Molecular sieves are inorganic crystal materials with regular pore structure formed by the connection of TO₄(T=Si, Al, etc.) tetrahedra with common vertices, which have abundant acid sites and ion exchange sites and can be used as catalyst supports. According to the size of the carrier pore, it can be divided into small pore molecular sieve, medium pore molecular sieve and large pore molecular sieve. Compare with that medium-macropore molecular sieve, the small-pore molecular sieve loaded with active component (Cu, Fe and the like) has more excellent NH₃-SCR activity. Typical small pore molecular sieves used for NH₃-SCR reactions are CHA, AEI, LTA, etc. In 2010, Kwak et al. First compared the catalyst performance of Cu-ZSM-5, Cu-Beta and Cu-SSZ-13, and Cu-SSZ-13 showed excellent catalytic activity and hydrothermal stability^[5]. Subsequently, Cu-based small pore molecular sieves, represented by Cu-CHA, have attracted wide attention and have been commercially used in diesel vehicle NO_x catalytic purification abroad^[6].

The molecular sieve with CHA configuration has a topology of double six-membered rings (d6r) connected by four-membered rings to form an ellipsoidal cage (CHA cage) with a three-dimensional eight-membered ring structure, and the size of the eight-membered ring orifice is ~ 3.88 Å^[7⇓⇓~10]. When copper is exchanged on the molecular sieve with CHA structure, the copper ion can migrate in the pore channel and can be stabilized in d6r at high temperature, so the molecular sieve has excellent NH₃-SCR activity.

in the early studies on the active site of Cu-CHA, Korhonen et al. Studied Cu-SSZ-13 based on in situ UV-vis DRS, X-ray diffraction (XRD) and X-ray absorption fine structure (XAFS), and first proposed that there was only one monomer in situ UV in the molecular sieve structure, and these monomer in situ UV species existing in the six-membered ring were the active center^[11]. Subsequently, Kispersky et al. Studied Cu-SSZ-13, Cu-SAPO-34 and Cu-ZSM-5 by X-ray absorption spectroscopy (XAS) and found that under the steady-state conditions of standard SCR reaction, the active centers of these copper-based molecular sieve catalysts are Cu⁺ and Cu²⁺,NH₃-SCR, and the reaction is based on the redox process of Cu⁺ and Cu²⁺, while the ratio of Cu⁺/Cu²⁺ depends on the structure of the molecular sieve^[12]. Deak et al. Found that the monomer Cu²⁺ located in the double six-membered ring was the active center of Cu-SSZ-13 by XAFS and XRD refinement^[13].

Kwak et al. Used H₂-TPR and Fourier transform infrared spectroscopy (FTIR) to study the Cu-SSZ-13 catalyst, and proposed that there are two distinct copper ion sites in the Cu-SSZ-13 molecular sieve structure. When the copper exchange amount is low, the copper ion preferentially occupies the relatively stable six-membered ring position, while when the copper exchange amount is large, the redundant copper ion needs to occupy the large cage position in the CHA structure^[14]. Gao et al. Used EPR and H₂-TPR to study the effect of different ion exchange capacity on the distribution of Cu²⁺ active species in Cu-SSZ-13. At low exchange capacity, Cu²⁺ is only located at the six-membered ring position (connected with two framework aluminum), which is the active center of SCR reaction^[15]. It continues to increase the amount of copper ion exchange, forming Cu₂O_y(y≥1), which is located in the eight-membered ring and promotes the non-selective oxidation of NH₃. Verma et Al. Studied the NO oxidation mechanism of a series of Cu-SSZ-13 (Si/Al ratio of 4.5) with different Cu/Al ratios, and found that when the monomer Cu²⁺ was only located in the six-membered ring, the maximum Cu/Al ratio could not exceed 0.2, and if it exceeded this limit, part of the monomer Cu²⁺ turned into copper dimer^[16]. This theoretical limit is mainly determined by the Si/Al ratio of the molecular sieve.

