Synthesis, textural and catalytic properties of nanosized Fe 2O3/MgO system

Article abstract of DOI:10.1016/j.materresbull.2013.06.025

FeMgO system was prepared by three different methods. The samples were nominated as FeMgO_IM, FeMgO_Co and FeMgO_HY due to preparation by impregnation, co-precipitation and hydrothermal, respectively. The catalysts were characterized by TGA, XRD, FTIR, EPR, S_BET and TEM techniques. The catalytic properties of these samples were investigated by using H₂O₂ decomposition at (25-35 C) and partial oxidation of methanol at (300-400 C). FeMgO_HY sample showed the highest catalytic activity toward H₂O₂ decomposition. FeMgO_Co sample showed the highest catalytic activity toward partial oxidation of methanol. The results showed the sensitivity of H₂O ₂ decomposition reaction to the surface concentration of active species. While partial oxidation of methanol is sensitive to surface texture, solids interaction between active phase and support. All catalysts are highly selective to formaldehyde at reaction temperature 300 C. FeMgO_Co and FeMgO_IM catalysts showed high catalytic activity and stability toward partial oxidation of methanol with reusing.

Full text of DOI:10.1016/j.materresbull.2013.06.025

Materials Research Bulletin 48 (2013) 4105–4111
Contents lists available at SciVerse ScienceDirect
Materials Research Bulletin
journal homepage: www.elsevier.com/locate/matresbu
Synthesis, textural and catalytic properties of nanosized
Fe O /MgO system
2
3
Sahar A. El-Molla *, Hala R. Mahmoud
Department of Chemistry, Faculty of Education, Ain Shams University, Roxy 11757, Cairo, Egypt
A R T I C L E I N F O
A B S T R A C T
Article history:
FeMgO system was prepared by three different methods. The samples were nominated as FeMgOIM,
Received 18 February 2013
Received in revised form 11 May 2013
Accepted 16 June 2013
FeMgOCo and FeMgOHY due to preparation by impregnation, co-precipitation and hydrothermal,
respectively. The catalysts were characterized by TGA, XRD, FTIR, EPR, SBET and TEM techniques. The
catalytic properties of these samples were investigated by using H
partial oxidation of methanol at (300–400 8C). FeMgOHY sample showed the highest catalytic activity
toward H decomposition. FeMgOCo sample showed the highest catalytic activity toward partial
oxidation of methanol. The results showed the sensitivity of H decomposition reaction to the surface
2 2
O decomposition at (25–35 8C) and
Available online 24 June 2013
2 2
O
Keywords:
2 2
O
A. Microporous materials
A. Nanostructures
concentration of active species. While partial oxidation of methanol is sensitive to surface texture, solids
interaction between active phase and support. All catalysts are highly selective to formaldehyde at
reaction temperature 300 8C. FeMgOCo and FeMgOIM catalysts showed high catalytic activity and
stability toward partial oxidation of methanol with reusing.
C. X-ray diffraction
D. Catalytic properties
D. Electronic paramagnetic resonance (EPR)
ß 2013 Elsevier Ltd. All rights reserved.
1
. Introduction
support for many materials having biomedical applications [13].
Fe
2
O
3
/MgO-based catalysts have been used for many redox
decompositions [4]. H is used as a green
Because of their unique properties, intensive research efforts
reactions as H
2
O
2
2 2
O
have been devoted to nanosized metal oxides preparation [1]. The
composition, preparation procedure and pretreatment conditions
strongly affect on the catalytic properties of nanosized materials
fuel/propellant instead of carcinogenic hydrazine in spaceflight
under certain conditions [14], and instead of liquid oxygen as an
2 2
oxidizer in fuel cells [15]. Decomposition of H O yielding oxygen
[
2,3]. Controlled particle size and shape metal oxides exhibit
over metal oxides and their mixtures has been investigated
previously [16]. Also, supported iron oxide was used as a catalyst
for methanol partial oxidation to give formate species [17].
Formaldehyde is an important intermediate in many chemical
industries, such as production of adhesives, molding compounds
and coating resins [18]. It has been reported that formaldehyde is
formed by a combination of oxidative and non-oxidative dehydro-
genation of methanol [19].
Production of formaldehyde from methanol and air is done with
either methanol-rich or methanol-lean feed using the silver and
oxide process, respectively [19]. Due to methanol price duplica-
tion, oxide process has been favored the last years [19,20,21]. Many
catalytic systems were investigated in methanol oxidation to
densely populated unsaturated surface coordination sites. These
sites are responsible for improving the catalytic performance [4].
Various methods have been proposed for metal oxides prepara-
tion, such as sol–gel [5], impregnation [6], co-precipitation [7] and
hydrothermal [8]. It has been reported that wet impregnation
method allows obtaining supported catalysts and co-precipita-
tion yields ‘‘bulk’’ mixtures [9]. While hydrothermal method
yields highly homogeneous crystalline product directly at a
relatively lower reaction temperature. It favors a decrease in
agglomeration between particles, phase homogeneity, uniform
composition, high product purity and controlled particle mor-
phology [10].
Nanosized iron oxide showed a great activity in the oxidation
reactions instead of noble metals [11]. Supporting iron oxide
resulted modifications of its textural, structural, and catalytic
properties [12]. MgO is a ceramic oxide used as a catalyst and
formaldehyde such as FeVO
supported Fe–V–O [23].
