Stabilized rhodium(0) nanoparticles: A reusable hydrogenation catalyst for arene derivatives in a biphasic water-liquid system

Article abstract of DOI:10.1002/(SICI)1521-3765(20000218)6:4<618::AID-CHEM618>3.0.CO;2-A

A colloidal system based on an aqueous suspension of rhodium(0) nanoparticles proved to be an efficient catalyst for the hydrogenation of arene derivatives under biphasic conditions. The rhodium nanoparticles (2 - 2.5 nm) were synthesized by the reduction of RhCl3·3 H₂O with sodium borohydride and were stabilized by highly water-soluble N-alkyl-N-(2- hydroxyethyl)ammonium salts (HEA-C(n)). These surfactant molecules were characterized by measurements of the surface tension and the aqueous dispersions with rhodium were observed by transmission electron cryomicroscopy. The catalytic system is efficient under ultramild conditions, namely room temperature and 1 atm H₂ pressure. The aqueous phase which contains the protected rhodium(0) colloids can be reused without significant loss of activity. The microheterogeneous behavior of this catalytic system was confirmed on a mercury poisoning experiment.

Full text of DOI:10.1002/(SICI)1521-3765(20000218)6:4<618::AID-CHEM618>3.0.CO;2-A

FULL PAPER
Stabilized Rhodium(0) Nanoparticles: A Reusable Hydrogenation Catalyst for
Arene Derivatives in a Biphasic Water± Liquid System
Jürgen Schulz, Alain Roucoux,* and Henri Patin[a]
Abstract: A colloidal system based on
an aqueous suspension of rhodium(0)
nanoparticles proved to be an efficient
catalyst for the hydrogenation of arene
derivatives under biphasic conditions.
The rhodium nanoparticles (2 ± 2.5 nm)
were synthesized by the reduction of
RhCl3 ´ 3H2O with sodium borohydride
and were stabilized by highly water-
soluble N-alkyl-N-(2-hydroxyethyl)am-
monium salts (HEA-Cn). These surfac-
tant molecules were characterized by
measurements of the surface tension
and the aqueous dispersions with rho-
dium were observed by transmission
electron cryomicroscopy. The catalytic
system is efficient under ultramild con-
ditions, namely room temperature and
1 atm H2 pressure. The aqueous phase
which contains the protected rhodium(0)
colloids can be reused without signifi-
cant loss of activity. The microheteroge-
neous behavior of this catalytic system
was confirmed on a mercury poisoning
experiment.
Keywords: arenes ´ biphasic cataly-
sis ´ colloids ´ hydrogenations ´
nanostructures ´ rhodium
prevent colloids from aggregating in water: i) electrostatic
stabilization with ionic species,[2a] ii) steric protection based
on the use of polymers,[11] and iii) electrosteric stabilization
generated by surfactants or polyoxoanions.[12]
Introduction
For economic and ecological reasons, biphasic catalysis based
on two immiscible liquid phases, water and hydrocarbons,
have attracted increasing attention.[1] In fact, such catalytic
systems offer a way to separate and recycle the catalyst by
simple decantation. The most classical approaches use water-
soluble homogeneous catalysts coordinated to phosphines
which contain hydrophilic functional groups (carboxylate,
sulfonate, hydroxy, etc.). Thus, several industrial processes are
essentially based on the exceptional water-solubility of ligands
such as the triphenylphosphinetrisulfonate sodium salt
(TPPTS-Na).
Our alternative is to use metallic nanoparticles that are
finely dispersed in water to allow an easy separation of the
catalyst from the reaction product.[2] During the last decade,
nanoparticles dispersed in liquid media have been studied[3] in
catalytic reactions, such as hydrogenation,[4±7] oxidation,[8]
hydrosilation,[9] or more recently, C ± C coupling.[10] These
microheterogeneous systems show a great potential because
of the large surface area of the particles. These small metal
particles can work efficiently as a catalyst if aggregation does
not occur. To prevent this phenomenon and to facilitate
recycling, particles must be stabilized by a highly water-
soluble protective agent. Three main methods are known to
We have chosen an ionic surfactant to prepare and to
protect the aqueous colloidal suspension of metallic particles.
We have synthesized N-alkyl-N-(2-hydroxyethyl)ammonium
salts (HEA), which provide sufficiently hydrophilic character
to maintain the catalytic species within aqueous phase.[2c]
The hydrogenation of benzene derivatives represents an
important industrial catalytic transformation.[13] Generally,
this reaction is carried out with heterogeneous catalysts.[14]
Some pure homogeneous[15] or microheterogeneous[5] systems
have been reported, but they require in many, if not most,
cases drastic conditions (high pressure and/or temperature).
However, colloidal catalysts do give satisfactory results for the
benzene/arene reduction in organic mixtures.[5e] In this paper,
we show that our catalytic system (Rh/protective agent) can
efficiently catalyze the hydrogenation of arenes in biphasic
liquid ± liquid (water/organic phase) media at room temper-
ature and under 1 atm H2 pressure. We demonstrate that
surfactant-protected colloids can be used in pure biphasic
conditions with a satisfactory recycling process.
