T.-S. Nguyen et al. / Journal of Catalysis 344 (2016) 136–140
137
1
00 mL of 15 wt% H
and placed in a sealed autoclave kept at 200 °C for 15 days. The
obtained solid was washed several times with 0.1 M NH NO to
2
SO
4
solution were mixed in a Teflon reactor
ing on the products, and used for checking the molar balance and
measuring the selectivities. Methane and methanol, which were
inherently formed during HDO reactions and detected online by
GC, together with water, are not accounted for in the selectivity
data in order to focus on the (most interesting) C -C products.
6 8
However, taking their presence into account, the molar balance
was close to 98% for all experiments.
4
3
remove adsorbed sulfate, then washed with distilled water, dried
at 100 °C overnight, and calcined at 350 °C in air for 2 h. AuRh
NPs, along with their monometallic counterparts, were prepared
by a colloidal chemical (co)reduction route [19] adapted from
Toshima, Prati, Hutchings, and coworkers [20–24]. The metal pre-
cursors were HAuCl
4
Á3H
2
O (Strem Chemicals, 99.9%, 49 wt% Au)
3
. Results
and RhCl O (Sigma-Aldrich, 99.9%, 38–40 wt% Rh). In a first
3
ÁnH
2
step, a 200 mL aqueous solution containing the two metallic pre-
cursors was prepared by adding the amounts of precursors neces-
sary for reaching an Au loading of 2 wt% and/or a Rh loading of 1 wt
3.1. Catalyst preparation and characterization
Au, Rh, and AuRh nanoparticles were synthesized by conven-
%
. Next, a 1 wt% aqueous solution of a stabilizing agent, polyvinyl
tional colloidal chemical co-reduction in water, using chloride salts
as metal precursors, polyvinyl alcohol (PVA) as a surfactant, and
NaBH as a reducing agent. Then, a powder of single-phase rutile
4
titania nanorods synthesized using a hydrothermal method was
added to the acidified colloidal suspension, which led to the immo-
bilization of PVA-embedded nanoparticles on the titania support
alcohol (PVA, Mw = 10,000) was added to the preceding solution
while keeping always a mass ratio mPVA/mmetal of 1.2. A solution
of 0.1 M NaBH , freshly prepared and kept at 0 °C before use, was
4
then dropped under stirring into the metallic precursor solution
with a molar ratio nNaBH4/nmetal of 5. Stirring was then maintained
for 30 min to allow the complete decomposition of the remaining
(
Section 2.1) [19]. The metal loadings of the resulting catalysts
NaBH
tion of 0.01 M HCl in order to favor the sol immobilization onto
the TiO support. The amount of support necessary for reaching
4
excess. The solution was then acidified to pH 3.5 by addi-
are 1.75 wt% Au, 0.62 wt% Rh and 1.58 wt% Au + 0.68 wt% Rh (i.e.
Au55Rh45 molar composition), as determined by ICP-OES. Fig. 1
shows TEM images of the as-prepared TiO -supported Au, Rh and
2
AuRh catalysts. In all cases, the nanoparticles are well distributed
on their support and exhibit a roundish shape with a relatively
sharp size distribution. The average metal particle sizes, as deter-
mined by analyzing several hundreds of nanoparticles on the
TEM images, are 2.8 ± 0.8 nm, 2.4 ± 0.6 nm, and 3.3 ± 1.0 nm for
Au, Rh and Au55Rh45, respectively. As previously reported follow-
ing infrared spectroscopy and scanning TEM, after a treatment at
2
the final metal loading was then added and stirring was continued
for 3 h. Finally, the material was filtered, washed several times
with hot distilled water (70 °C), and dried in air at 100 °C
overnight.
2.2. Catalyst characterization
Transmission electron microscopy (TEM) analyses were per-
3
2
50 °C in H flow similar to the in situ pretreatment performed in
formed with an aberration-corrected FEI ETEM Titan G2 (CLYM)
operated at 300 kV and a Jeol JEM 2010 operated at 200 kV. The
samples were crushed in ethanol and the solution was ultrasoni-
cally stirred before dropping on a holey carbon-covered copper
TEM grid, followed by drying.