Wang et al. Also pointed out that the silicon content in Cu-SAPO-34 catalyst would affect the position of Cu²⁺ species in the catalyst, that is to say, the existing state of copper species was not unique^[17]. At lower copper loadings, only monomeric Cu²⁺ was present in Cu-SSZ-13, whereas both CuO_x and Cu²⁺ were present in Cu-SAPO-34 even at very low copper loadings^[18]. Paolucci et Al. Considered that Cu²⁺ occupied the 2Al site first, and then occupied the 1Al site after the 2Al site was saturated, existing in the form of [Cu(OH)]⁺. In Cu-SSZ-13, the relative occupancy ratios of Cu²⁺-2Z and [Cu(OH)⁺]-Z were related to Si/Al and Cu/Al ratios^[19].

To sum up, a basic consensus has been reached on the copper active site distribution of copper-based small pore molecular sieves, that is, copper ions will preferentially occupy six-membered ring sites to form Z₂Cu species, and with the increase of copper content, redundant copper ions will occupy eight-membered ring sites to form ZCu (OH) species.

At present, it is generally believed that for small pore molecular sieves with medium and low copper content, the monomeric copper ion exists in the six-membered ring and is the active site for NH₃-SCR reaction. The preparation method, molecular sieve framework structure, Si/Al ratio, copper content and other factors may affect the location and existing state of metal active species in small pore molecular sieve catalysts, and the situation of active sites will be more complex under real NH₃-SCR reaction conditions, as well as after aging and poisoning. Although researchers have generally confirmed that monomeric copper ion is the active site of Cu-CHA, whether copper ion is the only active site and the understanding of its NH₃-SCR reaction mechanism are still controversial^{[6,9,19⇓ ~21]}.

Based on density functional theory (DFT) calculations and operando characterization, Paolucci et al. Further identified the Cu²⁺ of the six-membered ring in SSZ-13 as the active center for the low-temperature SCR reaction^[22]. It is believed that the SCR reaction is related to the redox cycle between Cu²⁺ and Cu⁺/H⁺, which are located in the six-membered ring and connected by two framework Al. The reaction mechanism of NH₃-SCR based on copper ion as a single active site was proposed by using {{\mathrm{NH}}_4}^{+} as the core reaction species. Then they verified through experiments and DFT simulations that the monomer Cu²⁺ first occupies a six-membered ring connected with two Al, and when these positions are saturated, the copper species may be connected with the OH ligand ([CuOH]⁺) and then bonded with Al). These copper active species are sensitive to the environment, and will be solvated by NH₃ and can migrate under NH₃-SCR conditions^[19]. They also verified that Cu⁺ was solvated by NH₃ to form mobile [Cu(NH₃)₂]⁺ transient ions under low temperature NH₃-SCR conditions through steady-state and transient experiments, XAS and first-principle theoretical calculations.O₂ activation step is the rate-determining step of the reaction. However, the migration of this part of Cu⁺ is carried out in a defined region under electrostatic binding, which leads to the reaction on the copper active site is neither a traditional heterogeneous reaction nor a homogeneous reaction, but a dynamic multi-site catalytic reaction^[23]. The low temperature SCR reaction rate of Cu-based molecular sieve is related to the distribution of copper species, which depends on the composition and topology of the molecular sieve. Therefore, optimizing the spatial distribution of copper species to promote copper migration can accelerate the redox reaction of Cu⁺ with Cu²⁺, thereby improving the low-temperature performance. At high temperature, Cu²⁺ loses the solvation of NH₃ and is anchored to the molecular sieve framework to become the active center. The change of the active site from Cu²⁺ (low temperature) to monomeric Cu²⁺ (high temperature) is also the reason for the seagull-type activity of Cu/SSZ-13 under standard SCR conditions (as shown in Fig. 1)^[24].