4 2 3 2 2
[22], a-Al O -, SiO - and TiO -
In this paper, we aimed to investigate the effect of preparation
methods of the Fe /MgO system on its physicochemical, surface
2 3
O
and the catalytic properties toward two redox reactions are
catalytic decomposition of hydrogen peroxide at 25–35 8C and
conversion of methanol at 300–400 8C using the flow system and.
The techniques employed were TG/DTG, XRD, EPR, TEM and
nitrogen adsorption at–196 8C.
*
Corresponding author. Tel.: +20 2 24558087/24558086; fax: +20 2 22581243.
E-mail address: saharelmolla@yahoo.com (S.A. El-Molla).
URL: http://elmolla.webs.com
0025-5408/$ – see front matter ß 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.materresbull.2013.06.025

4
106
S.A. El-Molla, H.R. Mahmoud / Materials Research Bulletin 48 (2013) 4105–4111
2
. Experimental
.1. Preparation of FeMgO system
2 3
Three Fe /MgO catalysts having the formula 0.045Fe O /
9.7 GHz (X band) for cavity operating. The catalysts were activated
at laboratory temperature under vacuum by a short evacuation.
The surface characteristics of the prepared catalysts were
determined from nitrogen adsorption isotherms measured at–
196 8C using a Quantachrome NOVA 2000 automated gas-sorption
apparatus model 7.11 (USA). The surface characteristics include
2
2
O
3
MgO were prepared by three different methods namely: wet
impregnation (IM), co-precipitation (Co) and hydrothermal (HY)
p
specific surface areas (SBET), total pore volume (V ) and average
methods. Fe
5.2%. All the employed chemicals were of analytical grade and
supplied by BDH Company (UK).
2
O
3
mass content (wt %) in these samples was fixed at
pore radius ( r´ ). All catalysts were degassed at 200 8C for 2 h under a
reduced pressure of 1.3 mPa before taking such measurements.
Morphology of the samples was investigated using transmis-
sion electron microscope (TEM) (JEM-2100CX (JEOL).
1
2
.1.1. Wet impregnation method (IM)
The catalytic activities of the various solid catalyst samples
were determined by using methanol conversion reaction at
different temperatures varying between 300 and 400 8C. The
The sample was prepared by impregnation a known amount of
magnesium hydroxide, which was prepared previously [4], with a
known amount of iron nitrate dissolved in the least amount of bi-
distilled water necessary to make a paste. The resulting paste was
dried at 110 8C and then calcined at 500 8C for 4 h. This sample was
catalytic reaction was conducted in
a flow reactor under
atmospheric pressure. Thus, a 50 mg catalyst sample was held
between two glass wool plugs in a Pyrex glass reactor tube 20 cm
long and 1 cm internal diameter packed with quartz fragments 2–
nominated as FeMgOIM
.
3
mm length. The temperature of the catalyst bed was regulated
2
.1.2. Co-precipitation method (Co)
The sample was prepared by co-precipitation iron and
magnesium hydroxides at room temperature using corresponding
amounts of iron and magnesium nitrates and 0.2 M NH OH at pH 8.
and controlled within ꢁ1 8C. Argon gas was used as the diluent and
the methanol vapor was introduced into the reactor through an
evaporator/saturator containing the liquid reactant at constant
temperature 35 8C. The flow rate of the argon was maintained at
25 ml/min. Before carrying out catalytic activity measurements each
catalyst sample was activated by heating at 300 8C in a current of
argon for 1 h then cooled to the catalytic reaction temperature. The
injection time of the sample products and the unreacted methanol
was fixed after 15 min, and many injections were carried out to give
constant conversion. The reaction products in the gaseous phase were
analyzed chromatographically using Perkin-Elmer Auto System XL
Gas Chromatograph fitted with a flame ionization detector. The
column used was stainless steel chromatographic columns, 4 m
length, packed with 10% squalane supported on chromosorb. The
reaction products were analyzed at a column temperature of 40 8C in
all conversion runs. Detector temperature was kept at 250 8C.
4
The precipitate was filtered and washed with bi-distilled water.
The obtained hydroxide was dried at 110 8C and then calcined in air
at 500 8C for 4 h. This sample was nominated as FeMgOCo
.
2
.1.3. Hydrothermal method (HY)
The coprecipitated iron–magnesium hydroxides sample at
5 8C, was suspended in 30 ml bi-distilled water. The obtained
2
suspension was placed into a 45 ml Teflon-lined autoclave and
then heated at 200 8C at vapor pressure15 atm for 1 h. Then
autoclave was quenched to room temperature and the resulting
colloidal iron/magnesium hydroxide was heated at 50 8C for
several hours to remove water [24]. The obtained solid was
calcined in air at 500 8C for 4 h. This sample was nominated as
The catalytic decomposition of H
catalysts was also determined. 100 mg of a given catalyst sample
was taken and 0.5 ml of H of known concentration diluted to
20 ml with distilled water was used at reaction temperature 25–
5 8C. The reaction kinetics was monitored by measuring the
volume of O liberated at different time intervals until no further
2 2
O in presence of the prepared
FeMgOHY
.
2 2
O
2
.2. Techniques
3
Thermogravimetric analysis (TGA) of the catalysts was per-
formed using a Shimadzu TGA-50H thermogravimetric analyzer
2
oxygen was liberated.
ꢀ
1
(Japan); the rate of heating was kept at 10 8C min and the flow
rate of nitrogen was 30 ml/min.
3. Results and discussions
X-ray diffractograms (XRD) were obtained using Brucker Axs
D8 Advance X-ray diffractometer (Germany), using Cu K
˚
irradiation (l = 0.15404 A) at a scan rate of 28 in 2u/min. The
a
1
3.1. Thermogravimetric analysis of the uncalcined FeMgO samples
accelerating voltage and applied current were 40 kV and 40 mA,
respectively. The crystallite sizes of the investigated phases were
calculated using Scherrer equation [25], Fourier transform infrared
Thermogravimetric analysis (TGA) of uncalcined FeMgOIM
,
FeMgOCo and FeMgOHY samples were determined and the data are
listed in Table 1. The TG curves of the investigated samples (not
given here), consist of three stages. The first step represents the
loss of surface-adsorbed water and some of crystallization water
from iron nitrate. The second step is indicative to the complete
decomposition of iron nitrate to Fe O along with decomposition
(FTIR) spectroscopy of different samples were recorded on a Jasco
IR 4100 spectrometer (Japan) using KBr pellets in the range of
ꢀ
1
4
000–400 cm region.
The EPR investigations were performed at laboratory tempera-
2
3
ture using a Brucker spectrometer (Germany) at a frequency of
most of the magnesium hydroxide. The final step indicates the
Table 1
Effect of changing the preparation method on the thermal behavior of uncalcined of Fe
2
O
3
/MgO samples.
Sample
T
1
(8C) (wt. loss%)
T
2
(8C) (wt. loss %)
T
3
(8C) (wt. loss %)
Total weight loss %
Calculated
Found
FeMgOIM
FeMgOCo
FeMgOHY
Rt ! 156 (8.6)
Rt ! 193 (10.2)
Rt ! 193 (11.4)
156 ! 422 (35.5)
193 ! 400 (20.1)
193 ! 356 (16.3)
422 ! 739 (4.5)
400 ! 764 (9.7)
356 ! 707 (10.4)
54.1
40.9
38.1
48.6
40.0
39.4
Rt: room temperature.

S.A. El-Molla, H.R. Mahmoud / Materials Research Bulletin 48 (2013) 4105–4111
4107
Fig. 1. XRD pattern of different prepared FeMgO catalysts calcined at 500 8C. Lines
1) refer to MgO phase.
Fig. 2. FTIR spectra of the catalysts: (a) FeMgOIM, (b) FeMgOCo and (c) FeMgOHY
calcined at 500 8C.
(
ꢀ
1
complete dehydroxylation of MgO and/or the removal of the
residual OH groups bonded to the MgO lattice [4,26]. The
difference in weight loss values of the investigated samples can
be assigned to the incomplete desorption of the residual O¯ H groups
bonded to the MgO lattice in this temperature range [26]. The
observed lower weight loss for uncalcined FeMgOCo and FeMgOHY
samples may be due to formation spherical and nanorod particles.
The observed higher weight loss for uncalcined FeMgOIM sample
may be due to formation microporous structure [27].
3600–3200 cm
due to adsorbed water molecules and O–H
stretching of surface hydroxyls disturbed by the hydrogen bonds
[30–32]. The observed broadening in this band in case of FeMgOHY
spectra (Fig. 2c) confirms a high degree of hydrogen bonding of
water molecules among themselves and with the surface of MgO
crystallites [33]. It is observed also presence band at 1640–
1610 cm is due to O–H bending [12]. The band at 1100 cm is
belonging to M–OH [31]. A broad band at 1460–1400 cm
corresponds to the Mg–O stretching as well as the Mg–O–Mg
deformation [32].
ꢀ
1
ꢀ1
ꢀ
l
3.2. X-ray diffractograms of the prepared catalysts
3.4. EPR of the prepared catalysts
X-ray diffraction technique was used to characterize FeMgOIM
,
FeMgOCo and FeMgOHY catalysts calcined at 500 8C. The phase
structure and the crystallite sizes of the investigated phases in these
solids were investigated as shown in Fig. 1 and Table 2. The results
showed that: (i) the diffractograms consist of MgO phase (periclase)
The EPR spectra of the prepared catalysts calcined at 500 8C
were investigated as shown in Fig. 3. Inspection of this figure there
is a very broad ESR signal at g ꢃ 2 with very low, moderate and
extremely high intensity for FeMgOIM, FeMgOHY and FeMgOCo
catalysts, respectively. This EPR signal could be due to Fe–O–Fe
species with ferri-, ferro-, and/or antiferromagnetic behavior [34].
(JCPDS 4-829). MgO phase appears with moderate degree of
ordering in case of FeMgOIM sample, and with very low degree of
crystallinity in case of FeMgOCo and FeMgOHY nanoparticles. The
2 3
It can originate from various iron oxide species such as g-Fe O or
absence of Fe
state beside its small amount to be detected by X-ray diffractometer
28]. So, magnesium oxide acts as a convenient support to hematite.
2
O
3
phase might indicate its presence in finely divided
Fe
3
O
4
(both ferrimagnetic), or
a
-Fe
2
O
3
2
(antiferromagnetic) [35]. It
ꢀ
can be related to superoxide ions (O ) associated with Fe ions
[36], or hydrated iron species [37]. The observed broadening of EPR
signal at g ꢃ 2 could be due to partial reduction of a very small
[
The crystallite size of the detected MgO phase is ꢂ21.7 nm. The
crystallite size values of MgO phase present in the FeMgOCo and
FeMgOHY catalystsweresmallerthanthatfoundinFeMgOIM sample.
The valueofcalculatedlatticeconstant ofMgO phasewas dependent
on the preparation technique. It is observed that the lattice
parameter of MgO in FeMgOHY sample has the smallest value. This
result could be explained on the light of the small particle size of
MgO phase in FeMgOHY sample [29].
3
+
2+
0
amount of Fe to EPR-silent Fe (or Fe ) species as shown in
FeMgOHY spectrum [38]. The observed increase in signal intensity
3.3. FTIR spectra of the prepared catalysts
The FTIR spectra of the prepared catalysts were investigated as
shown in Fig. 2. According to this figure there is a broad band at
Table 2
Intensity count, crystallite size and lattice parameter constant of MgO phase inthe
investigated samples calcined at 500 8C.
a-const [MgO] ( A˚ )a
Sample
MgO
Intensity
Crystallite
size (nm)
count (a.u.)
FeMgOIM
FeMgOCo
FeMgOHY
30.8
7.4
21.7
9.1
4.215
4.212
4.206
9.9
9.8
Fig. 3. Ambient temperature EPR spectra of different prepared FeMgO catalysts
calcined at 500 8C.
a
˚
The standard a value of MgO is 4.213 A.

4
108
S.A. El-Molla, H.R. Mahmoud / Materials Research Bulletin 48 (2013) 4105–4111
surface area. (iii) FeMgOCo sample consists of meso- and
micropores mixture while FeMgOHY sample is mesoporous
material. (iv) The values of SBET, S and S are close to each other
t
s
which justify the correct choice of standard t-curves used in the
analysis. According to the obtained results, presence a mixture of
meso- and micropores in FeMgOCo sample could reduce transport
limitations in catalysis, resulting in higher activities and better
controlled selectivities [43].
3.6. TEM of the prepared catalysts
To investigate the effect of preparation method on the FeMgO
nanoparticles morphology, TEM images were measured as shown
in Fig. 5. This figure showed that: (i) TEM of FeMgOIM solid contain
aggregates of uniform trigonal shapes from MgO nanoparticles
with average diameter 25 nm. (ii) TEM of FeMgOCo sample contains
uniform spherical nanoparticles with average diameter 7.9 nm.
(
iii) TEM of FeMgOHY sample contains some nano-rods and
uniform spherical nanoparticles of MgO. The average diameter
of the spherical nanoparticles is 3 nm and the average diameters
and lengths of the nano-rods are 9.1 and 43.5 nm, respectively. The
average particle size calculated from the TEM micrographs is
consistent with the average crystallite size obtained from XRD
measurement. (iv) There are very small darker spots highly
dispersed in the hydroxide or oxide layer. These spots correspond
to very small iron oxide particles completely dispersed [44]. The
presence relatively large sized uniform trigonal crystals in
FeMgOIM and FeMgOCo samples could be due to rapid growing
2
of Mg(OH) precursor during its preparation by precipitation. In
addition, calcination of this material affords MgO having corre-
spondingly large particle size [45,46]. So, FeMgOIM and FeMgOCo
samples calcined at 500 8C contain randomly distributed and
highly aggregated nanoparticles, these aggregated nanoparticles
exist in cluster format. These results confirm that the hydrother-
mal treatment play an important role in rod growing [47,48].
Fig. 4. Nitrogen adsorption/desorption isotherms (a) and V
prepared FeMgO catalysts calcined at 500 8C.
L
–t plots (b) of different
in case of FeMgOCo and FeMgOHY nanoparticles could be due to
presence of super paramagnetic iron oxide nanoparticles [39]. The
observed changes in the thermal, crystallographic, and spectral
properties of FeMgO system can affect both its surface and catalytic
properties.
3
.7. Catalytic activities of FeMgO catalysts
3
.7.1. Catalytic decomposition of H
2 2
O
The catalytic decomposition of H
2 2
O was studied at 25–35 8C
over FeMgOIM, FeMgOCo and FeMgOHY catalysts as shown in Fig. 6.
This figure shows the first-order plots, its slopes allowed a ready
determination of the reaction rate constant (k) at 30 8C. Inspection
of Fig. 6 the catalytic activity of the investigated samples expressed
3
.5. Surface characteristics of FeMgO catalysts
The nitrogen adsorption/desorption isotherms of FeMgO
ꢀ
1
as reaction rate constant k (min ) followed this order: FeMgO-
FeMgOCo > FeMgOIM. In other words, the solid prepared by
nanoparticles were carried out as shown in Fig. 4a. These isotherms
could be classified as type II of Brunauer’s classification. Table 3
includes the surface characteristics of FeMgO nanoparticles. The
HY
>
hydrothermal method showed the biggest catalytic activity. To
eliminate the role played by surface area, the values of reaction
specific surface areas (SBET) were calculated by the BET method
ꢀ
ꢀ1
ꢀ2
0
rate constant per unit surface area k (min
m
) at 30 8C were
[
40]. The total pore volumes, V
p
were taken at P/P = 0.95. The
calculated to as shown in Fig. 7. Due to this figure, there is a big
increase in the catalytic activity of FeMgOHY in spite of its small
surface area with comparison to FeMgOIM and FeMgOCo catalysts.
The obtained results could be explained on the light of the
following: (i) The observed catalytic activity of FeMgO system
average pore radius ( r´ ) was estimated from the relationship:
4
r´ = 10 (2V
and methods [41,42], respectively (as shown in Fig. 4b).
method is used also to calculate external and internal surface areas
ext and Sint) in the microporous samples [42]. Inspection of
p t s
/SBET). S and S surface areas were calculated by t-plot
a
s
a
s
(
S
Fig. 4 and Table 3: (i) SBET and average pore radius ( r´ ) of
nano-sized FeMgO catalysts follow this order: FeMgO-
HY < FeMgOIM < FeMgOCo. (ii) FeMgOIM sample is a microporous
material and its internal surface area is higher than the external
toward decomposition of H
Fe particles in nanosize MgO support as shown in XRD and TEM
sections. (ii) The observed higher catalytic activity of FeMgOHY
2 2
O refers to presence well-dispersed
2 3
O
2
+
3+
toward this reaction could be due to presence Fe and Fe species
Table 3
The specific surface areas and pore characteristics of the FeMgO samples calcined at 500 8C.
2
2
2
2
2
˚
Samples
S
BET (m /g)
S
t
(m /g)
S
s
(m /g)
S
ext. (m /g)
S
inter. (m /g)
V
p
(ml/g)
r´ (A)
FeMgOIM
FeMgOCo
FeMgOHy
95.2
99.1
43.9
94.6
93.3
42.2
–
51.1
–
0.1497
0.2722
0.1469
31.5
54.9
66.8
100.5
44.6
100.7
45.3
–
–

S.A. El-Molla, H.R. Mahmoud / Materials Research Bulletin 48 (2013) 4105–4111
4109
Fig. 5. TEM images of (a) FeMgOIM, (b) FeMgOCo and (c) FeMgOHY nanoparticles calcined at 500 8C.
2
+
[
38] as shown in EPR section. Presence of Fe ions increases the
rate of the investigated catalytic reaction. It seems that catalytic
decomposition of H is not much sensitive to the catalyst surface
present in FeMgOIM sample [49]. It has been reported that the
2 3
reducible transition metal oxides such as Fe O exhibit a
2
O
2
significantly enhanced activity for oxidation reaction, which was
attributed to their ability to provide reactive oxygen [50]. So, more
area enhancement while it is sensitive to presence transition metal
ions with different oxidation states. (ii) Also, presence nano-rods in
the morphology of FeMgOHY sample as shown in the TEM of this
sample could play an important role in enhancement of its
catalytic activity toward this redox reaction.
2 2
oxygen vacancies may facilitate the adsorption of H O and result
in increasing the catalytic activity [51]. (iv) The possible dis-
homogeneity of iron cations distribution [52] on the surface may
be responsible for the observed decrease in the catalytic activity of
FeMgOIM sample.
(
iii) The catalytic activity of FeMgOCo was higher than that of
FeMgOIM sample; this could be due to presence of various types of
surface active oxygen species in FeMgOCo sample more than that
3.7.2. Partial catalytic oxidation of methanol
The catalytic conversion of methanol, at reaction temperature
varied between 300 and 400 8C was carried out over fresh and
reused FeMgO catalysts calcined at 500 8C. The catalytic activity
ꢀ
Fig. 6. First-order plots of H
2
O
2
decomposition conducted at 30 8C over different
Fig. 7. Variation of (k ) toward H
2
O
2
decomposition conducted at 30 8C over FeMgO
prepared FeMgO catalysts calcined at 500 8C.
catalysts calcined at 500 8C as a function of preparation method.

4
110
S.A. El-Molla, H.R. Mahmoud / Materials Research Bulletin 48 (2013) 4105–4111
was expressed as a change in the total percentage conversion as
shown in Fig. 8 and Table 4. The values of catalytic selectivities
toward various products of methanol conversion in the gas phase
were calculated and added in Table 4. The detected products of
methanol conversion in presence FeMgO system were formalde-
hyde (FM), dimethyl ether (DME) and traces of methyl formate
(
MF). Inspection of Table 4 and Fig. 8: (i) the catalytic activity
increased with increasing the reaction temperature from 300 to
00 8C. (ii) FeMgOCo exhibited catalytic activity higher than that of
4
FeMgOIM and FeMgOHY catalysts. (iii) All catalysts were highly
selective to formaldehyde 100% at 300 8C; this selectivity was
slightly decreased with increasing the reaction temperature
especially in case of FeMgOCo and FeMgOIM catalysts. (iv) FeMgOHY
catalyst exhibited small selectivity toward DME in expense of
formaldehyde selectivity at 350 8C. (v) Very small amounts of
methyl formate and DME were detected at 400 8C. (vi) The total
conversion decreased by reusing the catalysts (FeMgOCo and
FeMgOIM) toward partial oxidation of methanol two times. The
activity retained of these samples at 400 8C after reusing for two
times was about 52%. (vii) The recycled FeMgOCo and FeMgOIM
samples showed high stability toward formaldehyde selectivity
and some fluctuation toward DME or methyl formate selectivity.
The obtained results can be explained as the following: (i) the
catalytic activity of FeMgO catalyst toward partial oxidation of
methanol to formaldehyde depends on presence nano-iron oxide
[
53,54] supported on magnesia. Presence these active species been
inferred from the observed decrease in the degree of ordering and
crystallite size of MgO as a support (cf. Table 2). (ii) The role of
surface area is important in the interpretation the high catalytic
activity of FeMgOCo and FeMgOIM with comparison to FeMgOHY
catalyst calcined at 500 8C (cf. Table 3). (iii) The most active one
was FeMgOCo catalyst. This activity is due to consistence meso and
micro structure as shown in surface section. So, combination of
smaller and larger mesopores could reduce transport limitations in
catalysis, resulting in higher activities and better controlled
selectivities [43]. (v) The observed low activity of FeMgOHY
calcined at 500 8C could be attributed to its smaller surface area as
shown in Table 3 or, possible smaller number of active surface sites
Fig. 8. Total conversion of methanol over (a) various catalysts as function of reaction
temperature (b) FeMgOCo and FeMgOHY samples as a function of reusing times.
x
Mg1ꢀxFe O1+0.5x with medium-strength basic sites is reported [59].
So, the observed decrease in the catalytic activity and selectivity of
FeMgOHY calcined at 500 8C could be due to ferrite formation. (iv)
Formaldehyde is produced according to dehydrogenation mecha-
[
56,57]. It has been reported by one of the authors that the surface
3
+
concentration of iron species measured by EDX technique [58] was
higher in case of the FeMgOCo and FeMgOHY samples with
comparison to FeMgOIM sample calcined at 500 8C. So, in spite
of this result the surface area parameter plays a dominant effect in
conversion of methanol, where the surface area of FeMgOHY was
half the value of other two samples. Other item one cannot neglect
was the ferrite formation, where FeMgOHY sample was the most
sample capable to form ferrite even at 500 8C [58]. It has been
nism and the active sites are nanosized Fe ions and medium-
strength basic active sites [Mg(M)–O] with high density [55]. The
observed high selectivity of the FeMgO catalyst for formaldehyde
production stresses on the role of magnesium ferrite in partial
oxidation to formaldehyde, and supports the idea of disappearance
2 3 2 3
Fe O phase. Fe O is known as an active catalyst to total oxidation
of methanol [60]. (vi) The observed decrease in the formaldehyde
selectivity in case of FeMgOHY sample could be due to such higher
3+
reported that Fe has higher electronegativity as compared to
degree of surface reduction during dehydrogenation process
2+
2+
Mg , and, formation of solid solutions having general composition
yielding Fe . This surface reduction would occur during
Table 4
Total conversion (T.C.) and selectivities to formaldehyde S
F
, methyl formate SMF and dimethyl ether SDME over various prepared catalysts calcined at 500 8C.
Reaction temp. (8C)
T.C. and selectivity (%)
Catalyst
FeMgOIM
Recycle1
Recycle2
FeMgOCo
Recycle1
Recycle2
FeMgOHY
3
00
50
T.C.
0.44
100
0.25
100
0.17
100
0.60
100
0.43
100
0.18
100
0.16
100
S
F
3
T.C.
4.0
98.5
0.75
0.75
3.9
98.7
1.3
0
2.0
98
7.3
98.9
0
6.5
99.2
0
3.8
97.1
0
3.0
88.5
0
S
S
S
F
MF
DME
1.5
0.5
1.1
0.8
2.9
11.5
4
00
T.C.
29.6
96.2
0.8
27.0
98
15.5
98.3
0.8
27.1
97.9
23.6
99.5
0
12.6
97.9
0.9
22.2
97.4
0.6
S
S
S
F
a
MF
DME
0.8
1.4
a
a
a
3.0
1.2
0.9
0.7
0.5
1.2
2.0
a
Others beside DME or MF.

S.A. El-Molla, H.R. Mahmoud / Materials Research Bulletin 48 (2013) 4105–4111
4111
[
9] A. Glisenti, J. Mol. Catal. A: Chem. 153 (2000) 169–190.
calcination too [61]. It has been reported that the surface of
coprecipitated sample has lower reducibility than other samples
[
[
10] K. Byrappa, T. Adschiri, Prog. Cryst. Growth Charact. Mater. 53 (2007) 117–166.
11] S.C. Kwon, M. Fan, T.D. Wheelock, B. Saha, Sep. Purif. Technol. 58 (2007) 40–48.
[
61]. The observed increase in the EPR signal intensity for FeMgOCo
[12] I.E. Wachs, Catal. Today 100 (2005) 79–94.
[13] Y.-H. Choa, J.-K. Yang,W.-J.Yang,K.-H. Auh, J. Magn.Magn.Mater. 266(2003)20–27.
[14] N. Teshima, Z. Genfa, P.K. Dasgupta, Anal. Chim. Acta 510 (2004) 9–13.
[15] A.E. Sanli, C.A. Ayta, Int. J. Hydrogen Energy 36 (2011) 869–875.
sample showed the high concentration of free iron species (not
reduced) in the course of preparation and calcination while the
lower intensity support the idea of lowering concentration of Fe
and possible formation Fe species in FeMgOHY catalyst through
preparation and calcination step. (vii) The observed selectivity of
FeMgOHY to DME could be due to formation nano-rods. It has been
reported that Fe possess significant acid properties because it
occupies low-coordinated crystallographic positions and the
coordination-non-saturated atom capable of creation more Lewis
acid sites on the surface of samples [57]. But the possible decrease
3
+
[16] N.K. Renuka, J. Mol. Catal. A: Chem. 316 (2010) 126–130.
[17] A. Khaleel, I. Shehadi1, M. Al-Shamisi, Colloids Surf. A 355 (2010) 75–82.
2+
[
[
[
[
18] G. Reuss, W. Disteldorf, A.O. Gamer, A. Hilt, 7th ed., Ullmann’s Encyclopedia of
Industrial Chemistry, vol. A11, Wiley-VCH, Weinheim, 2008, pp. 619–652.
