Wen
Cui†
ab,
Jiaqi
Wang†
a,
Yan
Li
a,
Pingqu
Liu
a and
Fan
Dong
*abc
aCollege of Resources and Environmental Engineering, Key Laboratory of Karst Georesources and Environment, Guizhou University, Guiyang 550025, China
bGuizhou Karst Environmental Ecosystems Observation and Research Station, Ministry of Education, Guiyang 550025, China
cResearch Center for Carbon-Neutral Environmental & Energy Technology, Institute of Fundamental and Frontier Sciences, University of Electronic Science and Technology of China, Chengdu 611731, China. E-mail: dongfan@uestc.edu.cn
First published on 26th September 2024
Nitrogen oxides (NOx, x = 1 and 2, the proportion of NO is about 95%), as one of the primary precursors for particulate matter and ozone, limit the continuous improvement of air quality. Photocatalytic NO purification technology has attracted significant attention, and much efforts have been devoted to realizing the complete photocatalytic oxidation and reduction of NO for toxic by-product inhibition and end-product recovery. This work presents a timely overview of the current research progress on the conversion of NO into nitrate/ammonia (NO3−/NH3), which can be further recycled and utilized. According to the principle of heterogeneous photocatalysis and considering the significance of the reaction microenvironment (surface active sites of photocatalysts, target pollutants and reaction media), this review systematically summarizes the progress on the strategies for controlling the surface structure of photocatalysts and reaction medium. Specifically, this critical overview is focused on various methods for the surface modification of photocatalysts, strategies to accelerate the mass transfer process of gaseous NO, and the effect of the additional introduction of a reductant/antioxidant in the reaction system. Furthermore, research trends and future prospects are discussed, aiming to provide insights into the breakthroughs and boost the development of photocatalytic NO removal technology.
Environmental significanceNO is one of the primary precursors for the formation of particulate matter and ozone. Considering its insoluble characteristic and the intermediate chemical valence state of nitrogen in NO, it can be removed and even upcycled via the redox reaction. Photocatalysis, which makes full use of inexhaustible solar energy, can be an eco-friendly and cost-effective method to realize the complete oxidation and reduction of NO into nitrate/ammonia (NO3−/NH3), which can be further utilized. Herein, a comprehensive critical overview and perspectives, focusing on the complete photocatalytic oxidation and reduction of NO for toxic by-product inhibition and end-product recovery, are presented, which will hopefully inspire ongoing efforts to optimize the sustainable conversion process of NO and advance the practical application of photocatalytic NO removal technology. |
Photocatalytic NO oxidation/reduction is a heterogeneous reaction, which is the reactant conversion process induced by surface/interface charge transfer between the reactants and photocatalyst, where the reaction microenvironment mainly consists of surface active sites on the photocatalyst, target pollutants and reaction media.30 In the gas–solid reaction, the photocatalyst is usually regarded as the core, where its surface structure determines the process of surface charge transfer and interfacial molecular transformation.31,32 Thus, the rational design of photocatalysts can be a fundamental way to regulate the surface charge arrangement and surface/interface charge transfer process, which is beneficial for optimizing the reaction pathway for toxic by-product inhibition, thus improving the selective catalytic conversion efficiency of reactants.30 In the gas–liquid–solid reaction, the reaction microenvironment is more complex, mainly reflecting the reaction medium.33 Besides the surface structure of the photocatalyst, the mass transfer process of gaseous reactants and the introduction of oxidants/reductants/sacrificing agents also need to be fully considered, which are the determinants in efficiently realizing selective catalytic conversion.
Photocatalytic NO oxidation is a reactive oxygen species-based (ROS-based) gas–solid reaction, and the diffusion of reactants (O2, H2O, NO) to the surface active sites of the photocatalyst for adsorption and activation is the premise of subsequent ROS formation and NO oxidation.30,34,35 Therefore, it is necessary to propose reasonable design and corresponding modification methods for the surface structure of photocatalysts, aiming to facilitate the adsorption and activation of the reactants and the separation of charge carriers for NO activation and the formation of ROS with deep oxidation ability to realize the efficient complete oxidation of NO.21,36–39 In the photocatalytic NO reduction process, the reaction microenvironment is more complex given that it generally proceeds in a gas–liquid–solid three-phase reaction system. The ultralow solubility of gaseous NO limits its transfer to the liquid phase for its subsequent conversion, and the reduction reaction is difficult to carry out spontaneously in an air environment.19,40 Therefore, in addition to optimizing the reactant activation process and charge carrier separation by precisely constructing the surface active sites in the photocatalyst, strategies for accelerating the mass transfer of gaseous NO and introducing applicable reductant/antioxidant into reaction system have also been proposed, aiming to achieve the efficient complete photocatalytic reduction reaction for NO upcycling.
In the past few decades, several excellent studies on photocatalytic NO removal have been published, and various critical reviews have emerged, summarizing the methods for the modification of the photocatalyst and the progress of typical photocatalysts for efficient N oxidation.19,41–47 However, photocatalytic NO complete oxidation/reduction with toxic by-product inhibition and end-product recovery has only recently attracted attention, and thus there is a lack of reviews focusing on this topic. Therefore, a comprehensive review is necessary, which can not only provide readers with a better understanding of the recent research progress in this field but also advance the development of photocatalytic NO removal technology. Herein, we attempt to present a critical review on photocatalytic NO removal, with the emphasis on summarizing the achieved progress in photocatalytic NO complete oxidation and reduction for toxic by-product inhibition and directed final product (NO3−/NH3) formation (Fig. 1). Some typical examples are also discussed in each section below, without bias towards studies from specific research groups. Meanwhile, the limitations and potential research directions to overcome the related challenges are also highlighted. It is hoped that this review will inspire ongoing efforts to precisely adjust the reaction microenvironment for efficient photocatalytic NO removal.
Fig. 1 Illustration of the complete photocatalytic oxidation and reduction of NO for toxic by-product inhibition and directed final product (NO3−/NH3) formation. |
The introduction of heteroatoms in graphitic carbon nitride (g-C3N4) has been widely reported for photocatalytic NO oxidation, which can achieve control of the interlayer or intralayer charge configuration of g-C3N4, promoting the activation of reactants and separation of charge carriers for the generation of ROS participating in the complete oxidation of NO. B-doped g-C3N4 enabled electron localization to accelerate the activation of the reactants and intermediates, and then facilitated photocatalytic NO oxidation reaction.54 Zhang et al. introduced Zn atoms into the g-C3N4 interlayer to form a Zn–N3 bridging structure (Zn-CN), which induced the adsorption of O2 and NO to fabricate the Zn–O2–NO structure, and then realized the direct formation of nitrate without NO2 generation (Fig. 3a).55 Besides, Sr-doped g-C3N4 with different doping modes was reported, and its corresponding functions were also elaborated (Fig. 3b).56 Sr-doping with simultaneous N atom replacement, cavity padding, and intercalation facilitated the activation of O2 to produce ˙O2−, endowing g-C3N4 with the ability to activate H2O2 for the generation of ˙OH and improving the transfer of photogenerated electrons, respectively. The different doping modes complement each other to realize efficient and deep NO removal.
