Topics in Catalysis

, Volume 60, Issue 3–5, pp 300–306 | Cite as

Catalytic Activity and Thermal Stability of LaFe1−xCuxO3 and La2CuO4 Perovskite Solids in Three-Way-Catalysis

  • Anke Schön
  • Jean-Philippe Dacquin
  • Christophe Dujardin
  • Pascal Granger
Original Paper

Abstract

LaFe1−xCuxO3 and La2CuO4 catalysts are investigated for three-way-catalysis application. LaFe1−xCuxO3 solids present interesting catalytic properties mainly with respect to hydrocarbon and CO oxidation and NO reduction in rich conditions with high N2 selectivity. The partial substitution of iron with copper leads to enhanced catalytic activity for oxidation reactions. This can be attributed to the higher reducibility of the solid as evidenced earlier by H2-temperature-programmed reduction. The orthorhombic structure associated with LaFe1−xCuxO3 solids (x up to 0.2) revealed good thermal stability during catalytic cycles in stoichiometric, lean and rich conditions whereas the Ruddlesden Popper phase associated with La2CuO4 completely decomposed in rich conditions.

Keywords

Perovskite Three-way catalysis Nitrogen oxide Oxidation 

1 Introduction

High conversion of pollutants emitted from gasoline car engines can be achieved via conventional three way catalysis (TWC). This has led to a decrease of pollutant emissions since the 1980s [1]. TWC catalysts are however the main industrial consumer of some Critical Raw Materials such as platinum group metals and rare earth elements which are listed by the European Union [2]. Research on alternatives to precious metal catalysts is expected to experience increased growth in the future. Perovskite-type materials were found to have interesting potential for three-way applications especially to prevent precious noble metals from thermal sintering [3]. This strategy led to interesting outcomes with lower noble metals contents for various applications [4]. The main drawbacks associated with perovskite-based solids lie in low specific surface area and a usual lanthanum enrichment which affects their catalytic properties [5]. It was also found that non-stoichiometric perovskites especially lanthanum-deficient solids led to significant enhancement of the catalytic performances [5, 6]. In this study, LaFeO3-based perovskites being non-toxic and thermally stable were investigated. It has been showed previously that catalytic properties of perovskites (ABO3 formula) are mainly influenced by the nature of the B-site cation [7]. Therefore, the influence of Fe substitution by a more reductive and more active metal was investigated. Cu was chosen as substituting cation due to its high reducibility. The aim of its incorporation in the perovskite structure was to profit from its catalytic properties while stabilizing it inside the LaFeO3 lattice. The incorporation of divalent cations could also destabilize the electroneutrality of the structure, hence leading to the formation of unusual Fe4+ species and/or oxygen vacancies [8, 9]. The latter are generally known to play an important role in the process of NO adsorption and desorption [8, 9] and in this study should give rise to an enhanced reduction performance. This investigation was performed on a series of LaFe1−xCuxO3 catalysts which were compared to a La2CuO4 catalyst.

2 Experimental

2.1 Catalyst Preparation and Related Physicochemical Characterization

A series of LaFe1−xCuxO3 (x = 0; 0.05; 0.1; 0.2) and La2CuO4 was prepared by a conventional citrate method [4, 10] from the corresponding metal nitrates (La(NO3)3·6H2O, Fe(NO3)3·9H2O and Cu(NO3)2, 2.5 H2O) respecting a molar ratio of \(\left( {{{{\text{n}}_{\text{La}} + {\text{n}}_{\text{Fe}} + {\text{n}}_{\text{Cu}} } \mathord{\left/ {\vphantom {{{\text{n}}_{\text{La}} + {\text{n}}_{\text{Fe}} + {\text{n}}_{\text{Cu}} } {{\text{n}}_{\text{citric acid}} }}} \right. \kern-0pt} {{\text{n}}_{\text{citric acid}} }}} \right) = 1\). Nitrate decomposition was done separately in a muffle oven before calcination at 600 °C for 8 h under air flow.

