Influence of Biochar Composition and Micro-Structure on the Denitration of Flue Gases at High Temperature

: Biochar materials are good reducers of nitrogen oxides. The composition and structure of biochar a ﬀ ect signiﬁcantly its ability to reduce C–NO. In order to study the denitration of ﬂue gases by biochar at high temperature, three kinds of biochar (bamboo charcoal (BC), rice husk ash (RHA), and straw charcoal (SC)) were mixed with cement raw meal in a ﬁxed-bed quartz reactor at the temperature of 800–900 ◦ C and O 2 concentration of 0.5%–2%. The results showed that the initial denitration rate of BC was higher than that of RHA, and that of SC was the lowest. RHA had the largest speciﬁc surface area, and BC the smallest. The elements C, N, and O and the functional groups of the three types of biochar had a greater inﬂuence on the denitration rate than their structures. The denitration rate decreased faster as the O / C ratio increased, and the increase in the relative content of the N element induced the formation of nitrogen-containing functional groups catalyzing C–NO reduction. The content of the C–C bond a ﬀ ected directly the rate of denitration, and both (NCO) x and C–O bonds had a positive e ﬀ ect on the reduction capability of biochar. It can be concluded that the composition of biochar has an important e ﬀ ect on the reduction of C–NO.


Introduction
The emission of NOx (including NO, N 2 O, NO 2 , etc.) harms seriously humans health and the environment by causing acid rain and photochemical reactions. With the expansion of construction in China, the NOx emissions of the cement industry have also increased year by year. In 2017, the amount of NOx emissions by the cement industry was 2.4 million tons [1,2]. The denitration technologies currently used in the cement industry are Selective Non-catalytic Reduction (SNCR) and Selective Catalytic Reduction (SCR). The SNCR technology does not require a catalyst and does not induce the conversion of SO 2 to SO 3 , which is suitable for the temperature reaction window of the cement kiln (800-1100 • C). However, the SNCR denitration efficiency can only reach about 50%, and SNCR causes the release of ammonia, thus promoting pollution [3], in contrast with the current emission regulations. The SCR technology requires precious metals or heavy metals as catalysts, such as titanium, vanadium, manganese, and other metals oxides, and uses ammonia or urea as a reducing agent in a low temperature range (350-450 • C). The SCR technology reduces nitrogen oxides to N 2 and H 2 O under the action of a catalyst [4]. The shortcomings of SCR are that the catalyst is extremely susceptible to poisoning, secondary pollution is caused by ammonia escape, and it requires high investments and operating costs [4]. Therefore, in order to meet the requirements of national environmental protection in the future, it is necessary to develop new technologies for nitrogen oxide reduction.
In recent years, biochar, a largely available and easily accessible renewable resource, has been used as a new energy source in various fields such as agriculture and industry [5]. Biochar-based materials are porous solid particulate materials produced by pyrolysis of a carbon-rich biomass material under anaerobic or anoxic conditions [6]. The main reactions leading to the reduction of nitrogen oxides by biochar are [7] 2C + 2NO → 2CO + N 2 (1) Recently, many scholars have studied the denitration process occurring when using biochar as a reducing agent to reduce nitrogen oxides. Garijo et al. [8] found that eucalyptus charcoal had an obvious reduction effect on NO. Due to its larger specific surface area and oxygen-containing functional groups on its surface, it is more active than charcoal and presents more active sites at high temperatures. Wang and Si [9] studied the C-NO reduction reaction with straw charcoal and found that carbon prepared at 1073 K exhibited the best pore structure, the largest specific surface area, and the best oxidation activity. The NO removal efficiency was more than 80%. Some researchers [10][11][12][13] found that the surface of biochar has highly catalytic and reactive sites and is rich in oxygen-containing functional groups such as carboxyl groups, hydroxyl groups, phenolic hydroxyl groups, which probably promote denitration.
A number of studies on the surface functional groups of carbonaceous materials have been published. It has been reported that the nitrogen oxide (NOx) adsorption capability of carbon materials is influenced greatly by the content of nitrogen and carbon-containing functional groups (CFGs) [10,11,14]. Pastor-Villegas et al. [15], Liu et al. [16], and Xia et al. [17] proved that the concentration of such oxygen-containing functional groups (OFGs) corresponded to the presence of acidic oxygen-containing groups, and the nitrogen-containing functional groups (NFGs) increased significantly after high-temperature treatment. Meng et al. [18] also analyzed the surface chemical composition of heat-treated bamboo slivers, showing that the O/C ratio decreased as temperature increased, while the amount of both C-C and C-O bonds increased. Wan [19] studied the effect of heat treatment and nitriding treatment on the reduction of NO in activated carbon materials. The results showed that the relative content of surface C-OH/C-O-C increased, while that of C=O decreased after heat treatment. The nitriding treatment resulted in an increase of nitrogen-containing compounds on the surface and significantly promoted the reduction of NO at the same time.
From the above studies, it can be found that the C-O bond is a reactive bond, and NFGs improve NO reduction. However, researches published so far mainly focused on a material, such as activated carbon or bamboo charcoal, and no systematic comparison of various biochar materials including their composition and structure has been conducted.
It was therefore the objective of this study to investigate the reduction of NO by different types of biochar, including rice husk ash (RHA), bamboo charcoal (BC), and straw charcoal (SC), mixed with cement raw material.