[a] Dr. A. Roucoux, J. Schulz, Prof. Dr. H. Patin
Laboratoire de Chimie Organique
Results and Discussion
et des Substances NaturellesÐCNRS ESA 6052,
Ecole Nationale Superieure de Chimie de Rennes
Avenue du Gal Leclerc, 35700 Rennes (France)
Fax : (‡ 33)299-871-332
We have already demonstrated that aqueous dispersions of
rhodium particles can efficiently catalyze alkene hydrogena-
tion in biphasic systems with i) electrostatic stabilization by
the use of trisulfonated triphenylphosphine oxide[2a] or
E-mail: alain.roucoux@ensc-rennes.fr
618
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0604-0618 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 4

618±624
ii) electrosteric stabilization based on the use of trisulfonated
molecules that contain a lipophilic side-chain and a bulky
trianionic hydrophilic group.[2b]
than 12 carbon atoms are surfactants and self-aggregate into
micelles above the critical micellar concentration (cmc) of 1 Â
2
3
3
1
10 , 2.5 Â 10 , 1 Â 10 , and 2.5 Â 10 4 molL for HEA-C12,
HEA-C14, HEA-C16, and HEA-C18, respectively. The decrease
in the surface tension of aqueous solutions of these surfactants
In this context, we have studied a new series of easily
synthesized surfactants that stabilize an active catalytic
system in the hydrogenation of arenes in a biphasic (liquid ±
liquid) medium.[2c] Thus, N-alkyl-N-(2-hydroxyethyl)ammo-
nium salts, HEA-Cn, have been prepared with an alkyl chain
bearing n ˆ 12 ± 18 carbon atoms. These molecules are ob-
tained by quaternization of N,N-dimethylethanolamine with
the appropriate bromoalkanes (Scheme 1).[16] All the HEA-
Cn compounds are highly soluble in water, but their properties
1
below the cmc (from 72 mNm for pure water to about
35 mNm for the surfactant-containing solutions) is related
1
to their adsorption at the water/air interface by the Gibbꢁs
law.[17] These series of HEA-Cn salts exhibited classical
2
behavior: the cmc values decreased from 1 Â 10 to 2.5 Â
4
10 as the hydrocarbon chain length increased and we
observed an usual linear variation of log cmc versus the
number of carbon atoms (Figure 2).
Scheme 1. Synthesis of N-alkyl-N-(2-hydroxyethyl)ammonium salts,
HEA-Cn.
depend on the number of carbons in the alkyl chain. Surface
tension measurements (Figure 1) demonstrated that all HEA-
Cn compounds which bear a lipophilic alkyl chain of more
Figure 2. Plot of log cmc versus alkyl chain length.
Based on these results, catalytically active aqueous suspen-
sions were made of metallic rhodium(0) colloids prepared at
room temperature by the reduction of rhodium trichloride
with sodium borohydride in dilute aqueous solutions of HEA-
Cn salts. Nevertheless, only the surfactants HEA-C16 and
HEA-C18 gave rise to stable monodispersed colloidal dis-
persions. With the compounds HEA-C12 and HEA-C14 the
rhodium suspensions were less stable and the metal particles
quickly aggregated. Accordingly, an effective electrosteric
stabilization requires a sufficiently lipophilic substituent,
namely, it must contain at least 16 carbon atoms in the side-
chain attached to the head group, a highly hydrophilic N-(2-
hydroxyethyl)ammonium salt.
Figure 1. Surface tensions of aqueous solutions of surfactants HEA-C12±18
.
The particle size of the optimized system Rh/HEA-C16 was
determined by transmission electron cryomicroscopy. We
compared the rhodium suspensions prior to catalysis and after
the first run (Figure 3). The histograms of the size distribution
of the nanoparticles were estimated after the original negative
had been digitally scanned for more accurate resolution.
Measurement of about 300 particles was made with a program
that counted the objects automatically based on shape
recognition.[18] The comparative TEM studies showed that
the catalytic suspensions had a similar average diameter
(2.1 nm versus 2.2 nm) and thus confirmed the conservation of
the size distribution (Figure 3). The ªnearly monodispersedº
colloidal suspension in water was highly stable under our mild
catalytic conditions (1 atm H2, 208C); this justifies a compar-
ison of catalytic behavior during recycling. More specifically,
our catalytic suspensions have good stability and can be used
up to 608C without degradation; although it should be noted
that the stability of our particles is somewhat less than that of
Abstract in French: Un syst›me catalytique colloïdal constituØ
dꢀune suspension de nanoparticules de rhodium(0) a montrØ
une bonne efficacitØ pour lꢀhydrogØnation de dØrivØs dꢀar›nes
en milieu biphasique. Des particules de rhodium dꢀenviron
2 nm sont obtenues par rØduction de RhCl3 ´ 3H2O par le
borohydrure de sodium en prØsence de sels hydrosolubles de
N-alkyl-N-(2-hydroxyØthyl)ammonium (HEA-Cn). Ces molØ-
cules tensioactives ont ØtØ caractØrisØes par des mesures de
tension de surface et les suspensions aqueuses de rhodium
observØes par cryomicroscopie Ølectronique à transmission. Le
syst›me catalytique est actif dans des conditions douces de
tempØrature et de pression (208C, 1 atm dꢀH2). La suspension
aqueuse de rhodium(0) sØparØe par simple dØcantation peut
Ã
etre rØutilisØe sans perte notable dꢀactivitØ. Enfin, le caract›re
microhØtØrog›ne du syst›me a ØtØ confirmØ par un test
dꢀempoisonnement au mercure.
