Limited Ce doped Ni–Co as a highly efficient catalyst for H₂-SCR of NO with good resistance to SO₂ and H₂O at low temperature

Abstract

The Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts doped with limited Ce were fabricated using a sol–gel method and employed to investigate the SO₂-resistance in the selective catalytic reduction of NO by hydrogen (H₂-SCR). The results indicated that the inclusion of a limited amount of cerium not only effectively enhanced the catalytic activity of Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts but also improved their resistance to SO₂. Notably, when the Ce doping was 0.07 at%, the catalyst showed high sulfur resistance and exhibited a broader temperature window of 140–350 °C with a maximum NO_X conversion of 98% at 190 °C. Also, addition of 5% H₂O into the feed gas slightly decreased the deNO_X performance of all catalysts, the NO_X conversion nearly recovered to the initial level with H₂O withdrawal, and the resistance to H₂O also increased gradually in correspondence to the increase in Ce doping. Meanwhile, an increase of GHSV from 4800 to 9600 mL h⁻¹ g⁻¹, the NO/H₂ ratios from 1 : 10 to 1 : 1 and the O₂ content from 0% to 6%, gradually decreased the removal efficiency of NO. The XPD and XRS analysis depicted fine cubic spinel-type structures and Ce bearing two chemical valence states doped in the spinels. In the H₂-TPR and NH₃-TPD analysis, the Ce doping resulted in the temperature shift of reductive peaks and weak acid sites to lower temperature and the increase of strong acid sites. These results enhanced the redox property and led to a positive effect on SCR reaction, which were consistent with the SCR activity data.

---

1. Introduction

Nitrogen oxides (NO_X) from fossil fuel combustion¹ are major air pollutants, which have serious environmental implications,² such as causing photochemical smog, acid rain, and greenhouse effects.³ The current efficient processing technology for NO_X emission control in industry is NH₃-SCR, however, many problems persist in NH₃-SCR, namely catalyst deterioration, NH₃-slip, ash odor and so on.⁴ To solve this issue, selective catalytic reduction by hydrogen (H₂-SCR) could greatly replace the NH₃-SCR, which is due to the efficient removal of NO_X without the emergence of any secondary pollutants at the lowest possible temperature.^5,6

Moreover, the noble metal catalysts as well as the precious metal doped catalysts in the process of H₂-SCR exhibit good catalytic activity for NO_X removal at relatively low temperatures.^7–10 For example, higher than 95% NO conversion with 78–92% N₂ selectivity at 100–400 °C was studied at Pt/MgO–CeO₂ catalyst,⁸ and G. L. Chiarello et al.¹⁰ researched the Pd/SiO₂ catalyst with 90% NO_X conversion at 170–300 °C. Nevertheless, noble metals catalysts have poor resistance to sulfur dioxide, and that lead to the formation of formed the sulfates and sulfites on the active sites of catalysts. Eventually such species results in the deactivation of SCR activity and reduces the NO removal at low temperature.^11,12 Such as, the activities of Pt/SiO₂ and Pd/SiO₂ were greatly influenced by 20 ppm SO₂ in the simulated gas stream.¹³ Thus, improving sulfur tolerance of the catalyst should be salient for practical usage.

For this problem, recently, we have used the limited cerium doping into Ni–Co spinel catalyst to improve the SO₂-resistance, the results exhibits that a high removal efficiency of NO_X along with excellent SO₂ resistance is achieved through the H₂-SCR in the presence of oxygen at relatively lower temperature. The good performance is due to the cerium particular properties, revealing two stable oxidation states of Ce⁴⁺ and Ce³⁺, that results in the reserve and release of oxygen into cerium by shifting between Ce⁴⁺ and Ce³⁺ under oxidizing and reducing conditions.^14,15 In our study, we also focus on the effects of GHSV, NO/H₂ ratios and O₂ content on the SCR activity. Moreover, the durability and tolerance of catalysts to H₂O is also investigated for practical consideration.

2. Experimental

2.1. Catalyst preparation

The catalysts were prepared using sol–gel method with inorganic salts, where the molar ratio of citric acid : Ni(NO₃)₃·6H₂O : Ce(NO₃)₃·6H₂O : Co(NO₃)₃·6H₂O : PdCl₂ was 3 : 1 − x : x : 1.95 : 0.05. Then these compounds were dissolved into deionized water, further addition of the ammonia solution was to adjust the pH of the mixture solution at 5–6, during this process, the polyethylene glycol (PEG) 400 was added to make the high specific surface area and uniform particle size.¹⁶ Then the mixture was stirred for 3 h at room temperature, and evaporated redundant water at 80 °C to form a wet gel, later on the gel was dried at 130 °C. Finally, after grinding the obtained material into fine powder, the powder was calcinated in muffle furnace at 400 °C for 4 h to obtain crystallized Ni_1−xCe_xCo_1.95Pd_0.05O₄.

2.2. Catalyst characterization

The samples X-ray diffraction (XRD) measurements have been conducted by using a Rigaku D/MAX 2550 diffractometer (Japan) with Cu Kα radiation (λ = 1.54056 Å) operating at 40 kV and 100 mA. The elemental composition of the samples was characterized by energy dispersive spectrometer (EDS) using a Falion 60s spectrometer. The morphological features of samples were studied by Scanning Electron Microscopy (SEM) using Nova NanoSEM 450.

