Catalytic performances of Ni and Cu impregnated MCM-41 and Zr-MCM-41 for hydrogen production through steam reforming of acetic acid

Article abstract of DOI:10.1016/j.cattod.2018.06.004

A new mesoporous catalyst support material was synthesized by incorporation of zirconia into the structure of MCM-41 through a one-pot procedure. Catalytic performances of MCM-41 and Zr incorporated MCM-41 (25Zr-MCM-41) supported Ni or Cu catalysts were investigated in steam reforming of acetic acid reaction, at 750 °C. Some deformation in the ordered pore structure of MCM-41 was observed as a result of Zr incorporation. The activity test results showed that the catalytic performances of the zirconia incorporated MCM-41 were more stable than the MCM-41 supported materials. The catalysts containing 5% and 10% Ni gave highly promising results to achieve high hydrogen selectivity. However, the performance of the catalysts containing 5% Ni was more stable. Comparison of performances of Cu and Ni based catalysts showed that Cu was not a good catalyst for the steam reforming reaction of acetic acid. In the presence of copper, mainly decarboxylation reaction of acetic acid took place, yielding large quantities of methane. Results proved that, 5% Ni impregnated 25Zr-MCM-41 was a highly promising catalytic material for hydrogen production through steam reforming of acetic acid.

Full text of DOI:10.1016/j.cattod.2018.06.004

Accepted Manuscript
Title: Catalytic Performances of Ni and Cu Impregnated
MCM-41 and Zr-MCM-41 for Hydrogen Production Through
Steam Reforming of Acetic Acid
Authors: Nurbanu Cakiryilmaz, Huseyin Arbag, Nuray Oktar,
Gulsen Dogu, Timur Dogu
PII:
S0920-5861(18)30724-7
DOI:
Reference:
https://doi.org/10.1016/j.cattod.2018.06.004
CATTOD 11493
To appear in:
Catalysis Today
Received date:
Revised date:
Accepted date:
19-1-2018
16-5-2018
4-6-2018
Please cite this article as: Cakiryilmaz N, Arbag H, Oktar N, Dogu G, Dogu T,
Catalytic Performances of Ni and Cu Impregnated MCM-41 and Zr-MCM-41 for
Hydrogen Production Through Steam Reforming of Acetic Acid, Catalysis Today
(2018), https://doi.org/10.1016/j.cattod.2018.06.004
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.

Catalytic Performances of Ni and Cu Impregnated MCM-41 and Zr-MCM-41
for Hydrogen Production Through Steam Reforming of Acetic Acid
Nurbanu Cakiryilmaz1, Huseyin Arbag1, Nuray Oktar1*, Gulsen Dogu1 and Timur Dogu2
1Department of Chemical Engineering, Gazi University, Ankara, Turkey
2Department of Chemical Engineering, Middle East Technical University, Ankara, Turkey
* Corresponding Author: nurayoktar@gazi.edu.tr; oktarnuray@gmail.com
Graphical abstract
Effect of Zr incorporation
100
80
60
40
20
0
40
35
30
25
20
15
10
5
5Ni@25Zr-MCM-41
5Ni@MCM-41
Decrease
in coke
deposition
Increase
in activity
0
30
60
90
120
150
180
Time, min

Highlights
 Zr-MCM-41 were synthesized successfully using a one-pot hydrothermal procedure
 Performances of Cu based catalyst are poor in terms of steam reforming reactions
 Ni-silica catalysts are highly promising in the production of hydrogen rich syngas
 Zr incorporation enhanced the activity of the Ni based MCM-41 catalyst
 Coke deposition over Ni based catalyst was decreased by Zr incorporation
Abstract
A new mesoporous catalyst support material was synthesized by incorporation of zirconia into
the structure of MCM-41 through a one-pot procedure. Catalytic performances of MCM-41 and
Zr incorporated MCM-41 (25Zr-MCM-41) supported Ni or Cu catalysts were investigated in
steam reforming of acetic acid reaction, at 750oC. Some deformation in the ordered pore
structure of MCM-41 was observed as a result of Zr incorporation. The activity test results
showed that the catalytic performances of the zirconia incorporated MCM-41 were more stable
than the MCM-41 supported materials. The catalysts containing 5% and 10% Ni gave highly
promising results to achieve high hydrogen selectivity. However, the performance of the
catalysts containing 5% Ni was more stable. Comparison of performances of Cu and Ni based
catalysts showed that Cu was not a good catalyst for the steam reforming reaction of acetic acid.
In the presence of copper, mainly decarboxylation reaction of acetic acid took place, yielding
large quantities of methane. Results proved that, 5% Ni impregnated 25Zr-MCM-41 was a
highly promising catalytic material for hydrogen production through steam reforming of acetic
acid.
Keywords: Hydrogen, Acetic Acid, Steam Reforming, Copper, Nickel, Zirconia

1. Introduction
Fast increase of energy demand and the increased consumption rate of fossil resources
accelerated the research activities for the development of environmentally safe alternative fuels
from renewable resources. Bio-waste components are regarded as promising resources for
hydrogen production. Catalytic processes for steam reforming of bio-oil components, bio-
ethanol and dry reforming of bio-gas yield synthesis gas, which may then be used as a resource
for the production of non-fossil fuels and valuable chemicals through Fischer-Tropsch
synthesis, methanol synthesis etc [1-3]. Hydrogen is considered as an attractive clean energy
carrier. While it can be produced from various raw materials, recent developments in fuel-cell
technologies caused significant increase in the research and development activities for its
production from renewable resources [4-7].
Flash pyrolysis of biomass (forest waste, agricultural waste, straw, industrial wood
waste etc.) yields about 75-80% bio-oil, which is considered as a potential resource for the
production of chemicals and/or fuels [7-10]. Catalytic steam reforming of bio-oil fractions is a
potential approach to produce synthesis gas, which may then be used for the synthesis of
alternative fuels or valuable chemicals. Synthesis gas produced through steam reforming of bio-
oil fractions may be used as a fuel directly in solid oxide fuel cells or may be further treated to
produce CO free hydrogen for PEM fuel cells.
The composition of bio-oil is quite complex and depends upon the source of biomass.
As it was reported in the literature, it may contain up to 20% acetic acid. Acetic acid has been
generally considered as a model compound of bio-oil for the production of synthesis gas or
hydrogen [11-16]. Acetic acid (AcOH) may also be produced through fermentation of bio and
fruit wastes. It is a non-toxic and non-flammable compound with quite high hydrogen content.
Steam reforming of AcOH may yield up to 4 moles of H2 per mole of AcOH reacted
(R.1). Hence, production of synthesis gas/hydrogen by steam reforming of acetic acid attracted
significant attention of researchers in recent years [17-24].
CH3COOH + 2H2O → 2CO2 + 4H2
(ΔHo = 131.4 kJ/mole)
(R.1)
Steam reforming of acetic acid is an endothermic reaction. Hence, it is favored at high
temperatures. Product distributions of acetic acid steam reforming process indicated formation
of mainly CO, CO2, CH4, as carbon containing compounds. Formation of some acetone was
also reported in some studies. Formation of CO and CH4 cause some decrease in hydrogen yield

