CHEMICAL ENGINEERING TRANSACTIONS
VOL. 57, 2017
A publication of
The Italian Association
of Chemical Engineering
Online at www.aidic.it/cet
Guest Editors: Sauro Pierucci, Jiří Jaromír Klemeš, Laura Piazza, Serafim Bakalis
Copyright © 2017, AIDIC Servizi S.r.l.
ISBN 978-88-95608- 48-8; ISSN 2283-9216
Partial Oxidation of Gasoline over Ni/Al2O3 Catalysts
using Nickel Aluminate as Precursor
Ruben López-Fonseca*, Cristina Jiménez-González, Miryam Gil-Calvo, Beatriz de
Rivas, Jose-Ignacio Gutiérrez-Ortiz
Department of Chemical Engineering, Faculty of Science and Technology, University of The Basque Country UPV/EHU,
P.O. Box 644, E-48080 Bilbao, Spain
ruben.lopez@ehu.eus
Gasoline, which currently presents a well-developed distribution network, represents an ideal fuel source for
H2 production onboard or in stationary facilities. In this work, a novel alumina supported Ni (NiAl2O4/Al2O3)
catalyst derived from the reduction of nickel aluminate at high temperature was investigated for the partial
oxidation of isooctane, which was selected as a gasoline surrogate. While being relatively active in the
process due to its small crystallite size, the catalyst did not show a stable behavior with time on stream since a
considerably coke formation was noticed. Interestingly a better performance in terms of conversion, yield of
hydrogen and stability was observed when water was added to the feedstream.
1. Introduction
Low-temperature fuel cells are considered viable candidates for direct electricity production from hydrogen for
transportation applications and also for distributed and portable power generation. However, the absence of a
feasible hydrogen storage option and a hydrogen marketing infrastructure, at least in the near term,
necessitate the search for an appropriate fuel. In this sense, gasoline, which already presents a well-
developed distribution network, represents an ideal fuel source for H2 production onboard or in stationary
facilities supplying refuelling stations with H2. Moreover, gasoline has both a higher energy density and larger
hydrogen content when compared with oxygenated hydrocarbons such as methanol and ethanol (Moon et al.,
2004). In this work, isooctane has been selected as a gasoline surrogate and its conversion to H2 by partial
oxidation has been explored for the following reasons: (i) the reaction, C8H18 + 4O2 8CO + 9H2,
(HR(25 ºC)= - 627 kJ∙mol
-1), is exothermic, making it much more energy-efficient than steam reforming; (ii) a
smaller reformer can be used to achieve a high conversion of the hydrocarbon selectively in favour of the
production of H2 at short contact times; and (iii) the partial oxidation setup is more compact and mechanically
simpler than the steam reforming, since no additional heating is required.
On the other hand, nickel-based catalysts offer a good compromise between cost and reforming behaviour in
spite of the fact that its operation is frequently accompanied by coking. Most studies related to this class of
catalysts use nickel oxide as a precursor to obtain active metallic nickel after reduction. An alternative
approach for synthesis is to stabilise nickel within a well-defined crystalline oxide nickel aluminate (NiAl2O4).
The potential of this precursor for methane reforming by partial oxidation, steam reforming or dry reforming
has been successfully reported (López-Fonseca et al., 2012). With the aim of optimising the mass catalytic
activity of the nickel active phase it seems reasonable to incorporate the spinel on a high specific surface
support. Alumina can be considered a suitable candidate given the structural and chemical compatibility with
reduced NiAl2O4, which results in Ni/Al2O3. In this regard our recent results on this type of catalyst formulation
prepared by precipitation pointed out that NiAl2O4/Al2O3 catalysts (10-24 w/w % Ni loading) exhibited a
comparable reforming activity than a bulk spinel sample (33 w/w % Ni) (Jiménez-González et al., 2015) for
methane reforming. It would be therefore of interest to further investigate the viability of this kind of samples
for processing more complex fuels by simpler reforming strategies such as partial oxidation.
DOI: 10.3303/CET1757160
Please cite this article as: Lopez-Fonseca R., Jimenez-Gonzalez C., Gil-Calvo M., De Rivas B., Gutierrez-Ortiz J.I., 2017, Partial oxidation of
gasoline over ni/al2o3 catalysts using nickel aluminate as precursor, Chemical Engineering Transactions, 57, 955-960
DOI: 10.3303/CET1757160
955
mailto:ruben.lopez@ehu.eus
2. Experimental
The spinel-derived Ni/Al2O3 catalyst was synthesised by co-precipitation according to the following procedure.