The Brunauer–Emmett–Teller (BET) surface area of the catalysts were measured with a nitrogen adsorption instrument (Micromeritics, Tristar II 3020) by using N₂ adsorption/desorption at liquid nitrogen temperature. Prior to BET analysis, the samples were degassed at 300 °C for 3 h and cooled to room temperature.

The H₂ Temperature Programmed Reduction (H₂-TPR) analysis were conducted on Auto Chem II 2920 equipped with a TCD detector. In a typical TPR experiment, about 50 mg sample was pretreated in a He stream at 300 °C for 1 h and cooled to room temperature before the H₂-TPR analysis. Then, the H₂-TPR runs were performed from room temperature to 800 °C at a heating rate of 10 °C min⁻¹ in a 10% H₂/Ar mixture with a flow rate of 40 mL min⁻¹.

Temperature Programmed Desorption of ammonia (NH₃-TPD) was performed on Auto Chem II 2920 analyzer. For NH₃-TPD experiments, 50 mg of the oven-dried sample was treated under helium flow at 300 °C for 60 min to remove the adsorbed H₂O and cooled to room temperature by other gases. Then, the saturation of NH₃ adsorption was achieved with 10% NH₃/He mixture at 50 °C for 60 min with a total flow rate of 40 mL min⁻¹. After NH₃ adsorption, the samples were purged by He (40 mL min⁻¹) for another 60 min. Finally, the NH₃-TPD was performed from room temperature to 800 °C at a heating rate of 10 °C min⁻¹ under a constant flow of He (50 mL min⁻¹). The NH₃ desorption was monitored by a Thermal Conductivity Detector (TCD).

X-ray Photoelectron Spectroscopy (XPS) were performed in a Quantum 2000 Scanning ESCA Microprobe (Physical Electronics) using a focused monochromatic Al Kα X-ray (1486.7 eV) source to analyze the surface atomic concentration and chemical states of the catalysts, the X-ray source operated at 100 W and 100 μm-diameter beam. The binding energy scale was calibrated by the carbon C 1s at 284.6 eV as known standards. The de-convolution of XPS peak was carried out using Casa XPS program.

The Raman spectroscopy experiments were conducted using an Iuvia microscope instrument (Iuvia Reflerx) with a wavelength of 514.5 nm.

2.3. Catalytic activity test

The activity measurements of the catalysts were carried out in a fixed-bed flow micro-reactor at 50–350 °C with powder catalyst of 1 g, the thermocouple controlled the catalyst bed temperature through interpolating into the fixed-bed reactor, and the fixed-bed reactor was heated with an artificial intelligence temperature controller of type Al-518/518P. The reactant gas typically consisted of 1071 ppm NO, 1071–10710 ppm H₂, 0–6% O₂, 5% H₂O (when used), 500 ppm SO₂ (when used), and while N₂ of high purity (99%) was used as a carrier gas. While the water vapor was obtained by passing N₂ through a heated gas-wash bottle including deionized water. Moreover, the total gas flow rate of the inlet gas was always maintained at 155 mL min⁻¹ with a GHSV of 9600 mL h⁻¹ g⁻¹, except the changes of GHSV. The gas effluent stream from the reactor was monitored by a NO–NO₂–NO_X Analyzer (Thermo Scientific, Model 42i), and the nitric oxide conversion was defined by using the following equation,where, X_NOX represented the NO_X conversion, [NO_X]_inlet and [NO_X]_outlet denoted the NO_X conversion at inlet and outlet of the reactor, respectively.

3. Results and discussion

3.1. Catalysts characterization

3.1.1. XRD analysis. In order to explore further the Ce doping effect and the arousal of possible structural changes in catalyst, the XRD patterns of Ni_1−xCe_xCo_1.95Pd_0.05O₄ were conducted. As shown in Fig. 1a, the characteristic diffraction peaks for NiCo_1.95Pd_0.05O₄ were observed over Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts, and the crystal planes (111), (220), (311), (222), (400), (422), (511) and (440) were comprised in the main diffraction peaks, which indicated that these spinels were cubic spinel-type structures and exhibit good crystallinity. While the diffraction angle theta of the strongest peak reflection at (311) slightly shifted to a lower angle due to increased amount of Ce doping. The greater amount of doping corresponded to larger shift in angle as illustrated in Fig. 1b. The main reason was the substitution of the Ni ions by the heavier and larger Ce ions in the structure of Ni_1−xCe_xCo_1.95Pd_0.05O₄ spinels, resulting in an angle shift of the intense peak (311) towards lower diffraction angle.¹⁷ When the Ce doping was 0.1 at%, the diffraction peaks at ca. 2θ = 28.7° can be indexed to (211) crystal planes of CeO₂ (JCPDS), indicating that the amount of doped Ce was so more that fraction of Ce dopant formed CeO₂. Moreover, as the Ce loading increased to 0.2 at%, the diffraction peak at (111) was very weak, it may be a significant change in the crystallinity of Ni_0.8Ce_0.2Co_1.95Pd_0.05O₄ caused by the redundant cerium. This resulted in the loss of SCR activity compared to the Ni_0.93Ce_0.07Co_1.95Pd_0.05O₄ catalyst.

Fig. 1 (a) XRD patterns of Ni_1−xCe_xCo_1.95Pd_0.05O₄ used in this study; (b) XRD peaks positions.

