Hydrogenation of nitro-compounds over rhodium catalysts supported on poly[acrylic acid]/Al2O3 composites

Article abstract of DOI:10.1016/j.apcata.2014.10.029

In this report, poly[acrylic acid] gels containing Al₂O₃ were prepared by simultaneous free-radical polymerization and sol-gel chemistry using different amounts of 3-(trimethoxysilyl)propyl methacrylate (TMPM) as a compatibilizer. The hybrid materials were used as supports for a rhodium catalyst in the chemoselective hydrogenation of 3-substituted aromatic nitro-compounds. The supported rhodium catalyst was prepared by an ion-exchange process. In situ H₂ flux was used to produce active species of the catalysts. The resulting materials were characterized by infrared spectroscopy, thermogravimetric analysis, solid-state ²⁹Si and¹³ C NMR, X-ray diffraction, transmission/scanning electron microscopy, and X-ray photoelectron spectroscopy. All materials exhibited simultaneous interpenetrating hybrid network structures (SIHNs). The morphologies and physicochemical properties depended on the amount of TMPM used. The catalysts were found to be effective for the reduction of nitrobenzene in ethanol at room temperature and a hydrogen pressure of 20atm. The most active and selective catalyst was used in the hydrogenation of different 3-substituted aromatic nitro-compounds. The hydrogenation reactions displayed high conversion levels and promoted exclusive-NO₂ group reduction, resulting in the sole formation of the corresponding amino-compound, with the exception of 1,3-dinitrobenzene, in which over-hydrogenation was detected. The presence of electron-donating/electron-withdrawing substituents at the 3-position resulted in different rates of-NO₂group hydrogenation. This effect was quantified in terms of the Hammett relationship, in which the catalyst displayed a linear correlation between the substituent constant (σ_i) and the hydrogenation rate, with the exception of-OH, -NH₂, and-OCH₃ groups. One explanation for this behavior is a proposed support-substrate hydrogen bond interaction during the catalytic reaction.

Full text of DOI:10.1016/j.apcata.2014.10.029

Accepted Manuscript
Title: HYDROGENATION OF NITRO-COMPOUNDS
OVERRHODIUM CATALYSTS SUPPORTED ON
POLY[ACRYLIC ACID]/AL O COMPOSITES
2
3
Author: Cristian H. Campos Edward Rosenberg Jos e´ L.G.
Fierro Bruno F. Urbano Bernab e´ L. Rivas Cecilia C. Torres
Patricio A. Reyes
PII:
S0926-860X(14)00652-8
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2014.10.029
APCATA 15061
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
17-1-2014
15-10-2014
19-10-2014
Please cite this article as: <doi>http://dx.doi.org/10.1016/j.apcata.2014.10.029</doi>
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

HYDROGENATION OF NITRO-COMPOUNDS OVERRHODIUM CATALYSTS
SUPPORTED ON POLY[ACRYLIC ACID]/Al O COMPOSITES
2
3
a
a
b
c
Cristian H. Campos , Edward Rosenberg , José L.G. Fierro , Bruno F. Urbano , Bernabé L.
c,*
d
d,*
Rivas , Cecilia C. Torres , Patricio A. Reyes
a
Department of Chemistry, University of Montana, Missoula, MT, 59812, USA
b
Grupo de Energía y Química Sostenibles (EQS), Instituto de Catálisis y Petroleoquímica-
CSIC, C/Marie Curie 2, 28049 Madrid, Spain
c
Facultad de Ciencias Químicas -Departamento of Polímeros, Universidad de Concepción,
Edmundo Larenas 129,4070371 Concepción, Chile
d
Facultad de Ciencias Químicas - Departamento of Fisicoquímica, Universidad de
Concepción, Edmundo Larenas 129, 4070371 Concepción, Chile
1
Page 1 of 54

HIGHLIGHTS


 New well-defined hybrid composites from acrylic acid polymer with Al O
2
3
 Rhnanoparticles and di-rhodium carboxylate complex deposition on PAA/Al O3
2
supports.


 Metallic Rh using as catalysts in nitro-compounds hydrogenation.
 Upper 98% selectivity was achieved over all the catalysts for nitrobenzene
hydrogenation

 Reaction over the most selective catalystshowed a linear Hammett relationship and
positive reaction constant (ρ = 1.24).
ABSTRACT
2 3
In this report, poly[acrylic acid] gels containing Al O were prepared by simultaneous free-
radical polymerization and sol-gel chemistry using different amounts of 3-
trimethoxysilyl)propyl methacrylate (TMPM) as acompatibilizer. The hybrid
materialswereused assupports fora rhodium catalyst in the chemoselectivehydrogenation of
-substituted aromatic nitro-compounds.The supported rhodium catalyst was prepared by
an ion-exchange process. In situ H
flux was used to produce active species of the catalysts.
The resulting materials were characterized by infrared spectroscopy,thermogravimetric
(
3
2
29
13
analysis, solid-state Si and C NMR, X-ray diffraction, transmission/scanning electron
microscopy,and X-ray photoelectron spectroscopy. All materials exhibitedsimultaneous
interpenetrating hybrid network structures (SIHNs). The morphologies and
2
Page 2 of 54

physicochemical properties depended on the amount of TMPM used. The catalysts were
found to be effective for the reduction of nitrobenzene in ethanol at room temperature and a
hydrogen pressure of 20 atm. The most active and selective catalyst was used in the
hydrogenation of different 3-substituted aromatic nitro-compounds. The hydrogenation
2
reactions displayed high conversion levels and promoted exclusive -NO group reduction,
resulting in the sole formation of the corresponding amino-compound,with the exception
of1,3-dinitrobenzene, in which over-hydrogenation was detected. The presence of electron-
donating/electron-withdrawing substituents at the 3-position resulted in different rates of –
NO
which the catalyst displayed a linear correlation between the substituent constant (σ
the hydrogenation rate, with the exception of-OH, -NH , and -OCH groups. One
2
grouphydrogenation. This effect was quantified in terms of the Hammett relationship,in
i
) and
2
3
explanation for this behavior is a proposed support-substrate hydrogen bond interaction
during the catalytic reaction.
Keywords:Rhodium, hybrid materials, hydrogenation, metallic catalysts.
3
Page 3 of 54

