Introduction

With the advancement of large-scale industrialization, much fossil energy coal is consumed, producing many air pollution gases. Among these pollutants, nitrogen oxides (NOx, including NO and NO2) are one of the severe air pollutants. The emission of NOx contributes to various environmental problems, such as fine particle matter and ozone (PM2.5-O3) compound pollution, the greenhouse effect, smog, and acid rain formation. Several technologies exist to purify NOx, selective catalytic reduction with ammonia (NH3-SCR) being the most extensively adopted and mature denitration method1, which selectively reacts NOx with NH3 to produce nitrogen and water. V-based catalysts are widely used in the thermal power industry because of their high nitrogen selectivity and strong sulfur poisoning resistance. The catalysts exhibit excellent performance within the mid-to-high temperature range (300–450 °C). However, their activity diminishes significantly at ultra-low temperatures (<150 °C), which hinders their application in treating low-temperature flue gas. Given the biological toxicity and hazardous waste characteristics of the V compound, how to enhance the performance of V-based catalysts with lower V content and lower temperature has emerged as a primary research focus2. He et al.3 found that the rate-determining step of NH3-SCR at low temperatures was NH3 dehydrogenation (corresponding lattice oxygen release process), so how to promote lattice oxygen activation at low temperatures was still a scientific research problem.

The auxiliary strategy of the external field (electricity, light, magnetism) can improve the performance of the corresponding catalysts4,5,6, in which the electric auxiliary strategy has better controllability, simpler operation, and a wider energy source. In recent years, Mei et al.7 successfully employed an electric-assisted programmed oxidation strategy to lower the ignition temperature of soot significantly. The strategy directly ignited soot through joule heat caused by current without external heating. This innovative approach, which utilized internal electric heating, provided a flexible, energy-efficient, effective means of thermal control and could enhance catalysis performance through electronic effects. Excitingly, the electric-assisted strategies had also found applications in other reactions, including hydrogen production through methane reforming8, the methanation of CO29, and water-gas shift10. Furthermore, various power sources were often available in NOx purification scenarios, such as the power plant itself, the self-provided power station of steel mills, and the self-provided power supply of diesel vehicles. Therefore, the electric-assisted NH3-SCR strategy had a unique operational advantage. The work put forward and studies the application of this strategy in the purification of NOx on the V-based catalysts with low V content and low temperature, which was of great significance to the development of new energy in the field of air pollutant prevention and control.

So far, this is still a challenge to study the weak current-assisted (WCA) catalysis deeply due to the lack of characterization means and insufficient understanding. In this study, our primary objective was to investigate the impact of the WCA (or called current-assisted, CA) strategy on V-based catalysts. Initially, the N-type semiconductor antimony tin oxide (ATO) with high conductivity and reducibility was used as the carrier for the NH3-SCR catalyst. A small amount of VOx was loaded on the ATO, serving as the primary active component. Subsequently, the acidic metals (Nb, Mo, W) were to augment the number and strength of acidic sites11,12,13 within the catalyst and ultimately obtained the optimized 2VNA catalyst. Different from the traditional electrocatalysis and NEMCA (non-faradaic electrochemical modification of catalysis activity) technology, the WCA strategy in this work only needs to apply electric current to the catalyst without complex reactor structure, and its core lies in the weakening of M-O chemical bond of the catalyst itself and the activation of molecular oxygen by flowing electrons. In the WCA strategy, the used catalyst has a high conductivity and a low potential difference between the two ends (<10 V at the laboratory level, which is beneficial to industrial application), corresponding to a current usually below 1 A. The catalyst carried out NH3-SCR by increasing the current without the external thermal field, and then the electrical and thermal effects in the electrification process were qualitatively distinguished. The results showed that the WCA strategy was very effective in the NH3-SCR catalytic system with low V content (1.36%), and NOx could be wholly converted at ultra-low temperature (T99 < 150 °C). Various methods were employed to characterize the overall morphology and material structure of the catalyst, in which high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) and X-ray absorption spectroscopy (XAS including XANES and EXAFS) together revealed that V atoms existed in the highly dispersed monoatomic states. The step-by-step exploration in mesoscopic and microscopic methods was to delve into the mechanism of the WCA strategy. At the mesoscopic level, the WCA-NH3-TPD and WCA-H2-TPR were used to study the NH3 adsorption and redox capacity of the catalyst. The oxygen transient experiments further characterized the enhancement catalysis of weak current on the oxidation-reduction stage of the catalytic process. Notably, the isotope oxygen exchange experiments, in situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) experiments, and kinetic studies were used to elucidate the mechanism of the strategy. At the microscopic level, the effects of electron migration on binding energy and chemical bonds were studied by the in situ XPS experiments and density functional theory (DFT) calculations. That deeply revealed the mechanism of WCA enhancement catalysis. In summary, the work confirmed the significant role played by the WCA strategy in enhancing ultra-low-temperature activity and elucidated the catalytic mechanism under the WCA strategy from the perspective of electron behavior and chemical bonds.