3.1.2. SEM, BET and EDS analysis. The morphological features of the catalyst samples prepared, were studied by SEM. Results obtained by SEM analysis are illustrated in Fig. 2. As it can be seen, the samples are particle shape, but many particles appear an agglomeration phenomenon. While the sample with 0.07 at% Ce exhibited a relatively good particle shape, when the Ce amount increased to 0.2 at%, the particle aggregation was intense, that could be attributed to the formation of CeO₂, which was confirmed with the use of XRD. The BET surface areas of the studied Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts are summarized in Table 1. The surface area of NiCo_1.95Pd_0.05O₄ was found to be 12.7 m² g⁻¹, while the 0.03 at% Ce doped catalyst, the BET surface area was just 6.7 m² g⁻¹. Table 1 showed an ascending trend of direct relation between dopant amount and catalysts surface area except the 0.07 at% Ce, which was 16.1 m² g⁻¹ larger than 0.1 at% Ce. When the amount increased to 0.2 at%, the BET surface area was 21.1 m² g⁻¹, which was largest calculated in this study. In order to further investigate the elemental composition of the samples, the EDX analysis was conducted, and the results obtained are presented on Table 1. It was found that the Co/Ni, Ce/Ni and Pd/Ni atomic ratios (Ni was selected for comparison reasons) of all samples were consistent with the atomic percentages of Ni_1−xCe_xCo_1.95Pd_0.05O₄ (0 ≤ x ≤ 0.2).

Fig. 2 SEM patterns of Ni_1−xCe_xCo_1.95Pd_0.05O₄ used in this study; (a) 0.03 at% Ce; (b) 0.07 at% Ce; (c) 0.1 at% Ce; (d) 0.2 at% Ce.

Table 1 Physical and chemical properties of the studied catalysts

Catalysts	SBET (m^2 g^−1)	Elemental composition (at%)
		Ni	Co	Pd	Ce	Pd/Ni	Co/Ni	Ce/Ni
NiCo1.95Pd0.05O4	12.1	20.29	38.97	1.00	0.00	0.049	1.92	0.000
Ni0.97Ce0.03Co1.95Pd0.05O4	6.7	21.73	39.90	1.43	0.94	0.066	1.84	0.043
Ni0.95Ce0.05Co1.95Pd0.05O4	7.4	20.93	41.03	1.41	1.86	0.067	1.96	0.089
Ni0.93Ce0.07Co1.95Pd0.05O4	16.1	19.71	39.85	1.53	2.13	0.078	2.02	0.108
Ni0.9Ce0.1Co1.95Pd0.05O4	12.7	19.18	38.56	1.04	2.43	0.054	2.01	0.126
Ni0.8Ce0.2Co1.95Pd0.05O4	21.1	15.63	36.02	1.14	4.35	0.079	2.31	0.278

---

3.1.3. Raman analysis. The Raman was conducted to further investigate the structural properties of Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts. As shown in Fig. 3, the peaks obtained at 183.57, 464.76, 505.47 and 652.86 cm⁻¹ correspond to E_2g, E_g, F_2g, and A_1g models of NiCo_1.95Pd_0.05O₄, respectively.^18,19 A broad band at 400–700 cm⁻¹ was detected, which was due to the vibrations of Ni–O in the octahedral sites and Co–O, Pd–O in the tetrahedral positions, these were attributed to the inverse spinel structure of NiCo₂O₄.²⁰ The vibration of Ce–O due to Ce doping in the catalyst, broadened the strong band at 450–650 cm⁻¹. As the amount of Ce doping increased, the intensity increased. Especially when 0.07 at% was added to Ce, the intensity was maximum, which was consistent with the higher catalytic activity of Ni_1−xCe_xCo_1.95Pd_0.05O₄. However, when the doping of At was increased further at 0.2 wt% on Ce, the Raman spectra became weak, probably due to the change in the crystallinity of Ni_0.8Ce_0.2Co_1.95Pd_0.05O₄, that was consistent with the XRD results.

Fig. 3 Raman profiles of Ni_1−xCe_xCo_1.95Pd_0.05O₄ used in this study.

3.2. Catalysts performance

3.2.1. SCR activity of the Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts. The results of the NO_X conversion over the Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts are illustrated in Fig. 4. The reaction condition was at a GHSV of 9300 mL g⁻¹ h⁻¹ at the temperature range of 50–350 °C with a feed stream concentration of 1070 ppm NO, 10 700 ppm H₂, 2% O₂ and N₂ as balance gas. As shown in Fig. 4a and b, initially the NO_X conversion over the Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts increased to maximum and later decreased as the temperature increased from 50 to 350 °C. While the NO_X conversion over NiCo_1.95Pd_0.05O₄ catalyst was over 80% with the temperature window of 170–260 °C, and the maximum conversion obtained was found to be 93% at 230 °C. The previous research²¹ indicated the Ce doping could effectively improve the SCR performance. Hence, in the present study, the effect of different concentrations of doped Ce on the conversion for H₂-SCR over the Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts was examined. The results shown in Fig. 4a, exhibit distinctly increased SCR activities in correspondence to the increase of amount of doped Ce. When the Ce doping was 0.03 at%, the maximum NO_X conversion was approximately 95% at 240 °C, while the temperature window over 80% was broadened in the range of 170–310 °C, which is illustrated in Fig. 4b. As the concentration of doped cerium increases, the temperature window of NO_X conversion became more and more wide, and the temperature of the highest NO conversion shifted to a lower temperature, except the Ce doping of 0.03 at% and 0.05 at%. When the Ce doping was 0.07 at%, the temperature window reached to the maximum of 140–350 °C, and the maximum NO_X conversion of 98% was achieved at 190 °C. Then the temperature window decreased but was still larger than the original value as upon the further increase in the amount of doped Ce. These results indicated that the Ce doping effectively enhanced the SCR activity of Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts, which was due to the contributive effect of cerium.²¹ The sequence of NO conversions for Ni_1−xCe_xCo_1.95Pd_0.05O₄ with different amount of doped cerium decreased as follow:

