NO x Removal by Selective Catalytic Reduction with Ammonia over a Hydrotalcite-Derived NiFe Mixed Oxide

: A series of NiFe mixed oxide catalysts were prepared via calcining hydrotalcite-like precursors for the selective catalytic reduction of nitrogen oxides (NO x ) with NH 3 (NH 3 -SCR). Multiple characterizations revealed that catalytic performance was highly dependent on the phase composition, which was vulnerable to the calcination temperature. The MO x phase (M = Ni or Fe) formed at a lower calcination temperature would induce more favorable contents of Fe 2+ and Ni 3+ and as a result contribute to the better redox capacity and low-temperature activity. In comparison, NiFe 2 O 4 phase emerged at a higher calcination temperature, which was expected to generate more Fe species on the surface and lead to a stable structure, better high-temperature activity, preferable SO 2 resistance, and catalytic stability. The optimum NiFe-500 catalyst incorporated the above virtues and afforded excellent denitration (DeNO x ) activity (over 85% NO x conversion with nearly 98% N 2 selectivity in the region of 210–360 ◦ C), superior SO 2 resistance, and catalytic stability. 2p 3/2 854.1, 855.9, Ni 2+ , Ni 3+ the Ni 3+ the of the NiFe-500 sample (75.5%) than of the NiFe-400 sample (72.5%), the content of the surface Fe 3+ species was opposite for the two catalysts, indicated that increasing the calcination temperature could result in a redox reaction between Ni 2+ and Fe 3+ to form Ni 3+ and Fe 2+ (Ni 2+ + Fe 3+ ↔ Ni 3+ + Fe 2+ ). However, the amount of surface Ni 3+ decreased to 60.1% with the increase in calcination temperature from 500 °C to 600 °C for the NiFe catalyst, which can be attributed to the formation of more high-crystallinity NiFe 2 O 4 spinels, so it was difficult for the reaction (Ni 2+ + Fe 3+ ↔ Ni 3+ + Fe 2+ ) to occur on the surface of the catalyst. Thus, comparatively, the redox reaction (Ni 2+ + Fe 3+ ↔ Ni 3+ + Fe 2+ ) may more easily occur for the NiFe-500 sample. Such interaction in the bimetallic oxide was similar to that previously reported [30,43] and it was considered to be favorable for the creation of a redox cycle and further promoted the catalytic activity, especially at low temperatures. temperature for the NiFe catalyst, be to the of more high-crystallinity spinels, for the reaction (Ni of the the redox reaction (Ni ) easily occur for the NiFe-500 sample. Such interaction in the bimetallic oxide was similar to that [30,43] and it was considered to be favorable for the creation of a redox cycle and promoted the catalytic activity, especially at low temperatures.


Introduction
Emission of nitrogen oxides (NO x ) is a main contributing factor to the induction of regional air pollution and ecosystem acidification [1]. During the past few decades, stringent regulations on NO x emissions have been issued worldwide and various measures have been devoted to diminish NO x , among which the removal of NO x from stationary sources by selective catalytic reduction of NO x with NH 3 (NH 3 -SCR) has been attracting significant attention due to the low cost and high efficiency [2][3][4]. The most adopted SCR catalysts in practical applications are V-based catalysts, while the toxicity of active vanadium species, together with high N 2 O formation at high temperatures, restricts their further application [5]. Therefore, for the current situation, developing environmentally friendly substitutive catalysts with excellent efficiency, strong stability, and high SO 2 resistance in NH 3 -SCR is imperative.
Recently, great efforts have been made to exploit novel NH 3 -SCR catalysts, and some transition metal oxide catalysts have received wide attention [6][7][8]. Notably, iron-based catalysts have been considered as a potential candidate for the NH 3 -SCR application due to their high reactivity, cost-effectiveness, superior SO 2 durability, and non-toxicity. However, pure iron catalysts are subject to to the limitations of a low acid amount [9] and inferior N 2 selectivity beyond 330 • C [10,11].
Up to now, a lot of reports have found that introducing other active ingredients (e.g., titanium oxides [12], manganese oxides [13], or tungsten oxides [2]) is an effective way to promote the denitration (DeNO x ) performance of iron-based catalysts. Especially, nickel species possess abundant surface Up to now, a lot of reports have found that introducing other active ingredients (e.g., titanium oxides [12], manganese oxides [13], or tungsten oxides [2]) is an effective way to promote the denitration (DeNOx) performance of iron-based catalysts. Especially, nickel species possess abundant surface acidity sites [14] and a favorable ability to increase the N2 selectivity of NH3-SCR reactions [15,16], which was expected to be introduced into the iron-based catalytic system [17]. Furthermore, nickel oxides are considered as valuable materials to induce a synergistic effect with other active species, such as NiMnOx [18,19] and NiTiOx [20], which are very sensitive to the synthesis process.