1
. INTRODUCTION
Metal nanoparticles continue to attract interest in different research areas due to their
different physical and chemical properties when compared to bulk metals[1, 2]. The surface
reactivity of metal nanoparticles results in theirpotential forapplication in a variety of fields,
such as medicine [3], electronics and optics[4] and catalysis [5]. Particle growth or
aggregation leads to a loss of the nano-scale size andthe related important properties[2, 5].
Essential methods have been developed in response to this challenge, for example:
application of polymeric stabilizing agents[3, 6, 7], electrochemical deposition[4], template
growth techniques[8], and functionalized support methods [9-13].Hybrid polymeric-
inorganic materials are a good alternative to pure nanoparticles because these materials can
exhibit greater stabilities when compared to their pure counterparts[14, 15].Of particular
interest for our work are simultaneous interpenetrating hybrid networks (SIHN). In this
case, the material is produced by a sol-gel process in combination with free-radical
polymerization of the organic phase[16-18]. This approach allows in-situ formation of the
inorganic network and thus the homogeneous incorporation of polymers that normally
would beimmiscible. The use of organic-inorganic coupling agents makes it possible to
enhance the compatibility of the two components. The coupling agents generally have
hydrolysable groups that participate in the sol-gel chemistry and organic-functionalized
endsfor incorporation into the organic polymer [19-21]. Optimizing the mixing of the
networks in SIHN materials results in the most homogeneous functional materials.Such
systems could result in supported metal nanoparticles that would be completely insoluble in
aqueous or organic solvents for use as heterogeneous catalysts.
4
Page 4 of 54

The focus of this study was on rhodium metal nanoparticles because they are especially
active in a supported form for a variety of catalytic applications, including CO
hydrogenation [22], hydrogen production [23, 24] and chemoselective hydrogenations [9,
1
1, 12, 25]. Generating such nano-sized particles on a hybrid material is of scientific
interest, especially if it results in particles or dispersions with novel properties. In previous
reports involving colloidal Rh particles,polymers were used as stabilizers. A variety of
well-known polymers are commonly used as protecting agents, such as
poly[vinylpyrrolidone] (PVP)[6, 26], poly[vinyl alcohol] (PVA)[6]andpoly[acrylic acid]
(PAA)[6, 11]. However, their efficiency and behaviors as protecting agents in mixed-media
colloids are hitherto unknown. Therefore, it is important to study how hybrid
materialsperform as protecting agents for rhodium nanoparticle support.
Aromatic amines are important starting materials and intermediates for the manufacture of
a great variety of chemicals. They are generally synthesized by the chemical reduction of
nitroarenes[27, 28]. Although a variety of methods have been well documented for this
purpose, some of these protocols have drawbacks, such as long reaction times, the use of
toxic and expensive catalysts, the requirement of carcinogenic solvents, and limited
reusability of catalysts[29, 30]. The selective reduction of nitro-groups in organic
compounds containing other reducible functional groups is a very challenging task in
organic synthesis. In addition, the reduction of aromatic nitro-compounds is often
associated with the formation of side products, including hydroxylamines, hydrazines, and
azoarenes, because the reaction can stop at an intermediate stage[29, 30]. Therefore,easily
separable, chemoselective, and effective catalysts for the reduction of organic compounds
are highly desirable.
5
Page 5 of 54

In the present work, the hydrogenation of nitrobenzene (NB) using SIHN prepared from
PAA/Al with different amounts of 3-(trimethoxysilyl)propyl methacrylate (TMPM) as a
2
O
3
coupling agent was investigated. Furthermore, the catalytic reduction of a series of 3-
substituted aromatic nitro-compoundswas investigated and the catalytic data weresubjected
to Hammett treatment.
2
2
. EXPERIMENTAL SECTION
.1 Materials
TMPM(98%), ammonium persulfate (APS; >98%), anhydrous 2-butanol (>99.5%), acrylic
acid (AA; anhydrous, containing 180-200 ppm of the monomethyl ether of hydroquinone as
an inhibitor, 99%),rhodium(III) chloride hydrate (RhCl
3 2
·xH O; 99.98% trace metal basis),
1
-chloro-3-nitrobenzene (NB-Cl; 98%), 1,3-dinitrobenzene (NB-NO
2
; 97%), nitrobenzene
®
(
NB), and 1-methyl-3-nitrobenzene(NB-CN) were obtained from Sigma–Aldrich and
®
Fluka . Aluminum tri-sec-butoxide(TSBAl; 97%), acetylacetone(acac), absolute ethanol,1-
methyl-3-nitrobenzene (NB-CH
3
),1-methoxy-3-nitrobenzene (NB-OCH
3
)
and 3-
®
nitrophenol (NB-OH) were supplied by Merck .AA was distilled in the presence of
terbutylcatechol at reduced pressure prior to use. All other chemicals were used without
further purification.All air-sensitive reactions were performed in a polymerization flask
®
using an inert N
2
atmosphere. H
2
(99.99%) was supplied by AGA - Chile.
6
Page 6 of 54

2
.2 SIHN synthesis
TSBAl (0.03mol) was dissolved in anhydrous 2-butanol and heated at353K. Subsequently,
.04 mol of acac, 0.02 mol of AA, and 0.002 mol of APS were added to the reaction
mixture. TMPMwas added in different mole ratiosofTMPM/(TMPM+TSBAl):0, 0.5, 0.25
and 0.125. The reaction was carried out at 353 K under an N atmosphere with continuous
0
2
stirring. Once the gel point was reached, the solution was cooled to room temperature and
deionized water(10 mL) was addedto the gel and mixed mechanically. The mixture was left
for 8h at room temperature and later dried in a vacuum oven at 353K until a fine white
powder was obtained.All composites were washed 3 times with deionized water with
magnetic stirring for 4 h and dried at 373K. All materials obtained were labeled as
2 3
PAA/Al O ─TMPM(x), where x corresponds to the TMPM/(TMPM+TSBAl) mole ratio.
2
.3 Synthesis of Catalysts
2 3
The catalysts (1.0 g) were prepared at 0.5wt% Rh using PAA/Al O -TMPM(x) as the
support. The methodology is analogous to that reported by Torres et al.[31]. The
appropriate amount of support (dry hybrid material) was placed in a round-bottom flask
containing 50 mL of water. Prior to the introduction of the metal precursor, a stoichiometric
-1
-
quantity of NaOH 0.5 molL (OH /Rh mole ratio= 3) was also added to the support
suspension to obtain a final pH = 11. A solution containing the required amount of
3 2
RhCl xH O was added to obtain the desired metal loading,and the system was placed in a
thermoregulated bathfor 12 h at 318K withmagnetic stirring at 300 rpm. After the sorption
7
Page 7 of 54

