Preparation of rhodium nanoparticles in carbon dioxide induced ionic liquids and their application to selective hydrogenation

Article abstract of DOI:10.1002/anie.200803773

(Figure Presented) In the trap: Catalytically active nanoparticles can be generated and stabilized in solid ionic matrices with the aid of supercritical CO₂. The method allows the use of simple ammonium salts that would not be classified as ion

Full text of DOI:10.1002/anie.200803773

Angewandte
Chemie
DOI: 10.1002/anie.200803773
Nanoparticle Catalysts
Preparation of Rhodium Nanoparticles in Carbon Dioxide Induced
Ionic Liquids and their Application to Selective Hydrogenation**
ˇ
Valentin Cimpeanu, Marijan Kocevar, Vasile I. Parvulescu,* and Walter Leitner*
There is currently great interest in the generation of metal
nanoparticles of controlled size and shape because of their
unique properties at the interface between molecular struc-
tures and bulk materials.[1,2] Ionic liquids (ILs) have emerged
as promising media for the synthesis, stabilization, and
utilization of metal nanoparticles for various applications,
including catalysis.[3,4] The stabilizing effect of ammonium
salts on metal nanoparticles is well-established in conven-
tional solvents[1,2,5] and is of course not restricted to materials
with melting points that fall within the definition of ILs (that
is, organic salts with melting points, Tm , below 1008C[6]). Most
recently, it has been demonstrated that common organic salts
can experience very significant melting point depression in
the presence of compressed CO2 with DTm values around or
above 1008C in certain cases.[7,8] Herein, we report a method
for the generation and entrapment of rhodium nanoparticles
in simple solid ammonium salts by exploiting their CO2-
induced melting to form ionic liquids. The utilization of the
resulting materials as selective catalysts for hydrogenation
reactions is exemplified, whereby a different catalytic behav-
ior from conventional homogeneous or heterogeneous cata-
lysts was noted for sterically encumbered aromatic olefins as
substrates.
The process for the generation of the matrix-embedded
nanoparticles is depicted in Figure 1. A mixture of the solid
ammonium salt [R4N]Br and the organometallic complex
Figure 1. Generation of matrix-embedded rhodium nanoparticles by
reduction in CO2-induced ionic liquids. a) Mixture of the ammonium
salt and solid molecular precursor complex [Rh(acac)(CO)2]; b) reduc-
tion under CO2/H2 in the CO2 induced ionic liquid phase (view into
the high pressure reactor including the magnetic stirrer bar); c) solid
material containing the embedded nanoparticles obtained after venting
the reactor.
[Rh(acac)(CO)2] (acac = acetylacetonate) as precursor (Rh
loading ca. 1% by weight) is placed into a window-equipped
autoclave. This mixture is then pressurized with CO2 and H2
and heated to 40–808C for the reduction step. Although most
simple ammonium salts have regular melting points around or
well above 1008C, they form liquid phases under the reaction
conditions because of the presence of the compressed CO2
phase.[7,8] This ensures the dissolution of the molecular
precursor and a homogeneous dispersion of the resulting
particles. Furthermore, the presence of CO2 is known to
enhance the availability of hydrogen in ionic-liquid phases,
which may further facilitate the reduction process.[9] Upon
venting the CO2/H2 mixture at the end of the reaction, the
nanoparticles are trapped in the solidifying matrix that is left
in the reactor.