Chem. Eur. J. 2000, 6, No. 4
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0604-0619 $ 17.50+.50/0
619

A. Roucoux et al.
FULL PAPER
alcohol) catalysts gave a monophasic system and, consequent-
ly, cannot be separated by simple decantation for the purpose
of recycling.
The molecular ratio R ˆ HEA-C16/Rh was optimized to
prevent aggregation and to provide a good activity. Specifi-
cally, the hydrogenation of anisole to the corresponding
methoxycyclohexane was studied with different ratios of
HEA-C16/Rh (Figure 4). Below a molar ratio R ˆ 2, the
rhodium particles aggregated with a total loss of activity for
hydrogenation. A molar ratio R ˆ 2 gave better activity and
was sufficient to maintain stable nanoparticles within the
aqueous phase during the catalytic process and hence allowed
its recycling. The activity decreased when R > 2 was used. In
this case, the quantity of surfactant present around the metal
particles was somewhat excessive and therefore prevented the
access of the substrates to the active site(s).
Figure 3. Comparison of colloids stabilized by HEA-C16 a) before and
b) after catalysis (scale barˆ 50 nm).
nanoclusters previously reported which are stable up to
1008C.[19]
Figure 4. Hydrogenation of anisole with various amounts of HEA-C16
(R ˆ HEA-C16/Rh).
Rhodium(0) colloids are known to reduce arenes,[20] but as
far as we know, our results provide the first example of an
efficient conversion under pure biphasic conditions, with
recycling at room temperature, and under 1 atm hydrogen
pressure. In preliminary studies, we compared the new
colloidal rhodium system with a standard Rh/C heteroge-
neous catalyst and with the classical water-soluble Rh/poly-
vinylpyrrolidone and Rh/poly(vinyl alcohol) systems synthe-
sized in hydroalcoholic media by Hirai.[21] These catalytic
systems were investigated by means of the reduction of
anisole; each reaction was performed under identical con-
ditions (Table 1). Our colloids exhibited the best catalytic
activity. Furthermore, only Rh/HEA-C16 can be reused for
further runs. The Rh/polyvinylpyrrolidone or Rh/poly(vinyl
The durability of the catalytic system was investigated by
employing it in several successive hydrogenations. For this,
anisole was again selected as a reference substrate. After a
1
first cycle (TOF ˆ 60 h ), the aqueous phase containing the
monodispersed colloidal suspensions of rhodium was sepa-
rated from the methoxycyclohexane product by slow (over-
night) decantation and then reused in a second run. In the
same way, the catalytic suspension was recovered for a third,
fourth, and fifth hydrogenation cycle. Figure 5 shows a
comparable turnover activity for all runs. We have observed
Table 1. Comparison of standard catalysts versus the Rh/HEA-C16 catalyst for
the hydrogenation of anisole.[a]
1st Run
t [h] TOF [h
2nd Run
t [h] TOF [h ]
1
[b]
1 [b]
Catalyst
]
[d]
Rh/C (5 wt.%; Degussa-type DG10) 15
20
33
±
±
±
57
[e]
Rh/PVA (M.W. 22000)[c]
Rh/PVP (M.W. 40000)[c]
Rh/HEA-C16
9
[e]
6.4 47
60
5
5.3
[a] Reaction conditions: metal (3.8 Â 10 5 mol), anisole (3.8 Â 10 3 mol), water
1
(10 mL), hydrogen pressure (1 atm), temperature (208C), stirred at 1500 min
.
[b] Turnover frequency defined as number of moles of consumed H2 per mole
of rhodium per hour. [c] Rh colloids synthesized as previously described by
Hirai.[21] [d] The Rh/C cannot be easily recycled by filtration. [e] Cannot be
recycled by decantation.
Figure 5. Activity of the recycled aqueous phase in the reduction of
anisole.
