LaOx modified MnOx loaded biomass activated carbon and its enhanced performance for simultaneous abatement of NO and Hg0

A battery of agricultural straw derived biomass activated carbons supported LaOx modified MnOx (LaMn/BACs) was prepared by a facile impregnation method and then tested for simultaneous abatement of NO and Hg0. 15%LaMn/BAC manifested excellent removal efficiency of Hg0 (100%) and NO (86.7%) at 180 °C, which also exhibited splendid resistance to SO2 and H2O. The interaction between Hg0 removal and NO removal was explored; thereinto, Hg0 removal had no influence on NO removal, while NO removal preponderated over Hg0 removal. The inhibitory effect of NH3 was greater than the accelerative effect of NO and O2 on Hg0 removal. The physicochemical characterization of related samples was characterized by SEM, XRD, BET, H2-TPR, NH3-TPD, and XPS. After incorporating suitable LaOx into 15%Mn/BAC, the synergistic effect between LaOx and MnOx contributed to the improvement of BET surface area and total pore volume, the promotion of redox ability, surface active oxygen species, and acid sites, inhibiting the crystallization of MnOx. 15%LaMn/BAC has the best catalytic oxidation activity at low temperature. That might be answerable for superior performance and preferable tolerance to SO2 and H2O. The results indicated that 15%LaMn/BAC was a promising catalyst for simultaneous abatement of Hg0 and NO at low temperature.


Introduction
With the implementation of the ultra-low emission and energy saving of coal-fired power plant plan since 2015 in China, electrostatic precipitator (ESP) or fabric filter (FF) system, wet flue gas desulfurization (WFGD) unit, and selective catalytic reduction (SCR) devices have been extensively adopted for corresponding pollutants abatement Zhao et al. 2019a;Gao et al. 2018a). Consequently, following SO 2 , NO x , and dust, mercury emission has triggered tremendous concerns due to its severe toxicity, high persistence, toilless biological fluidity, and strong biomagnification in the food chain and ecosystem after it is converted into more venomous methylmercury (Gao et al. 2018a;Xu et al. 2018a;Wang et al. 2017;Liu et al. 2019a). It is well recognized that elemental mercury (Hg 0 ), oxidized mercury (Hg 2+ ), and particle-bound mercury (Hg p ) coexist in coal-fired flue gas Yang et al. 2019a). Thereinto, Hg 2+ and Hg p can be readily captured by WFGD and ESP or FF, respectively (Xu et al. 2018a;Chi et al. 2017;Shan et al. 2019). However, Hg 0 (the major mercury form) is untoward to remove by existing environmental protection devices in consideration of its water insolubility and strong volatility (Gao et al. 2018b;Xu et al. 2016;Shen et al. 2018a;Jiang et al. 2018). Thus, the emphasis and difficulty of eliminating mercury pollution lie in controlling Hg 0 emission.

