The Design of Sulfated Ce/HZSM-5 for Catalytic Decomposition of CF4.

CF4 has a global warming potential of 6500 and possesses a lifetime of 50,000 years. In this study, we modified the HZSM-5 catalyst with Ce and sulfuric acid treatment. The S/Ce/HZSM-5 catalyst achieves 41% of CF4 conversion at 500 °C, which is four times higher than that over Ce/HZSM-5, while the HZSM-5 exhibits no catalytic activity. The effects of modification were studied by using NH3-TPD, FT-IR of pyridine adsorption, and XPS methods. The results indicated that the modification, especially the sulfuric acid treatment, strongly increased the Lewis acidic sites, strong acidic sites, and moderate acidic sites on catalysts, which are the main active centers for CF4 decomposition. The mechanism of acidic sites increases by modification and CF4 decomposition is clarified. The results of this work will help the development of more effective catalysts for CF4 decomposition.

1. Introduction

Perfluorocarbon (PFC) is a class of greenhouse-gas (GHG) in which all valences of carbon are satisfied with fluorine atoms [1,2]. CF4 is considered as the most abundant and harmful in PFCs, and it has a high global warming potential, which is about 6500 times higher than that of CO2 over a 100-year time scale [3]. Due to its symmetry and strong ionic character in the C–F covalent bond, which is the strongest bond (543 kJ mol−1) in organic chemistry, the CF4 molecule is extremely stable [4]. The aluminum production industry is considered as the main source of CF4 emission. When the anode effect occurs in the electrolytic cell, the flux Na3AlF6 will react with the C anode to form CF4 as in the equation below [5]. In 2018, the global CF4 emission from aluminum production is 4.408 Gg, accounting for approximately 4% of total GHG emission in CO2 equivalent [6]. However, the lifetime of CF4 in atmosphere is 50,000 years while that of CO2 is only 10–20 years. Therefore, it is necessary and important to remove CF4 presented in the exhaust flue gas from the aluminum production industry. 4Na

To date, several methods have been developed for CF4 abatement, such as fueled combustion, plasma, and catalytic hydrolytic decomposition [7,8,9]. Among them, the catalytic hydrolytic decomposition is considered as the ideal method for its high efficiency and mild reaction temperature compared to others.

HZSM-5 is a molecular sieve catalyst with Si, Al, and O elements as the framework, it has displayed excellent performance in chlorofluorocarbon and hydrofluorocarbon decomposition [10,11,12,13]. The catalytic behavior is related to its high specific surface area, strong acidity, and three-dimensional channel system. However, the amount of strong acidic sites and Lewis acidic sites are insufficient for CF4 decomposition.

Cerium (Ce) is a good promoter for catalyst, due to its high oxygen-storage capacity, high oxygen-vacancy concentration, and the facile Ce4+/Ce3+ redox cycle, which enhances the electron transfer that generates more acidic sites [14,15]. For example, de Rivas et al. [12] studied the 1,2-dichloroethane decomposition over the Ce/HZSM-5 catalyst, and they achieved a 90% conversion that was attributed to the addition of Ce which increased the Lewis acidic sites. Chen et al. [16] synthesized a [Ce-(1,3,5-benzenetricarboxylic acid)(H2O)6] catalyst, and achieved a good activity for toluene oxidation with conversion of T90% at 223 °C, the authors concluded that the catalytic activity was due to the great amount of acidic sites on catalyst. Moreover, some research suggests that the sulfuric acid treatment is an excellent strategy to significantly increase the Lewis acidic sites on catalysts, as the sulfate ions act as Lewis acidic sites; meanwhile, they attracted electrons to create more new acidic sites [17,18]. Song et al. [19] modified the γ-Al2O3 with Ce and sulfuric acid treatment for CF4 catalytic decomposition, and found that the addition of Ce increased the acidic sites, the sulfuric acid treatment further enhanced the increasement, and the CF4 conversions were consistent with the amount of acidic sites. Although varied Lewis acid type catalysts can be applied in CF4 decomposition, a high temperature (over 700 °C) is required to decompose the CF4 molecule [20,21,22]. Therefore, we focused on developing a HZSM-5 based catalyst by modification with element Ce and sulfuric acid treatment for CF4 decomposition below 700 °C.

