Three-Phase Microemulsion/Sol-Gel System for Aqueous Catalysis with Hydrophobic Chemicals

Article abstract of DOI:10.1002/chem.200305591

A facile three-phase transport process is described that allows to carry out catalytic reactions in water, whereby all components are hydrophobic. According to this process a hydrophobic substrate is microemulsified in water and subjected to an organometallic catalyst, which is entrapped within a partially hydrophobized sol-gel matrix. The surfactant molecules, which carry the hydrophobic substrate, adsorb/desorb reversibly on the surface of the sol-gel matrix breaking the micellar structure, spilling their substrate load into the porous medium that contains the catalyst. A catalytic reaction then takes place within the ceramic material to form the desired products that are extracted by the desorbing surfactant, carrying the emulsified product back into the solution. The method is general and versatile and has been demonstrated with the catalytic hydrogenations of alkenes, alkynes, aromatic C= C bonds, and nitro and cyano groups.

Full text of DOI:10.1002/chem.200305591

FULL PAPER
Three-Phase Microemulsion/Sol Gel System for Aqueous Catalysis with
Hydrophobic Chemicals
Raed Abu-Reziq,[a] Jochanan Blum,*[b] and David Avnir*[a]
Abstract: A facile three-phase trans-
port process is described that allows to
carry out catalytic reactions in water,
whereby all components are hydropho-
bic. According to this process a hydro-
phobic substrate is microemulsified in
water and subjected to an organome-
tallic catalyst, which is entrapped
within a partially hydrophobized sol
gel matrix. The surfactant molecules,
which carry the hydrophobic substrate,
adsorb/desorb reversibly on the surface
of the sol gel matrix breaking the mi-
cellar structure, spilling their substrate
load into the porous medium that con-
tains the catalyst. A catalytic reaction
then takes place within the ceramic
material to form the desired products
that are extracted by the desorbing sur-
factant, carrying the emulsified product
back into the solution. The method is
general and versatile and has been
demonstrated with the catalytic hydro-
genations of alkenes, alkynes, aromatic
C=C bonds, and nitro and cyano
groups.
Keywords: heterogeneous
catalysis
¥
hydrogenation
¥
microemulsion ¥ sol gel processes ¥
three-phases transport
Introduction
hydrophobic heterogeneous catalysis (and in fact in hetero-
geneous reactions in general).[5] It relies on a three-phase
emulsion/solid transport (EST) method illustrated in
Scheme 1. The hydrophobic substrate is emulsified in water
The three-phase emulsion/solid transport process: In this
study, we have merged two domains, each of which proved
to be very useful and fruitful for the general field of cataly-
sis. One domain is the application of sol gel materials[1] as
heterogeneous carriers of catalysts,[2] and the other is the
use of microemulsions for solubilization.[3] We show that by
the careful combination of these two methodologies, one
can carry out catalytic reactions in water, whereby both the
catalyst and its substrate are hydrophobic. Since the majori-
ty of catalytic reactions involve hydrophobic components, a
general method that allows the replacement of the tradition-
ally required organic solvents by water, may become useful
when environmental issues are to be considered.[4]
Towards this goal we developed recently a novel and gen-
eral approach that eliminates the need of organic solvents in
Scheme 1. A flow-chart of the emulsion/solid transport three-phase pro-
cess. The advantageous microemulsion process follows similar steps.
[a] R. Abu-Reziq, Prof. D. Avnir
Institute of Chemistry
(an oil-in-water (o/w) emulsion) with a carefully selected
surfactant and then is subjected to the (organometallic) cat-
alyst, which is entrapped within a partially hydrophobized
sol gel matrix. The surfactant molecules that carry the hy-
drophobic substrate adsorb/desorb reversibly on the surface
of the sol gel matrix,[5] breaking the micellar structure[6] and
spilling their substrate load into the porous medium that
contains the metal-complex catalyst. A catalytic reaction
then takes place within the ceramic material to form the de-
The Hebrew University of Jerusalem
Jerusalem 91904 (Israel)
Fax : (+972)2-620099
E-mail: david@chem.ch.huji.ac.il
[b] Prof. J. Blum
Institute of Chemistry
The Hebrew University of Jerusalem
Jerusalem 91904 (Israel)
Fax : (+972)2-6513832
E-mail: jblum@chem.ch.huji.ac.il
958
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/chem.200305591
Chem. Eur. J. 2004, 10, 958 962

958 962
sired products that are extracted by the desorbing surfac-
tant, carrying the emulsified product back into the solution.