To sum up, there are mainly two kinds of active sites, :Cu²⁺-2Z(Z₂Cu) and [Cu(OH)⁺]-Z(ZCuOH), which determine the catalytic activity of Cu-based small pore molecular sieves, and their distribution and existing state are affected by many factors, so it is of great significance to clarify the structure of Cu-based small pore molecular sieves and the distribution of active sites for the improvement of their sulfur poisoning mechanism and sulfur tolerance.

The sulfur poisoning of Cu-based small pore molecular sieves is closely related to the type, concentration, temperature and atmosphere of sulfur oxides, and different sulfur species will be formed under different conditions^[30]. Zhang et al. Studied the effect of SO₂ on the NH₃-SCR performance of Cu-SAPO-34. At low temperature (< 300 ℃), SO₂ obviously inhibited the catalytic performance of Cu-SAPO-34, because the formed ammonium sulfate covered the active sites and blocked the pore channels of the molecular sieve, while SO₂ and NO formed competitive adsorption, which affected the adsorption of NO_x on the catalyst^[31]. Brookshear et al. Also studied the effect of SO₂ on the NH₃-SCR properties of Cu-CHA, and obtained the same conclusion as Zhang et al^[32][31].

Jangjou et al investigated the effect of SO₂ on the oxidation of NH₃ during the NH₃-SCR of Cu-SAPO-34. When the temperature was below 350 ℃, SO₂ significantly inhibited the oxidation of NH₃, but had no significant effect at high temperature^[33]. They compared the effect of sulfate such as ammonium sulfate and copper sulfate on the SCR activity of Cu-SAPO-34 by changing the way of sulfurization, and found that ammonium sulfate and three different forms of copper sulfate would be formed on the surface of Cu-SAPO-34 when SO₂ and NH₃ were used at the same time, while only copper sulfate existed when only SO₂ was used. By in situ infrared spectroscopy of NO adsorption, it was found that the copper species located at the six-membered ring were completely poisoned after sulfidation, while the copper species located at the eight-membered ring were only partially poisoned, but the environment around the copper active site was changed. When the sulfur on the surface of Cu-SAPO-34 exists in the form of ammonium sulfate, the catalyst can be regenerated and recovered after heat treatment at 350 ℃, while the formed copper sulfate can be decomposed at a temperature higher than 480 ℃^[34].

They also further studied the effect of SO₂ on the two active sites of Cu-SSZ-13 catalyst, Z₂Cu and ZCuOH. It was found that at low temperature, at the Z₂Cu site, SO₂ reacted with NH₃ to form ammonium sulfate, while SO₂ reacted with ZCuOH to form copper bisulfite species, which was further oxidized to form copper sulfate species, resulting in the reduction of active components, thus inhibiting the activity of Cu-SSZ-13 catalyst. The active site of ZCuOH is more susceptible to SO₂ than Z₂Cu, generating stable copper bisulfate or copper sulfate species, resulting in a decrease in the catalytic performance of Cu-SSZ-13 (as shown in Fig. 2)^[35]. Wang et al. Comparatively studied the effect of SO₂ on Cu-LTA and Cu-SSZ-13, and found that the ZCuOH active site in Cu-LTA catalyst was also more easily affected by SO₂ than that in Z₂Cu, and the ZCuOH site in Cu-LTA was more easily poisoned by sulfur than that in Cu-SSZ-13 due to its weaker interaction with the framework^[36].