Fig. 3 (a) NO oxidization process over g-C3N4 and Zn-CN.55 Reproduced from ref. 55 with permission from Elsevier, Copyright 2023. (b) Heat absorption and release of various possible structures of Sr multi-site doped g-C3N4.56 Reproduced from ref. 56 with permission from Elsevier, Copyright 2019. |
Meanwhile, the band structure of photocatalysts can be tuned by element doping, which is conductive to extending their light adsorption range and enhancing their redox ability. Besides, the introduction of heteroatoms can induce the formation of new intermediate energy levels, facilitating the separation, migration and transfer of photogenerated carriers to enhance the photocatalytic performance. For example, Guo et al. constructed N-doped BiOCl (N-BiOCl) with intermediate energy levels (Fig. 4a), which optimized the migration process of charge carriers to effectively generate ˙O2− for the direct conversion of NO into NO3−.57 The surface charge distribution and band structure of the photocatalyst could also be optimized simultaneously by element doping. Fe ions in Fe-doped TiO2 (Fe-TiO2) not only provided more active sites but also induced new states due to the Fe 3d orbit between the band gap of Fe-TiO2 (Fig. 4b and c), which accelerated the adsorption and activation of the reactants, enhanced the light absorption range and promoted the separation of charge carriers for the deep oxidation of NO (Fig. 4d–g).58 The B atom in B-doped Bi2O2CO3 not only acted as a channel of charge transfer to promote the activation of reactants, but also contributed to the formation of intermediate energy levels for the separation of electron–hole pairs, which induced the deep oxidation of NO.59
Fig. 4 (a) Total density of states of BiOCl and N-BiOCl.57 Reproduced from ref. 57 with permission from Elsevier, Copyright 2022. Density of states for (b) TiO2 and (c) Fe-TiO2. Charge density difference distribution in optimized (d) NO adsorbed on TiO2, (e) NO adsorbed on Fe-TiO2, (f) O2 adsorbed on TiO2 and (g) O2 adsorbed on Fe-TiO2. Light blue and yellow represent charge loss and charge accumulation, respectively. The calculated positive Bader charge (Δq) indicates the electron transfer from the catalyst surface to adsorbed molecule.58 Reproduced from ref. 58 with permission from Elsevier, Copyright 2020. |
There is the possibility of forming defects in the original material during the introduction of heteroatoms. The presence of vacancies also has an effect on the surface charge distribution and band structure of photocatalysts. Zhang et al. found that La3+ doping narrowed the band gap of Bi5O7I and induced the formation of oxygen vacancies (OVs).60 The narrow band structure improved the electronic excitation in the photocatalyst, and the presence of OVs facilitated the activation of O2 to generate ˙O2− for complete NO oxidation. Besides, the co-doping of atoms in photocatalysts has been proposed, where the heteroatoms play a synergistic role in elevating the photocatalytic NO oxidation performance. In Na–Ca co-doped g-C3N4, sodium doping promoted the separation of electron–hole pairs and strengthened the redox capacity, while calcium doping enhanced the chemisorption of NO2 to inhibit the emission of toxic by-products.61 Also, Lu et al. further prepared K–Ca co-doped g-C3N4 and found that K and Ca resulted in a better improvement than other elements in the alkali and alkaline-earth family.62 K–Ca co-doped g-C3N4 possessed both electrons with strong reducibility and holes with strong oxidizability to generate ˙O2− and ˙OH, participating in photocatalytic NO removal, which was also was conductive to the chemisorption and physisorption of NO and NO2 on the photocatalyst surface for their further complete oxidation into NO3−.
Supported single-atom catalysts cannot only maximize the atomic efficiency of metals but also provide an alternative strategy to tune the activity and selectivity of catalytic reactions, which have also been employed to optimize the reaction pathway of photocatalytic NO oxidation. Guo et al. proposed the anchoring of atomically dispersed Ag1 atoms on a {001} facet-exposed BiOCl nanosheet (Ag1/BiOCl), creating Ag single atom sites with a triangle two-coordinated Cl–Ag1–Cl local configuration (Fig. 5a and b), which achieved >90% NO conversion to the favorable NO3− with high selectivity (>97%) (Fig. 5c and d).63 The triangle Cl–Ag1–Cl sites not only promoted light adsorption and electron transfer kinetics in BiOCl but also facilitated end-on O2 adsorption, contributing to the one-electron molecular oxygen activation to form ˙O2−. The boosted generation of ˙O2− could selectively react with NO to form nitrate with strong bidentate adsorption on Cl–Ag1–Cl sites, which inhibited the dissociation of nitrate to produce unfavorable NO2 (Fig. 5e). Also, single-atomic Na (Na-SA) played a “three-in-one” role in carbon nitride (CN).64 Specifically, the introduction of Na-SA induced electron transfer from Na to the complex N2C of CN and the formation of a built-in electric field between layers, which increased the light harvesting ability, facilitated the adsorption of reactants and accelerated the separation of carriers, thus improving the NO removal rate and preventing the production of toxic by-products. Highly dispersed Ni-incorporated C3N5 was facilely prepared for efficient photocatalytic NO oxidation.65 The surface anchored Ni sites not only optimized the optical property but also promoted the separation of photogenerated carriers to facilitate the activation of O2 for ROS formation, and it was also demonstrated that ˙O2− mainly contributed to NO → NO2 conversion, while 1O2 and ˙OH could further convert NO2 into the harmless NO3−.
Fig. 5 (a) Atomic-resolution ABF-STEM and (b) HADDF-STEM images of Ag1/BiOCl. (c) Photocatalytic NO oxidation of P25, BiOCl, Ag1/P25 and Ag1/BiOCl. (d) NO3− selectivity of Ag1/P25 and Ag1/BiOCl under visible light irradiation. (e) Schematic of photocatalytic NO oxidation at the Cl–Ag1–Cl sites under visible light.63 Reproduced from ref. 63 with permission from John Wiley and Sons, Copyright 2023. |
Besides, the introduction of OVs is an approach by which the optical property, charge separation, and surface structure of photocatalysts can be tuned.71–74 The OVs of blue TiO2 (TiO2-OV) with localized electrons not only facilitated the activation of O2 through the single-electron pathway to generate ˙O2− for the transformation of NO into nitrate, but also simultaneously induced photogenerated hole annihilation to inhibit the side-reaction between holes and NO for avoiding the formation of toxic by-products, which achieved the highly selective removal of NO (Fig. 6a–c).71 Besides facilitating the adsorption and activation of reactants, the OVs in Bi3TaO7 (OVs-BTO) induced the redistribution of local electrons to form a fast charge transfer channel between OVs and adjacent Ta atoms, increasing the transfer rate of photogenerated carriers, which was helpful to promote the deep oxidation of NO into nitrate (Fig. 6d).72 Zhang et al. synthesized g-C3N4 containing non-intrinsic OVs by oxygen pre-doping (Vo-CN) followed by elimination (Fig. 6e).73 The introduction of OVs induced the formation of a mid-gap state near the Fermi level to accelerate the separation of photogenerated carriers and provided favorable channels for carrier migration to accelerate O2 and NO adsorption, which promoted the generation of ROS, thus achieving efficient NO removal with high nitrate selectivity (Fig. 6f and g). Furthermore, the introduction of an intermediate energy level derived from the introduced vacancy can endow insulators with semiconductor-like photocatalytic activity, where BaSO4 with Ba-vacancy has already exhibited a photocatalytic NO oxidation performance.75 Ba-vacancy also induces the redistribution of surface charge to accelerate the activation of NO by donating electrons to electron-deficient areas, contributing to the conversion of NO into a higher valance state for further favorable photocatalytic oxidation.