Fig. 4 (a and b) Effects of the amount of doped Ce amount on NO_X conversion for H₂-SCR over the Ni_1−xCe_xCo_1.95Pd_0.05O₄ (0 ≤ x ≤ 0.2) catalysts. (c–e) Effects of GHSV, NO/H₂ ratios and O₂ content on NO_X conversion for H₂-SCR over the Ni_0.93Ce_0.07Co_1.95Pd_0.05O₄ catalyst. Reaction condition: [H₂] = 10 700 ppm (when examined the Ce content and GHSV), [NO] = 1070 ppm, [O₂] = 0%, 2% [when examined the Ce amount, GHSV and NO/H₂ ratios], 4% and 6%, [N₂] = balance; catalyst mass = 1 g, GHSV = 9300 mL g⁻¹ h⁻¹ (when examined the Ce amount and NO/H₂ ratios), T = 50–350 °C. (a and b) The Ce amount; (c) GHSV; (d) NO/H₂ ratios; (e) O₂ content.

Ni_0.93Ce_0.07Co_1.95Pd_0.05O₄ > Ni_0.9Ce_0.1Co_1.95Pd_0.05O₄ > Ni_0.8Ce_0.2Co_1.95Pd_0.05O₄ > Ni_0.95Ce_0.05Co_1.95Pd_0.05O₄ > Ni_0.97Ce_0.03Co_1.95Pd_0.05O₄ > NiCo_1.95Pd_0.05O₄.

Among them, the Ni_0.93Ce_0.07Co_1.95Pd_0.05O₄ catalyst exhibited high deNO_X performance with the maximum NO_X conversion of 98% at a relatively low temperature and wide temperature window of 140–350 °C.

3.2.2. Effect of GHSV, NO/H₂ ratios and O₂ content. Previous studies³ have reported that gas hourly space velocity (GHSV) could significantly influence the deNO_X performance. However, H₂ as a reducing agent could efficiently reduce NO_X without secondary pollution,²² so the H₂ concentration should not be ignored in SCR reaction. Moreover, experiments carried out in this study, Ni_0.93Ce_0.07Co_1.95Pd_0.05O₄ catalyst has good SCR activity and wide temperature window. Therefore, the NO_X conversions of Ni_0.03Ce_0.07Co_1.95Pd_0.05O₄ at different GHSVs from 4800 to 9300 mL h⁻¹ g⁻¹, NO/H₂ ratios from 1 : 1 to 1 : 10 and O₂ content from 0% to 6% were investigated at 50–350 °C range, and the results were shown in Fig. 4c–e.

As illustrated in Fig. 4c, the NO_X conversion at a low GHSV was higher than that at a high GHSV. The NO_X conversion first increased and remained over 90%, then finally declined as the temperature increased further. In the temperature range of 150–270 °C, Ni_0.93Ce_0.07Co_1.95Pd_0.05O₄ exhibited high NO_X conversion at GHSVs of 4800–9300 mL h⁻¹ g⁻¹, and the NO_X conversion faintly decreased upon the increase of GHSV. While, a high GHSV means requirement of the less catalyst that may subsequently reduce the cost of the SCR system. Hence, a high GHSV of 9300 mL h⁻¹ g⁻¹ was used in our work.

However, H₂ is a strong reducing agent and its concentration could significantly influence the SCR activity, so the effect of NO/H₂ ratios on Ni_0.93Ce_0.07Co_1.95Pd_0.05O₄ catalyst was studied, which was shown in Fig. 4d. The NO_X conversion first clung to the maximum around 200 °C, and then decreased as the temperature increased further at different NO/H₂ ratios. As the NO/H₂ ratios descended from 1 : 1 to 1 : 10, the H₂ concentration increased, resulting in the improvement of deNO_X performance, which showed that the maximum NO_X conversion enhanced from 60% to 96%. While the Ni_0.93Ce_0.07Co_1.95Pd_0.05O₄ catalyst exhibited high catalytic activity at low NO/H₂ ratio, and in industry, the lower NO/H₂ ratio is usually used to obtain a relatively higher NO_X conversion. Therefore, the Ce doping significantly improved the catalytic activity and economic performance.

Previous study has showed that the content of O₂ was a crucial parameter for the H₂-SCR reaction.^23,24 Generally, the oxygen content is 4–6% in the flue gas from circulating fluidized bed boiler.²⁵ Hence, in the present study the oxygen content was max 6%. Fig. 4e showed the O₂ influence on Ni_0.93Ce_0.07Co_1.95Pd_0.05O₄ catalyst, in the absence of O₂, the NO_X conversion ascended to the maximum of 100% and then remained stable at the high efficiency within a temperature range of 150–350 °C. When the O₂ content increased from 0% to 6%, the removal efficiency of NO_X gradually decreased from 100% to 75% at 200 °C, which was due to the consumption of the hydrogen reductant by oxygen. That was in consistence to the results of varying NO/H₂ ratios effects. When the O₂ content was 6%, the NO_X conversion still maintained higher removal level of over 75% at 170–250 °C range, which indicated that the addition of cerium efficiently enhanced the oxygen resistance of catalyst.