Results and discussion

Characterization of the catalyst optimization process

According to the principle of double sites14 (acid site and redox site), the intrinsic thermal catalysis performance was optimized. Specifically, by adjusting the VOx content and introducing the modified component (Nb), the 2VNA low-temperature denitration catalyst with the optimized thermal catalytic performance (Fig. S3) was obtained, in which the atomic mass percentage of V and Nb elements was 1.36 and 0.64 (Table S1) respectively. On the whole, after loading VOx and introducing the modified component (Nb), the specific surface area increased, the pore size decreased, and the surface defects increased on the premise of maintaining the original nanostructure of the catalyst (Fig. S4–S6 and S7a, b, and Table S2, S3). In the optimization process, introducing NbOx could optimize the acid site distribution of the catalyst while having no noticeable effect on the intrinsic oxidation ability of the catalyst (Fig. S8). Further characterization showed that sequential loading of V and Nb elements could promote the redox ability of the catalyst when the catalyst was in the conducting state (Fig. S9), which corresponded to the charge change of the valence distribution of Sn element due to the introduction of V and Nb elements15 (Fig. S7c).

Identification of V atoms sites on 2VNA (110)

The 2VNA was comprehensively characterized and observed, and it was observed that the V atoms were highly dispersed on the SnO2 (110) surface in the form of monoatomic states16,17. First, the dispersion state of V elements was characterized. Figure 1a shows that V atoms are highly distributed on the catalyst surface, and the quantification of each element is shown in Fig. S10 and Table S4. Moreover, the coordination of V on the catalyst was further explored. During the high-temperature calcination, Sb5+ ions (0.62 Å) could easily replace Sn4+ ions (0.69 Å) in the SnO2 lattice. The substitution resulted in ATO exhibiting superior conductivity compared to pure SnO2. Theoretically, V5+ ions (0.54 Å) could similarly replace Sn4+ ions (0.69 Å). The EXAFS, XANES, and HAADF-STEM were to verify the supposition. Analyzing the V K-edge Fourier transform (FT) EXAFS spectrums, Fig. 1b and Table S5 show that the first low-amplitude vibration on the left side corresponds to V = O, CN = 1. The second high amplitude corresponded to an average V-O distance of 2.0 Å, CN = 5 (inset of Fig. 1b). Two-dimensional wavelet transform (WT), which encapsulated three-dimensional information, could provide more precise signals and intuitive explanations than FT EXAFS spectrum18,19,20. Through the FT, the K space fitting of the 2VNA catalyst with standard samples (Fig. 1c) was transformed into R space. Figure 1h shows the R space WT EXAFS spectra between the catalyst with standard samples, and it could be known that the highest intensity at 7.0 Å−1 corresponded to the V-O coordination.

Fig. 1: Study on the fine structure of catalyst and the position of V atom.
figure 1

a HAADF TEM image and EDS mappings of the 2VNA catalyst. Scale bar, 2 nm; Scale bar in the EDS mapping images, 2 nm. b k3-weight FT EXAFS spectrum of the 2VNA catalyst. The dashed lines represent the V-O scattering paths. Inset: sketched structure of the V atom. c V K-edge XANES spectrums of the 2VNA, V foil, VO2, and V2O5. d Side-view HAADF-STEM image of the 2VNA catalyst. Inset: the ball-stacked model in the yellow dotted box. Scale bar, 1 nm. e, f Enlarged view of the red dotted box in d. The top-view and side-view structure model is shown in the upper (e). f1 and f2 give the atomic contrast curves at the green boxes, and V atoms anchored on 2VNA (110). g Characterization of valence distribution of vanadium by XPS. h V K-edge WT EXAFS spectrums of the 2VNA, V foil, VO2 and V2O5.

Full size image

The HAADF-STEM was used to directly observe the atoms anchored on the SnO2 (110) surface. A slight difference in atomic mass between Sb, Sn, and Nb atoms, distinguishing them based on atomic mass alone was not feasible. Therefore, when the electron beam was oriented parallel to the SnO2 (110) surface, the side-view HAADF-STEM image21 was captured. As demonstrated in Fig. 1d, the actual lattice spacing aligned with the theoretical value, indicating that the structure of the 2VNA catalyst remained consistent with the structure of SnO2. To further investigate the atomic arrangement, comparing the atomic intensity curve of the outermost layer with that of the sub-outer layer revealed that the V atoms in the outermost layer had replaced the Sn atom (Fig. 1e, f). Figure 1g shows the valence state distribution of the V element, primarily indicating a V5+ oxidation state, which indirectly confirmed that V was similar to Sb, and both replaced Sn with +5 valence.

