Introduction

Diesel engines, due to their higher power generation capability and good fuel economy, are extensively used in the transportation sector, especially in heavy-duty vehicles (Xin 2013). Particulate matter, i.e., soot produced during partial combustion of diesel fuel in fuel dense regions of the diesel engine (Pischinger et al. 1994) is a major contributor to air pollution and global warming (Ramanathan and Carmichael 2008) and is responsible for respiratory and cardiovascular diseases (Sydbom et al. 2001; Ristovski et al. 2012). Diesel particulate filter, an automotive exhaust emission control technology, traps soot from the exhaust of diesel engine and requires regeneration to avoid backpressure build-up due to blockage, which otherwise may adversely impact engine operation. Since soot ignites around 600 °C and the temperature of exhaust gases is much lower than 600 °C (Adler 2005), appropriate soot oxidation catalysts need to be developed to burn off soot at lower temperatures (Bueno-López et al. 2005; Andana et al. 2016).

CeO2 has been intensely researched for its catalytic activity in soot oxidation reactions. The exceptional redox properties of ceria (Ce3+/Ce4+) allow uptake and release of oxygen in oxygen-rich and lean conditions, respectively, resulting in good oxygen storage capacity (Setiabudi et al. 2004; Bueno-López et al. 2005; Machida et al. 2008). Gaseous oxygen may adsorb in the anionic vacancies of ceria forming active oxygen species, which react more spontaneously with soot as compared to gaseous oxygen and hence enhances soot oxidation reaction rate (Bueno-López et al. 2005). However, at high temperatures, it has been reported that aggregation of ceria leads to a reduction in surface area and changes its redox property and oxygen storage capacity (Atribak et al. 2008; Mukherjee and Reddy 2018). Incorporation of suitable dopants, like noble, transition and rare earth elements (Pt, Pd, Ru, Ag, Mn, Cu, Co, Fe, Zr, La, Pr, Hf, Gd, and  Nd) into the ceria lattice helps to modify properties like thermal stability, reducibility, and oxygen mobility of ceria to enhance soot oxidation activity (Bueno-López 2014; Liu et al. 2015; Andana et al. 2016; Mukherjee et al. 2016; Anantharaman et al. 2018; Patil et al. 2019). As per literature, noble metal (Ag, Ru, Pt, and Pd) supported ceria-based catalysts have been widely investigated in soot oxidation reactions (Aneggi et al. 2009; Kayama et al. 2010; Lim et al. 2011; Nascimento et al. 2014). Good metal-support interactions have shown to improve the reducibility of ceria and generate more active oxygen species, resulting in improved soot oxidation catalytic activity (Grabchenko et al. 2020). However, high cost and lesser availability of noble metals have encouraged research for other effective catalysts. Transition metals (Co, Cu, Fe, and Mn) on ceria-based support are studied for soot oxidation reaction as catalysts (Rao et al. 2011; Venkataswamy et al. 2014; Zhang et al. 2016; Xu et al. 2017). Due to the synergetic effect obtained by the addition of transitional metals to ceria-based catalysts, better catalytic performance is obtained in several oxidation reactions (Shen et al. 2013; Giménez-Mañogil et al. 2014; Sudarsanam et al. 2015; Putla et al. 2015).

Nickel-based catalysts are used to catalyze reactions like methane combustion reaction due to large oxygen adsorption capacity of NiO catalysts (Liu et al. 2017), water oxidation (Singh et al. 2013), CO oxidation (Singhania and Gupta 2017) and partial oxidation of methane (Alvarez-Galvan et al. 2019) reactions and are also used in fuel electrodes of solid oxide cells (Hauch et al. 2016). Recent studies on the Nickel-incorporated soot oxidation catalysts reported in the literature, such as NixCo3−xO4 spinel catalyst (Zhao et al. 2019), Mn1.9Ni0.1O3–δ (Neelapala et al. 2018), Ni/CoAl2O4 spinel (Álvarez-Docio et al. 2020) and SrMn0.98Ni0.02O3 perovskite catalysts (Uppara et al. 2019) have shown improved soot oxidation activity on nickel incorporation. Nickel incorporation in ceria lattice has also shown to have a positive impact on the soot oxidation activity of ceria (Bendieb Aberkane et al. 2019) due to an increase in the oxygen vacancies.