620
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0604-0620 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 4

Rhodium Nanoparticles
618±624
Table 2. Hydrogenation of benzene and monosubstituted derivatives under biphasic conditions.[a]
Stabilizing
1st Run
TOF [h
2nd Run
TOF [h ]
1
[c]
1 [c]
Substrate
agent
Substrate/Rh0
Product (yield%)[b]
t [h]
]
t [h]
benzene
benzene
benzene
toluene
toluene
toluene
ethylbenzene
propylbenzene
cumene
HEA-C16
HEA-C16
HEA-C18
HEA-C16
HEA-C16
CTAB
HEA-C16
HEA-C16
HEA-C16
100
200
100
100
200
100
100
100
100
cyclohexane (100)
cyclohexane (100)
cyclohexane (100)
methylcyclohexane (100)
methylcyclohexane (100)
methylcyclohexane (100)
ethylcyclohexane (100)
propylcyclohexane (100)
isopropylcyclohexane (100)
5.3
6.6
9.1
5.7
6.5
7.8
6.9
7.1
7.9
57
91
33
53
92
38
43
42
38
5.9
7.1
13.3
6.3
7.6
9.8
7.4
7.6
8.6
51
85
23
48
79
31
40
39
35
[a] Reaction conditions: catalyst (3.8 Â 10 5 mol), surfactant (7.6 Â 10 5), water (10 mL), hydrogen pressure (1 atm), temperature (208C), stirred at
1500 min 1. [b] Determined by GC analysis. [c] Turnover frequency defined as number of moles of consumed H2 per mole of rhodium per hour.
that the conservation of the TOF value during recycling
depends primarily on the amount of stable nanoparticles
remaining in the aqueous phase after separation; hence, a
meticulous and slow decantation is necessary in order to
conserve an efficient catalytic activity. Additionally, we
correlated the catalytic activity with the amount of rhodium
within the aqueous phase. An atomic absorption analysis (by
means of a rhodium lamp) of dilute solutions showed that the
concentration of the metal in the aqueous phase before and
selective catalytic reaction gave good results and the con-
version was usually complete after 7 h for a ratio S/Rh ˆ 100
(S ˆ substrate). Most importantly the ratio S/Rh (ˆ 100 or
200) has little effect on the reaction time. In contrast with the
N-alkyl-N-(2-hydroxethyl)ammonium salt, cetyltrimethylam-
monium bromide (CTAB) gave poor results. The turnover
frequency decreased during the first and second run as can be
seen for toluene. These results justify the use of the
hydroxyammonium function as the hydrophilic head group.
Unfortunately, we did not observe any cyclohexene or
cyclohexadiene derivatives as intermediates which ideally
would have been desirable. Nevertheless, steric substituent
effects of the arene influenced the reaction time. The
increasing steric hindrance of the alkyl group in the series
benzene, toluene, ethylbenzene, propylbenzene, and cumene
is clearly indicated by the increase in the reaction time and the
decrease in the catalytic TOF.
In a second series of studies, we investigated the hydro-
genation of functionalized arenes (Table 3). The reaction was
influenced by the electronic effects of the arene substituents.
Thus, arenes substituted by electron-withdrawing groups
reacted more slowly, whereas those with electron-donating
groups reacted more quickly. Here, electron-rich substrates
should be preferentially adsorbed on the rhodium colloids and
thus the reaction is accelerated[5d] for arenes such as anisole
(5 h). Surprisingly, halogenated benzenes were not hydro-
1
after catalysis was similar. A small loss (1.4 Æ 0.1 mgL ) was
1
observed between the original catalytic solution (11.7 mgL )
1
and that after the fifth run (10.3 mgL ); this corresponds to a
rhodium loss of ꢀ10%. Consequently, these water-soluble
surfactants HEA-C16±18 satisfactorily maintain the colloidal
rhodium particles within the aqueous phase during the
catalytic process and thus allow the reuse of the aqueous
suspensions in biphasic catalytic systems (water/hydrocarbon)
to within Æ10%.
We have also hydrogenated benzene and some of its
monoalkyl-substituted derivatives under biphasic conditions
(Table 2). The reaction was monitored by the volume of
hydrogen consumed and a gas chromatographic analysis. The
turnover frequency (TOF), defined as number of moles of
consumed H2 per mole of rhodium per hour, was determined
for the first run and for recycling after separation of the
aqueous phase. No induction period was observed. The very
Table 3. Hydrogenation of functionalized derivatives under biphasic conditions.[a]
Stabilizing
1st Run
TOF [h
2nd Run
TOF [h ]
1
[c]
1 [c]
Substrate
agent
Substrate/Rh0
Product (yield%)[b]
t [h]
]
t [h]
anisole
anisole
anisole
anisole
anisole
anisole
phenol
phenol
ethyl benzoate
bromobenzene[d]
aniline
styrene
a-methylstyrene
allylbenzene
HEA-C16
HEA-C16
HEA-C16
HEA-C16
HEA-C18
CTAB
HEA-C16
HEA-C18
HEA-C16
HEA-C16
HEA-C16
HEA-C16
HEA-C16
HEA-C16
100
200
500
1000
100
100
100
100
100
100
100
100
100
100
methoxycyclohexane (100)
methoxycyclohexane (100)
methoxycyclohexane (100)
methoxycyclohexane (100)
methoxycyclohexane (100)
methoxycyclohexane (100)
cyclohexanol (100)
cyclohexanol (100)
ethyl cyclohexanoate (100)
±
5
6.2
10
60
97
150
188
36
59
58
54
32
±
5.3
7.1
11.4
20
11.5
24
5.7
8.4
11.9
±
57
85
132
150
26
13
53
36
25
±
17
8.4
5.1
5.2
5.6
9.5
±
cyclohexylamine (100)
ethylcyclohexane (100)
isopropylcyclohexane (100)
propylcyclohexane (100)
10
30
55
43
39
±
±
7.3
9.2
10.3
8.4
10.1
11.2
48
40
36
[a] Reaction conditions: catalyst (3.8 Â 10 5 mol), surfactant (7.6 Â 10 5), water (10 mL), hydrogen pressure (1 atm), temperature (208C), stirred at
1500 min 1. [b] Determined by GC analysis. [c] Turnover frequency defined as number of moles of consumed H2 per mole of rhodium per hour.