Responsible Editor: Santiago V. Luis
Lei Yi and Jinke Xie contributed equally to this work and should be considered co-first authors.
In response to increasing environmental consciousness and rigorous mercury emission regulations, plentiful technologies, including catalytic oxidation and adsorption, have been tremendously researched for Hg 0 removal in recent years (Gao et al. 2018a;Zhang et al. 2017a;Yang et al. 2019a). To date, activated carbon injection (ACI) for Hg 0 abatement is a commercial technology (Shen et al. 2018b;Zhao et al. 2018;Shi et al. 2019). However, it suffers from some intractable bottlenecks, such as potential secondary contamination, huge operating costs, tardy regeneration rates, and the value deterioration of fly ash (Zhang et al. 2017aZhao et al. 2018). Furthermore, controlling NO and Hg 0 emissions by utilizing SCR and ACI independently confronts several inevitable shortcomings such as large equipment investment, high land requirement, huge maintaining, and operating costs (Gao et al. 2018a, b;Chen et al. 2020). It is essential to use the ameliorative SCR catalyst to achieve efficient reduction of NO x and oxidation of Hg 0 without subjoining equipment for considering comprehensive benefits; various SCR catalysts have been abundantly investigated for this purpose Zhang et al. 2021). It is well accepted that vanadium-based SCR catalyst can make part Hg 0 oxidize to Hg 2+ , but the converting ability from Hg 0 to Hg 2+ is relatively limited in low chlorine flue gas (Jiang et al. 2018;Zhang et al. 2018;Chen et al. 2018). However, such catalyst has to operate at a temperature range of 300-400°C and that urges SCR unit to be placed upstream of desulfurization and dedusting devices where catalyst is readily impaired by SO 2 and dust (Jiang et al. 2018;Zhang et al. 2021). Moreover, V 2 O 5 itself poses certain threat to the environment and human health Zhang et al. 2021). Notably, to overcome the abovementioned deficiencies, it is significantly needful to exploit preferable cryogenic catalysts with outstanding demercuration and denitration efficiencies, realizing simultaneous removal of Hg 0 and NO by the existing gas purification devices.
Notably, in order to achieve these objectives, numerous novel catalysts, such as Mn-Ce/TiO 2 , CuO-MnO x /AC-H, La 0.8 Ce 0.2 MnO 3 , have been researched for simultaneous abatement of NO and Hg 0 (Shen et al. 2018b;Zhao et al. 2019a, b, c;Zhang et al. 2017b). Thereinto, manganese oxides (MnO x ) catalysts with splendid cryogenic performance have been extensively investigated for NO and Hg 0 removal in view of preeminent properties of Mn species, such as the nature of labile oxygen, outstanding redox properties, diversiform oxidation states, and high oxygen storage/ release capacity as well as environmental friendliness, abundant reserves, and cheap price (Zhao et al. 2019b;Xu et al. 2018b;Fan et al. 2018). Nevertheless, some MnO x -based catalysts especially unsupported ones often bore with several shortcomings, such as poor tolerance to SO 2 and H 2 O, low thermal stability, and little specific surface area. These shortcomings impeded their actual applications Fan et al. 2018). Furthermore, La 2 O 3 can be used as an effective promoter to improve the dispersion of active components to obtain a catalyst with high stability and activity (Shen et al. 2018b). Studies have shown that the addition of La promotes the low temperature activity to Hg 0 and NO (Gao et al. 2018b;Yang et al. 2018a). Therefore, the manganesebased catalyst modified by lanthanum species may exhibit excellent performance.
As shown in literature, numerous carbon-based catalysts with activated carbon/biomass activated carbon (AC/BAC) carriers not only exhibited good performances for Hg 0 and NO simultaneous abatement at low temperature, but also often demonstrated great resistance to SO 2 and H 2 O (Gao et al. 2018a;Ren et al. 2017). That preeminent manifestations were possibly attributed to the excellent physicochemical characteristics of carbonaceous materials, and the good SO 2 tolerance was related to the large surface areas and abundant oxygencontaining functional groups (Ren et al. 2017;Guo et al. 2015), while the high H 2 O resistance might be associated with hydrophobic property of carbon materials (Abdelouahab-Reddam et al. 2015;Joung et al. 2014). As we well know, traditional ACs mainly come from non-renewable and relatively expensive resources such as wood and coal. Meanwhile, China suffers from intractable challenges from handling renewable agriculture straws in enormous quantities every year, owing to lack of reasonable approaches. As a result, BACs derived from agriculture straws have drawn attention with excellent application prospect. Until now, carbon materials including BACs have stimulated their potential in simultaneous removal of NO and Hg 0 , due to its sustainability, adjustable surface chemistry, and outstanding surface area (Shen et al. 2018a;Li et al. 2019). In addition, BACs with hierarchical porous structure usually show the advantages of excellent gas absorption and rapid mass transfer, which were also in favor of simultaneous removal of NO and Hg 0 . Therefore, manganese oxides loaded on BAC derived from agricultural straw wastes may be a promising catalyst for Hg 0 and NO simultaneous removal at low temperature. To the best of our knowledge, few studies related to LaO x modified MnO x /BAC for simultaneous abatement of Hg 0 and NO have been reported in publications, in which the synergistic effect between MnO x and LaO x might have a positive role on aggrandizing performance and tolerance to SO 2 and H 2 O. Consequently, a battery of systematic tests is performed to elucidate its performance for NO and Hg 0 simultaneous abatement over LaMn/BAC catalysts in this work.

Materials preparation
The manufacture method of BAC carrier was detailedly presented in our previous paper (Gao et al. 2018a). Lanthanum nitrates or manganese acetates acted as the precursors of LaO x and MnO x , respectively. The preparation steps of La/BAC, Mn/BAC, and LaMn/BAC were shown below. First, lanthanum nitrates or manganese acetates were dissolved into moderate deionized water to generate corresponding precursor solutions. Second, desired BACs were soaked in aforesaid precursor solutions for 24 h. Third, impregnated samples dried at 105°C and therewith calcined at 500°C for 4 h under N 2 atmosphere. The atomic ratios of La/Mn in XLaMn/BACs were 1:4 based on our preliminary experiments, where X corresponded to the mass fraction of LaMn mixed oxides. Meanwhile, for comparison, XMn/BAC and XLa/BAC were synthesized by the same procedures.