of the Rh solution, the temperature was raised to reflux with constant stirring until the color
changed from orange to black. The obtained solid was filtered and washed with deionized
water until a constant conductivity was obtained. Finally, the catalysts were dried in a
vacuum oven at 373 K for 1 h and stored in a desiccator under an N
2
atmosphere prior to
the catalytic tests. These were labeled as 0.5%Rh-PAA/Al -TMPM(x).
2
O
3
2
.4 Characterization
The metal content was quantitatively monitoredin triplicate by digesting0.05 g of loaded
composite in 10 mL of concentrated nitric/hydrochloric (1:3) acid solution using
microwave-assisted digestion. After the reduction process, the metal loading was measured
by inductively coupled plasma mass spectrometry (ICP-MS) on a Perkin Elmer Elas 6000S
instrument.The morphological and structural characteristics of the composites and catalysts
were determined by XRD (RigakuD/max-2500 diffractometer with Cu K
α
radiation at 40
kV and 100 mA). Spectroscopic analyses were performed using FT-IR spectroscopy
1
3
29
(
Nicolet 400D in KBr matrix in the range of 4000–400/cm) and solid-state C and Si CP-
MAS NMR spectra were recorded at 100.6 MHz and 79.49 MHz, respectively, using a
Bruker AV 400 WB spectrometer.Thermogravimetric analysis was performed with a
thermal analyzer (TGA (Netzsch STA 409 PC/PG [STA]))over a temperature range of 300–
-1
8
2 2
73 K and a heating rate of 10 K min in an N atmosphere.N -BET surface areas and pore
volumes were determined on a Micromeritics ASAP 2010 apparatus at 77 K.The specific
surface area was calculated using the Brunauer–Emmett–Teller (BET) method. The pore
size distributions were obtained from the adsorption and desorption branch of the nitrogen
isotherms by the Barrett–Joyner–Halenda (BJH) method. Morphological and chemical
8
Page 8 of 54

analyses were performed via SEM-EDS (JEOL JSM-6380), the samples were scanned by a
secondary electron scanner, and surface content was analyzed by energy-dispersive X-ray
spectroscopy (EDS) using the standards for quantification: C (CaCO
3 2
), Si and O (SiO ), Al
(
Al ), S (FeS ). The transmission electron microscope was carried out in a Philips model
2
O
3
2
CM200 equipment with an energy dispersive analyzer and a digital camera coupled to a
high speed TVIPS (FastScan F-114 model)at 1024x1024 pixels and 12 bits. The samples
for analysis were prepared by dispersion in ethanol/H
carbon/Cu grid (300 mesh). Up to 300 individual metal particles were measured for each
catalyst and the surface area-weighted mean Rh diameter (d ) was calculated using:
2
O (1:1) and deposited on a porous
p
i i
Wheren is the number of particles of diameter d . The size limit for the detection of Rh
particles on samples was ca. 1 nm. Assuming a population of monodispersed (i.e., with the
same size) metal particles, the dispersion, D (%), is related to the mean particle size by a
simple formula:
where D corresponds to the ratio of the number of surface metal atoms to the total number
of metal atoms[32]. X-ray photoelectron spectra (XPS) were recorded using an Escalab 200
R spectrometer equipped with a hemispherical analyzer and using non-monochromatic Mg
Kα X-ray radiation (hυ =1253.6 eV). The spectra were fitto a combination of Gaussian-
Lorentzian lines of variable proportion. The C 1s core-level of adventitious carbon at a
binding energy (BE) of 284.8 eV was used as an internal standard.
9
Page 9 of 54

2
.5 Catalytic activity
Catalytic activity assays of the hydrogenation of 3-substituted aromatic nitro-compounds
-
were performed in a stainless steel, Parr-type batch reactor at a concentration of 0.02 molL
1
of substrate using absolute ethanol as a solvent and stirring at 800 rpm under 20 bar H
2
pressure at 298K. All further experiments were carried out in the absence of external mass
transfer limitations[33]. The tested catalysts were in the form of a fine powder (>30
μm).For the substrate/Rh molar ratio studies, the concentration of substrate was kept
constant and the mass of the catalyst was varied.Small differences in initial velocity were
observed in the range of catalyst mass studied (0.01-0.07g) because the reaction rate was
proportional to the catalyst mass. In these studies, 0.050g of catalyst was used for kinetic
measurements. A non-invasive liquid sampling system with an in-line filter allowed the
3
controlled removal of aliquots (≤ 0.5 cm ) from the reactor. All the catalytic experiments
were made three times (with fresh catalyst every time) and the average values were
reported for kinetic data. The samples of reactions were analyzed in a GC-MS Shimadzu
(GCMS-QP5050) with helium as the carrier gas was used to analyze the reactants and
products. The turnover frequency (TOF) of the catalysts was calculated using:
1
0
Page 10 of 54

3
3
. RESULTS AND DISCUSSION
.1 Support Synthesis and Characterization
A washing procedure was performed to remove unreacted monomer and impurities in the
material. ThePAA/Al -TMPM(0) composite resulted in the lowest yield (56%); this
result can be attributed to the absence of the coupling agent. For the
compositesPAA/Al -TMPM(x) with x > 0, the yieldincreased as the TMPM content
increased (x: 0.125, 58%; x: 2.5, 61% and x: 0.5, 75%) because it acts as a crosslinking
agent between Al and the PAA chains.
Figure 1 displays FT-IR spectra of PAA/Al
bandsfor the silicon compounds areobservedat 1113 cm (C–Si–O, st). The absorption peak
2
O
3
2
O
3
2
O
3
2
3
O -TMPM(x). The expected absorption
-1
-1
of the Al–O stretching vibration often appears in the range of 1000–1200cm .However, for
the present system,this peak could not be resolved due to its overlap with the absorption
-1
peak of the Si–O–Si stretching vibration. The absorption band at3436cm is associated with
-1
the O–H bond vibration, while the very strong band at 1532 cm is attributed to the C=O
-1
carboxylic acid stretch of PAA. Theband at 1720 cm corresponds to theC=O ester
vibration of TMPM.
2 3
Figure 2 presents theTGAand DTGA curves ofPAA/Al O -TMPM(x). TGA (see Fig. 2.a)
exhibited two stages of degradation. The first step occurred at approximately 350Kand
corresponds to the evolution of water. The second stepappeared at higher temperatures in
the range of 620–720 K,andcan be attributed to degradation of the polymer, with
degradation ending at 780K.Additionally, it was observed that the mass loss depended on
1
1
Page 11 of 54