Chemical composition of solids was determined using inductively coupled plasma atomic emission spectroscopy ICP-AES (Institute of Analytical Sciences—CNRS). XRD patterns were recorded on a Bruker AXS D8 Advance diffractometer and unit cell parameters and crystallite sizes were obtained by full pattern matching with the FullProf suite [6]. Specific surface areas (SSA) were determined by N2 physisorption at −196 °C using the Micromeritics Tristar II [5]. SEM were recorded on a Hitachi SU70 SEM FEG. Samples were prepared by depositing the calcined powder on a conductive double-faced adhesive tape. H2-TPR experiments were performed on a Micromeritics Autochem II 2920 using a flow of 5 % H2/Ar (50 mL/min, 50 mg for LaFe1−xCuxO3 solids and 100 mL/min, 25 mg for La2CuO4 solid). XPS analyses were recorded either on an AXIS Ultra DLD Kratos spectrometer equipped with a monochromatised aluminium source (150 W) and charge compensation gun or on a VG Scientific Escalab 220i-XL equipped with a non-monochromatised aluminium source (300 W). All binding energies were calibrated with C 1 s core level at 285 eV.

2.2 Catalytic Measurements

Catalytic measurements were performed in a fixed bed reactor using 200 mg of catalyst (100–200 µm). The total flow rate was adjusted to 200 mL min−1 in order to get a gaseous hourly space velocity of 60,000 mL g−1 h−1 except for La2CuO4 (45,000 mL g−1 h−1). NOx and reductants concentrations were measured during temperature-programmed reactions from 110 to 500 °C successively under stoichiometric, lean (lower values of reductants) and rich (highest values of reductants) conditions. Realistic feed gas compositions consisted of 0.1 % NO, 15 % CO2, 10 % H2O, CO (0.5, 0.7, 0.9 %), CH4 (150, 225, 300 ppm), C3H6 (300, 450, 600 ppm), C3H8 (150, 225, 300 ppm), O2 (0.935, 0.777, 0.609 %), and H2 (0.167, 0.233, 0.3 %) and He balance as detailed in Ref. [5]. Conversions were calculated according to Eq. (1) and molar flow rates (Fi).
$$X_{i} = \frac{{{\text{F}}_{\text{i,inlet}} - {\text{F}}_{\text{i,outlet}} }}{{{\text{F}}_{\text{i,inlet}} }}$$
(1)

3 Results and Discussion

3.1 Characterization of LaFe1−xCuxO3 and La2CuO4 Solids

3.1.1 Bulk Properties

The series of LaFe1−xCuxO3 and La2CuO4 solids have been prepared according to the conventional citrate method. The chemical compositions determined with ICP-AES were in agreement with the theoretical bulk atomic ratio (La/(Fe + Cu)) for the LaFeO3 and LaFe0.8Cu0.2FeO3 samples (Table 1).
Table 1

Bulk atomic ratios of calcined LaFe0.8Cu0.2O3 and La2CuO4 determined with ICP-AES

Nominal composition

Bulk atomic ratios

xLa/xFe

xLa/x(Fe+Cu)

xCu/xFe

LaFeO3

1.01

1.01

LaFe0.8Cu0.2FeO3

1.26

1.03

0.23

The structural properties of LaFe1−xCuxO3 and La2CuO4 solids were assessed by XRD analysis. The orthorhombic structure was obtained up to x = 0.2 whereas the Ruddlesden Popper phase was obtained for La2CuO4. The segregation of another crystalline phase could not be put into evidence from XRD patterns. In addition, the relative intensities of the diffraction peaks were unchanged compared to the reference (PDF #37-1493), thus indicating that no preferential growth of the crystallites was observed. Previous studies investigating the structural modifications due to Fe-substitution by Cu in the B-site of LaFeO3 concluded that a solid solution was formed for Cu substitution degrees up to x = 0.2 [9].