Hence, the process was conducted by the drop-by-drop addition under constant stirring of a 0.6 M solution of
NH4OH into an aqueous slurry of a mixture of Ni(CH3-COO)2·4H2O and Al(NO3)3·9H2O (1:2 Ni/Al molar ratio)
and crushed -Al2O3 (133 m
2 g-1, 0.3-0.5 mm, SA 6173, Saint-Gobain) to obtain 17 w/w % nickel loading. The
temperature was kept at 25 ºC during the precipitation and the pH was fixed at 8. Afterwards the precipitate
was aged for 30 min before being filtered and washed with hot deionised water. The sample was dried at
110 ºC overnight and then calcined at 850 ºC in static air for 4 h at a heating rate of 10 ºC min-1. As a
reference reforming catalyst, a commercial powdered Rh-based catalyst was used (1 %Rh/Al2O3, BET surface
area = 132 m2 g-1, Alfa Aesar), which has been calcined at 700 ºC in static air for 4 h with the same heating
rate. Then pellets were prepared by compressing the powders into flakes in a hydraulic press (Atlas Series
Manual Hydraulic Press, Specac), crushing and sieving (0.3-0.5 mm).
2.1 Catalyst characterisation
The nickel-based catalyst was characterised by N2 physisorption at -196 ºC, wavelength dispersive X-ray
fluorescence (WDXRF), X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), transmission
electron microscopy (TEM), temperature programmed reduction with hydrogen (H2-TPR) and temperature
programmed desorption of NH3 (NH3-TPD). The spent samples were thoroughly characterised by BET
measurements, XRD, TEM, thermogravimetry coupled to mass spectrometry (TGA-MS) and Raman
spectroscopy. The experimental details of each technique are described elsewhere (Boukha et al., 2014).
2.2 Catalytic tests
The catalytic tests for partial oxidation of isooctane were performed using 125 mg of catalyst in a flow reactor
operating at atmospheric pressure and at a constant temperature of 600 ºC for 30 h. The reaction feed
consisted of a mixture of the hydrocarbon (2 vol. % C8H18 and oxygen (molar ratio O/C = 1)) diluted in N2 with
a total flow rate of 800 cm3 min-1. For the partial oxidation runs carried out under humid conditions water was
added in order to obtain a H2O/C molar ratio of 3. Prior to the reaction, the Ni-based catalyst was activated by
reduction with 5 vol. % H2/N2 at 850 ºC for 2 h whereas the Rh-based catalyst was reduced using the same
reducing stream at 700 ºC for 2 h. The liquid hydrocarbon and deionised water feeds have been delivered by
two different HPLC pumps and were vaporized separately, then mixed with the other gaseous components
(O2 and/or N2) in heated gas lines. Both hot box and feed lines have been heated to a temperature of 180 ºC
in order to vaporize the liquid feed. The obtained reaction products have been passed through a stainless
steel cold water condenser to collect excess water and the unreacted hydrocarbon before injection into the
gas chromatograph. Therefore, dry gas stream at the outlet of the reactor was analysed online by a gas
chromatograph (MicroGC 3000A, Agilent Technologies) equipped with a TCD detector. Hydrocarbon
conversion and yields of H2, CO, CO2 and CH4 have been determined according to these equations:
out out 2 out 4
in 8 18
F CO +F CO +F CH
, %
8 F C H
X
100 (1)
out 2
2
in 8 18
F H
Y(H )=
8 F C H
(2)
out
in 8 18
F CO
Y(CO)=
8 F C H
(3)
out 2
2
in 8 18
F CO
Y(CO )=
8 F C H
(4)
out 4
4
in 8 18
F CH
Y(CH )=
8 F C H
(5)
where Fin(C8H18) is the inlet molar flow rate of isooctane and Fout (i) is the outlet molar flow rate of the ith
gaseous component.
The thermodynamic data were calculated via the HSC Chemistry software package by the GIBBS programme
using the so-called Gibbs Energy Minimisation Method. In addition to solid carbon, the following gaseous
substances were considered: i-C8H18, CH4, O2, N2, CO, CO2, H2 and H2O.