from the maximum value of four per mole of AcOH reacted. Methane is formed mainly as a
result of decarboxylation of AcOH (R.2).
CH3COOH → CH4 + CO2
(ΔHo = -33.5 kJ/mole)
(R.2)
Overall stoichiometry of steam reforming of acetic acid is due to the sum of thermal
decomposition of AcOH (R.3) and water gas shift reaction (WGSR) (R.4).
CH3COOH → 2CO + 2H2 (ΔHo = 213.4 kJ/mole)
CO + H2O ↔ CO2 + H2 (ΔHo= -41 kJ/mole)
(R.3)
(R.4)
Water gas shift reaction (R.4) is an exothermic reaction and it is thermodynamically
favored at lower temperatures. Depending upon the reaction temperature and the reaction
scheme, water gas shift reaction may go through the forward or reverse directions. Methane
produced through (R.2) may also further react with steam (R.5) or with CO2 (R.6), to yield CO
and H2.
CH4 + H2O ↔ CO + 3 H2 (Steam reforming of methane)
CH4 + CO2 ↔ 2CO + 2H2 (Dry reforming of methane)
(R.5)
(R.6)
Coke formation is an undesired process which takes place during steam reforming of
AcOH. Coke formation may take place through Boudouard reaction (R.7) or through
dissociation of carbon containing products, like methane (R.8). Due to its exothermic nature,
Boudouard reaction is favored at lower temperatures, while decomposition of methane becomes
more significant at higher temperatures.
2CO ↔ C + CO2
CH4 ↔ C + 2H2
(ΔHo = -172.4 kJ/mole)
(ΔHo = 74.8 kJ/mole )
(R.7)
(R.8)
Achievement of high conversions of AcOH, to obtain high hydrogen yields and to
minimize coke formation is still important challenges of catalysis research. Catalytic
performances of noble metals, like Pt, Pd, Ru, Rh, were reported to be quite good in steam

reforming of acetic acid [9, 21, 23, 25-28]. Results reported for the comparison of catalytic
performances of noble metals and Ni based catalysts indicated that nickel was as active as noble
metals, giving very high hydrogen selectivity values in steam reforming of AcOH [9]. High
activity of Ni based catalysts is considered to be due to the ability of nickel to break the C-C,
C-H and O-H bonds. Recent results reported with Ni and/or Co based catalysts indicated high
activity for steam reforming of AcOH [13, 19, 20, 29-34]. Also, the performances of Ni-Co
based bi-metallic catalysts were generally considered as being better than the mono-metallic
catalysts, in terms of catalyst stability and coke minimization during steam or dry reforming
reactions [3, 34-36].
Copper based catalysts were also considered quite active for some reforming reactions,
like steam reforming of methanol [37, 38]. However, there are only a few studies in the
literature, reporting experimental results for steam reforming of AcOH using Cu based catalysts
[39-42]. In some of these studies, activities of Cu based catalysts were reported as being quite
low [40, 41]. Hu et al, reported that Cu was an important metal to suppress the formation of
CO, while Co was the main active metal for the reforming reaction, in the studies performed
over Co-Zn-Cu, Cu-Zn, Cu-Co and Co-Zn catalysts. However, it was reported in another study
on steam reforming of acetic acid that a Cu-Zn based catalyst which was supported on calcium
aluminate was quite active for this reaction, giving a hydrogen yield value of 80% [42].
Clarification of these conflicting results is needed in order to have a better understanding of
reaction paths over Cu and Ni based catalytic materials.
Support material of the catalysts used in reforming reactions is also expected to have
significant influence on the product distributions, as well as on coke formation and catalyst
stability [18, 19, 33, 43]. Different forms of alumina, zeolites, perovskites, zirconia etc. were
used as the support material in many of the studies investigating steam reforming of AcOH.
Discovery of the silicate structured mesoporous materials with ordered pore structures (MCM-
41, SBA-15 etc.) opened new avenues for the development of new catalysts, showing less
resistance for the transport of the reactants to the active sites and being less prone to deactivation
due to coke formation, than the conventional microporous catalytic materials. There are limited
number of publications in the literature, reporting the performances of MCM-41 type catalysts
in steam reforming of alcohols and dry reforming of methane [38, 44, 45], and wood vinegar
(contain 20% acetic acid) [46]. In order to improve the hydrothermal stability of silicate
structured SBA-15 at high temperatures, synthesis of zirconia incorporated materials and
testing of these materials as catalyst supports in reforming reactions were reported in the recent
literature [47-49]. However, information about the synthesis, characterization and testing of

zirconia incorporated MCM-41 type catalyst supports is not readily available and the use of
such zirconia incorporated MCM-41-like catalyst supports in reforming reactions need further
investigation. It is mentioned in the literature that the basic synthesis solution of MCM-41
(pH≥10) may not allow the incorporation of metal cation dopants into the silica framework and
may lead to precipitation of metal [50]. Main objectives of the present work were to obtain
detailed comparative information for the activities of Ni and Cu metals in steam reforming of
acetic acid, to have a better insight about the reaction paths over these catalytic materials, as
well as to investigate the suitability of MCM-41 and Zr-MCM-41 as the support materials of
the catalysts synthesized for this purpose.
2. Experimental
2.1.Catalyst Preparation
Mesoporous MCM-41 support was synthesized following a hydrothermal procedure
similar to the route described by Arbag et al [45]. In this procedure, a solution containing
dissolved surfactant (hexadecyltrimethylammonium bromide- CTMABr, Merck 99% pure) was
continuously stirred at 28-30°C, until a clear solution was obtained. Sodium silicate (%27 SiO2
+ %8 Na2O + %65 H2O, Merck), which was used as the silica source, was added dropwise to
the surfactant solution. Final pH of this solution was adjusted to 11 using sulfuric acid (4N,
Merck). A gel was formed and it was transferred into a Teflon-lined stainless steel autoclave
for hydrothermal synthesis for 96 h, at 120°C. The resulting solid product was filtered, washed
several times until the pH of the wash liquid was set to a constant value and then dried for 24h.
Calcination of the solid product was then performed in a tubular furnace, in a flow of dry air,
at 750oC. In the case of synthesis of Zr incorporated MCM-41-like material, a similar one-pot
hydrothermal route was followed. In this case, predetermined amount of zirconyl nitrate hydrate
(Aldrich) was added into the surfactant solution, just after the sodium silicate addition step. The
molar ratio of Zr/Si was selected as 0.25 basing on our earlier results, reported for zirconia
incorporated SBA-15 type catalytic materials [46, 51]. Hydrothermal synthesis, drying and
calcination steps were similar to the corresponding steps of MCM-41 synthesis route. This
catalyst support material, containing a Zr/Si molar ratio of 0.25 is denoted as 25Zr-MCM-41.
MCM-41 or 25Zr-MCM-41 supported Ni and Cu containing catalysts were then
prepared following a wet impregnation route. Nickel impregnated MCM-41 was prepared to
obtain catalysts containing 1.0 wt %, 5.0 wt % and 10.0 wt % Ni and they were denoted as