Materials characterization
The ASAP2460 volumetric sorption analyzer (Micromeritics Instrument Corp., USA) was applied to determine the specific surface areas and pore parameters of samples. Scanning electron microscopy photographs were taken to analyze sample surface structures and morphologies on the Hitachi S-4800 analyzer (Hitachi Limited, Japan). The X-ray diffraction (XRD) results embodying component crystallinity and dispersivity were collected on a Bruker D8-Advance X-ray diffraction device. NH 3 -TPD and H 2 -TPR (the abbreviations for ammonia-temperature programmed desorption and H 2temperature programmed reduction, respectively) were carried out using the Tianjin Xianquan TP-5080 automatic chemical adsorption instrument. The element chemical composition and chemical state of samples were conducted on the K-Alpha 1063 X-ray photoelectron spectrometer (Thermo Fisher Scientific, USA).
Experimental setup and procedure Figure 1 displays the experimental apparatus diagram. In each test, 250 mg sample was placed in the reactor with a quartz tube (600 mm length × 10 mm inner diameter). The total flow rate of simulated flue gas (SFG) was sustained at 500 mL/min (gas hourly space velocity (GHSV) = approximately 80,000 h -1 ), which mainly consisted of 5%O 2 , 500 ppm NO, 100 μg/ m 3 Hg 0 (g), 500 ppm NH 3 , and N 2 .
The mercury conversion tests were carried out to distinguish mercury speciations (Gao et al. 2018a;Ma et al. 2015a), in which the outlet gas of the reactor could be divided into two freely switching branches: one got through 10% SnCl 2 + HCl aqueous solution to reduce Hg 2+ to Hg 0 for measuring total mercury (Hg out 0 T ), whereas the other passed 10% KCl aqueous solution to dislodge Hg 2+ in order to accurately measure Hg 0 . Therefore, E oxi denoting the oxidation efficiency of Hg 0 could be determined by Eq. (1). Similarly, E NO and E Hg expressing their removal efficiencies were respectively calculated by Eqs. (2) and (3). Meanwhile, NO and Hg 0 concentrations were measured by Germany MGA 5 flue gas analyzer and Russia Lumex RA-915 M mercury analyzer, respectively. In which, NO in and NO out respectively indicated inlet and outlet NO concentrations. Likewise, inlet Hg 0 concentration and outlet Hg 0 concentration were independently represented as Hg 0 in and Hg 0 out . Moreover, in order to lessen test errors, E NO and E Hg were the average data of several parallel tests, and their relative errors were confined to 5%.

Results and discussion
Characterization of samples

BET analysis
The average pore diameters, total pore volumes, and BET surface areas of modified BACs and virgin BAC are summarized in Table 1. Further, their pore size distribution curves and N 2 adsorption-desorption isotherms are displayed in Fig.  2. According to IUPAC, these samples shared type IV isotherm with slit shaped pores, implying the presence of H3 hysteresis loops (Chen et al. 2017). As shown in Fig. 2b, 15%La/BAC and 15%Mn/BAC exhibited bimodal peaks centered at around 2.4 nm and 3.9 nm, while other samples possessed unimodal and narrow peaks located at about 2.4 nm. That clearly indicated these samples owned most mesopores and some micropores, which could provide more active sites and inner surface areas for SCR reaction (Guan et al. 2012). Virgin BAC owned the largest total pore volume (0.485 cm 3 / g) and the highest BET surface area (745.935 m 2 /g). Nevertheless, both total pore volume and BET surface area demonstrated noticeable decreases after loading metal oxides, and the descending trend became more and more severe with the augment of metal oxide impregnation. Especially, 30%LaMn/BAC exhibited the smallest total pore volume (0.306 cm 3 /g) and the poorest BET surface area (474.657 m 2 /g). That could be interpreted as the deposition of metal oxides in preexisting pores and the destruction of thin pore walls in the impregnation and calcination processes (Ma et al. 2015a;Xie et al. 2015). It was well-known that increasing the loading value of metal oxides might arouse their agglomerated effect, which was in accordance with XRD and SEM results. It was worth mentioned that 15%LaMn/BAC revealed bigger total pore volume and BET surface area than those of 15%La/BAC and 15%Mn/BAC, which might be ascribed to a strong interaction derived between LaO x and MnO x ; the phenomenon indicated that the addition of LaO x expresses a certain promotion effect on the BET surface area of the catalyst, which promoted a progressive dispersion of crystallites (Liotta et al. 2013). Similar appearance was found in other report (Jampaiah et al. 2015).

SEM analysis
Figure 3 revealed these SEM micrographs of virgin BAC and modified BACs with different loading values. The dark smoothness areas demonstrated carbon en-riched zones, while light pelletizing areas represented the presence of metal oxides. As you could see clearly, the introduction of metal oxides changed overwhelmingly pristine surface properties of virgin BAC. As shown in Fig. 3b, with regard to 7.5%LaMn/BAC, only a handful of agglomerates were discovered, and most metal oxides were highly dispersed. However, the presence of large dark areas indicated that 7.5%LaMn/BAC's surface could not obtain full application, and additional catalytic sites could be further provided through enhancing metal oxides loading (Gao et al. 2018a). For 15%LaMn/BAC, abundant metal oxides scattered plenarily on the BAC surface, although some agglomerates were still found. With regard to 22.5%LaMn/BAC and 30%LaMn/BAC, apparent and even serious agglomerates emerged, respectively. Consequently, metal oxide accumulation got more and more worse, resulting in the disappearance of available pores and catalytic active sites, exerting negative effect on its catalytic activity. This phenomena were in line with BET results. Figure 4 depicts the XRD patterns of modified BACs and virgin BAC. For virgin BAC, two strong diffraction peaks located at about 26.60°and 44.46°were detected, which were deemed to be the characteristic peaks of BAC (Gao et al. 2018a; Xie et al. 2015). Nevertheless, they weakened and even vanished with loading LaO x or MnO x , indicating the presence of a strong interaction between BAC and metal species (Gao et al. 2018b;Du et al. 2018). Besides, the fact that the peak intensity of metal oxides was much stronger than BAC might be also responsible for this phenomenon. With regard to 15%La/BAC, the diffraction peaks corresponded to La 2 O 3 (JCPDS 24-0508) and La(OH) 3 (JCPDS 05-0602) were found (Xu et al. 2016;Tang et al. 2004 and MnO x , thus resulting in smaller crystal sizes of metal oxides and higher surface areas due to the modification of LaO x , which might be favorable for catalytic reactions (Gao et al. 2018a;Li et al. 2012).

