Journal of Catalysis 338 (2016) 47–55

Contents lists available at ScienceDirect

Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat

Electronic properties and catalytic performance for DME combustion

of lanthanum manganites with partial B-site substitution
Robinson Dinamarca a, Catherine Sepúlveda a, Eduardo J. Delgado a, Octavio Peña b, J.L.G. Fierro c,
Gina Pecchi a,⇑
a
Departamento Físico-Química, Fac Ciencias Químicas, Universidad Concepción, Concepción, Chile
b
Institut Sciences Chimiques de Rennes, UMR 6226, CNRS, Université Rennes 1, 35042 Rennes Cedex, France
c
Instituto de Catálisis y Petroleoquímica, CSIC, 28049 Madrid, Spain

a r t i c l e i n f o a b s t r a c t

Article history: The effect on the geometric, electronic, and catalytic properties of LaMnO3 oxides of replacement of 50%
Received 18 December 2015 Mn with Cr, Fe, Co, and Ni in combustion reactions of dimethyl ether has been studied. The similarity in
Revised 6 February 2016 the ionic radius does not produce larger differences in the geometric properties of the compounds,
Accepted 9 February 2016
whereas the differences in the stability of their cation valences produce significant changes in their
electronic properties. The high stability of Cr3+ and Fe3+ species does not increase the Mn4+/Mn3+ redox
pair ratio in LaMn0.5Cr0.5O3 and LaMn0.5Fe0.5O3. Meanwhile, the high stability of Co2+ and Ni2+ species
Keywords:
favors the oxidation of Mn3+ to Mn4+ and the mobility of anionic oxygen in the LaMn0.5Co0.5O3 and
Lanthanum
Manganites
LaMn0.5Ni0.5O3 catalysts, which improves their catalytic performance.
DME combustion Ó 2016 Elsevier Inc. All rights reserved.
Reducibility
Perovskite

1. Introduction charge neutrality [12]. In contrast, the partial replacement of the

B metal with another B0 cation may induce structural modifications
The easy transport, low production of polluting particles, and similar to changes in the oxidation state of the original B cation.
high thermal efficiency of dimethyl ether (DME) make it a useful Pure LaBO3 (B = 3d transition metal) has been widely used as an
low-cost replacement for liquefied petroleum gas (LPG) and diesel oxidation catalyst for environmental applications [8,13,14], in par-
fuel [1]. Therefore, DME has emerged as a convenient energy car- ticular LaMnO3, which has been used as a catalyst for the oxidation
rier for distributive power generation and diesel replacement. of toluene [15], carbon monoxide [6] and NH3 [16]. Accordingly, in
The catalytic combustion of DME has been studied on noble this article, we report a systematic study of the effects of equimolar
metal-supported catalysts [1,2] and recently on mixed oxides B-metal substitution in LaMnO3 (LaMn0.5B0.5O3; B = Cr, Fe, Co, Ni)
such as Ce-doped manganese oxide octahedral molecular sieves on the catalytic performance in DME combustion.
(OMS-2) [3], cryptomelane oxides [4], and Al2O3 pillared layered
compounds [5]. Meanwhile, on perovskite-type oxides, well-
known total oxidation hydrocarbon catalysts, no previous reports 2. Experimental
can be found for DME combustion. Mixed oxides with
perovskite-type structures, which are represented by the general 2.1. Preparation of the catalysts
formula ABO3, where A is usually an alkaline earth or rare earth
element and B is a transition metal cation (Mn, Fe, Co) [6–8], can Pure LaMnO3 and substituted La(Mn0.5B0.5)O3 (B = Cr, Fe, Co, or
be highly active catalysts for the deep oxidation of hydrocarbons Ni) perovskites were prepared by the sol–gel self-combustion
[9–11]. The catalytic activities of perovskite-type oxides in com- method [17]. Glycine (H2NCH2CO2H), which was used as the igni-
bustion reactions are essentially controlled by the B cation. The tion agent, was added to an aqueous solution of metal nitrates with
partial replacement of the A metal may create oxygen vacancies the appropriate stoichiometry to reach NO3/NH2 (molar ratio) = 1.
and changes in the oxidation state of the B cation to preserve The resulting solution was slowly evaporated until a vitreous gel
was obtained. The gel was heated to approximately 250 °C, the
temperature at which the ignition reaction occurs, producing a
⇑ Corresponding author. powdered precursor. The solids were crushed and sieved to
E-mail address: gpecchi@udec.cl (G. Pecchi). obtain a particle size smaller than 200 lm prior to calcination.

http://dx.doi.org/10.1016/j.jcat.2016.02.011
0021-9517/Ó 2016 Elsevier Inc. All rights reserved.
48 R. Dinamarca et al. / Journal of Catalysis 338 (2016) 47–55