Given the above reasons, a series of modified HZSM-5 catalysts were developed to hydrolytic decompose CF4 at 500 °C. The aim of this work is to investigate the changes in the properties of the catalyst on the addition of Ce and acid treatment. The physicochemical properties of catalysts were characterized by different techniques. This study intended to elucidate the mechanisms of CF4 decomposition over HZSM-5 based catalysts.

2. Experimental Section

2.1. Catalyst Preparation

The Ce/HZSM-5 catalyst was prepared using the impregnation–calcination method from the commercial molecular sieve HZSM-5 (Si/Al = 18, mole ratio, Nankai Unv, Tianjin, China) calcined at 650 °C in N2 atmosphere for 5 h with an aqueous solution containing required amounts of cerium nitrate (Sinopharm Chemical Reagent Co., Ltd., Shanghai, China), followed by drying overnight at 80 °C. The Ce/HZSM-5 (10%Ce/HZSM-5) catalyst was crushed and sieved into 60–80 mesh. The Ce/HZSM-5 catalyst was impregnated with an aqueous solution containing H2SO4 (98%, Sinopharm Chemical Reagent Co., Ltd., Shanghai, China) under stirring for 24 h and dry for 12 h at 60 °C, the impregnating solution was adjusted to yield the same SO42− (12 wt%) as S/Ce/HZSM-5, finally, calcined at 650 °C for 5 h in N2 atmosphere. The S/Ce/HZSM-5 (12%S/10%Ce/HZSM-5) catalyst was crushed and sieved into 60–80 mesh.

2.2. Catalytic Activity Test

The hydrolytic decomposition of CF4 was carried out in a fixed-bed reactor. The temperature was maintained at 500 °C. In the aluminum production, the CF4 generated in electrolytic cell during the anode effect, the CF4 concentration in the flue gas is about 1%. Therefore, the gas flow consists of l% CF4, 35% H2O, and balanced by Ar. The water was pre-heated at 150 °C and then constantly introduced into the reaction system by using a syringe pump. The effluent gas was washed by aqueous potassium hydroxide to remove the produced HF, and then passed through a cold trap to remove water. Finally, the gas was analyzed by an online gas chromatography (GC-9790 II) equipped with a thermal conductivity detector (TCD).

2.3. Catalyst Characterization

The morphology of the samples was examined by scanning electron microscope (SEM, JSM-IT300LA, JEOL, Tokyo, Japan) with energy dispersive X-ray (EDX) analysis. The X-ray diffraction analysis was performed on a TTR III diffractometer (XRD, Rigaku, Tokyo, Japan). The chemical composition and state of the elements on catalyst surfaces were investigated by X-ray photoelectron spectroscopy (XPS, EscaLab 250Xi, Thermo Fisher, Waltham, MA, USA). The specific surface areas were measured with the N2 adsorption method on an ASAP analyzer (BET, ASAP2020, Micromeritics, Norcross, GA, USA). Ammonia temperature programmed desorption (NH3-TPD, AutoChem II 2920, Micromeritics, Norcross, GA, USA) analyses were performed using the following procedures. A 100 mg sample was pretreated at 550 °C, with helium flow of 30 mL/min for 1 h, and then cooled to 50 °C. Ammonia (10% NH3/He) was introduced to the catalyst for 1 h at 50 °C. After that, the sample was flushed with helium of 50 mL/min for 1 h to remove absorbed NH3, then the temperature was programmed to increase to 900 °C at a rate of 10 °C/min. The amount of ammonia desorbed from the catalyst was detected using TCD. Fourier transform infrared spectra (FT-IR, Nicolet iS50, Thermo Fisher, Waltham, MA, USA) of pyridine absorption were conducted using the following procedures. The sample was pressed and put into an IR cell, and then it was degassed at 400 °C under vacuum for 2 h to dehydrate. Next, the cell was cooled to room temperature, and the background signal was recorded. After that, pyridine vapor was introduced to the system until reaching adsorption equilibrium. The sample was evacuated out at 150 °C for 30 min followed by cooling down to 50 °C, and then spectral acquisition was performed.