Fig. 6 (a) EPR spectra for TiO2-OV and TiO2 under visible-light (420 nm) irradiation at 120 K. The corresponding simulated spectrum is indicated by the gray line. (b) Schematic of the interfacial charge-transfer path on TiO2-OV. The red and green ellipses represent the electron clouds, respectively. The oxygen vacancy with two localized electrons is essentially two unsaturated titanium ions (Ti3+). (c) Schematic drawing of the interfacial electron-transfer processes during photocatalytic NO oxidation on TiO2 and TiO2-OV.71 Reproduced from ref. 71 with permission from the American Chemical Society, Copyright 2019. (d) Mechanism of deep NO photo-oxidation by BTO and OVs-BTO.72 Reproduced from ref. 72 with permission from Elsevier, Copyright 2022. (e) Electronic location function of CN, OCN, and Vo-CN samples. (f) The pathway of photocatalytic NO oxidation on VO-CN. (g) Detailed photocatalytic reaction mechanism on the surface of VO-CN under visible light irradiation.73 Reproduced from ref. 73 with permission from Elsevier, Copyright 2024. |
Besides OVs, other types of defects have been fabricated to introduce extra active sites and optimize the band structure and surface electronic structure of photocatalysts, aiming to improve their performance. The metal-free layered conjugated semiconductor g-C3N4 has become one of the outstanding candidates for photocatalytic NO removal and different types of vacancies are easy to be constructed for further improving its photocatalytic performance.39,76–79 Carbon vacancy-modified C3N4 nanotubes were equipped with efficient performance for the selective oxidation of NO to nitrates, which was ascribed to the trapping of photo-induced electrons by carbon vacancies, and then reaction with surface-adsorbed O2 and H2O to form ˙O2− and ˙OH for NO complete oxidation, respectively (Fig. 7a).76 Li et al. prepared C3N4 with three-coordinate nitrogen vacancy (AC-CN4), where the introduction of nitrogen vacancy as the reactive sites either weakened the adsorption of intermediates/final products or facilitated the activation of O2 to generate 1O2, achieving an enhanced NO-oxidation performance and good reusability (Fig. 7b).77 Besides, the three-coordinate nitrogen vacancy in amorphous carbon nitride was beneficial to expand its visible light responsive range, decrease the activation barrier of the NO triple bond, and boost the generation of 1O2 and ˙O2− for complete NO removal (Fig. 7c and d).78 The integration of ion doping and vacancy engineering has also been proposed to boost the photocatalytic activity of g-C3N4 towards NO removal.79 The introduction of N vacancies and K+ ions induced the formation of charge channels and achieved a preferred pathway (NO → NO+ → NO3−) for NO removal.
Fig. 7 (a) Proposed photocatalytic oxidation mechanism of NO over C3N4 with carbon vacancies.76 Reproduced from ref. 76 with permission from Elsevier, Copyright 2020. (b) Reaction mechanism for NO oxidation of CN and AC-CN4.77 Reproduced from ref. 77 with permission from Elsevier, Copyright 2020. Calculated reaction pathways for NO2 oxidation on (c) CN and (d) amorphous carbon nitride with N3C-site vacancies (ACN3).78 Reproduced from ref. 78 with permission from John Wiley and Sons, Copyright 2021. |
It is worth mentioning that not all types of vacancies are beneficial to increase the photocatalytic performance of materials. Different types of defects including OVs clusters, surface OVs, and bulk OVs were introduced in TiO2.80 The surface OVs not only significantly promoted the adsorption of H2O and facilitated charge transfer to the adsorbed H2O, forming ˙OH, but also conducive to the adsorption and desorption of NO, contributing to the best photocatalytic NO removal activity. Amorphous TiO2 with OVs clusters had no photocatalytic ability, and bulk OVs neither helped the adsorption of H2O nor improved the charge transfer to H2O. Rao et al. constructed position-manipulated OVs (OVs1 and OVs2) on the surface of (BiO)2CO3 (BOC) to maximize the formation of ROS for photocatalytic NO deep oxidation (Fig. 8a and b).37 The introduction of OVs1 (OVs1-BOC) induced the formation of an intermediate energy level and promoted the formation of ˙O2− and ˙OH (Fig. 8c–e), and the light absorption edge of BOC was redshift after the construction of OVs2 (OVs2-BOC), and thus only ˙O2− could be generated. OVs1-BOC (50.0%) exhibited a higher photocatalytic NO removal efficiency than OVs2-BOC (41.6%) under identical conditions, but OVs2-BOC possessed more stable photocatalytic activity (Fig. 8f). Bi2Sn2O7−x with distinct Sn/Bi-adjacent OVs (VO1–Bi2Sn2O7−x/VO2–Bi2Sn2O7−x) was synthesized and exhibited different photocatalytic NO oxidation performances (Fig. 8g–i).81 VO1–Bi2Sn2O7−x served as the adsorption sites to boost the adsorption and activation of NO and facilitated electron–hole separation for efficient NO removal, whereas VO2–Bi2Sn2O7−x became the recombination centers of charge carriers, and thus led to a decrease in the photocatalytic performance (Fig. 8j–n). Therefore, the rational construction of surface vacancies is the key to maximizing the effect on elevating the photocatalytic performance.