The powders were then subjected to air thermal treatment calcina- ionization detector (FID), after separation on a Supelco 25462
tion at 700 °C for 9 h to eliminate the remaining carbon and obtain 30 m capillary and conversion to methane over a Ni catalyst at
a perovskite structure. 360 °C. External diffusion was excluded through a series of mass
transfer tests, and it was ignored because the particle size was less
2.2. Characterization than 0.4 mm (i.e., 40–60 mesh). To evidence the reproducibility of
the catalytic activity measurements and lack of deactivation pro-
Chemical analysis of the samples was conducted by atomic cesses, after the first lightoff curve, the catalysts were cooled down
absorption spectrometry (AAS) using a Perkin Elmer Model 3100 and new lightoff curve was conducted. The kinetic parameters
instrument. The presence of a B-site cation (IV) was followed by were evaluated on the basis of the specific and intrinsic reaction
a redox titration, dissolving the samples in a known excess of an rate and the apparent activation energy (Ea) by applying an Arrhe-
ammonium ferrous sulfate standard and titrating the excess of Fe nius plot.
(II) with potassium permanganate. The titration was performed
twice for each sample, with the reproducibility of results always
3. Results and discussion
within 2% [18]. The specific areas were calculated using the BET
method from the nitrogen adsorption isotherms obtained on a
3.1. Chemical composition and texture
Micromeritics Model ASAP 2010 apparatus at 196 °C. X-ray pow-
der diffraction (XRD) patterns were obtained with nickel-filtered
The elemental composition of the LaMnO3 (LaMn0.5B0 0.5O3;
Cu Ka1 radiation (k = 1.5418 Å) using a Rigaku diffractometer and 0
B = Cr, Fe, Co, Ni) perovskites determined by AAS shows metal con-
collected in the 2h range of 20–70°. Temperature-programmed
tents close to the nominal ones (Table 1). The specific BET areas
reduction (TPR), temperature-programmed desorption of O2 (O2
summarized in Table 1 show the largest surface area of
TPD), and temperature-programmed desorption of NH3 (NH3
11 m2 g1 for pure LaMnO3 perovskite, similar to other reported
TPD) experiments were performed in a TPR/TPD 2900 Micromerit-
SBET values for LaMnO3 (13–16 m2 g1) [19] that decreased upon
ics system with a thermal conductivity detector (TCD). Prior to the
B-site substitution [8,20].
reduction experiments the samples (30 mg) were thermally trea-
ted under an air stream at 300 °C to remove water and other con-
taminants. The reduction profiles were recorded under 5% H2/Ar 3.2. Redox titration
flow at 40 mL min1 at a heating rate of 10 °C min1 from room
temperature to 700 °C. For the O2 TPD experiments, the samples Considering the differences in thermodynamic stability with
were preheated in an O2 flow for 1 h at 700 °C, cooled to room tem- respect to the reduction of some of the first-row transition ele-
perature in the same atmosphere, and then switched to a helium ments (Cr, Fe, Co, and Ni) and the stability of Mn4+/Mn3+/Mn2+ ions,
flow with the oxygen desorption monitored using a TCD. For the we cannot neglect changes in the respective oxidation states of
NH3 TPD experiments, ammonia pulses were dosed to saturate manganese as a consequence of their partial replacement by B0
the catalyst surface at 100 °C, and then the samples were cooled ions. A redox titration, a well-known analytical method, in which
to room temperature; once the baseline was restored, the temper- stabilized Fe2+ Mohr salt can be oxidized to Fe3+ by another cation
ature was linearly increased to 500 °C. The XPS measurements (IV) present in the solution, has been used to quantify the Mn4+
were performed using a VG Escalab 200R electron spectrometer content. Between 32% and 35% of Mn(IV) has been reported for
equipped with a hemispherical electron analyzer and Mg Ka LaMnO3, which can increase upon B(II) cation substitution [21] or
(1253.6 eV) X-ray source. Prior to analysis, the samples were the preparation method [22]. Table 1 displays the Mn4+ content
degassed at 300 °C for 1 h inside the pretreatment chamber of of the perovskites carried out twice. For LaMnO3, an Mn4+ content
the spectrometer. Charging effects on the samples were corrected of 31% is closer to the reported values [23,24] and a decrease is
by taking the C1s peak of adventitious carbon at 284.8 eV. The detected upon B-substitution to a greater extent for Fe and Cr sub-
intensity of the peaks was estimated by calculating the integral stitution. This result can be explained by assuming that the nonsto-
of each peak after subtracting an S-shaped background and fitting ichiometry and amount of Mn4+ should decrease upon introduction
the experimental curve to a combination of Gaussian and Lorent- of a cation in a stable oxidation state (III) into the perovskite lattice,
zian lines. Atomic ratios were calculated using peak areas normal- whereas the larger stability of Co2+ and Ni2+ in comparison to Co3+
ized on the basis of acquisition parameters, sensitivity, and and Ni3+ species in air appears to account for the stabilization of a
transmission factors provided by the manufacturer and deter- larger content of Mn4+ ions in the perovskite structure.
mined from the corresponding peak intensities, corrected with tab-
ulated sensitivity factors, with a precision of ±7%. 3.3. X-ray diffraction

2.3. Catalytic activity In the XRD patterns shown in Fig. 1, well-defined diffraction
peaks appear at 2h of 22.8°, 32.6°, and 46.8°, characteristic of the
The evaluation of catalytic activity in the total combustion of (1 1 2), (1 1 0), and (2 2 0) planes of a perovskite structure. The
DME was performed in a conventional flow reactor under atmo- diffraction profiles of LaMnO3 indicate a rhombohedral geometry
spheric pressure using 0.3 g of the catalysts diluted in 1.5 g of
SiC. The activity was measured in the temperature range 150– Table 1
450 °C. The reactant mixture was fed into the reactor at Bulk composition (nominal values in parentheses) in wt.% for B = Cr, Fe, Co, Ni,
specific area, t factor, and Mn4+ content.
50 mL min1, and the temperature was linearly increased up to
the required conditions and then kept constant for 20 min. Subse- Mna Ba SBET t factor Mn4+
quently, it was increased to a new temperature using the same (m2 g1) content (%)

heating rate (1 °C min1). Several isothermal steps were performed LaMnO3 22.3 (22.7) – 11 0.957 31
until complete conversion was achieved using a reaction mixture LaMn0.5Cr0.5O3 11.2 (11.4) 9.5 (10.8) 7 0.964 18
LaMn0.5Fe0.5O3 11.9 (11.3) 10.9 (11.5) 6 0.957 17
of CH3OCH3, O2, and He (molar ratio 1:10:40) using a gas hourly
LaMn0.5Co0.5O3 11.5 (11.2) 11.6 (12.1) 4 0.960 24
space velocity (GHSV) of 10 dm3 g1 h1 [4]. The effluents from LaMn0.5Ni0.5O3 11.0 (11.3) 11.6 (12.0) 4 0.974 25
the reactor were analyzed using He as the carrier gas and an Perkin a
Estimated error below 5%.
Elmer Clarus Model 580 online gas chromatograph with a flame
 R. Dinamarca et al. / Journal of Catalysis 338 (2016) 47–55 49