the TMPM content; as the TMPM content increased, the mass loss increased. In this
case,the thermal decomposition of the polymer occurred at higher temperatures, which
2 3
demonstrates that PAA/Al O -TMPM(0) composites have higher thermal stabilities
andslower degradation rates than PAA in all cases. However, for the PAA/Al
2
3
O -
TMPM(x)composites with x > 0,significant changes in the TGA and DTGA (see figure 4.b)
wereobserved. Twodifferent temperatures in the mass loss steps were observed. The first
step, appearing at 654 K, corresponds to chemical transformations of the Si compounds. In
x y
similar systems, TMPM has been demonstrated to generateSi C O4-y species upon
thermolysis[34, 35]. For this reason, these results should notbe used to express the polymer
content in the composite. The second weight loss (at approximately 690–720 K), which
occurredto different extents, is related to degradation of the polymer (see Table 1).
Figure 3displays scanning electron microscopy (SEM) micrographs of PAA/Al
2 3
O -
TMPM(x) composite samples. The morphology and size of AA/Al -TMPM(x)
2
O
3
composite are described by SEM, and the results are shown in Fig. 3(A) – (D). The hybrid
particles exhibit a layered structure with a smooth surface. The SEM images of
PAA/Al2O3-TMPM(0) show a regular distribution of particles size. As the coupling agent
content increases an increment on micro-particle size was detected. This effect can be
explained for the uniformity provided by the TMPM. The polymers chains and metal oxide
network shows a higher interaction in the hybrid materials in presence of coupling agent.
2 3
The PAA/Al O -TMPM(x) particles obtained at higher nominal contents of TMPM are out
of shape, with a regular-uniform higher size (Fig. 5(D)), in agreement with the N
and SBET characterization.
2
isotherm
1
2
Page 12 of 54

2 3
The composition of PAA/Al O -TMPM(x) composites was quantitatively determined by
EDS analysis (Fig. 5(E)-(F)) is shown in Table 2. The result clearly confirms the presence
of Al and C on the surface of all the hybrids materials. Only for the PAA/Al
with x > 0, Si atom was detected.
2 3
O -TMPM(x)
Moreover, the weight percentages of C and Al were approximately 64% and 1-2%,
respectively. These indicated that Al O oxide phase was covered completely with PAA
2
3
and the -COOH pending groups are on the surface of hybrids materials. On the other hand,
Si was detected on PAA/Al O -TMPM(x) with x > 0, attributed to the TMPM in the
2
3
polymer backbone and provided bridges points during the sol-gel synthesis, improving the
metal oxide coating. The weight percentage of Si was lower than the nominal value. This
result indicates that Si atoms are strongly associated on the metal oxide network.
All the characterizations indicate that the composites are SIHN materials because the
incorporation of TMPM promotes the formation of laminar sites, which are characteristic of
this type of resin materials andare consistent with the decreased surface areas, as detected
by the SBET results (vide infra).
3
.2 Catalyst Synthesis and Characterization
The metal loadings shown in Table 3 are in the range of 0.45- 0.50 wt %. In general, the
composites displayed a high adsorbed affinity for Rh(III) during the catalyst preparation
process.
The FT-IR spectra, TGA, XRD,andN
2
adsorption-desorption isotherms of the supports and
catalysts did not show significant differences, and the crystalline nature of the catalystswas
1
3
Page 13 of 54

2 3
studied by XRD (see Fig. 4). The powder X-ray diffraction patterns of the PAA/Al O -
TMPM(x)catalysts exhibited characteristic spectral profiles of amorphous aluminates with
two amorphous halos present in the 2θ = 23 and 40° [17, 36]. This typical halo results from
the dispersion of the angles and bond distances between the basic structural units (silicates
and aluminates), which destroys the structural periodicity and produces a non-crystalline
material. Slight changes in the diffraction patterns were observed with increasing TMPM
content around 2θ = 10°. This result confirms the assumption that PAA hybridization
occurs on the solid surface without changing the structural form of the Al
PAA/Al -TMPM(0)), but that TMPM insertion modifies the inorganic network during
the syntheses. The characteristic Rh diffraction lines in the region of 2θ ~ 39.4° (plane 1, 1,
) and 45.6° (plane 2, 0, 0) were not detected.
2 3
O (composite
2
O
3
1
2 2 3
The N adsorption–desorption isotherms of 0.5%Rh-PAA/Al O -TMPM(x) composite
samples are shown in figure 5. The isotherms are similar to Type IV isotherms with H1-
type hysteresis loops at high relative pressures, according to the IUPAC classification,
which is characteristic of amorphous and macro- and mesoporous materials[16]. The
specific surface areas and pore sizes were calculated using the Brunauer–Emmett–Teller
(BET) and Barrett–Joyner–Halenda (BJH) methods, respectively.The structural data for all
materials are summarized in Table 3. It is clear that PAA/Al -TMPM(0) has the highest
2
O
3
2
-1
3 -1
BET surface area (140 m g ), pore volume (0.37 cm g ) and pore size (100 nm), indicative
that it was a host to PAA before washing. The values of these properties were lower for the
2 3
composites with x > 0 (see Table 3) than for composite PAA/Al O -TMPM(0); the TMPM
resulted in more highly cross-linked materials (polymeric-oxide network), promoting
smaller cavities and decreased SBET values.
1
4
Page 14 of 54

Metal cation sorption on chelating sorbents is a pHdependent process. The hydronium
concentration in water can change the structural properties of ligands and the speciation of
metal ions in aqueous solution. All of the synthesized catalysts were prepared by Rh(III)
sorption. It is important to consider that the -COOH group is a weak acid; at pH 11
thesegroups are totally ionized [37, 38].During the synthesis of rhodium nanoparticles, H
2
is
+
-
oxidized to H , resulting inan acidic medium that can alter the –COOH/–COO
equilibrium.The stoichiometric addition of NaOH to the synthesis system allowed us to
obtain the maximum ion-exchange/complexation process for rhodium retention and a
neutral medium at the end of the nanoparticle synthesis process (approx. pH = 7.0). The
metal particle sizes shown in Table 3were determined by TEM and can be observed in the
2 3
micrograph shown in Figure 5. The catalyst prepared on PAA/Al O -TMPM(0) displayed
large particle sizes, reaching approximately 4.6 nm, and a wide distribution of particle size
values. The average metal particle sizesof samples prepared with TMPM displayed an
inverse relationship with TMPM content. The polymeric network provided stability to the
Rh clusters formed during the metal reduction, thereby allowing the formation of smaller-
sized agglomerates. From these catalyst preparation results it is clear that a more cross-
linked support isoptimal.
The chemical changes in the rhodium particles were also investigated by XPS. Table 4
gives the Rh3d5/2 BEs for all of the catalysts under study, with the corresponding
contribution to the overall signal in parentheses. It was interesting to observe the
appearance of two peaks at 308.4 eV and 310.2 eV in allof the catalyst spectra. Rh 3d5/2zero
valentstates appearat 307.2eVin the 1.0%Rh/Al
2 3
O catalyst [39]. It is well known that very
small metallic particles, a strong interaction of metal particles with their supports, orthe
1
5
Page 15 of 54