Excellent catalytic activity under the WCA strategy

The catalytic performance under the WCA strategy in low-temperature NH3-SCR was investigated. The infrared thermal imaging was to confirm the accuracy of the temperatures measured by the thermocouple. As demonstrated in Fig. 2a, the temperatures recorded by infrared thermal imaging aligned with the actual temperatures measured by the thermocouple, verifying their consistency (Fig. S11). Moreover, comparing the WCA strategy activity of the ATO carrier revealed that V element loading could immensely enhance the catalysis performance (Fig. S12). To further evaluate the enhancement catalysis in the 2VNA catalyst under the WCA strategy, the NOx conversion of the two strategies (electrifying under the WCA strategy and mere heating) was compared. The red-shaded part indicated the enhancement catalysis of the weak current (ECWC) except for the thermal effect (Fig. 2b). As shown in Fig. S13, although the entire electrification process exhibited electrothermal synergistic catalysis7,10,22,23, the dominant role was played by the ECWC, which confirmed the effectiveness of the WCA strategy. In this electrifying process, the resistance of the catalyst was negatively correlated with the temperature (shown in Fig. S14). The energy can be converted between different forms24,25. Figure 2c shows that the power consumed by electrification gradually decreased with an increase in external heating temperature (Fig. S15). Therefore, the WCA strategy without the external thermal field demonstrated the best overall catalyst performance, achieving NOx removal at temperatures below 150 °C (T99 < 150 °C).

Fig. 2: Catalytic performance evaluation.
figure 2

a Visible image of the reactor without the weak current input. The thermocouple measured actual temperatures of 75 °C, 150 °C. b 2VNA catalyst activity by electrifying under the WCA strategy without the external thermal field and merely heating in an electric furnace. Commercial catalyst (2V/TiO2) by heating. c 2VNA catalyst activity by electrifying under the WCA strategy in different external thermal fields. d Under different external thermal fields, the catalyst activity was maintained at 90% by electrical compensation. e Stability of NOx reduction over 2VNA at 130 °C by electrifying (3.4 W, 144 h) under the WCA strategy; reaction conditions: 550 ppm NO, 550 ppm NH3, 5% O2, 50 ppm SO2 (when used) and balance N2; GHSV = 30,000 mL h−1 g−1. f Comparison of TOFs for the commercial and reported SCR catalysts at 200 °C under dry reaction conditions (see the Supplementary Information for TOF comparison under wet reaction conditions). The red solid dot indicates the weak current-assisted catalyst.

Full size image

The WCA strategy proved an effective method for expanding the low-temperature operating range of the catalyst. Figure 2d illustrates that the catalysis activity remained at 90% by electrical compensation under different external thermal fields, where the electrification effect could be dissected into the thermal catalysis and ECWC. The electric compensation increased as the external heating temperature decreased, which demonstrated that the strategy could widen the temperature window of the catalyst at low temperatures. Furthermore, the long-term stability of the SCR catalyst was crucial in evaluating its performance. As depicted in Fig. 2e, during the stability test for 100 h, the catalysis activity was maintained at ~95%, and the selectivity was also stable at 99%. After that, a sulfur resistance test conducted for more than 40 h showed that the activity remained at about 93%. At the same time, the continuous stability of WCA catalyst in the environment containing 7% H2O (Fig. S16a) showed that the WCA catalytic mode could still operate stably in the wet environment. The TOF comparison of various reported catalysts in the wet environment is shown in Fig. S16b. The catalytic cycle stability tests (Fig. S17) showed that the catalytic activity did not change after several cycles of 20 h with or without current. These results highlighted the excellent stability of the strategy. From an energy-saving perspective, the WCA strategy consumed only 15.7% of the energy required for electric furnace heating (power: 3.5 kW) while enhancing activity by 68% (Table S6). It is worth mentioning that as shown in Fig. 2f (dry reaction condition), Fig. S16b (wet reaction condition), and Table S7, S8, the 2VNA under WCA strategy showed the preeminent activity (TOF) among commercial and reported SCR catalysts at the same or higher WHSV. Based on the above results, it was concluded that the strategy reduced energy consumption and improved catalytic activity9, which could achieve a simultaneous focus on energy savings and emission reduction during the catalytic process, making it of significant practical value.

Characterization of mesoscopic mechanism under the WCA strategy

In the Arrhenius equation, the apparent activation energy (Ea) represented the energy barrier for a chemical reaction, while the pre-exponential factor (A) reflected the collision frequency between molecules. Kinetics studies were used to determine the specific impact of the ECWC. During the experiment, NOx conversion was controlled below 20% to avoid the influence of diffusion resistance26. As shown in Fig. 3c, the apparent Ea (25.9 kJ/mol) under the WCA strategy was nearly 20% lower than that under the merely heating mode (31.6 kJ/mol), which indicated that the WCA strategy could directly promote the activation of reactive molecules at active sites in this catalytic reaction. It was also worth noting that the A factor significantly increased from 219.9 μmol g−1 s−1 to 324.5 μmol g−1 s−1 under the WCA strategy. These results suggested that the ECWC reduced the reaction energy barrier and promoted the activation and effective collisions of reactive molecules (Table S9), which greatly facilitated the further reaction process and SCR activity27,28.