Fig. 8 Crystal structure models of (a) OVs1-BOC and (b) OVs2-BOC. Comparison of DOS/PDOS between (c) BOC and OVs1-BOC and (d) BOC and OVs2-BOC. (e) DMPO spin-trapping ˙O2− and ˙OH ESR spectra of BOC, OVs1-BOC, and OVs2-BOC under visible light irradiation. (f) Formation and photocatalytic mechanism on BOC, OVs1-BOC, and OVs2-BOC.37 Reproduced from ref. 37 with permission from the American Chemical Society, Copyright 2021. Structure model of (g) pristine Bi2Sn2O7, (h) VO1–Bi2Sn2O7−x (Sn-adjacent), and (i) VO2–Bi2Sn2O7−x (Bi-adjacent) with oxygen vacancies obtained using theoretical calculations. (j) Photocurrent responses, (k) electrochemical impedance spectra, (l) steady-state photoluminescence, (m) surface photovoltage spectra, and (n) transient fluorescence decay spectra of Bi2Sn2O7, VO1–Bi2Sn2O7−x, and VO2–Bi2Sn2O7−x.81 Reproduced from ref. 81 with permission from the Royal Society of Chemistry 2021. |
Surface defects have been universally regarded as the vital active sites for highly efficient photocatalytic reaction, but predesigned surface defects undergo dynamic transformations in response to the inevitable structural changes in catalysts during the reaction upon exposure to light field/electric field/temperature, and thus the reconstructed defective surfaces may act as the real active sites for catalytic reactions. Recently, Dong's group devoted their efforts to elaborating the dynamic evolution of surface defects and revealing the true reaction mechanism with real-time monitoring via in situ electron paramagnetic resonance (EPR) technology, which is a powerful tool for the quantitative characterization of vacancies based on accurate single-electron detection (Fig. 9a and b).82,83 TiO2 and BiOCl with predesigned surface defects were employed as the model catalyst, and the formation and evolution of photoexcited dynamic OVs were verified by in situ EPR technology (Fig. 9c and d).82,83 The photoswitchable OVs were found to be the true active sites during the reaction, which facilitated the activation of reactants, and thus realized the efficient conversion of NO into nitrates.
Fig. 9 (a) Schematic of operando EPR measurements. (b) Picture of the custom-made in situ reactor chamber and illustration of the obtained EPR spectra during different stages. (c) Cycling process of photoexcited OV (PE-Vo) generation and self-recovery on TiO2 (PD-Vo: predesigned OVs).82 Reproduced from ref. 82 with permission from the American Chemical Society, Copyright 2023. (d) Schematic illustration of photocatalytic NO oxidation in which PE-OVs on BiOCl serve as real active sites under UV irradiation and the reversible creation of PE-OVs on BiOCl.83 Reproduced from ref. 83 with permission from the American Chemical Society, Copyright 2022. |
Surface vacancies can normally increase the photocatalytic activity, but there is still a possibility that unstable vacancies suffer from deactivation during continuous reaction. With regards to preventing the deactivation of OVs in Bi-based photocatalysts, additional modification of Bi nanoparticles may be an accessible strategy (Fig. 10a).84,85 The surface-deposited Bi acts the active sites to activate O2 and H2O, which avoids O2 and H2O filling into OVs, thus preventing the deactivation of OVs (Fig. 10b). Besides, Ran et al. proposed the UV-light-induced destruction of Bi–O and Sb–O bonds to introduce OVs in BiSbO4 (BiSbO4-OV) (Fig. 10c); meanwhile, the mechanisms for the inactivation and regeneration of OVs were clarified to prevent the deactivation of photocatalysts.86 Although oxygen in the air can fill the OVs during photocatalytic reaction to lead to the consumption of OVs, and thus the deactivation of BiSbO4-OV, the deactivated photocatalyst was regenerated by re-irradiation with UV-light to reintroduce OVs for inducing the photocatalyst to return to its initial state (Fig. 10d).
Fig. 10 (a) Schematic of the co-effect of OVs and Bi on the photocatalytic performance of Bi-deposited porous Bi2Ti2O7 possessing rich OVs.84 Reproduced from ref. 84 with permission from Elsevier, Copyright 2022. (b) Charge difference distribution of optimized H2O and O2 adsorption on Bi2O2CO3 (BOC), Bi2O2CO3 nanosheets with OVs (OV-BOC) and Bi metal nanoparticle-decorated Bi2O2CO3 nanosheets with OVs (Bi@OV-BOC).85 Reproduced from ref. 85 with permission from Elsevier, Copyright 2020. (c) Solid-state EPR spectra of BiSbO4-OV samples. (d) Formation, inactivation and regeneration process of OVs on BiSbO4 and photocatalytic NO oxidation mechanism.86 Reproduced from ref. 86 with permission from the American Chemical Society, Copyright 2019. |
Fig. 11 (a) Schematic of photocatalytic NO removal by Au/Bi5FeTi3O15.88 Reproduced from ref. 88 with permission from Springer Nature, Copyright 2022. (b) Schematic of charge separation on the integrated effect of the Schottky junction and LSPR effect and corresponding photocatalytic reaction on Ag nanoparticle-decorated SrTiO3.92 Reproduced from ref. 92 with permission from the American Chemical Society, Copyright 2022. |
Besides, the introduction of plasmonic metals can not only induce the LSPR effect but also construct a Schottky junction to synergistically promote the separation and transfer of carriers.92,93 Ag nanoparticle-decorated SrTiO3 showed a six-fold increment in NO conversion rate and significant decline in toxic NO2 simultaneously.92 The elevated photocatalytic performance could be attributed to the following: (1) the LSPR effect of Ag nanoparticles promoted the light absorption and charge transfer in the photocatalyst, (2) the formation of a Schottky junction facilitated the charge separation and induced electron transfer from the silver particles to SrTiO3 to enhance the lifetime of excitons, and (3) the formation of an Ag–O bond between Ag nanoparticles and SrTiO3 increased the charge density of adjacent Ti, which offered a favorable channel for the adsorption and activation of reactants for the subsequent formation of ROS and complete oxidation of NO (Fig. 11b).
The introduced metal nanoparticles can also serve as active sites to promote the adsorption and activation of reactants, contributing to the optimization of the reaction pathway for the highly efficient complete photocatalytic oxidation of NO. g-C3N4 with monodisperse Au nanoparticle decoration (Au@CN) exhibited highly enhanced NO oxidation activity and superior NO2 inhibition ability.94 Besides the accelerated adsorption and activation of reactants and separation efficiency of carriers achieved by the modification of Au nanoparticles, the unique electronic structure of Au@CN tended to promote nitrate formation with a low rate-determining barrier, and thus largely suppressed the formation of the toxic intermediate NO2 (Fig. 12a–d). Furthermore, Li et al. designed Pd nanoparticle-decorated g-C3N4 (PdCN) by DFT simulation firstly, and it was found that the PdCN was equipped with a lower activation barrier in the rate-determining step and higher selectivity for the final product (nitrate) instead of the toxic intermediate (NO2) (Fig. 12e and f).95 PdCN was also fabricated and exhibited outstanding photocatalytic NO complete oxidation (Fig. 12g and h), which is consistent with the theoretical prediction. Therefore, the rational utilization of theoretical prediction can provide a quick and efficient method to screen and design efficient photocatalysts.