-
growth of metal nanoparticles in a –CCO environment may produce a slight shift towards
higher BEs.Based on thetypical binding energies for Rh(0), our values are slightly shifted
-
(ΔBE = +1.2 eV). In the case where the Rh nanoparticles were grown in a –COO
environment, the resulting complex could be covered with Rh surface layers, exhibiting a
different periodicity when compared to the Rh lattice, which agrees with similar studies
reported for Rh/MgO by Jin-Phillippet al.[40]. The binding energies of Rh 3d5/2 spectral
lines exhibit a single broad signal, which we suggest is due to an overlap of the two spectral
lines (surface layers and lattice) becausethis Rh spectral line is very sensitive to the
environment of the metal [11, 39-41].
A comparison of the integral intensities of these components further suggests that
Rh(0)species constitute approximately 34 – 60% of the total Rh in the catalyst and that the
contribution increases with increasing TMPM in the supports. The appearance of the
additional band at 310.2 eVcorresponds to Rh(II) carboxylate complexesformed during the
reduction process[42-45]. Overall, these XPS results clearly demonstrate the incomplete
reduction of Rh species onthe hybrid nanocomposite, which is consistent with previous
literature reports by Karakhanov et al.[11]. The main reason for incomplete Rh reduction is
most likely the high stability of the Rh – carboxylate ligand chelates formed during
complexation.
With higher TMPM contents, the band attributed toRh(0) becomesvery intense, mainly due
to the weakinteraction of the nanoparticles with the oxide surface of the inorganic
2 3
network.0.5%-PAA/Al O -TMPM(x) with x:0 displays a more intense Rh(II) carboxylate
complex contribution (66%), whichcan be attributed to the strong interactions between the
1
6
Page 16 of 54

Al
2 3
O and PAA networks. In the presence of oxygen and/or oxygen donors, these types of
di-rhodium species can form oxygen-bridged carboxylate complexes with the general
n+
n+
formulas [Rh
3
(μ
3
2 6 3 2 2 2 6 2
-O)(μ-O CR) (L) ] , [Rh (μ-O)(μ-O CR) ·(L) ] , and [M (μ-O)(μ-
n+
O
2
CR) (O CR) (L) ] [42]. These metal oxide – Rh complexes are the result of: (1)
2
2
2 2
reactions resulting in changes in the oxidation state of the metal atom only[46]; (2)
reactions involving the addition of extra electrons to terminal ligands L[42, 47]; and/or
reactions resulting in changes in the oxidation state of the metal and the rearrangement of
its core [47]. In our case, an increase in the (TMPM/(TMPM+TSBAl)) surface ratio in the
syntheses results in a change on the SIHN surface due to the coverage of the support by Si-
O-Al species. The binding energies for Rhd5/2 at 310 eV only exhibited a slight
displacement to minor energies (ΔBE = 0.3eV).This behavior can be attributed to changes
in terminal ligand electronic arrangements.
The binding energies assigned to each C 1s peaks are in agreement with accepted values for
these functionalities [48]. The peak representative of the carbon (i.e., 284.8 eV) at the α-
position of the carboxylic acid group is assumed to include contributions from the C single
bond of the PAA backbone. The binding energies at 286.7and 288.5 eV correspond to the
contribution of C from the carboxylic acid groups on the surface of the support and the
carboxylic acid ligands in the Rh – complex, respectively [48, 49]. The contribution of C1s
for the carboxylic acid groups decreases the C=OPAA/C-OPAAwithincreasing TMPM in the
support structure, in agreement with the change in Rhd5/2 binding energies (see table
4
).Finally, Si 2p binding energies at 101.3 eV for catalysts 0.5%-PAA/Al
2 3
O -TMPM(x)
with x ˃ 0.125 were detected, which is typical of organosilicon compounds. The
1
7
Page 17 of 54

preferential insertion of TMPM intothe bulk structure is confirmed by changes in the
intensity of this line when we used lower (TMPM/(TMPM+TSBAl)) nominal mole ratios.
1
3
The support structure and the Rh – PAA interactions were confirmed by solid-state C and
2
9
Si CP-MAS NMR of 0.5Rh%-PAA/Al
). Figure 6(A) displays the NMR spectra of PAA/Al
PAA shows signals at 178.7 ppm (C=O), 42.3 ppm (CH-), and 35 ppm (-CH
2
O
3
-TMPM(0.0) after metal reduction (see figure
-TMPM(x) with x: 0 to 0.50. The
-),which has
6
2 3
O
2
also been reported by da Silva et al.[50]. The hybrid materials exhibit similar signals, but
their intensities and shapes change with increasing TMPM in the structure of the materials.
PAA/Al
2
O
3
-TMPM(0) shows three resonances for the –COOH groups at 183.3, 190.8 and
-
1
91.5 ppm.The first corresponds to the COOH/COO pendant groups, and the other two are
2 2 4 2
attributed to[Rh (0 CR) ]L (characteristic doublet of this type of complex), in agreement
with similar results reported by Allen et al.[41]forrhodium catalysts prepared from
carboxylate-modified silicapolyamine composites. Convincing evidence that the Rh forms a
carboxylate complex with carboxylate-modified SIHN supports comes from the observation
2
13
103
of J C– Rh (70 MHz) doublets with values in agreement with the typical Rh (μ-O
2 2
CR)
or (μ-O CR)
2
4
complexes, as reported by Varshavskiiet al. [45]. We detected the same
signals and the same coupling constant values in all composites, but the intensity decreased
with increasing TMPM in the support structure. These results agree with the two types of
Rh species detected by XPS and confirm the coordination state of the Rh complex in the
catalysts. The signals at ̴ 100 ppm correspond to the quaternary carbon atoms formed by
cross-links between the linear chains of PAA formed during the free-radical polymerization
process. The AA/APS mole ratio was 10; the excess of initiator and the low solubility in 2-
butanol provides excellent conditions for reactivating the polymerization reaction during
1
8
Page 18 of 54