1Ni@MCM-41, 5Ni@MCM-41 and 10Ni@MCM-41, respectively. In the case of copper
impregnated material, Cu was impregnated to obtain a material containing 5 wt % Cu
(5Cu@MCM-41). As for the 25Zr-MCM-41 supported materials are concerned, either 5 wt %
Ni or 5 wt % Cu were impregnated and these materials were denoted as 5Ni@25Zr-MCM-41
and 5Cu@25Zr-MCM-41, respectively. Nickel nitrate hexahydrate (Ni(NO3)2.6H2O, Merck)
and copper nitrate hexahydrate (CuN2O6.3H2O, Sigma-Aldrich) were used as the Ni and Cu
precursors in the synthesis of these catalysts. After the evaporation step, these materials were
dried in an oven and then calcined in a tubular furnace under the flow of dry air. Temperature
of the calcination furnace was increased at a heating rate of 1oC/min, until 750oC was reached.
Calcination was then continued for 6 hours at 750oC. These catalytic materials were then
reduced at 750oC in a flow of hydrogen for 1 h, before the reaction tests.
2.2. Catalyst Characterization
Physical and structural properties of the synthesized catalysts were determined by
performing X-ray diffraction (XRD), N2 adsorption-desorption (BET), scanning electron
microscopy (SEM) and inductively Coupled Plasma-Mass Spectrometer (ICP-MS). SEM and
thermal analysis (TGA-DTA) of the spent catalysts were also performed to obtain information
about coke deposition and possible structural changes of the catalysts during reaction.
XRD analysis of the fresh and spent catalysts were performed by using a Rigaku
D/MAX 2200 instrument with a Cu K radiation source, at Middle East Technical University
(METU) Central Laboratory. Characterization of the pore structures of the catalytic materials
and BET surface area values were determined by means of the standard nitrogen
adsorption/desorption technique, using a Quanta Chrome-Autosorb-6 sorptometer. SEM
images of the fresh and the spent catalysts were obtained using a QUANTA 400F field emission
scanning electron microscope, coupled with EDX. ICP-MS analysis of the materials was
performed by a Perkin Elmer DRC II instrument. Thermogravimetric (TGA) and differential
thermal analysis (DTA) of spent catalysts were made by means of Setaram Setsys-1750 TG-
DTA System. These tests were performed in a stream of dry air by changing the temperature
from 30oC to 900oC, at a heating rate of 10oC/min.
2.3. Catalytic Activity Tests

Catalytic performances of the synthesized catalytic materials were tested in the steam
reforming reaction of acetic acid. These tests were performed in a quartz tubular flow reactor,
having an inner diameter of 6 mm, at 750oC. Most of the activity tests lasted for three hours.
Catalyst particles were packed into the center of the tubular reactor and supported from both
ends by quartz wool. Particle size range of the catalyst was 1-2 mm and 0.1 g of fresh catalyst
was used in each test. Acetic acid (AcOH) water mixture, with a pre-determined ratio, was
injected to the reaction system using a syringe pump. Acetic acid/water mixture was evaporated
within the evaporator of the reaction system at 140oC, where it was mixed with the carrier gas
Argon. This mixture was then sent to the tubular flow reactor. Total flow rate of the reactor
feed stream was adjusted to 82.5 ml/min (Space time = 0.072 g s/mL). Composition of the feed
stream was adjusted as (AcOH/H2O/Ar)=(1.0/2.5/2.0 in molar ratio). Composition of the
product stream leaving the reactor was analyzed using a gas chromatograph (GC, Agilent Model
6890N), which was connected on-line to the reactor exit stream. This gas chromatograph was
equipped with a thermal conductivity detector and Porapak-S column. A condenser was placed
between the tubular reactor outlet and gas chromatograph to collect the condensable
components of the reactor exit stream. Analysis of this condensed phase indicated that it
contained mainly water and small amount of unreacted acetic acid in some experiments.
Conversions of AcOH and product selectivity values were evaluated basing on the
analysis of the product stream compositions and the results were checked by a carbon balance.
AcOH Conversion:
XAcOH=[(AcOHin-AcOHout)/AcOHin]x100
Selectivity of Hydrogen:
SH2=[(Moles of H2 produced)/(4xMoles of AcOH reacted)]x100
Selectivities of CO, CO2, CH4:
Si= [(Moles of compound i produced)/(2xMoles of AcOH reacted)]x100
Moles of AcOH reacted= (AcOHin-AcOHout) =(1/2)(CH4 + CO2 + CO)in product stream
3. RESULTS AND DISCUSSIONS