3. Results and Discussion

3.1. Characterization of Catalysts

The morphology and elemental distribution of catalysts were monitored by SEM and EDS mapping, as shown in Figure 1. The microscopic images of HZSM-5 (Figure 1A) exhibit the nano-blocky particles with a smooth surface. The modified catalyst Ce/HZSM-5 (Figure 1B) and the S/Ce/HZSM-5 (Figure 1C,D) exhibit rough surface, the catalysts were covered by small solid granule. The EDS mapping for S/Ce/HZSM-5 (Figure 1E–I) exhibits the uniform distribution of O, Si, Al, S and Ce, respectively.

Figure 1

SEM of (A) HZSM-5, (B) Ce/HZSM-5, (C,D) S/Ce/HZSM-5, and SEM-EDS mapping for S/Ce/HZSM-5 (E–I).

The X-ray diffraction (XRD) was performed to investigate the crystal structure of the catalysts as in Figure 2. All the samples contain the HZSM-5 [23], and the peak of CeO2 (JCPDS:34-0394) is observed in modified samples, indicating the formation of CeO2 [24]. It should be noted that the peak of Ce2(SO4)3 is not observed in the S/Ce/HZSM-5 catalyst, which indicated the absence of reaction between Ce and sulfuric acid during the catalyst preparation step. The specific surface area and pore diameter of catalysts were analyzed, as listed in Table 1. The surface area of catalysts decreased from 248 m2 g−1 to 148 m2 g−1, and the pore diameter decreased from 2.38 nm to 2.28 nm after modified by Ce, these changes are in agreement with previous research [19,25], which may account for the formation of CeO2 on the surface of catalysts, in addition to element Ce, some cation ions were replaced by Ce, the CeO2 would block the original pore, results in the surface area and pore diameter decreasing. The surface area and pore diameter increased to 278 m2 g−1 and 2.28 nm, respectively, after the sulfate acid treatment. This may be due to the fact that acid treatment increased the amount of amorphous silica [26].

Figure 2

X-ray diffraction (XRD) patterns of the catalysts, (1) HZSM-5, (2) Ce/HZSM-5, and (3) S/Ce/HZSM-5.

Table 1

Specific surface area, pore diameter for the catalysts.

Catalysts	BET Surface Area (m2 g−1)	Pore Diameter (nm)
HZSM-5	248	2.38
Ce/HZSM-5	148	2.28
S/Ce/HZSM-5	278	2.63

The NH3-TPD experiments were conducted to analyze the different kinds of acidic sites in the catalysts. There are three kinds of acidic site according to the desorption temperature of NH3 on catalysts: weak acidic sites (T < 250 °C), moderate acidic sites (250 °C < T < 600 °C), and strong acidic sites (T > 600 °C) [27,28,29]. The TPD results are shown in Figure 3 and Table 2, the HZSM-5 catalyst exhibited the most weak acidic sites (663 µmol g−1) and the second highest amount of moderate acidic sites (267 µmol g−1), the strong acidic site was not observed. The addition of Ce decreased the amount of weak acidic sites to 290 µmol g−1 and increased the moderate acidic sites to 544 µmol g−1, but still, no strong acidic site was observed. After further acid treatment, the amount of weak acidic sites decreased to 273 µmol g−1 and the amount of moderate acidic sites increased to 855 µmol g−1, the amount of strong acidic sites dramatically increased to 1274 µmol g−1, which dominated over others. These findings strongly indicated that the impregnation with sulfate group influenced the acidic properties of catalysts, the amount of moderate and strong acidic sites significantly increased; moreover, the ratio of (strong + moderate)/weak was in the following order: S/Ce/HZSM-5(7.80) > Ce/HZSM-5(1.88) > HZSM-5(0.40).

Figure 3

NH3-TPD profiles of catalysts.

Table 2

NH3-TPD acidic sites results for the catalyst.