956
3. Results and discussion
3.1 Characterisation of the fresh catalyst
The XRD pattern of the calcined NiAl2O4/Al2O3 catalytic precursor is included in Figure 1 ((a) sample). A set of
diffractions peaks at 2 = 19.3º, 31.5º, 37.2º, 45.2º, 59.9º and 65.7º, assignable to a nickel aluminate phase,
could be identified. In addition to the signals corresponding to NiAl2O4, the pattern expectedly should contain
the signals attributable to the support. However, since the most intense signals of the alumina are close to
those of the spinel phase, the clearest evidence of the presence of the support was perhaps given by the
shoulder peak at 2 = 67.0º. On the other hand, highly crystalline NiO phase with distinct signals at 2 = 43.5º
and 62.9º (JCPDS 89-7131) was not found.
The formation of the spinel phase on the surface of the alumina was further corroborated by XPS and UV-vis
DRS. As for XPS analysis the Ni 2p3/2 spectrum of the calcined samples (Figure 1) was composed of two
contributions, namely the main peak corresponding to the Ni2+ ions and its satellite. The apparent symmetry of
this Ni signal suggested the presence of a single homogeneous phase. In fact, the binding energy of this band
was centred at about 855.8 eV, which was close to the theoretical value for the nickel aluminate phase
(856.0 eV). Moreover, the separation between the principal peak and its satellite (6.1 eV) matched with the
reference value (6.3 eV) corresponding to the nickel aluminate spinel (Heracleous et al., 2005).
Figure 1: XRD patterns of the calcined (a) and reduced (b) spinel catalyst (NiO (●), NiAl2O4 (■) and Al2O3 ()).
XPS spectrum of Ni 2p3/2 region of the calcined spinel catalyst.
The structural features of the sample were also examined by UV-Vis DRS (results are not shown here). Thus,
in the visible range the bands located at 380 and 720 nm were associated with Ni2+ ions hosted by octahedral
sites. The relative low intensity of these bands implied that the detected Ni2+ in an octahedral environment was
not ascribable to the presence of massively segregated NiO phase but to Ni2+ ions belonging to the nickel
aluminate lattice. The spectrum was also characterised by the presence of an additional intense band at 600-
645 nm and shoulders at 550 and 760 nm, thereby evidencing the presence of the tetrahedrally coordinated
Ni2+ ions in the nickel aluminate lattice (Rogers et al., 2016). Additionally, the precursor was analysed by H2-
TPR in order to identify the nickel species, their relative abundance and reducibility (Figure 2). It was observed
that the H2 uptake (2.8 mmol H2 g
-1, virtually identical to the theoretical reducibility) was noticeable in a wide
temperature range and the reduction process was complete at 950 ºC. The main contribution was found at
about 800 ºC corresponded to the reduction of the NiAl2O4 spinel. The shoulder identified at 650 ºC was
assigned to the presence of Ni2+ ions that were not completely integrated into the spinel while the small
consumption detected at 450 ºC (lower than 6 %) was related to free NiO species.
The reduced sample (submitted to a reduction step at 850 ºC for 2 h with 5 vol. %H2/N2) was analysed by
XRD, TEM and NH3-TPD. Recall that this was the activation procedure (NiAl2O4/Al2O3Ni/Al2O3) used in situ
in the reactor prior to reforming runs. The XRD pattern is shown in Figure 1 ((b) sample). It was found that Ni2+
species were massively reduced into metallic Ni (JCPDS 89-7128, peaks at 2 = 44.6º, 52.0º and 76.5º).
Moreover, the reduction of the spinel phase was complete and simultaneously provoked the formation of the
alumina phase. It is worth noting that this phase transformation involved a limited decrease in the surface area
from 94 to 84 m2 g-1. On the other hand, as evidenced by TEM analysis the particle size distribution (around
300 particles were measured) was characterised by a relatively symmetrical band. Most of the particles (99 %)
showed a size lower than 20 nm. An average value of 9.5 nm was thus estimated. This value was in good
agreement with the mean size calculated by XRD from the Ni(200) signal at 2 = 52º (9 nm). On the other
hand, the overall acidity (NH3-TPD) of the reduced catalyst was 302 mol NH3 g
-1, significantly lower than that
of the bare alumina support (630 mol NH3 g
-1). The desorption profile revealed the presence of a band at low
temperatures (175 ºC) accompanied by a second much more intense feature at relatively high temperatures
(300 ºC), the latter being due to a notable fraction of strong acid sites (about 85%) present at the surface of
the catalyst.