3.1. Characterization Results
Low angle XRD patterns of the synthesized silicate structured catalyst support clearly
showed that the ordered pore structure of MCM-41 was formed (Fig. 1a). The main XRD peak
observed at a 2θ value of 2.40o (corresponding to d100) and the three reflection peaks at 2θ
values 4.08o, 4.66o and 6.19o is a clear indication of MCM-41 structure. However, the low angle
XRD pattern of zirconia incorporated MCM-41 (25Zr-MCM-41) was quite different, indicating
that the ordered pore structure was not formed (Fig. 1a).
Nitrogen adsorption/desorption isotherm of MCM-41 support material also supported
the formation of the ordered pore structure in the mesopore range (Fig. 2a). The shape of the
isotherm (Type IV) supported the formation of an ordered mesoporous structure. On the other
hand, N2 adsorption/desorption behavior of zirconia incorporated MCM-41 did not show the
characteristic hysteresis loop of MCM-41 type material having ordered pore structure. These
results proved that, the zirconia incorporated silicate structured material (25Zr-MCM-41) did
not contain an ordered pore structure but it was mesoporous.
XRD patterns and nitrogen adsorption/desorption isotherms of Ni impregnated MCM-
41 type catalysts (Fig 1b and Fig 2b, respectively) showed that the ordered pore structure of
MCM-41 was not significantly altered as a result of Ni impregnation. However, significant
changes were observed both in the XRD patterns (Fig 1c) and N2 adsorption/desorption
isotherm (Fig. 2b) of the Cu impregnated materials. Ordered mesopore structure of MCM-41
was significantly deformed as a result of Cu impregnation. Sintering and strong interaction of
copper with the support material during the calcination/reduction steps is considered as the main
reasons of these structural changes. As it was also reported in the literature, copper was expected
to sinter and form agglomerates at temperatures higher than of 350oC [52-55]. Pore-size
distribution of 5Cu@MCM-41 was also found to be quite different than the pore-size
distributions of MCM-41 and Ni impregnated MCM-41 materials (Fig. 3). Formation of a bi-
modal pore-size distribution in the Cu impregnated MCM-41 is also considered to be due to
strong interaction of copper with SiO2 via OH groups on the surface of the support. Formation
of Cu-Si alloy was also reported in the literature at the calcination temperature of this material
[56].
Some physical properties of the synthesized catalysts are given in Table 1. The surface
area, average pore diameter and pore volume of MCM-41 decreased from 1213 m2/g to 761
m2/g, from 2.73 nm to 2.45 nm and from 1.15 cc/g to 0.64 cc/g, respectively, as a result of
impregnation of 1% Ni. These results indicated that most of the Ni was penetrated into the

mesopores and deposited on the pore surfaces of this catalyst. This was also supported by the
shift of pore-size distribution to smaller pores as a result of impregnation of Ni (Fig.3). It was
quite surprising to observe that increase of Ni content of the catalyst from 1% to 5-10% did not
cause further decrease in pore volume and pore diameter values of the synthesized materials.
Even some increase was observed in the surface area of the impregnated MCM-41 as a result
of increase of Ni content from 1% to 5-10%. These results indicated that some fraction of nickel
was deposited on the external surface of MCM-41 for 5Ni@MCM-41 and 10Ni@MCM-41.
XRD analysis results of the synthesized materials also supported these conclusions (Fig. 4).
The nickel peaks observed in the XRD patterns of 1Ni@MCM-41 were extremely small. This
result indicated that the crystal size of nickel was very small and nickel was very well dispersed
within the pores of MCM-4l. However, quite sharp Ni0 peaks were observed at 2θ values of
44.62o, 51.92o and 76.54o in the XRD patterns of the reduced catalysts containing 5% and 10%
nickel (5Ni@MCM-41 and 10Ni@MCM-41). Also, the absence of the peaks at 2θ values of
o
o
37.3 , 43.3 and 62.8o in these XRD patterns indicated that there were no NiO clusters
detectable by XRD in the reduced materials. The crystal sizes of nickel, estimated from the
XRD patterns using the Scherrer equation, were 15 nm and 22 nm for 5Ni@MCM-41 and
10Ni@MCM-41, respectively. These crystal sizes are larger than the average pore diameter of
MCM-41, supporting that significant fraction of nickel was deposited on the external surface
of the MCM-41 particles for these two catalytic materials. SEM-EDX mappings of Ni
impregnated MCM-41 type catalysts also clearly indicated well dispersion of this metal within
pore network (Fig. 5).
In the case of Cu impregnated MCM-41 (5Cu@MCM-41), decrease of surface area and
the pore volume values were more significant than the case observed for 5Ni@MCM-41 (Table
1). Low angle XRD patterns (Fig 1c) and nitrogen adsorption isotherms of 5Cu@MCM-41
supported the conclusion that the ordered pore structure of MCM-41 was significantly deformed
as a result of copper impregnation. As discussed above, sintering and of copper at temperatures
over 350oC and its interaction with the silicate structured support are considered as the main
reasons of this deformation. Wide angle XRD patterns of the 5Cu@MCM-41 showed typical
peaks corresponding to metallic Cu clusters at 2θ values of 43.38o, 50.51o and 74.16o (Fig. 4).
Similar peaks were reported by Dong et al [56]. The crystal size of copper was estimated as 20
nm in this catalyst.
Surface area values of Ni and Cu impregnated 25Zr-MCM-41 type catalytic materials
were also sufficiently high for catalytic purposes (Table 1). Their pore size distributions
indicated the presence of pores mainly in the range of 5-200 nm. These materials were also

mesoporous, but did not contain ordered pore structures. ICP-MS analysis of these materials
showed that zirconia was successfully incorporated into the silicate structure material.
As shown in Figure 4, the XRD patterns of Ni or Cu impregnated Zr-MCM-41 catalysts
are consistent with the results of ICP-MS analysis, indicating the presence of zirconia phase. In
the XRD patterns of these catalysts, the characteristic peaks of the tetragonal zirconia phase at
around 30.12o, 34.52o, 50.6o and 59.69o were observed.
Table 1. Physical properties of the synthesized catalysts
Sample
Procedure
Metal content in
Synthesis
Surface
Area,
m2/g
Mean Pore
Diameter,
nm
Pore
Volume,
cc/g
Solution (wt%)
-
MCM-41
Hydrothermal
Cu impregnated
Hydrothermal
Ni impregnated
Hydrothermal
Ni impregnated
Hydrothermal
Ni impregnated
Hydrothermal
One-pot
1213
2.73
1.17
5Cu@MCM-41
5% Cu
1% Ni
5% Ni
10% Ni
-
399
761
850
919
350
0.29
0.64
0.76
0.67
0.87
1Ni@MCM-41
5Ni@MCM-41
10Ni@MCM-41
25Zr-MCM-41
2.45
2.46
2.46
hydrothermal
Ni impregnated
Cu impregnated
5Ni@25Zr-MCM-41
5Cu@25Zr-MCM-41
5% Ni
5% Cu
88
0.48
0.45
110
450000
300000
(a)
25Zr-MCM-41
400000
350000
300000
250000
200000
150000
100000
50000
0
200000
100000
0
0 1 2 3 4 5 6 7 8 910
2θ
0
1
2
3
4
5
6
7
8
9
10
2θ