The free products are finally obtained by breaking the emul-
sion.
(CTAB),
N,N-dimethyl-N-dodecyl-N-(2-hydroxyethyl)am-
monium bromide (DDHAB)); and of partially hydrophobi-
cally derivatized silicas, namely octylated and phenylated
sol gel
matrices.
The
catalysts
[(CH3)(C8H17)3N]+
À
A major improvement we introduce here to this EST
process, and which we recommend as a standard procedure
over our previously reported procedure,[5] is to move from
an emulsion to a microemulsion. We recall that microemul-
sions are defined as thermodynamically stable and optically
transparent macroscopically homogeneous mixtures of oils,
water, and surfactants.[3,7] Being based on aqueous media
they are economically favored and environmentally safer
than conventional organic solvents, while overcoming re-
agent incompatibility with water. Microemulsions have been
used as microreactors,[8] and, owing to their large interfacial
area, were found to accelerate chemical processes and to
induce regioselectivity in organic reactions as a function of
their composition.[9] Microemulsions have also been used in
various catalytic processes. Examples include oxidation,[10]
hydroformylation,[11] carbonylation,[12] dehalogenation,[13]
and free-radical chain polymerization.[14] In the application
to the three-phase catalysis reported herewith the o/w mi-
croemulsion is used as a carrier of both the substrate and
the product. Since the catalysis takes place within the non-
soluble ceramic component of the system, it can be filtered
off and re-used. There are several advantages of replacing
the regular emulsion by a microemulsion in the EST
method.
[RhCl4]À¥nH2O and [Rh(cod)(PPh3)2]+PF6 were entrapped
within these matrices (and, for comparative purposes,
À
[Rh(cod)(PPh3)2]+PF6 was also entrapped in hydrophilic
silica) and used for the selective hydrogenation of unsaturat-
ed hydrophobic substrates.
Experimental Section
À[15]
Chemicals: [Rh(cod)(PPh3)2]+PF6
and N,N-dimethyl-N-dodecyl-N-(2-
hydroxyethyl)ammonium bromide[16] were synthesized according to pub-
lished procedures. All other chemicals are commercial.
Microemulsification procedures: Two general procedures were employed
for the microemulsification of the various substrates.
Procedure A: A mixture of an appropriate surfactant (1.0 1.4 g) in triply
distilled water (TDW; 10 mL) and the organic substrate (9.7 24 mmol)
was emulsified by vigorous stirring. The resulting emulsion was then ti-
trated with either n-butanol or n-amyl alcohol (typically, 2 5 mL) until a
clear transparent microemulsion was obtained. In case the exact end-
point had been passed by excess alcohol (turbidity appears), back-titra-
tion was carried out by adding more of the surfactant. The solution was
then sonicated for 5 10 min. (A simple test for a successful microemulsi-
fication, in addition to its transparency, is the shining of the light of a
laser pointer through the solution: the beam is clearly seen.) The micro-
emulsions so formed proved stable at the reaction temperatures of 80
1008C. Breaking of the emulsion at the end of the reaction was achieved
by agitation with NaCl (0.2g). If during the emulsification foam was
formed, sonication for 20 minutes solved this problem.
1) Temperature flexibility: a whole range of convenient
temperatures can be used (as opposed to the narrow
window around 808C needed for the regular emulsion
method[5]).
2) A wider range of surfactants can be used (unlike the reg-
ular emulsion described in our previous study,[5] which
we could affect only with cetyl(trimethylammonium) 4-
toluenesulfonate).
Procedure B (suitable in particular for polycyclic arenes): The substrate
(4 9 mmol) dissolved in either methylcyclohexane (2.0 mL) or warm bu-
tanol (3.0 mL) was added to a stirred solution of SDS (3 4 mmol) in
TDW (10 mL); the resulting emulsion was heated to 608C and titrated
with n-butanol (2 3 mL) to form the transparent microemulsion. It was
then sonicated for 5 min. Representative data for the formation of micro-
emulsions with several substrates are summarized in Table 1.