[*] Dr. V. Cimpeanu, Prof. Dr. W. Leitner
Institut fꢀr Technische und Makromolekulare Chemie
RWTH Aachen, Worringerweg 1, 52074 Aachen (Germany)
Fax: (+49)241-802-2177
E-mail: leitner@itmc.rwth-aachen.de
Homepage: http://www.itmc.rwth-aachen.de
Prof. Dr. W. Leitner
Max-Planck-Institut fꢀr Kohlenforschung
Kaiser-Wilhelm-Platz 1, 45470 Mꢀlheim an der Ruhr (Germany)
Prof. Dr. V. I. Parvulescu
Department of Chemical Technology and Catalysis
University of Bucharest
Organic by-products are inevitably formed from the
ligands during the preparation of metal nanoparticles by
hydrogenation of organometallic precursor complexes. Such
side products can be readily removed from the resulting
material by extraction with supercritical CO2 (scCO2). This
was demonstrated in the present method by passing a scCO2
stream through the reactor after reduction and venting
through an indicator solution of Fe2(SO4)3 (1% in H2SO4
(0.1n)). The immediate appearance of the characteristic red
color of [Fe(acac)3] proved the presence of acetylacetone in
the CO2 flow.
B-dul Regina Elisabeta 4-12, 030016 Bucharest (Romania)
E-mail: v_parvulescu@yahoo.com
ˇ
Prof. Dr. M. Kocevar
Faculty of Chemistry and Chemical Technology
University of Ljubljana (Slovenia)
[**] This work was supported by the Deutsche Forschungsgemeinschaft
(SPP 1191 “Ionic Liquids”, grant no. LE 930/11) and the Fonds der
Chemischen Industrie. We thank Dr. B. Tesche and Axel Dreier (Max-
Planck-Institut fꢀr Kohlenforschung, Mꢀlheim am der Ruhr) for
TEM measurements and Dr. Robert Kaufmann (Deutsches Woll-
forschunginstitut, RWTH Aachen) for XPS spectra. The Alexander
von Humboldt Foundation is gratefully acknowledged for making
this collaboration possible through a fellowship to V.I.P.
With this simple procedure, well-defined rhodium nano-
particles were obtained in various ionic matrices at mild
temperatures as exemplified for three different ammonium
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200803773.
Angew. Chem. Int. Ed. 2009, 48, 1085 –1088
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1085

Communications
bromide salts and the resulting materials Rh-1 to Rh-3
(Table 1; for further examples see the Supporting Informa-
tion). The brownish solid particles could be redissolved under
pressurized CO2 to form homogeneous solutions. Even after
storage of the solids over three months, the nanoparticles in
Table 1: Selected characteristic data for rhodium nanoparticles embed-
ded in solid ammonium salts that were generated by CO2-induced ionic
liquid phases.[a]
Cat. Matrix
M.p.
[8C]
T
p[b]
Particle
Surface/ Rh0/RhI
bulk
[8C] [bar] size
ratio[c]
[nm]
atom
ratio
Rh-1 [Bu4N]Br
124[d] 80
240
150
220
3.3Æ1.5 0.33
2.3Æ0.8 0.47
1.4Æ0.3 0.78
0.63
0.62
0.59
Rh-2 [Hex4N]Br 100[d] 40
Rh-3 [Oct4N]Br
98[c] 60
[a] Reaction conditions: ionic matrix (0.5 g), precursor [Rh(acac)(CO)2]
(1% Rh), H2 (40 bar), and scCO2 (density: ca. 0.7 gmLÀ1), 180 min.
[b] Total pressure at reduction temperature. [c] Calculated by using XPS.
[d] Melting points of the pure matrix under standard conditions.
Rh-1 to Rh-3 remained stable and no change in their catalytic
behavior was observed. Consistent with other ammonium
halide stabilized nanoparticles,[10] typical particle sizes of the
homogenously dispersed clusters were in the range 1–4 nm,
with relatively narrow distributions in most cases, as deter-
mined from TEM micrographs (Figure 2a).
X-ray photoelectron spectroscopic (XPS) analysis of the
new materials Rh-1 to Rh-3 showed the characteristic signals
for the matrix[11] and the embedded rhodium nanoparticles
(Figure 2b,c).[12] Deconvolution of the signals corresponding
to the Rh 3d5/2 level revealed two components; the minor
component is centered around 307.2 eV and can be assigned
to fully reduced Rh0 centers.[13] The binding energies of the
major signal are in the range 308.0–308.3 eV, that is between
Figure 2. a) Representative TEM micrograph (scale bar=50 nm).
b) Overview and c) expansion of the rhodium signals in XPS spectra of
rhodium nanoparticles generated from [Rh(acac)(CO)2] in [Hex4N]Br
(Rh-2).
the value of the RhI precursor ([Rh(acac)(CO)2], 309.2 eV[14]
)
and of reduced Rh0 species. Such values are typical for small
rhodium nanoparticles and indicate a partial oxidation to
RhI.[14]
Material Rh-1 was insoluble in nonpolar media and could
be readily separated and recycled by extraction with pentane.