Jiang et Al. Found that H-SSZ-13 did not adsorb SO₂, but in the presence of copper species, Si-OH-Al could adsorb SO₂ to form sulfuric acid species, and the ammonium sulfate formed by the adsorption of SO₂ on the active sites of Z₂Cu blocked the pores, resulting in the reduction of acid sites of the catalyst, while the copper bisulfate or copper sulfate species formed on ZCuOH caused the reduction of copper active sites, and the activity of this part could not be restored after regeneration at 550 ℃^[37]. Shen et al. Used different concentrations of SO₂ to vulcanize Cu-SAPO-34 for different time, and found that the catalytic activity of Cu-SAPO-34 decreased in different degrees after vulcanization treatment, which was due to the decrease of monomer Cu²⁺, but the activation energy of NH₃-SCR reaction of Cu-SAPO34 was not changed by SO₂ poisoning^[38]. Wijayanti et al. Studied the effect of SO₂ on standard SCR, fast SCR and slow SCR reactions of Cu-SAPO-34 and Cu-SSZ-13 catalysts, and found that the adsorption heat increased after sulfur poisoning by means of microcalorimeter measurement and hydrogen temperature-programmed reduction (H₂-TPR).Sulfur has little effect on the storage site of NH₃, but mainly affects the copper adsorption site and its redox ability. Sulfur has great effect on the standard SCR reaction, but has little effect on the fast and slow SCR reactions. This is because ammonium nitrate formed by NO₂ at low temperature covers the copper active site and reduces the SCR activity. When the temperature rises, ammonium nitrate decomposes and the SCR activity recovers^[39,40].

In diesel aftertreatment system, the regeneration of particulate filter (DPF) produces high temperature (> 650 ℃), which puts forward high requirements for the hydrothermal stability of NH₃-SCR catalyst downstream of DPF. Studies on the effect of H₂O on Cu-based small-pore molecular sieves show that the hydrothermal deactivation of Cu-based small-pore molecular sieves is mainly due to the hydrolysis of framework aluminum and the migration and transformation of active copper species Cu²⁺ into CuO_x and other species, resulting in the collapse of molecular sieve framework^{[41⇓⇓⇓⇓⇓~47]}.

When H₂O and SO₂ coexist, the effect on the NH₃-SCR performance of Cu-based small pore molecular sieves is greater. Shan et al. Found that when 100 ppm SO₂ was introduced during hydrothermal aging, the catalytic performance of the sample was lower than that of the sample aged alone, because the SO₂ accelerated the removal of framework aluminum, resulting in the reduction of acid sites and the generation of more CuO_x species^[48].

Zhang et Al. Also studied the synergistic deactivation mechanism of hydrothermal aging and SO₂ on Cu-SSZ-13, and found that hydrothermal aging alone mainly destroyed the bridge hydroxyl Si-OH-Al of the molecular sieve, resulting in the conversion of active species ZCu (OH) into CuO/CuAlO_x species^[49]. However, SO₂ aging alone makes ZCu (OH) transform into Z₂Cu, but when hydrothermal aging and SO₂ poisoning occur simultaneously, the destruction of Si-OH-Al sites is accelerated, more CuO/CuAlO_x species are formed, and even the collapse of molecular sieve framework is caused.

Current research on sulfur poisoning mainly focuses on the impact of SO₂, while the oxidation catalyst (DOC) located upstream of SCR catalyst in diesel exhaust after-treatment system will oxidize part of the SO₂ into SO₃,SO₃, resulting in more serious deactivation of copper-based molecular sieve catalyst^[50]. The SCR activity of Cu-SSZ-13 was more affected by SO₃ than that of mesoporous and macroporous molecular sieves^[51].