[d] Similarly, PhI, PhCl, and PhF are not reduced.
Chem. Eur. J. 2000, 6, No. 4
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0604-0621 $ 17.50+.50/0
621

A. Roucoux et al.
FULL PAPER
genated, not even slowly, as a consequence of catalyst
deactivation. Under our standard conditions, no selective
catalytic hydrogenation to afford a cyclohexyl product was
observed with i) a mixture of iodobenzene and benzene, or
ii) a recycled catalytic suspension which was used for the
hydrogenation of a halogenated compound; hence, the
electron-withdrawing effect of the halogen is not responsible
for the lack of reaction. Instead, it must be caused by a
poisoning effect of the substrate.
the catalyst (Figure 6). In light of prior reports,[5e] these results
provide good evidence that the Rh0 nanoparticles are the true
catalysts.
The chemoselectivity was good: for example, ester and
ether functions remained unaffected. Another case in point is
the hydrogenation of aniline which gave the cyclohexylamine;
here the catalyst cannot be recycled, since the colloidal
aqueous phase was insufficiently stable and generally aggre-
gated. We believe that the ammonium functionality was
reacting with the product, which resulted in a loss of
stabilization. In addition to the hydrogenation of the aromatic
ring, reducible components of the substituents can also be
hydrogenated, for example, exo-C ± C double bonds of
styrene, a-methylstyrene, and allylbenzene. In all cases, we
observed the reduction of the double bond prior to the
hydrogenation of the aromatic cycle. Nevertheless in the last
case, we detected (in addition to the hydrogenation of the
double bond to form propylbenzene) the partial isomerization
of the double bond with formation of b-methylstyrene. This
compound was then hydrogenated to propylbenzene and
finally to propylcyclohexane.
In a third set of studies, we examined the hydrogenation of
disubstituted benzene derivatives, such as xylene or methyl-
anisole compounds, with the optimized system HEA-
C16/Rh ˆ 2. A complete conversion was obtained in all cases
(Table 4). The cis products were largely the major products, as
is usually observed in heterogeneous catalytic systems.[22]
Finally, we tested the behavior of our catalytic system by
means of mercury poisoning experiments. Mercury is a well-
known heterogeneous-catalyst poison on account of its
classical adsorption onto the surface of transition metal(0)
catalysts.[5e] A large excess of Hg0 was added after 50%
conversion to a catalytically active solution containing
anisole. After stirring for 1 h, the solution was reconnected
to the hydrogenation apparatus. No catalytic activity was
detected, which indicated that Hg had completely inactivated
Figure 6. Hg0 poisoning of the catalyst in the reduction of anisole.
Conclusions
The results presented herein confirm that the Rh0 nano-
particles protected by N-alkyl-N-(2-hydroxyethyl)ammonium
salts are a very efficient catalyst for the reduction of arenes;
the results further demonstrate that our microheterogeneous
system can be used to catalyze hydrogenation in two liquid
phases, under mild conditions, and with recycling. These
experiments confirm that microheterogeneous colloidal cat-
alysis at room temperature and 1 atm hydrogen pressure, as
demonstrated by TEM observations and mercury poisoning
experiments, is a real alternative to biphasic liquid ± liquid
homogeneous catalysis, provided that the monodispersed
colloidal suspensions of the rhodium particles are prepared
and correctly stabilized with the correct, highly water-soluble
protecting agent.