H 2 -TPR analysis
It was noteworthy that the redox properties of catalysts usually played crucial roles in catalytic activities. The H 2 -TPR profiles of modified BACs and virgin BAC are illustrated in Fig.  5. As regards virgin BAC, the distinct reduction peak approximately at 680°C might be reasonably assigned to the gasification of BAC , which also existed in other Fig. 2 The N 2 adsorption/ desorption isotherms (a) and corresponding pore size distribution curves (b) of virgin BAC and modified BACs samples. As for modified BACs, the introduction of metal oxides would generate lattice oxygen. As a result, the broad higher peak at 675°C of modified BACs might be partly attributed to the reduction of lattice oxygen (Zhao et al. 2016;Lian et al. 2014). Therefore, the peak at round 700°C was greatly enhanced for modified BACs due to the superposition of BAC gasification and lattice oxygen reduction. With regard to 15%Mn/BAC, three remarkable reduction peaks located at 335°C, 439°C, and 517°C were observed in the low temperature range, which might belong to the stepwise reduction of MnO 2 → Mn 2 O 3 → Mn 3 O 4 → MnO (Ma et al. 2015a;Boningari et al. 2015;Zhang et al. 2014). By contrast, these reduction peaks of 15%LaMn/BAC slidden into lower temperature regions, which were discovered at 323°C, 425°C, and 505°C, respectively (Boningari et al. 2015;Zhang et al. 2014;Li et al. 2017). That suggested 15%LaMn/BAC obtained better redox ability than 15%Mn/BAC, which was beneficial to catalytic activity through introduction of LaO x Li et al. 2017). Meanwhile, these profile shifts might implicate the presence of La-Mn complexes, which was largely due to a synergistic effect stemmed from La species and Mn species (Gao et al. 2018a;Li et al. 2010). Thus, it was reasonable to infer that 15%LaMn/BAC embodied better catalytic activity than 15%Mn/BAC .

