These include a signal located
principal disadvantages of this procedure are (i) low selectivity
at high conversion and (ii) contamination of the end-product
with reaction modifiers. The present work details the synthesis
of a modifier-free catalyst based on Pt-containing HPS, which
(14) Sidorov, S. N.; Bronstein, L. M.; Davankov, V. A.; Tsyurupa, M.
P.; Solodovnikov, S. P.; Valetsky, P. M.; Wilder, E. A.; Spontak, R. J.
Chem. Mater. 1999, 11, 3210.
-1
-1
at 910 cm , as well as a weaker one at 990 cm , both of
which can be assigned to -CHdCH2 groups.
(
15) Lyazidi, H. A.; Benabdallah, M. Z.; Berlan, J.; Kot, C.; Fabre, P.-
L.; Mestre, M.; Fauvarque, J.-F. Can. J. Chem. Eng. 1996, 74, 405.
16) Cognet, P.; Berlan, J.; Lacoste, G.; Fabre, P.-L.; Jud, J.-M. J. Appl.
Electrochem. 1995, 25, 1105.
(
Incorporation of Pt into HPS is accompanied by a measurable
increase in oxygen content: up to 2.82 wt % in HPS-Pt-THF
and 4.49 wt % in HPS-Pt-ML (see Table 1). The solid-state
C NMR spectrum of the HPS-Pt-THF material is presented
in Figure 2c and exhibits a new signal (relative to the metal-
(
17) Mallat, T.; Br o¨ nnimann, C.; Baiker, A. Appl. Catal. A-Gen. 1997,
1
49, 103.
1
3
(
18) Mallat, T.; Br o¨ nnimann, C.; Baiker, A. J. Mol. Catal. A-Chem.
1
997, 117, 425.
(
19) Br o¨ nnimann, C.; Bodnar, Z.; Aeschimann, R.; Mallat, T.; Baiker,
A. J. Catal. 1996, 161, 720.
20) Br o¨ nnimann, C.; Mallat, T.; Baiker, A. J. Chem. Soc.-Chem.
Commun. 1995, 1377.
(21) van Dam, H. E.; van Bekkum, H. J. Catal. 1991, 131, 335.
(22) Coloma, F.; Sep u´ lveda-Escribano, A.; Fierro, J. L. G.; Rodr ´ı guez-
Reinoso, F. Langmuir 1994, 10, 750.
(

10504 J. Am. Chem. Soc., Vol. 123, No. 43, 2001
SidoroV et al.
sample to THF extraction for 2 h under agitation, which yielded
a dark brown THF solution that was subsequently evaporated
and dried in vacuo (1 mbar) at ambient temperature. The FTIR
-
1
spectrum of the resultant extract in the range of 400-250 cm
-
1
(
the far-infrared) reveals two signals at 330 and 305 cm . On
the basis of their intensity and position, they are assigned to
-
23
the [PtCl3L] ion, where L denotes the ligand. A counterion
+
should be [H3O] . Note that similar signals are observed in the
spectrum of HPS-Pt-THF. Several ligand structures can be
hypothesized. According to XPS data, the platinic acid used
throughout this study contains a small concentration of Pt(II)
2
4
25
species. Labinger et al. and Sen et al. report that a mixture
of Pt(IV) and Pt(II) compounds actively catalyzes THF oxidation
in the presence of water (in our case, the source of water is
H2PtCl6 × 6H2O). Upon doing so, several side products are
26
obtained, including those shown below. Moreover, the forma-
tion of alkyl chlorides due to the oxidation of alkanes by Pt
2
4
compounds has also been demonstrated. From these prior
Figure 1. X-ray photoelectron spectra of (a) HPS-Pt-THF and (b) HPS-
Pt-ML obtained with Mg KR radiation. The experimental data are
displayed as dashed lines, whereas deconvolution fits are shown as
solid lines.
findings, we expect that γ-butyrolactone (GBL), a major product
25,26
of this reaction,
will transform to the enole form, which is
stabilized by coordination with PtCl3 species. In this scenario,
Pt(IV) is reduced to Pt(II).
Table 1. Elemental Analysis Results from HPS and Its Pt
Nanocomposites
Another possibility to be considered is the coordination of
GBL through its keto-group, although such complexes with
PtCl3 fragments have not been previously reported. The FTIR
spectrum of the THF extract of HPS-Pt-THF shows signals at
composition (wt %)
designation
H
C
Cl
Pt
-
1
3
437 and 1178 cm (characteristic of hydroxyl groups), 1625
HPS
HPS-Pt-THF
HPS-Pt-ML
7.95
7.28
6.90
90.47
77.71
74.66
0.30
4.68
5.60
-
7.51
8.35
-
1
-1
-1
cm (water), 641 cm (C-Cl groups), 1028 cm (cyclic
compounds), and others. Thus, we conclude that the reaction
mixture resulting from interaction of THF with H2PtCl6 × 6H2O
contains several products, while free GBL in its keto-form is
not present (the identifying strong signal at 1770 cm- is absent).