Catalysts	Amount of Acid Site (µmol g−1)
	Weak	Moderate	Strong	Total	(Strong + Moderate)/Weak
HZSM-5	663	267	0	930	0.40
Ce/HZSM-5	290	544	0	834	1.88
S/Ce/HZSM-5	273	855	1274	2402	7.80

In order to gain more information on acidic properties of catalysts, the FT-IR spectra of pyridine adsorption experiments are conducted as shown in Figure 4 and Table 3. The spectra are recorded over a frequency range from 1400 cm−1 to 1700 cm−1, where characteristic vibration modes of adsorbed pyridine will appear. According to the references [30,31], the characteristic wave number of 1455 cm−1, 1612 cm−1 could be assigned to the pyridine species coordinatively adsorbed on Lewis acid sites, while the wave number of 1545 cm−1, 1635 cm−1 assigned to pyridinium ion formed on Brønsted acid sites, the 1490 peak was attributable to the L+B acid sites. The amounts of L-acid site on HZSM-5, Ce/HZSM-5 and S/Ce/HZSM-5 were 120.57 µmol g−1, 138.26 µmol g−1 and 156.84 µmol g−1, respectively. The amount of Brønsted acid sites showed the reversed trend, they were 233.91 µmol g−1, 83.61 µmol g−1 and 69.97 µmol g−1, respectively. The ratio of Lewis/Brønsted was in the following order: S/Ce/HZSM-5(2.24) > Ce/HZSM-5(1.65) > HZSM-5(0.52). The results indicated that the addition of Ce increased the amount of Lewis acidic sites while decreasing the Brønsted acidic sites; furthermore, the acid treatment enhanced the influence.

Figure 4

FT-IR spectra of pyridine adsorption on catalysts.

Table 3

FT-IR spectra of pyridine adsorption results for the catalyst.

Catalysts	Amount of Acid Site (µmol g−1)
	L-Acid Site	B-Acid Site	Lewis/Brønsted (L/B)
HZSM-5	120.57	233.91	0.52
Ce/HZSM-5	138.26	83.61	1.65
S/Ce/HZSM-5	156.84	69.97	2.24

3.2. Catalytic Performance of CF4 Decomposition

The conversion reactions of CF4 over catalysts were conducted at different temperature, as shown in Figure 5. The HZSM-5 exhibited no catalytic activity below 600 °C, and only 14% conversion of CF4 while the temperature increased up to 700 °C. The Ce/HZSM-5 exhibited a higher CF4 conversion compared to HZSM-5, the CF4 conversions at 500 °C, 550 °C, 600 °C, 650 °C and 700 °C were 10%, 41%, 52%, 60% and 63%, respectively. As for the S/Ce/HZSM-5 catalyst, the CF4 conversion was further enhanced, they were 41%, 50%, 63%, 66% and 67%, respectively. Moreover, at the temperature of 500 °C, the conversion of S/Ce/HZSM-5 was four times higher than that of Ce/HZSM-5. These results indicated that the acid treatment significantly increased the catalytic activity of Ce/HZSM-5, which showed a good agreement with the results of BET, NH3-TPD and FT-IR.

Figure 5

CF4 conversion reaction over catalysts at different temperatures.

The stability of catalysts was tested for 60 h at 500 °C as shown in Figure 6. The CF4 conversion over S/Ce/HZSM-5 decreased from 41% to 34% (17% loss), while the conversion over Ce/HZSM-5 decreased from 11% to 7% (27% loss). The catalytic stability of S/Ce/HZSM-5 was much higher than that of Ce/HZSM-5. This may be due to the increment of acidic sites by acid treatment.

Figure 6

CF4 conversion over catalysts with time on stream.

3.3. Mechanism Analysis of Hydrolytic Decomposition of CF4

In order to demonstrate the main factor that promoted the catalytic activity of catalysts, the close correlation between CF4 conversion and the ratio of (strong + moderate)/weak and Lewis/Brønsted were performed as shown in Figure 7. The correlation showed that the conversion is related to the ratio of (strong + moderate)/weak and the ratio of Lewis/Brønsted. Therefore, we concluded that the strong + moderate acidic sites and the Lewis acidic sites, are the main factor that promoted the activity of the catalysts. The HZSM-5 exhibited no catalytic activity due to its low ratio of (strong + moderate)/weak and Lewis/Brønsted. Similarly, it has been reported that the Lewis acid sites and strong + moderate acid sites played a promoting role in the decomposition of hydrofluorocarbons [32,33,34]. The results strongly suggest that the direct participation of the acid sites in CF4 decomposition is probably a step for subtracting fluorine atom from CF4 molecule by Lewis acid sites. The addition of element Ce and the acid treatment could significantly influence the acidic property, resulting in increasing the catalytic activity and stability of catalysts in CF4 decomposition.