the sol-gel process[51, 52]. The polymer backbone contains CH- and CH
2
- signals at 43 and
-TMPM(x)
2
8 ppm, which are typical chemical shifts for PAA[41, 50, 51, 53]. PAA/Al O
2 3
with x = 0.125 - 0.50 displays the same type of C signals,along with new resonances at 68
and 9.0 ppm. These resonances correspond to the insertion of TMPM into the backbone of
the polymer (methacrylate and propyl groups, respectively). The signals corresponding to
the C=O groups displaying a different intensity and a slight shift relative to PAA/Al
TMPM(0) because the increased TMPM on the surface forms Si-O-Al bonds, resulting in
steric blockage of the carboxylate – Al sites.The signal at 102 ppm displaysa decrease
2 3
O -
2
O
3
with TMPM insertion because the cross-links between polymeric phases and the inorganic
network decreases the mobility of the chains during the sol-gel process and results in more
linear PAA chains. This result agrees with the change in the intensity of the signals at 43
and 28 ppm for the PAA/Al
Figure 6(B) displays the solid-state Si CP-MAS NMR data of the PAA/Al
hybrids. In all supports, no residual unconverted TMPM was detected, as no peaksbetween
and −45 ppm were observed (TMPM should have a signal at −43.1 ppm) [54]. For the
PAA/Al -TMPM(x) hybrid, three resonances were detected between -47 and -68 ppm,
which correspond to the T , T and T anchors. The Al
2 3
O -TMPM(x) supports with x ˃ 0.125.
2
9
2
3
O -TMPM(x)
0
2
O
3
1
2
3
2
O
3
surface signals increase in
1
2 3
intensity withincreasing TMPM in the composites. For PAA/Al O -TMPM(0.125), only T
2
and T signals were detected at -47.8 and -59.4 ppm, respectively. The PAA/Al
2
O
3
-
TMPM(x) composites with x = 0.25 and 0.50exhibited three types of anchors. This is
attributed to the increase in the (TMPM/TMPM+TSBAl) nominal mole ratio in the
synthesis, which provides more sites for the formation of Si-O-Al bonds during the
hydrolysis-condensation process.
1
9
Page 19 of 54

3
.3 Nitrobenzene hydrogenation
Figure 7displays the activity of all catalysts used for the hydrogenation of NB. Increasing
the TMPM loading in the framework of the catalyst surface improved the activity. Pseudo-
g
first order kinetics (k ) with respect to NB was found in all cases. The turnover frequencies
(TOF) are given in Table 5.The possibility that the reaction reached equilibrium during the
measurements can be excluded, and all results shown are the maximum conversions before
reaching equilibrium. The TOF were calculated based on conversion of NB to An. All
catalysts were selective for aniline (An) and gave yields above 98%, as shown in Table 5.
2 3
Products from aromatic ring hydrogenation were not detected. 0.5%Rh-PAA/Al O -
TMPM(0.0) exhibited the lowest conversion of NB. These results confirm the synergistic
effect of using TMPM in the preparation of the rhodium nanoparticle supports. The catalyst
prepared without TMPM contained irregular Rh nanoparticles with respect to the other
supports. It might be expected thatthiswould lead to the preferential incorporation of metal
inside the tunnel framework, which would improve the activity of the catalyst. However,
the catalyst lacking TMPM contained a higher percentage of Rh carboxylate complexes, as
detected by XPS characterization (see table 4).This type of complex does not possess
catalytic activity in hydrogenation reactions [53]. Previous studies have demonstrated that
this reaction is a structure-sensitive reaction [31, 55]. In fact, the ability of the support
surface to disperse the metal nanoparticles results in a smaller metal particle size and
improves the activity and yield in NB hydrogenations[31]. In this case, the incorporation of
TMPM improves the reduction of the di-rhodium complex species remaining from the
reductionprocess. Table5displays some of the kinetic parameters for NB hydrogenation
andyields of An.Using Rh nanoparticles in the SIHN as a catalyst and ethanol as the solvent
resulted in the production ofphenyl-hydroxylamine(2 when X = H in scheme 1) and
2
0
Page 20 of 54

nitrosobenzene (3 when X = H in scheme 1)as the by-products (approximately 1-2%) (see
scheme 1 proposed by Kratky et al.[56]).The 0.5%Rh-PAA/Al -TMPM(0) catalyst
resulted in aslightly loweryieldof An, 98.2%,after 4 h of reaction. However, both the
reactivity and selectivity was higher than for the other catalysts. The 0.5%Rh-PAA/Al
2
O
3
2
3
O -
TMPM(0.125) catalyst (containing an overall metal component of 38% and 62% di-
rhodium complex on the surface) resulted in a greater conversion than the 0.5%Rh-
2 3
PAA/Al O -TMPM(0.0) catalystafter 7h of hydrogenation,as well an increase inselectivity
(see Table 6).Thesehydrogenation yields were higherwhen comparedto other reported metal
noble catalysts[57-60]. It has been reported that the rate-determining step of the
hydrogenation of NB derivatives is a nucleophilic attack of hydride ion, produced by
dissociative adsorption of a H
NO group [33, 55, 57]. The PAA on the surface of the support promotes the constant
polarizationof Rh nanoparticles. These active sites increase the electrophilic character of
the -NO group during the hydrogenation and decrease byproduct production. Maximum
activity and yield were achieved using 0.5%Rh-PAA/Al -TMPM(0.50),which was
2
molecule on the metal surface, on the nitrogen atom of the -
2
2
2
O
3
chosen for further study of the hydrogenation of 3-substituted aromatic nitro-compounds
and determination of the Hammett relationship for this reaction.
3
.4 Hydrogenation of Aromatic Nitro-compounds: Hammett Relationship
The 0.5%Rh-PAA/Al -TMPM(0.50)catalyst was tested in the hydrogenation reactions of
2
O
3
various 3-substituted nitrobenzene derivatives to their corresponding 3-substituted anilines.
Due to thedistinctive electrochemical and steric structural features of the tested 3-
substituted anilines, the results varied as shown in Table 6. In the previous section, we
noted that aniline was the only product in the hydrogenation of NBwiththis catalyst. In the
2
1
Page 21 of 54

hydrogenation of substituted 3-nitroarenes bearing –NH
and –NO in the 3-position, the catalyst wasalso 100% selective in generating the
corresponding amine product. In all the cases, intermediaries or other product of
hydrogenation (-CN, or in any case phenyl ring) or hydrotreatment (-Cl, -OH, -OCH
elimination) were not detected. Only in the case of NB-NO , 1,3-diamine benzene was
2 3 3
, –OH, –OCH , –CH , –Cl, –CN
2
3
2
detected at higher conversion levels.The higher activity observed for NB hydrogenation
also extended to all of the substituted 3-nitrobenzenes,where the following activity
sequence was established: NB-NO
OH>NB-NH . To assess the dependence of rate on the nature of the 3-substituent, we
applied the Hammett correlation. The Hammett equation evaluates the effect of a (3- or 4-
with -NO group as in nitrobenzene) substituent on reaction kinetics and can be used to
predict rate and equilibrium constants without prior experimental determination [33, 55,
1]. In this approach, the pseudo-first order constants for the substituted reactants (kg,i) can
2 3 3
>NB-CN > NB-Cl>NB>NB-CH >NB-OCH >NB-
2
2
6
be related to that obtained for the non-substituted (or reference) benzene derivative (kg,0
according to:
)
wherethe ρ term (reaction constant) is an estimation of the charge development during the
course of the reaction and provides a measurement of the susceptibility of the system to
i
substituent electronic effects[61, 62]. The σ factor is an empirical parameter (values used
in this study were taken from Campos et al. [55] and Jaffé[61]) that is dependent on the
substituent electron donor/acceptor character. This equation was initially conceived for
homogeneous systems, where good linear correlations have been established for the
2
2
Page 22 of 54