Since all the possible side products possess high boiling
temperatures (in the range of 180-203 °C), it is conceivable
that their evaporation under vacuum at ambient temperature may
be incomplete.
1
To confirm the formation of these side products, we have
reacted H2PtCl6 × 6H2O with THF for 24 h and recorded FTIR
spectra at different stages of product drying. The formation of
an orange-brown product accompanied by HCl evolution is
already detectable only 2 h after H2PtCl6 × 6H2O loading. The
FTIR spectrum, recorded after the reaction was complete (in
2
4 h) and purged with Ar to remove THF, exhibits strong signals
-
1
at 3401, 2942, 2870, 1106, 1051, and 1015 cm , as well as
moderate signals at 1733, 1635, 1444, 1388, and 647 cm .
Comparison of these FTIR spectral features with the IR
signatures of the compounds presumed to be present in the
mixture (assuming GBL exists as enole), as well as with the
spectrum of the THF extract, confirms the presence of oxidation
products and provides evidence that is consistent with the
reaction route described above. Thus, the interaction of H2PtCl6
-
1
2
7
(
23) Goodfellow, R. J.; Goggin, P. L.; Duddell, D. A. J. Chem. Soc. (A)
1
1
1
968, 504.
Figure 2. Solid-state 13C NMR spectra of (a) HPS, (b) HPS-Pt-ML,
and (c) HPS-Pt-THF. The asterisks identify spinning sidebands.
(24) Labinger, J. A.; Herring, A. M.; Bercaw, J. E. J. Am. Chem. Soc.
990, 112, 5628.
(25) Sen, A.; Lin, M. R.; Kao, L.-C.; Hutson, A. C. J. Am. Chem. Soc.
992, 114, 6385.
free HPS in Figure 2a) at 69 ppm, which can be assigned to
either ether or hydroxyl groups. To elucidate the source of these
oxygen-containing moieties, we have subjected the HPS-Pt-THF
(
26) Shi, M. J. Chem. Res.-S 1998, 592.
(27) Pouchert, C. J. The Aldrich Library of FT-IR Spectra; Aldrich
Chemicals: Milwaukee, WI, 1983; Vol. 1, pp 238, 248, 697.

L-Sorbose Oxidation
6H2O with THF in HPS is expected to fill the HPS
nanocavities with a mixture composed of Pt complexes and
residues of the products from THF oxidation.
J. Am. Chem. Soc., Vol. 123, No. 43, 2001 10505
×
In marked contrast, addition of H2PtCl6 × 6H2O to ML,
followed by 24 h of agitation and subsequent ML evaporation,
does not yield any chemical changes, according to FTIR
analysis. In this case, we conclude that both H2PtCl6 and ML
remain unchanged. This observation is consistent with the
reported25 susceptibility of alcohols to oxidation in the presence
of a mixture of Pt(II) and Pt(IV): oxidation becomes increas-
ingly more difficult as the length of the alkyl chain decreases.
It must be remembered, however, that incorporation of platinic
acid within HPS in ML is accompanied by the reduction of
Pt(IV) to Pt(II), as evidenced by XPS and confirmed by visual
observations (i.e., the HPS powder darkens). This result
corroborates our earlier hypothesis that HPS serves as a reducing
agent for Pt(IV) and is most likely oxidized, presumably by
the formation of hydroxyl groups.21,22 According to the FTIR
spectrum of metal-free HPS, HPS contains a small concentration
of double bonds that can easily coordinate with a Pt complex.
-
1
The presence of a signal at 330 cm in the far-IR spectrum of
the HPS-Pt-ML specimen suggests the formation of a trans-
PtCl2L2 complex,28 where L here denotes a functionality (i.e.,
a double bond or hydroxyl group) belonging to HPS. While
complexation of Pt(II) species with alcohols in solution has been
2
9
documented, such complexes have not been isolated. It is
therefore possible that the size restrictions of the HPS nano-
cavities facilitate complex formation and stabilization.
According to TEM analysis, the HPS-Pt-THF and HPS-Pt-
ML specimens contain no metal nanoparticles with a diameter
exceeding 0.5 nm (the estimated resolution of the microscope).
Electron micrographs of the HPS-Pt samples collected after the
precursor samples were reduced with H2 are presented in Figure
3
and clearly reveal the existence of small Pt nanoparticles
possessing mean diameters of 1.3 ( 0.3 nm for HPS-Pt-THF
Figure 3a) and 1.4 ( 0.3 nm for HPS-Pt-ML (Figure 3b). The
(
size histograms displayed in Figure 4 show that both samples
exhibit maxima at 1.2 nm for single nanoparticles. Note that
the H2-reduced HPS-Pt-ML specimen also consists of nanopar-
ticle aggregates ranging in equivalent diameter from 2.4 to 4.0
nm. Results obtained from XRD are included in Figure 5. Since
HPS-Pt-THF is an amorphous material, its XRD pattern (not
shown) is featureless. The scattering pattern of the reduced HPS-
Pt-THF nanocomposite is consistent with the presence of well-
defined Pt(0) particles possessing a mean diameter of about 1-2
nm, which is in good quantitative agreement with the TEM
results.