(32-0484) [25] with almost no changes in the well-defined La in the inset in Fig. 1, the typical doublet of the rhombohedral struc-
(Mn0.5B0.5)O3 (B = Cr, Fe, Co, or Ni) substituted perovskite structure. ture is sharper for LaMnO3 and the Cr-substituted perovskite. The
The absence of diffraction peaks belonging to single or mixed oxi- broader nature of this peak indicates a change to a rhombohe-
des suggests a high proportion of B cations [26]. Highly dispersed dral–hexagonal structure with Co, Fe, and Ni substitution. The
segregated phases cannot be rejected. On the other hand, a small crystal size values evaluated using the Debye–Scherrer equation
peak at 2h = 43° was observed in Co- and Ni-doped samples, corre- [27] applied to the diffraction line at 2h = 32.6° are summarized
sponding to La2CoO4 (34-1081) and La2NiO4 (11-557) spinels as in Table 1. Smaller crystallite sizes between 16 and 18 nm were
segregated phases. Moreover, a close-up between 67° 6 2h 6 70° observed for the rhombohedral perovskite structures, while larger
indicates some changes in the crystallographic structure. As shown sizes between 24 and 28 nm were observed for Fe-, Co-, and Ni-
substituted perovskites, suggesting a deformation of the well-
defined perovskite structure by the presence of these transition
metals.

3.4. Temperature-programmed reduction

Fig. 2 shows the TPR profiles of the perovskites studied. For

perovskite-type oxides, Isupova et al. [28] reported that in TPR
three main peaks can be distinguished: (i) the lowest-
temperature H2-consumption peak, which quantitatively corre-
sponds to the removal of no more than a monolayer of surface oxy-
gen (not present in the samples that mean that the pretreatment
was correct); (ii) H2-consumption peak in the range 400–600 °C
corresponding to the reduction of the B4+ to B3+ cations (or the
removal of oxygen from the coordination sphere of highly charged
cations with the formation of oxygen vacancies); (iii) consumption
at temperatures higher than 600 °C corresponding to deeper reduc-
tion of B3+ cations.
For pure LaMnO3, the reduction peak at 350 °C with a shoulder
at 420 °C corresponds to reduction of Mn4+ to Mn3+, and at 800 °C
Fig. 1. X-ray diffraction patterns of LaMnO3 and La(Mn0.5B0.5)O3 perovskites.
the destruction of the perovskite structure (Mn3+ to Mn2+) is
observed [29]. This last reduction step is clearly started and not
completed in Fig. 2 because it is outside of the temperature range
studied. For the substituted perovskites, two reducibility profiles
different from that of pure LaMnO3 are seen: two larger reduction
steps for Co- and Ni- and, only one and smaller reduction steps for
Cr- and Fe-perovskites. This behavior can be explained by consid-
ering the reducibility of the respective pure LaBO3 (B = Cr, Fe, Co,
or Ni) perovskite counterparts. The TPR profiles of the so-called
nonreducible LaCrO3 [30] and LaFeO3 [31] perovskites do not show
any reduction peaks, whereas two reduction peaks are reported for
LaCoO3 [32] and LaNiO3 [33,34] perovskites. The profiles observed
in Fig. 2 show intermediate behavior of the respective pure per-
ovskites, a decrease in the reduction peaks of Mn4+ to Mn3+ for
the Cr- and Fe-substituted perovskite structures, and an increase
for the Co- and Ni-substituted perovskite structures. Moreover,
the reduction profile attributed to manganese (Mn4+ to Mn3+) can
be related also to the substituted B-metal electronegativity accord-
ing to the Pauling scale, which follows the order 1.91 (Ni) > 1.88
(Co) > 1.83 (Fe) > 1.66 (Cr) > 1.55 (Mn). Therefore, the increase of
the electronegativity and then the weaker Mn–O bonds in the
Mn–O–B bridge could favor the reducibility of Mn in the substi-
tuted perovskites. The total hydrogen consumption between 200
and 500 °C, a measure of the amount of Mn4+, was evaluated by
Fig. 2. TPR profiles of LaMnO3 and La(Mn0.5B0.5)O3 perovskites. deconvolution of the area under the curve using a Lorentzian peak
and is given in Table 2. The H2 consumption decreased for Cr- and

Table 2
Crystal size (dhkl), hydrogen consumption in the TPR profiles, post-TPR specific area, a-oxygen, and NH3 desorbed content.

dhkl (nm) H2 TPR (mmol H2 g1) SBET (m2 g1) post-TPR a-oxygen (mmol O2 g1) Desorbed (mmol NH3 g1)
<200 °C 200–500 °C
LaMnO3 16 2.6 22 1.7 1.7 1.7
LaMn0.5Cr0.5O3 18 1.5 11 1.6 1.2 1.4
LaMn0.5Fe0.5O3 25 2.1 10 1.2 1.2 1.5
LaMn0.5Co0.5O3 28 1.7 7 1.0 1.3 1.9
LaMn0.5Ni0.5O3 25 2.9 8 1.2 0.5 3.0
50 R. Dinamarca et al. / Journal of Catalysis 338 (2016) 47–55