Entrapment of the catalysts in organically modified sol gel matrices
Catalyst 1: A solution of octyltriethoxysilane (2.1 mL, 6.72 mmol) in
EtOH (5.6 mL, 95.6 mmol) was hydrolyzed by TDW (0.36 mL) at 848C
for 4 d. The resulting solution was added to a mixture of tetramethoxysi-
lane (TMOS, 5.0 mL, 33.6 mmol) and TDW (4 mL) and stirred for 1 h
3) The large surface area of the microemulsion makes the
transport process into and out of the solid more effi-
cient.
(the molar ratio octyltriethoxysilane:TMOS is 1:5).
A solution of
À
[Rh(cod)(PPh3)2]+PF6 (40 mg, 0.0454 mmol) dissolved in a mixture of
Since the EST methodology
involves
several
equilibria
which take place in a rather
complex system, the main art
that had to be practiced is the
tailoring of the right HLB (hy-
drophilic/lipophilic balance) of
both the surfactant and the
solid matrix. The micro-EST
(for short, we shall continue to
use EST for microemulsions as
well) systems described below
are composed of either anionic
surfactants (sodium dodecylsul-
fate (SDS), sodium 4-dodecyl-
benzenesulfonate (SDBS)) or
cationic surfactants (cetyltri-
Table 1. Composition of representative microemulsions
Entry
Substrate (%)[a]
Surfactant (%)[b]
Precentage
of n-butanol
Percentage
of water
1
1-octene (10.1)
DDHAB (8.2)
11.4
67.3
19.7
13.5[c]
11.3
6.1
15.7
15.4[d]
21.4
24.3
19.6
70.3
21-octene (9.6)
SDS (7.8)
15.3
3
4
5
6
7
8
9
10
11
1-hexene (12.3)
1-undecene (10.4)
styrene (9.4)
phenylacetylene (9.8)
ethylbenzene (11.2)
naphthalene (3.0)
nitrobenzene (13.7)
diphenylamine (8.8)
benzonitrile (12.2)
SDS (7.3)
60.7
69.2
71.0
75.1
64.7
59.6
57.4
58.7
60.4
CTAB (6.9)
DDHAB(8.3)
DDHAB (9)
SDBS (8.4)
SDBS (7.7)
SDBS (7.5)
SDBS (8.2)
SDBS (7.8)
[a] Except for the microemulsions of naphthalene and diphenylamine that were prepared by procedure B, all
others were prepared by procedure A. [b] DDHAB: N,N-dimethyl-N-dodecyl-N-(2-hydroxyethyl)ammonium
bromide, SDS: sodium dodecylsulfate, CTAB: cetyltrimethylammonium bromide, SDBS: sodium 4-dodecyl-
methylammonium
bromide benzene sulfonate. [c] n-Pentanol was used here. [d] And 14.3% methylcyclohexane.
959
Chem. Eur. J. 2004, 10, 958 962
www.chemeurj.org ¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

D. Avnir, R. Abu-Reziq, and J. Blum
FULL PAPER
acetone (2mL) and THF (8 mL) was added. After stirring for 5 min a
solution of 0.1m tetrabutylammonium fluoride (0.2mL) in THF was
added. Gellation occurred within 1 h. The resulting material was dried at
0.5 mm for 24 h at 548C; the so-formed xerogel was treated with CH2Cl2
(20 mL) under reflux for 30 min. The material was dried at 0.5 mm for
ratios (lower hydrophobicities) were tested as well. Thus,
1:20, 1:15 and 1:10 molar ratios yielded n-octane and un-hy-
drogenized isomerization products at ratios of 89:9.5, 92:7
and 95:4%, respectively.
3 h, sonicated with the same solvent for 30 min and dried again (3 h).
