1 Introduction

Internal combustion engines (ICEs), such as diesel and gasoline engines, are used for primary power generation in the industry, heavy transport, marine, and aviation sectors [1, 2]. Diesel engines are preferred for these applications because their efficiency and durability are higher than that of gasoline engines [3, 4]. However, diesel engines generate pollutants such as particulate matter (also called soot particles) and nitrogen oxides (NOx), being both highly dangerous for human health and the environment [5,6,7]. Consequently, vehicle emission regulations are becoming more demanding, and the use of after-treatments systems is mandatory. Thus, a combination of diesel oxidation catalyst (DOC) and diesel particulate filter (DPF) is employed as diesel after-treatment system to meet emission regulations [8, 9]. However, DPF is made up of a ceramic honeycomb monolith coated with noble metals (Pt, Pt/Pd or Pd/Rh) [8, 9], being these PGMs scarce and expensive [10, 11].

Mixed oxides with perovskite-like structure (ABO3) present interesting properties to be used as catalysts for different reactions related to the control of pollutants generated by ICEs [12], and they are considered, among others, as a potential alternative to the use of noble metals for soot removal from diesel engines [13,14,15,16,17]. In a previous study, a series of BaFe1−xCuxO3 catalysts (x = 0, 0.1, 0.3 and 0.4) were synthesized, characterized, and tested for soot oxidation under gasoline direct injection (GDI) exhaust conditions [16, 18] and diesel exhaust conditions [16]. The BaFe1−xCuxO3 perovskites catalyze both, the NO2-assisted diesel soot oxidation and, to a lesser extent, the soot oxidation during the “fuel cut” stage of GDI exhaust conditions (1% O2). BFC0 is the most active catalyst as the activity seems to be mainly related with the amount of O2 evolved during an O2-TPD (related to lattice oxygen mobility) and decreases with the copper content. On the other hand, it has been published that the addition of Ni to Co3O4 induces a structural distortion, improving the density of oxygen vacancies and enhancing the lattice oxygen mobility [19]. Thus, considering this background, in this paper, the effect of the partial substitution of Fe by Ni in a BaFeO3 perovskite (BaFe1−xNixO3 catalysts, with x = 0, 0.1, 0.3 and 0.4) for the catalysis of NOx-assisted diesel soot oxidation reaction has been explored.

2 Experimental

2.1 Synthesis and Characterization of Catalysts

The BaFe1−xNixO3 (x = 0, 0.2, 0.4 and 0.8) catalysts were synthesized by sol–gel method using citric acid and EDTA as a chelating agent to facilitate the solution of metal precursors in the synthesis conditions [20, 21]. Briefly, a 40 ml solution of EDTA (in a 1:2 molar Fe: EDTA ratio) at 60 °C was prepared. Ba(NO3)2, Fe(NO3)3·9·H2O and Ni(NO3)2·6·H2O were added as metal precursors and, finally, citric acid was incorporated using the same molar ratio (1:2, Fe: citric acid). The pH of the solution is adjusted to 9 with an ammonia solution (30% wt). Subsequently, the temperature was increased to 80 °C to ensure the formation of a gel. The obtained gel was dried at 150 °C for 12 h and, as a final step, the solid was calcinated at 850 °C for 6 h.

For sample characterization, the following techniques were employed.

The nickel content was measured by ICP-OES on a Perkin-Elmer device model Optimal 4300 DV. For this analysis, the elements were extracted by the mineralization of the samples using a diluted aqua regia solution (HNO3: HCl, 1:3) and stirring at 60 °C for 1 h.

The textural properties were determined by N2 adsorption at −196 °C using an Autosorb-6B instrument from Quantachrome (Anton Paar Austria GmbH). The samples were degassed at 250 °C for 4 h before the N2 adsorption experiments.

The crystalline structure was obtained using X-ray Diffraction (XRD). The X-ray patterns were recorded between 20 and 80° 2θ angles with a step rate of 0.4°/min and using Cu Kα (0,15,418 nm) radiation in a Bruker D8-Advance device.