According to literature (Thrimurthulu et al. 2012; Guillén-Hurtado et al. 2015, 2020; Andana et al. 2016; Mukherjee et al. 2016), praseodymium-doped ceria performs better than pure ceria in catalyzing soot oxidation reactions. The addition of praseodymium as a dopant in ceria enhances the redox properties, leading to an increased number of oxygen vacancies (Andana et al. 2016; Guillén-Hurtado et al. 2020) and lattice oxygen mobility and also improves the thermal stability of ceria through solid solution formation (Thrimurthulu et al. 2012; Zhou et al. 2018). Effect of morphologies such as nanorod, nanocube, nanofiber (Andana et al. 2016; Jeong et al. 2019), and the effect of variation in Ce/Pr composition (Guillén-Hurtado et al. 2020) on soot oxidation activity have been extensively investigated. Copper loaded (Rico-Pérez et al. 2016) and Pt-impregnated (Andana et al. 2018) ceria-praseodymium mixed oxides have been investigated in literature for catalyzed soot oxidation reactions.

In the present study, different compositions of Nickel-modified praseodymium-doped ceria (xNi-PDC) (x = 0, 3, 5, 7, 10, 15, and 20 mol%) catalysts were synthesized by solution combustion synthesis (SCS) method and are studied for their catalytic activity towards diesel soot oxidation reaction.

Experimental

xNi-PDC (x = 0, 3, 5, 7, 10, 15, and 20 mol%) catalysts were synthesized by SCS method using glycine. Nickel Nitrate Hexahydrate (98%, Loba Chemie), Praseodymium (III) Nitrate Hexahydrate (99.9%, Aldrich), and Cerous Nitrate Hexahydrate (99.9%, SRL chemicals) were mixed in the stoichiometric ratio in water with continuous stirring and heating at 80 °C. Glycine (99.5%, SRL chemicals) was added to the mixture in the stirring condition with glycine to nitrate ratio as 0.12. The detailed synthesis procedure of solution combustion synthesis is similar to our earlier reports (Patil et al. 2019). The mixture was heated until a green-colored viscous gel was obtained and further heated at 250 °C in a hot air oven. The oven-dried powder obtained is calcined at 500 °C/2 h to obtain the desired xNi-PDC catalysts.

Characterization

The obtained catalysts were characterized by X-ray Diffraction (XRD, XPERT Pro diffractometer), Raman spectroscopy (Bruker RFS 27), high-resolution transmission electron microscopy (HR-TEM- JEOL/JEM 2100), and selected area electron diffraction (SAED) techniques as described in our earlier reports (Patil et al. 2019). XRD was carried out with Cu Kα radiation (λ  =  0.1540 nm) operating at 40 kV and 30 mA with 20°–80° 2θ range. Mean crystallite size (D) is determined using Scherrer’s equation [D = 0.9λ/β∙cosθ] and lattice strain (ε) is calculated using the Williamson-Hall equation [ε = β/(4tanθ)] (Bindu and Thomas 2014; Govindhan et al. 2018), where D is the crystallite size, λ is the wavelength of radiation, θ is the peak position, and β is the corrected peak width at half maximum intensity. Raman Spectroscopy with a 532 nm laser beam in the range of 200–3000 cm−1 wavelength was used to study the vibrational modes and oxygen vacancies present in the sample. HR-TEM with SAED analysis was used to understand the microstructure and morphology of the samples.