hydrogenation of aromatic nitro-compounds [63]. The involvement of adsorption
phenomena, the heterogeneous distribution and nature of active sites, and the
highercontribution of steric effects has limited the applicability of the Hammett expression
to heterogeneous catalytic systems[61]. Nevertheless, there have been some studies where
it has been successfully employed in the liquid-phase hydrogenation treatment of nitro-
aromatics over solid catalysts[33, 55, 57]. In agreement with a previously reported study of
the hydrogenation of 3-nitroarenes [55], at low loadings (≤ 1.0 w/w%), noble metals such
as Rhshowed higher selectivity for the corresponding 3-substituted anilines, with the
exceptionof NB-NO
2
. This substrate displayed only -NO
2
group hydrogenation with no
groupsto -NH groups
detection ofintermediates, but only a 40% of conversion of both -NO
was detected.
2
2
The fit of the experimental rate data to the Hammett relationship is presented in figure 8.
For the NB-OCH , NB-OH and NB-NH substrates,anon-linear Hammett relation was
3
2
detected. All other substratesdisplayed linear behavior and generated positive reaction
constants (ρ), consistent with a nucleophilic reaction mechanism. The ρ value generated
(1.24) indicates a greater dependence of rate on the electronic character of the substituent
and is diagnostic of a higher charge development in the reactant → intermediate step,
which isattributed to electrons flowing towards the aromatic ring in the rate-determining
step[33, 55]. The slope of the linear fit gives a positive value for ρ, with chemisorbed H
2
acting as nucleophilic agent that attacks the activated –NO group, resulting in the
2
formation of a negatively charged intermediate. This suggests an analogous reaction
mechanism for the formation of amines in these heterogeneous catalyst systems.
In the cases of theNB-OCH
3
2
, NB-OH and NB-NH substrates, the corresponding anilines
were obtained with the0.5%Rh-PAA/Al
2
O
3
-TMPM(0.50) catalyst, butdifferent catalytic
2
3
Page 23 of 54

performance was observed due to their functional group structures. Low kinetic pseudo-
first order constants were found for these three substrates with respect to the general
trend(see Table 6). The values calculated show a somewhat linear trend and similar
behavior with respect to the other substrates. For example,k(electron-withdrawing groups) ˃ k(NB)
(electron-donating groups), but the linear relationships did not cross the NB graph (zero point).
˃
k
We assume that substrates have different electronic contributionsfrom the3-position in the
negatively charged intermediate and that this could be one of the explanations for the low
conversions we observed. Thecarboxylate groups on the surface could be forming H-
bondinginteractions with the substrates. The reaction rates decrease because this H-
bondingenhancestheinductivedonatingeffect by polarization of the σ-bonds, as shown in
scheme 2. The negatively charged intermediate is less stable for these type of substituents
in the 3-position. NB-OH and NB-NH
2
can act as H donors or acceptors but NB-OCH
3
can
only act as an acceptor. As reported in other Hammett relation studies, NB-OCH
3
in the 3-
positiondisplays electron-withdrawing character with respect to the other substituents in the
aromatic ring[62, 64], but in our case this substrate displays an electron-donating effect
becausekNB ˃ kNB-OCH3
.
All of these results are indicative of a partially charged transition state with a ρ value
comparable to those reported for liquid-phase heterogeneous hydrogenations (Ir/ZrO
2
,
0
.639 [55], β-Mo
2
N 0.4 [33]and Pt/SiO
2
–AlPO
4
0.1–2.0 [57]). The pseudo-first order
constant displacement detected for NB-OCH
3
2
, NB-OH and NB-NH suggests a common
reaction mechanism for the formation of amines inthese heterogeneous catalyst systems.
4
. CONCLUSIONS
2
4
Page 24 of 54

2 3
A novel polymer–inorganic hybrid material, PAA/Al O
-TMPM, was prepared usingin situ
radical polymerization and a sol-gel process with acrylic acid and organometallic oxide
precursors. These new solid-based compositesare a SIHN and changes its properties when
different TMPM/(TMPM+TSBAl)molar ratios are used. IncreasingTMPM in the composite
results in an increase in thermal stability and a decrease in surface area. The Rh catalyst
was synthesized by an ionic exchange process and in situ H
RhCl as the inorganic precursor to produce catalysts with a mixture of Rh nanoparticles
and Rh(μ-O CR) complexes (where x: 2 or 4). The catalystsused at 1 wt %, exhibited
different residual Rh(μ-O CR) contents that decreasedwith increasing TMPM in the
support. All of the prepared systems were tested as catalysts in the liquid-phase
hydrogenation of NB and were 100% selective with respect to –NO group reduction. The
most active and selective catalyst was 0.5%Rh-PAA/Al -TMPM(0.50), which was used
as anoptimumcatalyst for the hydrogenation of a range of 3-substituted (–H, –OH, –OCH
CH , –Cl, –CN, and –NO ) nitrobenzenes. The reaction proceeds via a nucleophilic
reduction (to 398 K) using
2
3
2
x
2
x
2
2
O
3
3
,
–
3
2
mechanism where the presence of electron-withdrawing ring substituents served to increase
the rate, as demonstrated by the linear Hammett relationship and positive reaction constant
(ρ = 1.24). The obtainedresults demonstrate the potential of polymer hybrid composite-
supported nanoparticles to promote the clean production of amino-compounds by catalytic
nitrobenzene reduction.
2
5
Page 25 of 54

Acknowledgments
The authors thank FONDECYT (Grant No 1110079), FONDECYT (Grant No 1100259),
REDOC (MINEDUC Project UCO1202 at U. de Concepción). B. F. Urbano thanks to
FONDECYT Initiation (Grant No 11121291) and CIPA CONICYT Regional Project
R08C1002. C. H. Campos thanks CONICYT and BECAS CHILE for a postdoctoral
fellowship. We also acknowledge the generous support of the National Science Foundation
(USA, ER, CHE1049569)
2
6
Page 26 of 54

