The catalytic and physico-chemical properties of Ni/MgF2-MgO catalysts

Article abstract of DOI:10.1016/j.apcata.2012.09.033

The MgF₂-MgO system of different quantitative compositions was tested as a potential support for nickel catalysts to be used for hydrogenation of toluene. The catalytic performance of the system was studied as a function of the MgF₂/

Full text of DOI:10.1016/j.apcata.2012.09.033

Applied Catalysis A: General 449 (2012) 15–22
Contents lists available at SciVerse ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
The catalytic and physico-chemical properties of Ni/MgF2–MgO
catalysts
Michał Zielin´ ski∗
Adam Mickiewicz University, Faculty of Chemistry, Grunwaldzka 6, 60-780 Poznan´, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 17 May 2012
Received in revised form 3 August 2012
Accepted 18 September 2012
Available online 8 October 2012
The MgF2–MgO system of different quantitative compositions was tested as a potential support for nickel
catalysts to be used for hydrogenation of toluene. The catalytic performance of the system was studied
as a function of the MgF2/MgO ratio in the support and the reaction conditions. The hydrogenation was
performed at temperatures ranging from 75 to 225 ◦C on the catalysts activated in hydrogen at different
temperatures (400–550 ◦C). The activities of Ni/MgF2–MgO catalysts were compared to those of Ni/MgF2,
Ni/MgO and Ni/Al2O3. The highest activities were observed in the case of the catalysts supported on
MgF2–MgO (containing 40 mol.% MgO), especially after the activation at 500 ◦C. The catalytic activity of
nickel supported on magnesium oxo-fluoride tested in toluene hydrogenation was much higher than
that of Ni supported on the commonly used supports such as Al2O3 or MgO.
Keywords:
Nickel catalysts
MgF2
MgO
© 2012 Elsevier B.V. All rights reserved.
Magnesium oxo-fluoride
Al2O3
Toluene hydrogenation
1. Introduction
mesoporous support having well-developed porous structure, high
chemical inertness, thermal stability and hardness [29,30]. Unfor-
The performance of nickel catalysts in hydrogenation of aro-
matic hydrocarbons significantly depends on the type of support,
specific surface area of the metal, mechanical strength and ther-
mal stability as well as the compatibility with the nickel phase.
This is why the intensive work is conducted on the synthesis and
modification of the commonly known supports such as Al2O3 [1,2],
SiO2 [3,4], MgO [5,6], active carbon [7,8] and mesoporous mate-
rials [9–12]. The supported nickel catalysts with high dispersion
and loading can be prepared by sol–gel methods, but they usually
require complicated procedures [2,5,13]. Another method for the
synthesis of active catalysts of aromatic hydrocarbon hydrogena-
tion is based on the introduction of nickel onto binary oxide systems
such as SiO2–TiO2 [14,15] or SiO2–Al2O3 [16].
This study concerns catalysts consisting of nickel phase sup-
ported on magnesium fluoride or on MgF2–MgO binary system.
Magnesium fluoride has been used for many years as a support
for oxide and metal phases active and selective in processes impor-
tant from the environmental viewpoint such as NO reduction in the
presence of hydrocarbons as reducing agents [17,18], CO oxidation
[18,19], HDS reaction [20,21], hydrogenation of benzene [22,23]
and toluene [24–26], as well as selective reduction of chloroni-
trobenzene to chloroaniline [27,28]. Magnesium fluoride is a good
tunately, magnesium fluoride obtained by the reaction between
magnesium carbonate and aqueous solution of hydrofluoric acid
has a good thermal stability only up to 400 ◦C and its specific sur-
face area after calcination at 400 ◦C is not greater than 40 m2 g−1
.
Calcination at higher temperatures results in a significant decrease
in the specific surface area. However, when magnesium fluoride
was modified by combining it with magnesium oxide to form the
MgF2–MgO binary system, surface area considerably greater than
that of single components MgF2 or MgO was created and thermal
stability of the binary support appeared to be much better [31]. The
so-obtained support has been used for iridium catalysts active for
toluene hydrogenation to methylcyclohexane [32] and CO oxida-
tion [33]. However, the sol–gel method proved to be complicated
and expensive. One of the aims of this study was to simplify the
method of preparation of MgF2–MgO supports by applying the
direct reaction between hydrofluoric acid and basic magnesium
carbonate. The systems obtained in this way were used as supports
for the nickel phase active in toluene hydrogenation.
To the best of our knowledge this is the first report on the
potential application of MgF2–MgO system (obtained by the direct
reaction of hydrofluoric acid with basic magnesium carbonate) as
the catalyst support. The binary systems obtained by employing the
above direct reaction were tested as supports for the active nickel
phase. Nickel catalysts containing 5 wt.% of the active phase were
prepared by the impregnation method and supported on MgF2,
MgO and a mixed system of MgF2–MgO and tested for toluene
∗
Fax: +48 61 829 1505.