The soot oxidation experiment to evaluate the catalytic activity of samples was performed in thermogravimetric analysis equipment (TGA, TG–DTA 6300) as described in our previous report (Shenoy et al. 2019). TGA was operated up to 650 °C in an air atmosphere at a flow rate of 100 ml/min at a heating rate of 10 °C/min. Soot (Printex-U, Orion Engineered Chemicals) and xNi-PDC catalysts were mixed in a 1:10 ratio using electric mortar and pestle in tight contact mode for 30 min. The following formula is used for soot conversion (Zeng et al. 2020).

$${\text{Soot conversion }} = \, \left[ {\left( {W_{\text{i}} {-}W_{\text{o}} } \right) /\left( {W_{\text{i}} {-}W_{\text{e}} } \right)} \right] \, \times { 1}00$$

where, Wi is the initial sample weight, We is weight after the heating and Wo is the weight at each temperature point.

The soot oxidation kinetic data of PDC, 5Ni-PDC and 10 Ni-PDC catalysts is obtained at various constant heating rates (5, 10, 15 and 20 °C/min) and the activation energy is calculated using Ozawa equation [log β = B − 0.4567 (Ea/RTα)] (Zeng et al. 2020), where β is the constant heating rate, Tα is the absolute temperature at conversion α, Ea is the activation energy, B is a constant and R is the universal gas constant. A plot of log β versus 1/Tα graph gives a straight line for each percentage conversion (20, 30, 40, 50, 60, 70, 80, and 90%), and apparent activation energy can be calculated from the slope.

Results and discussion

XRD patterns of xNi-PDC catalysts synthesized by SCS and calcined in air at 500 °C/2 h are shown in Fig. 1, and the peaks correspond to the cubic fluorite structure of ceria (Chen and Chang 2005; Tok et al. 2007; Zhou et al. 2016). The peaks related to Ni (2θ = 44.4°, 51.8°, 76.4°) (Miniach et al. 2016)/NiO (2θ = 37.28°, 43.3°, 62.8°, 75.3) (Venugopal et al. 2007) and PrO2 (2θ = 28.6°, 33.2°, 47.6°, 56.5°, 59.3°, 69.6°) (Yadav et al. 2011)/Pr2O3 (2θ = 26.6°, 29.7°, 30.5°, 40.3°, 47.0°, 53.2°, 56.6°, 57.2°) (Lo Nigro et al. 2003)/Pr6O11 (2θ = 28.25°, 32.739°, 46.995°, 55.707°, 58.426°, 68.590°, 75.735°, 78.085°) (Abu-Zied et al. 2013; Matović et al. 2013) were not observed up to 15 mol% nickel doping in PDC samples, indicating Ni–Ce–Pr–O solid-solution formation. For the 20Ni-PDC catalyst, a small peak is observed at 2θ = 42.95°, which is attributed to the nickel oxide phase (Ding et al. 2018), indicating that some of the Nickel added as a dopant is present as NiO. Table 1 tabulates the crystallite size and lattice strain data of xNi-PDC catalysts. The average crystallite size varies from 6 to 8 nm for all catalysts. With the addition of Ni up to 7 mol%, as a dopant in PDC, the lattice strain is increased from 0.0131 to 0.0167. With a further 10–20 mol% increase in nickel content, the lattice strain is almost constant. Thus, it can be inferred that the addition of Ni from 10–20 mol% resulted in no significant change in lattice strain. In general, an increase in lattice strain due to lattice distortion on the incorporation of dopant results in better oxygen diffusion and oxygen migration, which enhances the oxidation reaction (Kehoe et al. 2011; Rushton and Chroneos 2014). The facet ratios, [100]/[111] and [110]/[111] were calculated and tabulated in Table 1. [100]/[111] is found to increase up to 10 mol% nickel doping and is highest for 10Ni-PDC sample. [110]/[111] facet ratio is found to decrease up to 7 mol% nickel doping in PDC, followed by an increase for 10 and 15 mol% nickel doping. Low facet ratio values observed for 20 mol% nickel doping may be attributed to the formation of the secondary NiO phase. In comparison to (111) plane, (100) and (110) planes have lower oxygen vacancy formation energy and hence are more reactive (Nolan et al. 2005; Aneggi et al. 2014; Andana et al. 2016; Capdevila-Cortada et al. 2016; Patil and Dasari 2019), thus indicating that 10 and 15 mol% Ni-doped PDC samples with highest [100]/[111] and [110]/[111] facet ratios respectively may show better soot oxidation activity.