Materials and Preparation
By comparing the physicochemical properties and surface topography of three biochar, this paper investigated the effects of carbon on the reduction of NO. Cement raw material was purchased from Beijing Building Materials Institute. In this experiment, the particle size of these carbon materials was 0.15 mm. Before the denitration experiments, the materials were dried at 100 • C for 24 h.
The cement raw material was mixed with the biochar and ground uniformly; the final amount of the mixture was 1 g. Then, the mixture was fed into a tube furnace and mixed well. The experimental setup is shown in Figure 1. Biochar with a low mass fraction could not provide a sufficient NOx conversion rate, and a higher mass fraction would affect the calcination and composition of the cement Appl. Sci. 2020, 10, 1920 3 of 12 raw material. Therefore, on the basis of previous experimental data [20], biochar at the concentration of 5 wt % was mixed with the cement raw material. materials was 0.15 mm. Before the denitration experiments, the materials were dried at 100 °C for 24 h.
The cement raw material was mixed with the biochar and ground uniformly; the final amount of the mixture was 1 g. Then, the mixture was fed into a tube furnace and mixed well. The experimental setup is shown in Figure 1. Biochar with a low mass fraction could not provide a sufficient NOx conversion rate, and a higher mass fraction would affect the calcination and composition of the cement raw material. Therefore, on the basis of previous experimental data [20], biochar at the concentration of 5 wt % was mixed with the cement raw material.

Experimental and Measurement Methods
Denitration experiments were carried out in a fixed-bed quartz reactor housed in an electrically heated oven, as illustrated in Figure 1. The quartz tube, 30 mm in inner diameter and 700 mm in length, contained a porous layer in the middle where the reaction by a reactor gas occurred. Three gases including N2, O2/N2, and NO/N2 were investigated. A nitrogen oxides analyzer was used to record the outlet concentration of NO: where NOin and NOout represent the inlet and outlet concentration of NO, respectively.
According to published literature [13][14][15][16][17] and previous experimental data [20,21], the optimal temperature of C-NO reduction is about 800-900 °C, so the temperatures selected in this experiment were 800 °C, 850 °C, and 900 °C. The O2 concentration was too low to be easily realized in an industrial atmospheric system. A higher O2 concentration produces a large amount of carbon oxides, which may affect the reduction ability of C-NO [20,21]. Therefore, we selected O2 concentrations of 0.5%, 1%, and 2%.

Characterization
The main elements of three carbon materials were measured by an Elemental Analyzer (EA, C440, made by Elementar Analysensysteme GmbH, Hanau, Germany). The surface functional groups in biochar were detected by X-ray Photoelectron Spectroscopy (XPS, Escalab 250Xi, made by Thermo Fisher, UK). The surface areas were determined by the N2-BET (C418, made by Micromeritics Instruments Corporation,Georgia, USA). The surface microscopic morphology of the carbon materials was detected by SEM (SU8020, made by Hitachi High-tech, Tokyo, Japan).