As mentioned above, an advantage of the microemulsion
approach over the emulsion approach is the possibility to
employ a wider range of surfactants. Thus, in addition to the
cationic surfactant DDHAB, we were able to hydrogenate
1-octene in a microemulsion prepared with an anionic sur-
factant. When 1-octene was microemulsified in the presence
of sodium dodecyl sulfate (SDS) (entry 2, Table 1), the hy-
drogenation with catalyst 1 yielded 96% of n-octane (along
with just 2% of isomerization products). One may wonder
how it is possible that both types of surfactants are suitable
for the EST process. As is well known,[6] the cationic and
anionic surfactants adsorb differently on the partially hydro-
phobic silica surface (Scheme 2a,b), namely with the cation-
À
The final weight of the ceramic catalyst, [Rh(cod)(PPh3)2]+PF6
@
(C8H17)-SG, (catalyst 1), was 3.9 g and the catalyst loading was
0.0116 mmolgÀ1. No detectable leaching of the entrapped catalyst was
evident neither in 1 nor in the catalysts described next.
À
Catalyst 2: For comparative purposes, [Rh(cod)(PPh3)2]+PF6 was also
entrapped in hydrophilic silica sol gel matrix by the following procedure.
A solution of TMOS (5.0 mL, 33.6 mmol) and TDW (4 mL) in EtOH
À
(5.6 mL) was stirred for 1 h.
A
solution of [Rh(cod)(PPh3)2]+PF6
(40 mg, 0.0454 mmol) in a mixture of acetone (2mL) and THF (8 mL)
was added. After stirring for 5 min a solution of 0.1m of tetrabutylammo-
nium fluoride (0.2mL) in THF was added and gellation occurred within
15 min. The resulting gel was treated as described for catalyst 1 to give
2.8 g of [Rh(cod)(PPh3)2]+PF6À@SG (catalyst 2). The catalyst loading was
0.0162mmolg À1
.
Catalyst 3: A solution of phenyltrimethoxysilane (0.63 mL, 3.37 mmol) in
EtOH (2.0 mL) was hydrolyzed with TDW (0.18 mL) by heating the solu-
tion at 608C for 24 h. The resulting solution of the hydrolyzed silane was
added to a stirred solution of TMOS (5.0 mL, 33.6 mmol; molar ratio of
phenyltrimethoxysilane:TMOS is 1:10), TDW (4.0 mL), methyltrioctyl-
ammonium chloride (65 mg, 0.16 mmol) and RhCl3¥3H2O (43 mg,
0.16 mmol) dissolved in MeOH (5.0 mL). The combined solutions were
stirred for 5 min and then a solution of 0.1m tetrabutylammonium fluo-
ride in tetrahydrofuran (0.2mL) was added. Stirring was continued until
the mixture gelled (2 3 d). The resulting material was dried at 0.5 mm
for 48 h at room temperature and then treated as above with CH2Cl2,
yielding 2.95 g of {[(CH3)(C8H17)3N]+[RhCl4]À}@(C6H5)-SG, (catalyst 3).
The catalyst loading was 0.0542mmolg À1
.
Catalyst 4: This catalyst was prepared similarly to catalyst 3, except that
the hydrophobic silane monomer used was octyltrimethoxysilane
(0.43 mL, 1.664 mmol, octyltrimethoxysilane:TMOS=1:20), which was
hydrolyzed with TDW (0.09 mL) in ethanol (2.0 mL) at 608C for 24 h.
An amount of 2.75 g of the ceramic catalyst, {[(CH3)(C8H17)3N]+
[RhCl4]À}@[C8H17)]-SG, (catalyst 4), was obtained. The catalyst loading
was 0.0581 mmolgÀ1
.
The catalytic reactions: A freshly prepared microemulsion (typically 15
20 mL) with the appropriate substrate (in the range of 3.6 24 mmol), pre-
pared according to either procedure A or B, was added to the catalytic
material (typically in the range of 0.02 0.08 mmol rhodium complex) in a
50 mL glass-lined stainless-steel autoclave equipped with a magnetic stir-
rer and a sampling device. After sealing the reaction vessel, it was
purged three times with hydrogen, pressurized to 200 psi H2 and heated
with stirring at 808C for the required length of time. Upon completion of
the reaction, the autoclave was cooled to room temperature and un-
sealed. The ceramic catalyst was separated from the reaction mixture by
filtration and the microemulsion broken by adding NaCl as described
above. The resulting two phases were separated, the organic phase was
worked up in the usual manner and the resulting products analyzed by
GC-MS, GC, and NMR spectroscopy. All products are known and were
compared with authentic samples. For re-use, the catalyst was refluxed in
MeOH (20 mL) for 30 min, dried at 0.5 mm for 3 h at room temperature,
sonicated with CH2Cl2 (20 mL) for 30 min and dried again at 0.5 mm for
2h at room temperature.