The turnover frequency value corrected for surface-exposed
rhodium centers (TOFsurface) dropped slightly after the first
run, but then remained stable at a value of approximately
14300 hÀ1 for at least four more runs. TEM micrographs and
XPS of a sample after three catalytic cycles indicated that the
primary particles remained largely unchanged (particle size:
(2.6 Æ 0.6) nm), although some partial agglomeration was
indicated by the TEM images and UV/Vis spectra. The other
materials were at least partially soluble in the reaction
mixtures, but could be readily precipitated and recovered in
active form by extraction of the products with scCO2.[15,16]
The promising results obtained with the simple test
reactions prompted us to evaluate the new materials as
catalysts for more challenging substrates. The sterically
encumbered (E)-2-(benzoylamino)-2-propenoic acid deriva-
tives 4a–e (Scheme 1) are precursors for nonnatural amino
acids and show interesting biological activities.[17] Selective
hydrogenation would be an interesting transformation to
The matrix-embedded nanoparticles Rh-1 to Rh-3 can be
used directly for catalytic hydrogenation reactions without
any further workup or purification. This was demonstrated by
first using cyclohexene (1) and benzene (2) as test substrates
under a standard set of reaction conditions (Table 2). The
catalyst materials were suspended or partly dissolved in the
neat substrate and hydrogen uptake was monitored online.
Turnover frequency (TOF) values were determined from the
slope within the first 20% of conversion. The reactions were
stopped after 30 minutes for 1 and after 120 minutes for 2, and
offline GC analysis confirmed quantitative conversion to 3 at
this stage. TOF values for the hydrogenation of 1 in the range
of 104 hÀ1 were observed with all materials under neat
conditions. Reaction rates were about two orders of magni-
tude lower for the reduction of the aromatic substrate 2.
These data are consistent with previously reported rates
obtained with rhodium nanoparticles in standard ionic
liquids.[4a]
1086 www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 1085 –1088

Angewandte
Chemie
Table 2: Representative catalytic results for benchmark hydrogenation
In contrast to the conventional catalysts, ammonium
reactions using matrix-embedded rhodium nanoparticles.[a]
halide stabilized rhodium nanoparticles were found to
combine good catalytic activities with significant differentia-
tion between the aromatic and olefinic moieties (Table 3). In
Table 3: Selective hydrogenation of (E)-2-(benzoylamino)-2-propenoic
acids 4a–e using Rh-1 as catalyst.[a]
Cat. Phase
behavior
TOFtotal TOFsurface
[hÀ1 [b] [hÀ1 [c]
Phase
TOFtotal TOFsurface
[hÀ1 [c]
]
]
behavior [hÀ1 [b]
]
]
Substrate T [8C] TOFtotal
TOFsurface
Selectivity[d]
[b]
[c]
Rh-1 immiscible 8800 26650
partially 35
miscible
106
17
[ꢁ10, hÀ1
]
[ꢁ10, hÀ1
]
5a–e
6a–e
7a–e
4a[e]
4b
4c
4d
4e
120
80
60
60
80
70
1550
50
160
10
210
4690
150
485
30
39
30
6
4
100
24
54
57
96
–
37
16
38
–
Rh-2 partially
miscible
6600 14050
partially
miscible
8
Rh-3 partially
miscible
35700 45800
partially 42
miscible
54
–
[a] Reaction conditions: T=408C, p(H2)=40 bar, neat; 1: Rh=1000:1,
2: Rh=100:1. [b] Total turnover frequency determined as mol substrate
per total amount of rhodium in matrix per hour, determined from
hydrogen uptake within the first 20% conversion; full conversion was
reached in all cases after appropriate reaction time. [c] Turnover
frequency corrected for surface-exposed rhodium centers by using the
dispersion data from Table 1.