Figure 3. Transmission electron micrographs of (a) HPS-Pt-THF and
2
(b) HPS-Pt-ML after H reduction. Groups of single Pt nanoparticles
are highlighted by circles, whereas nanoparticle aggregates in part b
are identified by arrowheads.
this case, nanoparticles form upon the thermal decomposition
of metal complexes at 200 °C. While thermal decomposition is
expected to facilitate atomic or tiny cluster migration, the size
of the Co nanoparticles remains primarily governed by the size
of the HPS nanocavities over a wide range of Co concentrations.
In the present work, H2 reduction of the Pt precursor greatly
lowers the probability of such intercavity migration at ambient
temperature, in which case the nanoparticle size is strictly limited
by the source of Pt atoms in each cavity.
In summary, we propose the following mechanism to explain
the production of Pt-containing HPS. Upon sorption of the
H2PtCl6 solution in THF by HPS and subsequent THF evapora-
tion, the solution becomes sufficiently concentrated so that the
oxidation-reduction reactions described earlier occur in the HPS
nanocavities. Both HPS and THF are oxidized, whereas the Pt
compound is reduced2 to a Pt(II) complex. Since the complex
effectively fills the HPS nanocavities, H2 reduction of HPS-Pt-
THF results in the formation of Pt nanoparticles with a mean
diameter of about 1.3 nm. This size is consistent with the
contraction of Pt complex clusters after ligand removal and the
formation of a metallic phase. To put this size scale in
perspective, the formation of Co nanoparticles in HPS yields
nanoparticles typically measuring about 2 nm in diameter. In
4,25
Thus, the amount of precursor (Pt complex), as well as the
size of the nanoparticles formed from the Pt atoms, is determined
exclusively by the HPS cavity size. On the basis of the van der
Waals radius of Pt (0.17 nm) and the corresponding atomic
3
volume (0.0213 nm ), it immediately follows that a spherical
Pt nanoparticle with a diameter of 1.3 nm (and a volume of
3
1
.15 nm ) should contain about 54 Pt atoms and should have
(
28) Adams, D. M.; Chatt, J.; Gerratt, J.; Westland, A. D. J. Chem. Soc.
964, 734.
29) Belluco, U.; Cattalin, L.; Basolo, F.; Pearson, R. G.; Turco, A. J.
Am. Chem. Soc. 1965, 87, 241.
2
an effective surface area of 5.30 nm . According to these results,
g of HPS-Pt-THF (with a composition of 7.51 wt % Pt from
Table 1) contains 3 × 10 nanoparticles with a total surface
1
1
(
1
8

10506 J. Am. Chem. Soc., Vol. 123, No. 43, 2001
SidoroV et al.
within the HPS nanocavities due to specific interactions with
HPS functionalities (e.g., double bonds and hydroxyl groups),
but a significant fraction of Pt(IV) species remains in the
polymer. After H2 reduction, the precursor complex transforms
into Pt nanoparticles. The existence of relatively large particles
may be attributed to the higher Pt concentration relative to the
H2-reduced HPS-Pt-THF specimen (see Table 1). As discussed
1
4
elsewhere, a limiting concentration appears to exist (8 wt %
for Co) below which nanoparticle size is not sensitive to the
loading level of the metal compound. At higher concentrations,
the formation of larger nanoparticles or nanoparticle aggregates
(as seen in Figure 3b) ensues.
Catalytic Properties of Pt-Containing HPS. In the direct
oxidation of L-sorbose to 2-keto-L-gulonic acid (according to
Scheme 1), four specimens have been examined: HPS-Pt-ML
and HPS-Pt-THF before and after H2 reduction. To ascertain
the optimal conditions favoring L-sorbose oxidation with HPS-
Pt-THF, the following reaction parameters have been varied:
catalyst concentration (Cc) from 0.007 to 0.027 M Pt, L-sorbose
concentration (C0) from 0.22 to 0.50 M, NaHCO3 concentration
(
CNaHCO ) from 0.22 to 0.50 M, reaction temperature (T) from
3
6
0 to 80 °C, reaction time (t) from 180 to 220 min, oxygen
-
6
-6
3
flow rate (VO) from 7.5 × 10 to 20 × 10 m /s, and stirring
rate from 200 to 1000 rpm. Several parameter relationships are
readily established from the data collected during the course of
this study and are presented in Table 2. For example, a decrease
in VO (compare runs 2 and 8) lowers both the L-sorbose
conversion and selectivity. This trend is explained by the
decreased amount of oxidizing agent present and by diffusion
limitations that become non-negligible at reduced flow rates (O2
bubbling provides additional mixing). Conversely, an increase
in VO (see runs 2 and 9) results in a lower final pH due to
accumulation of acid, which in turn slows down the reaction.