Kumar et al. Studied the effect of SO₂ and SO₃ on the catalytic performance of Cu-CHA. The ratio of SO₂/SO₃ was controlled by adjusting the temperature of DOC. At low temperature (200 ℃), SO₂ and SO₃ were completely and reversibly adsorbed on the same active site and had the same effect on the catalytic performance^[52]. At high temperature, the (400℃),SO₂ only affects the catalyst by adsorption, and its influence is very small, while the SO₃ causes serious deactivation of the catalyst, and it is difficult to recover, because the SO₃ not only adsorbs on the active site, but also reacts with the active site of the catalyst. Hammershøi et al. Studied the reversible and irreversible deactivation caused by SO₂ and SO₃ on Cu-CHA catalyst, and found that at 200 ℃, the presence of SO₃ in the reaction atmosphere would lead to more serious deactivation, but at 550 ℃, there was little difference between the presence and absence of SO₃ in the reaction atmosphere. The degree of irreversible deactivation was proportional to the amount of copper sulfate species formed in the catalyst, and reversible deactivation was mainly caused by lower SO₂/Cu ratio^[53]. They also further studied the effect of oxidation state of copper species in Cu-SAPO-34 on sulfur poisoning of catalyst, and found that the degree of sulfur poisoning deactivation of Cu-SAPO-34 mainly depends on the total exposure of SO₂, that is, it is related to the concentration of SO₂ in the atmosphere and exposure time.The results of DFT calculation show that the SO₂ is more easily adsorbed on Cu (Ⅰ), and Cu (Ⅱ) is easily reduced to Cu (Ⅰ) in low temperature SCR atmosphere, resulting in a higher S/Cu ratio of the catalyst, while the combination of SO₃ and Cu (Ⅱ) is stronger, resulting in a more serious sulfur poisoning of the catalyst^[54].

To sum up, previous studies have shown that the reduction of SCR activity of copper-based small pore molecular sieves after exposure to SO₂ is mainly due to the reduction of copper active sites, and the species and content of sulfur oxides generated may be different under different reaction conditions.ZCuOH located in the eight-membered ring is more likely to form copper sulfate and other related species, while Z₂Cu located in the six-membered ring is more likely to form ammonium sulfate, which blocks the pore and causes the decrease of activity^[12,55,56]. When SO₂ and H₂O coexist, the effect is greater, and SO₂ accelerates the removal of framework aluminum and forms more CuO/CuAlO_x species. The stronger combination of SO₃ and Z₂Cu located in the six-membered ring will also cause more serious deactivation of copper-based small pore molecular sieves.

By introducing a second metal including a rare earth metal and a transition metal as a sacrificial site on the copper-based microporous molecular sieve, the reaction between the SO₂ and the copper active center is avoided,The doped metal can also adjust the acidity and redox performance of the catalyst, and the increase of the acidity of the catalyst can reduce the adsorption of SO₂, thereby improving the NH₃-SCR catalytic activity and sulfur tolerance of the catalyst.

Fe-based catalysts have good sulfur resistance, and researchers have doped Cu-based small pore molecular sieves with Fe to improve their sulfur resistance. Yin et al. Synthesized Fe/Cu-SSZ-13 catalyst by a simple impregnation method, which has higher activity and resistance to SO₂ poisoning compared with Cu-SSZ-13. Through H₂-TPR, XPS, ESR and other characterization, it was found that the Fe species in the catalyst doped with Fe by impregnation method mainly exist in the form of Fe species in low valence oxidation state, which has better high temperature performance and sulfur resistance^[57]. Zhang et al. Introduced Fe and CeMn into Cu-SAPO-34 catalyst by impregnation method, among which 1% Fe/1% MnCe/3.86% Cu-SAPO-34 catalyst had the best SO₂ resistance and wider temperature window, and the increase in the number of Lewis acid sites and the existence of Fe species with lower oxidation state (Fe²⁺) were the reasons for the improved SCR activity of the catalyst^[58]. Zhang Weiping et al. Found that the bimetallic doped Cu/Fe-SSZ-39 catalyst has more and more stable ZCu (OH) species than Cu-SSZ-39 catalyst, and the presence of Fe species can inhibit the formation of ammonium sulfate, and as a sacrificial site, it can reduce the poisoning effect of Cu²⁺ active species by SO₂, thus improving its sulfur resistance^[59]. Yu et al. Prepared composite catalysts of Cu-SSZ-13 and oxides (Mn, Co, Ni, Zn, etc.) by sol-gel method, among which the ZnTiO_x-Cu-SSZ-13(1:5) catalyst had the best sulfur resistance, and ZnTiO_x as a sacrificial component reacted with SO₂ first, thus protecting the copper active sites from sulfation and improving the sulfur resistance of the catalyst^[60].

Li et al. Modified the surface of the catalyst and covered the surface of Cu-SAPO-18 catalyst with CeO₂ in the form of a thin film. The CeO₂ film can inhibit the oxidation of copper to copper oxide at high temperature, and also inhibit the formation and deposition of sulfate species, so that the modified catalyst has better catalytic activity and SO₂ resistance^[61]. Han et al. Prepared Ce-Cu-SAPO-18 catalyst by doping Ce into Cu-SAPO-18 catalyst, and found that sulfate would be preferentially adsorbed on Ce species, and the sulfate on Ce-Cu-SAPO18 catalyst was easier to decompose, thus improving the sulfur resistance and NH₃-SCR catalytic performance of the catalyst^[62]. Yu et al. Found that yttrium ions could inhibit dealumination during hydrothermal aging of Cu-SSZ-39 doped with rare earth metal yttrium, and could react with SO₂ as sacrificial sites to improve sulfur resistance. The NO_x conversion of Cu (0.25) -Y (0,12) -AEI (10) catalyst could be maintained at 47% after exposure to 50 ppm SO₂ atmosphere for 65 H, which was significantly better than that of undoped Cu-AEI and Cu-CHA molecular sieve catalysts^[63].

The special structure or morphology design of the catalyst can also effectively improve the sulfur tolerance of the catalyst.

Tang et al. Found that Fe₂O₃/MoO₃ catalyst can inhibit the deposition of ammonium bisulfate, acidic MoO₃ can capture the {{\mathrm{NH}}_4}^{+} in ammonium bisulfate, and because of its special lamellar structure,The {{\mathrm{NH}}_4}^{+} and the {{\mathrm{HSO}}_4}^{-} in the ammonium bisulfate are spatially separated, and the electrostatic force between the {{\mathrm{NH}}_4}^{+} and the {{\mathrm{HSO}}_4}^{-} is overcome, so that the {{\mathrm{NH}}_4}^{+} can react with NO at a low temperature, thereby inhibiting the deposition of the ammonium bisulfate^[64]. The research on promoting the decomposition of ammonium sulfate on the catalyst mainly focuses on reducing the stability of ammonium sulfate and promoting the decomposition of ammonium sulfate by adjusting the electronic interaction between ammonium sulfate and the catalyst. Through a series of studies, Li et al. Considered that through the interaction between ammonium sulfate and metal oxide, electrons could be transferred from metal oxide to sulfate, thus effectively promoting the decomposition of sulfate^[65]. They incorporated CeO₂ into V₂O₅-WO₃/TiO₂ and found that the sulfur tolerance of the catalyst was significantly increased, because the interaction between Ce and V caused the electrons around Ce to be biased towards ammonium sulfate, which reduced the stability of ammonium sulfate and made it easier to decompose^[66]. By adjusting the electronic interaction between ammonium sulfate and the molecular sieve support, the accumulation of ammonium sulfate (hydrogen) species on the surface of copper-based small pore molecular sieve may also be reduced.

In addition, studies have shown that the design of core-shell structure or hierarchical pores can also effectively improve the sulfur resistance of catalysts. Shi et al. Deposited CeO₂ nanoparticles on the surface of Cu-SSZ-13 with the assistance of cetyltrimethylammonium bromide (CTAB) to form a Cu-SSZ-13@CeO₂ core-shell structure, and its low temperature performance was significantly improved, and the sulfur and water resistance was also improved, which may be due to the fact that the CeO₂ wrapped on the outer layer of Cu-SSZ-13 inhibited the deposition of ammonium sulfate, while the interaction between Ce and Cu enhanced the redox performance of the catalyst^[67]. Peng et al. Prepared a hierarchical porous Cu-SSZ-13 catalyst by a dual-template method. Compared with the traditional Cu-SSZ-13 catalyst, it has better low-temperature activity and wider activity temperature window, and shows better SO₂ tolerance^[68]. Liang et al. Synthesized a Cu-SSZ-13 catalyst containing mesopores using carbon black as a template. Compared with the microporous Cu-SSZ-13 catalyst, it has better catalytic performance for NH₃-SCR and sulfur resistance^[69]. Zhang et al. Obtained a core-shell structure catalyst with mesoporous Cu-SSZ-13 as the core and mesoporous aluminosilicate as the shell by desilication method, which has better SCR performance because of more active copper species and lower diffusion limitation^[70]. In addition, the mesoporous structure formed by desilication also has higher acidity, which can reduce the adsorption of SO₂, and may also be an effective way to improve the sulfur resistance of Cu-SSZ-13 catalyst.