It was reported that layered double hydroxide (LDHs) materials are excellent precursors for the preparation of metal oxide catalysts with the desired chemical and phase composition [21]. Thermal decomposition of hydrotalcite-like materials could produce layered double oxides (LDOs) possessing a relatively high surface area, which could promote the catalytic performance.
Herein, a series of economical and eco-friendly NiFe mixed oxides catalysts were prepared via calcining a hydrotalcite-like compound at various temperatures. The NH3-SCR activity, SO2 resistance, and stability were evaluated for the different NiFe mixed oxides catalysts. The correlations among phase composition, structure, and catalytic performance were systematically explored by XRD (X-ray diffraction), FT-IR (Fourier Transform Infrared), TG (thermogravimetric analysis), N2 physisorption, NH3-TPD (NH3-Temperature Programmed Desorption), H2-TPR (H2-Temperature Programmed Reduction), and XPS (X-ray Photoelectron Spectroscopy). It was proved that the DeNOx performance was highly dependent on the phase composition of NiFe mixed oxides catalysts, which was available via adjusting the calcination temperature of NiFe-LDH. This work may serve as an important reference for LDH precursors to be tailored for potential applications in adjusting the phase component and catalytic performance.

Results and Discussion
NH3-SCR reactions over NiFe mixed oxides catalysts at various calcination temperatures (400 °C, 500 °C, and 600 °C) were performed during 150-450 °C and the results are given in Figure 1A. It is clear that the NiFe-400 sample showed rather low NH3-SCR efficiency with an obviously narrow operation temperature window. The DeNOx activity and temperature window for the NiFe-500 and NiFe-600 samples were promoted dramatically in comparison to that of the NiFe-400 catalyst. The NiFe-600 catalyst exhibited lower NOx conversion than the NiFe-500 catalyst at temperatures below 330 °C, while it behaved better in the higher temperature region, which implied that the higher calcination temperature was conducive to high-temperature SCR activity. As demonstrated in Figure 1B, all NiFe mixed oxides catalysts afforded superior N2 selectivity: nearly 98% below 360 °C. When temperatures exceeded 360 °C, the N2 selectivity of the NiFe-500 and NiFe-600 samples showed a slight decline compared with the NiFe-400 catalyst. Obviously, the NiFe-500 catalyst was the most promising candidate for NH3-SCR, affording above 85% DeNOx activity within the widest temperature region of 210-360 °C and superior N2 selectivity over the whole tested temperature range. The stability is an important indicator to estimate the properties of catalysts. The stability tests were performed at 300 °C for 50 h over each catalyst under the reaction conditions of 600 ppm NO, 600 ppm NH3, 5.0% O2, and balanced with N2. As presented in Figure 1C, the DeNOx activity of the NiFe-500 and NiFe-600 catalysts slightly fluctuated with only ± 1% variance, indicating fine stability. Comparatively, the NiFe-400 sample displayed the relatively worst stability. Such different stabilities can be ascribed to the different stabilities of the mixed oxides' structures. Thus, the NiFe-500 and NiFe-600 catalysts possessed a more stable structure in comparison with the NiFe-400 catalyst.
The presence of H2O and SO2 in the actual exhaust usually inhibits the NH3-SCR of NOx [22]. Therefore, it is of great importance to evaluate the H2O and/or SO2 tolerance under the reaction conditions. The H2O and/or SO2 tolerance of the serial NiFe mixed oxides catalysts was investigated at 300 °C under the reaction conditions of 600 ppm NO, 600 ppm NH3, 5% O2, 0 or 100 ppm SO2, 0 or 10.0% H2O, and balanced with N2. As depicted in Figure 1D, at the first stage, the NOx conversion for the NiFe-400, NiFe-500, and NiFe-600 catalysts was maintained at around 85%, 99%, and 96%, respectively. After adding 100 ppm SO2 into the reaction gas, the NOx conversion decreased slightly. However, the co-existence of 100 ppm SO2 and 10% H2O led to a gradual decrease to some extent in NOx conversion, where the decrease was only about 2% for the NiFe-500 and NiFe-600 samples while around a 5% decrease was displayed for NiFe-400. After purging off SO2 and H2O, the NOx conversion gradually recovered to the original values. Thus, the NiFe-500 and NiFe-600 catalysts possessed an excellent SO2 and H2O resistance ability in comparison to the NiFe-400 catalyst.