Fig. 3: In-situ DRIFTS experiments and reaction kinetic studies.
figure 3

a, b schematic diagram of the modified in situ cell, including front view (a) and top view (b). c Arrhenius plot for NH3-SCR by electrifying under the WCA (0.1 A) strategy and mere heating in the furnace. d Infrared thermal imaging under the WCA (0.1 A) strategy with the external thermal field. e, f In situ DRIFTS experiments between preabsorbed NO + O2 and NH3 over the 2VNA under the two strategies, respectively. The WCA (0.1 A) strategy with the external thermal field, the mere heating in the in situ cell.

Full size image

In-situ DRIFTS experiments were conducted to delve into the mechanism. Figure 3a, b provides schematic diagrams of the modified in situ cell, and the 2VNA catalyst was electrified and characterized on it. First, the in situ DRIFTS absorption spectra for the catalyst under the mere heating strategy was characterized. Figure 3e shows four bands at 1766 cm−1, 1705 cm−1, 1668 cm−1, and 1540 cm−1 corresponding to the adsorbed NO molecule, nitroso, bridged monodentate nitrite, and bidentate nitrate29,30, respectively. When NH3 was introduced, the adsorbed NO molecules desorbed, but nitroso, bridged monodentate nitrite, and bidentate nitrate partially reacted. In contrast, when the WCA strategy is adopted at 200 °C (Fig. 3d), the three bands at 1770 cm−1, 1665 cm−1, and 1539 cm−1 shown in Fig. 3f were assigned to absorbed NO molecule, bridged monodentate nitrite, and bidentate nitrate29,30, respectively. The reason why nitroso disappeared might be that it was oxidized to nitrite with the release of lattice oxygen under the WCA strategy. When NH3 was introduced, the reaction progressed gradually, achieving complete conversion in 25 min. These results indicated that the catalyst under the WCA strategy primarily followed the Langmuir-Hinshelwood (L-H) mechanism31. Combined with the kinetics data, it was evident that the ECWC could activate more inert sites, facilitate the decomposition of intermediate products like bridged monodentate nitrite and bidentate nitrate, and eventually enhance the SCR reaction.

For NH3-SCR, a deeper understanding of the reaction mechanism could be achieved by examining it from both acidic and redox site perspectives. To clarify the specific influence of the ECWC on the acid sites, Fig. 4a shows that the maximum desorption peak of NH3 shifts to high temperature as a whole, which indicates that the adsorption strength of NH3 on the catalyst surface slightly increased (Fig. S18). The redox sites were characterized. As shown in Fig. 4b, the catalyst was utterly reduced at 250 °C under the WCA strategy, whereas it required much higher temperatures (800 °C) for a total reduction under mere heating conditions (Fig. S19). These results showed that the ECWC significantly enhanced the oxidation ability of the catalyst, leading to an overall improvement in its catalytic activity. Furthermore, the two stages of oxidation (V4+ → V5+) and reduction (V5+ → V4+) were further characterized by the oxygen transient experiments. As presented in Fig. 4c and Table S10, at the same temperatures (150 °C), electricity substantially increased both the rate of lattice oxygen release and filling32,33, meaning the ECWC promoted these two processes. Additionally, the lattice oxygen release rate during the reaction process was significantly lower than the lattice oxygen filling rate, indicating it represented the rate-determining step in the catalytic cycle.

Fig. 4: Characterization results of catalyst surface acidity, redox ability, and oxygen circulation characteristics under the WCA strategy.
figure 4

a NH3-TPD and b H2-TPR profiles under the two strategies, respectively. The WCA strategy without the external thermal field (electrifying), the mere heating in the furnace. c Oxygen transient experiments: the oxygen was cutoff at 0 min and resumed at about 260 min under the two strategies. Experiments were maintained at different temperatures (75 °C, 150 °C). d, e Isotopic oxygen exchange experiments: the evolution of isotopic oxygen exchange (18O2 to 16O18O) under the two strategies. The WCA strategy without the external thermal field (the electric current ramp rate 0.06 A/min), the mere heating in the furnace (the temperature ramp rate of 13 °C/min). Profiles for the 16O18 O signal and temperatures were shown. MS, measured signal.

Full size image

To fully understand the catalytic reaction, it was essential to consider a holistic perspective rather than solely focusing on local effects. On the one hand, the production of the -NH2NO intermediate on V5+ with the release of lattice oxygen specifically encompassed the processes of V5+ → V4+ and NH3 → NH2− + H+3,34. Through isotopic oxygen exchange experiments (isotope O18 labeling), the lattice oxygen released process of the catalyst was characterized in detail35. As demonstrated in Fig. 4d, e, the signal for 16O18O (the exchange of 18O2 and lattice oxygen produced 16O18O.) begins to increase at 261.5 °C under the WCA strategy, whereas it required a much higher temperature (558.5 °C) when heating alone. At the same time, the signal of 16O2 decreased when the signal of 16O18O reached its highest value, indicating that the surface lattice oxygen of the catalyst will exchange with 16O18O for the second time to produce 16O2 (Fig. S20). These results validated that the ECWC promoted the release of lattice oxygen and also enhanced the mobility of lattice oxygen36,37,38. On the other hand, the promotion of lattice oxygen filling could be attributed to the activation of molecular oxygen under electrical conduction. XPS analysis revealed that the binding energy of the oxygen defect (Oβ) decreases by 0.28 eV (Fig. S21 and Table S11) under the WCA strategy. The decrease in binding energy indicated an increase in electron density at the oxygen defect site, which facilitated the activation process of molecular oxygen (O2 → O22−). The above results suggested that the WCA strategy can break the limitation of monoatomic catalysts to activate molecular oxygen at low temperature to some extent.