[a] Reaction conditions: 4 (Rh=500:1), iPrOH, H2 (100 bar). [b] Total
turnover frequency determined as mol substrate per total amount of
rhodium in the matrix per minute at 20–35% conversion. [c] Turnover
frequency corrected for surface-exposed rhodium centers by using the
dispersion data from Table 1. [d] Determined by using GC for compa-
rable conversions around 60%. [e] Debromination at the remote
aromatic ring occurred under hydrogenation conditions with Rh-1 and
Rh/Al2O3.
increase the structural diversity of these substances
(Scheme 1). However, typical homogeneous rhodium cata-
lysts such as Wilkinsonꢀs catalyst do not show any hydro-
genation activity for these substrates at temperatures and
pressures below their decomposition (< 5% conversion after
3 days at 1208C and 100 bar). Use of commercial Rh/Al2O3 as
a prototypical heterogeneous catalyst led to a quantitative
conversion under these conditions, but both the olefinic
double bond and the aromatic ring in the benzoyl protecting
group were fully hydrogenated to give 7a–e as the main
products. Monitoring the reaction by using offline GC showed
that both groups are hydrogenated with similar rates from the
beginning of the reaction, and it was not possible to achieve
significant selectivities for products of either type 5 or 6.
general, the hydrogenation rates are somewhat smaller over
Rh-1 than over the commercial Rh/Al2O3 catalyst at com-
parable loading. For substrates 4a–b, all three possible
hydrogenation products were formed in a network of parallel
and consecutive hydrogenation processes. Remarkably, prod-
uct 6b was formed preferentially over 5b, which indicated
that hydrogenation of the aromatic moiety is faster than that
of the olefinic double bond in this case. This trend is even
more pronounced for the sterically more encumbered sub-
strates 4c–d. Product 6d was obtained with excellent selec-
tivity reaching up to 96% at 60% conversion. In contrast,
substrate 4e, which contains a basic NH group in the
heteroaromatic substituent, was hydrogenated exclusively at
the double bond to provide the benzoyl amino acid 5e in good
yields.
In summary, we
have
shown
that
CO2-induced melting
to form ionic liquids
provides
a solvent-
free approach to the
generation and appli-
cation of catalytically
active nanoparticles
entrapped and stabi-
lized in simple ammo-
nium halide salts. The
use of the scCO2
phase has several
advantages
during
the preparation pro-
cess: firstly, the CO2-
induced melting-point
depression
greatly
expands the range of
organic salts that can
Scheme 1. Product spectrum for hydrogenation of (E)-2-(benzoylamino)-2-propenoic acids 4a–e.
Angew. Chem. Int. Ed. 2009, 48, 1085 –1088
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1087

Communications
Int. Ed. 2005, 44, 7852 – 7872; h) B. Chaudret, Top. Organomet.
Chem. 2005, 16, 233 – 259.
[3] A. Roucoux, K. Philippot in Handbook of Homogeneous
Hydrogenation (Eds.: J. G. De Vries and C. Elsevier), Wiley-
VCH, Weinheim, 2007, pp. 217 – 256.
be employed as stabilizing matrices and secondly, organic by-
products from the molecular precursor can be removed
efficiently by extraction into supercritical fluids. In catalytic
applications, the obtained Rh nanoparticles showed a signifi-
cantly distinct behavior as compared with standard homoge-
neous or heterogeneous catalysts and were readily separated
and recycled from the reaction mixtures.