Fig. 12 Calculated reaction pathway: calculated climbing image nudged elastic band (CI-NEB) reaction pathway (a and b) and structures of initial states, transitional states and final states (c and d) for NO photo-oxidation by ˙O2− on CN and Au@CN.94 Reproduced from ref. 94 with permission from Elsevier, Copyright 2019. Calculated reaction pathway for NO conversion by ˙O2− and optimized structures of adsorbed NO and NO2 on (e) CN and (f) PdCN. (g) NO conversion and (h) NO2 production on CN and PdCN.95 Reproduced from ref. 95 with permission from Elsevier, Copyright 2020. |
Non-noble plasmonic metal-based photocatalysts have been identified as alternatives to noble metal-based photocatalysts due to their advantages such as earth-abundance and cost effectiveness.96 Metal Bi, which is equipped with relatively small effective mass for the conduction of electrons, long mean free path, and high charge carrier mobility, exhibits interesting electronic properties. Correspondingly, metal Bi possesses a distinctive LSPR-response covering the entire UV-vis-NIR spectral region, and thus can directly drive the photocatalytic redox reaction. Dong et al. firstly found that Bi nanoparticles with a 100–200 nm size exhibited notable photocatalytic NO removal efficiency under 280 nm light irradiation, which was ascribed to the UV-mediated LSPR effect of Bi nanoparticles (Fig. 13a).97 Meanwhile, significant research has been directed to the rational design of Bi-decorated nanomaterials for photocatalytic NO removal, such as SiO2 nanoparticle-modified Bi microspheres,98 bismuth sphere-assembled graphene oxide,99 and Bi-decorated amorphous bismuth oxide.100 Although various Bi-decorated hybrid structures have been proposed in the earlier study, complete photocatalytic NO oxidation has only gradually attracted attention in recent years. Zhang et al. transformed Bi spheres with a surface amorphous Bi2O3 layer to a Bi@Bi2O2CO3 core–shell photocatalyst (Bi@C-BOC), realizing an increase in photocatalytic NO removal efficiency and the inhibition of NO2 formation (Fig. 13b).101 The crystalline Bi2O2CO3 layer was beneficial for the transfer of hot electrons on plasmonic Bi to Bi2O2CO3, contributing to the formation of ROS for the complete oxidation of NO (Fig. 13c).
Fig. 13 (a) Schematic of the plasmonic photocatalysis mechanism of Bi nanoparticles.97 Reproduced from ref. 97 with permission from the Royal Society of Chemistry, Copyright 2014. (b) Schematic of directional photogenerated electron transfer processes over Bi@C-BOC. (c) Proposed photocatalytic mechanism for enhancing the photocatalytic activity over Bi@C-BOC under visible light irradiation.101 Reproduced with permission from ref. 101. Reproduced from ref. 101 with permission from Elsevier, Copyright 2021. |
Besides, the integration of the LSPR effect and defect engineering has attracted widespread attention on Bi-decorated photocatalysts, which can synergistically promote the adsorption of reactants and the separation of carriers to increase the photocatalytic NO removal efficiency.102–106 Bi/BiOBr nanoflowers with OVs (BOB-25) were fabricated to achieve an enhanced photocatalytic performance and decreased NO2 generation.102 Concretely, NO tended to donate outer layer electrons to OVs to form NO+, and also additional active sites were introduced to promote the adsorption and activation of the reactants (Fig. 14a). Besides, plasmonic Bi with a larger work function than BiOBr served as an electron sink and the impurity level was formed because of the presence of OVs, and thus a directional transfer channel for electrons was formed, which inhibited the recombination of photoinduced carriers (Fig. 14b). Furthermore, Bi/Bi2O2−xCO3 nanosheets with surface OVs promoted the generation of H2O2 by capturing electrons from the defect states of Bi2O2−xCO3via the two-electron reduction of O2, and also the surface OVs in Bi–O layers provided a channel for electron transfer between Bi and Bi2O2−xCO3 to increase the charge separation efficiency, which induced the highly efficient complete oxidation of NO (Fig. 14c and d).103 A unique electron transfer covalent loop ([Bi2O2]2+ → Bi-metal → O2−) was formed in Bi-metal@Bi2O2[BO2(OH)] with OVs, which could guide the directional transfer of carriers to improve the separation efficiency of charge carriers and the yield of ROS (Fig. 14e).104 Simultaneously, the Bi metal functioned as an electron donor to activate NO, and thus induce a new reaction path of NO → NO− → NO3−, inhibiting the generation of toxic intermediates (Fig. 14f).
Fig. 14 (a) Proposed NO adsorption and photocatalytic conversion routes over BOB-25. (b) Plausible diagram for ROS generation under visible light over BOB and BOB-25.102 Reproduced from ref. 102 with permission from Elsevier, Copyright 2023. (c) Possible mechanisms of H2O2 formation over Bi/Bi2O2−xCO3 nanosheet. (d) Schematic of charge transfer in the Bi/Bi2O2−xCO3 system and the possible mechanism of photocatalysis.103 Reproduced from ref. 103 with permission from Elsevier, Copyright 2019. (e) Charge density difference in Bi-metal@Bi2O2[BO2(OH)] with OVs: electron accumulation is labeled in blue and depletion in yellow, and the isosurfaces are set to 0.002 eV Å−3. (f) Principal photocatalytic oxidation reaction mechanism of NO on Bi-metal@Bi2O2[BO2(OH)].104 Reproduced from ref. 104 with permission from Elsevier, Copyright 2022. |
The surface decoration of functional groups also has been proposed to improve the photocatalytic performance. The introduction of hydroxyl functional groups, which are equipped with the ability to extend the light absorption, provide extra active sites, and promote the adsorption and activation of reactants, is one of most widely used strategies.107,108 Yang et al. found that the intercalation of –OH group in BiOI could enhance visible light absorption, accelerate the carrier transfer, and reduce the energy barriers for the conversion of NO2 into NO2− and NO3−.108 Besides, cyano/hydroxyl group co-functionalized g-C3N4 realized efficient photocatalytic NO conversion with an extremely low NO2 production ratio (4.8%).109 The insertion of cyano groups induced the optimization of the band structure to promote the formation of ˙O2− in g-C3N4, and the intermediate NO2, which tended to adsorb on the surface of hydroxyl groups instead of being released, could be complete oxidized by changing the conversion pathway from NO → NO2 to NO → NO2 → NO3−.