400000
350000
300000
250000
200000
150000
100000
50000
0
(b)
5Ni@MCM-41
0
1
2
3
4
5
6
7
8
9
10
2θ
400000
350000
300000
250000
200000
150000
100000
50000
0
(c)
5Cu@MCM-41
0
1
2
3
4
5
6
7
8
9
10
2θ
Figure 1. Low angle XRD patterns of (a) MCM-41, 25Zr-MCM-41, (b) 5Ni@MCM-41,
(c) 5Cu@MCM-41
800
MCM-41
25Zr-MCM-41
(a)
700
600
500
400
300
200
100
0
0
0.2
0.4
0.6
0.8
1
P/P0

1Ni@MCM-41
10Ni@MCM-41
5Ni@MCM-41
5Cu@MCM-41
500
450
400
350
300
250
200
150
100
50
(b)
0
0
0.2
0.4
0.6
0.8
1
P/P0
Figure 2. Nitrogen adsorption/desorption isotherms
18
16
14
12
10
8
MCM-41
1Ni@MCM-41
5Ni@MCM-41
10Ni@MCM-41
5Cu@MCM-41
6
4
2
0
1
10
Pore diameter (nm)
Figure 3. Pore-size distributions of MCM-41, 1Ni@MCM-41 5Ni@MCM-41, 10Ni@MCM-
41 and 5Cu@MCM-41.

Figure 4. XRD patterns of (a) reduced Ni based MCM-41, 25Zr-MCM-41, (b) reduced fresh
and spent Cu based MCM-41, 25Zr-MCM-41, (c) spent 5Ni@MCM-41 and 5Ni@25Zr-
MCM-41catalysts.
( ●Ni; ٭Cu; ■ZrO2)

Si (green)
Ni (blue)
Figure 5. SEM image and Si, Ni EDX mappings of Ni impregnated MCM-41.
3.2.Activity Test Results
A series of catalytic activity tests were performed within a reaction period of 3 h at
750°C. The performances of the synthesized catalytic materials were evaluated in terms of
AcOH conversion and selectivity values of H2, CO2, CO and CH4. Before the activity tests,
which were performed with the synthesized catalysts, reaction experiments were made without
loading the catalysts into the quartz reactor, with the same feed composition and with the same
feed flow rate of 82.5 ml/min, at 750oC. In these experiments, close to 60% initial acetic acid
conversion values were obtained, due to thermal decarboxylation reaction. No hydrogen
formation was observed in these tests. Formation of only CO2 and CH4, with close to equal
mole fractions, was observed. These results indicated that the only reaction taking place in the
absence of catalyst was decarboxylation reaction of AcOH.
CH3COOH↔ CH4+CO2
Another set of initial experiments were also performed by packing the reactor with
quartz particles instead of synthesized catalysts. These experiments were performed with the
same feed composition and feed flow rate and at the same temperature of 750oC. The only
products observed in these tests were also CH4 and CO2, supporting the conclusion that the only
reaction that took place in the absence of a catalyst is decarboxylation of acetic acid. However,
the results of these tests indicated lower conversion values of acetic acid (about 22 %) than the
conversion obtained in an empty tubular reactor. This was simply because of lower mean
residence time of reacting species within the fixed bed than the empty tubular reactor. Since,
the inlet volumetric flow rates of the reactants in the experiments performed with the quartz
filled and empty tubular reactors were the same, higher velocity and hence lower residence time
of reactants is expected in the quartz filled system.

Acetic acid conversion values obtained with the MCM-41 supported catalysts are
reported in Fig. 6. As shown in this figure, AcOH conversions obtained with pure MCM-41 and
5Cu@MCM-41 were highly unstable and much lower than the conversion values obtained with
the Ni impregnated materials. In the presence of Cu impregnated catalyst, conversion of AcOH
decreased from 80% to about 40% within a reaction period of three hours. In the case of pure
MCM-41, conversion of AcOH decreased to a value of about 20%, which was about the same
as in the case of the quartz filled reactor. As discussed above, this conversion value was lower
than the conversion of AcOH observed in the absence of any packing in the reactor. This was
simply because of longer residence time of species in the empty tubular reactor than the fixed
bed reactors.
Highest activity was observed with the catalyst containing 5% Ni (5Ni@MCM-41). For
this catalyst, complete conversion of AcOH was observed at initial reaction times. Some
decrease of conversion observed at longer reaction times was considered to be mainly due to t
deactivation of the catalyst as a result of coke formation. Further increase of Ni content of the
catalyst from 5% to 10% did not cause further increase in activity. On the contrary, faster
deactivation of the catalyst was observed with 10Ni@MCM-41, than 5Ni@MCM-41. This was
considered to be due to larger crystal size of Ni in this catalyst. Catalytic performance of the
material containing 1% Ni was not as good as the performance of 5Ni@MCM-41, indicating
that amount of active sites for the steam reforming reaction were not sufficiently high within
this material.
Figure 6. Variation of AcOH conversions over the MCM-41 supported catalysts at 750oC.

Product distributions obtained over the MCM-41 supported catalysts indicated that,
hydrogen selectivity values were very low over MCM-41 and 5Cu@MCM-41, indicating that
the catalytic performances of these materials were poor for the steam reforming reaction of
acetic acid (Fig. 7). However, quite high and rather stable H2 selectivity values were obtained
over the 5Ni@MCM-41 and 10Ni@MCM-41. This result is due to high activity of Ni in steam
reforming reactions by breaking the C-C, C-H and O-H bonds.
MCM-41
5Ni@MCM-41
5Cu@MCM-41
10Ni@MCM-41
1Ni@MCM-41
80
70
60
50
40
30
20
10
0
30
60
90
120
150
180
Time, min
Figure 7. Hydrogen selectivity values over the MCM-41 supported catalytic materials.
Selectivity values of CO2, CH4 and CO reported in Fig.8 also supported the conclusion
obtained from the hydrogen selectivity values reported in Fig.7. With 5Ni@MCM-41 and
10Ni@MCM-41, quite stable CO2 and CO selectivity values of about 42% and 55% were
obtained, respectively. Experimental selectivity values of H2, CO and CO2 were very close the
equilibrium selectivities (71%, 56% and 43%, respectively), which were predicted from
thermodynamic calculations. These catalysts showed very high activity for steam reforming
reaction of AcOH. Methane selectivity was quite low (less than 5%) over these catalysts,
indicating further reforming of any methane formed in this processes with H2O or CO2 (R.5 and
R.6). A separate reaction test performed with a feed stream containing higher H2O/AcOH molar
ratio (30/1) showed that CH4 was completely disappeared in the product stream. In that test,
stable product distributions, composed of 66.7% H2, 30.9% CO2 and small amount of CO
(2.9%), were observed.
Results reported in Fig.8b showed that, quite high methane formation was observed over
5Cu@MCM-41. This result supported the conclusion that copper based catalyst prepared in this