NH 3 -TPD analysis
The surface acidity properties of modified BACs and virgin BAC were estimated by NH 3 -TPD. As shown in Fig. 6, virgin BAC exhibited a feeble peak at high temperature range, indicating that it also had a handful of acid sites for NH 3 adsorption and activation. The NH 3 desorption peak at high temperature range dramatically increased with the introduction of La, indicating that the introduction of La could significantly enhance its amount and strength of acidic sites. This phenomenon was plausibly assigned to that La could adjust the intensity distribution and amount of acid sites, stemming from its lanthanide contraction and 4f orbitals without full electron occupancy (Zhan et al. 2014;Liu et al. 2017), in which medium temperature peaks were in line with Brønsted acid sites, while high temperature peaks belonged to Lewis acid sites (Lónyi et al. 1996;Gu et al. 2010;Fang et al. 2018).
It was conjectured that Brønsted acid sites might be stemmed from surface hydroxyl groups (Gu et al. 2010;Ma et al. 2015b), while Lewis acid sites possibly originated from unsaturated metal sites (Gu et al. 2010;Zhu et al. 2017). Meanwhile, the intensity and areas of desorption peaks were associated with the strength and amount of acid sites (Gao et al. 2018b), which followed a descending trend: 15%LaMn/BAC > 15%La/BAC > 15%Mn/BAC > virgin BAC. Even more noteworthy was the fact that 15%LaMn/ BAC displayed the most acid sites for NH 3 adsorption and its succedent activation, thus boosting SCR activity, which was largely due to a synergistic effect stemmed between La Hence, it was reasonable to infer that 15%LaMn/BAC could behave the best NO removal efficiency. Figure 7 elucidates the XPS spectra of O 1s, Mn 2p, La 3d, and Hg 4f. The O 1s XPS spectra of uncirculated and used 15%LaMn/BAC were rendered in Fig. 7a, which were deconvoluted into three components centered 529.8~530.1 eV, 531.3~531.4 eV, and 532.7~532.9 eV. The lower binding energy peaks were put down to lattice oxygen (O a ), the medium binding energy peaks were regarded as weakly bonded oxygen and/or chemisorbed oxygen (O β ), and the higher binding energy peaks represented adsorbed water species and/or hydroxyl groups (O γ ) . However, virgin BAC only exhibited two kinds' peaks of O 1s; one peak was located at 531.61 eV, and the other one was sited at 533.9 eV. The former peak could be ascribed to O β , while the latter one might correspond to either-type oxygen, which was the primary product for carbonate-based electrolytes, stemming from reaction/reduction of carbonate-based solvents (Ma et al. 2015a;Veith et al. 2012;Freunberger et al. 2011). It was easy to accept that the introduction of metal oxides could provide abundant O α for modified BACs, which was beneficial for oxidating NO to NO 2 to boost E NO at low temperature (Kang et al. 2007;Boningari et al. 2015). In addition, the interaction between LaO x and MnO x could contribute to more oxygen vacancies and chemisorbed oxygens, which facilitated Hg 0 adsorption and oxidation . Combined with the analyses of BET and SEM, the introduction of metal oxides and its precursors (metal nitrates) could cover the surface of virgin BAC in the calcination and foregoing loading processes, resulting in the formation of lattice oxygen and disappearance of either-type oxygen (Gao et al. 2018a;Ma et al. 2015a Fig. 7b, the Mn 2p regions of uncirculated and used 15%LaMn/BAC consisted of two main peaks including Mn 2p3/2 and Mn 2p1/2 centered at around 641.8 eV and 653.5 eV, respectively. Moreover, the latter could be deconvoluted into three peaks at 641.1-641.5 eV, 642.5-643 eV, and 644.6-644.7 eV, which were ascribed to Mn 2+ , Mn 3+ , and Mn 4+ , respectively Zhang et al. 2014;Li et al. 2015c). It could be seen that the proportion of Mn 4+ descended distinctly from 23.7 to 19.1%, while the percentage of Mn 3+ enhanced slightly from 21.2 to 21.5%, and the ratio of Mn 2+ increased from 20.9 to 25.5%, indicating that some Mn 2+ generated from consumed Mn 4+ in reactions. It was demonstrated that Mn 4+ species and its redox cycle could boost the oxidation process from NO to NO 2 , which was beneficial for both NO conversion and Hg 0 oxidation (Boningari et al. 2015;Zhang et al. 2014;Li et al. 2015c). Furthermore, it was speculated that the large-span valence change from Mn 4+ to Mn 3+ and whereafter to Mn 2+ over Mn-based catalysts might be beneficial to NO and Hg 0 elimination; meanwhile, Mn 4+ was deemed to the most active species (Yang et al. 2018b;Liu et al. 2019b).

XPS analysis
The La 3d XPS spectra of uncirculated and used 15%LaMn/BACs are illustrated in Fig. 7c. Two essentially identical doublet peaks including main peaks and its satellite peaks were observed, where the peaks at 834.1 eV (834.4 eV) were the main peaks for La 3d5/2, 838.1 eV (838.3 eV) for the satellite peaks of La 3d5/2, and the peaks located at 851.1 eV (851.2 eV) were the main peaks of La 3d3/2, 855.0 eV (855.3 eV) for its satellite peaks of La 3d3/2 (Tholkappiyan and Vishista 2014). Meanwhile, the main peaks ascribed to 3d 9 4f 0 arose from the spin-orbit interaction whereas their satellite peaks due to the 3d 9 4f 1 final state were attributed to the electron transfer between the empty La 4f orbit and the oxygen valence (Dudric et al. 2014). The appearance of satellite peaks located at a higher binding energy of approximate 4 eV was a diagnose feature for confirming the presence of La 3+ compounds coordinated by other ligands (Rudyk et al. 2011). The energy separations between La 3d5/2 and La 3d3/2 core levels were approximately 17 eV, which was inconsistent with that of standard La 2 O 3 (Tholkappiyan and Vishista 2014). Moreover, the binding energy value of La 3d5/2 was 834.1 eV (834.4 eV), which was higher than that of bulk La 2 O 3 with the binding energy value of 831.9 eV (Rudyk et al. 2011). In addition, these binding energy values of La 3d were in concert with that in perovskites (Gao et al. 2018b;Rudyk et al. 2011;Blanchard et al. 2010). These phenomena indicated the transformation of La 2 O 3 into a perovskite-type structure of LaMnO x (Craciun and Dulamita 1997), which was in well accordance with H 2 -TPR analysis. It means that the addition of LaO x can control the morphological structure of the catalyst.
The Hg 4f XPS spectra of used 15%LaMn/BAC are presented in Fig. 7d; four peaks centered at 99.9 eV, 101.44 eV, 102.65 eV, and 104.6 eV were recorded. The medium peak at 102.65 eV corresponded to Si 2p (Tao et al. 2012). The weakest peak at 99.9 eV was attributed to the characteristic peak of adsorbed Hg 0 (Tao et al. 2012;Hutson et al. 2007). The strongest peak located at 104.6 eV belonged to Hg 4f5/2, and the other medium peak at 101.44 eV represented Hg 4 f7/ 2, which was associated with HgO (Tao et al. 2012;Hutson et al. 2007;Li et al. 2015b). As illustrated in Fig. 8, mercury conversion tests demonstrated that catalytic oxidation with the product of HgO and adsorption worked together for Hg 0 removal, in which their contributions varied with reaction time.