The metal concentrations were determined by inductively cou-
pled plasma optical emission spectroscopy (ICP-OES, Activa instru-
ment from Horiba Jobin Yvon). In order to dissolve them
this work before the catalytic tests, the PVA surfactant is com-
pletely removed and the bimetallic nanoparticles adopt a Janus-
type configuration with Au/Rh/TiO stacking [19,25]. The observed
2
segregation between Au and Rh is consistent with previous works
on the bulk-immiscible Au-Rh system [26,27], and has been
recently rationalized through DFT calculations on small clusters
[
25,28].
2 4
completely, the samples were treated with a mixture of H SO ,
aqua regia and HF at 250–300 °C.
3.2. Catalyst evaluation in guaiacol HDO
2
.3. Catalytic testing
Au/TiO
ated in gas-phase guaiacol hydroconversion in a flow-fixed bed
reactor under high pressure of hydrogen (4 MPa H , 2.7 kPa guaia-
col) between 240 °C and 300 °C. Table 1 and Fig. S1 (Supporting
Information) report guaiacol conversion rate data. The activities
of the three noble metal-containing catalysts appear similar and
2 2 2 2
, Rh/TiO , AuRh/TiO and bare TiO samples were evalu-
The guaiacol HDO reaction was performed in a gas-phase con-
tinuous flow reactor under H pressure (4 MPa) between 240 and
00 °C. Before the catalytic tests, the as-prepared catalysts were
2
2
3
reduced in the reactor under pure hydrogen flow (60 NmL/min)
at 300 °C for 1 h. The whole catalytic bench has been previously
described elsewhere [6]. The solid catalyst was placed on a frit
located in a tubular glass tube, which was itself inserted into a
tubular stainless-steel reactor. A 2.7 kPa partial pressure of guaia-
2
much higher than those of bare TiO . The turnover frequencies
are of 1.3–1.5 guaiacol molecules converted per surface metal
atom per second at 280 °C (Table 1). As shown in the Arrhenius
plot of Fig. S1, the apparent activation energy for the AuRh catalyst
col was generated by flowing H
(
(
between 50 and 100 mg, corresponding to weight hourly space
velocities (WHSV) between 6 and 3 h , respectively. The gas flows
and the total pressure were controlled using mass-flow controllers
and a back-pressure regulator, respectively. The products were
analyzed online using a gas chromatograph (HP 5890 II) equipped
with a flame ionization detector and a Varian CPSIL-5 column
2
(60 NmL/min) in a saturator
130 °C)/condenser (116 °C) system containing liquid guaiacol
Acros Organics, 99% purity). The tested catalyst weight comprised
(112 kJ/mol) is intermediate between those for Rh/TiO
2
(88 kJ/mol)
and Au/TiO (126 kJ/mol).
2
Guaiacol HDO has been previously shown [1] to proceed mostly
through demethylation, demethoxylation, hydrogenation and
dehydroxylation reactions, as illustrated in Scheme 1. In a first
step, guaiacol can lose a methyl group (demethylation) to form cat-
echol, or can be demethoxylated directly to phenol. Demethylation
and demethoxylation reactions proceed with the formation of
methane and methanol, whose ratio depends on the catalyst and
the temperature. Methylcatechol can also be formed, either from
the transfer of the guaiacol methyl group (from methoxy sub-
stituent) or from catechol methylation. On metal catalysts, hydro-
genation of the aromatic ring of guaiacol generally occurs, forming
methoxycyclohexanol. (Methyl)catechol is dehydroxylated to
À1
(
50 m  0.32 mm  5
2
lm). The condensed products (LN trap)
were also analyzed by gas chromatography coupled with mass
spectrometry (GC–MS Agilent 5975B). The product identification
was confirmed by manual injection of pure compounds. Response
factors were determined theoretically and experimentally depend-
Nguyen, Thanh-Son