In this case, L-sorbose conversion and selectivity likewise
decrease. As identified above, a reduction in the stirring rate
from 1000 to 200 rpm also promotes a decrease in conversion
and selectivity due to the onset of diffusion limitations (see runs
Figure 4. Particle size histograms derived from TEM images such as
those presented in Figure 3 for (a) HPS-Pt-THF and (b) HPS-Pt-ML.
2
and 12).
Comparison of the runs 2, 13, and 14 in Table 2 reveals that
L-sorbose conversion is incomplete at t ) 180 min. An increase
in t to 220 min yields 100% conversion, but a pronounced
decrease in selectivity due to the formation of 2-keto-L-gulonic
acid oxidation products. A similar situation occurs when Cc is
increased (as in runs 2 and 3) and when C0 is decreased (as in
c
2
Figure 5. X-ray diffraction patterns collected from (a) H -reduced
HPS-Pt-THF and (b) HPS-Pt-THF after the induction period. The
pattern in part a has been shifted vertically (by 6000 counts) to facilitate
scrutinization.
area of 15.9 m /g. For comparison, complementary studies30-32
2
runs 2 and 4). In marked contrast, opposite changes in C (runs
have shown that stable Ru, Rh, Pd, and Pt clusters containing
2 and 1) and C (runs 2 and 5) result in incomplete L-sorbose
0
55 atoms (the characteristic number of atoms to account for 2
conversion. From the data corresponding to runs 2, 6, and 7,
layers of atoms covering a central atom) can be prepared if they
are stabilized by ligands such as phosphines. We hasten to point
an increase or decrease in CNaHCO changes the pH of the reaction
3
medium, which has an overall negative effect on the selectivity
3
0
out that the diameter of a single Pt55 cluster is reported to be
about 1.3 nm, which coincides embarrassingly well with the
size of the Pt nanoparticles obtained in the present study. Thus,
HPS possesses a unique ability to stabilize an enormous
population of Pt clusters, each composed of 76% surface
33
of the transformation of L-sorbose to 2-keto-L-gulonic acid.
Temperature control is another adjustable parameter that must
be considered in the direct oxidation of L-sorbose. A reduction
in T by 10 °C (see runs 2 and 10) yields a decrease in L-sorbose
conversion, whereas an increase in T by 10 °C (see runs 2 and
3
2
atoms.
1
1) results in further oxidation of 2-keto-L-gulonic acid.
In the case of the HPS-Pt-ML material, the situation is slightly
different. Since ML does not reduce Pt(IV) during evaporation,
all the chemical changes in HPS-Pt-ML are induced solely by
the polymer matrix. Trans-PtCl2L2 complexes apparently form
According to the data listed in Table 2, it immediately follows
that direct L-sorbose oxidation is very sensitive to changes in
the reaction parameters investigated here.
The highest selectivity (98%) at 100% L-sorbose conversion
is achieved with the HPS-Pt-THF catalyst under the following
set of reaction conditions: Cc ) 0.021 M Pt, C0 ) 0.42 M,
(
(
30) Schmid, G.; Huster, W. Z. Naturforsch. (B) 1986, 41, 1028.
31) Barreteau, C.; Desjonqu e` res, M. C.; Spanjaard, D. Eur. Phys. J. D
2
000, 11, 395.
32) N u¨ tzenadel, C.; Z u¨ ttell, A.; Chartouni, D.; Schmid, G.; Schlapbach,
L. Eur. Phys. J. D 2000, 8, 245.
-
6
3
CNaHCO ) 0.42 M, T ) 70 °C, VO ) 14 × 10 m /s, stirring
(
3
rate ) 1000 rpm, and NaHCO3 added continuously for 180 min.

L-Sorbose Oxidation
J. Am. Chem. Soc., Vol. 123, No. 43, 2001 10507
Table 2. Reaction Activity and Selectivity of HPS-Pt-THF in the Direct Oxidation of L-Sorbosea
runb
C
(M)
C
(M)
C
NaHCO3 (M)
V
O
× 10 (m /s)
6
3
T (°C)
t (min)
pHc
conversion (%)
yield (%)
selectivity (%)
c
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
0.007
0.021
0.027
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.42
0.42
0.42
0.22
0.50
0.42
0.42
0.42
0.42
0.42
0.42
0.42
0.42
0.42
0.42
0.42
0.42
0.22
0.50
0.40
0.44
0.42
0.42
0.42
0.42
0.42
0.42
0.42
14.0
14.0
14.0
14.0
14.0
14.0
14.0
7.5
20.0
14.0
14.0
14.0
14.0
14.0
70
70
70
70
70
70
70
70
70
60
80
70
70
70
200
200
200
200
200
200
200
200
200
200
200
200
180
220
8.3
6.5
5.4
4.9
8.9
6.1
9.2
8.8
4.7
9.4
8.5
9.2
7.7
5.9
76
100
100
100
82
79
68
71
95
68
100
74
37
98
71
68
61
52
30
35
58
24
15
41
83
70
76.0
98
71.0
68.0
74.4
65.8
44.1
49.3
61.1
35.3
15.0
55.4
97.7
70.0
1
1
1
1
1
d
85
100
a
NaHCO
3
is loaded continuously over 180 min. b Stirring rate is 1000 rpm. c Final pH values. d Stirring rate is 200 rpm.