3.2.3. Effects of the presence of SO₂ and H₂O. Generally, the flue gas usually contains a certain amount of SO₂ and H₂O, and the former has poisoning effect on the SCR catalysts, which resulted in the reduction of NO_X removal due to the formation of sulfates and sulfites on the active sites of catalysts.^26,27 Cooperatively, the water vapor also influences the SCR activity, hence, the sulfur poisoning and H₂O are important factors that should not be ignored. The effects of SO₂ and H₂O on the NO_X conversion for the H₂-SCR over the Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts were displayed in Fig. 5. It was seen that the initial NO_X conversion was about 90% over NiCo_1.95Pd_0.05O₄ catalyst and kept steady for 60 min in absence of SO₂. When adding SO₂ to the feed gas, the NO_X conversion rapidly declined to 50% and kept steady for 150 min, while the SCR activity only recovered to 67.8% when the SO₂ was off, implying that the sulfur poisoning badly influenced the catalytic activity of NiCo_1.95Pd_0.05O₄ catalyst.

Fig. 5 Effects of SO₂ ([SO₂] = 500 ppm) and H₂O ([H₂O] = 5%) on NO_X conversion for H₂-SCR over the Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts. Reaction condition: [NO] = 930 ppm (950 ppm, when SO₂ was added), [H₂] = 9300 ppm (9500 ppm, when SO₂ was added), [O₂] = 2%, [N₂] = balance; catalyst mass = 1 g, GHSV = 9300 mL g⁻¹ h⁻¹, flow rate = 155 mL min⁻¹, T = 230 °C. (a) [SO₂] = 500 ppm; (b) [H₂O] = 5%.

Previous studies^21,28 have demonstrated that the doped cerium could improve the Mn–Ce/TiO₂ catalytic activity and resistance to SO₂. In the experiments performed in the study at hand, the main task was to adjust the amount of doped cerium to enhance the catalyst performance. As shown in Fig. 5a, the doped Ce Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts with doped Ce exhibited high NO_X conversion (92–97%) in 60 min, after addition of 500 ppm SO₂ to the gas feed stream, although there was obvious decrease in the deNO_X performance, yet the NO_X conversion showed comparatively a higher value of 61–72% than of 50% on NiCo_1.95Pd_0.05O₄ catalyst, which indicated the doped Ce catalysts exhibited less sensitivity to SO₂. In addition, as the Ce doping increased, the SO₂-resistance of Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts also increased, which was confirmed by the increase of NO_X conversion from 60.8% to 71.6% at 150 min in the presence of 500 ppm SO₂ with the doped Ce increased from 0.03 to 0.2. After the removal of SO₂, the NO conversion of Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts was greatly recovered, and it was also contributed to the successive increase in the amount of doped Ce, especially when the doped Ce increased to 0.07, 0.1 and 0.2, the NO_X conversion nearly recovered to initial value. Therefore, the addition of cerium weak sulfation on the catalyst surface, it was probable that the SO₂ preferentially reacted with cerium instead of active component, which inhibited the formation of sulfate on active sites of catalysts, it was consistent with the literature.²⁹ Thus these factors greatly improve the sulfur-resistance of Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts. Moreover, the Ni_0.93Ce_0.07Co_1.95Pd_0.05O₄ catalyst could quickly recover to the initial value whilst exhibiting high NO_X conversion of 95%.

The effect of H₂O on the selective catalytic reduction of NO_X with hydrogen over the Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts is demonstrated in Fig. 5b. It was found that the initial NO_X conversions of the catalysts were about 91–95% and stabilized for 60 min in absence of H₂O. After the addition of 5% H₂O in the gas feed stream, the deNO_X performance of all catalysts showed minimal decrease for 200 min, which was due to H₂O competitive adsorption with NO as well as with H₂,^30,31 resulting in the desorption of NO originally adsorbed on the catalyst.³² The NO_X conversion of the NiCo_1.95Pd_0.05O₄ couldn't recover to the initial level, that was attributed to the reduction in the number of the actives sites which was available for the adsorption and reaction of NO and H₂.³³ However, the catalyst with the addition of cerium showed better H₂O-resistance, the NO_X conversion rate continuously increased to almost the original level. As the Ce doped, the hydrogen adsorption of catalysts was promoted, which resulted in the decrease of H₂O adsorption, and increased the resistance to H₂O.

Therefore, the results manifested that the Ce doping could efficiently enhance the resistance to H₂O and SO₂, and among all, the Ni_0.93Ce_0.07Co_1.95Pd_0.05O₄ catalyst displayed good characteristics against the resistive of H₂O and SO₂ in the low-temperature SCR reaction.

3.3. Catalysts activity analysis

In order to study the reason for the improvement of catalytic activity, the H₂-TPR, NH₃-TPD analysis and XPS analysis were conducted, and the results were shown as follows.