Fig. 1
figure 1

XRD patterns of all Ni-PDC samples synthesized by SCS calcined at 500 °C/2 h in air

Full size image
Table 1 Crystallite size, lattice strain, XRD peak intensity ratios, Raman peak intensity ratio and T50 of all Ni-PDC samples synthesized by SCS and calcined in air at 500 °C/2 h
Full size table

Figure 2 demonstrates the Raman spectra of xNi-PDC catalysts carried out using a 532 nm laser beam. The Raman band at (450–456) cm−1 corresponds to the F2g mode of CeO2 with O–Ce–O symmetric breathing mode (Keramidas and White 1973; Reddy et al. 2009). The band at (558–573) cm−1 corresponds to the oxygen vacancy defects present in xNi-PDC catalysts (McBride et al. 1994; Luo et al. 2006b; Reddy et al. 2009; Martínez-Munuera et al. 2019). For NiO nanopowder, a one phonon band around 570 cm−1 corresponding to the presence of defects is the most pronounced, as reported by Mironova-Ulmane et al. (Mironova-Ulmane et al. 2007). Since it overlaps with the oxygen vacancy band of CeO2, it is difficult to identify bands corresponding to NiO. The intensity ratio of oxygen vacancy peak to the F2g mode peak (Iov/IF2g) tabulated in Table 1 is taken as a parameter to quantify the oxygen vacancies present in all synthesized samples (Luo et al. 2006a; Pu et al. 2007; Reddy et al. 2009). An increase in the relative intensity ratio is noted with the addition of Ni as a dopant. 10, 15, and 20 mol% of Ni doping resulted in almost similar values indicating that the addition of Ni over 10 mol% in the PDC system resulted in the generation of almost constant oxygen vacancies.

Fig. 2
figure 2

Raman spectra of all Ni-PDC samples synthesized by SCS calcined at 500 °C/2 h in air

Full size image

The morphology, particle size, and planes of PDC, 5Ni-PDC, and 10Ni-PDC samples were analyzed by TEM and SAED techniques and are shown in Figs. 3, 4, and 5, respectively. From Figs. 3a, 4a and 5a, agglomerated particles with an irregular elongated shape were observed from PDC, 5Ni-PDC and 10Ni-PDC, respectively. The particle sizes were found to be 6–10 nm for the PDC sample, 4–10 nm for the 5Ni-PDC sample, and 4–7 nm for the 10Ni-PDC sample, which is in the range of the crystallite size calculated from XRD analysis. The lattice spacing calculated from TEM images for PDC (Fig. 3b), 5Ni-PDC (Fig. 4b), and 10Ni-PDC (Fig. 5b) samples correspond to planes of ceria. The SAED patterns for all three samples depict the polycrystalline nature of the samples. (111), (200), (220) and (311) planes from the SAED patterns (Figs. 3c, 4c, and 5c) correspond to the fluorite structure of ceria. No plane corresponding to NiO is observed in both SAED and TEM patterns for PDC, 5Ni-PDC, and 10Ni-PDC samples, as observed in their XRD analysis. From XRD patterns (Fig. 1a), SAED patterns (Figs. 3c, 4c, and 5c) and lattice strain (Table 1), it can be observed that the addition of Ni in PDC catalysts up to 10 mol% resulted in no secondary phase of NiO (from XRD patterns and SAED patterns) and lattice strain increased up to 10Ni-PDC, and further addition has not increased lattice strain. The results confirm that the Ni form a solid-solution with PDC up to the 10Ni-PDC sample.