E-mail address: mardok@amu.edu.pl
0926-860X/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2012.09.033

16
M. Zielin´ski / Applied Catalysis A: General 449 (2012) 15–22
hydrogenation. The activity of the new catalysts for toluene hydro-
genation has been compared with that of nickel catalysts supported
on Al2O3, commonly used for this purpose.
at 2ꢀ = 36.92◦, 42.92◦, 62.29◦ and for MgF2 at 2ꢀ = 35.21◦, 53.48◦,
60.60◦.
2.2.3. Thermogravimetric analysis
2. Experimental
Thermogravimetric analysis for supports and nickel catalysts
was performed for fresh (uncalcined) samples in the temperature
range of 30–900 ◦C using a differential Thermoanalyzer Setaram
TGA equipped with a TG measurement unit. The experiments were
performed under nitrogen flow (99.995 % purity) at a heating rate
2.1. Catalyst preparation
2.1.1. Synthesis of supports
of 5 ◦C min−1
.
The MgF2 and MgF2–MgO supports with different MgF2
contents were obtained in the reaction of basic magnesium car-
bonate (4MgCO3·Mg(OH)2·5H2O) powder with controlled amounts
of 40 wt.% aqueous solution of hydrofluoric acid (both from
POCH–Polish Chemicals Reagents). Hydrofluoric acid solution was
used in the amounts enabling to result in 0, 40, 60 and 100 mol.%
of MgF2 content in the binary support. The resulting dense gels
of MgF2–Mg(OH)2 were subjected to ageing for 40 h at room tem-
perature under stirring, followed by drying at 80 ◦C for 24 h and
calcining under the air flow at 500 ◦C for 4 h. The MgF2–MgO sam-
ples were labelled as xMF, where x is the mol.% of MgF2. The MgF2
support was denoted as MF. The MgO support (denoted as MO) was
obtained by thermal decomposition of 4MgCO3·Mg(OH)2·5H2O at
500 ◦C for 4 h.
2.2.4. Temperature programmed reduction – TPR-H2
The temperature programmed reduction experiments were car-
ried out with an ASAP ChemiSorb 2705 (Micromeritics). Portions of
100 mg of a fresh catalyst were reduced in a mixture of 10 vol.% H2
in Ar (99.999 %, Linde) at the flow rate of 30 cm3 min−1 and heating
rate of 10 ◦C min−1 up to 800 ◦C.
2.2.5. The determination of metal dispersion by hydrogen
chemisorption
Prior to hydrogen chemisorption, the fresh catalysts were
reduced with H2 at 500 ◦C for 2 h. The reduced samples were placed
in an ASAP 2010C sorptometer and were evacuated for 15 min at
room temperature and then at 350 ◦C for 60 min, followed by addi-
tional reduction in hydrogen flow (40 cm3 min−1) at 350 ◦C and
evacuation for 120 min at 350 ◦C.
Hydrogen chemisorption measurements were carried out at
35 ◦C. Nickel dispersion was calculated from total amount of
chemisorbed hydrogen.
Aluminium oxide was prepared by the reaction of aluminium
isopropoxide with the excess of water. The resulted aluminium
hydroxide was washed out to remove isopropyl alcohol, dried at
110 ◦C for 24 h and calcined at 500 ◦C for 4 h.
2.1.2. Synthesis of nickel catalysts
Nickel was deposited on MgF2, MgO and MgF2–MgO supports
by wet impregnation using aqueous solution of nickel(II) acetate
tetrahydrate (98%, Aldrich). The supports were mixed with a solu-
tion containing a desired amount of nickel(II) acetate tetrahydrate
and allowed to stand for 2 h with occasional gentle shaking. The
synthesis was carried out using such an amount of the nickel
precursor that after the pretreatment the final catalyst contained
5 wt.% of Ni. After evaporation of water in a rotary vacuum evapo-
rator at 90 ◦C, the catalysts were dried at 110 ◦C for 24 h. They were
denoted as fresh catalysts – Ni/support-f.
The Ni/Al2O3 catalyst was prepared by impregnating the sup-
port with an aqueous solution of nickel(II) acetate tetrahydrate in
the same way as described for the catalysts mentioned above.
The catalysts were labelled as Ni/support-x, where x is the reduc-
tion temperature.
Metallic surface area S was calculated using the following equa-
tion [34]:
vm · NA · n · am · 100
S =
(m2 g−1
)
Ni
22414 · m · wt.
where vm is expressed in cm3, NA is Avogadro’s number
(6.022 × 1023 mol−1), n is the chemisorption stoichiometry, am is
the surface area (m2) occupied by a metal atom, m is the sample
mass (g), wt. (%) is the metal loading.
The dispersion of active phase can be calculated from the for-
mula:
S · M
D =
am · NA
where S is the metallic surface area, M is nickel atomic weight, NA
is Avogadro’s number and am is the surface covered by one nickel
atom.
2.2. Catalyst characterization
2.2.1. Determination of surface area and porosity
The Brunauer–Emmet–Teller (BET) surface areas (SBET) were
determined by N2 adsorption at −196 ◦C using a Micromeritics
ASAP 2010 sorptometer. Total pore volume and average pore radius
were determined by the Barrett–Joyner–Halenda (BJH) method.
