Full Papers
then dispersed in the Ce(NO3)3 solution. The solvent was then re-
moved by using a rotary evaporator under vacuum. Subsequently,
the sample was dried further in a vacuum oven at 808C overnight.
After subjecting to calcination treatment, the support prepared by
the impregnation route was denoted as IM (impregnation).
7008C with heating ramp of 58Cminꢀ1, then kept at that end tem-
perature for 30 min before cooling down to RT (108Cminꢀ1).
Raman spectroscopy data were obtained with a Bruker Senterra
Raman spectrometer (laser wavenumber 532 cmꢀ1) at ambient con-
ditions. The spectra were recorded with a laser power of 5 mW, ex-
position time of 1–2 s. The spectra represent 50–100 scans with
Solid samples obtained after drying in the vacuum oven, were
then heated in air with a flow rate of 50 cm3 minꢀ1. The furnace
was first heated to 1508C with a temperature ramp of 58Cminꢀ1
and kept at 1508C for 1 h. Then, the furnace temperature was in-
creased to 7258C with a heating rate of 58Cminꢀ1, kept at that
temperature for 6 h prior to cooling down to RT at a rate of
a spectral resolution of 3–5 cmꢀ1
.
Catalytic testing
58Cminꢀ1
.
The catalysts or supports (15 mg) with a grain size of 0.3–0.6 mm
were packed in an a-alumina fixed-bed reactor (i.d. 4 mm) and
held in place between two quartz wool plugs. The catalyst/support
was first reduced in situ in 10 vol.% H2/Ar (flow rate 50 cm3 minꢀ1
,
Deposition of nickel
temperature ramp 108Cminꢀ1) at 6508C for 1 h. The reactor was
then purged with Ar for approximately 30 min before the furnace
temperature was raised to 7008C. An aqueous solution of m-cresol
(ꢁ20 gLꢀ1) was sent to an evaporator by using a syringe pump
(ISCO Model 500D). The solution was vaporised at 1408C. The
m-cresol/steam mixture was carried to the reactor by an Ar flow
(flow rate 25 cm3 minꢀ1). Ar was also used as an internal standard
gas for the outlet gas mixture. With such flow rates, the weight
hourly space velocity (WHSV) of m-cresol equalled 7.89 hꢀ1. The
outlet gas stream passed a flash separator in which the condensed
liquid was sampled every hour and the gas was sent directly to an
online Varian gas chromatograph (GC, CP-3800) with three parallel
separation channels using multiple GC columns (channel 1: Haye-
sep Q, Hayesep T, Molsieve 13X; channel 2: Hayesep Q, Molsie-
ve 5A; channel 3: CP Wax 52CB). The GC was equipped with ther-
mal conductivity detectors for quantification of permanent gases
(CO, CO2, H2 and Ar) and a FID for quantification of hydrocarbons.
Nickel was deposited on these supports by the homogeneous dep-
osition–precipitation technique. Typically, Ni(NO3)3·6H2O (99%,
0.1734 g) was dissolved in demineralised water in a round flask
with vigorous agitation. Then the support (ꢁ1 g) was dispersed in
the solution. The mixture was preheated to 758C. A solution con-
taining approximately 4.5 g urea was poured into the mixture at
this point. The total volume of liquid was 250 mL. Subsequently,
the temperature was increased to 908C and the mixture was kept
at this temperature until the deposition was completed (4–5 h,
final pH of 7.5). Samples were then filtered and washed with de-
mineralised water prior to drying in the vacuum oven (808C) for
3 h. After that, the solid samples were calcined (air flow rate
50 cm3 minꢀ1) at 5008C for 3 h with a heating rate of 58Cminꢀ1. All
catalysts were reduced in 10 vol.% H2/N2 at 6508C (ramping tem-
perature 108Cminꢀ1) for 3 h. After the reduction step, catalysts
were denoted as Ni/CP, Ni/HT, Ni/IM, respectively.
Condensed liquid from the flash separator was analysed offline
with a Shimadzu reverse phased High Performance Liquid Chroma-
tograph Prominence system equipped with a Hypesil Gold column
and a UV detector (l=254 nm). A mobile phase containing water/
methanol 40:60 (v/v) with a flow rate of 0.5 mLminꢀ1 was used for
analysis. The column oven was kept at 408C. Typically, it took
15 min to perform an analysis.
Catalyst characterisation
The textures of these catalysts (specific area, pore size distribution)
were determined from liquid nitrogen adsorption/desorption iso-
thermal curves acquired with a Micromeritics TriStar instrument.
Samples were degassed at 3008C prior to the analysis. X-ray dif-
fraction data were recorded with a Bruker D2 Phaser diffractometer
using CuKa radiation, l=0.1544 nm. Elemental compositions of the
catalysts were analysed by an X-ray fluorescence spectrometer
(Philips PW 1480). XPS analysis was performed with a Quantera
SXM (scanning XPS microprobe) spectrometer from Physical Elec-
tronics. X-ray (Al Ka) power of 50 W, 20 mA and a beam size
200 mm were used.
m-Cresol conversion was calculated as the mole of m-cresol react-
ed divided by the moles of m-cresol in the feed. The amounts of
unconverted m-cresol were calculated from HPLC data with the as-
sumption that the amount of m-cresol in the gas stream to GC
was negligible. Thus the condensed liquid volume was theoretical-
ly estimated, based on the feed concentration and water con-
sumed in Reactions (1) and (2). Hydrogen yield was defined as the
mole of hydrogen produced divided by the maximum amount of
H2 that can be produced based on Reaction (1):
TPR or TPO was performed in a homebuilt set-up. For TPO and TPR
measurement, the catalysts or supports (10–50 mg), respectively,
with grain sizes of 0.3–0.6 mm were packed between two quartz
plugs in a 4 mm (i.d.) a-alumina tube. Before TPR or TPO analysis,
samples were pre-treated in Ar at 1508C for 30 min. Then the oven
was cooled to RT. For TPR analysis, a 5 vol.% H2/Ar mixture with
a flow rate of 25 cm3 minꢀ1 was used. H2 consumption was moni-
tored with a standard thermal conductivity detector calibrated by
H2 reduction of NiO (purity 99.999%). For TPO analysis, a 1 vol.%
O2/He mixture with a flow rate of 75 cm3 minꢀ1 was sent through
the reactor. The gas outlet containing CO or CO2 was sent (split
ratio 6%) to an online methanizer (Model 110 Chassis, SRI Instru-
ments Europe GmbH) equipped with a flame ionisation detector
(FID). The COx produced was calibrated by using Al2(CO3)3 as
a carbon source standard (the C content of Al2(CO3)3 was deter-
mined by a CHNS elemental analyser mentioned elsewhere[8]). In
a typical TPR or TPO measurement, the furnace was heated to
C7H8O þ 13 H2O ! 7 CO2 þ 17 H2
C7H8O þ 6 H2O ! 7 CO þ 10 H2
ð1Þ
ð2Þ
Therefore, yields of and selectivities to H2, COx and phenol were
calculated as shown in Equations (3)–(7):
mole of hydrogen produced
ð3Þ
ð4Þ
ð5Þ
Yield of H2 ð%Þ ¼
ꢃ 100
17 ꢃ mole of m-cresol feed
6 ꢃ mole of phenol
Yield of phenol ð%Þ ¼
Yield of COx ð%Þ ¼
ꢃ 100
7 ꢃ mole of m-cresol feed
mole of COx
ꢃ 100
7 ꢃ mole of m-cresol feed
ChemCatChem 2015, 7, 468 – 478
477
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim