Study of the electrical and catalytic properties of spinels with CuFe2-xMnxO4 composition (x = 0, 0.4, 0.8, 1.6 and 2) Dedicated to Jacques Védrine.

Article abstract of DOI:10.1016/j.apcata.2014.09.048

CuFe_2-xMn_xO₄ spinels (x = 0, 0.4, 0.8, 1.6 and 2) were used as catalysts for the total oxidation of methane in the temperature range 450-800 °C and characterized by in situ electrical conductivity measurements. Their electrical conductivity was studied as a function of temperature and oxygen partial pressure and temporal responses during sequential exposures to air, methane-air mixture (reaction mixture) and pure methane in conditions similar to those of catalysis were analyzed. The catalytic activity decreased with increasing value of x in the CuFe_2-xMn_xO₄ spinels as follows: CuFe₂O₄ > CuFe_1.6Mn_0.4O₄ > CuFe_1.2Mn_0.8O₄ > CuFe_0.4Mn_1.6O₄ > CuMn₂O₄. CuFe₂O₄ appeared to be a p-type semiconductor, while CuMn₂O₄ a n-type semiconductor under air in the reaction temperature range but CuFe₂O₄ became n-type while CuMn₂O₄ p-type when contacted with methane. Both types of conduction mechanisms are present in CuFe_2-xMn_xO₄ spinels with x = 0.4, 0.8 and 1.6, being predominant one or another as a function of temperature. Nevertheless, they were of p-type under air in the reaction temperature range and became n-type when contacted with methane. The conducting behavior and redox properties of the spinels in the reaction temperature range were correlated with their catalytic behavior. For CuMn₂O₄ catalyst, the reaction mainly involves surface lattice O^- species, whereas for the other surface lattice O^2- anions are mainly involved. In all cases, the overall reaction mechanism can be assimilated to a Mars and van Krevelen mechanism.

Full text of DOI:10.1016/j.apcata.2014.09.048

Accepted Manuscript
Title: Study of the electrical and catalytic properties of spinels
with CuFe2−xMnxO4 composition (x = 0, 0.4, 0.8, 1.6 and 2)
Author: Ionel Popescu Amel Boudjemaa Nassima Helaili
˘ ˘
Yassine Bessekhouad Madalina Tudorache Khaldoun Bachari
Ioan-Cezar Marcu
PII:
S0926-860X(14)00610-3
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2014.09.048
APCATA 15029
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
22-7-2014
24-9-2014
27-9-2014
Please cite this article as: I. Popescu, A. Boudjemaa, N. Helaili, Y. Bessekhouad, M.
Tudorache, K. Bachari, I.-C. Marcu, Study of the electrical and catalytic properties of
spinels with CuFe2−xMnxO4 composition (x = 0, 0.4, 0.8, 1.6 and 2), Applied Catalysis
A, General (2014), http://dx.doi.org/10.1016/j.apcata.2014.09.048
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.

Dedicated to Jacques Védrine.
Study of the electrical and catalytic properties of spinels with CuFe2-xMnxO4
composition (x = 0, 0.4, 0.8, 1.6 and 2)
Ionel Popescu,a Amel Boudjemaa,b Nassima Helaili,b,c Yassine Bessekhouad,c, d Mădălina
Tudorache,e Khaldoun Bachari,b Ioan-Cezar Marcu,a, e, *
a
Research Center for Catalysts and Catalytic Processes, Faculty of Chemistry, University of
Bucharest, 4-12, Blv. Regina Elisabeta, 030018 Bucharest, Romania
b
Research Center in Analytical Chemistry and Physics (CRAPC), BP 248, Algiers RP 16004,
Algeria
c
Laboratory of Storage and Valorization of Renewable Energies, Faculty of Chemistry,
U.S.T.H.B., BP 32, 16111 El-Alia, Algiers, Algeria
d National Veterinary School, BP 161 El-Harrach, Algiers, Algeria
e
Laboratory of Chemical Technology and Catalysis, Department of Organic Chemistry,
Biochemistry and Catalysis, Faculty of Chemistry, University of Bucharest, 4-12, Blv. Regina
Elisabeta, 030018 Bucharest, Romania
Abstract
CuFe2-xMnxO4 spinels (x = 0, 0.4, 0.8, 1.6 and 2) were used as catalysts for the total oxidation of
methane in the temperature range 450-800 °C and characterized by in situ electrical conductivity
measurements. Their electrical conductivity was studied as a function of temperature and oxygen
partial pressure and temporal responses during sequential exposures to air, methane – air mixture
(reaction mixture) and pure methane in conditions similar to those of catalysis were analyzed.
The catalytic activity decreased with increasing the value of x in the CuFe2-xMnxO4 spinels as
follows: CuFe2O4 > CuFe1.6Mn0.4O4 > CuFe1.2Mn0.8O4 > CuFe0.4Mn1.6O4 > CuMn2O4. CuFe2O4
appeared to be a p-type semiconductor while CuMn2O4 a n-type semiconductor under air in the
reaction temperature range but CuFe2O4 became n-type while CuMn2O4 p-type when contacted
* Corresponding author. Tel.: +40 213051464; fax: +40 213159249. E-mail addresses:
ioancezar.marcu@g.unibuc.ro ; ioancezar_marcu@yahoo.com
Page 1 of 31

with methane. Both types of conduction mechanisms are present in CuFe2-xMnxO4 spinels with x
= 0.4, 0.8 and 1.6, being predominant one or another as a function of temperature. Nevertheless,
they were of p-type under air in the reaction temperature range and became of n-type when
contacted with methane. The conducting behavior and redox properties of the spinels in the
reaction temperature range were correlated with their catalytic behavior. For CuMn2O4 catalyst
the reaction mainly involves surface lattice O– species, whereas for the other surface lattice O2–
anions are mainly involved. In all cases the overall reaction mechanism can be assimilated to a
Mars and van Krevelen mechanism.
Keywords: electrical conductivity; CuFe2-xMnxO4 spinels; methane combustion
1. INTRODUCTION
The most widely used catalysts for volatile organic compounds (VOCs) abatement are noble
metals, generally supposed to be more active than metal oxides [1]. Nevertheless they have some
disadvantages like high sintering rates, volatility, poisoning in the presence of water or sulfur
compounds and high price [2]. Transition metal oxides which possess high chemical and thermal
stability, resistance to poisoning and low costs become very attractive to obtain active oxidation
catalysts of VOCs [3]. Moreover, it was already shown [1] that metal oxides can compete with
noble metals in terms of catalytic performances. Therefore, in recent years, many efforts were
directed towards the design of catalytic materials based on transition-metal oxides to replace
noble metal catalysts for VOCs abatement [4-17]. In that context, several studies showed that
spinel ferrites could be effective catalysts for the total oxidation of VOCs [18-24].
Ferrites are potential catalysts for this reaction due to their redox activity and their stability under
oxidative conditions [25]. A spinel-type ferrite structure presents two types of interstitial sites,
tetrahedral and octahedral, surrounded by four and six oxygen atoms, respectively [26]. The
formula of the spinel ferrites is AB2O4 where the AII ions occupy the tetrahedral and BIII ions the
octahedral positions. Such a structure is called normal spinel. If divalent cations AII occupy the
octahedral sites and BIII cations are distributed among tetrahedral and octahedral sites, the
structure is called inverse spinel. Most of the ferrite properties depend on the cation distribution
in the tetrahedral or octahedral sites in the spinel structure. The cation distribution is responsible
Page 2 of 31

of the number of vacancies and the non-stoichiometry, altering bulk and surface properties of the
corresponding system, which is of special interest for catalysis [27]. These structural properties
provide opportunities to extensively tune redox and electronic properties by varying the chemical
composition of the ferrites to optimize their catalytic properties. It is noteworthy that in the
spinel structure the octahedral sites are almost exclusively exposed at the surface of the solid
and, therefore, the octahedral cations play a key role in determining the catalytic activity [28].
Spinel ferrites are transition metal oxide semiconductors and their oxidation or reduction can be
evidenced by following the evolution of their electrical conductivity as a function of the nature
of the gas phase in contact with the solid. Thus, exposure of a p-type semiconducting oxide to
oxygen leads to the increase of its electrical conductivity with respect to that in inert atmosphere,
according to the following equilibrium:
O2(gas) ' O2−(ads) + h• ' 2O(−ads) + h• ' 2O2− + 2h• ' 2O×O
(1)
(ads)
where h• represents a hole and O×O a lattice oxygen anion of the solid. The electrons trapped by
the adsorbed oxygen are provided by the valence band of the oxide where new positive holes are
created. On the contrary, exposure to a reducing gas such as methane leads to the consumption of
lattice oxygen with the subsequent formation of oxygen vacancies. In this case electrons are
released in the valence band of the oxide resulting in the decrease of the positive holes
concentration and thus in the decrease of the electrical conductivity.
In contrast with the p-type oxides, the exposure of a n-type semiconducting oxide to oxygen
leads to the decrease of its electrical conductivity with respect to that in inert atmosphere,
according to the equilibrium depicted in Eq. (2):
O2(gas) + e' ' O2−(ads) + e' ' 2O(−ads) + 2e' ' 2O2− ' 2O×O
(2)
(ads)
where e' represents a free electron. The electrons trapped by the adsorbed oxygen are in this case
provided by the conduction band of the oxide and, consequently, the concentration of the charge
carriers decreases. The exposure to a reducing gas such as methane leads to the consumption of
lattice oxygen with the subsequent formation of oxygen vacancies but, in this case, electrons are
released in the conduction band of the oxide resulting in the increase of the free electrons
concentration and thus in the increase of the electrical conductivity.
It is well known that manganese and copper are key active components of the most active
combustion catalysts based on transition metal oxides [29, 30]. On the other hand, among the
Page 3 of 31

ferrites studied as catalysts for VOC abatement, CuFe2O4 shows superior catalytic activities [19].
Moreover, Cu-containing mixed spinels were shown to be more active than CuFe2O4 simple
spinel in CO oxidation [31]. Therefore, it was of interest to modify copper ferrite with
manganese for obtaining CuFe2-xMnxO4 mixed spinels and to study their catalytic properties in
the complete oxidation of methane as a VOC model compound. As these oxide materials are
expected to be semiconductors functioning in the catalytic oxidation via a redox mechanism, it
was also of interest to study their semiconductive and redox properties using in situ electrical
conductivity measurements with the aim of explaining their catalytic behavior.
2. EXPERIMENTAL
2.1. Catalysts preparation and characterization
The CuFe2-xMnxO4 spinels were prepared by solid-state reactions between appropriate amounts
of finely powdered oxides CuO 99.99 % (Aldrich), Fe2O3 99.98 % (Sigma-Aldrich) and Mn2O3
≥ 99.99 % (Aldrich). The reactions were carried out in alumina boats under atmospheric air at
800 °C for 24 h. Such a high temperature is necessary for the complete formation of spinel phase
[32].
Powder X-ray diffraction (XRD) patterns of the prepared spinels were recorded on a Philips
diffractometer using Cu Kα radiation (λ = 1.5418 Å) at a scanning speed of 2° min-1 in the 15-
80° 2θ range.
Surface areas of the catalysts were measured from the adsorption isotherms of nitrogen at −196
°C using the BET method with a Micromeritics ASAP 2020 sorptometer. Before analysis the
samples were outgassed at 200 °C overnight.
2.2. Catalytic measurements
The total oxidation of methane was carried out in a fixed bed quartz tube down-flow reactor
operating at atmospheric pressure. The internal diameter of the reactor tube was 18 mm. 1 cm3 (~
1.5 g) of catalyst supported by quartz wool was always used. The axial temperature profile was
measured using an electronic thermometer placed in a thermowell centered in the catalyst bed.
Quartz chips were used to fill the dead volumes before and after the catalyst bed to minimize
potential gas-phase reactions at high reaction temperatures. The reaction mixture consisted of 1
vol. % methane in air. The methane flow rate was controlled and measured using a Vögtlin Red-
Page 4 of 31

y smart controller. The air flow rate was controlled by a fine needle valve and was measured by a
capillary flow-meter. Steady-state catalytic measurements were performed at total gas hourly
space velocity (GHSV) of 15000 h-1, in the temperature range 450-800 °C.
Before each activity test, the reactor was heated to the desired temperature at a heating rate of 5
°C/min, and the reactant flows were set to give the desired reaction mixture. The system was
allowed to stabilize for about 30 min at the reaction temperature before the first product analysis
was made. Each run was carried out over a period of 1-2 hours, until two consecutive analyses
were identical. The feed and the reactor outflow gases were analyzed on-line by a Clarus 500
Gas-Chromatograph equipped with a thermal conductivity detector (TCD), using two packed
columns in series (6 ft Hayesep and 10 ft molecular sieve 5Å). After the catalytic test the reactor
was cooled down under the flow of reactants.
The amount of raw material transformed in reaction divided by the amount that was fed to the
reactor gave the conversion. Complete selectivity to CO2 and H2O was always observed. The
carbon balance was satisfactory in all runs varying within the margin of error of ± 2 %.
2.3. Electrical conductivity measurements
7
The samples were compressed at ca. 2.76 x 10 Pa using a Carver 4350.L pellet press to ensure
good electrical contacts between the catalyst grains. The obtained pellet was then placed in a
horizontal quartz tube between two platinum electrodes. Flow rates of gases flowing over the
sample were controlled by fine needle valves and were measured by capillary flow meters. The
temperature was controlled using thermocouples soldered to the electrodes and, when by passed,
they were used to determine the electrical conductivity σ of the samples, which can be expressed
by the formula:
1
t
σ =
×
(3)
R S
where R is the electrical resistance and t/S is the geometrical factor of the pellet including the
thickness t (ca. 3 mm) and the cross sectional area S of the pellet whose diameter was equal to 13
mm. The electrical resistance was measured with a megaohmmeter (FLUKE 177 Digital
Multimeter).
To compare the electrical conductivities of the samples, it is required that the solids have similar
textures and surface states. This requirement is easily fulfilled since all the samples have similar
Page 5 of 31

textures. Indeed, the electrical conductivity of semiconducting oxide powders can be written as:
σ = An
(4)
where n is the concentration of the main charge carriers and A is a coefficient of proportionality
which includes the mobility of the main charge carriers and the elementary charge of the electron
and depends on the compression of the powder and on the number and quality of contact points
between particles [33]. Since the samples were compressed at the same pressure and have similar
BET surface areas and since the electrical measurements were standardized, A can be considered
as identical for all the samples under identical conditions.
The common reference state for σ determination has been chosen under air at atmospheric
pressure and at 640 °C. At this temperature, which is in the range used in the catalytic reactions,
most of the ionically adsorbed species such as H3O+, HO– which would produce an additional
surface conductivity are eliminated. The solid was initially heated from room temperature to the
desired temperature at a heating rate of 5 °C/min.
3. RESULTS AND DISCUSSION
3.1. Catalysts characterization
The XRD patterns of the CuFe2-xMnxO4 spinels are presented in Fig. 1. Obviously, all the
samples showed a cubic spinel structure identified with PDFs 34-0425 and 71-1142
corresponding to CuFe2O4 and CuMn2O4 spinels, respectively. The well-defined sharp peaks
indicate a good crystalline quality of the materials. Nevertheless, weak diffraction lines
corresponding to CuO (PDF 80-1268) and α-Fe2O3 (PDF 33-0664) or α-Mn2O3 (PDF 73-1826)
side phases were observed in all the samples. This means that the initial amounts of single oxides
were incompletely consumed during the crystallization of spinel structure obviously due to the
moderate temperature (800 °C) used for the solid state reaction. Based on Scherer’s equation
[34] the crystallite size was estimated to be in the range 42-74 nm (Table 1). This demonstrates
that spinel phase with nanosized crystallites has been obtained.
As expected based on the method of preparation used, the specific surface areas of the CuFe2-
xMnxO4 spinels (Table 1) were relatively low and comparable with those of similar spinels
calcined at high temperatures [19]. The low surface areas observed suggests the agglomeration of
the nanosized crystallites [19].
Page 6 of 31

3.2. Catalytic properties
The CuFe2-xMnxO4 spinels with x = 0, 0.4, 0.8, 1.6 and 2, were tested in the total oxidation
reaction of methane in the temperature range from 500 to 770 °C and 1 % vol. CH4 in air (GHSV
= 15000 h-1). The methane conversions as a function of the reaction temperature were plotted in
Fig. 1a. It can be observed that the ignition temperature was around 530 °C for all the catalysts
and none of them gave a complete conversion of methane in the temperature range considered.
CuFe2O4 was the most active, while CuMn2O4 was the least active in the series studied. The
catalytic activity decreased with increasing the value of x in the CuFe2-xMnxO4 spinels as
follows: CuFe2O4 > CuFe1.6Mn0.4O4 > CuFe1.2Mn0.8O4 > CuFe0.4Mn1.6O4 > CuMn2O4. This order
was confirmed by comparison of the temperatures corresponding to 10 % methane conversion
(T10) and of the specific reaction rates of the different catalysts measured at 710 °C presented in
Table 2. Based on such comparisons, a detrimental effect of Mn incorporation in CuFe2O4 is
observed.
The specific reaction rate was calculated according to Eq. (5)
FCH ,inlet (1− x)
4
r = k
(5)
Q0
where x is the degree of methane conversion at the reaction temperature T (K) considered, Q0 the
total flow rate (L s-1), FCH the molar flow rate of methane at the inlet of the reactor (mol s-1)
4 ,inlet
and k the reaction rate constant (L g-1 s-1).
It is well known that methane total oxidation follows a pseudo first order kinetics [5, 7, 35, 36]
and the reaction rate constant can be calculated from the integrated equation of first order
kinetics for the plug-flow reactor
Q
k = 0 ln
W
1
(6)
1− x
where W is the mass of catalyst (g). The apparent activation energies (Ea) on the CuFe2-xMnxO4
catalysts were calculated from the slope of the lnk versus 1000/T plots shown in Fig. 1b. Only
methane conversions lower than ca. 30 % were considered to avoid diffusion limitations. For all
the catalysts studied a linear dependency was observed confirming the pseudo first order kinetics
for the methane total oxidation over the catalysts studied. The apparent activation energies (Ea)
on the different catalysts are reported in Table 2. They were in the range from 101.5 kJ mol-1 to
114.5 kJ mol-1. The lowest value calculated for CuFe2O4 can explain its highest catalytic activity.
Page 7 of 31

The introduction of Mn in the CuFe2O4 structure increases the apparent activation energy, then it
decreases with increasing the value of x in CuFe2-xMnxO4 spinels. At the same time, a
compensation effect is observed as the pre-exponential factor decreases following the same way
(Table 2) probably due to a decrease in the density of active sites with increasing Mn content.
This effect explains the decrease of the specific activity with increasing the Mn content.
3.3. Electrical conductivity measurements
a) Variations of the electrical conductivity as a function of temperature
The electrical conductivities of the ferrite samples have been measured as a function of
temperature to determine the activation energy of conduction Ec, both under air and under
nitrogen at atmospheric pressure, in the temperature range from 400 to 760 °C. The semi-log
plots [logσ = f(1/T)] obtained are given in Fig. 3. The linear variations observed showed that all
the solids behaved as semiconductors both under air and under nitrogen and their electrical
conductivities varied exponentially with temperature. The slopes of the semi-log plots enabled
one to calculate the Ec values presented in Table 1. It can be observed that the Ec values under
nitrogen were lower than under air for all the oxides but varied similarly, as follows: they
increased by substituting Mn to Fe in CuFe2O4 and by increasing its content in ternary spinels
CuFe2-xMnxO4 up to x = 1.6, then decreased for CuMn2O4. This behavior can be attributed to the
modification of the nature of the energy band structure, as well as to the shift of the energy level
impurities into the forbidden band gap with increasing Mn content in ternary spinels. It is
noteworthy that the activation energies of the CuFe2O4 and CuMn2O4 samples under air are
comparable with the values reported in the literature [37, 38].
The electrical conductivities of the catalysts under air and under nitrogen were compared in Fig.
3 for determining the nature of the semiconductor type of the solid by applying the Heckelsberg
criterion [39]. According to this criterion, if the conductivity of an oxide is higher under oxygen
(air) than under an inert gas, i.e. nitrogen, then it is a p-type semiconductor. In contrast, if the
conductivity of an oxide is lower under air than under nitrogen, then it is a n-type semiconductor.
This is due to fact that the electrons trapped by the adsorbed oxygen, which always adsorbs as
anionic species, are provided by the valence band of the oxide in the case of a p-type
semiconductor or by its conduction band in the case of a n-type semiconductor. In the first case
new positive holes are created and, thus, the electrical conductivity increases while in the second
Page 8 of 31

case the concentration of charge carriers (free electrons) is diminished and the electrical
conductivity decreases. As it can be seen in Fig. 3, the electrical conductivity was higher in air
flow than in nitrogen flow for CuFe2O4 and lower for CuMn2O4 within all the temperature range
considered. This clearly suggests that, according to the Heckelsberg criterion, the first compound
is a p-type semiconductor and the second one a n-type semiconductor. For the CuFe2-xMnxO4
samples, with x = 0.4, 0.8 and 1.6, the electrical conductivity in air flow was higher than in
nitrogen flow at high temperatures and lower at low temperatures. This suggests that both types
of conduction mechanisms are present in the ternary spinels, being predominant one or another
as a function of temperature. Thus, at low temperatures they behave as n-type semiconductors
while at high temperatures as p-type. The change from one type of semiconducting behavior to
another was observed at 540, 520 and 480 °C for CuFe1.6Mn0.4O4, CuFe1.2Mn0.8O4 and
CuFe0.4Mn1.6O4, respectively. This change took place at temperatures lower or around the
ignition temperature of the catalysts, i.e. 530 °C, and therefore all the ternary spinels were of p-
type in the reaction temperature range.
Note that the existence of both n- and p-type conduction mechanisms was already observed for
CuFe2O4 [37], the two competing mechanisms being associated to the following reactions:
Fe2+ ' Fe3+ + e'
Cu2+ ' Cu+ + h•
(7)
(8)
where e' stands for electrons and h• for holes. The type of charge carriers and their
concentration are determined by the kinetics of the reactions (7) and (8) [37].
b) Variations of the electrical conductivity under air as a function of oxygen partial
pressure
Fig. 4 shows the variations of σ as a function of oxygen pressure at 640 °C in a log–log plot for
the CuFe2-xMnxO4 samples with x = 0, 0.8 and 2. It clearly appeared that no dependence of σ on
the oxygen pressure was observed for all the samples. This would rather agree with the fact that
these materials are intrinsic semiconductors (∂σ/∂PO = 0). If that was true, the theory would
2
imply that the band-gap energy EG should be equal to twice the activation energy of conduction
Ec (EG = 2Ec). The UV spectra of the samples presented in Fig. 5 were used to determine their
band-gap energies. Indeed, the band-gap energies were obtained from the plots (αhν)2 vs. hν by
extrapolating the straight line which intercept the energy axes hν (Fig. S1). The absorption
Page 9 of 31

coefficient α, which depend on the wavelength, is proportional to ln[(Rmax − Rmin)/(R − Rmin)],
where the reflectance decreased from Rmax to Rmin due to the semiconductor absorption and R is
the intermediate for any intermediate energy photons [40]. Table 1 shows the band-gap energies
obtained. It can be observed that the band-gap energies determined from the UV spectra were
different from those calculated. Consequently, the solids cannot be considered intrinsic but
extrinsic semiconductors under our conditions with sources of electrical charge carriers
independent of oxygen pressure.
c) In situ electrical conductivity measurements under catalytic conditions.
To get information on the solids under conditions as close as possible to those of catalysis, the
electrical conductivity of the CuFe2-xMnxO4 samples with x = 0, 0.8 and 2 was measured at 640
°C, i.e. within the reaction temperature range, during sequential periods under air, under
methane-air mixture (reaction mixture) and under methane-nitrogen mixture (Fig. 6). The solids
were heated from room temperature to 640 °C, at a heating rate of 5 °C min-1 in air flow at
atmospheric pressure. After reaching the steady state under the air flow, a methane-air mixture
was passed over the samples. For the samples with x = 0 and 0.8 the electrical conductivity
slightly increased while for the sample with x = 2 it slightly decreased. The behavior of the
samples with x = 0 and 0.8 corresponds to the n-type semi-conducting character since, for oxide
semiconductors, the n-type criterion is ∂σ/∂PO2 < 0 or, considering methane as a reductant,
∂σ/∂PCH > 0, while that of the sample with x = 2 corresponds to the p-type semi-conducting
4
character since the p-type criterion is ∂σ/∂PO2 > 0 or ∂σ/∂PCH < 0. After reaching the steady
4
state under the methane-air flow, the atmosphere was again switched to air. The electrical
conductivity decreased and reached a plateau for the CuFe2-xMnxO4 samples with x = 0 and 0.8
confirming their n-type semi-conducting character, while for the CuMn2O4 sample (x = 2) it
increased confirming its p-type character. After reaching again the steady state under air flow,
the atmosphere was switched to a methane-nitrogen mixture before the air sequence was
repeated for testing the reversibility of the phenomena. Again, an important increase of the
electric conductivity was observed for the CuFe2-xMnxO4 samples with x = 0 and 0.8
corresponding to their n-type semi-conducting character. Under air their electrical conductivities
decreased immediately and reached a plateau confirming the reversibility of the phenomena
observed. However, for the ternary spinel sample (x = 0.8) the reversibility was not complete as
Page 10 of 31

the plateau corresponding to the new steady state under air was not the same as before the
methane-nitrogen mixture sequence, suggesting a new state of the solid that obviously
corresponds to a reduced oxide. For the CuMn2O4 sample (x = 2), the electrical conductivity
decreased under the methane-nitrogen flow suggesting again a p-type character for this sample.
Under air, its electrical conductivity increased and reached a plateau identical to that reached
before reduction suggesting a completely reversible reduction of the solid.
It is interesting to note that the introduction of methane in the flow in contact with the solid
changed the nature of the semiconductivity of all the solids. The CuFe2-xMnxO4 samples with x =
0 and 0.8 passed from p-type in air (Fig. 3) to n-type conduction, while CuMn2O4 passed from n-
type in air (Fig. 3) to p-type conduction. This change from p-type in air to n-type in the presence
of a hydrocarbon was already observed for Ti and Zr pyrophosphates [41].
The reversible redox process observed in the case of the CuFe2-xMnxO4 samples with x = 0 and
0.8 during successive sequences under methane-air mixture or methane-nitrogen mixture and
pure air can be explained by the oxidation of methane by lattice oxygen anions on the catalyst
surface, which leads to a partially reduced surface as already suggested for the oxidative
dehydrogenation of n-butane over metal pyrophosphates [41, 42] and for propane ammoxidation
over VAlO mixed oxides [43]. Thus, the increase of σ can be ascribed to the creation of anionic
vacancies:
O× + CH4 ꢀ Oxidation products + V×
(9)
O
O
followed by their spontaneous ionization:
VO× ' VO• + e'
(10)
(11)
VO• ' VO•• + e'
where O×O represents an oxygen anion of the solid in regular lattice points (O2− ), VO× a filled
anionic vacancy, VO• a singly ionized anionic vacancy and VO•• a doubly ionized anionic
vacancy. The release of free electrons into the conduction band of the solid accounts for the
observed increase in electrical conductivity. The subsequent introduction of air over the samples
restores the initial value of σ indicating the refilling of the previously created anionic vacancies
by dissociation of oxygen and capture of free electrons:
1/2O2 + VO• + e' ' O×O
(12)
Page 11 of 31

1/2O2 + VO•• + 2e' ' O×O
(13)
This behavior is in agreement with a Mars – van Krevelen type mechanism [44].
The case of CuMn2O4 sample (x = 2) is different as it behaves as a p-type semiconductor with
holes, h• , as the main charge carriers under the reaction conditions. If one considers that from
the chemical point of view, a positive hole corresponds to an electron vacancy in the valence
band of lattice O×O anions, i. e. the "chemical site" of a positive hole corresponds in fact to a
lattice O•O anion [45, 46], according to the reaction:
O×O + h• ' O•O
(14)
(15)
the initial activation step of methane over CuMn2O4 catalyst can be written:
– •
CH4 + O• ꢀ CH3 + (HO )
•
O
as already suggested for titanate-based catalysts [47] and LaCo1−xFexO3 perovskites [48].
Water elimination generates oxygen vacancies, according to the reaction:
2(HO–)• ' H2O(g) + O× + V••
(16)
O
O
Carbon dioxide formation also involves positive holes consumption and generates oxygen
vacancies, according to the simplified reactions:
C + 2O• + 2 h• ' CO2(g) + 2 V••
(17)
O
O
or
C + 2O× + 4 h• ' CO2(g) + 2 V••
(18)
O
O
where, for simplicity, we considered a carbon atom, C, instead of CHx species. The formation of
oxygen vacancies is in line with the observed reduction of the solid under methane-air and
methane-nitrogen flows. These vacancies must be filled in by gaseous oxygen in order to
reoxidize the solid:
1/2O2(g) + V•• ꢀ O× + 2 h•
(19)
O
O
This suggests that the overall reaction mechanism over the CuMn2O4 catalyst can also be
assimilated to a Mars – van Krevelen type mechanism [44], but in this case the methane
transformation involves mainly surface lattice O– species (O• ), whereas for the other CuFe2-
O
xMnxO4 catalysts, surface lattice O2– anions (O×O ) are mainly involved. Nevertheless, as the
CuFe2-xMnxO4 catalysts with x = 0, 0.4, 0.8 and 1.6 behave as p-type semiconductors under air,
Page 12 of 31

the initial activation step of methane involving O– species according to Eq. (15) cannot be
excluded.
It is interesting to note that as the O– species are known to be quite active species for alkanes’
oxidation [45], a high activity would be expected for the CuMn2O4 catalyst. The weak variations
of the electrical conductivity observed in that case (Fig. 6) suggest a low concentration of O–
species on the catalyst surface explaining its low catalytic activity.
Finally, we note that no direct correlations between the catalytic reaction and conduction
parameters of the studied ferrites were observed. This may be due to the complex nature of their
electrical behavior, which changes from a conduction type to another both as a function of
temperature and of the nature of the gas phase in contact with the solid.
4. CONCLUSION
Nanosized cubic spinels with CuFe2-xMnxO4 composition (x = 0, 0.4, 0.8, 1.6 and 2) have been
prepared by solid-state reactions between the corresponding oxides. Their catalytic activity in the
total oxidation of methane in the temperature range from 500 to 800 °C, decreased with
increasing the Mn content, i.e. the value of x, as follows: CuFe2O4 > CuFe1.6Mn0.4O4 >
CuFe1.2Mn0.8O4 > CuFe0.4Mn1.6O4 > CuMn2O4. None of the catalysts gave complete conversion
of methane in the temperature range chosen. A pseudo first order kinetics of the methane total
oxidation reaction was noted for all the catalysts. CuFe2O4 appeared to be a p-type
semiconductor while CuMn2O4 a n-type semiconductor under air in the reaction temperature
range but CuFe2O4 became n-type while CuMn2O4 p-type when contacted with methane. Both
types of conduction mechanisms are present in CuFe2-xMnxO4 spinels with x = 0.4, 0.8 and 1.6,
being predominant one or another as a function of temperature. Nevertheless, they were of p-type
under air in the reaction temperature range and became of n-type when contacted with methane.
In situ changes in the electrical conductivity under different gaseous atmospheres showed that all
the catalysts are partially reduced during catalysis and provide clear and direct evidence for a
redox mechanism with consumption of surface lattice oxygen. It can be assimilated to a Mars
and van Krevelen mechanism involving mainly surface lattice O– species for CuMn2O4 catalyst
and surface lattice O2– anions for all other catalysts studied.
ACKNOWLEDGMENTS
Page 13 of 31

Ionel Popescu expresses his gratitude to the management staff of the University of Bucharest,
especially to the Vice Rector Prof. Emil Barna, for the financial support within the frame of the
project UB 256/2014.
REFERENCES
1. M.R. Morales, B.P. Barbero, L.E. Cadús, Appl. Catal. B 67 (2006) 229-236.
2. R.M. Heck, R.J. Farrauto, Catalytic Air Pollution Control. Commercial Technology, Van
Nostrand Reinhold, New York, 1995.
3. R. Craciun, B. Nentwich, K. Hadjiivanou, H. Knozinger, Appl. Catal. A 243 (2003) 67-79.
4. T. Machej, E.M. Serwicka, M. Zimowska, R. Dula, A. Michalik-Zym, B. Napruszewska, W.
Rojek, R. Socha, Appl. Catal. A 474 (2014) 87-94.
5. M. Răciulete, G. Layrac, D. Tichit, I.C. Marcu, Appl. Catal. A 477 (2014) 195-204.
6. N. T. Binh, C. Gennequin, R. Cousin, E. A. Zhilinskaya, H. L. Tidahy, N. M. Son, V. T. Thu
Ha, L. T. Son, N. T. Minh Thu, P. X. Nui, S. Siffert, Chimica Oggi / Chemistry Today 31
(2013) 51-54.
7. A. Urdă, I. Popescu, T. Cacciaguerra, N. Tanchoux, D. Tichit, I.C. Marcu, Appl. Catal. A
464-465 (2013) 20-27.
8. J. Ludvíková, K. Jirátová, J. Klempa, V. Boehmová, L. Obalová, Catal. Today 179 (2012)
164- 169.
9. A. Kaddouri, N. Dupont, P. Gélin, P. Delichère, Catal. Lett. 141 (2011) 1581-1589.
10. S. Tanasoi, G. Mitran, N. Tanchoux, T. Cacciaguerra, F. Fajula, I. Săndulescu, D. Tichit, I.C.
Marcu, Appl. Catal. A 395 (2011) 78-86.
11. F. Kovanda, K. Jirátová, Appl. Clay Sci. 53 (2011) 305-316.
12. R. Dziembaj, M. Molenda, L. Chmielarz, M. Drozdek, M. M. Zaitz, B. Dudek, A. Rafalska-
Łasocha, Z. Piwowarska, Catal. Lett. 135 (2010) 68-75.
13. A. Urdă, I. Popescu, I.C. Marcu, G. Cârjă, N. Apostolescu, I. Săndulescu, Rev. Chim. 61
(2010) 267-271.
14. S. Rousseau, S. Loridant, P. Delichère, A. Boreave, J.P. Deloume, P. Vernoux, Appl. Catal.
B 88 (2009) 438-447.
15. I. Popescu, A. Urda, T. Yuzhakova, I.C. Marcu, J. Kovacs, I. Săndulescu, C. R. Chim. 12
(2009) 1072-1078.
Page 14 of 31

16. Y. Li, Y. Guo, B. Xue, Fuel Proc. Technol. 90 (2009) 652-656.
17. S. Tanasoi, N. Tanchoux, A. Urdă, D. Tichit, I. Săndulescu, F. Fajula, I.C. Marcu, Appl.
Catal. A 363 (2009) 135-142.
18. N. Rezlescu, E. Rezlescu, P.D. Popa, C. Doroftei, M. Ignat, Appl. Catal. B 158-159 (2014)
70-75.
19. N. Rezlescu, E. Rezlescu, P.D. Popa, E. Popovici, C. Doroftei, M. Ignat, Mater. Chem. Phys.
137 (2013) 922-927.
20. C. Munteanu, M. Caldararu, D. Gingasu, M. Feder, L. Diamandescu, N. I. Ionescu, Reac.
Kinet. Mech. Catal. 104 (2011) 357-368.
21. T. Tsoncheva, E. Manova, N. Velinov, D. Paneva, M. Popova, B. Kunev, K. Tenchev, I.
Mitov, Catal. Commun. 12 (2010) 105-109.
22. M. Florea, M. Alifanti, V.I. Parvulescu, D. Mihaila-Tarabasanu, L. Diamandescu, M. Feder,
C. Negrila, L. Frunza, Catal. Today 141 (2009) 361-366.
23. A. Urdă, A. Herraïz, Á. Rédey, I.C. Marcu, Catal. Commun. 10 (2009) 1651-1655.
24. J.E. Tasca, C.E. Quincoces, A. Lavat, A.M. Alvarez, M.G. González, Ceram. Int. 37 (2011)
803-812.
25. J.B. Silva, C.F. Diniz, R.M. Lago, N.D.S. Mohallem, J. Non-Cryst. Solids 348 (2004) 201-
204.
26. A. Goldman, Modern Ferrite Technology, 2nd ed., Springer, Pittsburg, 2006.
27. J.A. Toledo-Antonio, N. Nava, M. Martínez, X. Bokhimi, Appl. Catal. A 234 (2002) 137-
144.
28. J.P. Iacobs, A. Maltha, J. Drimal, V. Ponec, H.H. Brongersma, J. Catal. 47 (1994) 294-300.
29. T.V. Choudhary, S. Banerjee, V.R. Choudhary, Appl. Catal. A 234 (2002) 1-23.
30. W.B. Li, J.X. Wang, H. Gong Catal. Today 148 (2009) 81-87.
31. C.U. Aniz, T.D.R. Nair, Open J. Phys. Chem. 1 (2011) 124-130.
32. K. Faungnawakij, Y. Tanaka, N. Shimoda, T. Fukunaga, S. Kawashima, R. Kikuchi, K.
Eguchi, Appl. Catal. A 304 (2006) 40-48.
33. J.-M. Herrmann, in: Catalyst Characterization, Physical Techniques for Solid Materials, B.
Imelik and J. C. Védrine (eds.), Plenum Press, New York, 1994, ch. 20.
34. H. Klung, L. Alexander, X-ray Diffraction Procedures, Wiley, New York, 1962.
Page 15 of 31

35. Z. Jiang, J. Yu, J. Cheng, T. Xiao, M.O. Jones, Z. Hao, P. P. Edwards, Fuel Proc. Technol. 91
(2010) 97-102.
36. S. Pengpanich, V. Meeyoo, T. Rirksomboon, K. Bunyakiat, Appl. Catal. A 234 (2002) 221-
233.
37. R.K. Selvan, C.O. Augustin, L.J. Berchmans, R. Saraswathi, Mat. Res. Bull. 38 (2003) 41-
54.
38. D.B. Ghare, A.P.B. Sinha, L. Singh, J. Mater. Sci. 3 (1968) 389-394.
39. L. F. Heckelsberg, A. Clark, G. C. Bailey, J. Phys. Chem. 60 (1956) 559-561.
40. V. Kumar, T.P. Sharma, J. Phys. Chem. Solids 59 (1998) 1321-1325.
41. I.C. Marcu, J.M.M. Millet, J.M. Herrmann, Catal. Lett. 78 (2002) 273-279.
42. I.C. Marcu, I. Săndulescu, Y. Schuurman, J.M.M. Millet, Appl. Catal. A 334 (2008) 207-216.
43. J. Bilde, C. Janke, C. Lorentz, P. Delichere, I. Popescu, I.C. Marcu, S. Loridant, A. Brückner,
J.M.M. Millet, J. Phys. Chem. C 117 (2013) 22926-22938.
44. S. Mars, N. van Krevelen, Chem. Eng. Sci., Special Sup. 9 (1954) 41-57.
45. J.M. Herrmann, P. Vernoux, K.E. Béré, M. Abon, J. Catal. 167 (1997) 106-117.
46. J.M.M. Millet, Top. Catal. 38 (2006) 83-92.
47. I. Popescu, I. Săndulescu, Á. Rédey, I.C. Marcu, Catal. Lett. 141 (2011) 445-451.
48. I. Popescu, Y. Wu, P. Granger, I.C. Marcu, Appl. Catal. A 485 (2014) 20-27.
Page 16 of 31

FIGURES CAPTION
Figure 1. XRD patterns of the CuFe2-xMnxO4 spinels: (a) CuFe2O4; (b) CuFe1.6Mn0.4O4; (c)
CuFe1.2Mn0.8O4; (d) CuFe0.4Mn1.6O4; (e) CuMn2O4. Symbols: ▼ – cubic ferrite; ■ – α-Fe2O3; # –
α-Mn2O3; ● – CuO.
Figure 2. The light-off curves (a) and the corresponding Arrhenius plots (b) for the combustion
of methane over the spinel catalysts (reaction conditions: GHSV = 15000 h-1, 1 vol. % CH4 in
air): „ – CuFe2O4;  – CuFe1.6Mn0.4O4; z – CuFe1.2Mn0.8O4; ¾ – CuFe0.4Mn1.6O4; S –
CuMn2O4.
Figure 3. Arrhenius plots for the electrical conductivity σ of CuFe2O4 (a), CuFe1.6Mn0.4O4 (b),
CuFe1.2Mn0.8O4 (c), CuFe0.4Mn1.6O4 (d) and CuMn2O4 (e) under air (¾ and filled symbols) and
under nitrogen (ª and open symbols) (σ in ohm-1 cm-1).
Figure 4. Variation of σ as function of the oxygen pressure for CuFe2O4 („), CuFe1.2Mn0.8O4
(z) and CuMn2O4 (S) at 640°C (PO in atm; σ in ohm-1 cm-1).
2
Figure 5. UV-VIS spectra of the CuFe2-xMnxO4 spinels: (a) CuFe2O4; (b) CuFe1.6Mn0.4O4; (c)
CuFe1.2Mn0.8O4; (d) CuFe0.4Mn1.6O4; (e) CuMn2O4.
Figure 6. Variation of the electrical conductivity under sequential exposures to air, methane – air
mixture (reaction mixture) and methane – nitrogen mixture for CuFe2O4 („), CuFe1.2Mn0.8O4
(z) and CuMn2O4 (S) at 640°C (σ in ohm-1 cm-1).
Page 17 of 31

Table 1. Physico-chemical characteristics of the CuFe2-xMnxO4 spinels.
c
Catalyst
SSA a (m2 g-1) D b (nm) Ec (kJ mol-1)
Theoretical Experimental
EG d (eV)
EG d (eV)
in air
in N2
CuFe2O4
2.2
74
42
47
43
58
54.8
69.8
71.4
77.3
67.1
51.2
54.4
59.4
70.0
52.5
1.14
1.44
1.48
1.60
1.40
1.86
1.81
1.80
1.66
1.78
CuFe1.6Mn0.4O4 3.4
CuFe1.2Mn0.8O4 2.6
CuFe0.4Mn1.6O4 2.5
CuMn2O4
5.5
a Specific surface area.
b Average crystallite size.
c Activation energy of conduction in the temperature range 400-760°C.
d Band-gap energy.
Table 2. Catalytic performances in methane total oxidation of the CuFe2-xMnxO4 spinel catalysts.
Catalyst
T10 (°C) Reaction rate at 710 °C
lnk0 (k0 in Ea
L g-1 s-1)
(kJ mol-1)
Specific
Intrinsic
(×10-7 mol g-1 s-1) (×10-7 mol m-2 s-1)
CuFe2O4
637
2.7
2.1
1.9
1.8
1.6
1.2
0.6
0.7
0.7
0.3
5.3
6.5
6.1
5.5
5.1
101.5
114.5
111.2
106.8
104.2
CuFe1.6Mn0.4O4 667
CuFe1.2Mn0.8O4 667
CuFe0.4Mn1.6O4 673
CuMn2O4
680
Page 18 of 31

Figure 1.
#
(e)
#
(d)
(c)
(b)
▼
▼
■
▼
▼
▼
▼
■
▼
▼
▼
(a)
●
■
●
15
25
35
45
55
65
75
2-theta (°)
Page 19 of 31

Figure 2.
(a)
70
60
50
40
30
20
10
0
500
550
600
650
700
750
800
Temperature (°C)
Page 20 of 31

(b)
-6
-7
-8
-9
-10
-11
0.9
1
1.1
1.2
1.3
103/T (K-1)
Page 21 of 31

Figure 3.
(a)
-1
-2
-3
-4
0.8
1
1.2
1000/T (K-1)
1.4
1.6
Page 22 of 31

(b)
-1
-2
-3
-4
0.8
1
1.2
1.4
1.6
1000/T (K-1)
Page 23 of 31

(c)
-1
-2
-3
-4
0.8
1
1.2
1.4
1.6
1000/T (K-1)
Page 24 of 31

(d)
-2
-3
-4
-5
0.8
1
1.2
1.4
1.6
1000/T (K-1)
Page 25 of 31

(e)
-1
-2
-3
-4
0.8
1
1.2
1000/T (K-1)
1.4
1.6
Page 26 of 31

Figure 4.
-2.1
-2.2
-2.3
-2.4
-2.5
-1.5
-1.4
-1.3
-1.2
-1.1
-1
-0.9
-0.8
-0.7
-0.6
lg PO2
Page 27 of 31

Figure 5.
(e)
(d)
(c)
(b)
(a)
200
400
600
800
1000
Wavelength (nm)
Page 28 of 31

Figure 6.
← Air →│← CH4 + Air →│← Air →│← CH4 + N2 →│← Air →│
-1.9
-2
-2.1
-2.2
-2.3
-2.4
-2.5
0
20
40
60
80
100
120
140
160
180
200
Time (min)
Page 29 of 31