The performance of samples
Effect of active ingredient Figure 9a, b reveals the effect of active ingredients on E NO and E Hg at 60~340°C, respectively. It was clearly to see that both reaction temperature and active ingredients exerted significant influences on NO and Hg 0 removal. Apparently, E NO and E Hg enhanced significantly after loading Mn or La species onto virgin BAC, indicating that active ingredients were conducive to NO and Hg 0 removal. Particularly, 15%LaMn/BAC exhibited preferable E NO and E Hg than those of 7.5%LaMn/BAC and 30%LaMn/BAC. Besides, 15%LaMn/BAC also yielded better performance and broader active temperature window compared with 15%Mn/BAC and 15%La/BAC, which exerted the highest E Hg of 100% at 100~180°C and the best E NO of 88.6% at 220°C with a second-best E NO of 86.7% at 180°C. The former appearance demonstrated that active ingredients could not always display decisive role in NO and Hg 0 removal, which might be also affected by BET surface areas and total pore volumes (Gao et al. 2018a, b). The later appearance could be assigned to a synergistic effect originated from La and Mn species derived from the LaO x modified Mn supported catalyst, which contributed to smaller crystal sizes and better dispersion of metal oxides, bigger BET surface area, bigger total pore volume, higher redox ability, and more acid sites, and those properties were responsible for better performance and broader active temperature window Liotta et al. 2013;Zhang et al. 2014). Notably, the E Hg of La/BAC and Mn/BAC significantly decreased after 220°C in Fig. 9b, which was probably ascribed to their relatively poorer catalytic activity for Hg 0 oxidation than that of LaMn/BAC, due to Hg 0 oxidation dominating for Hg 0 removal at that time since this samples had suffered from uninterrupted tests of 20 h. Simultaneously, incorporating La 2 O 3 into MnO x lattice could generate more bulk oxygen species and oxygen vacancies with high mobility, which were beneficial to Hg 0 oxidation. Hou et al. 2020). Similar appearances were also discovered in previous works (Gao et al. 2018a, b).
In addition, E NO and E Hg of modified BACs showed similar trends with increasing reaction temperature, respectively. Meanwhile, E Hg exhibited a slight increase in the temperature range of 60 to 180°C and whereafter declined dramatically with further elevating reaction temperature, whereas E NO yielded an apparent increase from 60~220°C and afterwards displayed a very slight dip except for 15%La/BAC. That phenomenon demonstrated that elevating reaction temperature often boosted Hg 0 and specially NO abatement until a certain Fig. 8 The results of mercury conversion tests over 15%LaMn/ BAC temperature, after which further enhancing reaction temperature exerted negative effects on Hg 0 and NO abatement. With respect to 15% LaMn/BAC, E NO aggrandized from 78.5 to 88.6% until reaction temperature augmented from 60 to 220°C , and further increasing reaction temperature resulted in E NO declining to 83.8% at 340°C, while E Hg reached 100% from originally 98.3% at 60°C and maintained 100% removal efficiency at 100~180°C, and therewith, it decreased slightly from 100 to 90.1% at 340°C. It was recognized that increasing reaction temperature would boost catalytic activity at low temperature owing to enhancing catering activation energy and chemisorption stemmed from generating more chemical bonds (Tao et al. 2012;Zeng et al. 2004). Moreover, the manifest declines of E Hg and E NO at high temperatures might be interpreted by two reasons. On the one hand, the adsorption of reactant molecules like Hg 0 onto adsorption sites would be inhibited by high temperature (Li et al. 2011). On the other hand, the structures of carbon-based catalysts might be destroyed by high temperature due to active ingredients catalytic oxidizing carbon matrix (Gao et al. 2018a, b;Lu et al. Fig. 9 Effect of active ingredients on simultaneous NO and Hg 0 removal at 60~340°C. a NO. b Hg 0 2010). It was noteworthy that 15%LaMn/BAC respectively exhibited 100% and 86.7% removal efficiency for Hg 0 and NO at 180°C, indicating it has excellent application potential at low temperature compared with 15%Mn/BAC without LaO x modification.