Fe-substitution compared with pure LaMnO3, in accordance with 3.5. Temperature-programmed desorption of oxygen
previous results [23]. On the other hand, a similar effect of
lanthanum substitution for divalent alkaline-earth cations is The evolution of oxygen during temperature-programmed des-
observed for Co- and Ni-substitution, where a decrease in the num- orption (O2 TPD) experiments is closely related to the redox prop-
ber of O2 ions in the lattice [35] is reported. To understand the erties [38]. Fig. 4 shows the O2 TPD profiles of the perovskites
correlation between the Mn4+ content and H2 consumption in the calcined at 700 °C. Basically the O2 TPD of perovskite-type oxides
TPR profiles, the theoretical hydrogen consumption of indicate three types of desorbed oxygen: (i) a first desorption peak
2.1 mmol H2 g1 for the complete reduction of Mn3+ to Mn2+ in a at temperatures lower than 100 °C, which corresponds to physi-
stoichiometric LaMnO3 perovskite can be considered. Lisi et al. cally adsorbed oxygen species; (ii) a second peak, which desorbs
[36] reported for LaMnO3 a consumption of 2.70 mmol H2 g1 in between 200 and 500 °C, attributed to chemically adsorbed oxy-
the TPR profile, indicative of 23% of Mn4+ presence, whereas Liu gen, called a oxygen; and (iii) lattice oxygen that desorbs at tem-
et al. [21] reported 2.41 mmol H2 g1, which indicates closer to peratures up to 600 °C, so-called b oxygen, which is associated
100% content of Mn4+. The number of Mn4+ cations calculated from with oxygen species occupying inner vacancies. So mixing of type
redox titration follows the same trend as the H2 consumption asso- (i) and type (ii) species was avoided. For LaMnO3, physically
ciated with reduction of Mn4+ to Mn3+, indicative of the correct adsorbed oxygen at ca. 80 °C, chemically adsorbed oxygen that
assignment of the reduction peak. To corroborate the modification desorbs between 200 and 500 °C [30], and lattice oxygen or oxygen
in the perovskite structure during the TPR experiments, once the
TPR was finished, the reducing gaseous mixture was changed to
helium flow and the samples were cooled down to room tempera-
ture. The sample was kept under inert atmosphere and transferred
to the XRD camera. Fig. 3 shows the diffraction profiles recorded
for the prepared perovskites under these conditions. The XRD
post-TPR (Fig. 3) of the 50% B-cation-substituted perovskites dis-
plays diffraction lines similar to those for the calcined perovskites
(Fig. 1), suggesting a high thermal stability under hydrogen. More-
over, changes of the nonstoichiometric rhombohedral LaMnO3.15
(Fig. 1) to a stoichiometric orthorhombic LaMnO3 perovskite struc-
ture (Fig. 3) confirm that the hydrogen consumption between 200
and 500 °C in the TPR profiles (Fig. 2) corresponds to the reduction
of Mn4+ to Mn3+. This result agrees with those reported by Irusta
et al. [37], where the loss of oxygen is due to reduction and the sig-
nal is shifted toward smaller 2h angles. Similar behavior under
hydrogen is observed for the Ni-, Cr-, and Co-substituted per-
ovskites, where a cubic and also a stoichiometric LaMnO3 per-
ovskite structure are detected. Even though the post-TPR
diffraction profiles of the Fe-substituted perovskite present many
diffraction lines, all of them fit quite well with an orthorhombic
LaFeO3 perovskite structure. The absence of new diffraction lines
associated with segregated phases means that if they are present,
they are in a small quantity or high degree of dispersion. The speci- Fig. 4. Oxygen TPD profiles of LaMnO3 and La(Mn0.5B0.5)O3 perovskites.
fic area of the perovskites post-TPR shown in Table 2 indicates an
increase after the reduction treatment. This is an unexpected result
because the surface area of these oxides suffers sintering, decreas-
ing their surface areas to lower than 1 m2 g1 [26]; meanwhile the
decrease in the intensity of the peaks is consistent with less crys-
talline solids.

Fig. 3. XRD patterns after TPR profiles of LaMnO3 and La(Mn0.5B0.5)O3 perovskites. Fig. 5. Ammonia TPD profiles of LaMnO3 and La(Mn0.5B0.5)O3 perovskites.
 R. Dinamarca et al. / Journal of Catalysis 338 (2016) 47–55 51

species occupying inner vacancies that desorbs at temperatures up 3.6. Temperature-programmed desorption of ammonia
to 600 °C [39] were detected. As the catalytic DME combustion
reaction occurs at low temperatures, it is expected that the pres- The acidity of perovskite-type oxides is related to Lewis acid
ence of lattice oxygen has little influence compared with that of sites due to the presence of La3+ and B3+ cations and the number/
chemisorbed oxygen, that corresponds to one to a few tens of oxy- ionic oxidation ratios [42], which depend on the oxidation number
gen monolayers, also associated with oxygen in the coordination of the cations [43]. The strength of the acid sites is classified
sphere of the highly charged Mn4+ content [13]. The oxygen vacan- according to NH3 temperature desorption as weak acid sites
cies in the perovskite structure can activate molecular oxygen [40], (T < 250 °C) and strong (T > 250 °C) [44] or NH3 acid site-bound
forming active oxygen species in the ABO3 structure [11] during [45]. Considering that the Mn(IV) surface has higher Lewis acidity,
oxygen adsorption, to improve the catalytic activity in the combus- the acid strength of perovskite-type oxides increases with the pres-
tion reaction. Two different behaviors can be observed from the ence of Mn4+ [46]. The NH3 desorption profiles of the studied per-
TPD profiles in Fig. 4. For Fe- and Cr-substituted perovskites, ovskites shown in Fig. 5 indicate different overlapped desorbed
behavior similar to that of pure LaMnO3 was obtained; the pres- peaks attributed to the presence of Lewis acid sites with different
ence of chemisorbed oxygen was overlapped by the physically strengths [47]. The desorbed ammonia content, shown in Table 2,
adsorbed oxygen peak and lattice oxygen species beginning at was calculated by the deconvolution of the peaks at a given tem-
600 °C, which reach a maximum below 700 °C. Conversely, for perature. The obtained values are higher than previously reported
the for Co- and Ni-substituted perovskites, the absence of lattice values of ammonia desorption for pure LaCoO3 and LaFeO3 per-
oxygen species at temperatures lower than 700 °C was detected. ovskites [46] and similar to those for nickel phosphide catalysts
Because TPD-MS experiments confirm that the evolved gas and [44], and they follow the order Co  Ni > Mn > Cr  Fe, the same
the He flow only contain oxygen, the deconvolution of the oxygen as the Mn4+ content trend.
desorption curves using Lorentzian lineshapes allows the calcula-
tion of the desorbed oxygen in the corresponding temperature 3.7. Surface analysis by X-ray photoelectron spectroscopy
ranges. The amounts of desorbed O2 in Table 2 are comparable to
those for Pd-, Pt-, and Rh-integrated LaMnO3 perovskites [41] XPS spectra were obtained to evaluate surface compositions;
and Cu-substituted manganites [36]. meanwhile, considering the low specific surface of the prepared