The redox properties and acidic properties of catalysts are an indicator to determine the catalytic activity, thus the Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts were demonstrated using H₂-TPR and NH₃-TPD analysis, respectively, and the results are exhibited in Fig. 6. As shown in Fig. 6a, the profile of NiCo_1.95Pd_0.05O₄ catalyst had a sharp reduction peak at 128 and a broad peak at 352 °C. According to the literature,^34,35 the reduction peak at 128 °C was assigned to the reduction of both Pd²⁺ and Co³⁺, which was the form of Pd²⁺ to Pd⁰ and Co³⁺ to Co²⁺, respectively.^36,37 While the broad peak appeared at 352 °C, corresponding to the co-reduction of Co²⁺ to Co⁰ and Ni²⁺ to Ni⁰.^36,37 However, when the Ce doped in catalyst, the temperature of the reduction peaks shifted to low temperature, which was due to the interaction between Pd²⁺ and Ce⁴⁺.³⁸ The peak around 100 °C corresponded to the co-reduction of Pd²⁺, Co³⁺, Ce⁴⁺ and Ce³⁺, while the peak around 252 °C was assigned to the reduction of both Co²⁺ and Ni²⁺. Upon the increase of Ce doping content, the temperature of peaks shifted greatly to low temperature, which was advantageous to the hydrogen adsorption, resulted in the enhancement of the redox property of Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts, hence improving the SCR activity.

Fig. 6 H₂-TPR and NH₃-TPD profiles of Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts used in this study. (a) H₂-TPR; (b) NH₃-TPD.

Temperature programmed desorption of ammonia (NH₃-TPD) has been employed to investigate the acidic properties of all catalysts, and the results obtained are shown in Fig. 6b. All catalysts displayed one broad NH₃ desorption peak spanned in the temperature range of 100–350 °C, and the desorption of NH₃ at different temperature regions was determined by the strength of acid sites and the amount of acid, while the peak was corresponded to the weak and medium strength of acid sites.^15,39 For the NiCo_1.95Pd_0.05O₄ catalyst, a shoulder over 600 °C was observed, and no peak formation may be in result of the temperature of NH₃-TPD not exceeding 700 °C. However, when the Ce was doped in the catalyst, the temperature of weak acid sites shifted to low temperature, which eventually further weakened the strength of weak acid, and the result was facilitating to NO stripping and beneficial to SCR reaction. Meanwhile, one broad peak from 550 °C to 700 °C was observed, that was assigned to the chemisorbed NH₃ molecules adsorbed by strong acid sites.⁴⁰ As the increased of Ce doping, the intensity and area of the peak centred at 650 °C was enlarged, which resulted in a positive effect on SCR reaction. Therefore, the Ce doping could improve the SCR activity and thus widened the NO_X conversion temperature window,²⁷ which was in good agreement with the SCR activity of the Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts (Fig. 4).

In order to further analyze the Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts, in this study, the Ce content of 0 at%, 0.03 at%, 0.07 at% and 0.1 at% were selected to investigate the surface characteristics. Fig. 7 shows the surface chemical sate of Ni, Co, Pd and Ce in the Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts, which was conducted by XPS analysis. The Ni 2p spectra is given in Fig. 7a, all samples showed two spin–orbit doublets characteristic of Ni²⁺ and Ni³⁺ with two shakeup satellites. The Ni²⁺ peaks could be found at the binding energy of 854.4 eV and 871.6 eV, while the fitting peaks with an energy of 856.43 eV and 873.8 eV corresponded to Ni³⁺, respectively.^18,41

Fig. 7 XPS spectra of Ni_1−xCe_xCo_1.95Pd_0.05O₄ used in this study. (a) Ni 2p, (b) Co 2p, (c) Pd 3d, (d) Ce 3d.

Fig. 7b demonstrated the Co 2p spectra with two types of species of 2p_3/2 and 2p_1/2, the peaks at 779.4 eV and 794.8 eV were from Co³⁺, while the binding energy at 781.4 eV and 796.7 eV were associated with Co²⁺, respectively,^42–44 which indicated that the low spin Co³⁺ species were dominantly positioned in the octahedral sites than the Co²⁺ species.^45,46

The Pd 3d XPS pattern indicated the fitting peaks at the binding energy of 337.1 eV and 342.5 eV assigned to 3d_5/2 and 3d_3/2, respectively.⁴⁷ The Pd_5/2 binding energy of 337.1 eV was in between metallic Pd and PdO, which demonstrated the Pd doping stably dissolve into the spinel crystal lattice.⁴⁸

The Ce 3d XPS spectra was illustrated in Fig. 7d, where v and u represented the spin–orbit coupling of 3d_5/2 and 3d_3/2, respectively. The binding energy peaks of v, v′′, v′′′, u, u′′ and u′′′, locating at 881.9 eV, 888.7 eV, 897.7 eV, 900.32 eV, 907.6 eV and 916.5 eV, respectively, corresponded to Ce⁴⁺. And the peaks v⁰, v′ and u′ at the binding energy of 879.3 eV, 884.8 eV and 900.32 eV were assigned to Ce³⁺.⁴⁹ While the Ce⁴⁺ peaks were more over that of Ce³⁺, it indicated that the Ce co-existed as Ce⁴⁺ and Ce³⁺ and the Ce⁴⁺ was dominant in the Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts.⁵⁰ Meanwhile, the low-temperature deNO_X performance was efficiently enhanced, which was due to the possibility of oxygen storage and release by the coexistence of Ce⁴⁺ and Ce³⁺ in the catalysts, and that was consistent with the results of SCR activity test of Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts. A shift was observed in Ni, Co and Pd XPS spectra with an addition of doping Ce, which was due to the big electronegativity of cerium, it resulted in a shift to lower bingding energy of Ni 2p, Co 2p and Pd 3d.