The surface chemistry was assessed by X-ray Photoelectron Spectroscopy (XPS) using a K-Alpha Photoelectron Spectrometer by Thermo-Scientific with an Al Kα (1486.7 eV) radiation source. To obtain XPS spectra, the pressure of the analysis chamber was maintained at 5 × 10–10 mbar. The binding energy (BE) and kinetic energy (KE) scales were adjusted by setting the C 1 s transition at 284.6 eV, and the BE and KE values were then determined with the peak-fit software of the spectrometer. The OL/(Ba + Fe + Ni), Ni/(Ba + Fe + Ni), and Fe(IV)/Fe(III) XPS ratios were calculated using the area under the suggested deconvoluted bands of O 1 s, Mn 2p3/2, Fe2p3/2, Ni 2p3/2 and Ba 3d5/2.

The reducibility of catalysts was determined by Temperature Programmed Reduction with H2 (H2-TPR) in a Pulse Chemisorb 2705 (from Micromeritics) fitted with a Thermal Conductivity Detector (TCD) and using 30 mg of the sample which was heated at 10 °C/min from 25 to 1000 °C in 5% H2/Ar atmosphere (40 ml/min). The quantification of the H2 consumption was carried out using a CuO reference sample.

Temperature Programmed Oxygen Desorption (O2-TPD) experiments were performed in a TG-MS (Q-600-TA and Thermostar from Balzers Instruments (Pfeiffer Vacuum GmbH, Germany) respectively), with 16 mg of the sample heated at 10 °C/min from room temperature to 950 °C under a 100 ml/min of helium atmosphere. The 18, 28, 32 and 44 m/z signals were followed for H2O, CO, O2 and CO2 (respectively) evolved during the experiments. The amount of evolved oxygen was estimated using a CuO reference sample.

TEM images were obtained in a 120 kv JEM-1400 PLUS transmission electron microscope from JEOL, available at the technical services of the University of Alicante, equipped with an ORIUS GATAN image acquisition camera. The equipment reaches a resolution of 0.38 nm between two points and 0.2 nm between lines.

2.2 Activity Tests

The activity for NO to NO2 and NOx-assisted diesel soot oxidation was determined by Temperature Programmed Reaction in a quartz fixed-bed reactor, heated up from 25 to 800 °C (10 °C/min), under a gas flow mixture (500 ml/min) containing 500 ppm NOx, 5% O2, balanced with N2. For NO to NO2 oxidation experiments (NOx-TPR), 80 mg of catalyst was diluted with 320 mg SiC. Soot oxidation tests (Soot-NOx-TPR) were performed by mixing 80 mg of catalyst and 20 mg of Printex-U (the carbon black used as model soot) with a spatula to ensure loose contact, and the mixture was diluted with 300 mg of SiC. The most active catalysts were also tested under isothermal soot oxidation conditions at 450 °C for 180 min. The gas composition was monitored by specific NDIR-UV gas analysers for NO, NO2, CO, CO2 and O2 (Rosemount Analytical Model BINOS 1001, 1004 and 100, Emerson Electric Co., St. Louis, MO, USA). The NOx conversion and the NO2 generation percentages were calculated using the following equations:

$${\text{NO}}_{{\text{x}}} {\text{ conversion}}\left( {\text{\% }} \right) = \frac{{\left( {{\text{NO}}_{{{\text{x}},{\text{in}}}} - {\text{NO}}_{{{\text{x}},{\text{out}}}} } \right)}}{{{\text{NO}}_{{{\text{x}},{\text{in}}}} }} \times 100$$
(1)
$${\text{NO}}_{2} {\text{ generation}}\left( {\text{\% }} \right) = \frac{{{\text{NO}}_{{2,{\text{out}}}} }}{{{\text{NO}}_{{{\text{x}},{\text{out}}}} }} \times 100$$
(2)

where NO2, out and NOx,out are the NO2 and NOx (NO + NO2) concentrations measured at the reactor exit, being NOx,in the NO + NO2 concentrations measured at the reactor inlet.

