Sulfur resistance of Ce-Mn/TiO2 catalysts for low-temperature NH3–SCR

Ce-Mn/TiO2 catalyst prepared using a simple impregnation method demonstrated a better low-temperature selective catalytic reduction of NO with NH3 (NH3–SCR) activity in comparison with the sol-gel method. The Ce-Mn/TiO2 catalyst loading with 20% Ce had the best low-temperature activity and achieved a NO conversion rate higher than 90% at 140–260°C with a 99.7% NO conversion rate at 180°C. The Ce-Mn/TiO2 catalyst only had a 6% NO conversion rate decrease after 100 ppm of SO2 was added to the stream. When SO2 was removed from the stream, the catalyst was able to recover completely. The crystal structure, morphology, textural properties and valence state of the metals involving the novel catalysts were investigated using X-ray diffraction, N2 adsorption and desorption analysis, X-ray photoelectron spectroscopy, scanning electron microscopy and energy dispersive spectroscopy, respectively. The decrease of NH3–SCR performance in the presence of 100 ppm SO2 was due to the decrease of the surface area, change of the pore structure, the decrease of Ce4+ and Mn4+ concentration and the formation of the sulfur phase chemicals which blocked the active sites and changed the valence status of the elements.


Introduction
Nitrogen oxides NO x (NO, NO 2 , N 2 O) are global by-products of high-temperature combustion [1]. NO x abatement from emissions is necessary because emitted nitrogen oxides cause severe adverse health effects, acid rain and ozone layer depletion [2][3][4][5][6]. Among the NO x treatment approaches, the technology involving the selective catalytic reduction (SCR) of NO x with NH 3 (NH 3  used method because of its high denitration efficiency and wide operable temperature range [7]. The SCR method refers to the process in which the reducing agent NH 3 reacts with NO x with the assistance of catalyst, generating N 2 and H 2 O [8]. The reactions involved in this process can be seen in the following equations 4NH 3  Reactions (1.1) and (1.2) are designated as the main reactions, and reaction (1.1) is often regarded as the standard SCR reaction. Elevated temperature may lead to the occurrence of NH 3 oxidation side effects, partly shown in equations (1.3) and (1.4). The coexistence of NH 3 , NO and NO 2 will lead to a rapid SCR reaction, shown in equation (1.5). Studies have shown that the reaction rate of equation (1.5) is faster than that of standard SCR reactions.
The fundamental goal of NH 3 -SCR technology is to develop a SCR catalyst that possesses a high activity at relatively low temperature ranges, strong anti-sulfur performance and vanadium-free to be environmentally friendly. Metal oxide catalyst groups that are Mn-based [9][10][11] and Ce-based [12][13][14] have high efficiencies at low temperature ranges because of the reduction of Mn 4+ to Mn 3+ in the Mn phase, and CeO 2 has a considerably large oxygen storage capacity and excellent redox properties. Composite catalysts prepared using these two kinds of active components have favourable low-temperature denitration activity and are stable when introduced to typical catalyst poisons. In addition, the vast surface area, developed pore structure and acid sites of TiO 2 are more conducive to the adsorption of NH 3 and can accelerate the reaction [15][16][17]. Qi et al. [18] prepared a non-load-type MnO x -CeO 2 , low-temperature SCR catalyst using co-precipitation. The removal of NO was boosted by improving the ability of the redox catalyst by infusing Mn in the CeO 2 lattice. This generated a large number of oxygen vacancies. Lee et al. [19] prepared a MnO x /CeO 2 -TiO 2 catalyst. They found that Ce doping increased the surface area of the catalyst, improved the Mn 4+ concentration and increased the overall catalytic activity. Liu et al. [3] proved that the environmentally benign Mn-Ce-Ti catalyst had a high affinity for NO x removal because of the dual redox properties and the amorphous structure of the catalyst. Moreover, the Mn-Ce-Ti catalyst displayed a high resistance towards H 2 O and SO 2 .
The principal objective of this study involves determining how well the co-doped Ce-Mn/TiO 2 catalyst can tolerate sulfur while undergoing low-temperature SCR. A series of Ce-Mn/TiO 2 catalysts were prepared using a single impregnation (IP) method and investigated for their effectiveness of lowtemperature SCR of NO x with NH 3 . The possible mechanism of the best performing, low-temperature SCR catalyst in this work is discussed in detail using various characterization methods.