XRD is an essential technology to evaluate the structure of catalysts. The XRD pattern of the NiFe-LDH precursor is presented in Figure 2A The FT-IR spectrum of NiFe-LDH is illustrated in Figure 2B. An intense and board peak at 3429 cm -1 corresponded to the stretching vibration of the -OH group in the NiFe brucite-like layers, the interlayer water molecules, and the lattice water [23]. The peak that occurred at 2987 cm -1 can be assigned to a CO2-H2O bridging bond [24]. The vibration of the angular deformation of H2O molecules was observed at 1601 cm -1 [23]. The peaks at 1385 cm -1 belonged to the stretching vibrations of carbonate ions [25]. Additionally, the peak located at 676 cm -1 might be correlated to the vibration The stability is an important indicator to estimate the properties of catalysts. The stability tests were performed at 300 • C for 50 h over each catalyst under the reaction conditions of 600 ppm NO, 600 ppm NH 3 , 5.0% O 2 , and balanced with N 2 . As presented in Figure 1C, the DeNO x activity of the NiFe-500 and NiFe-600 catalysts slightly fluctuated with only ±1% variance, indicating fine stability. Comparatively, the NiFe-400 sample displayed the relatively worst stability. Such different stabilities can be ascribed to the different stabilities of the mixed oxides' structures. Thus, the NiFe-500 and NiFe-600 catalysts possessed a more stable structure in comparison with the NiFe-400 catalyst.
The presence of H 2 O and SO 2 in the actual exhaust usually inhibits the NH 3 -SCR of NO x [22]. Therefore, it is of great importance to evaluate the H 2 O and/or SO 2 tolerance under the reaction conditions. The H 2 O and/or SO 2 tolerance of the serial NiFe mixed oxides catalysts was investigated at 300 • C under the reaction conditions of 600 ppm NO, 600 ppm NH 3 , 5% O 2 , 0 or 100 ppm SO 2 , 0 or 10.0% H 2 O, and balanced with N 2 . As depicted in Figure 1D, at the first stage, the NO x conversion for the NiFe-400, NiFe-500, and NiFe-600 catalysts was maintained at around 85%, 99%, and 96%, respectively. After adding 100 ppm SO 2 into the reaction gas, the NO x conversion decreased slightly. However, the co-existence of 100 ppm SO 2 and 10% H 2 O led to a gradual decrease to some extent in NO x conversion, where the decrease was only about 2% for the NiFe-500 and NiFe-600 samples while around a 5% decrease was displayed for NiFe-400. After purging off SO 2 and H 2 O, the NO x conversion gradually recovered to the original values. Thus, the NiFe-500 and NiFe-600 catalysts possessed an excellent SO 2 and H 2 O resistance ability in comparison to the NiFe-400 catalyst.
XRD is an essential technology to evaluate the structure of catalysts. The XRD pattern of the NiFe-LDH precursor is presented in Figure 2A The FT-IR spectrum of NiFe-LDH is illustrated in Figure 2B. An intense and board peak at 3429 cm −1 corresponded to the stretching vibration of the -OH group in the NiFe brucite-like layers, the interlayer water molecules, and the lattice water [23]. The peak that occurred at 2987 cm −1 can be assigned to a CO 2 -H 2 O bridging bond [24]. The vibration of the angular deformation of H 2 O molecules was observed at 1601 cm −1 [23]. The peaks at 1385 cm −1 belonged to the stretching vibrations of carbonate ions [25]. Additionally, the peak located at 676 cm −1 might be correlated to the vibration of M-O (Fe-O, Fe-O-Ni or Ni-O) on the layer of the NiFe-LDH samples [26]. Both XRD and FT-IR data confirmed that NiFe-LDH was prepared successfully with CO 2− 3 as an interlayer anion.  [26]. Both XRD and FT-IR data confirmed that NiFe-LDH was prepared successfully with CO 2-3 as an interlayer anion. The process of thermal decomposition of the NiFe-LDH precursor into mixed oxides was studied by TG coupled with MS (mass spectrometer) analysis (TG-MS). From the TG and DTG (derivative thermogravimetry) curves in Figure 3A, three major weight-loss stages were observed visually. The initial reduction in mass occurred between 80 °C and 180 °C (approximately 9 wt.%) and arose from the desorption of the surface adsorbed and interlayer water and CO2 molecules. The second mass loss stage occurred at the temperature range of 190-350 °C (about 13 wt.