The soot conversion and CO2 selectivity were determined as:

$${\text{Soot conversion }}\left( {\text{\% }} \right) = \frac{{\mathop \sum \nolimits_{0}^{{\text{t}}} {\text{CO}}_{2} + {\text{CO}}}}{{\mathop \sum \nolimits_{0}^{{{\text{final}}}} ({\text{CO}}_{2} + {\text{CO}})}} \times 100$$
(3)
$${\text{CO}}_{2} {\text{ Selectivity }}\left( {\text{\% }} \right) = \frac{{{\text{CO}}_{2} }}{{\mathop \sum \nolimits_{0}^{{{\text{final}}}} ({\text{CO}}_{2} + {\text{CO}})}} \times 100$$
(4)

where \(\sum_{0}^{t} ({\text{CO}}_{2} + {\text{CO}})\) is the amount of CO2 and CO evolved at a time t, while \(\sum_{0}^{{{\text{final}}}} {\mkern 1mu} {\text{CO}}_{2}\) and \(\sum_{0}^{{{\text{final}}}} ({\text{CO}}_{2} + {\text{CO}})\) are the total amount of CO2 and CO + CO2 evolved during the test.

3 Results and Discussion

3.1 Chemical and Morphological Properties

Table 1 presents the nomenclature used for the catalysts, the actual nickel content of the catalysts, determined by ICP-OES, as well as the nominal content calculated from the stoichiometric formula.

Table 1 Nomenclature, nickel content and BET surface area of catalysts
Full size table

In general terms, the actual Ni contents are similar to the nominal ones, so, the synthesis method allows the incorporation of the required amount of this metal. On the other hand, BET data reveal that all BaFex−1NixO3 catalysts present a very low surface area, as it is expected for mixed oxides with perovskite like structure [22].

3.2 Crystalline Structure

XRD diffractograms obtained for BaFe1−xNixO3 catalysts are shown in Fig. 1. The unique crystalline phase detected for the BaFeO3 catalyst is the oxygen-deficient hexagonal perovskite-type structure (BaFeO2.67). Note that the incorporation of nickel significantly modifies the XRD patterns of BF-E. First, the catalysts with nickel present the oxygen-deficient hexagonal perovskite structure BaNiO2.36 so, it seems that nickel is not inserted into the BaFeO3 perovskite structure and instead of that, the hexagonal BaNiO3 perovskite is formed as main crystal phase. This founding can be justified based on the tolerance factor value (t) for perovskites series [12] (featured in Table 2) that have been calculated using the ionic radius of the components of perovskites (Ba+2: 175 ppm, Fe3: 64 ppm, Ni2+:78 ppm and O2−:140 ppm), which is closer to 1 for BaNiO3 than for any of the BaFe1−xNixO3 perovskites.

Fig. 1
figure 1

XRD patterns of the BaFe1−xNixO3 catalysts

Full size image
Table 2 Tolerance factor value
Full size table

However, note that the main peak corresponding to hexagonal BaNiO2.36 is overlapped with the main peak of BaFeO3 perovskite and this fact hinders the assignation of XRD peaks. On the other hand, BFN2E, BFN4E, and BFN8E also present NiO and BaCO3 peaks showing the heterogeneity of samples after nickel addition. It is remarkable that, in the presence of Ni, the structure of perovskite does not change to cubic, as observed for BaFe1−xCuxO3 series [16, 18], thus, it seems to support that Ni is not inserted into the BaFeO3 perovskite. Instead of that, Ni is forming a new BaNiO3 perovskite because, as commented above, the tolerance factor (t) for nickel perovskite is closer to the ideal value (see Table 2).

3.3 Surface Properties

The XPS technique allows the study of the surface composition of the samples (up to 5 nm in depth) and the oxidation state of the elements [23]. The XPS spectra for the transition Ni2p3/2 (Fig. 2a), Fe2p3/2 (Fig. 2b) and O1s (Fig. 2c) for the BaFe1−xNixO3 catalysts are presented in Fig. 2.