Crystallite sizes estimated from XRD (Table 2) decreased slightly after Cu addition compared to LaFeO3. Besides, the volume of the unit cell decreased linearly while Cu-content in the perovskite structure increased (Fig. 1). When assuming a simple steric model, the observed unit cell contraction could not be explained by comparing the ionic radii of Cu2+ and Fe3+ of 0.730 Å and 0.645 Å (high-spin) in six-coordinate environment respectively [9]. In this case, a unit cell expansion was to be expected. Such an apparent discrepancy emphasized the fact that a charge compensation associated to the creation of anionic vacancies or the stabilization of Fe in an unusual tetravalent oxidation state could occur. The difference in ionic radii between Cu2+ and Fe3+ of 0.085 Å was not counterbalanced by the difference in radii between Fe3+ and Fe4+ (0.06 Å). This suggests that a large amount of defects in the form of anionic vacancies is required to ensure the electroneutrality of this system. Similar results have been reported in literature suggesting a negligible amount of Fe4+ species measured by titration [9].
Table 2

Comparison of specific surface areas and cumulative pore volume of LaFe1−xCuxO3 and La2CuO4 solids

Composition

SSA (m2 g−1)

Vpore (cm3 g−1)

dcryst (nm)a

Stheoretical (m2 g−1)a

LaFeO3

13.9

0.08

20.1 ± 0.1

45

LaFe0.95Cu0.05O3

15.0

0.09

15.7 ± 0.1

57

LaFe0.9Cu0.1O3

18.4

0.09

15.3 ± 0.1

59

LaFe0.8Cu0.2O3

13.8

0.07

13.5 ± 0.1

66

La2CuO4

12.2

0.07

22.1 ± 0.1

38

aDetermined from XRD measurements; Stheoretical = 6.103/(ρ.dcryst)

Fig. 1

Evolution of the unit cell volume of calcined LaFe1−xCuxO3 (x = 0, 0.05, 0.1, 0.2)

The textural properties were studied by N2-physisorption (Fig. S1). All solids of the LaFe1−xCuxO3 series presented isotherms which are indicative of macroporous or weakly porous solids (type II) with a small additional contribution of mesopores (type IV). No significant changes in hysteresis associated with a broad pore size distributions were observed in the series. The formation of mesopores (~4 nm) was previously observed on a series of La1−xFeO3 [5]. The proportion of mesopore was found to increase while x-value decreased on La1−xFeO3 solids [5]. The SEM image of LaFeO3 is presented in Fig. 2. Aggregates of approximately 50 nm were observed. This size is larger than crystallite size (Table 2) calculated from XRD patterns. Table 2 also reports the specific surface areas and the cumulative pore volumes of the Cu-substituted perovskites. Textural properties were very similar to those of the initial LaFeO3 solid. Only slight increases in specific surface area were observed for LaFe0.95Cu0.05O3 and LaFe0.9Cu0.1O3. The theoretical surface area can be roughly estimated from the crystallite size using simple geometrical model which assumes spherical particles. The deviation between the theoretical and experimental specific surface area indicates that the solids here consist of an agglomeration of a small number of crystalline particles. This result is in agreement with the deviation previously discussed between aggregate size and crystalline size.
Fig. 2

SEM image of LaFeO3 catalyst

The reducibility of the Cu-substituted perovskites was investigated with H2-TPR analyses. The H2 consumption profiles vs. temperature are collected in Fig. 3. On the opposite to LaFeO3 which is only reducible at temperatures above 700 °C [5], the profiles of Cu-substituted samples clearly showed an additional low temperature reduction peak with its maximum at ≈240 °C as reported elsewhere [8]. The H2 consumption increased with increasing Cu-content. This low temperature signal consisted of two overlapping peaks which developed with increasing Cu-content. The comparison with the H2-TPR profile of La2CuO4 confirmed that the additional peaks observed for LaFe1−xCuxO3 (x = 0.05, 0.1, 0.2) could be attributed to the Cu2+ reduction in the perovskite structure. For all Cu-substituted perovskites, Cu2+ reduction was stopped in a lower temperature range (up to 280 °C) compared to the Ruddlesden Popper phase (up to 350 °C). This is in agreement with literature [9] and confirms that anionic defects contribute to compensate for the lower oxidation state of Cu cation compared to Fe cation in the perovskite lattice.
Fig. 3