In
te
ns
ity
, a
.u
. Satellite
Ni 2p3/2 NiAl2O4
875 870 865 860 855 850 845
Binding energy, eV
R
el
at
iv
e
in
te
ns
ity
, a
.u
.
(a)
(b)
10 20 30 40 50 60 70 80
Angle, 2
In
te
ns
ity
, a
.u
. Satellite
Ni 2p3/2 NiAl2O4
875 870 865 860 855 850 845
Binding energy, eV
In
te
ns
ity
, a
.u
. Satellite
Ni 2p3/2 NiAl2O4
875 870 865 860 855 850 845
Binding energy, eV
R
el
at
iv
e
in
te
ns
ity
, a
.u
.
(a)
(b)
10 20 30 40 50 60 70 80
Angle, 2
R
el
at
iv
e
in
te
ns
ity
, a
.u
.
(a)
(b)
10 20 30 40 50 60 70 80
Angle, 2
957
Figure 2: H2-TPR profile of the calcined spinel catalyst.
3.2 Catalytic behaviour
Figure 3 compares the evolution of isooctane conversion by partial oxidation in absence or presence of water
(POX and WET POX, respectively) as a function of time over the Ni and Rh catalysts at 600 ºC (7500 mLC8H18
g-1 h-1; O/C=1). Table 1 shows the values of Y(CO), Y(CO2), Y(CH4), H2/CO and CO/CO2 molar ratios for the
two investigated catalysts taken the experimental data at the beginning and the end of the runs.
For the POX process, the nickel catalyst showed a relatively low initial conversion (60 %) and a less stable
performance with time, while the commercial Rh catalyst was slightly more active (73 %) and exhibited a lower
rate of loss of activity. Thus, conversion steadily decreased from 60 to a relatively constant value of 48 %
during the first 10h-interval. The loss of activity over the Rh catalyst was also visible but an unchanged
conversion (57 %) was attained sooner.
Correspondingly, the yield of hydrogen, Y(H2), decreased from 0.49 to 0.30 for Ni catalyst and from 0.50 to
0.44 for Rh catalyst. While, no significant differences were found in the H2/CO molar ratio over the two
catalysts, the product distribution for the Ni catalyst was characterised by a lower yield of CO (0.25 compared
with 0.35 over the Rh catalyst) and a low CO/CO2 molar ratio (1.1 compared with 1.8 over the Rh catalyst).
This was associated with a higher activity of the noble metal catalyst for the reverse water gas shift reaction
that induced a larger presence of CO and/or a favoured transformation of CO into carbon for the nickel
catalyst. On the other hand, the Y(CH4) was very low (0.01-0.02) for both catalysts.
Figure 3: Evolution of isooctane conversion with time in the investigated partial oxidation processes.
The observed deactivation of the Ni catalyst could be related to several causes including coke formation,
probably caused by unwanted reactions such as the Boudouard reaction and/or thermal cracking of isooctane,
nickel sintering and partial nickel oxidation of the metallic nickel to NiO. Therefore, in order to analyse the
extent of these potential deactivating phenomena, the spent Ni catalyst was characterised by TGA-MS, TEM,
BET measurements, XRD and Raman spectroscopy. As revealed by TGA-MS a notable formation of coke of
about 83 wt % was observed (Figure 4). This finding was related to the substantial strong surface acidity of the
catalyst. In accordance with Westrich et al. (2010), the two distinct oxidation peaks at 530 and 570 ºC (Figure
4) can be attributed to the combustion of filamentous carbon present on the catalyst. The absence of oxidation
peaks below 450 ºC seems to indicate that the sample did not contain appreciable amounts of coating carbon.