Experimental and Measurement Methods
Denitration experiments were carried out in a fixed-bed quartz reactor housed in an electrically heated oven, as illustrated in Figure 1. The quartz tube, 30 mm in inner diameter and 700 mm in length, contained a porous layer in the middle where the reaction by a reactor gas occurred. Three gases including N 2 , O 2 /N 2 , and NO/N 2 were investigated. A nitrogen oxides analyzer was used to record the outlet concentration of NO: where NO in and NO out represent the inlet and outlet concentration of NO, respectively. According to published literature [13][14][15][16][17] and previous experimental data [20,21], the optimal temperature of C-NO reduction is about 800-900 • C, so the temperatures selected in this experiment were 800 • C, 850 • C, and 900 • C. The O 2 concentration was too low to be easily realized in an industrial atmospheric system. A higher O 2 concentration produces a large amount of carbon oxides, which may affect the reduction ability of C-NO [20,21]. Therefore, we selected O 2 concentrations of 0.5%, 1%, and 2%.

Characterization
The main elements of three carbon materials were measured by an Elemental Analyzer (EA, C440, made by Elementar Analysensysteme GmbH, Hanau, Germany). The surface functional groups in biochar were detected by X-ray Photoelectron Spectroscopy (XPS, Escalab 250Xi, made by Thermo Fisher, Waltham, MA, USA). The surface areas were determined by the N 2 -BET (C418, made by Micromeritics Instruments Corporation, Norcross, GA, USA). The surface microscopic morphology of the carbon materials was detected by SEM (SU8020, made by Hitachi High-tech, Tokyo, Japan).

BET and SEM Results
The structural parameters of the three selected carbon materials are shown in Table 1. The average values were determined by performing three tests. The BET surface areas of RHA, BC, and SC were 354.1 m 2 g −1 , 9.2 m 2 g −1 , and 148.3 m 2 g −1 , respectively. RHA had the largest specific surface area, and BC has the least specific surface area. The surface morphologies of RHA, BC, and SC are shown in Figure 2. At 100 k magnification, the surface morphology of the three types of biochar were similar. The surface morphology appeared cracked, characterized by a densely, randomly arranged network structure. This structure could easily contact the reaction gas. The active elements and related chemical bonds on the carbon surface were activated and formed more reactive sites, so that the biochar achieved a higher nitrogen oxide conversion rate at high temperature. As shown in Table 1, the pore volumes of RHA, BC, and SC were 0.24 cm 3 g −1 , 0.02 cm 3 g −1 , and 0.12 cm 3 g −1 , respectively. According to the data, the pore volume of RHA was relatively larger than those of SC and BC is. The volume of the surface pores may be related to the denitration rate, which is affected by the concentration of reaction gas. An experimental analysis was performed. SC were 354.1 m 2 g −1 , 9.2 m 2 g −1 , and 148.3 m 2 g −1 , respectively. RHA had the largest specific surface area, and BC has the least specific surface area. The surface morphologies of RHA, BC, and SC are shown in Figure 2. At 100 k magnification, the surface morphology of the three types of biochar were similar. The surface morphology appeared cracked, characterized by a densely, randomly arranged network structure. This structure could easily contact the reaction gas. The active elements and related chemical bonds on the carbon surface were activated and formed more reactive sites, so that the biochar achieved a higher nitrogen oxide conversion rate at high temperature. As shown in Table  1, the pore volumes of RHA, BC, and SC were 0.24 cm 3 g −1 , 0.02 cm 3 g −1 , and 0.12 cm 3 g −1 , respectively. According to the data, the pore volume of RHA was relatively larger than those of SC and BC is. The volume of the surface pores may be related to the denitration rate, which is affected by the concentration of reaction gas. An experimental analysis was performed.    Figure 3 shows the denitration rate of the three biochars at different reaction temperatures. From Figure 3a, at 800 • C, 850 • C, and 900 • C, the initial denitration rates of RHA were 87.7%, 92.7%, and 89.7%, respectively. The denitration rates of RHA were higher at the start of the experiment in the three temperature conditions, which indicates that RHA has high reduction ability at high temperatures. As shown in Figure 3a, the denitration rate of RHA was the highest at the temperature of 850 • C and decreased slowly with time. When the temperature increased to 900 • C, the denitration rate of RHA decreased. The reason for this may be that at 850 • C, the reduction of NO by C reached an optimal state. At high temperatures, carbon oxides are more likely to be produced [22]. When the temperature was 900 • C, more carbon oxides were generated. On the one hand, the carbon oxides produced consumed C and O, and on the other hand, it suppressed the reduction reaction. Figure 3b shows the denitration rate of BC at different temperatures. The initial denitration rates of BC were 70.9%, 81.6%, and 98.9%, corresponding to the environmental temperatures of 800 • C, 850 • C, and 900 • C, respectively. The differences in the initial denitration rates at the three reaction temperatures were larger compared to those of RHA, and the highest denitration rate was measured at 900 • C. From the data, it can be seen that BC is a good reduced of NO when the temperature is above 850 • C. This may be due to the fact that BC contains more C-O bonds and (NCO)x bonds than RHA, which are promote the reduction reaction. These chemical bonds are more active at high temperatures, so the denitration rate was the highest at 900 • C. Figure 3c shows the denitration rate of SC. The initial denitration rates at 800 • C, 850 • C, and 900 • C were 85.3%, 88.0%, and 91.3%, respectively. The differences in the denitration rates of SC at different temperatures were small. SC displayed a higher reducing effect on nitrogen oxides at higher temperatures. In summary, the optimum reaction temperatures for RHA, BC, and SC were 850 • C, 900 • C and 900 • C, respectively. 89.7%, respectively. The denitration rates of RHA were higher at the start of the experiment in the three temperature conditions, which indicates that RHA has high reduction ability at high temperatures. As shown in Figure 3a, the denitration rate of RHA was the highest at the temperature of 850 °C and decreased slowly with time. When the temperature increased to 900 °C, the denitration rate of RHA decreased. The reason for this may be that at 850 °C, the reduction of NO by C reached an optimal state. At high temperatures, carbon oxides are more likely to be produced [22]. When the temperature was 900 °C, more carbon oxides were generated. On the one hand, the carbon oxides produced consumed C and O, and on the other hand, it suppressed the reduction reaction. Figure 3b shows the denitration rate of BC at different temperatures. The initial denitration rates of BC were 70.9%, 81.6%, and 98.9%, corresponding to the environmental temperatures of 800 °C, 850 °C, and 900 °C, respectively. The differences in the initial denitration rates at the three reaction temperatures were larger compared to those of RHA, and the highest denitration rate was measured at 900 °C. From the data, it can be seen that BC is a good reduced of NO when the temperature is above 850 °C. This may be due to the fact that BC contains more C-O bonds and (NCO)x bonds than RHA, which are promote the reduction reaction. These chemical bonds are more active at high temperatures, so the denitration rate was the highest at 900 °C. Figure 3c shows the denitration rate of SC. The initial denitration rates at 800 °C, 850 °C, and 900 °C were 85.3%, 88.0%, and 91.3%, respectively. The differences in the denitration rates of SC at different temperatures were small. SC displayed a higher reducing effect on nitrogen oxides at higher temperatures. In summary, the optimum reaction temperatures for RHA, BC, and SC were 850 °C, 900 °C and 900 °C, respectively. (c) SC