Scheme 2. Illustration of the adsorption of surfactants: a) a cationic sur-
factant on a partially hydrophobic silica surface; b) an anionic surfactant
on a partially hydrophobic silica surface; c) a cationic surfactant on a hy-
drophilic silica surface; d) an anionic surfactant on a hydrophilic silica
surface.
ic head pointing to the surface or the anionic head pointing
away from the surface, respectively. However, the final out-
come is similar for the EST purpose: Reversible rupture of
the (hemi)micellar[6] structure, spilling its content into the
porous material, or collecting the product from it.
When–for comparative purposes–[Rh(cod)(PPh3)2]+
À
PF6 was entrapped in the nonhydrophobized silica sol gel
matrix (catalyst 2) and used for the hydrogenation of 1-
octene (808C and 200 psi of H2) in heptane as solvent, 63%
of n-octane and 37% of isomerization products were ob-
tained after 4 h (compared to the hydrogenation by catalyst
1 under the microemulsion conditions which produced, as
described above, n-octane in an almost quantitative yield).
Thus, under these ™standard∫ organic-solvent conditions, the
isomerization products do not reside long enough within the
hydrophilic catalytic material to be converted to n-octane.
Similar results, but with some improvement, were obtained
Results and Discussion
Hydrogenation of 1-octene: [Rh(cod)(PPh3)2]+PF6 entrap-
À
ped in octylated silica (prepared from a 1:5 molar ratio of
octyltriethoxysilane/TMOS; catalyst 1) promoted quantita-
tive transformation (98%) of a microemulsion of 1-octene
(see entry 1 in Table 1), to n-octane within 4 h under the
conditions given in the Experimental Section. Lower molar
960
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 958 962

Microemulsion/Sol Gel Systems
958 962
with the hydrophilic catalyst 2 when heptane was replaced
by the cationic (DDHAB) microemulsion: a fast reaction
took place (full conversion within 4 h); however, again sub-
stantial isomerization competed with the hydrogenation
process [29% of isomerization products (cis- and trans-2-
octene and trans-3-octene) and 70% n-octane].
Hydrogenation of the aromatic ring: Ion pair
[(CH3)(C8H17)3N]+[RhCl4]À¥nH2O entrapped in a phenylat-
ed silica sol gel matrix (catalyst 3) was found most suitable
for the EST hydrogenation of the aromatic C=C bonds of
ethylbenzene. Structural similarity between the component
of the catalyst and the substrate seems to be necessary for
successful catalytic reactions in the EST system. In fact the
hydrogenation of the arene was found to proceed best when
the surfactant contained a phenyl ring as well (cf. the re-
quirement of a longer surfactant for the long chained 1-un-
decene than that for 1-octene). Thus, ethylbenzene was mi-
croemulsified with SDS (entry 7, Table 1). Naturally, the
arene ring hydrogenation proceeded slower than the satura-
tion of the above-mentioned alkenes, so that ethylbenzene
afforded 63% of ethylcyclohexane (as the only reaction
product) after 24 h under the conditions described in the
Expermental Section [Eq. (3)]. Aromatic polycylics reacted
even slower. For example, the microemulsion of naphtha-
lene (entry 8, Table 1) gave after 24 h at 808C only a mix-
ture of 9% tetralin, 11% of cis- and 7% of trans-decalin
[Eq. (4)].
It can be concluded therefore that the process of hydro-
phobization of the silica surface by surfactant adsorption is
less efficient than the hydrophobization by partial surface
derivatization (catalyst 1). The existing hydrophobic chains
on catalyst 1 interact with the chains of the cationic surfac-
tant, promoting its adsorption. Interestingly, however, when
the anionic SDS microemulsion was used with catalyst 2, the
percentage of full reduction to n-octane increased to 90
(along with 10% of isomerization). We are currently in a
position to only speculate as to the origin of this surprising
apparent pronounced hydrophobization. When a cationic
micelle opens and adsorbs, the individual surfactant mole-
cules are dispersed on the surface (Scheme 2c) in a way that
follows the dispersion of the silanols (about 4 5 sila-
nolsnmÀ2[17] ). However, when anionic surfactant molecules
adsorb, heterogeneous clusters of hydrophobic chains are
energetically preferred over homogeneous dispersion, and,
thus, on a local microscopic level, adsorbed anionic surfac-
tants form islands of pronounced hydrophobicity
(Scheme 2d); this facilitates the EST process at these sites.