2.3. Catalytic test – hydrogenation of toluene
Toluene hydrogenation was performed at atmospheric pres-
sure using a fixed-bed flow reactor and H2 as a carrier gas. Fresh
dried catalyst (0.05 g) was placed in the reactor and reduced in situ
for 2 h in a flow (100 cm3 min−1) of pure hydrogen (99.99%, pur-
chased from Messer) in the temperature range of 400–550 ◦C
prior to starting the reaction. The reduction temperature was
included in the catalyst code. The reaction mixture was prepared
by passing hydrogen (50 cm3 min−1) through a saturator filled with
toluene. The concentration of toluene in the feed was stable and
amounted to 0.45 ␮mol cm−3. The catalysts were heated at the rate
of 10 ◦C min−1 in a flow of pure hydrogen and the catalytic activ-
ities were measured at temperatures between 75 ◦C and 225 ◦C in
several steps over the same catalyst. The reaction products were
analysed on a gas chromatograph equipped with a capillary column
RESTEK-MXT-1. The catalytic activity was expressed as turnover
frequency (TOF), s−1 (in moles of toluene reacted per surface Ni
2.2.2. X-ray diffraction analysis
The X-ray powder diffraction was performed on a Bruker AXS
D8 Advance diffractometer with Ni-filtered CuK␣ radiation over
the 2ꢀ range of 20–80◦.
The compositions of MgF2–MgO supports were determined
from XRD data for the samples calcined at 900 ◦C for 4 h (in order to
develop well crystallized phases of MgF2 and MgO). The contents
of MgO and MgF2 were determined on the basis of comparison of
intensities of reflections in the diffraction patterns of the samples
studied with those in the diffraction patterns of the mechanical
mixtures of MgF2 and MgO calcined at 900 ◦C, containing 15, 30
or 70 mol.% of MgO. The calculations were made with the use of
DQuant program, taking into account the XRD reflections for MgO

M. Zielin´ski / Applied Catalysis A: General 449 (2012) 15–22
17
Table 1
Physical characterization of supports and nickel catalysts.
Sample codea
MgF2 content,b mol. %
Pretreatment conditions
BET surface area,c m2 g−1
Total pore volume, cm3 g−1
Average pore diameter, nm
MO
0
43.6
62.8
100
Calcination (air, 4 h, 500 ◦C)
Calcination (air, 4 h, 500 ◦C)
Calcination (air, 4 h, 500 ◦C)
Calcination (air, 4 h, 500 ◦C)
142.7
85.2
51.0
32.0
0.29
0.36
0.30
0.17
8.19
16.83
23.79
21.26
40MF
60MF
MF
Al2O3
–
Calcination (air, 4 h, 500 ◦C)
198.2
0.41
8.27
Ni/MO-500
Ni/40MF-500
Ni/60MF-500
Ni/MF-500
0
43.6
62.8
100
Reduction (H2, 2 h, 500 ◦C)
Reduction (H2, 2 h, 500 ◦C)
Reduction (H2, 2 h, 500 ◦C)
Reduction (H2, 2 h, 500 ◦C)
152.4
123.9
69.0
1.09
0.62
0.45
0.18
28.59
20.00
26.31
28.49
27.8
Ni/Al2O3-500
–
Reduction (H2, 2 h, 500 ◦C)
196.3
0.39
7.95
a
See Section 2 for labelling of supports and catalysts.
b
Calculated with the use of DQuant program taking into account the XRD reflections for MgO at 2ꢀ = 36.92◦, 42.92◦, 62.29◦ and for MgF2 at 2ꢀ = 35.21◦, 53.48◦, 60.60◦.
The method of calculation is specified in Section 2.
c
The Brunauer–Emmet–Teller surface areas were determined by N2 adsorption at −196 ◦C using a Micromeritics ASAP 2010 sorptometer. The conditions of measurements
are specified in Section 2.
atoms in 1 s) or as apparent rate (rt) calculated by the following
3.1.2. X-ray diffraction analysis
equation:
The catalysts were subjected to XRD study to determine the
content of MgO in the mixed supports MgF2–MgO (Fig. 1a) and to
identify Ni metal phase present in the catalysts prepared (Fig. 1b).
After calcination at 500 ◦C, the supports studied revealed the
presence of crystalline MgF2 and MgO phases (Fig. 1a), although
well-crystallized MgF2 and MgO phases were detected only after
the calcination at 900 ◦C. As follows from the diffraction patterns
of MgF2–MgO systems after calcination at 900 ◦C, MgF2 and MgO
do not react but form separate crystalline phases. The diffrac-
tion patterns did not show the presence of undecomposed basic
magnesium carbonate. MgO contents determined by XRD were
slightly higher than the desired values and equalled to 43.6 and
62.8 mol.%, while the intended values were 40 and 60 mol.%, respec-
tively (Table 1). As follows from our results, the proposed method
of magnesium oxo-fluoride support synthesis ensures a reasonably
good control of MgO content in the samples.
ꢀ
ꢁ
FYC
N
molTl
molNi min
rt =
where F is the total flow rate of feed (cm3 min−1); Y is the frac-
tional conversion; C is the concentration of toluene in the feed
(molTl cm−3) and N is the nickel content (molNi) in a sample.