As expected from earlier findings,33 gradual loading of the
alkaline agent ensures higher selectivity than one-shot loading
because it provides better control over the pH of the reaction
medium. Since the oxidation rate is effectively limited by the
rate of NaHCO3 addition, continuous loading is not employed
in the kinetic experiments described in the next section. While
the highest selectivity of the HPS-Pt-THF catalyst is 98%, the
highest selectivity of the HPS-Pt-ML catalyst, on the other hand,
does not exceed 40%. This significant difference in selectivity
can be explained in terms of the different catalytic species
present. After H2 reduction, comparable Pt nanoparticles form
in both materials, but their selectivity does not exceed 88-
9
0% at 100% L-sorbose conversion. While we recognize that
other explanations for such decreased selectivity may exist, we
attribute this observation to a difference in the surface properties
of Pt nanoparticles formed by H2 versus in situ reduction.
According to the data presented here, this difference appears
sufficient to alter the sorption-desorption equilibria of the
L-sorbose, oxygen, and 2-keto-L-gulonic acid.
The kinetic characteristics of selective L-sorbose oxidation
have been studied with regard to HPS-Pt-THF and HPS-Pt-ML
catalysts in which the substrate/catalyst concentration ratio (q
)
C0/Cc) has been varied as follows: 12.1 to 32.2 mol/mol Pt
for HPS-Pt-THF and 10.2 to 30.1 mol/mol Pt for HPS-Pt-ML.
These experiments have been conducted with an O2 flow rate
Figure 6. Dependence of L-sorbose conversion on reaction time (t) in
the presence of (a) HPS-Pt-THF and (b) HPS-Pt-ML at different
-
6
3
of 14 × 10 m /s and a stirring rate of 1000 rpm at 70 °C. As
discussed above, the reaction medium is charged with a single
shot of the alkaline agent (in an equimolar amount with respect
to the L-sorbose) before the beginning of the reaction. The
dependence of L-sorbose conversion on reaction time is
presented in Figure 6 and demonstrates that an increase in q
L-sorbose concentrations (C , in M): 0.05 (b), 0.11 (O), 0.16 (2), 0.22
0
0
(4), and 0.32 ((). Parts a and b do not include every C for which
symbols have been assigned. The solid lines are guides for the eye.
overall mean diameter of the nanoparticles in this specimen is
measured to be 1.8 ( 1.1 nm. According to the corresponding
histogram provided in Figure 8, the size distribution of the
nanoparticles exhibits a maximum in the vicinity of 1.0 nm and
becomes fairly continuous thereafter. If we assign those nano-
particles with a diameter less than or equal to 2.0 nm as
representative of the initial nanoparticles in the system (see
Figure 4a) and those larger than 2.0 nm as enlarged due to
L-sorbose oxidation, then the mean diameter and fraction of
initial nanoparticles are 1.2 ( 0.5 nm and 71%, respectively.
The population of enlarged particles measures 3.2 ( 0.9 nm
in diameter, in reasonably good agreement with the XRD results.
Formation of these nanoparticles in this catalyst can be explained
by two concurrent events: (i) slow, but continued, reduction
by L-sorbose and (ii) facilitated diffusion due the swollen state
of HPS during the oxidation reaction in aqueous media. Since
the first consideration should hinder nucleation, fewer nuclei
and, generally speaking, larger particles are anticipated to
(or, alternatively, an increase in C0 at constant Cc) promotes an
increase in reaction time (t), which is typical for most catalysts.
These kinetic curves also reveal the existence of an induction
period of about 100 min, during which time L-sorbose conver-
sion is negligible.
A sample of the HPS-Pt-THF catalyst collected after the
induction period has been isolated and subjected to examination
by XRD, TEM, and XPS. The XRD pattern acquired from this
material indicates the presence of Pt crystallites possessing a
mean diameter of ca. 4 nm, along with some unresolved
structure that might be assigned to amorphous Pt nanoclusters
(see Figure 5b). An electron micrograph of this sample (Figure
7) shows the coexistence of small nanoparticles with a mean
diameter approaching the resolution of the microscope (0.5 nm)
and larger particles with a mean diameter of up to 6.0 nm. The
5
(33) Kiyoura, T. U.S. Patent No. 4, 599,446, July 8, 1986.
develop. The image in Figure 7, however, displays a large

10508 J. Am. Chem. Soc., Vol. 123, No. 43, 2001
SidoroV et al.
Figure 9. XPS spectrum of HPS-Pt-THF after the induction period of
the catalyst during the direct oxidation of L-sorbose in aqueous media.