In addition to the above two main modification methods, there are also studies on reducing the sulfur poisoning of catalysts by changing the reaction atmosphere. Wang et al. Found that ammonium bisulfate deposited on the V₂O₅-WO₃/TiO₂ catalyst can react with NO and NO₂ in the reaction atmosphere at temperatures above 100 ℃, and the {{\mathrm{NH}}_4}^{+} in ammonium bisulfate reacts with NO/NO₂/O₂.It promotes the rapid decomposition of ammonium bisulfate, but in NO/O₂ atmosphere, the reaction needs to occur above 300 ℃, which indicates that the SCR catalytic activity of ammonium bisulfate poisoned catalyst can be effectively restored by adjusting the proportion of NO₂ in the reaction atmosphere^[71,72].

To sum up, although the sulfur tolerance of copper-based small pore molecular sieve catalysts has been improved by doping with Fe, Ce, Y and other metals or controlling the surface morphology, the deactivation caused by sulfur poisoning is still very serious in the presence of SO₂ and H₂O. The construction of core-shell structure on Cu-based small pore molecular sieves as sacrificial sites may be an effective way to improve the sulfur tolerance of Cu-based small pore molecular sieves, but the effects on catalyst performance and hydrothermal stability need to be considered comprehensively. In addition to the design and modification of the catalyst itself, the study of the regeneration method and regeneration mechanism of sulfur-poisoned copper-based small pore molecular sieve catalyst can also promote the efficient utilization of copper-based small pore molecular sieve.

Brookshear et al. Studied the regeneration of Cu/SSZ-13 catalyst sulfided at 250 and 400 ℃. When the catalyst was regenerated at 500 ℃ for 10 min, only part of the activity was recovered. When the regeneration time was extended to 30 min, the performance of the catalyst could be basically recovered^[32]. Wang et al. studied the SO₂ aging and regeneration of Cu-SAPO-34, and found that all the sulfurized catalysts had sulfate, and the activity of the sulfurized catalysts could be basically restored after being treated at 600 ℃^[73]. Zhang et al. Studied the effect of SO₂ poisoning on Cu-SAPO-34 catalyst, and regenerated the SO₂ poisoned catalyst at high temperature. When the temperature was lower than 500 ℃, the activity of the catalyst was recovered by 90%, which may be due to the decomposition of ammonium sulfate. When the temperature continued to rise to 700 ℃, the catalytic activity was completely recovered, even better than that of the fresh catalyst, which may be due to the solid copper ion exchange during high temperature treatment^[31]. Therefore, after the copper-based small pore molecular sieve is treated at 500 ℃, the active copper species deactivated by ammonium sulfate coverage can be completely restored to activity, while the decomposition temperature of copper sulfate species is higher (it needs 700 ℃ to be completely decomposed), and the deactivation caused by copper sulfate species requires a higher regeneration temperature^[74].