Effect of O 2 Concentration
Based on the above conclusions, RHA, BC, and SC were selected to investigate the effect of 0.5%-2% O 2 on the denitration rate at the reaction temperatures of 850 • C, 900 • C, and 900 • C, as shown in Figure 4. The initial denitration rates of RHA were 94.2% at 0.5% of O 2 concentration, 92.7% at 1%, and 84.5% at 2%. As shown in Figure 4a, when O 2 concentration was 0.5% and 1%, the denitration rate decreased slowly with time. However, the initial denitration rate was lower when the oxygen concentration was 2%, and its decline was faster, so that a 36% denitration rate was reached after 10 min. As shown in Figure 4b, the denitration rate of BC increased first and then decreased with the increase of O 2 concentration. The initial denitration rates of BC were 84.9% at 0.5%, 98.9% at 1%, and 72.7% at 2% of O 2 concentration, respectively. When the oxygen concentration was low, nitrogen oxides were adsorbed on the surface of BC, occupying some reaction sites and resulting in a weaker C-NO reaction. When the oxygen concentration was high, more reaction sites caused C-NO to react more easily. The initial denitration rate was lower at 2% oxygen concentration. As shown in Figure 5b, as the oxygen concentration increased, the relative content of CO 2 increased. At 2% oxygen concentration, the reaction produced more CO 2 , which suppressed the reduction reaction and decreased CO production. As shown in Figure 4c, the initial denitration rates were almost the same t different oxygen concentrations. The denitration rate declined faster with time under higher O 2 concentrations.
The presence of oxygen molecules helps to promote the C-NO reaction under high temperature. However, there is a competition between C-O2 and C-NO reactions, and the former is thermodynamically easier. From the energy changes of the three types of reaction involving C, it appears that the easiest oxidation of C is the one that leads to the formation of CO or CO2. In addition, the enthalpy change of CO generation is lower than that of the generation of other oxidation products of C. The enthalpy changes of C directly reducing NO is higher. This means that the presence of O2 causes a large consumption of carbon and lowers the selectivity of the reaction [20]. Therefore, the optimal choice of O2 concentration is necessary in the reduction of C-NO.    Based on the above data, the optimal reaction conditions and denitration rates of the three biochar are shown in Table 2. Both RHA and SC presented the optimal denitration rates when O 2 concentration was 0.5%, whereas the O 2 concentration allowing the optimal denitration rate by BC was 1%. From the above analysis, it appears that the specific surface area and the surface morphology have no important effect on the denitration rate. The SEM pictures of the three biochars show that they had similar surface morphology. The specific surface area and pore volume of BC are smaller than those of the other two biochars. The O element was hardly absorbed on the surface of this biochar at low O 2 concentration, resulting in lower efficiency. At a higher concentration, O 2 reacted with the biochar, which generated large amounts of carbon oxides and inhibited C-NO reduction. Therefore, the optimal O 2 concentration was 1% for BC. In contrast, RHA and SC have a large specific surface area and pore volume, which allowed them to achieve a higher denitration efficiency at a lower O 2 concentrations. BC presented the largest initial denitration rate with the least specific surface area, while RHA and SC behaved the opposite. Therefore, at a high reaction temperature, the specific surface area has no important effect on the denitration rate of biochar. The larger the specific surface area, the lower the O 2 concentration required for the biochar to reach the optimal denitration rate.