Surface Ni atoms were determined by hydrogen chemisorption.
Catalytic measurements carried out on pure supports MgF2, MgO
or MgF2–MgO and Al2O3 proved the inactivity of the supports for
the reduction of toluene.
3. Results and discussion
The same method was used to identify the Ni phase present in
the reduced catalysts (Fig. 1b). The measurements were performed
for the catalysts reduced with hydrogen at 500 ◦C. Only in the case
of the Ni/MF-500 catalyst, the weak reflections assigned to metal-
lic nickel were detected (Fig. 1b). The XRD pattern of Ni/MF-500
catalyst showed two distinct reflections at 2ꢀ = 44.59◦ (1 1 1) and
51.90◦ (2 0 0) characteristic of metallic nickel of the fcc structure
[26,35]. The absence of reflections assigned to metallic nickel in the
XRD patterns of Ni/60MF-500, Ni/40MF-500 and Ni/MO-500 sug-
gests the amorphous state or well-dispersed nickel particles. The
intensity of the diffraction reflections at ∼43.0◦ for Ni/40MF-500
and Ni/MO-500 was higher than that in the diffraction patterns of
pure supports 40MF and MO which at the absence of the diffraction
lines assignable to metallic nickel can testify for the formation of
a solid solution of NiO and MgO [6,36]. These phases are difficult
to distinguish owing to their similar crystal structures [37]. More-
over, the relatively low chemisorption of hydrogen (Table 2) at the
absence of XRD reflections assignable to nickel suggest that nickel
particles are surrounded by magnesium oxide.
3.1. Catalyst characterization
3.1.1. Determination of surface area and porosity
The textures of supports after calcination at 500 ◦C and nickel
catalysts after reduction at 500 ◦C were characterized by the low-
temperature nitrogen adsorption (Table 1). In the same Table data
were given on the commonly used alumina support. According to
the results presented in Table 1, the content of magnesium oxide
has a considerable effect on the surface area of the MgF2–MgO sys-
tem. The surface area of the supports containing 43.6 and 62.8 mol.%
MgO was greater than that of pure MgF2 by a factor of two or
three, respectively. In general, the surface area of mixed supports
increased with increasing amount of MgO introduced. The BET sur-
face area of MO was almost five times greater than that of MF. All the
samples obtained were mesoporous with the mean pore diameters
ranging from 8 to 24 nm. The impregnation of MgF2–MgO, MgF2
and MgO supports with aqueous solution of nickel(II) acetate, fol-
lowed by evaporation of water, drying and reduction in hydrogen
(2 h at 500 ◦C), resulted in an increase in the specific surface areas
of all catalysts containing MgO in the support. The greatest increase
in the specific surface area by almost 50% compared to that of the
support free of the active phase was noted for samples 40MF and
60MF. Only in the case of Ni/MF-500 a small decrease in the spe-
cific surface area was found after the active phase introduction. The
same was observed after the impregnation of alumina support with
aqueous solution of nickel(II) acetate. However, the changes were
smaller and related to the blocking of some pores by the deposited
nickel phase.
3.1.3. Thermogravimetric analysis
The synthesis of MgF2–MgO supports by using the reaction of
basic magnesium carbonate with controlled amounts of hydroflu-
oric acid involves the formation of MgF2 and leaves some unreacted
basic magnesium carbonate. Upon thermal treatment the latter
decomposes to MgO. Thermogravimetric study revealed that all
MgO-containing supports give three mass loss maxima (Fig. 2a).
The first two, recorded also for MgF2, are related to the release of
physisorbed water (in the range of 50–100 ◦C, with a maximum

18
M. Zielin´ski / Applied Catalysis A: General 449 (2012) 15–22
Fig. 1. X-ray diffraction patterns of supports treated at 500 and 900 ◦C (a) and supports and nickel catalysts after reduction at 500 ◦C (b) in the range of Ni signals (41–60
2ꢀ◦).
at 75 ◦C) and the desorption of chemisorbed water or the decom-
position of Mg(OH)2 (in the range 210–330 ◦C, with a maximum
at 274 ◦C). The mass loss with a maximum at 274 ◦C recorded for
40MF and 60MF supports and for MO support (obtained by ther-
mal decomposition of basic magnesium carbonate) is related to
Mg(OH)2 decomposition to MgO. The mass loss above 400 ◦C is
interpreted as the result of the decomposition of unreacted mag-
nesium carbonate, taking place in the range of 330–570 ◦C [38] and
dehydroxylation of the surface [39,40].