[4] a) J. Dupont, G. S. Fonseca, A. P. Umpierre, P. F. P. Fichtner,
S. R. Teixeira, J. Am. Chem. Soc. 2002, 124, 4228 – 4229; b) M.
Antonietti, D. Kuang, B. Smarsly, Y. Zhou, Angew. Chem. 2004,
116, 5096 – 5100; Angew. Chem. Int. Ed. 2004, 43, 4988 – 4992;
c) D. Zhao, Z. Fei, T. Geldbach, R. Scopelliti, P. J. Dyson, J. Am.
Chem. Soc. 2004, 126, 15876 – 15882; d) L. S. Ott, M. L. Cline, M.
Deetlefs, K. R. Seddon, R. G. Finke, J. Am. Chem. Soc. 2005,
127, 5758 – 5759; e) G. Clavel, J. Larionova, Y. Guari, C. Guerin,
Chem. Eur. J. 2006, 12, 3798 – 3804; f) Y. S. Chung, J. Y. Shin,
C. E. Song, S.-g. Lee, Chem. Commun. 2008, 942 – 944; g) G. A.
Somorjai, F. Tao, J. Y. Park, Top. Catal. 2008, 47, 1 – 14; h) R.
Umeda, A. Ryuhei, N. Hiroshi, T. Nakahodo, H. Fujihara, J. Am.
Chem. Soc. 2008, 130, 3240 – 3241; i) B. Lꢃger, A. Denicourt-
Nowicki, A. Roucoux, H. Olivier-Bourbigou, Adv. Synth. Catal.
2008, 350, 153 – 159.
[5] a) J. Kiwi, M. Grꢄtzel, J. Am. Chem. Soc. 1979, 101, 7214 – 7217;
b) H. Bꢂnnemann, W. Brijoux, R. Brinkmann, E. Dinjus, T.
Joussen, B. Korall, Angew. Chem. 1991, 103, 1344 – 1346; Angew.
Chem. Int. Ed. Engl. 1991, 30, 1312 – 1314; c) M. T. Reetz, W.
Helbig, J. Am. Chem. Soc. 1994, 116, 7401 – 7402.
[6] Ionic Liquids in Synthesis (Eds.: P. Wasserscheid and T. Welton),
Wiley-VCH, Weinheim, 2008.
[7] For earlier reports on moderate melting-point depressions in IL/
CO2 systems, see S. G. Kazarian, N. Sakellarios, C. M. Gordon,
Chem. Commun. 2002, 1314 – 1315.
[8] a) A. Scurto, W. Leitner, Chem. Commun. 2006, 3681 – 3683;
b) A. M. Scurto, E. Newton, R. R. Weikel, L. Draucker, J.
Hallett, C. L. Liotta, W. Leitner, C. A. Eckert, Ind. Eng. Chem.
Res. 2008, 47, 493 – 501.
[9] M. Solinas, A. Pfaltz P. G. Cozzi, W. Leitner, J. Am. Chem. Soc.
2004, 126, 16142 – 16147.
[10] M. T. Reetz, W. Helbig, S. A. Quaiser, U. Stimming, N. Breuer,
R. Vogel, Science 1995, 267, 367 – 369.
[11] E. F. Smith, I. J. Villar Garcia, D. Briggs, P. Licence, Chem.
Commun. 2005, 5633 – 5635.
[12] J. Rothe, J. Pollmann, R. Franke, J. Hormes, H. Bꢂnnemann, W.
Brijoux, K. Siepen, J. Richter, Fresenius J. Anal. Chem. 1996,
355, 372 – 374.
[13] Practical Surface Analysis by Auger and X-Ray Photoelectron
Spectroscopy (Eds.: D. Briggs and M. P. Seah), Wiley, Chi-
chester, 1983, p. 612.
Experimental Section
Safety Precaution: Working with compressed gases can be hazardous
and must be conducted only with suitable apparatus.