Catalyst preparation
Catalysts were prepared using previously discussed IP and sol-gel (SG) methods [20,21]. The IP method involved using a molar ratio of 2 : 3 manganese acetate to cerium nitrate. The components were dissolved in deionized water and stirred for 1 h. Subsequently, approximately 2.5 g of TiO 2 was added to the mixture and stirred for 2 h. The suspension was transferred to a rotary evaporator and heated at 55°C for 1 h to obtain the viscous liquid. This sample was dried at 105°C overnight in an oven. Subsequently, the sample was calcined at 550°C for 6 h and ground to a mesh size of 20-40. The mass ratio of Ce/TiO 2 was tuned and recorded as Ce(wt%)-Mn/TiO 2. The molar ratio of Mn and Ce element maintains constant in each catalyst.
The SG method involved using 2.65 g of tetra-n-butyl titanate and 21.2 g of ethanol. The components were mixed and stirred for 50 min to form a yellow-tinted solution referred to as 'solution A'. A separate solution of manganese nitrate and cerium nitrate was dissolved in deionized water, and then anhydrous ethanol and acetic acid were added. This solution was stirred for 1 h to obtain a red-tinted solution referred to as 'solution B'. Subsequently, solution B was slowly added to solution A, and the mixture was stirred for 12 h. The gel formed at a room temperature for 2 h. The gel was then dried at 105°C for 24 h and calcined at 500°C for 2 h in a 20% O 2 flow. The final sample was ground to a mesh size ranging from 20 to 40. The ratio of tetra-n-butyl titanate, ethanol, deionized water and acetic acid was 1 : 8 : 6 : 3. The molar ratio of manganese element and cerium was 2 : 3 (the same ratio used with the IP method outlined above).

Catalyst activity measurement
SCR activity of the catalyst was measured using a fixed bed, stainless steel tube reactor with an inner diameter of 11 mm and an outer diameter of 14 mm. Gas purchased to be used in the experiment contained specific concentrations of components to simulate flue gas. The feed gas mixture consisted of 500 ppm NH 3 , 500 ppm NO, 3% O 2 (volume fraction), 0-200 ppm SO 2 (depending on the experiment) and a balance of N 2 . The simulated gas flow rate was 1000 ml min −1 . A volume of 6 ml of catalyst was loaded into the reactor for each experiment. The temperature ranged 100°C and 300°C at a heating rate of 3°C/min with a gas hourly space velocity (GHSV) of 10 000 h −1 . The concentrations of NO were monitored at the inlet and outlet in real time using a gas analyser (Testo 340) to calculate the conversion rate using the following equation.
where [NO] in and [NO] out referred to the NO concentration at the reactor inlet and outlet, respectively. The concentration was measured when the reaction reached a steady-state condition (approx. 20-40 min) at each temperature, which reduced the measurement errors caused by instability.

Catalyst characterization
The powder X-ray diffraction (XRD) characterization of the samples was performed using a Bruker D8-Advance X-ray powder diffractometer with a Cu Kα radiation source (λ = 1.5406 Å), a pulverized sample with scattering angles (2θ ) of 5-85°and a 0.0197 step size operated at 50 kV and 50 mA. The diffraction lines were identified by matching them with reference patterns in the Joint Committee on Powder Diffraction Standards (JCPDS) database. A ThermoFisher Escalab 250Xi X-ray powder photoelectron spectrometer was used to qualitatively analyse the X-ray photoelectron spectroscopy (XPS) characterization of the sample surface composition using an Al Kα radiation source with a scattering of 0-5000 eV. The binding energy was calibrated using the C 1s peak contaminate carbon (BE = 284.6 eV) as an internal standard. N 2 adsorption and desorption of each sample were measured at −196°C using the ASAP 2020 automatic rapid surface area and mesoporous/microporous analyser with a N 2 adsorption gas. The samples were degassed at 200°C for 12 h before the measurement took place. The specific surface area was calculated according to the Brunauer-Emmett-Teller (BET) method. The total pore volume was determined based on the amount of the adsorbed N 2 volume at a relative pressure of approximately p/p 0 = 0.99.
The surface morphology of the catalysts was characterized with scanning electron microscopy (SEM) using a Hitachi SU8010 with an acceleration voltage of 15 kV. Qualitative and quantitative analysis of catalyst elements were carried out using energy dispersive X-ray spectroscopy (EDS) on a HORIBA EX-350 with an acceleration voltage of 15 KV.