4. Conclusions

The nickel cobaltite spinels doped with cerium were investigated for the SO₂-resistence in the process of NO selective reduction by hydrogen. It is confirmed that the addition of a small amount (0.03–0.07 at%) of cerium not only effectively enhances the catalytic activity of Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts but also improves their resistance towards SO₂. On introducing 500 ppm SO₂ to the flue gas, the deNO_X performance results in a significant decrease but still shows a higher value of 61–72% than that of NiCo_1.95Pd_0.05O₄ catalyst. Further rising Ce inclusion in the catalyst increased the SO₂-resistance and in varying dopant amount also exhibited a linear relation towards SO₂-resistance. When the SO₂ was removed, the NO_X conversion of Ni_1−xCe_xCo_1.95Pd_0.05O₄ catalysts greatly recovered, especially when the Ce doping was 0.07 at%, the NO_X conversion quickly recovered to the initial value, and exhibited a broad temperature window of 140–350 °C with a maximum NO_X conversion of 98% at 190 °C. Meanwhile, addition of 5% H₂O to inlet gas, slightly decreases the deNO_X performance of all catalysts for 200 min. The NO_X conversion nearly recovered to the initial level upon the withdrawal of H₂O, among of all, the Ni_0.93Ce_0.07Co_1.95Pd_0.05O₄ catalyst showed better H₂O-resistance.

The effect of GHSV, NO/H₂ ratios and O₂ content on the NO_X conversion over the Ni_0.93Ce_0.07Co_1.95Pd_0.05O₄ catalysts were also investigated. As the increase of GHSV from 4800 to 9600 mL h⁻¹ g⁻¹, the NO_X conversion faintly decreased from 98% to 96% in the temperature range of 150–270 °C. A decrease of NO/H₂ ratios from 1 : 1 to 1 : 10, resulted in the increase of deNO_X performance, and showed the maximum NO_X conversion increase from 60% to 96%. As the O₂ content increased from 0% to 6%, the removal efficiency of NO_X gradually decreased from 100% to 75% at 200 °C, although the O₂ content reached up to 6%, the NO_X conversion still maintained high removal level of over 75% at 170–250 °C range, these results indicated that the addition of cerium efficiently enhanced oxygen resistance of the catalyst.

The XPD and XRS analysis showed the good cubic spinel-type structures, and the Ce doped in the spinels with two chemical valence states of Ce³⁺ and Ce⁴⁺, that was conducive to oxygen store and release, enhancing the SCR activity. In the H₂-TPR and NH₃-TPD analysis, when the Ce was doped in catalyst, resulted in the shift of reductive temperature to low temperature, that enhanced the redox property; weakened the strength of weak acid, which was facilitating to NO stripping and beneficial to SCR reaction; enhancing the strong acid sites, added positive effect on SCR reaction. These results were consistent with the SCR activity data.

Acknowledgements

This research is based on work supported by the National Natural Science Foundation of China (21277045). Thanks to Zhenhua Zhou for providing solutions to the experimental process and writing assistance.