The experimental data are displayed as dashed lines, whereas decon-
volution fits are shown as solid lines.
According to the XPS data provided in Figure 9, the Pt 4f7/2
binding energy changes from 73.6 to 72.4.eV after the induction
period in the HPS-Pt-THF catalyst. Deconvolution of this
spectrum reveals four different species with different binding
energies. While one of them possesses a Pt 4f7/2 binding energy
of 71.3 eV and can be ascribed to Pt(0), additional species with
higher binding energies also appear to exist, which is contrary
to the XRD results indicating the presence of Pt(0) nanoparticles.
One possible explanation for this apparent contradiction is that
L-sorbose partially reduces Pt(II) to Pt(0), in which case mixed-
valence nanoparticles form. Alternatively, the presence of
“oxidized” Pt species may reflect the strong chemical interaction
between the surface atoms of the Pt nanoparticles and 2-keto-
L-gulonic acid. Such interaction can strongly change the binding
energy of Pt. In both scenarios, catalytic species are undoubtedly
formed in situ during the induction period. Therefore, HPS
appears to serve two crucial roles: one as a support for the Pt
catalyst during L-sorbose oxidation and the other as the
nanostructured matrix responsible for controlling nanoparticle
growth.
As is evident from the data listed in Table 2, the reaction
temperature constitutes an important reaction parameter in the
present study. Here, we examine the effect of T on the kinetics
of L-sorbose oxidation over the range 60-80 °C (the conditions
employed in the previous kinetic experiments are retained).
Resultant data are presented in Table 3 and show that an increase
in T promotes an increase in the rate of L-sorbose oxidation. It
is interesting to note that the reaction selectivity decreases
markedly for both the HPS-Pt-THF and HPS-Pt-ML catalysts
at 80 °C. A reasonable explanation for this observation is further
oxidation of 2-keto-L-gulonic acid under the reaction conditions
Figure 7. TEM image of HPS-Pt-THF after the induction period during
the direct oxidation of L-sorbose in aqueous medium. Groups of single
Pt nanoparticles are highlighted by circles, whereas substantially
enlarged nanoparticles are identified by arrows.
(which eventually results in the formation of CO2). At the
highest selectivity of the HPS-Pt-THF catalyst measured in this
work (98% at 100% conversion), the value of the turnover
Figure 8. Particle size histogram generated from TEM images such
as the one displayed in Figure 7 for HPS-Pt-THF after the induction
period.
-4
frequency (TOF) is found to be 5.4 × 10 [mol/mol Pt‚s],
19
which is comparable in magnitude to previously reported data
at significantly lower conversions.
population of the initial (small) nanoparticles, thereby indicating
that the nucleation rate of the nanoparticles does not significantly
influence their size. In this case, nanoparticle growth is restricted
by both the size of the HPS nanocavities and strong chemical
interactions with L-sorbose. On the basis of this conclusion, it
follows that the presence of enlarged nanoparticles in this
catalyst is most likely the result of the second consideration:
facilitated aggregation of small Pt nanoclusters in water-swollen
HPS. This explanation is furthermore consistent with the greater
population of enlarged nanoparticles, and correspondingly lower
activity, observed in the HPS-Pt-ML catalyst after its induction
period (data not shown).
The T dependence of L-sorbose oxidation can be further
examined if Arrhenius behavior is obeyed. To ascertain the
applicability of the Arrhenius equation in the present work, we
have measured the reaction time yielding 10% L-sorbose
conversion (τ0.1). A graphical representation of 1/τ0.1 as a
function of 1/T is displayed on semilogarithmic coordinates in
Figure 10 and demonstrates that Arrhenius behavior is exhibited
by both catalysts over the temperature range investigated.
Regression of the data shown in this figure yields apparent
activation energy (Q) values of 39 ( 3 and 41 ( 7 kJ/mol for
HPS-Pt-THF and HPS-Pt-ML, respectively. Values of the
frequency factor (V0) derived from the ordinate intercept would

L-Sorbose Oxidation
J. Am. Chem. Soc., Vol. 123, No. 43, 2001 10509
Table 3. Influence of Temperature on L-Sorbose Oxidation with HPS-Pt-THF and HPS-Pt-MLa
b
temperature (°C)
reaction outcome
60
65
70
75
80
L-sorbose conversion (%)
15.0/15.5b
12.0/10.5
80.0/67.7
1.7/1.2
22.0/18.5
19.0/15.3
86.4/82.7
2.5/1.9
47.0/26.0
45.0/24.5
95.7/94.2
5.4/2.8
63.0/38.0
55.0/33.0
87.3/86.8
7.2/3.8
82.0/75.0
59.0/56.3
72.0/75.1
9.4/6.5
2
-keto-L-gulonic acid (%)
selectivity (%)
4
TOF × 10 (mol/mol Pt‚s)
a
-6
3
Reaction conditions: V
O
) 14 × 10 m /s; NaHCO
3
is charged one-shot at the beginning of the reaction; C
0
) 0.107 M; q ) 13.8 mol/mol
b
Pt; t ) 200 min. The data are presented in the format of HPS-Pt-THF/HPS-Pt-ML.