Elemental Analysis Results
The main elements of the three carbon materials were detected by Elemental Analysis (EA) of C, H, N, O, S, and the ash, as shown in Table 3. The ash contents of RHA and BC were, 6.13% and 8.38%, respectively, lower than that of SC, which was 38.51%. The reactive element C contents of RHA, BC, and SC were 87.00%, 72.30%, and 52.18%, respectively. The C content in RHA and BC was higher, which resulted in their high denitration rate. Although the C relative content of SC was the least, the initial denitration rate of SC was much higher than those of RHA and BC, which means that the C element in this biochar is a better reducer of C-NO. This may result from the presence of a higher number of aromatic hydrocarbons and various reactive functional groups in SC. As for the O/C ratio which will inhibit the reduction of C-NO [17], the relative content in BC and SC was relatively high. The concentration of CO and CO 2 produced during the C-NO reduction reaction of the three kinds of biochar is shown in Figure 5. As shown in Figure 5a, with the increase of O 2 concentration and reaction time, the relative content of CO produced in the reduction of nitrogen oxides by RHA decreased. The concentration of CO 2 increased with increasing O 2 concentration and reaction time. The relative changes in the concentration of CO and CO 2 indicated that CO participated in the reduction Appl. Sci. 2020, 10, 1920 8 of 12 reaction to generate CO 2 . Figure 5b shows the concentration of COx produced by the reduction of NO by BC. The concentration of CO was the highest when O 2 was 1%, decreased in the presence of 0.5% O 2 , and further diminished with 2% O 2 . Under high temperature conditions, carbon is more likely to generate CO as the oxygen concentration decreases. However, as the oxygen concentration increases, CO 2 is more easily generated [22]. Thus, at the optimal oxygen concentration, there is more CO and less CO 2 . The concentration of CO 2 increased with the increase of O 2 concentration and reaction time. As shown in Figure 5c, the COx concentration trend in the presence of SC was similar to that obtained with RHA. From the analysis of the COx concentration generated from the three biochars in the reduction of NO, some conclusions can be drawn. CO participates in the reduction reaction, and its involvement is related to O 2 concentration during the conditions. With the increase of O 2 concentration, a large amount of CO 2 will be generated and will inhibit the activity of the reaction sites on the carbon surface, thus suppressing the reduction of C-NO.
The presence of oxygen molecules helps to promote the C-NO reaction under high temperature. However, there is a competition between C-O 2 and C-NO reactions, and the former is thermodynamically easier. From the energy changes of the three types of reaction involving C, it appears that the easiest oxidation of C is the one that leads to the formation of CO or CO 2 . In addition, the enthalpy change of CO generation is lower than that of the generation of other oxidation products of C. The enthalpy changes of C directly reducing NO is higher. This means that the presence of O 2 causes a large consumption of carbon and lowers the selectivity of the reaction [20]. Therefore, the optimal choice of O 2 concentration is necessary in the reduction of C-NO.