Effect of synthetic method on the NH 3 -SCR activity
SCR activities of the Ce(20)-Mn/TiO 2 catalysts synthesized by the SG and IP methods are shown in figure 1. These results reveal that the IP method had a better low-temperature denitration activity than the SG method. At a temperature range of 140-260°C, the NO conversion rate was higher than 90% in the presence of the IP catalyst. At a temperature range of 160-300°C, the NO conversion rate was higher than 90% in the presence of the SG catalyst. The NO conversion rate was approximately 99.7% at a temperature of 180°C using either method. The catalyst made using the IP method showed better lowtemperature denitration activity, was easy to prepare, and had a low-cost preparation process. Therefore, the following Ce-Mn/TiO 2 catalysts used in this study were prepared using the IP method.

Resistance to SO 2 poisoning on Ce-Mn/TiO 2 catalysts
The addition of active components has a crucial effect on the catalyst performance in NH 3 -SCR. As shown in figure 2a, the NO x conversion rate on pure TiO 2 was less than 10% in the temperature range of 100-300°C, indicating that the carrier was inactive for denitration. After adding Ce and Mn, the catalytic  activity of Ce-Mn/TiO 2 was enhanced substantially. Catalysts were loaded with 10%, 20% and 30% Ce and denoted as Ce (10), Ce (20) and Ce(30), respectively. The temperature windows for obtaining a denitration rate higher than 90% included 160-220°C for the Ce(10)-Mn/TiO 2 catalyst, 140-260°C for the Ce(20)-Mn/TiO 2 catalyst and 160-240°C for the Ce(30)-Mn/TiO 2 catalyst. Therefore, the Ce(20)-Mn/TiO 2 catalyst had the highest NO conversion rate of 99.7% with a wide denitration low-temperature window.
In the absence of SO 2 , all three Ce-Mn/TiO 2 catalysts had a NO conversion of approximately 100% (during the initial 1 h shown in figure 2b). After adding 100 ppm of SO 2 , the NH 3 -SCR activity of Ce(10)-Mn/TiO 2 and Ce(30)-Mn/TiO 2 decreased rapidly to 0% and 68% in 5 h, respectively. After SO 2 was removed from the stream, the Ce(10)-Mn/TiO 2 catalyst lost all of its activity, while the Ce(30)-Mn/TiO 2 catalyst only maintained an activity of approximately 65%. However, the Ce(20)-Mn/TiO 2 catalyst had only a small decrease of 6% in the NO conversion after 100 ppm of SO 2 was added in 5 h. After SO 2 was removed from the stream, the NH 3 -SCR activity involving the Ce(20)-Mn/TiO 2 catalyst recovered completely and reached 100% NO conversion.
The effect of reaction temperature on the sulfur resistance of the Ce(20)-Mn/TiO 2 catalyst was investigated, and the results are shown in figure 2c. The addition of SO 2 at various reaction temperatures all had an inhibitory effect on the catalytic activity. The conversion rate of NO constantly decreased with the addition of 100 ppm SO 2 . The decrease in catalytic activity at 140°C was the most pronounced, while the catalyst activity at 180°C had the best sulfur resistance.
The effect of the SO 2 concentration on the activity of the best Ce(20)-Mn/TiO 2 catalyst was investigated, and the results are shown in figure 2d. When the catalyst was at 180°C for 1 h, the addition of SO 2 in different concentrations decreased the catalytic activity, and there were different levels of recovery after SO 2 was removed from the stream. SO 2 had a toxic effect on the Ce(20)-Mn/TiO 2 catalyst, and as the concentration of SO 2 increased, the catalyst activity decreased. The addition of 100, 150 and 200 ppm SO 2 decreased the NO conversion rate from 99.7% to 94%, 90% and 83%, respectively. After removing 100 ppm of SO 2 , the catalyst activity fully recovered; however, only partial recovery was seen after the catalyst was submitted to the higher SO 2 concentrations.