Thermogravimetric study was also carried out for fresh catalysts
– Ni/support-f (Fig. 2b). Only for Ni/MO catalyst a strong mass loss
signal was observed and related to the decomposition of Mg(OH)2
with a maximum at 299 ◦C, which is a consequence of MgO partial
hydrolysis during impregnation procedure. Much smaller mass loss
signals with maxima shifted towards higher temperatures by about
30 ◦C were recorded for Ni/40MF-f and Ni/60MF-f, and this fact can
be a result of a smaller content of MgO in the supports. Magnesia in
these supports undergoes a partial hydrolysis to Mg(OH)2 by reac-
ting with the nickel precursor deposited from aqueous solution, but
the presence of hydrophobic MgF2 in these supports significantly
inhibits the process of hydrolysis. The signal was not recorded for
the catalyst supported on MgF2. Additional mass loss at 395 ◦C
probably corresponds to the decomposition of nickel(II) acetate
terahydrate (Fig. 2b). The shift to higher temperatures, from 355
to 395 ◦C, can be explained by the inhibition of the decomposition
of nickel precursor due to the interaction with the support. In this
temperature range, some contribution from the decomposition of
magnesium carbonate, formed during drying of the catalysts as a
result of the reaction between Mg(OH)2 and CO2 from the air, is
also possible. This could explain a small increase in the area of the
relevant peak with increasing MgO content in the support.
According to X-ray powder diffraction and thermogravimetric
results, after reduction at 500 ◦C the supports were characterized by
well-developed crystalline structure. The above temperature was
also found to be sufficient for the complete decomposition of the
active phase precursor.
3.1.4. Temperature programmed reduction – TPR-H2
Additional information on the state of the supported oxide
phases has been provided by the temperature-programmed reduc-
tion with hydrogen. Studies of TPR-H2 were carried out on samples
of fresh catalysts dried at 110 ◦C. Fig. 3 presents the reduction
Table 2
H2-chemisorption data and TOFs of toluene hydrogenation on Ni/support-500 catalysts.
−1
Catalyst codea
Total H2 adsorbed, 10−5 [mol gc−a1t
]
Ni dispersion, %
Ni surface area, m2
g
Mean particle size,b nm
TOF,c s−1
Ni
Ni/MO-500
Ni/40MF-500
Ni/60MF-500
Ni/MF-500
2.36
4.31
4.66
3.78
5.6
10.1
10.9
8.9
37.07
67.49
73.09
59.24
18.2
9.9
9.2
0.04
0.07
0.08
0.07
11.4
Ni/Al2O3-500
0.71
1.7
11.12
60.6
0.02
a
See Section 2 for labelling of supports and catalysts.
Determined by H2 adsorption.
b
c
Turnover frequency (TOF, s−1) was calculated for surface nickel atoms determined by hydrogen chemisorption. Reaction conditions: 125 ◦C, cTl = 0.45 ␮mol cm−3, activity
after 1 h.

M. Zielin´ski / Applied Catalysis A: General 449 (2012) 15–22
19
profiles of the Ni/support-f catalysts. The peak observed in the
first step at ∼370 ◦C is attributed to the reduction of Ni2+ ions
weakly bonded to the support. Black colour of the reduced mate-
rial suggests the reduction of Ni-species to Ni0. For the catalysts
Ni/60MF-f, Ni/40MF-f and Ni/MO-f the second peak appears at a
higher temperature (∼500 ◦C). The intensity of this peak increases
with increasing MgO content and simultaneously the intensity of
the peak with the maximum at 370 ◦C decreases. Probably, the high
temperature reduction peak originates from the reduction of Ni2+
located inside MgO to form NiO–MgO solid solution [41,42]. The
presence of a solid solution NiO–MgO is also indicated by the ear-
lier discussed XRD results. The amount of hydrogen consumed in
the temperature range of 300–600 ◦C was used for the calculation
of extent of nickel reduction in the catalyst. For Ni/support sam-
ple, the latter is equal to ∼95%, which means that almost all nickel
present in the catalyst can be reduced to the zero valent state.
3.1.5. The determination of metal dispersion by hydrogen
chemisorption
An important parameter characterizing metallic catalysts is the
dispersion of metallic phase. The techniques most frequently used
for the determination of metal dispersion are X-ray powder diffrac-
tion, transmission electron microscopy, and chemisorption. The
latter was employed in the present study (Table 2), because the
amount of chemisorbed hydrogen permits to determine not only
the average size of metallic particles, but also the number of nickel
surface atoms that is required for the calculation of turnover fre-
quency (TOF, s−1) of toluene hydrogenation. We did not carry out
TEM measurements because their interpretation is difficult due
to relatively high electron density of MgF2. On the other hand,
XRD experiments show the presence of reflections originated from
nickel metal only for Ni/MF-500, but the method of estimation
of the crystallite size on the basis of X-ray diffraction patterns
according to the Scherrer equation is charged with a consider-
able error (apparatus line broadening), particularly when the signal
from metallic nickel has a low intensity. Blank experiments proved
that the H2-pretreated magnesium fluoride and magnesium oxide
supports did not adsorb hydrogen. The nickel surface area was cal-
culated assuming the hydrogen adsorption stoichiometry of one
hydrogen atom per one surface nickel atom.
Fig. 2. DTG curves recorded for supports and basic magnesium carbonate (precursor
of MgO) (a); and nickel catalysts and precursor of nickel (b).