(a)
LaMnO 3 LaMn 0.5Cr 0.5O3 LaMn 0.5 Fe0.5O3
Counts/s (u.a)

Counts/s (u.a)

Counts/s (au)

830 840 850 860 830 840 850 860 830 840 850 860
BE (eV) BE (eV) BE (eV)

LaMn 0.5Co 0.5O3 LaMn0.5 Ni 0.5O3

Counts/s (u.a)

Counts/s (au)

830 840 850 860 820 840 860 880

BE (eV) BE (eV)

Fig. 6. (a) La3d5/2; (b) Mn2p3/2, and (c) Cr, Fe, Co, and Ni2p3/2 core-level spectra of LaMnO3 and La(Mn0.5B0.5)O3 perovskites.
52 R. Dinamarca et al. / Journal of Catalysis 338 (2016) 47–55

(b)
LaMnO3 LaMn0.5Cr0.5O3 LaMn0.5Fe0.5O3
Counts/s (au)

Counts/s (au)
632 640 648 656 664 Counts/s (au)
632 640 648 656 664 632 640 648 656 664
BE (eV) BE (eV) BE (eV)

LaMn0.5Co0.5O3 LaMn0.5Ni0.5O3
Counts/s (au)

Counts/s (au)

632 640 648 656 664

630 640 650 660
BE (eV)
BE (eV)

Fig. 6 (continued)

perovskites, it can be extrapolated that similar surface and bulk 529.8 eV; (ii) surface hydroxyls/carbonate groups at 531.0–
compositions would be present. The C1s, O1s, Mn2p3/2, and 531.4 eV; and (iii) weakly bonded O2  species at 532.4–532.6 eV
La3d5/2 core-level spectra were recorded for all perovskites, as well [29]. In the studied perovskites, O1s species at BE higher than
Cr2p3/2, Fe2p3/2, Co2p3/2, and Ni2p1/2 for their respective counter- 532 eV were not detected, indicating that the degassing procedure
parts. Fig. 6a–c shows several representative spectra, and Table 3 (300 °C for 1 h) to remove the total molecular water from the sur-
presents the respective binding energies (BE). The BE of C1s at face of the samples was appropriate. Although large differences in
289.8 eV (not included) was observed in all perovskites studied, the O1s core-level spectra are not seen in Table 3, some differences
and corresponds to the presence of surface carbonates attributed in the oxygen as stronger electrophilic species weakly adsorbed at
to La content. The most intense peak of La3d5/2 (Fig. 6a) shows 531 eV can be related to the electronic changes proposed. In
two components at 834.3 and 838.1 eV, both attributed to La3+ in Table 3, the chemisorbed oxygen appears at 531.1 eV for LaMnO3
the perovskite structure [48,49]. In relation to Mn2p peaks, and Cr- and Fe-substitution, whereas for Co- and Ni-substitution,
Fig. 6b shows a single doublet for all the samples. Unfortunately, the shift to BE of 530.8 eV is detected. This last behavior suggests
it is not possible to distinguish between Mn3+ or Mn4+ species a weaker attraction of the electrons by the B cation, supporting
due to closer BE of the most intense peak 2p3/2 at 641.3 eV for the previous discussion related to a change of the oxidation state
Mn3+ and 642.4 eV for Mn4+ [24]. For Cr, the two components of of Co and Ni surface species from +3 to +2. The presence of larger
the most intense peak of 2p3/2 at 576.6 and 579.4 eV are assigned Co2+ and Ni2+ surface phases extrapolated to bulk composition is
to surface Cr3+ and Cr6+ [50], respectively; for Fe, the 2p3/2 located an indirect measurement of a higher Mn4+ content.
at 710.8 eV and 2p1/2 at 724.8 eV are due to Fe3+ species [51,52].
Meanwhile, for Co2p3/2, two peaks are detected at 780.6 and 3.8. Catalytic activity
779.6 eV, associated with Co3+ and Co2+ species [51]. Because the
Ni2p3/2 signal coincides with the La3d3/2 signal, the displacement The combustion of DME using a flow reactor with excess oxygen
of the BE of Ni2p1/2 to 873.2 eV indicates the presence of Ni2+ spe- was studied. Under the experimental conditions, the noncatalytic
cies [53]. In relation to O1s, it is generally deconvoluted into three reaction (not shown) starts at 250 °C and total conversion was
peaks for perovskite-type oxides: (i) lattice oxygen of the O2 at reached at 650 °C. Fig. 7 shows the experimental ignition curves
 R. Dinamarca et al. / Journal of Catalysis 338 (2016) 47–55 53