Hydrogenation of other alkenes and alkynes: 1-Hexene and
1-undecene were hydrogenated best in microemulsions by
[(CH3)(C8H17)3N]+[RhCl4]À¥nH2O entrapped in octylated
silica sol gel (catalyst 4). 1-Hexene was microemulsified as
specified in entry 3 (Table 1). After 6 h under the conditions
given in the Expermental Section, 90% of the hexene was
converted into alkene-free n-hexane. Unlike 1-hexene and
1-octene, 1-undecene could be neither microemulsified by
SDS nor by DDHAB, but required a surfactant with a
longer chain, such as cetyltrimethylammonium bromide. It is
also worth noting the fact that n-butanol proved unsuitable
as the co-surfactant in this system, but n-amyl alcohol led to
the formation of a stable microemulsion (entry 4, Table 1)
that afforded a quantitave yield of pure undecane after a 4 h
period of hydrogenation.
Hydrogenation of the nitro and cyano moieties: The EST
methodology was also applied to the hydrogenation of nitro
and cyano moieties. The microemulsion described in entry 9
(Table 1) was hydrogenated gradually to give initially aniline
(after 13 h at 80 8C 98% of aniline was isolated), which was
further transformed to cyclohexylamine and dicyclohexyl-
amine. The latter becomes finally the major product. After
24 h the mixture of products consisted of 69% dicyclohexyl-
Alkenes and alkynes substituted by aromatic moieties
could be microemulsified by the same procedure as em-
ployed for octene. The amount of co-surfactant required for
these substrates was, however, smaller than that needed for
the non-aromatic alkenes (see, for example, entries 5 and 6,
Table 1). The microemulsified styrene was smoothly hydro-
genated to ethylbenzene [74% within 4 h, Eq. (1)] and mi-
croemulsified phenylacetylene formed, in a stepwise way,
styrene and ethylbenzene [Eq. (2)].
amine, 25% cyclohexylamine, and 6% aniline, [Eq. (5)]. Di-
phenylamine seems not to be a reaction precursor of dicy-
clohexylamine. Careful continuous analysis of the reaction
neither revealed any traces of this compound nor of its par-
tial hydrogenation products. The hydrogenation of diphenyl-
amine listed in Table 1 (entry 10) is slow and affords after
17 h 12% of dicylohexylamine and 4% of phenylcyclohexyl-
amine [Eq. (6)].
961
Chem. Eur. J. 2004, 10, 958 962
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

D. Avnir, R. Abu-Reziq, and J. Blum
FULL PAPER
[4] a) F. Currie, K. Holmberg, G. Westman, Colloids Surf. A. 2003, 215,
51; b) V. Nardello, M. Hervÿ, P. L. Alsters, J. M. Aubry, Adv. Synth.
Catal. 2002, 344, 184.
[5] R. Abu-Reziq, D. Avnir, J. Blum, Angew. Chem. 2002, 114, 4311;
Angew. Chem. Int. Ed. 2002, 41, 4132.
The hydrogenation of a microemulsion of benzonitrile,
(see entry 11) afforded after 12h in the presence of catalyst
3 43% of dibenzylamine [Eq. (7)]. We recall that the ion
[6] a) D. Myers, Surfaces, Interfaces, and Colloids: Principles and Appli-
cations, Wiley-VCH, Weinehim, 1999; b) R. Atkin, V. S. J. Craig,
E. J. Wanless, S. Biggs, Adv. Colloid Interface Sci. 2003, 103, 219.
[7] F. M. Menger, Angew. Chem. 1991, 103, 1104; Angew. Chem. Int.
Ed. Engl. 1991, 30, 1086.
[8] See, for example, a) E. A. Lissi, E. Abuin, Langmuir 2000, 16,
10084, and references therein; b) B. Orlich, H. Berger, M. Lade, R.