Characterization of the microcosmic mechanism under the WCA strategy

Although the superficial reason for ECWC was to promote the release of lattice oxygen, the deeper underlying mechanisms still required further exploration. First, the essence of the ECWC was studied. Figure S22 suggests that the weak current had a promoting effect on the catalyst only when it was in the conducting state. Then, the most critical embodiment in the conducting state was current, and the essence of current was electron migration, which showed that the essence of the ECWC was electron migration. The approximate in situ XPS was characterized by an argon atmosphere under the WCA strategy to delve into the mechanism by which electron migration affected lattice oxygen. As shown in Fig. S21 and Table S11, at the same temperatures (150 °C), the binding energy of V and Sn elements under the WCA strategy decreased by 0.25 eV and 0.26 eV39,40,41, respectively. These results indicated that electronic migration weakened the strength of V-O and Sn-O bonds, and lattice oxygen was easier to detach. The approximate in situ XPS analysis of NH3-SCR with the WCA strategy was also characterized (Fig. S23 and Table S12). The increase in the proportion of V5+ and Sn4+ and the decrease in the ratio of lattice oxygen (Oα) could be attributed to activated molecular oxygen oxidizing V4+ and Sn2+ (activated molecular oxygen translating into O22-) under the conducting state, which probably promoted the lattice oxygen filling42.

To further validate the impact of electron migration on the binding energies of the metal elements, the in situ XPS was tested under applied bias voltage in a vacuum environment. The in situ XPS test with bias voltage modification was based on standard in situ XPS, with the sample table and the sample connected to the positive voltage of a power supply, while the vacuum chamber and the negative electrode of the power supply were grounded at zero potential (Fig. S24a, b). Figure S24c, d demonstrates that the application of external voltage can weaken the binding of V atoms and Sn atoms to the surrounding electrons, especially for V atoms43. More importantly, in order to prove that the V-O binding energy was reduced due to electric current rather than the existence of only potential difference, electrons emitted by the electron gun were used on the surface of the sample to simulate the conduction state of the sample in the experiment. Figure 5a, b shows the schematic diagram of in situ XPS equipment with electron emission function. As can be seen from Fig. 5c, d, with the increase of weak current, the binding energy of V 2p decreased gradually, but the binding energy of Sn 3d did not change, which suggested that when the current flowed through the catalyst, the mobile electrons were mainly concentrated around the V single atoms on the surface of 2VNA catalyst (“electron enrichment effect” based on monoatomic electron bridge), or the directional flow of electrons reduced the energy required for the excitation of electrons around embedded monatomic vanadium (“electron escape effect” mediated by directional flow of electrons), which eventually leaded to the weakening of V-O bond (it can be vividly called “electron scissors effect”). Combined with the results of activity and isotopic oxygen exchange experiments in the non-conducting state (Fig. S22 and Fig. 3b; Fig. S25 and Fig. 5d), it could be seen that the existence of electric current was the key of WCA strategy.

Fig. 5: In situ XPS characterization with weak current applied by electron gun.
figure 5

a, b Schematics of the in situ XPS cell created in-house for electrified binding energy deflection, with front (a) and top-view (b). c, d In situ XPS results during different binding energies 522–512 eV (c) and 500–482 eV (d). The 2VNA sample was tableted, and different current were applied under vacuum conditions, the operating voltages of 0, 2, 4, and 6 V correspond to the currents of 0, 1.35, 12, and 22.8 μA on the sample surface, respectively.

Full size image

DFT calculation

The internal mechanism of lattice oxygen activation through DFT calculations could be elucidated. First, the catalyst model was optimized to obtain the 2VA catalyst, and it kept the structure under the electric field (Fig. 6a–c). Based on the optimized model, the energy band structure of the entire catalyst was calculated in two situations before and after the application of an electric field. Figure 6d shows that the maximum valence band and the minimum conduction band were situated at the Γ point of the Brillouin zone, and the Fermi level (EF) shifted to the bottom of the conduction band. These results indicated that the 2VA catalyst was a direct gap N-type semiconductor. Additionally, the energy band gap after electron migration became smaller, implying that electron migration improved the electronic state of the entire catalyst44. The improved electronic state may manifest in the weakening of chemical bonds45.

Fig. 6: DFT calculation for electronic state over monoatomic vanadium-based catalyst.
figure 6

Side-view and top-view of optimized structural model of (a) ATO, (b, c) 2VA. Comparative analysis of the 2VA catalyst performance before and after electric field application via DFT calculations: characterizations including (b, c) crystalline structure, (d) band Structure, (e, f) density of states, and (g, h) COHP. The electric field is set to 0.01 V/Å, and the direction is omnidirectional.