(c)
LaMn0.5Cr0.5O3 LaMn0.5Fe0.5O3

Counts/s (a.u)

Counts/s (a.u)
568 576 584 592 704 712 720 728
BE (eV) BE (eV)

LaMn0.5Co0.5O3 LaMn0.5Ni0.5O3
Counts/s (a.u)

Counts/s (a.u)

770 780 790 800 810 820 840 860 880

BE (eV) BE (eV)

Fig. 6 (continued)

of the catalysts starting from 150 °C, where the initial catalytic shown by pure LaMnO3 perovskite even though larger differences
activity of perovskites was observed, up to total conversion at in the intrinsic reaction rate values are detected. In Table 4, it
380 °C. To evidence the reproducibility of the catalytic activity can be seen that the Co- and Ni-substituted perovskites display a
measurements and lack of deactivation processes, after the first large increase in the intrinsic reaction rate compared with that
lightoff curve, the catalyst was cooled and a new lightoff curve for LaMnO3. Also, the catalyst with the highest Mn4+ content and
was obtained, without differences from the original ignition curve. reducibility has the best activity, while the catalysts with lower
Moreover, under the experimental conditions, only water and car- values have the lowest activity in DME combustion. Due to the
bon dioxide products were detected. The catalytic activity perfor- absence of previous reports of perovskite-type oxides in DME com-
mance of the perovskites during DME oxidation related to bustion, the corresponding intrinsic reaction rate values should be
ignition temperature values (T50), defined as the temperature compared with those for similar mixed oxide catalysts. The
required to reach 50% conversion, and to specific (mol g1 h1) obtained values are larger than the corresponding ones of doped
and intrinsic (mmol m2 h1) reaction rates are summarized in cryptomelane manganese oxides reported by Sun et al. [4] with
Table 4. To ensure the absence of mass and heat transfer limita- the same order of decreasing activity in DME combustion, i.e.,
tions, the catalytic activity was evaluated at low conversion levels Ni > Co > Fe > Cr, and are comparable to Ce-doped manganese
(<10%). It seems that B-site substitution does not indicate a posi- oxide octahedral molecular sieves [3]. For CO oxidation, Yan et al.
tive impact on the lightoff curves because the lower T50 is still [54] reported a decrease in the catalytic activity for B-site substi-
54 R. Dinamarca et al. / Journal of Catalysis 338 (2016) 47–55

Table 3
BE of La3d5/2, Mn2p3/2, B (Cr, Fe, Co, Ni) 2p3/2, and O1s.

La3d5/2 (eV) Mn2p3/2 (eV) B2p3/2 (eV) O1s (eV)

LaMnO3 834.4 642.3 – 529.4 (60)
531.1 (40)
LaMn0.5Cr0.5O3 834.4 642.2 576.5 (49) 529.6 (65)
579.5 (51) 531.1 (35)
LaMn0.5Fe0.5O3 834.3 642.4 710.9 529.1 (58)
531.1 (42)
LaMn0.5Co0.5O3 834.4 642.1 780.4 529.4 (59)
530.9 (41)
LaMn0.5Ni0.5O3 834.1 642.1 855.2 529.0 (49)
530.8 (51)

Fig. 7. Stationary-state conversion of DME as a function of reaction temperature of

(j) LaMnO3; (s) LaMn0.5Cr0.5O3; (d) LaMn0.5Fe0.5O3; (▲) LaMn0.5Ni0.5O3; and (4)
LaMn0.5Co0.5O3.

Table 4
Ignition temperature T50, specific and intrinsic reaction rate at 180 °C, and apparent
activation energy.

T50 (°C) Reaction rate Eact app (kJ mol1)

1 1 2 1
mol g h mmol m h
LaMnO3 252 0.016 5.2 82
LaMn0.5Cr0.5O3 297 0.007 3.6 122
LaMn0.5Fe0.5O3 305 0.009 5.4 124
LaMn0.5Co0.5O3 303 0.013 11.7 80
LaMn0.5Ni0.5O3 299 0.013 11.7 75

Fig. 8. Linear correlation between the specific reaction rate of the perovskites and
(a) hydrogen consumed; (b) Mn4+ content; (c) weakly bound oxygen.

tuted cobaltite in the following order: Co  Ni > Fe  Cr. Alter-

nately, Machocki et al. [55] reported that a higher Mn4+ content perovskites being almost isostructural, a next step is to look for a
leads to higher methane oxidation rates than for manganese lan- correlation of the redox properties with the catalytic activity. As
thanum oxides modified with silver, and in A-site Sr-substituted the specific BET areas of the perovskites are low and inherent error
LaMnO3, the increases in the redox properties produce an enhance- may be important, the intrinsic rate values could not be properly
ment of the catalytic activity in oxidation reactions [22]. Better compared. This is why reaction rates were reported in terms of
reducibility is also related to higher catalytic activity, as noted by specific reaction rates. The almost linear trends of the plots of reac-
Chen et al. [56] for pure LaMeO3 (Me = Mn, Fe, Co) perovskites. tion rate against hydrogen consumption in the TPR profiles, Mn4+
Liu et al. [21] also found that the higher oxygen species concentra- content, and alpha oxygen desorbed, shown in Fig. 8a–c, corrobo-
tion is responsible for the superior catalytic activity in the total rate the previous discussion, in line with Rezlescu et al. [58],
oxidation of carbon monoxide and toluene relative to La0.6Sr0.4- who reported that the presence of Mn4+ ions allows the availability
MnO3-modified perovskites [57]. To make good use of the prepared of less anchored surface oxygen species closely related to the cat-
 R. Dinamarca et al. / Journal of Catalysis 338 (2016) 47–55 55