Fig. 15 (a) Proposed process of NO + O2 reaction over CdS/BaSnO3 under visible light irradiation.118 Reproduced from ref. 118 with permission from Elsevier, Copyright 2023. Schematic of the BiOI/Bi2O2SO4 p–n heterojunction: (b) before contact, (c) after contact, (d) under visible-light irradiation, and (e) visual representation of electron migration process from p-BIO to n-BSO (the background image is the atomic structure of the BiOI/Bi2O2SO4 p–n heterojunction).121 Reproduced from ref. 121 with permission from Elsevier, Copyright 2021. (f) Schematic of photocatalytic NO removal over 4% β-Bi2O3/CeO2−δ under visible light irradiation.122 Reproduced from ref. 122 with permission from Elsevier, Copyright 2021. |
Based on type II heterojunctions, the p–n heterojunction has been proposed, which can accelerate the electron–hole migration across the heterojunction by providing an additional internal electric field (IEF) for the optimization of photocatalytic performance.110,111 Generally, combining theoretical calculations and experimental characterization, the directional transfer of charge under the influence of IEF was confirmed.121–123 Geng et al. constructed the BiOI/Bi2O2SO4 p–n heterojunction, and thus a directional charge transfer channel induced by the IEF was established to allow photogenerated electrons to migrate from BiOI to Bi2O2SO4 (Fig. 15b–e).121 The boosted charge separation and transfer contributed to an electron-rich interfacial environment, promoting the reactant activation and ROS generation, and thus promoting the photocatalytic complete oxidation of NO. Besides, the synergistic effects of OVs and p–n heterojunctions was demonstrated, and the well-designed β-Bi2O3/CeO2−δ p–n heterojunction with OVs exhibited efficient deep oxidation of NO.122 Concretely, the light responsive range was broadened to the visible region and the spatial charge transfer was achieved by the IEF, and also the OVs could further promote the migration of charge carriers at the interface, which contributed to the production of ˙OH and ˙O2− participating in NO removal (Fig. 15f).
Another mainstream Z-scheme heterojunction has been widely acknowledged, aiming to not only accelerate the separation of carriers but also maximize the redox potential of the heterojunction system.124,125 For example, the covalent organic skeleton (COF)/TiO2 Z scheme heterojunction was constructed to achieve the highly selective oxidation of NO (Fig. 16a).126 The introduced COF improved the light capture capability and separation of photogenerated carriers, and the –CN– structure of COF with accumulated photogenerated electrons was identified as the activation sites of O2 to form ˙O2−; meanwhile, the photogenerated holes were inhibited to react with adsorbed water to generate ˙OH, further hindering the critical path of NO oxidation to produce toxic by-products NO2 (Fig. 16b–f). Noteworthily, other COF/MOx (M = Zn, Zr, Sn, Ce, and Nb) catalysts also exhibited superior selectivity and activity, meaning that this scheme is credited with universality. Besides, layered 2D/2D W18O49/g-C3N4 heterostructures were equipped with well-developed interfaces, thus showing extended light absorption from the visible light to NIR region and improved photocatalytic NO oxidation performances.127 The combination of a Z-scheme heterojunction and OVs also could synergistically promote the deep oxidation of NO.128–130 Besides, Lu et al. reported the preparation of the ternary Z-scheme photocatalyst SnO2/NCDs/ZnSn(OH)6 (NCDs = N-doped carbon quantum dots), which possessed strong redox ability for NO complete oxidation.131 The NCDs, as an electron transport bridge, improved both the broad-spectrum light-harvesting ability and the rapid separation of photoinduced electrons, contributing to the formation of ROS for NO removal (Fig. 16g).
Fig. 16 (a) Photoexcited carrier transfer process in COF and TiO2. (b) Mechanism of photocatalytic NO oxidation on COF/MOx. Charge difference distribution of adsorbed O2 and NO on (c and d) TiO2 and (e and f) COF, respectively.126 Reproduced from ref. 126 with permission from John Wiley and Sons, Copyright 2023. (g) Pathway of photocatalytic NO transformation and the reaction mechanism on the surface of the SnO2/NCDs/ZnSn(OH)6 Z-scheme nanohybrid under vis-NIR irradiation.131 Reproduced from ref. 131 with permission from the Royal Society of Chemistry, Copyright 2019. |
The S-scheme heterojunction, which is composed of reduction photocatalysts (RP) and oxidation photocatalysts (OP) with staggered band structures, has been proposed given that the pointless photogenerated charge carriers are recombined and a strong redox potential is introduced.132–134 Recently, the S-scheme heterojunction was also employed in photocatalytic NO removal.135–137 Moreover, Lv et al. developed a synergistic strategy to improve the photocatalytic NO oxidation efficiency with doping elements, building an S-scheme heterojunction and introducing OVs (0.1BTO/LTO-OV), which realized highly efficient NO removal with NO2 inhibition (Fig. 17a).135 Specifically, the doped electronic states and oxygen defect states narrowed the band gap to extend the light absorption range, and the S-scheme heterojunction formed between Bi-doped La2Ti2O7 and La-doped Bi4Ti3O12 enhanced the separation efficiency of photogenerated carriers and improved the transfer rate of electrons across the interface, which was conductive to the activation of the reactants and ROS formation for photocatalytic NO purification (Fig. 17b). A ternary g-C3N4/TiO2/Ti3C2 MXene photocatalyst was constructed, which possessed an interconnected nanosheet structure to increase the light absorption and provide a reaction interface between the reactants and photocatalysts.136 Simultaneously, the separation of photogenerated carriers was promoted by the formed S-scheme heterojunction and the Ti3C2 tightly bonded with TiO2 could accelerate the transfer of photogenerated holes, resulting in a high NO removal efficiency and low NO2 generation (Fig. 17c).
Fig. 17 (a) Schematic of charge transfer and radical generation over 0.1BTO/LTO-OV S-scheme heterojunction for NO removal by different routes. (b) Schematic generation of reactive oxygen species of 0.1BTO/LTO-OV.135 Reproduced from ref. 135 with permission from Elsevier, Copyright 2024. (c) Schematic of the energy band structure and electron–hole separation of g-C3N4/TiO2/Ti3C2.136 Reproduced from ref. 136 with permission from Elsevier, Copyright 2021. |
Furthermore, insulator–semiconductor heterostructures, as a new family of photocatalysts, have been investigated because earth-abundant insulators possess the advantages of low cost, easy availability and good stability. Dong's group constructed a series of insulator-based heterojunctions for photocatalytic NO oxidation, challenging the traditional opinion that free electrons cannot be transferred to insulators. For example, the SrCO3–BiOI core–shell structure exhibited an enhanced visible light absorbance between 400–600 nm and could efficiently convert NO into harmless nitrate via the NO → NO+ and NO2+ → nitrate or nitrite routes (Fig. 18a).138 As confirmed by DFT simulation and experimental characterization, the covalent interaction between the O 2p orbital of insulator (SrCO3, n-type) and the Bi 6p orbital of the photocatalyst (BiOI, p-type) provided an electron transfer channel between SrCO3 and BiOI by the formation of a p–n heterostructure, allowing photogenerated electrons to transfer from the photocatalyst to the conduction band of the insulator, thus promoting the formation of ˙OH and ˙O2− for photocatalytic NO oxidation (Fig. 18b and c). Similarly, the photogenerated electrons from the semiconductor (BiOI) could directly transfer to the insulator (BaCO3) through the electron delivery channel (Bi atoms → adjacent carbonate layer) due to the potential difference between the Bi layer of BiOI and the carbonate layer of BaCO3 under visible light irradiation, contributing to the utilization of free electrons on BaCO3 to produce ROS participating in deep oxidation of NO (Fig. 18d–f).139 Besides, alkaline earth metal sulfate-based heterostructures also could induce electron transfer from the semiconductor to insulator, and thus accelerate the separation of charge carriers for efficient NO removal.140,141