The removal of sulfur species on the surface of copper-based small pore molecular sieve catalyst can be accelerated when a low concentration of reducing gas is introduced under an oxidizing atmosphere to form a local reducing environment on the surface of the catalyst. Kumar et al. Found that when the reducing gas propylene was introduced into the reaction atmosphere, the sulfate species on the Cu-CHA catalyst could be completely removed at a temperature below 500 ℃, thus basically restoring the catalyst activity, while the catalyst activity could not be completely restored by simple heat treatment even at a temperature above 700 ℃^[52]. This may be because the reducing gas forms a local reducing environment on the surface of the catalyst, which changes the valence of copper species, reduces the binding capacity between sulfate and active sites, decomposes copper sulfate species, and regenerates the catalyst. Mesilov et al. Found that the regeneration of sulfur-poisoned Cu-SSZ-13 at 500 ° C under inert atmosphere and reducing atmosphere (H₂) could effectively remove ammonium sulfate and restore the catalyst activity^[75]. Regeneration in a reducing atmosphere can achieve desulfurization at lower temperatures than in an inert atmosphere, while regeneration in an oxidizing atmosphere may result in the conversion of a portion of the ammonium sulfate to a more stable metal sulfate.

To sum up, the regeneration of copper-based small pore molecular sieve catalyst is mainly to partially or completely restore the activity by decomposing copper sulfate and other related species through high temperature treatment. Generally, the treatment temperature is 500 ~ 700 ℃. When reducing gas is introduced into the reaction atmosphere, the decomposition temperature of copper sulfate and other related species can be reduced.

Hammershøi et al. Found that after sulfurization at different temperatures for 120 H, the activity of Cu-CHA catalyst was recovered by 80% after regeneration at 550 ℃, and the S/Cu ratio after regeneration was not more than 0. 2, indicating that thermal regeneration could effectively solve the problem of sulfur poisoning^[76]. They also sulfided the Cu-SSZ-13 catalyst at 550 ° C for different times and then regenerated it at 550 ° C, and found that most of the ZCuOH sites could be recovered by desorption and sulfur redistribution, but part of the sulfur would migrate back to the Z₂Cu sites to form more stable copper sulfate species^[77]. Dahlin et al. Studied the poisoning of Cu-SSZ-13 catalyst exposed to SO₂ at different temperatures (220 and 400 ℃), and the deactivation of SO₂ at low temperature has a greater impact, because the catalyst adsorbs more sulfur species at low temperature, and the activity of the catalyst can be recovered by more than 80% after regeneration at 500 ℃, and can be completely recovered at 700 ℃, and can be completely recovered at 500 ℃ under rapid SCR conditions^[78]. They found that the sulfur content of the catalyst aged directly on the engine was close to that of the catalyst aged in the laboratory, but the activity of the aged sample in the laboratory could be completely restored after regeneration, while the aged catalyst on the engine could be completely restored after regeneration.The catalyst activity is only recovered by 10%, which may be due to the fact that the catalyst aged on the engine contains S, P, Ca, Zn, etc., and there may be different deactivation mechanisms such as phosphorus poisoning in addition to sulfur poisoning.

Bergman et al. Studied Cu-SSZ-13 poisoned by SO₂, and found that the deactivation of the catalyst was mainly due to the formation of copper bisulfate species. The mobility of copper ions in the catalyst poisoned by sulfur became lower, and the activity could be completely restored after regeneration at 700 ℃, mainly due to the decomposition of copper bisulfate species during heating^[79]. Shen et al. Further studied the regeneration conditions and regeneration mechanism of Cu-SSZ-13 catalyst poisoned by sulfur, and found that 700 ℃ was the best regeneration temperature, and the catalyst could basically recover its activity after 16 H of regeneration^[80]. The regeneration mechanism of Cu-SSZ-13 poisoned by sulfur mainly includes two processes: the decomposition of copper sulfate into copper oxide and the migration of copper. When the temperature is higher than 700 ℃, the dealumination of the molecular sieve inhibits the migration of copper species, resulting in part of the activity can not be restored.

In a word, the regeneration of copper-based small pore molecular sieve is mainly the decomposition of sulfate species, in which ammonium sulfate (hydrogen) species can be completely decomposed at a temperature below 500 ℃, while the decomposition of copper sulfate or copper hydrogen sulfate requires a higher temperature to be completely decomposed, and the copper ions produced by the decomposition are redistributed as new active centers.