The highest dispersions, D = ∼11 and 10%, and the smallest nickel
particle sizes were found for Ni/60MF-500 and Ni/40MF-500 cat-
alysts (reduced at 500 ◦C), respectively. The dispersion was by
10% higher than that obtained for the catalyst supported on pure
MgF2 (D = 8.9%) and by almost 50% higher than that for Ni/MO-500
(D = 5.6%). In the latter case, the low dispersion of nickel (retarda-
tion of hydrogen chemisorption) is a result of Ni species embedding
into magnesia, as indicated both by XRD (the formation of NiO–MgO
solid solution) and TPR-H2 studies (the dominant peak of hydrogen
consumption is that at ∼500 ◦C).
3.2. Catalytic activity
The activation procedure for metallic catalysts is one of the
most important factors influencing their hydrogenation activity,
so determination of the optimum activation conditions is very
important while designing new catalysts. The methods of activa-
tion of supports and catalysts are described in Table 1. At first the
effect of the activation temperature on the catalysts performance
in toluene hydrogenation was studied. Measurements were car-
ried out on fresh catalysts. The supports did not show any activity
in the gas phase toluene hydrogenation, while the reduced Ni-
catalysts were active in this process. Methylcyclohexane was the
only product in all catalytic tests. Samples of Ni/MF-f, Ni/MO-f,
Ni/40MF-f and Ni/60MF-f were reduced in the temperature range of
400–550 ◦C at 50 ◦C intervals. Catalytic tests in the hydrogenation
Fig. 3. TPR-H2 profiles of fresh nickel catalysts.

20
M. Zielin´ski / Applied Catalysis A: General 449 (2012) 15–22
Fig. 4. The effect of the activation temperature on the apparent rate of toluene hydrogenation after 1 h at 125 ◦C over Ni/MF (a), Ni/60MF (b), Ni/40MF (c) and Ni/MO (d)
catalysts. Activation of catalyst: H2 = 100 cm3 min−1, 2 h; Reaction: cTl = 0.45 ␮mol cm−3, H2 = 50 cm3 min−1
.
of toluene to methylcyclohexane were carried out in a fixed-bed
flow reactor at 125 ◦C. Fig. 4a–d shows the hydrogenation activity
of the catalysts as a function of the activation temperature. The
catalytic activity of Ni/MF (Fig. 4a) decreased significantly after
activation at temperatures above 500 ◦C, whereas between 500
and 550 ◦C it was almost two times lower. The catalytic perfor-
mance of nickel catalysts containing MgO in the support (Ni/60MF,
Ni/40MF and Ni/MO, Fig. 4b–d) was different – it increased with
the temperature of activation and reached maximum values (after
in hydrogenation activity for different processes with increasing
number of basic sites has already been reported in several papers
[5,25,43–45].
The second parameter to be optimized was the reaction temper-
ature. Fig. 5 illustrates the effect of the reaction temperature on the
activity of Ni/MF-500, Ni/60MF-500 and Ni/MO-500 catalysts. The
activity increased with a rise in temperature and reached maxima at
125 ◦C, 150 ◦C and 175 ◦C for Ni/60MF-500, Ni/MF-500 and Ni/MO-
500, respectively. The only sample that maintained high activity in
the whole range of 125–175 ◦C was Ni/60MF-500. The reduction in
the catalytic activity at higher temperatures is explained in the lit-
erature by dehydrogenation [46] or cracking of methylcyclohexane
formed [47].
reduction at 500 ◦C) of 0.52 molTl mol−1 min−1, for Ni/60MF-500
Ni
and 0.43 molTl mol−1 min−1, for Ni/40MF-500. The lowest activity
was observed in thNei case of the catalyst supported on pure MgO
(Fig. 4d). Its catalytic activity was low, almost four-times lower
than that of the most active catalyst Ni/60MF-500. A comparison of
the performance of the nickel catalysts on different supports after
reduction at 500 ◦C gives the following series:
Ni/60MF-500 > Ni/40MF-500 > Ni/MF-500 > Ni/MO-500
The catalysts tested in toluene hydrogenation were reduced at
400–550 ◦C, and the maximum activity was observed after the acti-
vation at 500 ◦C for Ni/60MF-500 and Ni/40MF-500. The optimum
temperature of reduction (500 ◦C) correlates well with the TPR-H2
data. As follows from Fig. 3, at this temperature the second stage
of catalyst reduction takes place. Only in the case of Ni/MF sample,
the complete reduction of the active phase proceeds up to 400 ◦C,
so the maximum conversions in toluene hydrogenation over this
catalyst were observed after its reduction at this temperature. Fur-
ther increase in the reduction temperature leads to the active phase
sintering, which results in the hydrogenation activity decrease.
The most active catalysts are the ones on mixed supports and
their activity falls down with increasing MgO content in the sup-
port. This decrease can be explained by growing number of basic
sites of the support with increasing content of MgO. The reduction
Fig. 5. The effect of the reaction temperature on the apparent rate of toluene hydro-
genation after 1 h at each temperature over nickel catalysts. Activation of catalyst:
500 ◦C, H2 = 100 cm3 min−1, 2 h; Reaction: cTl = 0.45 ␮mol cm−3, H2 = 50 cm3 min−1
.

M. Zielin´ski / Applied Catalysis A: General 449 (2012) 15–22
21
surface of the MgF2–MgO binary system is more active for hydro-
genation of toluene to methylcyclohexane than that supported on
single component supports such as MgF2, MgO as well as on the
commonly used Al2O3.