Effect of flue gas components
The effects about flue gas components on NO and Hg 0 simultaneous abatement over 15%LaMn/BAC are presented in Fig.  10. It could be seen that both E Hg and especially E NO declined dramatically when 5% O 2 was removed from SFG. The poor performance might be profited from preexisted O α and O β on the sample surface (Gao et al. 2018a, b;Xie et al. 2015), as demonstrated in the O 1s XPS analysis. Nevertheless, compared with that under SFG, additional 5% O 2 joining the SFG engendered negligible influences on NO and Hg 0 removal. This phenomenon indicated that O 2 could play a positive effect on E NO and E Hg under oxygen-poor conditions. It was well-known that O α and O β would be expended in NO reduction and Hg 0 oxidation reactions; meanwhile, active metal species were reduced in that processes. Noteworthily, gaseous O 2 could supplement consumed O α and O β through oxidizing aforesaid reduced metal species, thus facilitating these reactions continuing (Bueno-López et al. 2005;Grabowski et al. 2002).
As shown in Figs.10 and 11, both SO 2 and H 2 O restrained E NO and E Hg , and the synchronous presence of SO 2 and H 2 O aroused more serious declines of E Hg and E NO than that under separate SO 2 and H 2 O. The prohibitive influences of SO 2 and H 2 O could be interpreted by several reasons. First of all, the existence of SO 2 and H 2 O might compete with Hg 0 , NH 3 , NO, and O 2 for adsorption and catalytic sites (Li et al. 2011(Li et al. , 2008Casapu et al. 2009). Secondly, the emerging ammonium sulfates or bisulfates could cover activated sites and destroy porous structure; additionally, the possible generating metal sulfates stemmed from metal oxides reacting with SO 2 might lead activated adsorption or catalytic sites to inactive ones (Li et al. 2015b), thus suppressing E NO and E Hg . Most notably, 15%LaMn/BAC exhibited better tolerance to SO 2 and H 2 O than that of 15%Mn/BAC in preliminary experiments because of introduction of LaO x , which might be ascribed to bigger BET surface area and higher metal oxides dispersion as well as other advantageous synergistic effects between LaO x and MnO x species .

The interaction between NO removal and Hg 0 removal
It was indispensable to inspect the possible interaction effects between NO removal and Hg 0 removal allowing for practical application. As illustrated in Fig. 12, E NO seemed hardly changed when 100 μg/m 3 Hg 0 was suddenly removed from SFG, and E NO only manifested some indistinctive changes even if additional 100 μg/m 3 Hg 0 was added into SFG. That suggested Hg 0 removal exerted almost no impact over NO removal, which might be explained by that Hg 0 concentration was too small to have little impact on NO removal. On the contrary, E Hg elevated remarkably when 500 ppm NO and 500 ppm NH 3 were concurrently subtracted from SFG, indicating NO removal displayed detrimental influence on Hg 0 removal. It was inferred that NO removal might take precedence of Hg 0 removal in high NO and NH 3 concentrations (Zhang et al. 2017c;Niksa and Fujiwara 2005).
To further investigate the separate effect of NO and NH 3 on Hg 0 removal, systematic tests were performed, as presented in Fig. 12. E Hg manifested an obvious drop when 500 ppm NO was precluded from the SFG, and the similar result was also Fig. 10 The effects of flue gas components on NO and Hg 0 simultaneous removal over 15%LaMn/BAC observed when additional 500 ppm NH 3 was joined in the SFG. Clearly, that indicated NH 3 had evidently suppressive influence on Hg 0 abatement, which was probably assigned to rapidly adsorbed NH 3 occupying some active sites; therewith, part active oxygen was expended, thus restraining Hg 0 removal (Chen et al. 2017;Qi et al. 2004;He et al. 2016). Moreover, E Hg exhibited significant increase when 500 ppm NH 3 was precluded from SFG, and the positive appearance was also detected when additional 500 ppm NO was added into the SFG, which demonstrated NO displayed promotional role on Hg 0 removal. That appearance might be attributed to the fact that some weakly adsorbed NO could be oxidized to NO 2 , which was in favor of Hg 0 oxidation (Zhao et al. 2019b;Li et al. 2010).