work was not active for the AcOH reforming reaction. Sintering and aggregation of copper
species at temperatures over 350oC are expected to cause irreversible deactivation of copper
based catalysts [52-55]. Nitrogen adsorption-desorption isotherm of 5Cu@MCM-41 catalyst
supported this conclusion, indicating that Cu incorporation decreased the pore volume of
MCM-41 material significantly. Although the XRD patterns of the Cu impregnated materials
synthesized in the present study did not show the characteristic peaks of copper silicate, possible
formation of a Cu-Si alloy and copper phyllosilicate were reported in the literature at the
synthesis and reaction temperatures of the present study [52,53, 55].
Comparison of the AcOH conversion values of the Cu and Ni based catalysts supported
on either MCM-41 or 25Zr-MCM-41 showed that the catalytic performances of the zirconia
incorporated MCM-41 type catalysts were better. As shown in Fig.9, quite stable complete
conversion of AcOH was observed over 5Ni@25Zr-MCM-41 during a reaction period of three
hours. However, some decrease of AcOH conversion was observed with the 5Ni@MCM-41
catalyst, within the same reaction period. Results proved that Zr incorporation improved the
stability of the MCM-41 supported Ni catalyst.
Comparison of hydrogen selectivity values obtained using Ni or Cu impregnated MCM-
41 and 25Zr-MCM-41 catalysts also supported the conclusion that 5Ni@25Zr-MCM-41 gave
the best performance, giving the highest H2 selectivity and the most stable activity (Fig. 10).
Product distributions obtained with 5Ni@25Zr-MCM-41 also supported this conclusion
(Fig.11). Highly stable hydrogen (~60%), CO (~21%) and CO2 (~19%) compositions were
obtained over this catalyst. Absence of CH4 in the product stream is another important
superiority of this catalyst over the MCM-41 supported catalytic materials. The absence of CH4
in the product stream was also reported by Vagia and Lemonidou over a Ni supported ceria-
zirconia catalyst at 750oC [57].
Comparison of the AcOH conversion values obtained with 5Cu@MCM-41 and
5Cu@25Zr-MCM-41 catalysts (Fig. 9) also supported the conclusion that the activity of the Zr
incorporated MCM-41 type catalyst was higher than the activity of MCM-41 supported catalyst.
However, the stability of the 5Cu@25Zr-MCM-41 was not good.

Figure 8. Product selectivities over MCM-41 supported catalysts with a feed stream
composition of AcOH/H2O/Ar=1.0/2.5/2.0: (a) CO2 ; (b) CH4 ; (c) CO

5Ni@MCM-41
5Cu@MCM-41
5Ni@25Zr-MCM-41
5Cu@25Zr-MCM-41
100
80
60
40
20
0
30
60
90
120
150
180
Time, min
Figure 9. Comparison of AcOH conversions obtained with Ni and Cu based catalysts over
MCM-41 and 25Zr-MCM-41 type catalytic materials. (AcOH/H2O/Ar=1.0/2.5/2.0; T=750oC)
Figure 10. Comparison of hydrogen selectivity values of MCM-41 and 25Zr-MCM-41
supported Ni and Cu catalysts.
Figure 11. Product distributions over 5Ni@25Zr-MCM-41 catalyst

All of the catalytic activity tests reported above were performed for a reaction period of
three hours, using 0.1 g catalyst. In order to observe catalyst deactivation at longer times, time-
on stream tests extending to 12 hours were performed using 0.05 g 5Ni@25Zr-MCM-41. Due
to decrease of catalyst amount from 0.1 g to 0.05 g, slight decrease of initial conversion of
AcOH from 100% to 98% was observed in these tests. More importantly, decrease of
conversion of AcOH to about 70% was observed after a reaction period of 12 hours. This was
considered to be caused by deactivation of the catalyst due to coke formation at longer reaction
times. However, product distributions were highly stable during the 12 hour tests.
After the 12 hour activity test of 5Ni@25Zr-MCM-41, spent catalyst was regenerated
at 750oC, using air flow for 90 minutes and then using hydrogen flow for 60 minutes. Catalytic
performance of the regenerated catalyst was tested again at 750oC, using 0.05 g catalyst. After
the regeneration step, initial AcOH conversion reached to about 98%, which was about the same
as it was observed with the fresh catalyst. Decrease of this conversion to about 60% was
observed after a reaction period of four hours, which was again due to coke formation.
However, the product selectivity values observed with the regenerated catalyst were highly
stable within the reaction period of four hours (Fig. 12).
60
50
40
H2
CO
CH4
CO2
30
20
10
0
30
60
90 120 150 180 210 240
Time, min
Figure 12. Product distributions observed with the regenerated 5Ni@25Zr-MCM-41 catalyst at
750oC.
3.3.Coke Formation and Characterization of Spent Catalysts
Thermogravimetric and differential thermal analysis (TG-DTA) and SEM tests were
performed to obtain more information about coke formation and structural properties of spent
catalysts. TGA–DTA was performed in a flow of dry air. Results of TGA-DTA of the spent

catalysts (5Ni@MCM-41, 5Cu@MCM-41 and 5Ni@25Zr-MCM-41) are shown in Figures 13
and 14. Significant weight loss was observed in the TGA of spent catalysts in the temperature
range of 500–800oC. This weight loss is due to combustion of coke deposited on the catalyst.
Results showed more coke formation over Ni impregnated MCM-41 than Cu impregnated
MCM-41. Zr incorporation decreased the crystal size of Ni and Cu in the catalyst (Table 2). In
parallel to this observation, coke deposition over Ni based catalyst was also decreased as a result
of Zr incorporation into the MCM-41 like support material.
Table 2. Crystal sizes of the Ni and Cu in MCM-41 and Zr-MCM-41 catalysts before activity
tests
Catalysts
Crystal Size, nm
5Ni@MCM-41
5Ni@25Zr-MCM-41
5Cu@MCM-41
5Cu@25Zr-MCM-41
15.5
11.4
20.0
15.0
Figure 13. TG analysis of spent 5Ni@MCM-41, 5Cu@MCM-41 and 5Ni@25Zr-MCM-41
catalysts after 3 h of reaction time.
DTA analyses of the spent 5Ni@MCM-41, 5Cu@MCM-41 and 5Ni@25Zr-MCM-41
are reported in Figure 14. As shown in this figure, coke removal due to combustion takes place
in the temperature range of 450-800oC for MCM-41 supported Ni and Cu catalysts. In fact, two
distinct peaks were observed in the temperature ranges of 450-700oC and 750-800oC in the
DTA of spent 5Ni@MCM-41. Amorphous and filamentous carbon was expected to be