Fig. 3
figure 3

a TEM image, b HR-TEM image and c SAED pattern of PDC sample synthesized by SCS calcined at 500 °C/2 h in air

Full size image
Fig. 4
figure 4

a TEM image, b HR-TEM image and c SAED pattern of 5Ni-PDC sample synthesized by SCS calcined at 500 °C/2 h in air

Full size image
Fig. 5
figure 5

a TEM image, b HR-TEM image and c SAED pattern of 10Ni-PDC sample synthesized by SCS calcined at 500 °C/2 h in air

Full size image

Figure 6a illustrates the soot oxidation activity of xNi-PDC catalysts, and Table 1 tabulates its corresponding T50 temperatures. With the addition of the Ni from 0 to 10 mol% in the PDC system, the T50 temperature decreased from 374 to 360 °C, and further addition of Ni (i.e., 15 and 20 mol%) in PDC system haven’t decreased the T50 temperature further. The soot oxidation activity order is 20Ni-PDC = 15Ni-PDC = 10Ni-PDC > 7Ni-PDC > 5Ni-PDC > 3Ni-PDC > PDC. The T50 temperature for catalyzed soot oxidation reactions carried out in the present study is lower than that for uncatalyzed soot oxidation reaction by 200–230 °C.

Fig. 6
figure 6

a Soot conversion versus temperature plots for uncatalyzed soot oxidation and catalyzed soot oxidation of xNi-PDC catalysts at 10 °C/min heating rate. b Plot of T50, lattice strain and Iov/F2g Raman intensity ratio versus Ni mol%. c 10Ni-PDC catalyst reproducibility test for three runs for soot oxidation activity

Full size image

From the soot oxidation activity results of xNi-PDC catalysts, it can be interpreted that the soot oxidation activity of xNi-PDC catalysts increased with the increase in the Ni mol% from 0 to 10 mol% and a further increase in the Ni mol% have almost no impact on the catalytic activity, thus indicating that the catalytic activity has been saturated and only 10 mol% of Ni is enough to show a better catalytic activity. From XRD analysis and Raman spectroscopy analysis, the lattice strain and relative intensity ratio (Iov/F2g) (Fig. 6b) showed similar trend indicating that the catalytic activity trend followed the lattice strain and oxygen vacancy (relative intensity ratio) and hence the catalytic activity is controlled by the lattice strain and oxygen vacancy xNi-PDC catalysts for soot oxidation.

Table 2 summarizes the information related to the Ni/Ce/Pr catalysts activation energy, T50 temperature, and soot oxidation experimental conditions obtained from the literature and compared with 10Ni-PDC catalyst performance. Under tight contact mode, 10Ni-PDC catalysts showed better T50 temperature than the Ce0.5Pr0.5O2 (nano-fiber) (Jeong et al. 2019), Ce0.9Pr0.1O2 (Jeong et al. 2019), CeO2 (nanocube) (Jeong et al. 2019), Ce–Pr oxide (Jeong et al. 2019), Ce–Pr oxide (50:50) (nanocube/nanorod) (Andana et al. 2016), CeO2–Pr2O3 (8:2) (Mukherjee et al. 2016), Ce0.8Pr0.2O2 (Thrimurthulu et al. 2012), NiO/CeO2 (Bendieb Aberkane et al. 2019) and CeO2 (Palmisano et al. 2006) catalysts. The reason for such kind of better activity can be due to the synthesis method, Ni, and Pr loading in ceria (which controls the physio-chemical properties). xNi-PDC catalysts (360–374 °C) showed much better T50 temperature than the perovskite catalysts [BSCF (460 °C), LSCF (500 °C) and LSCM (474 °C)] prepared by reverse co-precipitation method (Shenoy et al. 2019) and Nd and Gd-doped ceria catalysts (427 °C) prepared by glycine-nitrate-process (Anantharaman et al. 2018; Patil et al. 2019). The present study also shows that promotional effect of nickel addition on soot oxidation activity (T50 = 360–374 °C) of Ce0.9Pr0.1O2 is better than the promotional effect on soot oxidation activity (T50 = 386–410 °C) of cobalt addition to Ce0.5Zr0.5O2 (Zhang et al. 2016).