Experimental Section
Starting materials: Rhodium chloride hydrate and mercury were obtained
from Strem Chemicals. Rh/C (5 wt.%; Degussa-type G10), poly(vinyl
Table 4. Hydrogenation of disubstituted benzene derivatives under biphasic conditions.[a]
Stabilizing
1st Run
TOF [h
2nd Run
TOF [h ]
1
[c]
1 [c]
Substrate
agent
Substrate/Rh0
Product (yield%)[b]
t [h]
]
t [h]
o-xylene
HEA-C16
100
1,2-dimethylcyclohexane
cis (95), trans (5)
1,3-dimethylcyclohexane
cis (87), trans (13)
1,4-dimethylcyclohexane
cis (70), trans (30)
1-methoxy-2-methylcyclohexane
cis (97), trans (3)
1-methoxy-4-methylcyclohexane
cis (92), trans (8)
3-methylcyclohexanol
cis (99), trans (1)
7.5
40
41
42
32
33
37
8.9
48
35
37
29
30
33
m-xylene
HEA-C16
HEA-C16
HEA-C16
HEA-C16
HEA-C16
100
100
100
100
100
7.3
7.1
9.5
9.2
8.2
8.5
8.2
p-xylene
o-methylanisole
p-methylanisole
m-cresol
10.3
10
9.1
[a] Reaction conditions: catalyst (3.8 Â 10 5 mol), surfactant (7.6 Â 10 5), water (10 mL), hydrogen pressure (1 atm), temperature (208C), stirred at
1500 min 1. [b] Determined by GC analysis. [c] Turnover frequency defined as number of moles of consumed H2 per mole of rhodium per hour.
622
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0604-0622 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 4

Rhodium Nanoparticles
618±624
alcohol) (PVA; MWˆ 22000), polyvinylpyrrolidone (PVP; MWˆ 40000),
sodium borohydride, various bromoalkanes, N,N-dimethylethanolamine,
and all aromatic substrates were purchased from Aldrich or Fluka and were
used without further purification. Water was distilled twice before use by a
conventional method. The surfactants HEA-C12±18 were prepared and fully
characterized as previously reported.[16]
Engl. 1995, 34, 1575 ± 1577; c) F. Joo, A. Katho, J. Mol. Catal. A 1997,
116, 3 ± 26; d) B. Cornils, W. A. Herrmann, in Aqueous-Phase
Organometallic Catalysis, Wiley-VCH, Weinhem, 1998.
[2] a) C. Larpent, H. Patin, J. Mol. Catal. 1988, 44, 191 ± 195; b) C.
Larpent, E. Bernard, F. Brisse-Le Menn, H. Patin, J. Mol. Catal. A
1997, 116, 277 ± 288; c) J. Schulz, A. Roucoux, H. Patin, Chem.
Commun. 1999, 535 ± 536.
[3] a) G. Schmid, Chem. Rev. 1992, 92, 1709 ± 1727; b) L. N. Lewis, Chem.
Rev. 1993, 93, 2693 ± 2730; c) G. Schmid, in Clusters and Colloids,
VCH, Weinhem, 1994, p. 459.
[4] For the hydrogenation of C ± C double or triple bonds, see for
example: a) H. Bönnemann, W. Brijoux, K. Siepen, J. Hormes, R.
Franke, J. Pollmann, J. Rothe, Appl. Organomet. Chem. 1997, 11, 783 ±
796; b) H. Bönnemann, W. Brijoux, A. Schulze Tilling, K. Siepen, Top.
Catal. 1997, 4, 217 ± 227; c) W. Yu, M. Liu, H. Liu, X. Ma, Z. Liu, J.
Colloid Interface Sci. 1998, 208, 439 ± 444; d) H. Hirai, N. Yakura, Y.
Seta, S. Hodoshima, React. Funct. Polym. 1998, 37, 121 ± 131; e) E.
Sulman, Y. Bodrova, V. Matveeva, N. Semagina, L. Cerveny, V. Kurtc,
L. Bronstein, O. Platonova, P. Valetsky, Appl. Catal. A 1999, 176, 75 ±
81.
Analytical procedures: The surface tension measurements were performed
at 208C by means of the ring method with a Du Nouy tensiometer
(Krüss K10T).
The transmission electronic cryomicroscopic studies were conducted on a
PHILIPS CM12 transmission electron microscope at 100 KeV. Samples
were prepared by a dropwise addition of the stabilized colloid in water onto
a Cu mesh covered with carbon. The colloidal dispersion was removed after
1 min with cellulose and the samples were quickly frozen in liquid ethane
before their transfer to the microscope.
Gas chromatography was performed on a Carlo Erba GC 6000 with a FID
detector equipped with an AlltechAT1 column (30 m long, 0.25 mm inner
diameter). Parameters were as follows: initial temperature 408C; initial
time 3 min; ramp 88Cmin 1; final temperature 1408C; final time 5 min;
injector temperature 2208C; detector temperature 2508C; injection volume
0.3 mL.
[5] For the hydrogenation of arenes, see for example: a) P. Drognat Lan-
Â
dre, M. Lemaire, D. Richard, P. Gallezot, J. Mol. Catal. 1993, 78, 257 ±
The atomic absorption measurements were performed on a Varian AA-
1275 spectrometer with the following parameters: lamp current 12 mA,
lRh ˆ 343.5 nm, slit width 0.5 mm, air/acetylene flame. The sample (3 mL)
was diluted in water (100 mL). The concentration (mgL 1) was determined
by means of a calibration graph.
Â
261; b) K. Nasar, F. Fache, M. Lemaire, J.-C. Beziat, M. Besson, P.
Â
Gallezot, J. Mol. Catal. 1994, 87, 107 ± 115; c) P. Drognat Landre, D.