H2 consumption curves during temperature-programmed reduction of calcined LaFe1−xCuxO3 compared to La2CuO4

The sample with the lowest substitution degree (x = 0.05) also showed a weak third reduction peak at 390 °C. The possible assignment to bulk CuO or supported CuO phase can be proposed. However bulk CuO reduction proceeds generally in a two-step process via the formation of Cu2O around 110 °C [11, 12]. On the other hand the reduction of supported CuO can be ruled out since supported CuO reduction should occur at lower temperature than the one of bulk CuO [11]. The H2 consumption is detailed in Table 3. The complete reduction of the Cu2+ cations proceeds with H/Cu ratio of 2.04 which is equal to the theoretical H/Cu ratio for La2CuO4. By contrast copper is only partially reduced in the substituted Fe-based perovskites with H/Cu ratio values between 1.64 and 1.07. While the reduction of Cu2+ to Cu+ may be complete for all solids below 280 °C as reported in literature [8], the reduction of Cu+ to metallic Cu only partially proceeds for LaFe1−xCuxO3 (x = 0.05, 0.1, 0.2). This was in agreement with the enhancement of the stability of the structure in reducing atmosphere.
Table 3

Comparison of H2-TPR results of LaFe1−xCuxO3 and La2CuO4 solids

Nominal composition

H2 consumptiona

Atomic ratioa

Tmax reduction peaks

(mmol H2.gcat−1)

H/Cu

(°C)

(°C)

(°C)

LaFeO3

>1100

LaFe0.95Cu0.05O3

0.24

1.07

237

(368)

>1100

LaFe0.9Cu0.1O3

0.29

1.40

245

1084

LaFe0.8Cu0.2O3

0.67

1.64

241

1080

La2CuO4

2.51

2.03

300

525

667

aH2 consumption was integrated over the temperature range of 150–400 °C

3.1.2 Surface Properties

XPS analyses were conducted on the LaFe1−xCuxO3 and La2CuO4 series. The spectra of La 3d, Fe 2p and Cu 2p (Fig. S2) exhibited the characteristic multiplet splitting and satellites of trivalent La and Fe and divalent Cu cations. In the case of Fe 2p photopeak, the presence of Fe3+ was confirmed by the positions of the satellite (≈719 eV as for the non-substituted perovskite). The stabilization of Cu2+ cations was confirmed by the positions of the Cu 2p3/2 photopeak at approximately 933.1 eV and the presence of the characteristic satellite. The ratio between the area of the Cu 2p3/2 satellite and the area of the main peak was approximately one half which confirms the predominance of Cu2+ species. The surface atomic ratios of the LaFe1−xCuxO3 and La2CuO4 catalysts were calculated from XPS analysis (Table 4). Only 0–15 at.% of Cu+ species were estimated by spectral decomposition of the Cu 2p3/2 core level. The stabilization of copper in perovskite structure was previously related to a lower reducibility. The lanthanum-based perovskite usually exhibits a surface lanthanum enrichment [5, 6]. The \({{{\text{x}}_{\text{La}} } \mathord{\left/ {\vphantom {{{\text{x}}_{\text{La}} } {{\text{x}}_{{({\text{Fe + Cu}})}} }}} \right. \kern-0pt} {{\text{x}}_{{({\text{Fe + Cu}})}} }}\) values for the series of LaFe1−xCuxO3 were above the theoretical values (one for LaFe1−xCuxO3 solids and two for La2CuO4 solid). The surface lanthanum enrichment was then put into evidence for all solids and is in agreement with previous studies [13]. This lanthanum excess was only slightly attenuated by the Cu substitution. Another interesting result concerned the evolution of the atomic Cu/Fe ratio. The xCu/xFe values were higher than theoretical ones hence indicating a stronger decrease of surface concentration for iron than for copper. In fact the surface concentration of Cu as calculated by \({{{\text{x}}_{\text{Cu}} } \mathord{\left/ {\vphantom {{{\text{x}}_{\text{Cu}} } {\left( {{\text{x}}_{\text{La}} + {\text{x}}_{\text{Fe}} } \right)}}} \right. \kern-0pt} {\left( {{\text{x}}_{\text{La}} + {\text{x}}_{\text{Fe}} } \right)}}\) was quite close to the theoretical one. The examination of the O 1s photopeak (Fig. S2) revealed two contributions at 531.5 and 529.5 eV which were previously ascribed to adsorbed oxygen-containing species and lattice oxygen species (O2−) respectively. The formation of surface carbonate was put into evidence by the parallel development of O 1s photopeak at 531.1 eV and C1s photopeak at 289.9 eV. The presence of carbonate species may be related to the lanthanum excess which can also decrease the accessibility to the active site [6].
Table 4