In line with the temperature-programmed oxidation profiles, TEM images confirmed that the spent sample was
0 5 10 15 20 25 30 35
10
30
50
70
90
time, h
X
, %
POX
WET POX
Ni catalyst
Rh catalyst
0 5 10 15 20 25 30 35
10
30
50
70
90
time, h
X
, %
POX
WET POX
Ni catalyst
Rh catalyst
H
2
up
ta
ke
, m
m
ol
g-
1
100 200 300 400 500 600 700 800 900 10 20 30
Temperature, ºC Time, min
Isothermal
step
450 ºC
650 ºC
800 ºC
H
2
up
ta
ke
, m
m
ol
g-
1
100 200 300 400 500 600 700 800 900 10 20 30
Temperature, ºC Time, min
Isothermal
step
450 ºC
650 ºC
800 ºC
958
mainly covered with filamentous carbon which in turn had an effect on its textural properties as well. Indeed,
the specific surface area markedly increased (from 84 to 117 m2 g-1) owing to the contribution of the porosity
of the deposited carbon filaments. For the Rh catalyst, it was observed that the considerably lower coke
deposition (slightly larger than 2 wt %) did have a clear negative influence on stability. In this case,
encapsulating coke was mainly formed given the relatively low combustion temperatures (about 375 ºC), the
absence of filamentous carbon (as revealed by TEM images of the spent sample) and the observed decrease
in specific surface area (from 132 to 114 m2 g-1).
Table 1: Behaviour of the investigated catalysts with time for the partial oxidation of isooctane in the absence
or presence of water.
POX WET POX
Ni/Al2O3 Rh/Al2O3 Ni/Al2O3 Rh/Al2O3
t=0 t=30 h t=0 t=30 h t=30 h t=30 h
X, % 60 48 73 57 81 46
Y(H2) 0.49 0.30 0.50 0.44 1.30 0.41
Y(CO) 0.36 0.25 0.46 0.35 0.16 0.13
Y(CO2) 0.22 0.22 0.41 0.20 0.59 0.32
Y(CH4) 0.05 0.01 0.20 0.02 0.06 0.01
H2/CO 1.5 1.3 1.4 1.4 8.9 3.6
CO/CO2 1.6 1.1 2.0 1.8 0.3 0.4
On the other hand, the diffraction pattern of the post-run nickel catalyst also revealed the presence of large
amounts of coke as evidenced by the characteristic peak of graphitic carbon at 2 = 26.3º (JCPDS 89-8487).
However, since the absence of amorphous carbon could not be ruled out by XRD, the spent catalyst was
characterised by Raman spectroscopy as well. Hence, two distinct bands were detected at 1340 cm-1 (the so-
called D band) attributed to the defective and disordered structures and 1580 cm-1 (the so-called G band)
attributed to ordered graphitic coke (Yi et al., 2006). Thus, it was found that both amorphous and graphitic
filamentous carbon were present with an ID/IG intensity ratio of about 1.2. Further, XRD analysis of the used
samples also suggested the partial oxidation of the active phase into inactive NiO as evidenced by the signal
at 2 = 43.5º. Besides, the average Ni particle size estimated by TEM suggested that a slight sintering
occurred with an increase in the size from 9 to 11 nm. Although these two findings may also induce a
decrease in conversion with time on stream, it is however believed that, judging from the extent of these
phenomena, these were of secondary importance in comparison with coke formation.
In spite of the fact that the behaviour of the Ni catalyst in the reforming process pointed out that the synthesis
route based on the use of nickel aluminate as precursor was attractive, it was also clear that the reforming of
liquid hydrocarbons by partial oxidation would surely require higher reaction temperatures to partially avoid
coking and/or the addition of water to notably favour the gasification of deposited coke (Vivanpatarakij et al.,
2014). Particularly our attention was focused on this last strategy. Therefore new catalytic runs were carried
out (C8H18 + 3.5O2 + H2O 8CO + 10H2; HR(25 ºC)= - 385 kJ mol
-1) in presence of water (H2O/C = 3) in the
feedstream resulting in the so-called WET POX process (which also could be seen as an oxidative steam
reforming). The same operation conditions were examined, namely 7500 mLC8H18 g
-1 h-1; O/C=1; 600ºC, 30 h.
Figure 4: DTG signal of the combustion of deposited coke on the spent Ni catalyst.
200 300 400 500 600 700 800
Temperature, ºC
POX
Wet POX
460 ºC
530 ºC 570 ºC
dT
G
si
gn
al
, m
g
m
in
-1
200 300 400 500 600 700 800
Temperature, ºC
POX
Wet POX
460 ºC
530 ºC 570 ºC
dT
G
si
gn
al
, m
g
m
in
-1
959
Results showed in Figure 1 were very promising since a noticeably higher conversion (81 %) and, perhaps
more importantly, a marked stability were noticed for the Ni catalyst. This performance revealed that WET
POX reforming of isooctane over the spinel-derived nickel catalyst was a viable strategy. Additional beneficial
effects related to the presence of water were also observed in terms of a higher H2/CO molar ratio (close to 9)
and a lower CO/CO2 ratio (<0.3) owing to the promotion of the water gas shift reaction.