Fig. 2
figure 2

XPS spectra of the a Ni 2p3/2 transition, b Fe 2p3/2 transition and c O1s transition for the BaFe1−xNixO3 catalysts

Full size image

Figure 2a shows the XPS spectra for the Ni2p3/2 transition. According to the literature [24], binding energy values between 854 and 856 eV indicate the presence of Ni(II) species on the surface. The presence of Ni(II) is confirmed by the satellite peaks detected at 860 and 864 eV. The main signal is deconvoluted in two components corresponding to: (i) Ni (II) in BaNiO3 perovskite, at higher binding energy and (ii) Ni(II) in NiO particles on the catalyst surface, at lower binding energy. Note that the amount of Ni in BaNiO3 increases with the Ni content showing that the formation of this perovskite is promoted.

Table 3 compares the percentage of nickel on the surface calculated by XPS and the percentage of nickel determined by ICP-OES. In the BFN2-E and BFN8-E catalysts, the XPS value is much lower than that of ICP-OES, so the fraction of nickel on the surface is low. However, in the BFN4-E catalyst, the XPS value is more similar to the nickel content determined by ICP-OES than for the other samples, indicating that Ni is more homogeneously distributed. Therefore, Ni species are present both on the surface and inside the BaNiO3 perovskite lattice. Note that the BFN4-E catalyst presents the highest percentage of surface nickel (around 90% vs. around 40–50% for the other two catalysts).

Table 3 XPS characterization data of nickel
Full size table

Figure 2b shows the band deconvolution of the Fe2p3/2 transition. The lower energy band at ca. 710 eV indicates the presence of Fe(III), which is confirmed by the satellite peak of this oxidation state at 718 eV. The band with the highest binding energy, detected between 712 and 713 eV, indicates the presence of Fe(IV) [18, 25]. Table 4 shows the Fe(IV)/Fe(III) surface ratios, calculated from the corresponding peak areas. Note that, for the raw perovskite BaFeO3, Fe should be present as Fe(IV), according to stoichiometry to compensate the positive and negative charges of ions in the perovskite structure. However, as both Fe(III) and Fe(IV) coexist on the catalyst surface, a positive charge imbalance is generated which has to be compensated by the formation of oxygen vacancies. The data indicate that Fe(III) is present on the surface of all catalysts and, consequently, these perovskites should present oxygen vacancies.

Table 4 XPS characterization data of iron and oxygen
Full size table

Figure 2c shows the spectra for the O1s transition for all catalysts. According to the literature [18, 25], the O1s spectrum is usually deconvoluted into three bands corresponding to three oxygen species: (i) lattice oxygen (OL) (around 530 eV) (ii) surface oxygen species (adsorbed oxygen, hydroxyl and carbonate groups) around 531 eV and (iii) water adsorbed on the surface (at around 533 eV). Note that these bands are detected for all catalysts, being the main band corresponding to the surface oxygen species. Additionally, Table 4 includes the OL/(Ba + Fe + Ni) ratio, calculated from the atomic percentages obtained by XPS and the area of OL peak. The corresponding nominal ratio (calculated from the stoichiometric composition of the BaFe1−xNixO3 catalysts) is 1.5, but all catalysts present an OL/Fe + Ba + Ni ratio lower and far to the nominal one, which indicates that the perovskite structure presents a very low proportion of lattice oxygen, that is, it is a very distorted structure. The use of EDTA as additional chelating agent during sol–gel process seems to promote the distortion of the structure, as BaFeO3 obtained avoiding the use of EDTA [16, 18] presents a higher OL/Fe + Ba + Ni ratio, i.e., a less distorted structure. Thus, the main oxygen band is the Oads and OH−/CO32− bands as the formed perovskite is non-stoichiometric (due to the presence of Fe(III) and Ni(II)). Therefore, there is a positive charge imbalance that is compensated by the presence of oxygen vacancies. In the presence of Ni, the amount of oxygen vacancies seems to decrease as OL/Fe + Ba + Ni ratio increases probably as the BaNiO3 perovskite presents a less distorted structure.