combusted at lower temperatures [51]. On the other hand, graphitic carbon was expected to be
burned at temperatures higher than 750oC. This result showed that both filamentous and
graphitic carbons were formed over the MCM-41 supported catalysts. However, in the case of
25Zr-MCM-41 supported catalyst, coke removal was achieved in the temperature range of 450-
670oC, indicating that graphitic carbon was not formed over this material. Sharp XRD peak
observed at 2θ=26o in the XRD pattern of spent 5Ni@MCM-41 (Fig.4) supported the formation
of graphitic carbon over this material.
Figure 14. DT analysis of spent 5Ni@MCM-41, 5Cu@MCM-41, and 5Ni@25Zr-MCM-41
catalysts after 3 h of reaction time.
Comparison of SEM images of fresh and spent 5Ni@MCM-41 catalyst also showed the
formation of carbon filaments on the surface of used 5Ni@MCM-41 after the 3 h of reaction
test (Figure 15). Analysis of the dimensions of these carbon filaments showed that their
diameters were in the range of 11−26 nm.

a)
b)
Figure 15. SEM images of a) 5Ni@MCM-41 b) used 5Ni@MCM-41
4. CONCLUSIONS
A new mesoporous catalyst support material was synthesized by the incorporation of Zr
into the MCM-41 structure through a one-pot procedure. Ni impregnated 25Zr-MCM-41 was
shown to give highly stable performance in steam reforming of acetic acid. Experimental results
obtained for steam reforming acetic acid at 750oC over Ni impregnated MCM-41 and the

zirconia incorporated MCM-41-like mesoporous material gave very high AcOH conversions
and hydrogen yields. Product compositions obtained with these catalysts were shown to be close
to the equilibrium compositions predicted from thermodynamic analysis of the system. It was
concluded that incorporation of zirconia into the structure of MCM-41 through a one-pot
procedure enhanced the activity, as well as the stability of the Ni impregnated catalysts for
steam reforming of acetic acid. Minimization of CH4 in the product stream was an added
advantage of this catalyst. Amount of impregnated Ni was also shown to be quite important in
terms of activity and stability of the synthesized catalytic materials. Optimum Ni content of the
catalysts was found as 5%. Catalytic performance of the material containing 10% Ni was not
as good as the catalyst containing 5% Ni, just because of the formation of larger Ni crystals.
Another advantage of zirconia incorporated MCM-41 support over MCM-41 is that graphitic
carbon was not formed over 5Ni@25Zr-MCM-41. Coke formed over this material was easily
removed from the catalyst surface by combustion at temperatures lower than 670oC. It was
concluded that 5Ni@25Zr-MCM-41 was a highly promising catalyst for hydrogen production
through steam reforming of acetic acid.
Catalytic performances of Cu impregnated silicate structured mesoporous materials
MCM-41 and 25Zr-MCM-41 were poor in terms of steam reforming reactions. Large fractions
of CH4 and CO2 present in the product streams indicated the occurrence of decarboxylation
reaction of AcOH rather than the reforming reactions in the presence of Cu impregnated
catalysts. Hydrogen yield values were also quite low over the Cu impregnated materials.
Sintering of copper species and strong interaction of Cu with the silicate structured support
caused catalyst deactivation, as well as deformations in the ordered pore structure of the support
materials.
5. ACKNOWLEDGMENT
Financial support from TUBITAK (No: 214M578) for collaborative research between Gazi
University, Middle East Technical University and Slovenian National Academy of Chemistry,
Gazi University Research Fund (06/2017-03) and TUBA are gratefully acknowledged. The
authors also thank to the Central Laboratory of METU for the characterization results of the
synthesized materials.
6. REFERENCES
1. T. Dogu, D.Varisli, Turk. J. Chem. 31 (2007) 551-567.
2. J.X. Chen, J.M. Sun, Y. Wang, Ind. Eng. Chem. Res. 56 (2017) 4627-4637.

3. H. Arbag, S. Yasyerli, N. Yasyerli, G. Dogu, T. Dogu, Appl. Catal. B: Environ. 198 (2016)
254-265.
4. W. Nabgan, T.A.T Abdullah, R. Mat, B. Nabgan, Y. Gambo, M. Ibrahim, A. Ahmad, A.A.
Jalil, S. Triwahyono, I. Saeh, Renew. Sustainable Energy Reviews. 79 (2017) 347-357.
5. A.N. Fatsikostas, D.I. Kondarides, X.E. Verykios, Catal Today 75 (2002) 145-155.
6. S. Gunduz, T. Dogu, Appl. Catal. B: Environ. 168-169 (2015) 497-508.
7. A.A. Lemonidou, P. Kechagiopoulos, E. Heracleous, S. Voutetakis, Steam reforming of
bio-oils to hydrogen. In: Kostas Triantafyllidis Angelos Lappas Michael Stöcker (Eds.), The
Role of Catalysis for Sustainable Production of Bio-fuels and Bio-chemicals, New York:
Elsevier Science, 2013, pp. 467-493.
8. A.V. Bridgwater, J. Anal. Appl. Pyrol. 51 (1999) 3-22.
9. A.C. Basagiannis, X.E.Verykios, Int. J. Hydrogen Energy. 32 (2007) 3343-3353.
10. F. Bimbela, J. Abergo, R. Puerto, L. Garcia, J. Arauzo, Appl. Catal. B: Environ. 209 (2017)
346-357.
11. C. Branca, P. Giudicianni, C. Blasi, Ind. Eng. Chem. Res. 42 (2003) 3190–3202.
12. F. Guilherme, E. Nogueira, G.M. Assaf, H. Carvalho, E. Assaf, Appl. Catal. B: Environ.
160-161 (2014) 188-199.
13. R. Trane-Restrup, A.D. Jensen, Appl. Catal. B: Environ. 165 (2015) 117-127.
14. Z. Li, X. Hu, L. Zhang, G. Lu, J Molecular Catal. A: Chem. 355 (2012) 123-133.
15. F. Bimbela, M. Oliva, J. Ruiz, L. Garcia, J. Arauzo, J. Anal. Appl. Pyrol. 79 (2007) 112-
120.
16. X. Hu, G. Lu, Appl. Catal. B: Environ. 88 (2009) 376-385.
17. K.A. Resende, C.N. Avila-Neto, R.C. Rabelo-Neto, F.B. Noronha, C.E. Hori, Catal. Today.
242 (2015) 71-79.
18. F. Zhang, M. Wang, L. Zhu, W. Shurong, J. Zhou, Int. J. Hydrogen Energy. 42 (2017) 3667-
3675.
19. S. Goicoechea, E. Kraleva, S. Sokolov, M. Schneider, M.M. Pohl, N. Kockmann, H. Ehrich,
Appl. Catal. A:Gen. 514 (2016) 182-191.
20. A.C. Basagiannis, X.E. Verykios, Appl. Catal. A:Gen. 308 (2006) 182-193.
21. A.C. Basagiannis, X. E. Verykios, Appl. Catal. B: Environ. 82 (2008) 77–88.
22. S. Goiceechea, E. Kraleva, H. Ehrich, Catal. Letters. 147 (2017) 1403-1410.
23. R.B. Da Silva, S.T. Brando, A. Lucotti, M.S. Tommasini, C. Castiglioni, G. Groppi, A.
Beretta, Catal. Today. 289 (2017) 162-172.
24. G.Y. Chen, J.Y. Tao, C.X. Liu, B.B Yan., W.Q. Li, X.P. Li, Renew. Sustainable Energy