The post-run spent sample subjected to WET POX was also thoroughly characterised. The most important
feature was that the amount of coke was considerably reduced to about 24 % (Figure 4). Although XRD
analysis and TEM images evidenced that the morphology was filamentous as well, its chemical nature was
less recalcitrant as oxidation at high temperatures (570 ºC) was not observed. In view of these results, coke
formation in this reforming process apparently did not lead to deactivation, thereby suggesting that the
accessibility to active nickel sites was not affected to a large extent.
4. Conclusions
The production of H2-rich streams by partial oxidation of isooctane combined with the presence of water was
feasible over a 17 %/Ni/Al2O3 catalyst derived from nickel aluminate as precursor. The good performance was
related to the relatively small size of nickel particles (9 nm). Water was required in order to minimise coke
formation owing to thermal cracking of the feed and/or the Boudouard reaction. Filamentous carbon was
formed although accessibility to active sites was not negatively impacted since high conversion and
remarkable stability were noted. Further, the Ni-based catalyst exhibited a better performance in terms of
isooctane conversion and yield of H2 than that shown by a commercial Rh-based catalyst (1 % Rh/Al2O3).
Acknowledgments
The authors wish to thank the financial support for this work provided by the Ministry of Economy and
Competitiveness (ENE2013-41187-R), the Basque Government (PRE_2013_2_453, IT657-13) and the
University of The Basque Country (UFI 11/39). Technical and human support from SGIker is also gratefully
acknowledged.
Reference
Boukha Z., Jiménez-González C., de Rivas B., González-Velasco J.R., Gutiérrez-Ortiz J.I., López-Fonseca R.,
2014, Synthesis, characterisation and performance evaluation of spinel-derived Ni/Al2O3 catalysts for
various methane reforming reactions, Appl. Catal. B: Environ., 158-159, 190-201.
Heracleous E., Leeb A.F., Wilson K., Lemonidou A.A., 2005, Investigation of Ni-based alumina-supported
catalysts for the oxidative dehydrogenation of ethane to ethylene: structural characterization and reactivity
studies, J. Catal., 231, 159-171.
Jiménez-González C., Boukha Z., de Rivas B., Delgado J.J., Cauqui M.A., González-Velasco J.R., Gutiérrez-
Ortiz J.I., López-Fonseca R., 2013, Structural characterisation of Ni/alumina reforming catalysts activated
at high temperatures, Appl. Catal. A: Gen., 466, 9-20.
Li Y., Zhang B., Xie X., Liu J., Xu Y., Shen W., 2006, Novel Ni catalysts for methane decomposition to
hydrogen and carbon nanofibers, J. Catal., 238, 412-424.
López-Fonseca R., Jiménez-González C., de Rivas B., Gutiérrez-Ortiz J.I., 2012, Partial oxidation of methane
to syngas on bulk NiAl2O4 catalyst. Comparison with alumina supported nickel, platinum and rhodium
catalysts, Appl. Catal. A: Gen., 437-438, 53-62.
Moon D.J., Ryu J.W., Lee S.D., Lee B.G., Ahn B.S., 2004, Ni-based catalyst for partial oxidation reforming of
isooctane, Appl. Catal: A: Gen, 272, 53-60.
Rogers J.L., Mangarella M.C., D’Amico A.D., Gallagher J.R., Dutzer M.R., Stavitski E., Miller J.T., Sievers C.,
2016, Differences in the nature of active sites for methane dry reforming and methane steam reforming
over nickel aluminate catalyst, ACS Catal., 6, 5873-5886.
Vivanpatarakij S., Rulerk D., Assabumrungrat S., 2014, Removal of tar from biomass gasification process by
steam reforming over nickel catalysts, Chemical Engineering Transactions, 37, 205-210, DOI:
10.3303/CET1437035
Westrich T.A., Chen X., Schwank J.W., 2010, Isooctane decomposition and carbon deposition over ceria-
zirconia supported nickel catalysts, Appl. Catal. A: Gen., 386, 83-93.
960