alytic activity. The apparent activation energies determined from [9] R. Sumathi, K. Johnson, B. Viswanathan, T.K. Varadarajan, Appl. Catal. A: Gen.
172 (1998) 15–22.
the Arrhenius plots, compiled in Table 4, support the hypothesis
[10] J. Niu, J. Deng, W. Liu, L. Zhang, G. Wang, H. Dai, H. He, X. Zi, Catal. Today 126
regarding the higher reaction rate being attributed to redox prop- (2007) 420–429.
erties between surface oxygen reactivity and reducibility of the B- [11] G. Pecchi, P. Reyes, R. Zamora, C. Campos, L.E. Caduus, B.P. Barbero, Catal.
site cations. Moreover, the combination of reducibility and oxygen Today 133 (2008) 420–427.
[12] J.J. Liang, H.S. Weng, J. Catal. 140 (1993) 302–310.
mobility gives the order of redox capability in the activity order in [13] H. Falcón, J.A. Barbero, J.A. Alonso, M.J. Martínez-Lope, J.L.G. Fierro, Chem.
close relation to the weakening of the B–O bond strength. Consid- Mater. 14 (2002) 2325–2333.
ering that the combustion reaction involves reversibly chemi- [14] M.R. Morales, B.P. Barbero, L.E. Cadús, Fuel 87 (2008) 1177–1186.
[15] J.J. Liang, H.S. Weng, Ind. Eng. Chem. Res. 32 (1993) 2563–2572.
sorbed oxygen atoms that are continuously supplied from the gas [16] Y. Wu, T. Yu, B.S. Dou, C.X. Wang, X.F. Xie, Z.L. Yu, S.R. Fan, Z.R. Fan, L.C. Wang, J.
phase and surface oxygen originating from the lattice, it is pro- Catal. 120 (1989) 88–107.
posed that at lower reaction temperatures (<400 °C), the catalytic [17] L.A. Chick, L.R. Pederson, G.D. Maupin, J.L. Bates, L.E. Thomas, G.J. Exarhos,
Mater. Lett. 10 (1990) 6–12.
activity is largely determined by the number of weakly bound sur- [18] P. Porta, S. De Rossi, M. Faticanti, G. Minelli, I. Pettiti, L. Lisi, M. Turco, J. Solid
face oxygen species that can activate the gas phase reactant mole- State Chem. 146 (1999) 291–304.
cule, and the gas phase oxygen or bulk oxygen to fill surface [19] S. Royer, H. Alamdari, D. Duprez, S. Kaliaguine, Appl. Catal. B: Environ. 58
(2005) 273–288.
oxygen vacancies, presumably through a Mars van Krevelen [20] N. Russo, D. Fino, G. Saracco, V. Specchia, J. Catal. 229 (2005) 459–469.
(MVK) mechanism, largely accepted for these types of crystalline [21] Y. Liu, H. Dai, J. Deng, X. Li, Y. Wang, H. Arandiyan, S. Xie, H. Yang, G. Guo, J.
structures [4]. Finally, it is also emphasized that surface acidity Catal. 305 (2013) 146–153.
[22] H. Arandiyan, H. Dai, J. Deng, Y. Liu, B. Bai, Y. Wang, X. Li, S. Xie, J. Li, J. Catal.
does not play a significant role in the combustion reaction, because
307 (2013) 327–339.
acidity values were similar for all the samples. [23] F. Patcas, F.C. Buciuman, J. Zsako, Thermochim. Acta 360 (2000) 71–76.
[24] S. Ponce, M.A. Pena, J.L.G. Fierro, Appl. Catal. B: Environ. 24 (2000) 193–205.
[25] G. Pecchi, C. Campos, O. Pena, Mater. Res. Bull. 44 (2009) 846–853.
4. Conclusions [26] G. Pecchi, C. Campos, React. Kinet. Catal. Lett. 91 (2007) 353–359.
[27] V. Uvarov, I. Popov, Mater. Charact. 58 (2007) 883–891.
The effect of the B-site substitution of 50% of the first-row tran- [28] L.A. Isupova, I.S. Yakovleva, G.M. Alikina, V.A. Rogov, V.A. Sadykov, Kinet. Catal.
46 (2005) 729–735.
sition metal (B = Cr, Fe, Co, or Ni) in LaMnO3 does not significantly [29] L.G. Tejuca, J.L.G. Fierro, J.M.D. Tascon, Adv. Catal. 36 (1989) 237–328.
change the geometric properties. The similar ionic radius allows [30] S. Ifrah, A. Kaddouri, P. Gelin, G. Bergeret, Catal. Commun. 8 (2007) 2257–2262.
almost complete substitution and isostructural crystalline struc- [31] R. Jimenez, R. Zamora, G. Pecchi, X. Garcia, A.L. Gordon, Fuel Process. Technol.
91 (2010) 546–549.
tures. The differences in the stability of the oxidation state of the
[32] N. Escalona, S. Fuentealba, G. Pecchi, Appl. Catal. A: Gen. 381 (2010) 253–260.
B-substituted metal produce larger differences in the electronic [33] P. Ciambelli, S. Cimino, L. Lisi, M. Faticanti, G. Minelli, I. Pettiti, P. Porta, Appl.
properties of the Cr-, Fe-, Co-, and Ni-substituted perovskites. Catal. B: Environ. 33 (2001) 193–203.