Figure 7

Correlation between CF4 conversion and ratio of (strong + moderate)/weak and Lewis/Brønsted.

In order to acquire insights into the surface chemical composition of catalysts, X-ray photoelectron spectroscopy (XPS) analysis of catalysts was carried out and shown in Figure 8 and Table 4. A curve-fitting for this analysis was carried out after Shirley-type background subtraction using a combination of Gaussian and Lorentzian functions. Figure 8A presented the S 2p core level spectra, the two major binding energy values were located at 168.90 eV and 170.10 eV, attributing to S 2p1/2 and S 2p2/3 respectively, which were ascribed to S6+ from SO42−. According to references [17,18,35], the S6+ from SO42− could significantly increase the amount of Lewis acidic sites, which showed a good agreement with the results of NH3-TPD and FT-IR. The O 1s core level spectra of catalysts were shown in Figure 8B, there are three major binding energy values located at range 528.9 to 529.5 eV, range 530.0 to 531.6 eV, and range 531.9 to 533.2 eV, attributing to Olat, Osur and Oads, respectively [36,37]. After the acid treatment, the peak of Olat disappeared, the intensity of peak for Oads increased, it could be concluded that the Olat were transformed to Oads, meanwhile, some of the Olat bonded with S6+ to form SO42−, then increased the amount of Lewis acidic sites. The Ce 3d core level spectra of catalysts are shown in Figure 8C, the peak v2 (885.9 eV–886.3 eV) and peak u2 (904.0 eV–904.5 eV) were ascribed to Ce3+, the peak v1 (883.0 eV), peak v3 (887.3 eV–889.1 eV), peak v4 (898.8 eV), peak u1 (901.5 eV), peak u3 (905.9 eV–907.7 eV) and peak u4 (917.3 eV) were ascribed to Ce4+ [38,39], after the sulfuric acid treatment, the peak u4 and peak v4 disappeared, the Ce3+/Ce increased from 25.9% to 31.0%, indicating that the sulfuric acid treatment reduced the Ce4+ to Ce3+ because they attracted electrons to create more Lewis acidic sites, which showed a good agreement with the CF4 conversion performance and the results of NH3-TPD and FT-IR.

Figure 8

X-ray photo electron spectroscopy (XPS) spectra of catalyst, (A): S 2p for catalyst, (B): O 1s for catalysts, and (C): Ce 3d for catalysts. (1) HZSM-5, (2) Ce/HZSM-5, and (3) S/Ce/HZSM-5.

Table 4

Surface atomic distribution of element O and Ce.

Catalysts	Distribution (%)
	Olat	Osur	Oads	Ce3+/Ce
HZSM-5	0	0	100	/
Ce/HZSM-5	10.5	22.0	67.5	25.9
S/Ce/HZSM-5	0	10.1	89.9	31.0

4. Conclusions

We modified the commercial HZSM-5 with the addition of Ce(Ce/HZSM-5) and further sulfuric acid treatment (S/Ce/HZSM-5). The S/Ce/HZSM-5 catalyst achieved a 41% CF4 conversion at 500 °C which is four times higher than that over Ce/HZSM-5, while the HZSM-5 exhibited no catalytic activity. The close correlation between CF4 catalytic conversion and acidic property indicated that the Lewis acidic sites and moderate + strong acidic sites are the main factors that promoted the activity of the catalysts. The effects of modification were confirmed by using NH3-TPD, FT-IR of pyridine adsorption and XPS methods. These results indicate that the acidity properties of catalysts are strongly influenced by the addition of Ce and further sulfuric acid treatment, and then significantly improved the catalytic activity in the CF4 decomposition process, which is an acid-catalyzed reaction.