Synthesis of Rh-1: Tetrabutylammonium bromide (0.5 g,
1.54 mmol) and [Rh(acac)(CO)2] (12.5 mg, 4.86 ꢁ 10À2 mmol, corre-
sponding to 5 mg Rh) were placed in a 10 mL stainless-steel high-
pressure reactor. The reactor was evacuated three times, pressurized
to 40 bar with H2 followed by CO2 (ca. 7 g) using a compressor. The
autoclave was heated to 808C, at which temperature the total pressure
reached about 240 bar. A deeply colored liquid phase in equilibrium
with a colorless supercritical phase was formed under these con-
ditions. After stirring for 3 h, the reactor was cooled to ambient
temperature and carefully vented. The purification step was per-
formed at 808C with a constant flow of CO2 (180 bar) until no
acetylacetone was observed in the scCO2 stream. After cooling and
final depressurization, the dark-brown solid was used directly for
catalysis or transferred to a Schlenk tube for further storage, analysis,
and catalytic application. The materials Rh-2 and Rh-3 were obtained
by using the same procedure (see Table 1).
General procedures for catalytic hydrogenation: Hydrogenation
of cyclohexene (1) and benzene (2) (200 mL, ca. 2 mmol) were
performed using Rh-1 (20 mg, 1.94 mmol Rh) in an autoclave
heated on a metal block at 408C, allowing 30 min for temperature
equilibration before pressurization with H2 (40 bar). The products
were isolated by filtration or extraction with scCO2 and analyzed by
GC.
The hydrogenation of substrates 4a–e (0.1 mmol) was performed
using Rh-1 (20 mg, 1.94 mmol Rh) and iPrOH (1 mL) in an autoclave
heated on a metal block at temperatures between 60–1208C, allowing
30 min for equilibration before pressurization with H2 (100 bar). In
this case, the products were isolated by column chromatography on
silica gel using mixtures of EtOAc and iPrOH as eluents and analyzed
by 1H and 13C NMR spectroscopy.
Received: July 31, 2008
Published online: December 30, 2008
[14] J. Evans, B. E. Hayden, M. A. Newton, Surf. Sci. 2000, 462, 169 –
180.
Keywords: carbon dioxide · heterogeneous catalysis ·
[15] Of course, distillation would also allow the isolation of
cyclohexane in the present examples. However, extraction with
scCO2 from ILs is known to be a more general isolation method
allowing also the separation of low volatility products, see
a) L. A. Blanchard, D. Hancu, E. J. Beckmann, J. Brennecke,
Nature 1999, 399, 28 – 29; b) L. A. Blanchard, J. F. Brennecke,
Ind. Eng. Chem. Res. 2001, 40, 287 – 292.
[16] For a similar precipitation–recycling of PEG-stabilized Co-
nanoparticles in Pauson–Khand reactions with scCO2, see J.-L.
Muller, J. Klankermayer, W. Leitner, Chem. Commun. 2007,
1939 – 1941.
.
ionic liquids · nanostructures · supercritical fluids
[1] Nanoparticles. From Theory to Applications (Ed.: G. Schmid),
Wiley-VCH, Weinheim, 2004.
[2] For selected recent reviews, see: a) J. D. Aiken III, R. G. Finke,
J. Mol. Catal. A 1999, 145, 1 – 44; b) H. Bꢂnnemann, R. M.
Richards, Eur. J. Inorg. Chem. 2001, 2455 – 2480; c) A. Roucoux,
J. Schultz, H. Patin, Chem. Rev. 2002, 102, 3757 – 3778; d) M.
Moreno-Manas, R. Pleixats, Acc. Chem. Res. 2003, 36, 638 – 643;
e) M. T. Reetz, J. G. de Vries, Chem. Commun. 2004, 1559 –
1563; f) C. Burda, X. Chen, R. Narayanan, M. A. El-Sayed,
Chem. Rev. 2005, 105, 1025 – 1102; g) D. Astruc, F. Lu, J. R.
Aranzaes, Angew. Chem. 2005, 117, 8062 – 8083; Angew. Chem.
[17] a) U. Schmidt, A. Lieberknecht, J. Wild, Synthesis 1988, 159 –
1972; b) C. Bonauer, T. Walenzyk, B. Kꢂnig, Synthesis 2006, 1 –
20.
1088
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 1085 –1088