[34] H. Provendier, C. Petit, C. Estournes, S. Libs, A. Kiennemann, Appl. Catal. A: Gen.
The replacement of Mn3+ by a B3+ cation with a stable (III) valence,
180 (1999) 163–173.
such as Cr3+ or Fe3+, decreases their redox properties, whereas [35] F.C. Buciuman, F. Patcas, J. Zsakó, J. Therm. Anal. Calorim. 61 (2000) 819–825.
when the replacement is by a B cation with stable (III) or (II) [36] L. Lisi, G. Bagnasco, P. Ciambelli, S. De Rossi, P. Porta, G. Russo, M. Turco, J. Solid
valence, such as Co3+/Co2+ or Ni3+/Ni2+, their redox properties are State Chem. 146 (1999) 176–183.
[37] S. Irusta, M.P. Pina, M. Menéndez, J. Santamarı´a, J. Catal. 179 (1998) 400–412.
increased. The decrease in the reaction rate of DME combustion [38] H.M. Zhang, Y. Shimizu, Y. Teraoka, N. Miura, N. Yamazoe, J. Catal. 121 (1990)
for LaMn0.5Cr0.5O3 and LaMn0.5Fe0.5O3 and the increase for LaMn0.5- 432–440.
Co0.5O3 and LaMn0.5Ni0.5O3 perovskites are attributed to the lower [39] R.D. Zhang, A. Villanueva, H. Alamdari, S. Kaliaguine, J. Catal. 237 (2006) 368–
380.
and higher redox being in close relation to the Mn4+/Mn3+ redox [40] S. Royer, F. Berube, S. Kaliaguine, Appl. Catal. A: Gen. 282 (2005) 273–284.
pair content. [41] C.-L. Li, Y.-C. Lin, Appl. Catal. B: Environ. 107 (2011) 284–293.
[42] M.A. Pena, J.L.G. Fierro, Chem. Rev. 101 (2001) 1981–2017.
[43] M.I. Zaki, M.A. Hasan, F.A. Al-Sagheer, L. Pasupulety, Colloids Surf. A:
Acknowledgments Physicochem. Eng. Aspects 190 (2001) 261–274.
[44] J.A. Cecilia, A. Infantes-Molina, E. Rodriguez-Castellon, A. Jimenez-Lopez, S.T.
CONICYT, Fondecyt 1130005, Red Doctoral REDOC and the Oyama, Appl. Catal. B: Environ. 136 (2013) 140–149.
[45] D. Liu, P. Yuan, H. Liu, J. Cai, D. Tan, H. He, J. Zhu, T. Chen, Appl. Clay Sci. 80–81
MINEDUC project UCO1202. The authors belong to the Chile-
(2013) 407–412.
France Joint International Laboratory, LIA-MIF 836. [46] N. Hosseinpour, Y. Mortazavi, A.A. Khodadadi, Appl. Catal. A: Gen. 487 (2014)
26–35.
[47] M.C. Sanchez-Sanchez, R.M. Navarro, J.L.G. Fierro, Int. J. Hydrogen Energy 32
Appendix A. Supplementary material (2007) 1462–1471.
[48] B. Kucharczyk, W. Tylus, Catal. Today 90 (2004) 121–126.
Supplementary data associated with this article can be found, in [49] B. Kucharczyk, W. Tylus, Appl. Catal. A: Gen. 335 (2008) 28–36.
[50] K. Rida, A. Benabbas, F. Bouremmad, M.A. Peña, E. Sastre, A. Martínez-Arias,
the online version, at http://dx.doi.org/10.1016/j.jcat.2016.02.011. Appl. Catal. B: Environ. 84 (2008) 457–467.
[51] W.-Y. Lee, H.J. Yun, J.-W. Yoon, J. Alloys Compd. 583 (2014) 320–324.
References [52] M. Markova-Velichkova, T. Lazarova, V. Tumbalev, G. Ivanov, D. Kovacheva, P.
Stefanov, A. Naydenov, Chem. Eng. J. 231 (2013) 236–244.
[53] M.E. Rivas, C.E. Hori, J.L.G. Fierro, M.R. Goldwasser, A. Griboval-Constant, J.
[1] A. Ishikawa, E. Iglesia, J. Catal. 252 (2007) 49–56.
Power Sources 184 (2008) 265–275.
[2] A. Ishikawa, M. Neurock, E. Iglesia, J. Am. Chem. Soc. 129 (2007) 13201–13212.
[54] X. Yan, Q. Huang, B. Li, X. Xu, Y. Chen, S. Zhu, S. Shen, J. Ind. Eng. Chem. 19
[3] L. Yu, G. Diao, F. Ye, M. Sun, J. Zhou, Y. Li, Y. Liu, Catal. Lett. 141 (2011) 111–119.
(2013) 561–565.
[4] M. Sun, L. Yu, F. Ye, G. Diao, Q. Yu, Z. Hao, Y. Zheng, L. Yuan, Chem. Eng. J. 220
[55] A. Machocki, T. Ioannides, B. Stasinska, W. Gac, G. Avgouropoulos, D. Delimaris,
(2013) 320–327.
W. Grzegorczyk, S. Pasieczna, J. Catal. 227 (2004) 282–296.
[5] J. Zhou, L. Yu, M. Sun, G. Diao, Y. Li, X. Cheng, Micropor. Mesopor. Mater. 181
[56] J. Chen, M. Shen, X. Wang, J. Wang, Y. Su, Z. Zhao, Catal. Commun. 37 (2013)
(2013) 105–110.
105–108.
[6] Y. Teraoka, H. Nii, S. Kagawa, K. Jansson, M. Nygren, Appl. Catal. A: Gen. 194
[57] Z. Li, M. Meng, Q. Li, Y. Xie, T. Hu, J. Zhang, Chem. Eng. J. 164 (2010) 98–105.
(2000) 35–41.
[58] N. Rezlescu, E. Rezlescu, P.D. Popa, C. Doroftei, M. Ignat, Compos. B: Eng. 60
[7] C. Moure, D. Gutierrez, O. Pena, P. Duran, J. Solid State Chem. 163 (2002) 377–
(2014) 515–522.
384.
[8] N.A. Merino, B.P. Barbero, P. Grange, L.E. Cadús, J. Catal. 231 (2005) 232–244.