3.4 Reducibility

The reducibility of the catalysts was determined by Temperature Programmed Reduction with H2 (H2-TPR) following the procedure described in Sect. 2.1. H2-TPR profiles (Fig. 3) show several reduction peaks between 200 °C and 900 °C, which corresponds to the reduction of Ni and Fe. Hosseinpour et al. [26] analysed the Fe2O3 reduction and observed three reduction peaks corresponding to the reductions of Fe2O3 to Fe3O4 (partial reduction of Fe(III) to Fe(II)) at 480 °C, and from Fe3O4 to FeO at 615 °C and finally from Fe(II) to Fe (0) at 825 °C. On the other hand, Bai et al. [27] studied the reduction of nickel perovskites and observed several reduction peaks: (i) at 230–240 °C, attributed to the reduction of NiO species present on the perovskite surface, with a weak interaction Ni-perovskite; (ii) between 410 °C and 500 °C, appear the reduction peaks due to NiOx species on surface with an intense interaction with perovskite surface and subsurface, and (iii) peaks at temperatures above 510 °C, attributed to reduction of Ni(II) present on the bulk perovskite structure.

Fig. 3
figure 3

H2-TPR profiles of the BaFe1−xNixO3 catalysts

Full size image

For BF-E sample, two well overlapped peaks between 200 °C and 480 °C are observed. According to the literature [26], these peaks correspond to the partial reduction of Fe(IV) to Fe(III) and to the reduction of Fe(III) to Fe(II). BFN2-E presents a reduction peak between 230 °C and 240 °C which could be assigned to the reduction of NiO species with a weak interaction with the perovskite [27] as, for this sample, XRD peaks corresponding to NiO have been detected (Fig. 1). For BFN4-E, the reduction profile shows three peaks around 400 °C that should correspond not only to the Fe reduction but also to the reduction of NiOx species on surface and subsurface of perovskite, that is to the reduction of NiOx species with more intense interaction with perovskite [27]. Thus, Fe and Ni reduction peaks seem to be overlapped for this sample. Finally, two well defined peaks appear on the BFN8-E catalyst: (i) the peak at a lower temperature (approximately at 430 °C) corresponds to the reduction of nickel on the perovskite surface and also to the reduction of Fe(IV) to Fe (III) and; (ii) the highest temperature peak (about 480 °C) corresponds to the nickel of the BaNiO3 perovskite and the reduction of Fe (III) to Fe (II). The absence of low-temperature Ni reduction peak for BFN4-E and BFN8-E samples is in agreement with the absence of XRD NiO peak in the diffractogram of these two samples (Fig. 1).

3.5 O2 Desorption

Figure 4 shows the O2 emission profiles obtained during the temperature programmed desorption (O2-TPD) for catalysts. According to the literature [28], perovskite-type mixed oxides evolve O2 at three temperatures that correspond to three types of oxygen: (i) the peak with a maximum below 400 °C is assigned to desorption of oxygen adsorbed on surface vacancies, called α-oxygen; (ii) the peaks with a maximum between 400 °C and 700 °C correspond to the desorption of oxygen coming from lattice defects such as dislocations, which is called α’-oxygen and, finally and; (iii) the peaks with a maximum above 700 °C correspond to the desorption of lattice oxygen, denoted as β-oxygen, which is related with the reducibility of the B cation perovskite and informs about the mobility of oxygen through the lattice.