Elemental Analysis Results
The main elements of the three carbon materials were detected by Elemental Analysis (EA) of C, H, N, O, S, and the ash, as shown in Table 3. The ash contents of RHA and BC were, 6.13% and 8.38%, respectively, lower than that of SC, which was 38.51%. The reactive element C contents of RHA, BC, and SC were 87.00%, 72.30%, and 52.18%, respectively. The C content in RHA and BC was higher, which resulted in their high denitration rate. Although the C relative content of SC was the least, the initial denitration rate of SC was much higher than those of RHA and BC, which means that the C element in this biochar is a better reducer of C-NO. This may result from the presence of a higher number of aromatic hydrocarbons and various reactive functional groups in SC. As for the O/C ratio which will inhibit the reduction of C-NO [17], the relative content in BC and SC was relatively high. This partly explain why the denitration rate decreased rapidly with time. RHA and BC contain a little amount of N element, which does not exist in SC. According to the literature [16][17][18], the N element induces the formation of NFGs, which promote C-NO reduction. The positive effect of NFGs on the reduction of biochar is attributed to two properties: NFGs promote C-NO reduction under aerobic conditions and inhibit the reaction of biochar with O 2 . In summary, the denitration rate decreases faster in the presence of a high O/C ratio, and the high relative content of N element can induce NFGs, which catalyze C-NO reduction.

XPS Results
According to the above experimental data, the optimum denitration rates of RHA, BC, and SC were 850 • C, 900 • C, and 900 • C, respectively. Under the protection of N 2 , the three materials were held at their optimum denitration temperature for 10 min, and then we tested the type and relative content of chemical bonds on the surface of their samples by XPS. XPS allowed the quantitative analysis at the micrometer-scale of the surface C1s of the samples, which contained both organic and inorganic C element; in contrast, EA tested the organic C element of the whole sample. Therefore, the rests of the two analyses were very different, as shown in Tables 3 and 4. The chemical bond species of the surface C1s of the three materials as well as their relative contents are shown in Figure 6 and Table 4. The relative content was obtained by dividing the area of each single peak area by the total peak area.