%) and mainly released H2O, CO2, and NO2, which were detected by online MS analysis ( Figure 3B). This rapid and major mass loss can be assigned to the dehydroxylation of the inorganic layers and decomposition of interlayer anions (NO -3 and CO 2-3 ) [27] accompanied by the collapse of the hydrotalcite structure. A small DTG exothermic peak at about 395 °C observed in the Figure 3A corresponds to the release of CO2 and NO2 in the MS curves, which was attributed to the degradation of nitrates and carbonates with high thermal stability. Based on the above, stable NiFe mixed oxides catalysts can be obtained when hydrotalcite-like precursors are calcined at over 400 °C. An XRD analysis was performed for the serial NiFe mixed oxides catalysts, which were produced by calcining NiFe-LDH at different temperatures (400-600 °C). As exhibited in Figure 2C, the calcination temperature was an important factor affecting the formation of crystalline phase. As The process of thermal decomposition of the NiFe-LDH precursor into mixed oxides was studied by TG coupled with MS (mass spectrometer) analysis (TG-MS). From the TG and DTG (derivative thermogravimetry) curves in Figure 3A, three major weight-loss stages were observed visually. The initial reduction in mass occurred between 80 • C and 180 • C (approximately 9 wt.%) and arose from the desorption of the surface adsorbed and interlayer water and CO 2 molecules. The second mass loss stage occurred at the temperature range of 190-350 • C (about 13 wt.%) and mainly released H 2 O, CO 2 , and NO 2 , which were detected by online MS analysis ( Figure 3B). This rapid and major mass loss can be assigned to the dehydroxylation of the inorganic layers and decomposition of interlayer anions (NO − 3 and CO 2− 3 ) [27] accompanied by the collapse of the hydrotalcite structure. A small DTG exothermic peak at about 395 • C observed in the Figure 3A corresponds to the release of CO 2 and NO 2 in the MS curves, which was attributed to the degradation of nitrates and carbonates with high thermal stability. Based on the above, stable NiFe mixed oxides catalysts can be obtained when hydrotalcite-like precursors are calcined at over 400 • C.  [26]. Both XRD and FT-IR data confirmed that NiFe-LDH was prepared successfully with CO 2-3 as an interlayer anion. The process of thermal decomposition of the NiFe-LDH precursor into mixed oxides was studied by TG coupled with MS (mass spectrometer) analysis (TG-MS). From the TG and DTG (derivative thermogravimetry) curves in Figure 3A, three major weight-loss stages were observed visually. The initial reduction in mass occurred between 80 °C and 180 °C (approximately 9 wt.%) and arose from the desorption of the surface adsorbed and interlayer water and CO2 molecules. The second mass loss stage occurred at the temperature range of 190-350 °C (about 13 wt.%) and mainly released H2O, CO2, and NO2, which were detected by online MS analysis ( Figure 3B). This rapid and major mass loss can be assigned to the dehydroxylation of the inorganic layers and decomposition of interlayer anions (NO -3 and CO 2-3 ) [27] accompanied by the collapse of the hydrotalcite structure. A small DTG exothermic peak at about 395 °C observed in the Figure 3A corresponds to the release of CO2 and NO2 in the MS curves, which was attributed to the degradation of nitrates and carbonates with high thermal stability. Based on the above, stable NiFe mixed oxides catalysts can be obtained when hydrotalcite-like precursors are calcined at over 400 °C. An XRD analysis was performed for the serial NiFe mixed oxides catalysts, which were produced by calcining NiFe-LDH at different temperatures (400-600 °C). As exhibited in Figure 2C, the calcination temperature was an important factor affecting the formation of crystalline phase. As An XRD analysis was performed for the serial NiFe mixed oxides catalysts, which were produced by calcining NiFe-LDH at different temperatures (400-600 • C). As exhibited in Figure 2C, the calcination temperature was an important factor affecting the formation of crystalline phase. As the calcination temperature increases from 400 • C to 600 • C, the diffraction peaks were found to significantly sharpen, representing the mushrooming of crystallinity and particle size. Meanwhile, new crystal phases progressively appeared. For the NiFe-400 sample, all diffraction peaks at 37 4 , in which half of the Fe 3+ occupies the tetrahedral sites with the octahedral sites comprised of a 1:1 ratio of Ni 2+ and Fe 3+ cations. It was reported that such an inverse NiFe 2 O 4 spinel is a valuable material with structural stability and the possibility for bifunctional redox properties [28,29], Combining the results in the catalytic performance tests, the NiFe 2 O 4 spinel could be confirmed to be the expected phase in this system.