Comparison of surface atomic ratios calculated from XPS analyses and theoretical values of calcined LaFe1−xCuxO3 and La2CuO4 solids

Nominal

xLa/x(Fe+Cu)

xCu/xFe

xCu/(xLa + xFe)

Olatticea

C(CO32−)b

Cu+/Cutotal

Composition

Theor.

Experim.

Theor.

Experim.

Theor.

Experim.

LaFeO3

1

2.13

0

0

55

13

LaFe0.95Cu0.05O3

1

1.94

0.05

0.08

0.026

0.024

52

15

5

LaFe0.9Cu0.1O3

1

2.08

0.11

0.14

0.053

0.043

51

23

0

LaFe0.8Cu0.2FeO3

1

1.82

0.25

0.39

0.11

0.099

47

23

8

La2CuO4

2

2.98

0.50

0.211

25

25

15

aOlattice + Oads = 1

bC(CO32−) + Cadventitious = 1

3.2 Catalytic Performances of LaFe1−xCuxO3 and La2CuO4 Samples in Three-Way-Catalysis

The catalytic activity was measured during successive experiments with reaction mixtures corresponding to stoichiometric, rich and lean conditions. Residual outlet NOx and reductants concentrations were measured during temperature-programmed reactions. Due to the complexity of reaction mixture, the conversion profiles of CO vs. T are reported in Fig. 4. The conversion profiles of other components are reported in Fig. S3 and Fig. S4. The reference LaFeO3 catalyst exhibited poor catalytic performances. By contrast, a significant enhancement of CO conversion in stoichiometric conditions was obtained with the partial substitution of iron by copper. Best CO oxidation results were obtained by both samples with the highest Cu-substitution degree, LaFe0.8Cu0.2O3 and La2CuO4, which are also the most reducible solids. T50 values corresponding to the temperature required for 50 % conversion are reported in Table 5. The performances of LaFe0.8Cu0.2O3 (T50 (CO, stoic1) = 295 °C) even exceeded those observed on La2CuO4 especially when we consider the lower GHSV used for La2CuO4. The CO oxidation performance enhancement also followed the trend of increasing substitution degree in lean and rich conditions. The high activity for CO oxidation can be interesting especially in the presence of platinum which is generally inhibited by CO at low temperature. It should be pointed out that CO conversion decreased on La2CuO4 around 380 °C in rich conditions due to the occurrence of additional reforming and/or water–gas shift (WGS) reactions (Fig. S3 and Fig. S4). Another remark concerns the reduction of NO. Even in absence of noble metals, the reduction of NO was obtained in rich conditions above 450 °C on LaFe1−xCuxO3 catalyst. The main product of NO reduction obtained on LaFe1−xCuxO3 is N2 whereas LaFeO3 mainly formed NH3. The samples had low N2O selectivity in the temperature range of 440–500 °C with slight differences in the N2 and NH3 selectivity in stoichiometric conditions (Table 5). The high N2 selectivity of the Cu-substituted orthoferrites and the high catalytic activity of the LaFe0.8Cu0.2O3 catalyst are in agreement with previous reports on the catalytic reduction of NO by propylene over LaFe0.8Cu0.2O3 in the presence of O2 by Zhang et al. [14]. The reduction of NO under stoichiometric and lean conditions remained limited for the LaFe1−xCuxO3 catalysts. The conversion was marginal accounting for a maximum of 5 % at temperatures above 430 °C under lean conditions and in the temperature range of 360–380 °C under stoichiometric conditions (Fig. S3). The sole reduction product observed under lean conditions is N2 whereas nitrous oxide was formed as side product under stoichiometric conditions.
Fig. 4