Full size image

The partial density of states (PDOS) and crystal orbital Hamiltonian population (COHP) were calculated to evaluate the metal-oxygen bond strength46. Figure 6e, g reveals that after electron migration, the center of the p-band had shifted upward relative to the Fermi level, resulting in a reduced energy gap between the vanadium 3d and anion 2p band centers47, thereby improving the intrinsic electronic conductivity of the catalyst. Moreover, the integrated COHP (ICOHP) increased from −2.346 to −2.246, which indicated that electron migration resulted in an elevation of the energy of bonded states and an increase in the proportion of antibonding orbitals48. As a result, the V-O bond was weakened, leading to an average bond length of 1.824 Å (Fig. S26). The trend was also observed in the Sn-O bond. After electron transfer, the center of the p-band for Sn moved upward, the corresponding ICOHP increased to −2.119, and the average bond length of Sn-O expanded to 2.122 Å (Fig. 6f, h and Fig. S24). However, compared to V-O, the electron migration effect on Sn-O was less pronounced, indirectly indicating that V was the primary active component of the 2VA catalyst. In summary, the above experimental and theoretical studies demonstrated that electrons migrated and occupied the anti-bond orbital of metal-oxygen bonds, which weakened the chemical bond strength of V-O and Sn-O, especially for V-O bond, thereby promoting the activation of lattice oxygen. Figure S27 illustrates a schematic comparison between traditional thermal catalysis and CA catalysis, and Fig. 7 shows a schematic diagram of the weakening of the chemical bond energy of V-O-Sn, revealing the reason for the excellent performance at low temperatures from the perspective of valence electron state. It was estimated that if the WCA was introduced on the basis of traditional thermal catalysis, 30.63 million tons of standard coal could be saved annually (As shown in the Supplementary Methods of the supplementary information). Moreover, the renewable power energy used in the WCA catalysis strategy was extensive, such as solar energy, wind energy, hydrogen energy and so on, which was very conducive to the large-scale promotion of the WCA catalysis.

Fig. 7: Schematic diagram of current-assisted catalysis and prospects for the electrification of catalyst.
figure 7

Average distance between metal element sites (V or Sn) near monatomic V active sites and O atoms is calculated according to DFT calculations. Energy-saving data is estimated according to the annual flue gas emissions and the proportion of NOx emission in low-temperature flue gas.

Full size image

To summarize, it is vital for V-based catalysts to achieve high NH3-SCR efficiency with low V content and ultra-low temperature operating condition. The catalyst (2VNA) was systematically characterized, and it was found that the V atoms in the catalyst were highly dispersed and anchored on the surface of the ATO lattice in the form of monoatomic states, which significantly reduced the V content. Then, the catalysis activity was tested under the WCA (or called CA) strategy. The results showed that the ECWC played a significant role in the WCA strategy, and it was very effective for NH3-SCR (T99 < 150 °C). In practical application, the WCA strategy significantly reduced energy consumption, dramatically broadened the low-temperature temperature window of V-based catalyst (maintaining high activity and high N2 selectivity), and the catalyst had excellent stability. In-situ DRIFTS experiments and kinetic studies were characterized under the WCA strategy, and it was found that ECWC significantly promoted the effective collisions of reactive molecules and the decomposition of intermediate products (nitrite, nitrate). Oxygen transient and isotope oxygen exchange experiments showed that ECWC promoted the production and decomposition of -NH2NO intermediate on V5+ (accompanied by the release of lattice oxygen) and activated molecular oxygen to some extent. Finally, in situ XPS results indicated that the directional flowing electrons mainly concentrated around monodisperse V atoms in the electron migration process, which reduced the binding energy of V, leaded to the increase of bonding energy, and eventually promoted the release of lattice oxygen. DFT calculation showed that the proportion of antibonding orbitals increased after electron migration, and correspondingly, the chemical bonds of V-O and Sn-O became longer. These results directly proved that electron migration weakened chemical bond strength (electron scissors effect), thus improving the activation of lattice oxygen. Moreover, in addition to the lattice oxygen-mediated mechanism mentioned above, CA catalysis also promotes electron transfer over the catalyst, and the enhancement mechanism may be potentially intrinsically related to the electron delocalized mechanism49, which needs to be specifically explored in the future.

The WCA catalysis strategy breaks through the limitation of traditional thermal catalysis and provides a solution to the problem of low activity for existing low-temperature catalytic systems. In particular, the characteristics of reducing cost, improving efficiency, simple reactor structure and low operating voltage and power (conducive to practical application) align with the future development trend, which can effectively promote the industrial application of low-temperature catalytic systems and promote the development of catalyst electrification. This work provides a new strategy for achieving greater activity with less energy consumption and can play a very positive role in promoting the realization of global carbon neutrality.