Experimental Section
Materials. The HPS was synthesized according to the procedure
described in detail elsewhere,12 whereas H
hydrogencarbonate (NaHCO ) were obtained from Reakhim (Moscow,
PtCl × 6H O and sodium
2
6
2
3
Russia). Reagent-grade tetrahydrofuran (THF) and methanol (ML) were
purchased from Aldrich and used as received, as was the L-sorbose
from Fluka. Water was purified with a Milli-Q water purification
system.
HPS-Pt Preparation. Particles of HPS were inserted into a Schlenk
tube capped with a rubber septum. Following evacuation at 1.5 mbar,
2 6
the tube was filled with Ar. In a typical synthesis, 0.4 g of H PtCl
was dissolved in 4 mL of THF or ML, which promoted complete
swelling of 1 g of HPS without excess solvent. The solution was
subsequently added to the Schlenk tube by syringe through the rubber
septum. After the HPS was allowed to swell for 30 min in the H PtCl
2 6
solution, it was evacuated for 3 days at 1.5 mbar. Table 1 lists elemental
analysis data of the resultant HPS-Pt samples, which were later reduced
Figure 10. The temperature dependence of the reaction time yielding
0% conversion of L-sorbose in the presence of HPS-Pt-THF (O) and
HPS-Pt-ML (4). The solid lines denote regressed fits of the Arrhenius
equation and yield the activation energy (Q) values listed in the
text.
1
2
in a steady flow of H for 3 h at ambient temperature.
L-Sorbose Oxidation. The reaction was conducted batchwise in a
specially constructed apparatus that permits control over such param-
eters as the L-sorbose and NaHCO concentrations, catalyst concentra-
3
tion, temperature, (pure) oxygen feed rate, and stirring rate. A solution
of catalyst and L-sorbose (100 mL) prepared at a predetermined
concentration was placed in the temperature-controlled apparatus
equipped with a magnetic stir bar and reflux condenser. The rate of
oxygen feed was controlled by a rotameter. An equimolar quantity of
Table 4. Activity and Selectivity of HPS-Pt-THF after Repeated
Catalytic Cycles
a
4
no. of catalytic cycles TOF × 10 (mol/mol Pt‚s) selectivity (%)
1
7
5
5.4
5.3
5.5
95.7
95.0
95.1
1
a
See Table 3 (T ) 70 °C) for the reaction conditions.
3
alkalizing agent (NaHCO ) was fed to the apparatus in one shot or
continuously over 180 min (to maintain a constant pH of 7.7) with use
of an automatic dispenser. The high stirring rates employed here ensured
good mixing without diffusion limitation. Samples of the reaction
mixture were periodically removed for analysis. At the end of the
experiment, the catalyst was separated by filtration, and the filtrate was
analyzed for the presence of L-sorbose and 2-keto-L-gulonic acid. The
amount of residual L-sorbose was discerned by gas chromatography
provide a measure of the number of collisions or active centers
on the surface of a catalyst surface,34 but suffer from inaccuracy
due to scatter in the data. Since both catalysts possess almost
identical activation energies, we infer that the number of active
catalytic centers is greater for HPS-Pt-THF than HPS-Pt-ML
to explain the significantly higher activity of the HPS-Pt-THF
catalyst. This result is consistent with the greater fraction of
large nanoparticles, with lower surface area, in the HPS-Pt-ML
material.
(GC) through the use of a “Chrom-5” chromatograph, operated
isothermally with a flame-ionization detector and glass column filled
with 5% SE-30 on Chromaton N-AW. The quantity of 2-keto-L-gulonic
3
7
acid was determined by the classical iodometric method of Heyns.
For runs 2, 7, and 13 listed in Table 2, the quantity of 2-keto-L-gulonic
In addition to selectivity and activity, another important
feature of any promising catalyst is the stability of its catalytic
properties. Unfortunately, catalysts often lose their activity after
only two or three repeated uses, typically due to either loss of
38
acid was determined by direct isolation via amine complexation and
3
9
subsequent transformation to the sodium salt of ascorbic acid, the
purity of which (98.5%) was measured at Belgorodvitaminy (Belgorod,
Russia) with a “Millichrom 4” HPLC equipped with a UV detector, a
PLRS-s 100A solid polymer phase, and a 0.2 M NaH PO (aq) liquid
35
the catalytic metal or contamination of the catalytic surface
by reaction products.36 As is seen from the data tabulated in
Table 4, neither the activity nor the selectivity of the HPS-Pt-
THF catalyst deteriorates even after 15 catalytic cycles. This
result demonstrates that nanoparticles can exhibit remarkably
stable catalytic properties. Such stability is attributed to what
we refer to as “double protection”, in which (i) the location of
the nanoparticles within HPS nanocavities of comparable size
prevents leaching of catalytic metal; and (ii) chemical interaction
of the nanoparticle surface with 2-keto-L-gulonic acid avoids
surface contamination.