CO conversion curves of LaFe1−xCuxO3 and La2CuO4 catalysts under stoichiometric (stoic1), lean, rich and during second cycle stoichiometric (stoic2) conditions respectively. LaFeO3 (blackfilled circle), LaFe0.95Cu0.05O3 (redopen circle), LaFe0.9Cu0.1O3 (bluefilled triangle),LaFe0.8Cu0.2O3 (greenopen square) and La2CuO4 (orangefilled square)

Table 5

Temperatures corresponding to 50 % conversion of CO, C3H6 C3H8, and H2 (stoichiometric 1) and NO (rich composition) of LaFe1−xCuxO3 catalysts and selectivities of NO reduction products

Condition

Stoichiometric

Rich

T50 (°C)

 

Selectivities at TS

CO

C3H6

C3H8

H2

NO

S N2

S NH3

S N2O

TS (C)

LaFeO3

438

463

478

501

a

0.05

0.95

0

507

LaFe0.95Cu0.05O3

328

510

n.a.

440

485

0.85

0.11

0.03

499

LaFe0.9Cu0.1O3

305

469

467b

444

489

0.51

0.43

0.07

493

LaFe0.8Cu0.2O3

294

497

505b

421

501b

0.94

0

0.06

513

La2CuO4

300

415

293

479

1

0

0

489

aConversion lower than 50 %

bT25

Different reaction mechanisms for NO reduction were proposed on LaFe0.8Cu0.2O3 catalyst depending on the reductive agent. According to Zhang et al. nitrosyl (NO) species are the main adsorbed species during NO reduction by CO [8] whereas the NO reduction by propylene in the presence of oxygen involved the formation of nitrates (NO3) at the perovskite surface [13]. Regarding the light-off curves, it appears unlikely that CO + NO reaction occurred under three-way catalytic conditions. The NO reduction by propylene or hydrogen was more probable in this case.

Specific reaction rates calculated at 250 °C are reported in Table 6 for CO oxidation under stoichiometric conditions. The catalytic performance of LaFe1−xCuxO3 catalysts in CO oxidation was markedly promoted with increasing Cu-content. Kinetic data in Table 6 also revealed a copper content dependency with regards to the specific rate of CO oxidation. Further comparison of the reaction rate could be obtained by calculating intrinsic reaction rates. The intrinsic reaction rates values expressed per surface Cu atoms recorded on LaFe1-xCuxO3 with x = 0.05, 0.1 and 0.2 could lead to a rough estimation of TOF values. The relative concentration of copper atom at the surface was based on the surface atomic \({{{\text{x}}_{\text{Cu}} } \mathord{\left/ {\vphantom {{{\text{x}}_{\text{Cu}} } {\left( {{\text{x}}_{\text{La}} + {\text{x}}_{\text{Fe}} } \right)}}} \right. \kern-0pt} {\left( {{\text{x}}_{\text{La}} + {\text{x}}_{\text{Fe}} } \right)}}\) ratio previously estimated from XPS analysis. A proportional relationship was observed by plotting the intrinsic rate vs surface atomic \({{\text{Cu}} \mathord{\left/ {\vphantom {{\text{Cu}} {\left( {{\text{La}} + {\text{Fe}}} \right)}}} \right. \kern-0pt} {\left( {{\text{La}} + {\text{Fe}}} \right)}}\) ratio as showed in Fig. 5. This is consistent with the same coordination and symmetry of copper ion in La2CuO4 and LaFe1−xCuxO3. This calculation seems to emphasize that no significant segregation/agglomeration of copper at the surface would occur during the preparation of those samples especially during the calcination. Finally, the weak influence of copper content on the propylene conversion would suggest the involvement of different active sites on the substituted orthoferrites.
Table 6