The nitrogen adsorption and desorption isotherms of NiFe mixed oxides catalysts are displayed in Figure 4A, in which can clearly be observed the typical shape of type IV isotherms curves (IUPAC classification), indicating the presence of 2-50 nm mesopores [30,31]. The NiFe-400 and NiFe-500 samples displayed a typical type H2 hysteresis loop that verified the existence of an "ink bottle" state structure, whereas the NiFe-600 catalyst presented the narrow type H3 hysteresis loop, demonstrating the existence of a "slit"-shaped mesopore [9,32], Meanwhile, the pore diameter distribution of the serial catalysts can be visually observed from Figure 4B, where NiFe-500 exhibited the more intensive distribution in the region of 1-6 nm. The Brunauer-Emmett-Teller (BET) surface area and the average pore diameter of the as-prepared catalysts are listed in Table 1. The BET specific surface areas calculated by N 2 desorption isotherms were clearly seen to drop sharply with the increase of the calcination temperature in Table 1. However, the order of the catalysts by average pore diameter was NiFe-600 (15.5 nm) > NiFe-400 (5.1 nm) > NiFe-500 (4.0 nm). As discussed in the XRD analysis, with the increasing of calcination temperature, the crystallinity and particle size were both increased. Thus, the BET surface areas were decreased, which is in agreement with the literature [33]. In addition, in comparison to NiFe-400, NiFe-500 showed a smaller pore diameter, which may be due to the shrinkage of pores accompanied by the increase in crystallinity. However, for NiFe-600, a high crystallinity NiFe 2 O 4 spinel was formed, which may result in the formation of a "stacked hole" with a larger pore diameter. The relatively larger specific surface area and even porous structures can provide more catalytic centers, which might account for the better DeNO x performance.  The surface acidity and acid strength distribution of the NiFe mixed oxides catalysts were investigated by NH3-TPD experiments. As illustrated in Figure 5A, the NH3-TPD curves showed two ammonia desorption peaks for each sample in the 70-700 °C range. The first desorption peak centered at around 120 °C corresponds to the physisorbed NH3 or weakly acidic sites [34]. The other peak located at around 480 °C can be assigned to the chemical adsorbed ammonia on strong Lewis and Brønsted acidity sites [35]. The amount of total NH3 adsorption and the proportion for every type of acid site were calculated from the integrated areas of the corresponding peaks. The calculation results were normalized and are listed in Table 2. It is clear that the increase of calcination temperature The surface acidity and acid strength distribution of the NiFe mixed oxides catalysts were investigated by NH 3 -TPD experiments. As illustrated in Figure 5A, the NH 3 -TPD curves showed two ammonia desorption peaks for each sample in the 70-700 • C range. The first desorption peak centered at around 120 • C corresponds to the physisorbed NH 3 or weakly acidic sites [34]. The other peak located at around 480 • C can be assigned to the chemical adsorbed ammonia on strong Lewis and Brønsted acidity sites [35]. The amount of total NH 3 adsorption and the proportion for every type of acid site were calculated from the integrated areas of the corresponding peaks. The calculation results were normalized and are listed in Table 2. It is clear that the increase of calcination temperature resulted in a dramatic reduction of ammonia adsorption and surface acidity. The desorption amount of NiFe-500 and NiFe-600 decreased by 35% and 73%, respectively. In addition, with the calcination temperature increasing from 400 • C to 600 • C, the proportion of weakly acidic sites shows an obvious rise from 34.5% to 52.4%, while the proportion of the strongly acidic sites decreased from 65.5% to 47.6%. Combined with the results of XRD and BET, we speculated that the increase of calcination temperature resulted in severe agglomeration of iron species and the acid center was covered in this process. More acid sites should be conductive to promote DeNO x activity. However, the NiFe-400 catalyst afforded the worst NH 3 -SCR performance though it owned the maximum acid quantity, suggesting that the number of acid sites was not the only factor to affect catalytic performance. Moreover, to our great interest, the NiFe-600 catalyst possessed the lowest number acid sites, but its high-temperature DeNO x activity was satisfactory. The DeNO x mechanism was further analyzed as below. It was confirmed that an iron-based mixed oxide catalyst would follow the Langmuir-Hinshelwood mechanism under 250 • C, where both NH 3 and NO would be adsorbed to all of the acid sites and then transformed to the corresponding intermediate [36]. The dramatic decrease of acid quantity for NiFe-600 resulted in the absence of adsorption of the reactant; accordingly, the low-temperature activity decreased. At medium and higher temperatures, the catalyst would fit the Eley-Rideal mechanism [31,37], where only the adsorption and activation process of NH 3 was conducted, generating the reaction between the adsorbed intermediate and gaseous-state NO. Therefore, the residual acid sites were still enough to absorb the majority of the NH 3 despite the evident decrease.  