XPS Results
According to the above experimental data, the optimum denitration rates of RHA, BC, and SC were 850 °C, 900 °C, and 900 °C, respectively. Under the protection of N2, the three materials were held at their optimum denitration temperature for 10 min, and then we tested the type and relative content of chemical bonds on the surface of their samples by XPS. XPS allowed the quantitative analysis at the micrometer-scale of the surface C1s of the samples, which contained both organic and inorganic C element; in contrast, EA tested the organic C element of the whole sample. Therefore, the rests of the two analyses were very different, as shown in Table 3; Table 4. The chemical bond species of the surface C1s of the three materials as well as their relative contents are shown in Figure  6 and Table 4. The relative content was obtained by dividing the area of each single peak area by the total peak area.
The content of functional groups of C is determined by the formula in which Co is the original relative content determined by the XPS test, and C1s is total carbon content.   The content of functional groups of C is determined by the formula in which C o is the original relative content determined by the XPS test, and C1s is total carbon content. Figure 6 and Table 4 show that the three biochars contain multiple carbon-containing functional groups. The relative content of graphite carbon C-C bond (284.5~284.9 eV) in RHA was larger with respect to the content in the other two biochars, which may partly explain why the denitration rate of RHA decreased slowly with time at different reaction temperatures. However, due to their high contents of (NCO) x bonds (292.1-297.2 eV) which can catalyze the reduction reaction of C-NO and increase the denitration rate 1 [17,18], also BC and SC showed a high denitration rate, although less C-C bonds were observed in BC and SC than in RHA. According to published studies [12,[17][18][19], COOH (289.2-289.8 eV), an acid group in the reaction process, will inhibit the reduction reaction. For the relative content of C-O bonds (286.0-286.4 eV), BC has the highest, followed by RHA and lastly by SC. This order is consistent with the optimal denitration rates of the three biochars. According to the experimental study of Xia and Wan [17][18][19], C-O bonds are reactive sites and play a very important role in the reaction. On the one hand, the binding energy of the C-O bond is low, so it can easily break and participate in the reaction; on the other hand, the C-O bond can play a reducing role in the reaction process. There are other functional groups in SC, as shown by XPS of C1s, such as the C-N bond (285.0-286.1 eV), C=O bond (287.6-288.0 eV), and π-π bond (291.3 eV) [19,23]. The binding energy of the C-N bond is between those of the C-C and C-O bonds, so it could also be involved in the reduction of C-NO.
These data indicate that the C-C bond, (NCO)x bond, and C-O bond promote or catalyze the reduction reaction of C-NO, while the COOH bond has an inhibitory effect. The mechanism is presented in Figure 7. At high temperature in the presence of active gas, the surface of biochar contains a large number of chemical bonds that participate in the process of C-NO reduction, promoting or inhibiting the reduction of nitrogen oxides.
(289.2-289.8 eV), an acid group in the reaction process, will inhibit the reduction reaction. For the relative content of C-O bonds (286.0-286.4 eV), BC has the highest, followed by RHA and lastly by SC. This order is consistent with the optimal denitration rates of the three biochars. According to the experimental study of Xia and Wan [17][18][19], C-O bonds are reactive sites and play a very important role in the reaction. On the one hand, the binding energy of the C-O bond is low, so it can easily break and participate in the reaction; on the other hand, the C-O bond can play a reducing role in the reaction process. There are other functional groups in SC, as shown by XPS of C1s, such as the C-N bond (285.0-286.1 eV), C=O bond (287.6-288.0 eV), and π-π bond (291.3 eV) [19,23]. The binding energy of the C-N bond is between those of the C-C and C-O bonds, so it could also be involved in the reduction of C-NO.
These data indicate that the C-C bond, (NCO)x bond, and C-O bond promote or catalyze the reduction reaction of C-NO, while the COOH bond has an inhibitory effect. The mechanism is presented in Figure 7. At high temperature in the presence of active gas, the surface of biochar contains a large number of chemical bonds that participate in the process of C-NO reduction, promoting or inhibiting the reduction of nitrogen oxides.

Conclusions
In this work, by comparing the reduction of nitrogen oxides by three biochar and analyzing their physical and chemical morphology, we found that: (1) The structure of biochar will affect the process of O2 adsorption on the surface and its participation in the reaction. The larger the specific surface area, the lower the O2 concentration that allows to reach the optimal denitration rate of the biochar. (2) The elemental composition and type of surface chemical bonds play an important role in promoting or inhibiting the reduction of C-NO. The C element in the biochar promotes the reduction of C-NO. The denitration rate of biochar decreases faster as the O/C ratio increases, and the higher the relative content of N element, the larger the production of NFGs which

Conclusions
In this work, by comparing the reduction of nitrogen oxides by three biochar and analyzing their physical and chemical morphology, we found that: (1) The structure of biochar will affect the process of O 2 adsorption on the surface and its participation in the reaction. The larger the specific surface area, the lower the O 2 concentration that allows to reach the optimal denitration rate of the biochar. (2) The elemental composition and type of surface chemical bonds play an important role in promoting or inhibiting the reduction of C-NO. The C element in the biochar promotes the reduction of C-NO. The denitration rate of biochar decreases faster as the O/C ratio increases, and the higher the relative content of N element, the larger the production of NFGs which catalyze C-NO reduction.
The C-C bond directly affects the rate of denitration, and the (NCO) x bond and C-O bond have a positive effect on the reduction of nitrogen oxides by biochar. (3) The denitration rates and reaction conditions of the three types of biochar are different. Their specific surface area and surface morphology affect the reaction conditions, and the chemical elements and functional groups present in each one of them have an important effect on the denitration rate.
Over the past decade, environmental and air pollution issues have attracted more and more attention. Due to its unique structure and richness in surface chemical bonds, biochar, a reusable product of agricultural waste, has important applications in the adsorption and reduction of pollutants. The research in this paper shows that biochar is very effective in the reduction of nitrogen oxides. Therefore, the application of biochar has great potential for the elimination of harmful industrial gases and environmental protection.