Schom‰cker, Biotechnol. Bioeng. 2000, 70, 638; c) E. Franqueville,
H. Stamatis, H. Loutrari, A. Friboulet, F. Kolisis, J. Biotechnol. 2002,
97, 177.
[9] See a) B. Jˆnsson, B. Lindman, K. Holmberg, B. Kronberg: Surface
and Polymers in Aqueous Solution, Wiley, Chichester, UK, 1998,
Chapter 18; b) K. Holmberg, Adv. Colloid Interface Sci. 1994, 51,
137; c) A. S. Chhatre, D. A. Joshi, J. Wang, B. D. Kulkarni, J. Colloid
Interface Sci. 1993, 158, 183; d) D. A. Jaeger, J. Wang, Tetrahedron
Lett. 1992, 33, 137.
[10] a) M. H‰ger, K. Holmberg, A. M. d×A. Rocha Gonalves, A. C.
Sevra, Colloids Surf. A. 2001, 183, 247; b) R. W. Hay, T. Clifferd, N.
Govan, Transition Met. Chem. 1998, 23, 619.
[11] a) W. Tic, I. Miesiac, J. Szymanowski, J. Colloid Interface Sci. 2001,
244, 423; b) M. Haumann, H. Koch, P. Hugo, R. Schom‰cker, Appl.
Catal. A 2002, 225, 239; c) M. Haumann, H. Yildiz, H. Koch, R.
Schom‰cker, Appl. Catal. A 2002, 236, 173.
pair [(CH3)(C8H17)3N]+[RhCl4]À¥nH2O encapsulated in a hy-
drophilic sol gel promotes hydrogenation of nitrobenzene
and benzonitrile to give mainly aniline and benzylamine, re-
spectively.[18] Here we observed that modification of the
sol gel matrix with the hydrophobic group changes the se-
lectivity of the catalyst. (Modification of selectivity was also
observed in the regular emulsion system.[5])
Finally, it should be noted that in all of the hydrogenation
catalyses we were able to recover the solid-state components
of the system simply by filtration and re-use the catalyst for
three consecutive runs.
In conclusion, we have shown the versatility of the three-
phase emulsion-surface transport (EST) methodology by uti-
lizing microemulsified substrates. The concept is general and
we believe that carrying out many-hydrophobic reactions in
aqueous media will prove feasible and be of economical and
environmental benefit.
[12] A. V. Cheprakov, N. V. Ponomareva, I. P. Beletskaya, J. Organomet.
Chem. 1995, 486, 297.
[13] a) S. Zhang, J. F. Rusling, J. Colloid Interface Sci. 1996, 182, 558;
b) G. N. Kaman, J. F. Rusling, Langmuir 1996, 12, 2645.
[14] a) M. Dreja, B. Tieke, Langmuir 1998, 14, 800; b) X. Fu, S. Qutu-
buddin, Langmuir 2002, 18, 5058; c) M. GÛmez-Cisneros, R. D. Per-
alta, L. C. Cesteros, I. Katime, E. Mendizµbal, J. E. Puig, Polymer
2002, 43, 2993.
Acknowledgement
[15] R. R. Schrock, J. A. Osborn, J. Am. Chem. Soc. 1971, 93, 2397.
[16] J. Schulz, A. Roucoux, H. Patin, Chem. Eur. J. 2000, 6, 618.
[17] R. K. Iler, The Chemistry of Silica, Wiley, New York, 1987.
[18] J. Blum, A. Rosenfeld, N. Polak, O. Israelson, H. Schumann, D.
Avnir, J. Mol. Catal. A 1996, 107, 217.
This work was supported by the Ministry of Science, Art, and Sport
through the Tashtiot project and by the Israel Science Foundation
through Grant No. 143/00 12.0.
[1] J. D. Mackenzie, J. Sol-Gel Sci. Technol. 2003, 26, 2 3.
[2] J. Blum, D. Avnir, H. Schumann, Chemtech 1999, 29, 32 .
[3] Handbook of Microemulsion Science and Technology, (Eds.: P.
Kumer, K. L. Mittal), New York, Marcel Dekker, 1999.
Received: October 1, 2003 [F5591]
962
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 958 962