Table 3. For all samples, the amount of total theoretical H2 consumption was higher than that of the total actual H2 consumption, implying the existence of an incomplete reduction of active species. Therein, the actual H2 consumption amount of Ni species (peak 1) was higher than the theoretical values for each catalyst (A1 > E1), while Fe species (peak 2) were the opposite (A2 < E2). Therefore, the first reduction peak may be ascribed to the co-reduction of Ni as well as partial Fe species and the second peak may be only attributed to the Fe species. From Table 3, it was also clear that the NiFe-400 and NiFe-500 catalysts exhibited a similar position of reduction peaks and the NiFe-600 catalyst significantly shifted to high-temperature zones, presenting the worst reducibility for NiFe-600, which can well-explain its degraded DeNOx activity at a low temperature compared with the NiFe-400 and NiFe-500 catalysts.  where T1 and T2 refer to the reduction peak temperature; A1 and A2 refer to the actual H2 consumption of peak 1 and peak 2, respectively; and E1 and E2 refer to the theoretical H2 consumption of peak 1 and peak 2, respectively. XPS analysis was performed to determine the surface components and valence states of elements presented on the catalyst surfaces. The XPS spectra of Fe 2p and Ni 2p3/2 are shown in Figure 6 and the corresponding surface atomic concentrations as well as the proportion of different valence states are listed in Table 4. The H 2 -TPR technique was employed to determine the reducibility of the as-prepared NiFe mixed oxides catalysts. Over all samples, two noticeable reduction peaks are exhibited in Figure 5B. According to the literature [38,39], the reduction peak located at the lower temperature region (200-400 • C) may be assigned to the reduction of Ni 2+ (Ni 3+ ) to Ni 0 (Peak 1), and the reduction peak at the higher temperature region (400-450 • C) could represent the reduction of Fe species (Peak 2). However, it was reported that the reduction peaks of mixed oxides usually do not correspond to only one kind of metal oxide [40]. Hence, the theoretical (denoted as E) and actual (denoted as A) H 2 consumption of each catalyst were calculated and are shown in Table 3. For all samples, the amount of total theoretical H 2 consumption was higher than that of the total actual H 2 consumption, implying the existence of an incomplete reduction of active species. Therein, the actual H 2 consumption amount of Ni species (peak 1) was higher than the theoretical values for each catalyst (A 1 > E 1 ), while Fe species (peak 2) were the opposite (A 2 < E 2 ). Therefore, the first reduction peak may be ascribed to the co-reduction of Ni as well as partial Fe species and the second peak may be only attributed to the Fe species. From Table 3, it was also clear that the NiFe-400 and NiFe-500 catalysts exhibited a similar position of reduction peaks and the NiFe-600 catalyst significantly shifted to high-temperature zones, presenting the worst reducibility for NiFe-600, which can well-explain its degraded DeNO x activity at a low temperature compared with the NiFe-400 and NiFe-500 catalysts.  where T 1 and T 2 refer to the reduction peak temperature; A 1 and A 2 refer to the actual H 2 consumption of peak 1 and peak 2, respectively; and E 1 and E 2 refer to the theoretical H 2 consumption of peak 1 and peak 2, respectively.
XPS analysis was performed to determine the surface components and valence states of elements presented on the catalyst surfaces. The XPS spectra of Fe 2p and Ni 2p 3/2 are shown in Figure 6 and the corresponding surface atomic concentrations as well as the proportion of different valence states are listed in Table 4. According to Table 4, the amounts of surface Fe species showed a monotonic increase for the NiFe-400, NiFe-500, and NiFe-600 catalysts. As discussed in the XRD analysis section, when NiFe-LDH was calcined at 400 °C, iron species existed in the form of FeOx, which should be mostly covered by Ni species. With the increment of calcination temperature, NiFe2O4 spinels were generated. Fe species were gradually exposed or transferred to the surface of the catalysts, and the ideal model is displayed in Figure 6. It is well-known that more Fe species could result in better high-temperature activity; thus, DeNOx activity at high temperatures (>330 °C) displaying the order of NiFe-600 > NiFe-500 > NiFe-400 can be reasonable.  For the XPS spectra of Fe 2p in Figure 7A, two main peaks assigned to Fe 2p3/2 and Fe 2p1/2 can be observed centered at a binding energy of approximately 711 eV and 724 eV, respectively. By performing peak-fitting deconvolution, the Fe 2p3/2 spectra were fitted into sub-bands located at approximately 710.9 eV and 712.7 eV, which corresponded to Fe 2+ and Fe 3+ , respectively [41]. As shown in Figure 7B, the Ni 2p3/2 spectra can be separated into three characteristic peaks at 854.1, 855.9, and 861.4 eV, which can be ascribed to the Ni 2+ , Ni 3+ , and satellite peaks, respectively [42]. Combined with the data in Table 2, it was clear that the Ni 3+ content on the surface of the NiFe-500 sample (75.5%) According to Table 4, the amounts of surface Fe species showed a monotonic increase for the NiFe-400, NiFe-500, and NiFe-600 catalysts. As discussed in the XRD analysis section, when NiFe-LDH was calcined at 400 • C, iron species existed in the form of FeO x , which should be mostly covered by Ni species. With the increment of calcination temperature, NiFe 2 O 4 spinels were generated. Fe species were gradually exposed or transferred to the surface of the catalysts, and the ideal model is displayed in Figure 6. It is well-known that more Fe species could result in better high-temperature activity; thus, DeNO x activity at high temperatures (>330 • C) displaying the order of