References

1. C. N. Costa and A. M. Efstathiou, J. Phys. Chem. C, 2007, 111, 3010–3020.
2. A. Fritz and V. Pitchon, Appl. Catal., B, 1997, 13, 1–25.
3. G. Qi, R. T. Yang and F. C. Rinaldi, J. Catal., 2006, 237, 381–392.
4. G. G. Olympiou and A. M. Efstathiou, Chem. Eng. J., 2011, 170, 424–432.
5. M. Machida, S. Ikeda, D. Kurogi and T. Kijima, Appl. Catal., B, 2001, 35, 107–116.
6. S. S. Kim and S. C. Hong, J. Ind. Eng. Chem., 2010, 16, 992–996.
7. F. J. P. Schott, P. Balle, J. Adler and S. Kureti, Appl. Catal., B, 2009, 87, 18–29.
8. C. N. Costa and A. M. Efstathiou, Appl. Catal., B, 2007, 72, 240–252.
9. Y. U. Qing, F. Kong, L. I. Landong and W. U. Guangjun, Chin. J. Catal., 2010, 31, 261–263.
10. G. L. Chiarello, D. Ferri, J. D. Grunwaldt, L. Forni and A. Baiker, J. Catal., 2007, 252, 137–147.
11. B. Q. Jiang, Z. B. Wu, Y. Liu, S. C. Lee and W. K. Ho, J. Phys. Chem. C, 2010, 114, 4961–4965.
12. H. Chang, X. Chen, J. Li, L. Ma, C. Wang, C. Liu, J. W. Schwank and J. Hao, Environ. Sci. Technol., 2013, 47, 5294–5301.
13. T. Yoshinari, K. Sato, M. Haneda, Y. Kintaichi and H. Hamada, Appl. Catal., B, 2003, 41, 157–169.
14. X. Li, S. Zhang, Y. Jia, X. Liu and Q. Zhong, J. Nat. Gas Chem., 2012, 21, 17–24.
15. Z. Wu, R. Jin, Y. Liu and H. Wang, Catal. Commun., 2008, 9, 2217–2220.
16. J. Fan and W. Huang, Chin. J. Catal., 2011, 32, 139–143.
17. H. Kavas, A. Baykal, M. S. Toprak, Y. Köseoğlu, M. Sertkol and B. Aktaş, J. Alloys Compd., 2009, 479, 49–55.
18. Z. Yu, H. Li, X. Zhang, N. Liu, W. Tan, X. Zhang and L. Zhang, Biosens. Bioelectron., 2016, 75, 161–165.
19. Z. Q. Liu, K. Xiao, Q. Z. Xu, N. Li, Y. Z. Su, H. J. Wang and S. Chen, RSC Adv., 2013, 3, 4372–4380.
20. M. N. Iliev, P. Silwal, B. Loukya, R. Datta, D. H. Kim, N. D. Todorov, N. Pachauri and A. Gupta, J. Appl. Phys., 2013, 114, 033514–033515.
21. S. Y. Jiang and R. X. Zhou, Fuel Process. Technol., 2015, 133, 220–226.
22. W. U. Peng, Y. Liu, F. Zhang, W. U. Guangjun, L. Baocheng and N. Guan, Chin. J. Catal., 2008, 29, 191–196.
23. L. Yuan, X. Zheng, K. Duan, H. Hu, J. Wang, S. I. Woo and Z. Liu, Front. Environ. Sci. Eng., 2013, 7, 457–463.
24. S. Yang, X. Wang, W. Chu, Z. Song and S. Zhao, Appl. Catal., B, 2011, 3, 380–385.
25. L. Huilin, Z. Guangbo, B. Rushan, C. Yongjin and D. Gidaspow, Fuel, 2000, 79, 165–172.
26. H. Chang, J. Li, J. Yuan, L. Chen, Y. Dai, H. Arandiyan, J. Xu and J. Hao, Catal. Today, 2013, 201, 139–144.
27. K. J. Lee, M. S. Maqbool, P. A. Kumar, K. H. Song and H. P. Ha, Catal. Lett., 2013, 143, 988–995.
28. S. Yun, T. Aikebaier, Q. Xie, S. Chen and H. Yu, Appl. Catal., B, 2014, 150–151, 630–635.
29. R. Jin, L. Yue, W. Yan, W. Cen, Z. Wu, H. Wang and X. Weng, Appl. Catal., B, 2014, 148–149, 582–588.
30. R. Burch and M. D. Coleman, Appl. Catal., B, 1999, 23, 115–121.
31. M. Leicht, F. J. P. Schott, M. Bruns and S. Kureti, Appl. Catal., B, 2012, 117–118, 275–282.
32. X. Lu, C. Song, S. Jia, Z. Tong, X. Tang and Y. Teng, Chem. Eng. J., 2015, 260, 776–784.
33. M. D. Amiridis, I. E. Wachs, G. Deo and J.-M. Jehng, J. Catal., 1996, 161, 247–253.
34. F. Bustamante, F. Córdoba, M. Yates and C. M. D. Correa, Appl. Catal., A, 2002, 234, 127–136.
35. J. M. Giraudon, A. Elhachimi, F. Wyrwalski, S. Siffert, A. Aboukaïs, J. F. Lamonier and G. Leclercq, Appl. Catal., B, 2007, 75, 157–166.
36. T. H. Lim, S. J. Cho, H. S. Yang, M. H. Engelhard and D. H. Kim, Appl. Catal., A, 2015, 505, 62–69.
37. Y. Gou, X. Liang and B. Chen, J. Alloys Compd., 2013, 574, 181–187.
38. H. Zhu, Z. Qin, W. Shan, W. Shen and J. Wang, J. Catal., 2004, 225, 267–277.
39. M. Imran, D. H. Kim, W. A. Al-Masry, A. Mahmood, A. Hassan, S. Haider and S. M. Ramay, Polym. Degrad. Stab., 2013, 98, 904–915.
40. Z. Huang, Z. Zhu, Z. Liu and Q. Liu, J. Catal., 2003, 214, 213–219.
41. J. H. Li, C. C. Wang, C. J. Huang, Y. F. Sun, W. Z. Weng and H. L. Wan, Appl. Catal., A, 2010, 382, 99–105.
42. Z.-Q. Liu, K. Xiao, Q.-Z. Xu, N. Li, Y.-Z. Su, H.-J. Wang and S. Chen, RSC Adv., 2013, 3, 4372.
43. J. Yang, M. Cho and Y. Lee, Biosens. Bioelectron., 2016, 75, 15–22.
44. X. Wang, W. Wen, J. Mi, X. Li and R. Wang, Appl. Catal., B, 2015, 176–177, 454–463.
45. J. G. Kim, D. L. Pugmire, D. Battaglia and M. A. Langell, Appl. Surf. Sci., 2000, 165, 70–84.
46. G. A. Babu, G. Ravi and Y. Hayakawa, International Journal of Science and Engineering Applications, 2014, 1, 17–20.
47. J. M. Giraudon, A. Elhachimi, F. Wyrwalski, S. Siffert, A. Aboukaïs, J. F. Lamonier and G. Leclercq, Appl. Catal., B, 2007, 75, 157–166.
48. G. C. M. Rodríguez and B. Saruhan, Appl. Catal., B, 2010, 93, 304–313.
49. S. Watanabe, X. Ma and C. Song, J. Phys. Chem. C, 2009, 113, 14249–14257.
50. B. Shen, Y. Wang, F. Wang and T. Liu, Chem. Eng. J., 2014, 236, 171–180.

---