120
Z. Skoufa et al. / Journal of Catalysis 322 (2015) 118–129
9 mm, whereas the internal diameter in the pre-catalytic and post-
catalytic sections was 4 mm. The temperature of the catalyst was
measured by a thermocouple placed in a quartz capillary well
located in the middle of the catalytic bed. The reactor was situated
in a cylindrical furnace controlled by a programmable temperature
controller.
acquired 30 min after MeOH introduction. The samples were sub-
sequently purged with He and heated up to the next higher tem-
perature, and the experimental procedure was repeated.
2.4. Temperature-programmed oxygen desorption (O2-TPD) studies
Experiments with deuterated ethane C2D6 (2% C2D6/He, Cortec-
Net; 99% atom enrichment) and ethylene C2D4 (2% C2D4/He, Cor-
tecNet; 99% atom enrichment) were conducted over NiO and
Ni0.85Nb0.15Ox in order to investigate the existence of kinetic iso-
tope effect. For exploring ethane activation, experiments with
1%C2D6/1%O2/He, 1%C2H6/1%O2/He, and 0.5%C2D6/0.5%C2H6/1%O2/
He feed were carried out. Additional isothermal experiments were
also conducted over both samples that involved switching from
1%C2H6/1%O2/He feed to 1%C2D6/1%O2/He feed. Temperature-pro-
grammed experiments with 1%C2D4/1%O2/He, 1%C2H4/1%O2/He,
and 0.5%C2D4/0.5%C2H4/1%O2/He feeds were conducted to study
ethylene conversion pathways. The kinetic isotope effects were
calculated as the ratio of normal hydrocarbon (C2H6 or C2H4) con-
sumption rate to deuterated hydrocarbon (C2H6 or C2D4) consump-
tion rate under identical conditions. The experimental procedure
involved catalyst pretreatment under 10% O2/He flow at 450 °C
for 30 min, followed by subsequent cooling to room temperature
under He flow. Finally, the temperature was raised to 450 °C at a
heating rate of 10 °C/min under a total flow of 50 cm3/min of the
appropriate composition. The reactor exit was monitored online
by mass spectrometry (Omnistar, Balzers) by following m/z signals:
4 (He), 36 (18O2, C2D6), 34 (18O16O), 32 (16O2), 30 (C2H6), 24 (C2H4),
18 (H2O), 44 (CO2), 26 (C2D4). Overlapping fragmentation contribu-
tions of various gas compounds were taken into account.
Oxygen desorption properties of fresh and used catalysts were
studied by O2-TPD measurements. The catalyst sample (200 mg)
was pretreated in a flow of He at 450 °C for 0.5 h and cooled to
room temperature under helium flow. The system was subse-
quently flushed with He for 1 h and the temperature was raised
to 850 °C at a heating rate of 15 °C/min in He (50 cm3/min). The
reactor exit was monitored online by a quadrupole mass analyzer
(Omnistar, Balzers) and the desorbed oxygen was detected by fol-
lowing the 32 (m/z) fragment. Calibration of the mass analyzer was
performed with O2/He mixtures of known concentration. Quantita-
tive estimation of the amount of desorbed oxygen was performed
by integration of the corresponding oxygen flow rates with time.
3. Results and discussion
3.1. Isotopic studies
3.1.1. Kinetic isotope effects in C2H6/O2 reaction
Two separate temperature-programmed experiments with nor-
mal (1%C2H6/1%O2/He) and deuterated (1%C2D6/1%O2/He) ethane
were conducted over both NiO and Ni0.85Nb0.15Ox. Due to the large
difference in surface area between the two catalysts, the sample
weight was adjusted accordingly in order to achieve similar con-
version levels. As expected [5], pure nickel oxide presents signifi-
cantly higher specific surface activity than Ni0.85Nb0.15Ox. The
main reaction products were ethylene (unlabeled and deuterated
labeled), carbon dioxide, and water, with no important variation
of the product distribution (selectivity) between conventional
and isotopic feed within experimental error. However, the intro-
duction of C2D6 in the feed led to significant changes in the temper-
ature profiles of ethane surface consumption rate compared to
C2H6, both over NiO and Ni0.85Nb0.15Ox as shown in Fig. 2a and b,
respectively.