NiFe-600 > NiFe-500 > NiFe-400 can be reasonable. For the XPS spectra of Fe 2p in Figure 7A, two main peaks assigned to Fe 2p 3/2 and Fe 2p 1/2 can be observed centered at a binding energy of approximately 711 eV and 724 eV, respectively. By performing peak-fitting deconvolution, the Fe 2p 3/2 spectra were fitted into sub-bands located at approximately 710.9 eV and 712.7 eV, which corresponded to Fe 2+ and Fe 3+ , respectively [41]. As shown in Figure 7B, the Ni 2p 3/2 spectra can be separated into three characteristic peaks at 854.1, 855.9, and 861.4 eV, which can be ascribed to the Ni 2+ , Ni 3+ , and satellite peaks, respectively [42]. Combined with the data in Table 2, it was clear that the Ni 3+ content on the surface of the NiFe-500 sample (75.5%) was higher than that of the NiFe-400 sample (72.5%), but the content of the surface Fe 3+ species was opposite for the two catalysts, which indicated that increasing the calcination temperature could result in a redox reaction between Ni 2+ and Fe 3+ to form Ni 3+ and Fe 2+ (Ni 2+ + Fe 3+ displayed in Figure 6. It is well-known that more Fe species could result in better high-temperature activity; thus, DeNOx activity at high temperatures (>330 °C) displaying the order of NiFe-600 > NiFe-500 > NiFe-400 can be reasonable.  For the XPS spectra of Fe 2p in Figure 7A, two main peaks assigned to Fe 2p3/2 and Fe 2p1/2 can be observed centered at a binding energy of approximately 711 eV and 724 eV, respectively. By performing peak-fitting deconvolution, the Fe 2p3/2 spectra were fitted into sub-bands located at approximately 710.9 eV and 712.7 eV, which corresponded to Fe 2+ and Fe 3+ , respectively [41]. As shown in Figure 7B, the Ni 2p3/2 spectra can be separated into three characteristic peaks at 854.1, 855.9, and 861.4 eV, which can be ascribed to the Ni 2+ , Ni 3+ , and satellite peaks, respectively [42]. Combined with the data in Table 2, it was clear that the Ni 3+ content on the surface of the NiFe-500 sample (75.5%) was higher than that of the NiFe-400 sample (72.5%), but the content of the surface Fe 3+ species was opposite for the two catalysts, which indicated that increasing the calcination temperature could result in a redox reaction between Ni 2+ and Fe 3+ to form Ni 3+ and Fe 2+ (Ni 2+ + Fe 3+ ↔ Ni 3+ + Fe 2+ ). However, the amount of surface Ni 3+ decreased to 60.1% with the increase in calcination temperature from 500 °C to 600 °C for the NiFe catalyst, which can be attributed to the formation of more highcrystallinity NiFe2O4 spinels, so it was difficult for the reaction (Ni 2+ + Fe 3+ ↔ Ni 3+ + Fe 2+ ) to occur on the surface of the catalyst. Thus, comparatively, the redox reaction (Ni 2+ + Fe 3+ ↔ Ni 3+ + Fe 2+ ) may more easily occur for the NiFe-500 sample. Such interaction in the bimetallic oxide was similar to that previously reported [30,43] and it was considered to be favorable for the creation of a redox cycle and further promoted the catalytic activity, especially at low temperatures.
Ni 3+ + Fe 2+ ). However, the amount of surface Ni 3+ decreased to 60.1% with the increase in calcination temperature from 500 • C to 600 • C for the NiFe catalyst, which can be attributed to the formation of more high-crystallinity NiFe 2 O 4 spinels, so it was difficult for the reaction (Ni 2+ + Fe 3+ LDH was calcined at 400 °C, iron species existed in the form of FeOx, which should be mostly covered by Ni species. With the increment of calcination temperature, NiFe2O4 spinels were generated. Fe species were gradually exposed or transferred to the surface of the catalysts, and the ideal model is displayed in Figure 6. It is well-known that more Fe species could result in better high-temperature activity; thus, DeNOx activity at high temperatures (>330 °C) displaying the order of NiFe-600 > NiFe-500 > NiFe-400 can be reasonable.  For the XPS spectra of Fe 2p in Figure 7A, two main peaks assigned to Fe 2p3/2 and Fe 2p1/2 can be observed centered at a binding energy of approximately 711 eV and 724 eV, respectively. By performing peak-fitting deconvolution, the Fe 2p3/2 spectra were fitted into sub-bands located at approximately 710.9 eV and 712.7 eV, which corresponded to Fe 2+ and Fe 3+ , respectively [41]. As shown in Figure 7B, the Ni 2p3/2 spectra can be separated into three characteristic peaks at 854.1, 855.9, and 861.4 eV, which can be ascribed to the Ni 2+ , Ni 3+ , and satellite peaks, respectively [42]. Combined with the data in Table 2, it was clear that the Ni 3+ content on the surface of the NiFe-500 sample (75.5%) was higher than that of the NiFe-400 sample (72.5%), but the content of the surface Fe 3+ species was opposite for the two catalysts, which indicated that increasing the calcination temperature could result in a redox reaction between Ni 2+ and Fe 3+ to form Ni 3+ and Fe 2+ (Ni 2+ + Fe 3+ ↔ Ni 3+ + Fe 2+ ). However, the amount of surface Ni 3+ decreased to 60.1% with the increase in calcination temperature from 500 °C to 600 °C for the NiFe catalyst, which can be attributed to the formation of more highcrystallinity NiFe2O4 spinels, so it was difficult for the reaction (Ni 2+ + Fe 3+ ↔ Ni 3+ + Fe 2+ ) to occur on the surface of the catalyst. Thus, comparatively, the redox reaction (Ni 2+ + Fe 3+ ↔ Ni 3+ + Fe 2+ ) may more easily occur for the NiFe-500 sample. Such interaction in the bimetallic oxide was similar to that previously reported [30,43] and it was considered to be favorable for the creation of a redox cycle and further promoted the catalytic activity, especially at low temperatures.