Methods

Sample synthesis

Active components and modified active components were loaded on ATO by the impregnation method. Based on the loading percentage, ATO, ammonium metavanadate solution and acidic components (if any) were added to the solution, stirred and evaporated at 70 °C, and then calcined at 500 °C for 4 h, and finally catalysts such as xV-yW/ATO, xV-yMo/ATO and xV-yNb/ATO were prepared. The x and y represent the mass percentages of V2O5 and modified active components (W, Mo, Nb) in the catalyst. In addition, for the convenience of main text description, 2 V/ATO and 2V-1Nb/ATO catalysts are called 2VA and 2VNA, respectively.

Catalyst evaluation

The catalytic activity of the catalyst in NH3-SCR was measured in an 8 mm fixed bed quartz tube reactor with a 0.15 g sample. The total flow rate was kept at 75 mL/min, and the reaction conditions corresponded to a GHSV of 30,000 mL/(g·h). During the test, the D07-19B mass flow controller was used to control the gas flow. The inlet airflow includes 550 ppm NH3, 550 ppm NO, 5% O2, 50 ppm SO2 (when used), 7% H2O (when used), and the rest is N2. The Antaris IGS gas analyzer was used to analyze inlet and outlet gas components (NO, NO2, NH3, N2O). The catalytic test was carried out at ambient pressure, and the temperature range was 25 °C to 250 °C. The NO conversion and N2 selectivity under steady-state conditions were calculated by formulas (1) and (2), respectively.

$${{NO}}_{x}{conversion}\left(\%\right)=\frac{{[{{NO}}_{x}]}_{{in}}-{[{{NO}}_{x}]}_{{out}}}{{[{{NO}}_{x}]}_{{in}}}\times 100\%$$
(1)
$${N}_{2}{selectivity}(\%)=\left(1-\frac{{2[{N}_{2}O]}_{{out}}}{{[{{NO}}_{x}]}_{{in}}+{[{{NH}}_{3}]}_{{in}}-{[{{NO}}_{x}]}_{{out}}-{[{{NH}}_{3}]}_{{out}}}\right)\times 100\%$$
(2)

where [NH3]in, [NOx]in, [NH3]out, [NOx]out, and [N2O]out were the concentrations of NH3 and NOx (including NO and NO2) at the inlet and those at the outlet.

In addition, the WCA strategy refers to the auxiliary strategy with a voltage below 10 V and a current of 0–1 A, and the whole catalytic reaction was affected by the current. During the electrical performance test, the weak current auxiliary strategy was adopted, and the external direct current (DC) power supply (eTM-3020P) was used to increase the applied auxiliary current gradually. The actual temperature of the catalyst was monitored by an external K-thermocouple. The schematic diagram of the catalytic activity test is shown in Fig. S1 and Fig. 2a.

Characterization

X-ray diffraction was performed using a Cu-Kα (40 kV, 40 mA, λ = 0.15418 nm) excitation source on a D8 Advanced X-ray diffractometer from Bruker AXS GmbH, Germany. XRD data were recorded over 10-minute intervals in the range of 2θ = 5°–80°. N2 adsorption and desorption isotherms of the samples were measured at 77 K using Micromeritics Kubo 1200 surface area and porosity analyzer instrument. Samples were previously evacuated at 250 °C for 4 h. During the testing process, the relative pressure range was set to P/P0 = 0–0.99. Pore volume (Vp), pore size distribution (BJH), and specific surface area (BET) of the sample were assessed via N2 adsorption and desorption isotherm measurements. The elements of the sample were tested by the inductively coupled plasma atomic emission spectroscopy (Agilent 5110 instrument) with a fixed volume of 25 mL mixed solution (nitric acid and hydrofluoric acid) in the dilution factor of 10. The high-resolution transmission morphology inside the material was observed with the Titan Cubed Themis G2300 instrument. Among them, the sample was prepared by ultrasonic treatment in ethanol and then dropped on the ultra-thin carbon film of copper mesh and finally observed. Raman spectra in the range of 50–4000 cm−1 were obtained by a Renishaw’s inVia confocal Raman microscope with a grating of 532 nm, objective lens of, and optical intensity set at 1. The spectra results are conducted to detect the atomic bonding states within the catalyst’s interatomic bonds. The resistivity of the sample powder was measured by the ST-2258C instrument, in which the resistivity and conductivity were reciprocal. X-ray photoelectron spectroscopy (XPS) was used to analyze the surface element state and content of the catalyst using Thermo Scientific K-Alpha instrument. The redox performance of the catalyst was analyzed by cyclic voltammetry (CV) with the Zahner PP211 electrochemical workstation instrument. Under the action of ultrasonic wave, 20 mg of sample powder, 2 mg of carbon powder, 50 ml of Nafion and 2 ml of ethanol were thoroughly mixed, then dropped on the platinum-carbon electrode, and the test was completed in 0.5 mol/L Na2SO4 electrolyte solution with a scanning rate of 100 mV/s. XAS comprises XANES and EXAFS spectra. XAS measurement was conducted at Japan’s Spring8 BL 14B2 beamline with an electron beam energy of 8 GeV and an electron beam current of 99.5–99.6 mA. Data was collected using two flat-crystal Si (111) crystal fixed outlet monochromators, and the raw data was finally analyzed using Athena software.