Comparison of specific reaction rates rspecific of CO and C3H6 oxidation for LaFe1−xCuxO3 catalysts under stoichiometric conditions

Catalyst

rspecifica

rintrinsica

(mol.s−1 g−1)

(mol.s−1.at. Cu surf−1)

CO

CO

LaFeO3

≈0

0

LaFe0.95Cu0.05O3

1.1 × 10−8

0.57 × 10−4

LaFe0.9Cu0.1O3

8.4 × 10−8

2.50 × 10−4

LaFe0.8Cu0.2O3

22.5 × 10−8

3.08 × 10−4

La2CuO4

28.5 × 10−8

5.54 × 10−4

aSpecific reaction rates were calculated from conversions at 250 °C for CO and 350 °C for C3H6

Fig. 5

Intrinsic reaction rate for CO oxidation for LaFe1−xCuxO3 and La2CuO4 catalysts under stoichiometric conditions

Finally the stability of perovskite structure was investigated with a second cycle in stoichiometric conditions (stoic2) and with ex situ XRD. All solids exhibited the same crystalline structure than before the catalytic measurements except for the La2CuO4 solid which completely decomposed to La2O2CO3, La2O3 and CuO. For this reason, the lack of thermal stability of La2CuO4 in rich conditions completely limits its applicability to TWC. On the other hand, the optimal catalytic solution for practical application seems to be the LaFe1−xCuxO3 catalyst with the x value comprised between 0.05 and 0.2 and which shows an appropriate thermal stability.

4 Conclusions

A series of Cu-substituted orthoferrites was prepared by citrate method and was compared to LaFeO3 and La2CuO4 references. The catalytic properties were investigated for three-way catalysis application in stoichiometric, lean and rich conditions. LaFe1−xCuxO3 catalysts (x = 0.05, 0.1, 0.2) present an orthorhombic structure similar to the LaFeO3 solid whereas the Ruddlesden Popper phase is obtained for La2CuO4. The partial substitution of iron by copper in B-site of perovskite led to an enhancement of reducibility at between 150 and 350 °C as evidenced by H2-TPR. XPS revealed higher amounts of adsorbed oxygen and carbonate species on the surface of the Cu-substituted perovskites. The Ruddlesden Popper phase La2CuO4 however completely decomposed in rich conditions which limit any practical application of La2CuO4 for TWC. On the other hand the LaFe1−xCuxO3 catalysts (x = 0.05, 0.1, 0.2) were found stable even after one cycle. The partial substitution of iron by copper LaFe1−xCuxO3 led to increase significantly the catalytic activity especially for CO oxidation. Intrinsic reaction rate for CO oxidation was related to the concentration of copper atoms. Finally the LaFe0.9Cu0.1O3 catalyst yielded to the most interesting results concerning HC and CO oxidation while the highest catalytic activity for CO oxidation was observed for LaFe0.8Cu0.2O3 catalyst.

Notes

Acknowledgments

The research leading to these results has received funding from the European Union’s 7th Framework Programme under grant agreement no 280890-NEXT-GEN-CAT. The “Fonds Européen de Développement Régional (FEDER)”, “CNRS”, “Région Nord Pas-de-Calais” and “Ministère de l’Education Nationale de l’Enseignement Supérieur et de la Recherche” are acknowledged for funding X-ray diffractometers.