2
4
phase. Each experiment reported here was repeated at least 3 (often 5)
times to ensure reproducibility.
Material Characterization. X-ray fluorescence (XRF) measure-
ments were performed with a Zeiss Jena VRA-30 spectrometer (Mo
anode, LiF crystal analyzer and SZ detector). Analyses were based on
the Co KR line and a series of standards prepared by mixing 1 g of
polystyrene with 10-20 mg of standard compounds. The time of data
acquisition was constant at 10 s. Fourier-transform infrared (FTIR)
spectra were recorded with a Nicolet spectrometer in the spectral region
-
1
-1
ranging from 250 to 4000 cm at a resolution of 2 cm . X-ray
diffraction (XRD) measurements were acquired with a Rigaku D/max-
(34) Sakakini, B. H.; Verbrugge, A. S. J. Chem. Soc., Faraday Trans.
1
997, 93, 1637.
(37) Heyns, K. Ann. Chem. 1947, 558, 171, 177.
(38) Schaper, M. German Offen. DE 3831070 A1, 1990.
(39) Crawford, T. C.; Crawford, S. A. AdV. Carbohydr. Chem. Biochem.
1980, 37, 79.
(
35) Bakos, I.; Mallat, T.; Baiker. A. Catal. Lett. 1997, 43, 201.
36) Schuurman, Y.; Kuster, B. F. M.; van der Wiele, K.; Marin, G. B.
(
Appl. Catal. A-Gen. 1992, 89, 47.

10510 J. Am. Chem. Soc., Vol. 123, No. 43, 2001
SidoroV et al.
RC diffractometer operated at 12 kV with Cu KR radiation. X-ray
photoelectron spectroscopy (XPS) was conducted with a two-chamber
Kratos XSAM-800 spectrometer. Powdered samples were evacuated
restricted by the size of the HPS nanocavities. After H reduction
in dry, solid polymer, the nanoparticles possess a narrow size
2
distribution and a mean diameter of 1.3-1.4 nm. Following the
reduction of Pt(II) species during the induction period of
catalytic L-sorbose oxidation, the nanoparticles exhibit a sig-
nificantly broader size distribution. The small nanoparticles
possess a mean diameter of about 1.2 nm and are restricted in
size by the HPS nanocavity and chemical interaction with
2-keto-L-gulonic acid. Enlarged nanoparticles measuring up to
6.0 nm in diameter and having a mean diameter of about 3.2
nm form due to aggregation in water-swollen HPS. The catalytic
activity and selectivity of these nanocomposite materials have
been measured under a wide variety of reaction conditions.
These catalysts exhibit high selectivity (98% at 100% conver-
sion) and remarkably stable catalytic properties that remain
practically unchanged after 15 catalytic cycles. In this regard,
the nanostructured HPS matrix serves as both a nanoscale reactor
and catalyst support to (i) restrict the growth of the nanoparticles,
-
9
-10
in the spectrometer for 12 h at 10 -10
mbar, and spectra were
collected in the medium resolution regime at 25 °C. For photoelectron
excitation, characteristic Mg KR radiation (hυ ) 1253.6 eV) was used.
The power of the X-ray gun never exceeded 75 W (15 kV, 5 mA).
Transmission electron microscopy (TEM) was performed with a Zeiss
EM902 electron spectroscopic microscope operated at 80 kV and an
energy loss of 0 eV. Insoluble HPS-Pt powders were embedded in epoxy
resin and subsequently microtomed at ambient temperature. Images of
the resulting thin sections (ca. 50 nm thick) were collected on plate
negatives, digitized at 600 dpi, and analyzed with the Adobe Photoshop
software package and the companion Image Processing Toolkit. Solid-
1
3
state C magnetic resonance (NMR) spectra were collected with a 4.0
mm Bruker Magic Angle Spinning Probe in a Bruker Avame DSX
spectrometer. Data were obtained by using cross polarization from
protons (4 µs proton π/2 pulse, 500 µs contact time) under 12.000 kHz
magic angle spinning. Two-pulse phase modulation (TPPM) decoupling
4
0
was employed during acquisition.
(ii) prevent leaching of the nanoparticles, and (iii) ensure a high
degree of chemical interaction.
Conclusions
Acknowledgment. L.M.B., S.N.S., P.M.V., V.G.M., N.V.L.,
and E.M.S. were supported by the Russian Foundation for Basic
Research (Grant No. 01-03-32937) and the NATO Science for
Peace Program (SfP-974173-Colloidal Catalysts).
A robust strategy for synthesizing Pt-containing hyper-cross-
linked PS is described. The accumulation of Pt(II) complexes,
as well as subsequent Pt nanoparticle formation, is effectively
(
40) Auger, M.; Lakshmi, K. V.; Griffin, R. G. J. Chem. Phys. 1995,
1
03, 6951.
JA0107834