Richard, M. Draye, P. Gallezot, M. Lemaire, J. Catal. 1994, 147, 214 ±
222; d) F. Fache, S. Lehuede, M. Lemaire, Tetrahedron Lett. 1995, 36,
885 ± 888; e) K. S. Weddle, J. D. Aiken III, R. G. Finke, J. Am. Chem.
Soc. 1998, 120, 5653 ± 5666.
Synthesis of the aqueous Rh0 colloidal suspensions: The suspensions were
prepared under nitrogen at 208C. Sodium borohydride (36 mg, 9.5 Â
10 4 mol) was added to an aqueous solution of surfactant (95 mL) with
[6] For the hydrogenation of C ± O double bonds, see for example: a) W.
Yu, H. Liu, Q. Tao, Chem. Commun. 1996, 1773 ± 1774; b) H. Feng, H.
Liu, J. Mol. Catal. A 1997, 126, L5 ± L8; c) H. Bönnemann, G. A.
Braun, Angew. Chem. 1996, 108, 2120 ± 2123; Angew. Chem. Int. Ed.
Engl. 1996, 35, 1992 ± 1995; d) H. Bönnemann, G. A. Braun, Chem.
Eur. J. 1997, 3, 1200 ± 1202; e) W. Yu, H. Liu, X. An, J. Mol. Catal. A
1998, 129, L9 ± L13; f) Ref. [4c]; g) W. Yu, H. Liu, M. Liu, Q. Tao, J.
Mol. Catal. A 1999, 138, 273 ± 286; h) W. Yu, M. Liu, H. Liu, X. An, Z.
Liu, X. Ma, J. Mol. Catal. A 1999, 142, 201 ± 211.
[7] For the hydrogenation of nitro compounds, see for example:
a) M. Liu, W. Yu, H. Liu, J. Mol. Catal. A 1999, 138, 295 ± 303;
b) Ref. [6h].
[8] a) F. Launay, H. Patin, New J. Chem. 1997, 21, 247 ± 256; b) Ref. [4b];
c) F. Launay, A. Roucoux, H. Patin, Tetrahedron Lett. 1998, 39, 1353 ±
1356; d) H. Bönnemann, W. Brijoux, R. Brinkmann, A. Schulze Til-
ling, T. Schilling, B. Tesche, K. Seevogel, R. Franke, J. Hormes, G.
Köhl, J. Pollmann, J. Rothe, W. Vogel, Inorg. Chim. Acta 1998, 270,
95 ± 110; e) Y. Shiraishi, N. Toshima, J. Mol. Catal. A 1999, 141, 187 ±
192.
1
various concentrations (7.6 Â 10-3 molL gave the best results). This
solution was quickly added, under vigorous agitation, to an aqueous
solution (5 mL) of the precursor RhCl3 ´ 3H2O (100 mg, 3.8 Â 10 4 mol) to
give an aqueous Rh0 colloidal suspension (100 mL). The reduction
occurred instantaneously and was characterized by a color change from
red to black. The suspensions obtained were stable for months, as
confirmed by TEM (the sizes of the particles remain unmodified over this
time frame).
General hydrogenation procedure: All hydrogenation reactions were
carried out under standard conditions (208C, 1 atm of H2). A round-
bottom flask (25 mL), charged with the chosen aqueous suspension of Rh0
(10 mL) and a magnetic stirrer, was connected to a gas burette (500 mL)
with a flask to balance the pressure. The flask was closed by a septum, and
the system was filled with hydrogen. The appropriate aromatic substrate
(3.8 Â 10 3 mol) was injected through the septum, and the mixture was
stirred (1500 min 1). The reaction was monitored by the volume of gas
consumed and by gas chromatography. At the end of the reaction, the two
phases were separated by decantation and the aqueous phase was reused in
a second run. The turnover frequencies (TOF) were determined for 100%
conversion.
Mercury poisoning experiment:[5e] A standard hydrogenation was con-
ducted with anisole (411 mg, 3.8 Â 10 3 mol) as the reference substrate.
Subsequently, a second hydrogenation of anisole (411 mg, 3.8 Â 10 3 mol)
was conducted. After 50% conversion, the reaction was stopped and
mercury was added (2.29 g, 1.14 Â 10 2 mol, 300 equiv; as described
elsewhere[5e]). The mixture was stirred for 1 h and was then reconnected
to the hydrogenation apparatus. The two hydrogenation reactions were
compared.
[9] G. Schmid, H. West, H. Mehles, A. Lehnert, Inorg. Chem. 1997, 36,
891 ± 895.
[10] a) M. Beller, H. Fischer, K. Kühlein, C.-P. Reisinger, W. A. Herr-
mann, J. Organomet. Chem. 1996, 520, 257 ± 259; b) M. T. Reetz, G.
Lohmer, Chem. Commun. 1996, 1921 ± 1922; c) M. T. Reetz, R.