Ni 3+ + Fe 2+ ) to occur on the surface of the catalyst. Thus, comparatively, the redox reaction (Ni 2+ + Fe 3+ According to Table 4, the amounts of surface Fe species showed a monotonic increase for the NiFe-400, NiFe-500, and NiFe-600 catalysts. As discussed in the XRD analysis section, when NiFe-LDH was calcined at 400 °C, iron species existed in the form of FeOx, which should be mostly covered by Ni species. With the increment of calcination temperature, NiFe2O4 spinels were generated. Fe species were gradually exposed or transferred to the surface of the catalysts, and the ideal model is displayed in Figure 6. It is well-known that more Fe species could result in better high-temperature activity; thus, DeNOx activity at high temperatures (>330 °C) displaying the order of NiFe-600 > NiFe-500 > NiFe-400 can be reasonable.  For the XPS spectra of Fe 2p in Figure 7A, two main peaks assigned to Fe 2p3/2 and Fe 2p1/2 can be observed centered at a binding energy of approximately 711 eV and 724 eV, respectively. By performing peak-fitting deconvolution, the Fe 2p3/2 spectra were fitted into sub-bands located at approximately 710.9 eV and 712.7 eV, which corresponded to Fe 2+ and Fe 3+ , respectively [41]. As shown in Figure 7B, the Ni 2p3/2 spectra can be separated into three characteristic peaks at 854.1, 855.9, and 861.4 eV, which can be ascribed to the Ni 2+ , Ni 3+ , and satellite peaks, respectively [42]. Combined with the data in Table 2, it was clear that the Ni 3+ content on the surface of the NiFe-500 sample (75.5%) was higher than that of the NiFe-400 sample (72.5%), but the content of the surface Fe 3+ species was opposite for the two catalysts, which indicated that increasing the calcination temperature could result in a redox reaction between Ni 2+ and Fe 3+ to form Ni 3+ and Fe 2+ (Ni 2+ + Fe 3+ ↔ Ni 3+ + Fe 2+ ). However, the amount of surface Ni 3+ decreased to 60.1% with the increase in calcination temperature from 500 °C to 600 °C for the NiFe catalyst, which can be attributed to the formation of more highcrystallinity NiFe2O4 spinels, so it was difficult for the reaction (Ni 2+ + Fe 3+ ↔ Ni 3+ + Fe 2+ ) to occur on the surface of the catalyst. Thus, comparatively, the redox reaction (Ni 2+ + Fe 3+ ↔ Ni 3+ + Fe 2+ ) may more easily occur for the NiFe-500 sample. Such interaction in the bimetallic oxide was similar to that previously reported [30,43] and it was considered to be favorable for the creation of a redox cycle and further promoted the catalytic activity, especially at low temperatures.
Ni 3+ + Fe 2+ ) may more easily occur for the NiFe-500 sample. Such interaction in the bimetallic oxide was similar to that previously reported [30,43] and it was considered to be favorable for the creation of a redox cycle and further promoted the catalytic activity, especially at low temperatures.

Catalyst Preparation
The NiFe hydrotalcite-like precursor with the molar ratio of Ni to Fe being 4 was prepared by the urea hydrolysis method [44]. First, a urea solution and a mixed solution with Ni(NO3)2·6H2O, data were recorded when the steady-state reaction was maintained after 30 min at each temperature. The NO x conversion and N 2 selectivity were calculated according to the following expression:

Conclusions
A series of NiFe mixed oxides catalysts derived from hydrotalcite-like precursors were prepared and novelly employed as NH 3 -SCR catalysts. The result proposed that the ideal phase composition of NiFe mixed oxides catalysts was available via adjusting the calcination temperature of NiFe-LDH, which had a significant effect on DeNO x activity. The oxide phase formed at a lower calcination temperature is propitious for a better redox property and low-temperature activity, while the presence of NiFe 2 O 4 spinels could contribute to high-temperature activity as well as SO 2 resistance and stability. We testify to NiFe-500 being the optimal catalyst incorporating NiO and NiFe 2 O 4 spinels, where the virtues of the two phases were synergistically exerted and the best NH 3 -SCR performance was achieved.
Author Contributions: X.W. and Y.D. designed and administered the experiments. C.Z. and X.L. performed experiments. R.W. collected and analyzed data. R.W., X.W., and Y. D wrote the manuscript.