Mechanism exploration
According to the analyses of the experimental results, the feasible mechanism of simultaneous removal of NO and Hg 0 Fig. 11 The performance for NO and Hg 0 simultaneous abatement over 15%LaMn/BAC under the conditions with 300 ppm SO 2 and 3% H 2 O Fig. 12 The interaction between NO removal and Hg 0 removal over 15%LaMn/BAC on LaMn/BACs was speculated. The total low-temperature activity of the catalyst was facilitated as the introduction of La species (Wang et al. 2019a), which is consistent with the results of H 2 -TPR. It was indicated that La 2 O 3 promoted the electron movement between Mn 4+ , Mn 3+ , and Mn 2+ attributed to the synergistic effect emerged between La and Mn. The XPS characterization of Mn 2p demonstrated that the conversion of Mn 4+ to Mn 2+ was facilitated under the catalytic reaction; LaO x might existed as an electron promoter in the catalyst. This alteration in the valence state of manganese species on the catalyst is a possible mechanism for NO reduction and Hg 0 oxidation (Yang et al. 2018a;Liu et al. 2019b;Yang et al. 2019b;Gao et al. 2019). The possible NH 3 -SCR reaction process of NO could be summarized as follows: gaseous NH 3 was adsorbed on the Lewis and Brønsted acid sites on the surface of the catalyst, and formative coordinated NH 3 and NH 4 + reacted with gaseous NO and adsorbed NO 2 via the Langmuir-Hinshlwood mechanism to convert into innocuous N 2 and H 2 O (Zhang et al. 2017a;Zhao et al. 2018;Fu et al. 2014). Moreover, we detected tiny amount of CO 2 and CO in the outlet gas, as shown in Eqs. (19)-(21) (where KO was denoted as lanthanum-manganese composite metal oxide). Notably, this phenomenon of some side reactions might occur during the NH 3 -SCR process on the carbon-supported metal oxide because of the carbon support as reducer had also been reported in other studies (Shen et al. 2018b;Lu et al. 2010). Notwithstanding, the dominance of reactions (8)-(18) relative to reactions (19)-(21) was not affected. The specific reaction formulas were as follows: The conversion pathway of Hg 0 on the catalyst prepared in this work can be divided into adsorption including physisorption, chemisorption, and catalytic oxidation, which have also been notarized in our previous researches (Gao et al. 2018a, b). As shown in Fig. 8, the contribution of the two modes of mercury conversion varied with the reaction time. It was potty to observe from Fig. 8 that the contribution of catalytic oxidation was relatively small at the beginning, but Hg 0 oxidation prevailed gradually with the increase of reaction time. The oxidation of Hg 0 was ascribed to the Mars-Maessen mechanism under the atmosphere of N 2 + O 2 that both O α and O β participated in Hg 0 oxidation (Zhao et al. 2019b;Wang et al. 2019b). O α and O β were easy to combine with adsorbed Hg 0 to form HgO (Zhao et al. 2019c) (as shown in Eq. (23)). However, the oxidation of Hg 0 was mainly attributed to the consumption of O α , which was also consistent with the analysis results of XPS O 1s (Gao et al. 2018a;Shen et al. 2018b). Therefore, the oxidation path of Hg 0 was speculated that Hg 0 (g) was first adsorbed on the sample surface and oxidized with O α and O β to generate HgO (ad) . Nevertheless, part of HgO (ad) might be converted into HgO (g) . The reaction process was summarized as follows: Furthermore, it is very valuable to explore the possible mechanisms about the effect of gas components on Hg 0 removal. Figure 10 indicates that both H 2 O and SO 2 had inhibitory effects on the removal of Hg 0 . The inhibitory effect of H 2 O was due to its competitive adsorption with Hg 0 , hindering the adsorption of Hg 0 Xu et al. 2016). Similarly, SO 2 also possessed the competitive adsorption with Hg 0 . In addition, SO 2 not only reduced HgO (ad) to Hg 0 (g) but also reacted with active metal oxides MnO and Mn 2 O 3 to form  2with the participation of O β , which caused the active sites to be covered and reduced the removal efficiency of Hg 0 (Liu et al. 2021;Zhao et al. 2017;Zhang et al. 2017d). Therefore, the possible reaction of SO 2 could be described as follows; Eqs. (25)-(27) were shown. Moreover, Fig. 12 reveals that NH 3 had an adverse effect on E Hg . On one hand, this might also be related to the competitive adsorption between NH 3 and Hg 0 . On the other hand, NH 3 was adsorbed on the sample surface and reacted with SO 3(ad) or H 2 O (ad) to form (NH 4 ) 2 SO 4 or NH 4 HSO 4 crystals, which would block the microporous structure of the sample, thereby reducing the removal efficiency of Hg 0 (as shown in Eq. (28)) (Xu et al. 2016Sun et al. 2020). On the contrary, NO showed a promotional effect on the removal of Hg 0 ; it could be attributed to the NO 2 /NO active species formed by the oxidation of NO via Eqs. ( + 29)-(30), which would promote Hg 0 oxidation ). In summary, the possible mechanisms about the effect of gas components on Hg 0 removal could be described as follows: MnO2 Conclusions A battery of LaMn/BACs adopting a facile ultrasoundassisted impregnation method was prepared for simultaneous abatement of NO and Hg 0 . 15%LaMn/BAC manifested excellent performance for NO and Hg 0 simultaneous removal and superior resistance to SO 2 and H 2 O at low temperature due to bringing LaO x in the catalyst; meanwhile, it exhibited splendid Hg 0 removal efficiency (100%) and extraordinary NO removal efficiency (86.7%) at 180°C. The interaction between Hg 0 removal and NO removal was explored; thereinto, Hg 0 removal had no influence on NO removal, while NO removal preponderated over Hg 0 removal. The accelerative effects of NO and O 2 on Hg 0 removal could not offset the inhibitory influence of NH 3 . SEM, XRD, BET, H 2 -TPR, NH 3 -TPD, and XPS were exploited to characterize the physicochemical properties of relevant samples. After incorporating suitable LaO x into 15%Mn/BAC, the synergistic effect between LaO x and MnO x emerged in 15%LaMn/BAC, which contributed to the enhancement of BET surface area and total pore volume, the improvement of redox ability, surface active oxygen species, and acid sites, suppressing the crystallization of MnO x . That might be answerable for superior performance and preferable tolerance to SO 2 and H 2 O under the modified effect of LaO x . The results indicated that 15%LaMn/BAC was a promising catalyst for simultaneous abatement of Hg 0 and NO at low temperature. Data availability The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Declarations
Ethics approval and consent to participate Not applicable.
Consent for publication Not applicable.

Competing interests
The authors declare no competing interests.