Fig. 4
figure 4

O2-TPD profiles of the BaFe1−xNixO3 catalysts

Full size image

In the O2-TPD profiles, the observed peaks correspond to the desorption of α-oxygen, α’-and β-oxygen are detected. BF-E catalyst presents two O2 desorption peaks: (i) one between 200 °C and 300 °C, which corresponds to the desorption of oxygen adsorbed on surface vacancies (α-O2) and (ii) another more intense and defined peak with a maximum over 720 °C, which corresponds to the desorption of oxygen from the lattice (β-O2). These results confirm the presence of oxygen vacancies on BaFeO3 perovskite (the main crystal phase in BF-E), that have been created to compensate the positive charge imbalance due to the presence of Fe (III). BFN2-E and BFN4-E present also a peak, between 200 °C and 300 °C, which is more intense than the observed for BF-E, especially for BFN4-E. This means that BFN4-E presents a higher oxygen mobility at low temperatures which is expected to help the soot oxidation. This is because when Fe (III) is partially substituted by Ni(II), the positive charge imbalance increases, so, more oxygen vacancies are generated to achieve the electroneutrality. Note that for these samples, the high temperature peak (corresponding to the desorption of lattice oxygen) appears at higher temperatures because it also includes, the desorption of lattice oxygen coming from the BaNiO3 perovskite. This shift suggests that the desorption of oxygen coming from BaNiO3 perovskite is more difficult than the coming from BaFeO3 perovskite. Finally, BFN8-E catalyst presents not only a high temperature peak similar to BFN4-E, (which corresponds to the desorption of lattice oxygen from both perovskites) but also three peaks assigned to the desorption of oxygen chemisorbed in the lattice defects of hexagonal BaNiO2.67.

3.6 Catalytic Activity Tests

3.6.1 NO to NO2 Oxidation (NOx-TPR)

The activity of catalysts for the oxidation of NO to NO2 was studied by Temperature Programmed Reaction (NOx-TPR) experiments carried out using a gas mixture which simulates the composition of the exhaust gases of a diesel vehicle (500 ppm NOx and 5% O2 in N2). Figure 5a shows the NOx conversion profiles as a function of temperature.

Fig. 5
figure 5

a NOx conversion profiles under NOx-TPR conditions for the BaFe1−xNixO3 catalysts, b NO2 generation profiles obtained under NOx-TPR conditions for the BaFe1−xNixO3 catalysts

Full size image

Considering Eq. (1), a positive conversion value indicates that the amount of NOx detected at the reactor outlet is lower than that in the inlet, i.e., the NOx is being adsorbed on the catalyst. On the contrary, a negative value indicates that NOx is being desorbed. All catalysts show a certain NOx adsorption capacity, which decreases in the presence of Ni. Note that, as observed for other perovskite-based catalysts [15,16,17,18,19], the peak corresponding to NOx desorbed is larger than the peak corresponding to NOx adsorbed due to the amount of NOx adsorbed at room temperature before starting the NOx-TPR experiment which is not registered.

In Fig. 5b the NO2 generation profiles as a function of temperature during NOx-TPR experiments are shown. The percentage of NO2 generation (Eq. (2)) represents the fraction of NO2 at the outlet of the reactor. However, these profiles cannot indicate the real value of NO2 generated because, as indicated by Fig. 5a, a fraction of the NO2 is being adsorbed on the catalyst surface, so, it cannot be detected. Note that NO2 generation profiles indicate that all catalysts are active for the oxidation of NO to NO2 at temperatures below 450 °C, but the oxidation capacity is not directly related to the nickel content as, except for the BFN4-E catalyst, nickel-containing catalysts have a lower activity than the nickel-free catalyst (BF-E). Thus, BF-E and BFN4-E catalysts are the catalysts presenting the highest activity to generate NO2 with a maximum of around 38% at 400 °C. These two samples present also higher NOx adsorption capacity, so, their activity for NO to NO2 oxidation should be much higher. In fact, BFN4-E catalyst is more active for NOx adsorption than BFN2-E and BFN8-E, probably because it presents the highest proportion of surface nickel detected for this sample and the highest oxygen mobility at low temperatures (see Table 3 and Fig. 4).

3.6.2 Soot Oxidation (Soot-NOx-TPR)

Soot oxidation tests were carried out in Temperature Programmed Reaction (Soot- NOx-TPR) conditions using the same gas mixture described in Sect. 2.2 and Printex-U in loose contact, as model soot. As BF-E and BFN4-E are the most active catalysts for NO2 generation (see Fig. 5) and, considering that NO2 is much more active than O2 for soot oxidation [15,16,17,18,19], these two catalysts have been selected to carry out several consecutive cycles of Soot- NOx-TPR, in order to determine their stability in reaction conditions. Additionally, they have been used in isothermal soot oxidation reaction at 450 °C.