19] M. Massa, R. Haggblad, S. Hansen, A. Andersson, Appl. Catal. A: Gen. 408 (2011)
¨
3+
63–72.
20] Methanex Corporation, Vancouver, Canada (methanol prices from 2002 to 2008,
www.methanex.com).
21] B. Crichton, in: Informally Speaking (newsletter from Perstorp Formox, www.per-
storpformox.com), spring/summer 2003, pp. 12–13.
3+
2+
[22] R. H a¨ ggblad, J.B. Wagner, S. Hansen, A. Andersson, J. Catal. 258 (2008) 345–355.
in surface concentration of Fe (to give Fe ) as shown in EPR and
redox reaction of H may decrease the external surface Lewis
[
[
23] R. H a¨ ggblad, M. Massa, A. Andersson, J. Catal. 266 (2009) 218–227.
24] L. Diamandescu, M. Feder, D. Tarabasanu-Mihaila, F. Vasiliu, Appl. Catal. A: Gen.
325 (2007) 270–275.
2 2
O
acidity. It has been reported that the selectivity to DME was
enhanced as the ZSM-5 crystal size of samples were reduced and it
depends on the acid site distribution, and if there is less or even no
acid site on the external surface, all of methanol could be converted
to DME. It was also found that uniform ZSM-5 nanocrystals are the
most selective catalyst to DME production [62]. According to these
reported results one can explain the observed selectivity for
FeMgOHY sample toward DME due to presence some observed
nano-rods with possible low concentration of surface acidity. (viii)
Absence DME in case of FeMgOCo and FeMgOIM samples with
comparison to FeMgOHY is due to the role of dilution effect of iron
oxide species by MgO support (due to increasing surface area and
degree of dispersion as shown in XRD section). Where, it was
profound on the dimethyl ether formation suggesting that the iron
oxide surface atoms must be closely situated to promote the
adsorption of two neighboring methanol groups as initially
methoxy groups that subsequently combine to give dimethyl
ether.
[
[
[
[
25] H.P. Klug, L.E. Alexander, X-ray diffraction pro-cedures for polycrystalline and
amorphous materials John Wiley and Sons, New York, 1966, pp. 491.
26] J.A. Wang, O. Novaro, X. Bokhimi, T. L o´ pez, R. G o´ mez, J. Navarrete, M.E. Llanos, E.
L o´ pez-Salinas, Mater. Lett. 35 (1998) 317–323.
27] L. Kumari, W.Z. Li, C.H. Vannoy, R.M. Leblanc, D.Z. Wang, Ceram. Int. 35 (2009)
3355–3364.
28] T. Tsoncheva, J. Roggenbuck, M. Tiemann, L. Ivanova, D. Paneva, I. Mitov, C.
Minchev, Micropor. Mesopor. Mater. 110 (2008) 339–346.
[29] W.H. Qi, M.P. Wang, J. Nanopart. Res. 7 (2005) 51–57.
[
30] M.C. Roco, R.S. Villiams, P. Alivisatos (Eds.), Nanotechnology Research Directions,
Vision for Nanotechnology R&D in Next Decade, WGN Workshop Report, Kluwer,
Dordrecht, 2001.
[31] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination
Compounds, 4th ed., Wiley, USA, 1986.
[32] I.F. Mironyuk, V.M. Gun’ko, M.O. Povazhnyak, V.I. Zarko, V.M. Chelyadin, R.
Leboda, J. Skubiszewska-Zi e˛ ba, W. Janusz, Appl. Surf. Sci. 252 (2006) 4071–4082.
[33] A. Kumar, J. Kumar, J. Phys. Chem. Solids 69 (2008) 2764–2772.
[
[
34] A. Br u¨ ckner, R. L u¨ ck, W. Wieker, B. Fahlke, H. Mehner, Zeolites 12 (1992) 380–385.
35] T. Ida, H. Tsuiki, A. Ueno, K. Tohji, Y. Udagawa, K. Iwai, H. Sano, J. Catal. 106 (1987)
428–439.
[36] H.L. Chen, L.G. Wu, C.L. Zhu, Chem. Eng. J. 78 (2000) 159–164.
[
[
[
[
37] G. Berlier, G. Spoto, S. Bordiga, G. Ricchiardi, P. Fisicaro, A. Zecchina, I. Rossetti, E.
Selli, L. Forni, E. Giamello, C. Lamberti, J. Catal. 208 (2002) 64–82.
38] P. Decyk, M. Trejda, M. Ziolek, J. Kujawa, K. G?aszczka, M. Bettahar, S. Monteverdi,
M. Mercy, J. Catal. 219 (2003) 146–155.
4
. Conclusions
39] M. Mahmoudi, S. Sant, B. Wang, S. Laurent, T. Sen, Adv. Drug Deliv. Rev. 63 (2011)
24–46.
In conclusion, the physicochemical and catalytic properties of
40] S. Brunauer, P.H. Emmett, E. Teller, J. Am. Chem. Soc. 60 (1938) 309–319.
FeMgO system depend mainly on the preparation method, i.e.
impregnation (IM), coprecipitation (Co) and hydrothermal (HY).
The catalytic activity of the FeMgO system toward redox reaction
[41] B.C. Lippens, J.H. de Boer, J. Catal. 4 (1965) 319–323.
[
[
[
42] K.S.W. Sing, Chem. Ind. (1967) 829–830.