Catalyst characterization
The Ce(20)-Mn/TiO 2 catalyst had the best NH 3 -SCR performance, so only fresh and used Ce(20)-Mn/TiO 2 catalyst were analysed for determining specific catalyst characteristics. The used catalyst was extracted after the reaction was carried out in a 100 ppm SO 2 atmosphere for 5 h at 180°C. The XRD patterns in figure 3a showed that no new peaks were observed in the used catalyst, indicating that the catalyst did not form any (or negligible) crystalline states of sulfate or the amount of the formed phase was too small in the SO 2 atmosphere. Based on the activity discussion mentioned above, the NO conversion rate of the Ce(20)-Mn/TiO 2 catalyst was approximately 94% after the addition of 100 ppm SO 2 . Therefore, the Ce(20)-Mn/TiO 2 catalyst had an ideal sulfur resistance.
The specific surface area, pore volume and average pore size of the different Ce(wt%)-Mn/TiO 2 catalysts are summarized in table 1. Owing to the large specific surface area of CeO 2 , the loading amount    -d and table 3) verified the existence of S and N on the surface, implying that the ammonium sulfate material formed on the surface of the used catalyst. This covered the active sites and decreased the catalytic activity. However, the XRD patterns imply that the ammonium sulfate material was not obviously detected because there is no formation or only a relatively small amount was present and therefore accounted for only a negligible decrease of the catalytic activity in the SO 2 atmosphere.
XPS spectra of the Ce 3d, Mn 2p and O 1s of the fresh and used Ce(20)-Mn/TiO 2 catalysts are shown in figure 5a-c. Figure 5a shows the Ce 3d in the fresh and used catalysts had eight peaks, where the V , U were the characteristic peaks of Ce 3+ (Ce 2 O 3 ), and V, V , V , U, U , U were the characteristic peaks of Ce 4+ . The leading form of the Ce oxide in both catalysts was CeO 2 (Ce 4+ ) which was beneficial for oxygen storage capacity. The conversion between Ce 3+ and Ce 4+ accomplished a storage-release cycle of oxygen for better promoting the process of denitration. Figure 5b illustrates that the peak of Mn      used catalyst. The high concentration of the surface adsorption of oxygen had a strong oxidation effect, which not only completed the oxidation and reduction cycle but also enhanced the oxidation process of NO to NO 2 . This promoted a rapid response to the SCR reaction [25,26]. XPS spectra of the S 2p and N 1s of the used Ce(20)-Mn/TiO 2 catalysts are shown in figure 5d-e. Figure 5d shows that the leading form of S in the used catalyst was S 4+ and S 6+ . The peak at 168.5 eV belonged to SO 3 2− , while the peaks at 169.9 and 169.1 eV were assigned to SO 4 2− [24]. S 6+ accounted for approximately 63% of S and included some sulfate material which formed on the catalyst surface after undergoing reactions in the SO 2 atmosphere. Figure 5e demonstrates the peaks of N located in the vicinity of 400.8 eV and 407.3 eV were characterized as NH 4 + and NO 3 − , respectively [27]. NH 4 + accounted for approximately 64% of the total N. The conversion of NH 3 to NH 4 + and NO 3 − was indicated by the results.

Conclusion
NH 3 -SCR activity was measured after using an IP method and a SG method to form a Ce-Mn/TiO 2 catalyst. The catalyst formed with the IP method had a better low-temperature denitration activity than the catalyst synthesized using the SG method. The Ce-Mn/TiO 2 catalyst loaded with 20% Ce had the best low-temperature activity with a NO conversion rate of 99.7% at 180°C. Despite the addition of 100 ppm SO 2 at 180°C for 5 h, the NO conversion rate of the Ce(20)-Mn/TiO 2 catalyst was still as high as 94% and recovered when SO 2 was removed from the stream. The SO 2 resistance of the Ce(20)-Mn/TiO 2 catalyst is attributed to the widely distributed elements of Mn and Ce. This led to the inability of the sulfate material to remain on the surface. Characterization of fresh and used Ce(20)-Mn/TiO 2 catalysts was performed using XRD, BET, XPS and SEM-EDS. These results indicated that there was a decrease in Ce 4+ and Mn 4+