Breinbauer, K. Wanninger, Tetrahedron Lett. 1996, 37, 4499 ± 4502;
d) S. Klingelhöfer, W. Heitz, A. Greiner, S. Oestreich, S. Förster, M.
Antonietti, J. Am. Chem. Soc. 1997, 119, 10116 ± 10120.
[11] a) C. W. Chen, M. Akashi, Langmuir 1997, 13, 6465 ± 6472; b) A. B. R.
Mayer, J. E. Mark, Eur. Polym. J. 1998, 34, 103 ± 108; c) L. M.
Bronstein, S. N. Sidorov, A. Y. Gourkova, P. M. Valetsky, J. Hartmann,
M. Breulmann, H. Cölfen, M. Antonietti, Inorg. Chim. Acta 1998, 280,
348 ± 354.
[12] a) Y. Lin, R. G. Finke, J. Am. Chem. Soc. 1994, 116, 8335 ± 8353; b) Y.
Lin, R. G. Finke, Inorg. Chem. 1994, 33, 4891 ± 4910; c) T. Yonezawa,
T. Tominaga, N. Toshima, Langmuir 1995, 11, 4601 ± 4604; d) M. T.
Reetz, W. Helbig, S. A. Quaiser, U. Stimming, N. Breuer, R. Vogel,
Science 1995, 267, 367 ± 369; e) J. D. Aiken III, Y. Lin, R. G. Finke, J.
Mol. Catal. A 1996, 114, 29 ± 51; f) Y. Berkovich, N. Garti, Colloids
Surf. A. 1997, 128, 91 ± 99.
Acknowledgments
We thank CNRS and the Region Bretagne for financial support. We would
also like to thank Dr. Rolland (University of Rennes 1) for helpful
comments and the TEM experiments.
[13] K. Weissermel, H. J. Arpe, in Industrial Organic Chemistry, 2nd ed.,
VCH, New York, 1993, p. 343.
[14] a) S. Siegel, in Comprehensive Organic Synthesis, Vol. 5 (Eds.: B. M.
Trost, I. Fleming), Pergamon Press, New York, 1991; b) M. S. Eisen,
T. J. Marks, J. Am. Chem. Soc. 1992, 114, 10358 ± 10368; c) M. A.
[1] a) W. A. Herrmann, C. W. Kohlpaintner, Angew. Chem. 1993, 105,
1588 ± 1609; Angew. Chem. Int. Ed. Engl. 1993, 32, 1524 ± 1544; b) B.
Cornils, Angew. Chem. 1995, 107, 1709 ± 1711; Angew. Chem. Int. Ed.
Chem. Eur. J. 2000, 6, No. 4
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0604-0623 $ 17.50+.50/0
623

A. Roucoux et al.
FULL PAPER
Keane, J. Catal. 1997, 166, 347 ± 355; d) H. Gao, R. J. Angelici, J. Am.
Chem. Soc. 1997, 119, 6937 ± 6938.
[19] For examples of colloidal suspensions stable above 1008C, see: a) U.
Kolb, S. A. Quaiser, M. Winter, M. T. Reetz, Chem. Mater. 1996, 8,
1889 ± 1894; b) See also ref. [10].
[20] a) K. R. Januszkiewicz, H. Alper, Organometallics 1983, 2, 1055 ±
1057; b) J. Blum, I. Amer, K. P. C. Vollhardt, H. Schwarz, G. Höhne,
J. Org. Chem. 1987, 52, 2804 ± 2813; c) See also ref. [5].
[21] a) H. Hirai, Y. Nayao, N. Toshima, J. Macromol. Sci. Chem. 1978,
A12, 1117 ± 1141; b) H. Hirai, J. Macromol. Chem. 1979, A13, 633 ±
649; c) H. Hirai, Y. Nakao, N. Toshima, J. Macromol. Chem. 1979,
A13, 727 ± 750.
[15] a) I. P. Rothwell, Chem. Commun. 1997, 1331 ± 1338; b) L. Plasseraud,
G. Süss-Fink, J. Organomet. Chem. 1997, 539, 163 ± 170; c) E.
Garcia Fidalgo, L. Plasseraud, G. Süss-Fink, J. Mol. Catal. A 1998,
132, 5 ± 12; d) P. J. Dyson, D. J. Ellis, D. G. Parker, T. Welton, Chem.
Commun. 1999, 25 ± 26.
[16] G. Cerichelli, L. Luchetti, G. Mancini, G. Savelli, Tetrahedron 1995, 51,
10281 ± 10288.
[17] D. Mayers, in Surfactant Science and Technology, VCH, Weinhem,
1992.
[22] H. Nagahara, M. Ono, M. Konishi, Y. Fukuoka, Appl. Surf. Sci. 1997,
121/122, 448 ± 451.
[18] J.-P. Rolland, P. Bon, D. Thomas, Comput. Appl. Biosci. 1998, 13, 563 ±
564.
Received: May 3, 1999 [F1761]
624
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0947-6539/00/0604-0624 $ 17.50+.50/0
Chem. Eur. J. 2000, 6, No. 4