43] F. Meshkani, M. Rezaei, Powder Technol. 196 (2009) 85–88.
44] K. Bachari, A. Touileb, A. Saadi, D. Halliche, O. Cherifi, J. Porous Mater. 17 (2010)
573–581.
2 2
as H O decomposition is mainly determined by surface concen-
tration of active species and presence transition metal ions with
different oxidation states while behavior this system toward
partial oxidation of methanol is mainly dependent on surface
texture, solid–solid interaction between active phase and support.
The results revealed that the catalytic activity of nanosized FeMgO
[45] L. Huang, D.-Q. Li, Y.-J. Lin, M. Wei, D.G. Evans, X. Duan, J. Inorg. Biochem. 99
2005) 986–993.
(
[
[
46] W. Wang, X. Qiao, J. Chen, F. Tan, H. Li, Mater. Charact. 60 (2009) 858–862.
47] M. Saif, H. Hafez, A.I. Nabeel, Chemosphere 90 (2013) 840–847.
[48] L. Yan, J. Zhuang, X. Sun, Z. Deng, Y. Li, Mater. Chem. Phys. 76 (2002) 119–122.
[
49] P. Li, C. He, J. Cheng, C.Y. Ma, B.J. Dou, Z.P. Hao, Appl. Catal. B: Environ. 101 (2011)
70–579.
5
system toward H
FeMgOHY > FeMgOCo > FeMgOIM
toward partial oxidation of methanol can be ranked as: FeMgO-
2
O
2
decomposition follows the sequence as:
[50] M.M. Schubert, S. Hackenberg, A.C. van Veen, M. Muhler, V. Plzak, R.J. Behm, J.
Catal. 197 (2001) 113–122.
.
While behavior this system
[
[
51] D. Guin, B. Baruwati, S.V. Manorama, J. Mol. Catal. A 242 (2005) 26–31.
52] F. Basile, G. Fornasari, M. Gazzano, A. Vaccaria, J. Mater. Chem. 12 (2002) 3296–
Co > FeMgOIM >FeMgOHY
.
3303.
[53] J. Ge, T. Huynh, Y. Hu, Y. Yin, Nano Lett. 8 (2008) 931–934.
References
[54] A.C. Silva a, D.Q.L. Oliveira, L.C.A. Oliveira, A.S. Anast a´ cio, T.C. Ramalho, J.H. Lopes,
H.W.P. Carvalho, C.E.R. Torres, Appl. Catal. A: Gen. 357 (2009) 79–84.
[
[
[
55] G. Torres, C.R. Apesteguia, J.I. Di Cosimo, Appl. Catal. A: Gen. 317 (2007) 161–170.
56] D. Kulkarni, I.E. Wachs, Appl. Catal. A: Gen. 237 (2002) 121–137.
57] I. Shchelokov, E. Asabina, M. Sukhanov, M. Ermilova, N. Orekhova, V. Pet’kov, G.
Tereshchenko, Solid State Sci. 10 (2008) 513–517.
[
1] J.H. Adair, T. Li, T. Kido, K. Havey, J. Moon, J. Mecholsky, A. Morrone, D.R. Talham,
M.H. Ludwig, L. Wang, Mater. Sci. Eng. R 23 (1998) 139–342.
2] A. Glisenti, G. Favero, G. Granozzi, J. Chem. Soc., Faraday Trans. 94 (1998) 173–182.
3] G.A. El-Shobaky, N.R.E. Radwan, M.S. El-Shall, A.M. Turky, M.A. Hassan, Colloids
Surf. A 311 (2007) 161–169.
[
[
[
[
[
[
[
58] S.A. El-Molla, G.A. Fagal, N.A. Hassan, G.M. Mohamed, Res. Chem. Intermed.
(2013), http://dx.doi.org/10.1007/s11164-013-1220-2.
[
[
[
[
[
4] S.A. El-Molla, L.I. Ali, N.H. Amin, A.A. Ebrahim, H.R. Mahmoud, Chem. Pap. 66
59] N. Ballarini, F. Cavani, L. Maselli, A. Montaletti, S. Passeri, D. Scagliarini, C. Flego, C.
Perego, J. Catal. 251 (2007) 423–436.
60] A.M. Beale, S.D.M. Jacques, E. Sacaliuc-Parvalescu, M.G. O’Brien, P. Barnes, B.M.
Weckhuysen, Appl. Catal. A: Gen. 363 (2009) 143–152.
(
2012) 722–732.
5] N. Rakmaka, W. Wiyaratnb, C. Bunyakana, J. Chungsiriporn, Chem. Eng. J. 162
2010) 84–90.
6] W. Suprun, M. Lutecki, R. Glaser, H. Papp, J. Mol. Catal. A: Chem. 342–343 (2011)
1–100.
7] K. Maaz, S. Karim, A. Mashiatullah, J. Liu, M.D. Hou, Y.M. Sun, J.L. Duan, H.J. Yao, D.
Mo, Y.F. Chen, Physica B 404 (2009) 3947–3951.
8] X. Liu, J. Liu, Z. Chang, X. Sun, Y. Li, Catal. Commun. 12 (2011) 530–534.
(
61] A.P. Vieira Soares, M.F. Portela, A. Kiennemann, L. Hilaire, Chem. Eng. Sci. 58
9
(
2003) 1315–1322.
62] A.A. Rownaghi, F. Rezaei, M. Stante, J. Hedlund, Appl. Catal. B: Environ. 119–120
2012) 56–61.
(