Fig. 18 (a) Photocatalytic NO transformation pathway and (b) visible light photocatalysis mechanism of the insulator-based SrCO3–BiOI core–shell heterojunction. (c) Charge difference distribution between BiOI and SrCO3: charge accumulation is given in purple and depletion in green, and the isosurface is set to 0.003 eV Å−3.138 Reproduced from ref. 138 with permission from the American Chemical Society, Copyright 2018. (d) Charge difference distribution between BiOI and BaCO3: charge accumulation is shown in blue and depletion in yellow, and the isosurface is set to 0.002 eV Å−3. (e) Schematic of the band configuration and the charge separation at the interface of BaCO3/BiOI heterojunctions under visible light irradiation; Φ is the work function, VD is the contact potential, E0 is the vacuum level, Ec is the bottom of the conduction band, Ev is the top of the valence band, Eg is the band gap, Ef1 and Ef2 are the Fermi levels of BiOI and BaCO3, respectively, and Ef is the Fermi level of the BaCO3/BiOI heterojunction. (f) Proposed schematic diagram of the separation and transfer of photogenerated carriers and the photocatalytic process over the insulator–semiconductor heterojunction.139 Reproduced from ref. 139 with permission from the Royal Society of Chemistry, Copyright 2018. |
Fig. 19 (a) Number of acid sites on Zr-T and Zr-M. (b) Evolution process of ROS and photocatalytic NO oxidation process over Zr-M.27 Reproduced from ref. 27 with permission from the American Chemical Society, Copyright 2023. (c) Energy level diagram arrangement of ZnO prepared from different precursors and calcination temperatures. (d) Photocatalytic mechanism of ZnO with mixed exposure of {101(−)1} and {0002} planes.143 Reproduced from ref. 143 with permission from Elsevier, Copyright 2022. |
Numerous facet-dependent photocatalysts also have been designed to optimize the band structure and surface electronic structure, which can adjust the adsorption and activation of reactants on the active sites and the behavior of carrier dynamics for achieving photocatalytic NO deep oxidation. The ZnO crystal with mixed exposure of {101(−)1} and {0002} planes exhibited a superior photocatalytic performance to that with exposed {0002} facets for NO removal. The nanocrystals with predominant {0002} crystal plane led to a shrinkage in band structure, and thus the low valence band position and poor affinity for NO decreased the NO removal efficiency and selectivity for NO3− production (Fig. 19c and d).143 SrBi2Ta2O9 with a higher {200} crystal plane exposure ratio showed higher photocatalytic NO deep oxidation activity compared to {001}-BiO and {001}-TaO facets.144 The promoted conversion of NO into NO3− was attributed to the greater exposure of {200} planes, which (1) facilitated the adsorption of NO; (2) endowed electron-enriching ability between the two types of the {001} layers to induce polarized electro-field for charge separation; and (3) provided more active sites for the formation for ˙O2−.
Crystal structure control of Bi-based photocatalysts has been widely employed, and photocatalytic NO deep oxidation has also been reported.145–149 The different exposed facets of Bi-based photocatalysts can provide various activate sites for the adsorption of reactants, which have an effect on ROS formation and the subsequent NO oxidation reaction. Bi2O2CO3 with {001} facets (001-BOC) exhibited much higher photocatalytic activity than that with {110} facets (110-BOC) given that the different adsorption and activation patterns directed the conversion pathway of NO deep oxidation.145 In combination with time-dependent in situ DRIFTS spectra and DFT calculation results, 001-BOC enabled the transformation of NO into NO− or cis-N2O22−, while 110-BOC induced NO to be activated into NO+ or N2O3 during the adsorption and activation process, which is conductive to the subsequent oxidation of the intermediates to the final products (NO3−) and the desorption of NO3− on 001-BOC (Fig. 20a). Li et al. fabricated BiOCl with tailored crystal planes, where efficient ROS generation and smoother NO conversion on the {010} facet-exposed BiOCl promoted the complete oxidation of NO.146 Specifically, in comparison with the {001} facet-exposed BiOCl (BOC-001), the alternating distribution of [Bi2O2]2+ and Cl− on the {010} facet of BiOCl (BOC-010) not only promoted H2O2 dissociation, which is the rate-determining step of ˙O2− generation (˙OH → H2O2 → ˙O2−), but also decreased the activation energies of NO2 to suppress NO2 accumulation on the photocatalyst (Fig. 20b–d).
Fig. 20 (a) DFT-calculated adsorption energy and bond length of several major intermediate adsorption products on 110-BOC and 001-BOC; all lengths are given in Å.145 Reproduced from ref. 145 with permission from the Royal Society of Chemistry, Copyright 2019. Calculated reaction pathways and activation energies (Ea) for (b) ˙OH and (c) ˙O2− induced NO conversion on BOC-001 and BOC-010. (d) Calculated reaction pathways and activation energies (Ea) for ROS generation and transformation on BOC-001 and BOC-010.146 Reproduced from ref. 146 with permission from Elsevier, Copyright 2019. |
Besides, the geometric structures of OVs on different crystal faces of BiOCl are totally different, affecting the transfer and transformation of surface charge to adjust the photocatalytic NO oxidation reaction.150 BiOCl with exposed {010} facets displayed an open channel structure with O, Bi, and Cl atoms exposed, and OVs could localize two electrons to form an archetypal F center . BiOCl with exposed {001} facets possessed a closely packed O-exposed structure, and OV was a single electron-trapped center . During the photocatalytic NO oxidation process, induced two-electron charging and subsequent one-electron decharging , and the back-donated electron was retrapped by VO to produce a new single-electron-trapped for a second round of selective NO oxidation (Fig. 21b). Noteworthily, could directly oxidized NO by O22− into more stable bidentate nitrate in an exothermic process with a small energy barrier, but the -mediated NO oxidation product possessed a remarkable geometric difference as monodentate nitrate (Fig. 21a). Besides, one could oxidize two molecules of NO, resulting in a remarkably higher atomic vacancy utilization efficiency (Fig. 21c and d). The interfacial charging–decharging strategy by the synergistic effect of crystal structure control and vacancy engineering realized highly efficient NO complete oxidation.