Acknowledgements
Financial support of the projects N N204 141339 and N N204
214140 from the Ministry of Scientific Research and Information
Technology (Poland) is gratefully acknowledged.
References
[1] J. Choi, S. Zhang, J.M. Hill, Catal. Sci. Technol. 2 (2012) 179–186.
[2] P.G. Savva, K. Goundani, J. Vakros, K. Bourikas, Ch. Fountzoula, D. Vattis, A.
Lycourghiotis, Ch. Kordulis, Appl. Catal. B 79 (2008) 199–207.
[3] M. Xue, S. Hu, H. Chen, Y. Fu, J. Shen, Catal. Commun. 12 (2011)
332–336.
Fig. 6. Changes in the activity of Ni/MF-500, Ni/MO-500 and Ni/60MF-500 cat-
alysts in toluene hydrogenation as
function of the reaction time at 150 ◦C.
Activation of catalyst: 500 ◦C, H2 = 100 cm3 min−1, 2 h; Reaction: cTl = 0.45 ␮mol cm−3
2 = 50 cm3 min−1
a
,
[4] A.-G. Boudjahem, M. Pietrowski, S. Monteverdi, M. Mercy, M.M. Bettahar, J.
Mater. Sci. 41 (7) (2006) 2025–2030.
H
.
[5] S. Hu, M. Xue, H. Chen, J. Shen, Chem. Eng. J. 162 (2010) 371–379.
[6] H. Chen, M. Xue, J. Shen, Catal. Lett. 135 (2010) 246–255.
[7] R. Wojcieszak, M. Zielin´ ski, S. Monteverdi, M.M. Bettahar, J. Colloid Interface
Sci. 299 (2006) 238–248.
The stability of activity of the catalysts Ni/MF-500, Ni/MO-
[8] R. Wojcieszak, M. Zielin´ ski, S. Monteverdi, M.M. Bettahar, Pol. J. Environ. Stud.
500 and Ni/60MF-500 for toluene hydrogenation was tested at
150 ◦C and results of the tests are given in Fig. 6. In 24-h tests all
catalysts showed a stable work. The activity of Ni/MF-500 cata-
lyst increased for the first 40 min of the reaction to the level of
0.46 molTl mol−1 min−1, then it gradually decreased and stabilized
after 5 h, wherNeias the activity of Ni/MO-500 catalyst diminished
considerably during the first 2 h and then stabilized. The best sta-
bility was found for the catalyst Ni/60MF-500. Its activity in 24-h
test decreased by less than 3%, while the activity of Ni/MF-500 fell
down by 15% and that of Ni/MO-500 by 28%.
The performance of the catalysts tested after their reduction at
500 ◦C was compared with that of the nickel catalyst obtained by
the impregnation of alumina (Table 2). It results from these data
that pure oxides (MgO, Al2O3) applied as Ni supports are the least
promising among the catalysts studied for toluene hydrogenation.
Calculations of TOF (s−1), performed using the values of disper-
sion, have shown that the most active catalyst was Ni/60MF-500,
whose activity was by over 10% higher than that of Ni/40MF-500
and Ni/MF-500 and twice as high as that of Ni/MO-500. The activity
of Ni/Al2O3 was lower by a factor of 2–4 than that of Ni supported
on MgF2 or MgF2–MgO binary supports. The introduction of MgF2
to MgO results in a jumpwise activity increase.
14 (2005) 67–71.
[9] A. Masalska, Appl. Catal. A 294 (2005) 260–272.
[10] Á. Szegedi, M. Popova, J. Porous Mater. 17 (2010) 663–668.
[11] Á. Szegedi, M. Popova, V. Mavrodinova, M. Urbán, I. Kiricsi, C. Minchev, Micro-
porous Mesoporous Mater. 99 (2007) 149–158.
[12] Á. Szegedi, V. Mavrodinov, M. Popova, C. Minchev, React. Kinet. Catal. Lett. 87
(2) (2006) 405–412.
[13] M.A. Ermakova, D.Y. Ermakov, Appl. Catal. A 245 (2003) 277–288.
[14] J.R. Grzechowiak, I. Szyszka, J. Rynkowski, D. Rajski, Appl. Catal. A 247 (2003)
193–206.
[15] J.R. Grzechowiak, I. Szyszka, A. Masalska, Catal. Today 137 (2008)
433–438.
[16] V.L. Barrio, P.L. Arias, J.F. Cambra, M.B. Güemez, B. Pawelec, J.L.G. Fierro, Appl.
Catal. A 242 (2003) 17–30.
[17] M. Wojciechowska, M. Zielin´ ski, W. Przystajko, M. Pietrowski, Catal. Today 119
(2007) 44–47.
[18] M. Wojciechowska, A. Malczewska, B. Czajka, M. Zielin´ ski, J. Goslar, Appl. Catal.
A 237 (2002) 63–70.
[19] M. Wojciechowska, W. Przystajko, M. Zielin´ ski, Catal. Today 119 (2007)
338–341.