Over both catalysts, the consumption of C2H6 commences at
260 °C and the consumption of C2D6 at 280 °C, indicating lower
temperature of CAH (C2H6 activation) compared to CAD (C2D6 acti-
vation) bond cleavage. Provided that only reaction rates at differ-
ential conditions (C2H6/C2D6 conversion <10%) are used so that
the reactants’ partial pressure remains almost constant, it is possi-
ble to calculate the apparent activation energy of the reaction with
the Arrhenius equation. The calculated values for ethane and deu-
terated ethane are tabulated in Table 1. In accordance with previ-
ous results [5], similar apparent activation energies for C2H6/O2
reaction were found for NiO and Ni0.85Nb0.15Ox. Moreover, for both
catalysts, the apparent activation energy for deuterated ethane
consumption is higher than the values for normal ethane, revealing
relatively easier activation of normal against deuterated ethane.
The comparison of reaction rates obtained with C2H6/O2 mix-
ture with those attained with C2D6/O2 at 330 °C (ethane conversion
<10%) allowed the quantification of the kinetic isotope effect upon
H/D isotopic switch. As tabulated in Table 1, normal kinetic isotope
effects (KIE > 1) are observed for both catalysts. The strong kinetic
isotope effect for ethane oxidative dehydrogenation, much larger
than 1, indicates clearly that CAH bond cleavage is the rate deter-
mining step for ethane activation. Moreover, it is very interesting
that similar KIE values were observed for pure nickel oxide and
nickel–niobium mixed oxide, indicating that both catalysts share
similar active sites. The calculated values are in good agreement
with published results for ethane oxidative dehydrogenation over
2.3. CH3OH sorption–desorption experiments
Methanol sorption/desorption experiments were conducted in a
TGA setup (SDT Q600, TA Instruments). A nitrogen flow was satu-
rated with methanol by being passed through a methanol-contain-
ing condenser (CH3OH, 99.8% PanReac) kept at appropriate
temperature in order to obtain ꢂ10% v/v methanol concentration.
During the sorption step, the methanol-saturated stream was
passed over the catalytic sample at 100 °C for 60 min. Subse-
quently, the sample was purged with N2 for 60 min to remove
physisorbed methanol. Finally, temperature-programmed desorp-
tion (TPD) was performed by heating the sample to 450 °C with a
heating rate of 10 °C/min under nitrogen flow.
Temperature-programmed methanol desorption was addition-
ally studied in the homemade flow apparatus described above.
The catalytic samples were placed in a U-shaped quartz reactor
and MeOH adsorption was performed at 100 °C via passing a 10%
v/v methanol-saturated helium stream for 1 h. After flushing with
He until room temperature, the catalyst sample was heated at a
constant rate of 10 °C/min under helium flow, and the reactor exit
composition was constantly monitored by online mass spectrome-
try following m/z signals 4 (He), 15 (CH4), 18 (H2O), 29 (CH2O), 31
(CH3OH), 44 (CO2), 46 (DME), 60 (methyl formate). Overlapping
fragmentation contributions of various gas compounds were taken
into account.
Finally, in situ DRIFT spectroscopy was employed for investigat-
ing surface species during methanol adsorption over NiO and
Ni0.85Nb0.15Ox catalysts. The powdered samples were placed inside
a Zn/Se window chamber (Specac), and in situ DRIFT spectra were
collected by a Bruker Tensor 27 FT-IR spectrometer. The chamber
temperature was controlled by a Specac temperature controller.
At each temperature studied (20, 50, 100, 150, 200, and 250 °C),
after acquisition of stable reference spectra under He flow, gas
phase methanol was introduced to the environmental chamber
via a MeOH-saturated helium flow. In situ DRIFT spectra were