Figure 6 shows the soot conversion profiles (Soot- NOx-TPR) for BF-E and BFN4-E catalysts. Note that, at temperatures below 600 °C, which is the interval of interest for the removal of soot from diesel engines [29], the two catalysts present conversion values higher than the observed in the absence of catalysts (blank in Fig. 6), therefore, both catalysts are active confirming that the NO2 generated is very effective for soot oxidation [15,16,17,18,19].

Fig. 6
figure 6

Soot conversion profiles obtained under TPR conditions for BF-E and BFN4-E catalysts

Full size image

From these profiles, the temperature at which 50% conversion is reached (T50%), was determined and the data is included in Table 5, where also the selectivity to CO2 (calculated by Eq. (4) based on the amount of CO and CO2 coming from soot oxidation) is shown.

Table 5 T50% and soot oxidize selectivity to CO2 for BF-E and BFN4-E catalysts
Full size table

As expected, the two catalysts show a high selectivity to CO2 [30, 31] and a lower T50% value than in the absence of catalyst (blank). To obtain more information about the performance of catalysts, four consecutive soot oxidation cycles were carried out, and Fig. 7 compares the values of T50% (°C). BF-E presents the highest deactivation during four cycles since the T50% (°C) continuously increases. BFN4-E shows a similar deactivation but only after the first cycle, being T50% values almost constant for the following cycles. Thus, it seems that the partial substitution of Fe by Ni in this catalyst enhances the catalytic stability [32].

Fig. 7
figure 7

T50% values in Soot- NOx-TPR- cycles

Full size image

BFN4-E shows the most stable performance during soot oxidation reaction because, as revealed by the soot conversion profiles at 450 °C features in Fig. 8, this catalyst presents a higher soot oxidation rate (25 µmol/min) than the BaFeO3 catalyst (11 µmol/min), minimizing the accumulation of soot during cycles and, thus, the deactivation. These soot deposits remain at the end of the experiment due to the high thermal stability of the carbon black used as model soot [33]. It is remarkable that, as revealed by XPS results above discussed (Fig. 2 and Table 4), BFN4-E catalyst presents the highest proportion of surface Ni(II) species that are able to increase the soot oxidation rate and, hence, helps to avoid the catalyst deactivation.

Fig. 8
figure 8

Isothermal reaction profiles for BF-E and BFN4-E catalysts at 450 °C

Full size image

In order to confirm that the deactivation of BF-E is due to the accumulation of unreacted soot during reaction cycles, BF-E and BFN4-E catalysts used in isothermal soot oxidation at 450 °C were characterized by TEM. Thus, the images feature in Fig. 9 evidence that BF-E accumulates more soot than BFN4-E because of the lower soot oxidation rate.

Fig. 9
figure 9

TEM images for catalysts after being used in the isothermal soot oxidation at 450 °C: a BF-E and b BFN4-E catalysts

Full size image

4 Conclusions

In this work, a series of BaFe1−xNixO3 (x = 0, 0.2, 0.4 and 0.8) catalysts were synthesized to determine the effect of partial substitution of Fe by Ni on the NO to NO2 oxidation and on the NO2-assisted diesel soot oxidation activity.

From the above discussed results, the following conclusions can be drawn:

  1. (i)

    Ni is not inserted in the BaFeO3 perovskite and, instead of that, forms a new BaNiO3 perovskite.

  2. (ii)

    Fe(III) and Fe(IV) coexist on catalysts surface, being Fe(III) the main oxidation state.

  3. (iii)

    Oxygen vacancies are present on catalysts to compensate the positive charge imbalance due to the presence of Fe(III) for BaFeO3 and of Fe(III) and Ni(II) for BaFe1−xNixO3.

  4. (iv)

    All catalysts are effective for NO oxidation to NO2, being BaFeO3 and BaFe0.6Ni0.4O3 the most active,

  5. (v)

    BaFe0.6Ni0.4O3 is the most stable catalyst for NOx-assisted diesel soot oxidation as it presents the highest proportion of Ni(II) surface species, which allows the highest initial soot oxidation rate and minimizes the accumulation of unreacted soot during reaction.