Reviews. 79 (2017) 1091-1098.
25. A. Lemonidou, E.C. Vagia, J.A. Lercher, ACS Catal. 3 (2013) 1919-1928.
26. K. Takanabe, K. Aika, K. Inazu, T. Baba, K. Seshan, L. Lefferts, J. Catal. 243 (2006) 263-
269.
27. Q. Wang, S. Wang, X. Li, L. Guo, Bio Resources. 8 (2013) 2897-2909.
28. F. Bossola, C. Evangelist, M. Allieta, R. Psaro, S. Recchia, V. Del Santo, Appl. Catal.
B:Environ. 181 (2016) 599-611.
29. S. Goicoechea, E. Kraleva, S. Sokolov, M. Schneider, M.M. Pohl, N. Kockmann, H. Ehrich,
Appl. Catal. A: Gen. 514 (2016) 182-191.
30. X. Hu, L. Zhang, G. Lu, Appl. Catal. A: Gen. 427- 428 (2012) 49-57.
31. R. Hu, C. Yan, X. Zheng, H. Liu, Z. Zhou, Int. J. Hydrogen Energy. 38 (2013) 6033-6038.
32. P.G.M. Assaf, F.G.E. Nogueira, E.M. Assaf, Catal. Today. 213 (2013) 2-8.
33. G.Y. Chen, J.Y. Tao, C.X. Liu, B.B. Yan, W.Q. Li, X.P. Li, Int. J. Hydrogen Energy. 42
(2017) 20729-20738.
34. S.C.M. Mizuno, A.H. Braga, C.E. Hori, J.B.O. Santos, J.M.C. Bueno, Catal. Today. 296
(2017) 144-153.
35. F. Zhang, N. Wang, L. Yang, M. Li, L. Huang, Int. J. Hydrogen Energy. 39 (2014) 18688-
18694.
36. K.K. Pant, P. Mohanty, S. Agarwal, A.K. Dalai, Catal. Today. 207 (2013) 36-43.
37. M. Turco, G. Bagnasco, C. Cammarano, P. Senese, U. Costantino, M. Sisani, Appl. Catal.
B: Environ. 77 (2007) 46–57.
38. V. Deshmane, R. Abrokwah, D. Kuila, Int. J. Hydrogen Energy. 40 (2015) 10439-10452.
39. B.P. Karaman, N. Cakiryilmaz, H. Arbag, N. Oktar, G. Dogu, T. Dogu, Int. J. Hydrogen
Energy, 42 (2017) 26257-26269.
40. X. Hu, G.X. Lu, Appl. Catal. B: Environ. 99 (2010) 289-297.
41. X. Hu, L. Zhang, G. Lu, Int. J. Hydrogen Energy. 41 (2016) 13960-13969.
42. P. Mohanty, M. Patel, K.K. Pant, Bioresource Technol. 123 (2012) 558-565.
43. N. Iwasa, T. Yamane, M. Takei, J. Ozaki, M. Arai, Int. J. Hydrogen Energy. 35 (2010) 110-
117.
44. S. Gunduz, T. Dogu, Ind. Eng. Chem. 51 (2012) 8796-8805.
45. H. Arbag, S. Yasyerli, N. Yasyerli, G. Dogu, Int. J. Hydrogen Energy. 35 (2010) 2296–
2304.
46. X. Xu, E. Jiang, Y. Sun, Z. Li, Energy Fuels. 31 (2017) 1678-1688.
47. A. Arslan, S. Gunduz, T. Dogu, Int. J. Hydrogen Energy. 39 (2014) 18264-18272.

48. J. Anderson, C. Fergusson, I. Rodriguezramos, A. Guerreroruiz, J. Catal. 192 (2000) 344–
354.
49. L. Fuxiang, Y. Feng, L. Yongli, L. Ruifeng, X. Kechang, Micropor Mesopor Mat. 101(1-2)
(2007) 250-255.
50. X. Yang, L. Zhou, C. Chen, J. Xu, Mat. Chem. Physics 120 (2010) 42-45.
51. N. Cakiryilmaz, H. Arbag, N. Oktar, G. Dogu, T. Dogu, Int. J. Hydrogen Energy, 43 (2018)
3629-3642.
52. I. Sierra, J. Erena,, A. Aguayo, J. Arandes, J. Bilbao, Appl. Catal. B: Environ. 94 (2010)
108–116.
53. T. Ding, H. Tian, J. Liu, W. Wu, J. Yu, Chinese J. Catal. 37 (2016) 484–493.
54. M.V. Twiggy, M.S. Spencer, Appl. Catal. A:Gen. 212 (2001) 161-174.
55. G. Celik, A. Arinan, A. Bayat, H.O. Ozbelge, T. Dogu, Topics in Catal. 56 (2013) 1764-
1774.
56. X. Dong, X. Ma, H. Xu, Q. Ge, Catal. Sci. Technol. 6 (2016) 4151-4158.
57. E. C. Vagia, A.A. Lemonidou, J. Catal. 269 (2010) 388-396.