Supplementary material

11244_2016_615_MOESM1_ESM.docx (1.2 mb)
Supplementary material 1 (DOCX 1275 kb)

References

  1. 1.
    Granger P, Parvulescu VI (2011) Catalytic NOx abatement systems for mobile sources: from three-way to lean burn after treatment technologies. Chem Rev 11(5):3155–3207CrossRefGoogle Scholar
  2. 2.
    Ad hoc working group on defining critical raw materials, Report on critical raw materials for the EU, (2014)Google Scholar
  3. 3.
    Dacquin JP, Cabié M, Henry CR, Lancelot C, Dujardin C, Raouf S, Granger P (2010) Structural changes of nano-Pt particles during thermal ageing: support-induced effect and related impact on the catalytic performances. J Catal 270(2):299–309CrossRefGoogle Scholar
  4. 4.
    Dacquin JP, Lancelot C, Dujardin C, Cordier-Robert C, Granger P (2011) Support-induced effects of LaFeO3 perovskite on the catalytic performances of supported Pt catalysts in DeNOx applications. J Phys Chem C 115(5):1911–1921CrossRefGoogle Scholar
  5. 5.
    Schön A, Dujardin C, Dacquin JP, Granger P (2015) Enhancing catalytic activity of perovskite-based catalysts in three-way catalysis by surface composition optimization. Catal Today 258:543–548CrossRefGoogle Scholar
  6. 6.
    Wu Y, Ni X, Beaurain A, Dujardin C, Granger P (2012) Stoichiometric and non-stoichiometric perovskite-based catalysts: consequences on surface properties and on catalytic performances in the decomposition of N2O from nitric acid plants. Appl Catal B Environ 125:149–157CrossRefGoogle Scholar
  7. 7.
    Peña MA, Fierro JLG (2001) Chemical structures and performance of perovskite oxides. Chem Rev 101(7):1981–2018CrossRefGoogle Scholar
  8. 8.
    Zhang R, Alamdari H, Kaliaguine S (2006) Fe-based perovskites substituted by copper and palladium for NO + CO reaction. J Catal 242:241–253CrossRefGoogle Scholar
  9. 9.
    Caronna T, Fontana F, Sora IN, Pelosato R (2009) Chemical synthesis and structural characterization of the substitution compound LaFe1−xCuxO3 (x = 0–0.40). Mater Chem Phys 116:645–648CrossRefGoogle Scholar
  10. 10.
    Taguchi H, Matsu-ura S, Nagao M, Choso T, Tabata K (1997) Synthesis of LaMnO3+δ by firing gels using citric acid. J Solid State Chem 129:60–65CrossRefGoogle Scholar
  11. 11.
    Anderson JA, Fierro JLG (1994) Bulk and surface properties of copper-containing oxides of the general forumula LaZr1−xCuxO3. J Solid State Chem 108:305–313CrossRefGoogle Scholar
  12. 12.
    Terribile D, Trovarelli A, de Leitenburg C, Primavera A, Dolcetti G (1999) Catalytic combustion of hydrocarbons with Mn and Cu-doped ceria-zirconia solid solutions. Catal Today 47:133–140CrossRefGoogle Scholar
  13. 13.
    Twagirashema I, Frere M, Gengembre L, Dujardin C, Granger P (2007) Structural regeneration of LaCoO3 perovskite-based catalysts during the NO + H2 + O2 reactions. Top Catal 42–43:171–176CrossRefGoogle Scholar
  14. 14.
    Zhang R, Villanueva A, Alamdari H, Kaliaguine S (2006) Cu- and Pd-substituted nanoscale Fe-based perovskites for selective catalytic reduction of NO by propene. J Catal 237:368–380CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2016

Authors and Affiliations

  • Anke Schön
    • 1
  • Jean-Philippe Dacquin
    • 1
  • Christophe Dujardin
    • 1
  • Pascal Granger
    • 1
  1. 1.Univ. Lille, CNRS, ENSCL, Centrale Lille, Univ. Artois, UMR 8181–UCCS-Unité de Catalyse et Chimie du SolideLilleFrance

Personalised recommendations