Table 2 Comparison of T50 temperature, activation energy, and soot oxidation experimental conditions of various Ni/Ce/Pr catalysts
Full size table

The activity tests were conducted three times on each catalyst to check the reproducibility of the soot oxidation activity, and T50 is almost similar to an error range of less than ± 1.5 °C (see Table 1). Figure 6c shows the soot oxidation activity of 10Ni-PDC catalysts carried out for three runs to check the reproducibility, and each run showed similar activity (change T50 is with ± 1.2 °C) and Figure S1 in supplementary information provides the reproducibility of the soot oxidation activity of xNi-PDC catalysts. Since soot oxidation experiments in the present study have been performed inflow of air, the active oxygen mechanism can occur during the soot oxidation process (Guillén-Hurtado et al. 2015).

Soot oxidation activity kinetics is further explored on the PDC, 5Ni-PDC, and 10Ni-PDC catalysts to obtain activation energy, which can further help in analyzing the catalytic activity of synthesized oxides. Figure 7 shows the sigmoidal soot conversion versus temperature curves for all experiments carried out at 5, 10, 15, and 20 °C/min heating rates over PDC (Fig. 7a), 5Ni-PDC (Fig. 7b) and 10Ni-PDC (Fig. 7c) catalysts. The mean activation energy values calculated from linear fits of Ozawa plots shown in Fig. 8 are 133.2 kJ/mol, 133.7 kJ/mol, and 137.3 kJ/mol for PDC (Fig. 8a), 5Ni-PDC (Fig. 8b) and 10Ni-PDC (Fig. 8c) samples. The Ozawa plots have parallel lines for each percentage conversion, indicating a single-step reaction. The activation energy values obtained in the present study are lower than that of non-catalyzed soot oxidation reaction (168 kJ/mol) (Neeft et al. 1997). However, the values of all analyzed catalysts are very close, indicating similar nature and reactivity of active oxygen species facilitated by the synthesized catalysts (Russo et al. 2005; Shimizu et al. 2010). The activation energies of various catalysts for soot oxidation activity is showed in Table 2. The activation energy of 10Ni-PDC (137 kJ/mol) is in a similar range than compared to other catalysts (under tight contact mode) such as Ni/CeO2 (Bendieb Aberkane et al. 2019), 20Mn2O3/CeO2 (Jampaiah et al. 2019). The present study shows that the Ni plays a promotional effect in ceria-based catalysts by improving the descriptors such as lattice strain and oxygen vacancies, which further enhances the soot oxidation activity.

Fig. 7
figure 7

Soot conversion versus temperature plots at 5, 10, 15, 20 °C/min heating rate for a PDC, b 5Ni-PDC, c 10Ni-PDC catalysts

Full size image
Fig. 8
figure 8

Ozawa plots for a PDC, b 5Ni-PDC, c 10Ni-PDC catalysts

Full size image

Conclusion

The addition of Ni to Ce0.9Pr0.1O2 oxide catalysts synthesized by solution combustion synthesis method showed a promotional effect on soot oxidation activity. It is shown that nickel doping in PDC positively affects the soot oxidation activity with optimized doping to be 10 mol% of Nickel. 10 mol% nickel doping in PDC catalyst reduced the T50 temperature from 374 °C for PDC to 360 °C for 10Ni-PDC, which has been attributed to increased lattice strain and oxygen vacancies. From XRD and SAED patterns, it is confirmed that up to 10 mol% Ni addition in PDC catalysts resulted in the formation of solid-solution and thus increased the oxygen vacancies, which is observed from Raman spectroscopy. For the 10Ni-PDC catalyst, from soot oxidation kinetic studies, the activation energy obtained from Ozawa plots is 137 kJ/mol.