REFERENCES
[1]
C. Burda, X. Chen, R. Narayanan, M.A. El-Sayed, Chem. Rev. 105 (2005) 1025-
1
102.
[
2]
3]
P. Serp, P. Kalck, Chem Rev 102 (2002) 3085-3128.
[
H. Valle, B.L. Rivas, M.R. Aguilar, J.S. Román, J. Appl. Polym. Sci. 129 (2013) 537-
5
48.
[4]
K.S. Lokesh, Y. Shivaraj, B.P. Dayananda, S. Chandra, Bioelectrochemistry 75
(
2009) 104-109.
5] H. Bönnemann, G. Braun, W. Brijoux, R. Brinkmann, A.S. Tilling, K. Seevogel, K.
Siepen, J OrganometChem 520 (1996).
[
[
6]
7]
J.E. Jaine, M.R. Mucalo, J. Colloid Interf. Sci. 375 (2012) 12-22.
[
A.M. Lyakhovich, S.S. Mikhailova, S.I. Pomogailo, G.I. Dzhardimalieva, A.D.
Pomogailo, Macromol. Symp. 204 (2003) 251-256.
[
8]
9]
T.-H. Tran, T.-D. Nguyen, Colloids Surf., B 88 (2011) 1-22.
[
M.J. Jacinto, P.K. Kiyohara, S.H. Masunaga, R.F. Jardim, L.M. Rossi, Appl. Catal., A
3
38 (2008) 52-57.
[10] W. Liu, X. Yang, W. Huang, J. Colloid Interf. Sci. 304 (2006) 160-165.
[11] A.L. Maksimov, S.N. Kuklin, Y.S. Kardasheva, E.A. Karakhanov, Pet. Chem. 53
(
2013) 157-163.
12] I. Nakamula, Y. Yamanoi, T. Imaoka, K. Yamamoto, H. Nishihara, Angew. Chem.,
Int. Ed. 50 (2011) 5830-5833.
[
2
7
Page 27 of 54

[13] Z. Nazarpoor, K. Khivantsev, E. Kyriakidou, C. Kubicki, S. Ma, P.T. Fanson, O.S.
Alexeev, M.D. Amiridis, J. Colloid Interf. Sci. 398 (2013) 22-32.
[
[
[
[
[
14] J. Pyun, K. Matyjaszewski, Chem. Mater. 13 (2001) 3436-3448.
15] B.A. Rozenberg, R. Tenneb, Prog. Polym. Sci. 33 (2008) 40–112.
16] C.H. Campos, B.F. Urbano, B.L. Rivas, Composites, Part B (2013).
17] R.J. Kalbasi, M. Kolahdoozan, M. Rezaei, J. Ind. Eng. Chem. 18 (2012) 909-918.
18] R.J. Kalbasi, M. Kolahdoozan, S.M. Vanani, J. Solid State Chem. 184 (2011) 2009–
2
016.
[19] S. Altmann, J. Pfeiffer, Monatsh. Chem. 134 (2003) 1081–1092.
[20] M. Castellano, A. Gandini, P. Fabbri, M.N. Belgacem, J. Colloid Interface Sci. 273
(
2004) 505-511.
21] L.V. Habekost, G.B. Camacho, G.S. Lima, F.A. Ogliari, E. Piva, R.R. Moraes, J. Appl.
Polym. Sci. 127 (2013) 3467-3473.
22] S. Kim, K. Qadir, S. Jin, A. Satyanarayana Reddy, B. Seo, B.S. Mun, S.H. Joo, J.Y.
Park, Catal. Today 185 (2012) 131-137.
23] D. Çelik, S. Karahan, M. Zahmakıran, S. Özkara, Int. J. Hydrogen Energy 37
2012) 5143-5151.
24] L. De Rogatis, T. Montini, M.F. Casula, P. Fornasiero, J. AlloysCompd. 451 (2008)
16-520.
25] L. Huang, P. Luo, W. Pei, X. Liu, Y. Wang, J. Wang, W. Xing, J. Huang, Adv. Synth.
Catal. 354 (2012) 2689-2694.
26] T. Ashida, K. Miura, T. Nomoto, S. Yagi, H. Sumida, G. Kutluk, K. Soda, H.
Namatame, M. Taniguchi, Surf. Sci. 601 (2007) 3898-3901.
[
[
[
(
[
5
[
[
2
8
Page 28 of 54

[27] V. Höller, D. Wegricht, I. Yuranov, L. Kiwi-Minsker, A. Renken, Chem. Eng.
Technol. 23 (2000) 251-255.
[
[
[
28] R. Xu, T. Xie, Y. Zhao, Y. Li, Nanotechnology 18 (2007) 1-5.
29] H.-U. Blaser, M. Studer, Appl. Catal., A 189 (1999) 191-204.
30] F.T. Zangeneh, S. Sahebdelfar, M.T. Ravanchi, J. Nat. Gas Chem. 20 (2011) 219-
2
31.
31] C. Torres, C. Campos, J.L.G. Fierro, M. Oportus, P. Reyes, Catal. Lett. 143 (2013)
63-771.
[
7
[
32] J.J.F. Scholten, A.P. Pijpers, A.M.L. Hustings, Catal. Rev. 27 (1985) 151-206.
33] F. Cárdenas-Lizana, D. Lamey, S. Gómez-Quero, N. Perret, L. Kiwi-Minsker, M.A.
[
Keane, Catal. Today 173 (2011) 53-61.
34] R.O. Pinho, E. Radovanovic, I.L. Torriani, I.V.P. Yoshida, Eur. Polym. J. 40 (2004)
15–622.
35] M.A. Schiavon, S.U.A. Redondo, S.R.O. Pina, I.V.P. Yoshida, J. Non-Cryst. Solids
04 (2002) 92–100.
36] B. Friederich, A. Laachachi, M. Ferriol, D. Ruch, M. Cochez, M.Toniazzo, Polym.
Degrad. Stab. 95 (2010) 1183-1193.
[
6
[
3
[
[37] B.L. Rivas, M. Palencia, Sep. Purif. Technol. 81 (2011) 435–443.
[38] B.L. Rivas, L.N. Schiappacasse, E. Pereira, I. Moreno-Villoslada, Polymer 45
(
2004) 1771–1775.
[39] J. Kiss, A. Oszkó, G. Pótári, A. Erdőhelyi, Vacuum 86 (2012) 594-598.
[40] N.Y. Jin-Phillipp, P. Nolte, A. Stierle, H. Dosch, Surf. Sci. 603 (2009) 2551-2555.
2
9
Page 29 of 54