[20] M. Wojciechowska, M. Pietrowski, B. Czajka, S. Łomnicki, Catal. Lett. 87 (3–4)
(2003) 153–157.
[21] M. Wojciechowska, M. Pietrowski, S. Łomnicki, Chem. Commun. 5 (1999)
463–464.
[22] M. Zielin´ ski, A. Wajnert, M. Wojciechowska, Pol. J. Environ. Stud. 18 (2009)
269–273.
[23] M. Zielin´ ski, M. Pietrowski, M. Wojciechowska, Pol. J. Environ. Stud. 18 (5)
(2009) 965–969.
[24] M. Zielin´ ski, M. Pietrowski, M. Wojciechowska, ChemCatChem
1653–1658.
[25] M. Zielin´ ski, M. Wojciechowska, Stud. Surf. Sci. Catal. 175 (2010) 429–432.
[26] M. Zielin´ ski, M. Wojciechowska, Catal. Today 169 (2011) 175–180.
3 (2011)
Therefore, we can conclude that nickel on MgF2-containing sup-
ports shows the highest activity. On the other hand, the content of
a certain amount of MgO in the support is necessary to ensure large
specific surface area and a better dispersion of nickel, which deter-
mine the catalyst activity for toluene hydrogenation. In our opinion,
this new class of MgF2–MgO mixed supports offers an attractive
alternative to other binary supports.
[27] M. Pietrowski, M. Zielin´ ski, M. Wojciechowska, ChemCatChem
3 (2011)
835–838.
[28] M. Pietrowski, M. Zielin´ ski, M. Wojciechowska, Catal. Lett. 128 (1-2) (2009)
31–35.
[29] M. Wojciechowska, B. Czajka, M. Pietrowski, M. Zielin´ ski, Catal. Lett. 66 (2000)
147–155.
[30] M. Wojciechowska, M. Zielin´ ski, M. Pietrowski, J. Fluorine Chem. 120 (2003)
1–11.
[31] M. Wojciechowska, A. Wajnert, I. Tomska-Foralewska, M. Zielin´ ski, B. Czajka,
4. Conclusions
Catal. Lett. 128 (2009) 77–82.
A
highly active nickel catalyst for toluene hydrogenation
[32] M. Zielin´ ski, M. Wojciechowska, Catal. Commun. 18 (2012) 1–4.
[33] I. Tomska-Foralewska, M. Zielin´ ski, M. Pietrowski, W. Przystajko, M. Woj-
ciechowska, Catal. Today 176 (2011) 263–266.
[34] G. Bergeret, P. Gallezot, G. Ertl, H. Knözinger, J. Weitkamp, Handbook of Het-
erogeneous Catalysis, Wiley, Weinheim, 1997.
mounted on an unconventional support MgF2–MgO was developed
as a result of the study on the preparation and characteriza-
tion of MgF2-containing supports. This is the first application of
Ni/MgF2–MgO catalyst to the reaction of toluene hydrogenation to
methylcyclohexane.
The introduction of magnesia into magnesium fluoride by the
precipitation method allows to obtain catalyst supports that have
much larger surface area and also better physico-chemical proper-
ties which ensure a higher dispersion of the Ni phase compared to
separate components. In general, the nickel phase supported on the
[35] 1999 JCPDS 01-1260.
[36] T. Furusawa, A. Tsutsumi, Appl. Catal. A 278 (2005) 207–212.
[37] M. Serra, P. Salagre, Y. Cesteros, F. Medina, J.E. Sueiras, Solid State Ionics 134
(2000) 229–239.
[38] K.H. Lautenschläger, W. Schröter, A. Wanniger, The Modern Compendium of
Chemistry, Polish Scientific Publishers PWN, Warsaw, 2007.
[39] B.V. King, F. Freund, Phys. Rev. B 29 (1984) 5814–5824.
[40] M. Serra, P. Salagre, Y. Cesteros, F. Medina, J.E. Sueiras, Phys. Chem. Chem. Phys.
6 (2004) 858–864.

22
M. Zielin´ski / Applied Catalysis A: General 449 (2012) 15–22
[41] S. Takenaka, S. Sato, R. Takahashi, T. Sodesawa, Phys. Chem. Chem. Phys. 5
(2003) 4968–4973.
[45] S. Scirè, C. Crisafulli, R. Maggiore, S. Minicò, S. Galvagno, Appl. Surf. Sci. 136
(1998) 311–320.
[42] A. Parmaliana, F. Arena, F. Frusteri, N. Giordano, J. Chem. Soc. Faraday Trans. 86
(14) (1990) 2663–2669.
[43] A.Yu. Stakheev, L.M. Kustov, Appl. Catal. A 188 (1999) 3–35.
[44] J. Zhao, M. Xue, Y. Huang, J. Shen, Catal. Commun. 16 (2011) 30–34.
[46] M.H. Peyrovi, M.R. Toosi, React. Kinet. Catal. Lett. 94 (1) (2008) 115–119.
[47] S. Loiha, K. Föttinger, K. Zorn, W. Klysubun, G. Rupprechter, J. Wittayakun, J.
Ind. Eng. Chem. 15 (2009) 819–823.