WCA NH3 temperature-programmed desorption (WCA-NH3-TPD)

The acidity of electrified samples was determined by the WCA-NH3-TPD. A 150 mg catalyst sample was loaded into a quartz tube inside a furnace. Pretreatment in Ar at 200 °C for 30 min. The temperature was maintained at 100 °C by electrifying or heating. After the initial pretreatment, NH3 (10% NH3-He) was adsorbed onto the catalyst at 100 °C for 1 h. Subsequently, the sample cooled down, and the desorption test was conducted by heating from 50 °C to 650 °C at a rate of 10 °C per minute in pure Ar. Data collection throughout the process was carried out by mass spectrometer (MS), HPR-20 EGA.

WCA H2 temperature-programmed reduction (WCA-H2-TPR)

The redox properties of the electrified samples were characterized by the WCA- H2-TPR. A 150 mg catalyst sample was placed into a quartz tube within a furnace. Pretreatment in Ar at 200 °C for 30 min, and then cooling to room temperature. Finally, with the gas flow of 10% H2-Ar, the test was completed in the process of rising from 30 °C to 255 °C at the rate of 5.5 °C/min. Data collection throughout the process was carried out by MS, HPR-20 EGA.

In situ DRIFTS

The wires were fixed to the surface of ITO substrate by silver paste, and then they were placed in an oven at 150 °C for 5 h. Subsequently, ultrasonic dispersed sample ethanol suspension was dripped onto the conductive substrate. A custom-made electrically connectable sample holder was placed in an in situ cell. Two heating methods were employed: one using a 0.1 A microcurrent in addition to a resistive heater and the other using only a resistive heater. Both methods maintained a controlled temperature of 200 °C. A Nicolet IS50 FTIR spectrometer was used for in situ DRIFTS within the range of 4000 to 1000 cm−1.

In situ XPS

Under vacuum conditions, the positive bias voltage of 0.5 V and 4 V were applied by the Kratos axis ultra-DLD XPS system to study the influence of electron migration on the binding energy of V-O and Sn-O in the sample, and the carbon peak (C 1 s = 284.8 eV) was used as the standard value to calibrate the binding energy, so that the final data could be obtained. Furthermore, in order to further confirm the role of electron migration, the electrons excited by the electron gun on the sample surface are used to simulate the weak current of the conductive sample in the actual test.

Isotopic oxygen exchange experiments

The lattice oxygen in the catalytic process was studied by isotope oxygen exchange experiment. A 150 mg catalyst sample was placed into a quartz tube within a furnace. The catalyst underwent a 30-minute purge with pure Ar, followed by switching the gas flow to 5% 18O2 diluted with Ar. The total flow rate was 75 mL/min (4.7 mL/min of 80% 18O2 and 70.3 mL/min of Ar). The current was gradually increased at a rate of 0.06 A/min up to 1.8 A and maintained. Similarly, to make the heating rate consistent, the temperature ramp rate of the simple heating mode was 13 °C/min. Monitor molecular weight with MS: 32 (16O2), 34 (16O18O), 40 (Ar).

Computational details

First-principles calculations were carried out in Vienna ab initio simulation package (VASP)48,50,51,52 using the framework of DFT. DFT calculation can be used to study the states of energy bands, chemical bonds and active electrons in the samples during the catalytic process at the electronic level. A periodic flat plate with three atomic layers was established on tin dioxide (110) as the base of the SCR catalyst surface. Then, the V atoms randomly replaced Sn atoms on the surface with a loading percentage of 1.36% as the catalyst model. Wherein the thickness of the vacuum layer on the surface of the tin dioxide (110) is set to be at least 30 Å. When the plane wave cutoff energy was set to 450 eV, the lattice parameters had little difference after geometric optimization using generalized gradient approximation Perdew-Burke-Ernzerhof (GGA-PBE) functional. PBE functional based on GGA is widely used to explain the exchange-correlation between the catalyst and selective catalytic reduction of ammonia53,54. The DFT-D3 dispersion correction method was used to describe the weak interaction between slabs and active species55. Therefore, the interaction between ions and electrons is described by the Projected Augmented Wave (PAW) method56,57. In this model, the electric field parameters are as follows: the electric field intensity was 0.01 V/Å, and the electric field direction was applied in all directions. Precisely, the energy band structure of the catalyst as a whole, the PDOS of vanadium 3d and oxygen ion 2p, the PDOS of Sn 3d, and the COHP of the chemical bond between vanadium oxide and tin oxide were calculated by VASP. The purpose of these calculations is to study further the influence of the catalytic enhancement of weak current on the electronic states and anti-bond orbitals in the catalyst so as to further explain the specific function of weak current in NH3-SCR at a deeper level. Among them, the calculation result of COHP was carried out using Lobster software58,59.