Fig. 21 (a) Free energy change during O22−- and ˙O2−-mediated NO oxidation on the VO of BOC-010 and BOC-001, respectively. (b) Schematic illustration of the charging–decharging scheme on the of BOC-010. (c) Corresponding NO oxidation rate dynamics and corresponding diagram of reaction process (inset). (d) Schematic illustration of NO removal and corresponding electronic excitation process on .150 Reproduced from ref. 150 with permission from the American Chemical Society, Copyright 2019. |
Goldstein et al. firstly demonstrated the potential feasibility of photocatalytic reduction NO into NH3 (Fig. 22a).159 TiO2 loaded with 10–130 electrons per particle, which was constructed by using γ-irradiation of acidic TiO2 colloid solutions containing 2-propanol, was studied for NO reduction under anaerobic conditions. There was competition among the reaction paths to produce NH3, N2O and N2, where the complete reduction of NO into NH3 proceeded via five consecutive one-electron transfer reactions. Inspired by the electrocatalytic NO reduction process, a 0D/3D g-C3N4 QD/three-dimensional macroporous and mesoporous (3DOMM) TiO2−x electrode was developed and its performances in both electrocatalytic and photocatalytic reduction of NO to NH3 were investigated (Fig. 22b).160 The g-C3N4 QD/3DOMM-TiO2−x electrode exhibited a much higher NH3 yield of 3841.5 μg h−1 mg−1 during the electrocatalytic test, and the photocatalytic NO reduction reaction was carried out in a single-compartment cell using g-C3N4 QD/3DOMM-TiO2−x as the photoelectrode. The NH3 yield in the photocatalytic test was estimated to be 95.07 μg h−1 mg−1, which was attributed to the fact that g-C3N4 QDs/3DOMM-TiO2−x possessed the following features: (1) a large specific surface area to provide surface active sites for the adsorption of the reactants, (2) macroporous and mesoporous channels to accelerate the mass and charge transfer, and (3) S-scheme heterostructure to facilitate charge separation. However, although the photocatalytic NO reduction route was initially developed, the photocatalytic efficiency and product selectivity still need to be further optimized.
Fig. 22 (a) Proposed mechanism of NO reduction to NH3 by TiO2 electrons in colloid solution via consecutive one-electron transfer steps.159 Reproduced from ref. 159 with permission from the American Chemical Society, Copyright 2015. (b) Schematic diagram of photocatalytic NO reduction to NH3 on g-C3N4 QD/3DOMM-TiO2−x electrode.160 Reproduced from ref. 160 with permission from Elsevier, Copyright 2023. |
It is worth noting that the ultralow solubility of gaseous NO limits its transfer to liquid-phase reaction systems for its subsequent conversion and upcycling, leading to an unsatisfactory performance. Therefore, integrated processes combining continuous chemical absorption and electrocatalytic/biological reduction have emerged for NO removal from flue gas.161–163 Recently, Li et al. developed an on-site coupling system by combining continuous chemical absorption and photocatalytic reduction reaction to increase the solubility of NO for facilitating subsequent NO upcycling, achieving outstanding conversion efficiency of NO (89.05% ± 0.71%) and production selectivity of NH3 (95.58% ± 0.95%) (Fig. 23a).29 In the synchronous NO absorption and conversion system, an Fe(II) ethylenediaminetetraacetic acid complex [Fe(II)EDTA] was used as the absorbent to generate Fe(II)EDTA–NO, and the Au nanoparticle-decorated TiO2 photocatalyst was fabricated by an operando deposition method during the photocatalytic reaction. Also, formaldehyde (HCHO) was introduced as an antioxidant to avoid the transformation of Fe(II) into Fe(III) and a quenching agent of photogenerated holes, which realized continuous NO reduction and Fe(II)EDTA regeneration.
Fig. 23 (a) Illustration of the on-site coupling system of continuous chemical absorption and catalytic reduction of NO.29 Reproduced from ref. 29 with permission from the American Chemical Society, Copyright 2023. (b) Illustration for the coupled continuous absorption and conversion process of NO assisted by SO2 poison.158 Reproduced from ref. 158 with permission from the American Chemical Society, Copyright 2023. |
Generally, the NO released by anthropogenic sources is produced from the process of fossil fuel combustion, together with multicomponent air pollutants.164,165 Sulfur dioxide (SO2) is recognized as the main coexisting species with NO, and thus the potential impact of SO2 on NOx purification should be seriously considered. Correspondingly, the simultaneous removal and utilization of NOx and SO2 in the flue gas has attracted interest from researchers.166,167 A redox pair of SO2–NO, namely the synergetic SO2 oxidation and NO reduction driven by absorption–photocatalysis system, was proposed, using Fe(II)EDTA as the absorbent and commercial TiO2 as the photocatalyst (Fig. 23b).158 The constructed redox pairs of SO2–Fe(III) hindered the side reaction of Fe(II)-to-Fe(III) oxidation to guarantee the sufficient concentration of Fe(II)EDTA for continuous NO absorption in the absorption process, and with regard to the photocatalytic reaction process, the redox pair of SO2–NO facilitated the separation of photogenerated charge carriers by h+-induced SO2 oxidation reaction and e−-promoted NO reduction reaction. This contributed to ultrahigh selectivity for both NO-to-NH3 upcycling and SO2-to-SO42− purification under ambient conditions.
(1) Photocatalysts are not indestructible and their deactivation becomes noticeable at certain timescales, which can affect the reaction efficiency and even hinder their commercial viability. However, deactivation studies have not mirrored the recent advances in improving the stability of photocatalysts and exploring available regeneration methods, which deserves more attention and effort for advancing practical applications.
(2) In actual scenarios where NO pollution exists, multicomponent pollutants inevitably coexist, while most research only focused on the single NO pollutant. The interaction between different pollutants should be considered, including the synergistic/antagonistic effect and corresponding mechanism, which can lay the foundation for the application of photocatalytic technology in real situations.
(3) In the current design route of photocatalysts, powder materials are the most common form. However, powdery photocatalysts suffer from loss under the action of air/water flow during the reaction process, which not only decreases the reaction efficiency and increases the running cost, but also leads to secondary pollution. In this case, supporting photocatalysts by loading them on a carrier material may be a feasible strategy, which can induce the powder to be firmly supported, and also provide a good reaction site.
(4) With regard to the ultralow solubility of gaseous NO, integrated processes combining chemical absorption and reduction have been proposed to promote the absorption of NO for the convenience of subsequent photocatalytic reduction process. In the process of practical application, the maturity, economy and operability of the coupling technologies should be the primary consideration.
(5) In the current research on NO reduction reaction, it is preferred to pursue the excellent synthesis rate of NH3; however, continuous and efficient NO conversion should be the most important target given that it is the primary obstacle in the further development of photocatalytic NO purification and upcycling technology. It is suggested that the rational design of the reaction system (including photocatalyst, reaction medium, and reactor) will facilitate the synergistic promotion of NO conversion efficiency, ammonia selectivity, ammonia yield, ammonia recovery and long-term stability.
(6) It is important and urgent to establish in situ characterization methods with high spatial and temporal resolution to dynamically monitor the reaction process, and thus reveal the true active sites and conversion pathway of reactants, which can essentially guide the precise design of photocatalysts and the reaction pathway.
(7) Theoretical simulation methods have been widely employed to clarify the crystal structure of photocatalysts and reaction pathways, but it is worth noting that the theoretical simulation should be based on experimental results. Thus, optimizing the matching degree between theoretical simulation methods and experiments should be seriously considered.
Footnote |
† These authors contributed equally to this work. |
This journal is © The Royal Society of Chemistry 2025 |