English

• 

  Petroleum-based fuels, refined from different crude oils under various technological conditions, are typically a mixture of diverse organic compounds. Combustion of fuels containing high content of sulfur results in production of SO_x compounds, which can damage the environment and cause health problems^{[1, 2]}. Ultra-deep desulfurization of transportation fuel, therefore, has increasingly become a worldwide research subject and is now a mandatory request of China V standard.

  Hydrodesulfurization (HDS) method is one of the most widely used desulfurization technologies^{[2, 3]}. However, the major sulfur compounds in current Chinese commercial diesel fuel are the refractory alkyl dibenzothiophenes (DBTs), which show low hydrogenation activity because of the steric hindrance caused by their large ring size and alkyl substitutes in positions 4 and 6 such as 4, 6-dimethyldibenzothiophene. Therefore, the HDS method for ultra-deep desulfurization of Chinese diesel fuel, which theoretically requires more hydrogen consumption on a much larger catalyst bed at higher temperature and pressure, becomes less efficient and more expensive^[4]. Hence, it is necessary to develop an alternative or supplementary process for lower-cost deep desulfurization of Chinese diesel fuel^[5].

  The adsorptive desulfurization method has been investigated as one of the most competitive and efficient techniques for producing ultraclean fuel^[4-7]. A variety of adsorbents have been reported as potential desulfurization adsorbents, including silver^{[8, 9]}, copper^[10-12], nickel^[13-15] and cerium^[16] supported on porous materials such as carbon materials^[17-19], oxides and mixed oxides^{[20, 21]}, zeolites^[22-24] and metal organic frameworks (MOFs)^{[25, 26]}. However, the desulfurization efficiencies of most of these adsorbents are not high enough for commercial application due to the steric hindrance effect of alkylated sulfur compounds and the strong competitive adsorption of polyaromatic hydrocarbons^{[4, 5]}, except few adsorbent formulations like silver supported on TiO₂-Al₂O₃ and TiO₂-SiO₂, which show effective desulfurization capability for diesel supplied by USA companies, including Off-Road Diesel (452 mg/kg) and Ultra Low Sulfur Diesel (7.5 mg/kg) at ambient conditions^{[5, 8, 27]}. Silver (I) oxide (Ag₂O) was the active Ag state and the interaction was proved to be the π-interactions both experimentally and theoretically^{[8, 9]}.

  However, it is known that the concentrations of refractory alkyl dibenzothiophenes, nitrogen compounds and polycyclic aromatic hydrocarbons included in Chinese diesel fuels is quite different from that of USA diesels. It is proposed here to incorporate Ce component for deeper desulfurization because it might have several advantages such as: doping the surfaces with Ce would promote the formation of coordinative unsaturated sites; cerium oxide would interact with metallic silver to generate more active sites with Ag oxides; and more defects of crystal structure due to incorporation of Ce would allow silver components to be better dispersed onto the supports^{[21, 28-30]}. TiO₂-CeO₂/MCM-48 has been reported as a high-efficiency adsorbent for the ultra-deep desulfurization of diesel fuel through an irradiation oxidation-adsorption desulfurization approach^{[4, 31]}.

  In the present study, CeO_x/TiO₂-SiO₂ was prepared through a novel co-impregnation method and utilized as the support for silver based desulfurization adsorbents. N₂ absorption-desorption, X-ray photoelectron spectroscopy (XPS), X-ray diffraction (XRD) and transmission electron microscope (TEM) were employed to characterize active species and the dispersion. Finally, the impacts of CeO_x doping on the desulfurization performance of the adsorbents was also studied. The tests were performed using Chinese II standard (CN-II), Chinese III standard (CN-III), and Chinese IV standard (CN-IV) transportation fuel. The effect of CeO_x doping on adsorptive capacity of the adsorbents for organosulfur was examined in both static batch test and dynamic breakthrough test.

  1   Experimental

  1.1   Preparation of adsorbent

  The SiO₂ powder support was purchased from Qingdao Xinchanglai silica gel Co., Ltd.. Titanium butoxide (99%+) and titanium isopropoxide (97%+) were acquired from Aladdin and Alfa Aesar, respectively. Ce (NO₃)₃·6H₂O and AgNO₃ were analytical-grade reagents purchased from Shanghai Colloid Chemical Plant and were used as received.

  The CeO_x/TiO₂-SiO₂ was prepared through a novel co-impregnation method with 1:2:44.3 molar ratio of Ce:Ti:Si. Briefly, after undergoing calcination at 823 K, 1.2 g SiO₂ powder was immersed in 20 mL anhydrous ethanol solution of tetrabutyl titanate (0.045 mol/L) and stirred in an ice-water bath for 60 min to form suspension. Then, 10 mL of 0.045 mol/L cerium nitrate in 80% ethanol solution was dropped into the suspension solution (0.5 mL/min). The slurry was treated by ultrasonic (50 W) for 30 min subsequently, and then oven-dried overnight at 333 K. Finally, the powdery samples were calcined in air at 823 K with ramp heating rates of 2 K/min for 2 h to prepare the CeO_x/TiO₂-SiO₂ support. TiO₂-SiO₂ support was prepared by the similar method but without addition of Ce species.

  Silver was dispersed onto the above supports through typical incipient wetness method^{[5, 8]}, where aqueous AgNO₃ was used as the silver precursor. The incipient wetness volume of supports was about 2.33 mL/g and the corresponding Ag metal loading was 4%. The impregnated samples were dried at 383 K for 6 h and then calcined in air at 723 K with the ramp heating rates of 2 K/min for 2 h.

  Ag-TiO₂-Al₂O₃ was also prepared according to the reference for comparison with the previous studies^[5]. Here, TiO₂-Al₂O₃ (1:4.4 as the weight ratio of Ti to Al) was prepared with Al₂O₃ purchased from Alfa Aesar and titanium isopropoxide dissolved in isopropanol through incipient wetness method. Then, a 4% of Ag was loaded on the TiO₂-Al₂O₃ through incipient wetness method to obtain the final adsorbent.

  1.2   Characterization of adsorbent

  N₂-physisorption was utilized to determine pore structure of adsorbents and supports. The N₂-physisorption isotherms were obtained at 77 K with a Micromeritics ASAP 2020 adsorption porosimeter. XRD measurements were performed on a PANalytical X'pert Pro Super X-ray diffractometer operating in the reflection with Co Kα radiation (40 kV, 30 mA). Diffraction patterns were taken over a 2θ range of 5°-90° and scanned at a speed of 0.208 8(°)/s with a step size of 0.016 7°. XPS measurements were performed with ESCALAB 250 (Thermo Scientific) spectrometer with a monochromatized Al Kα radiation. XPS spectra were obtained with pass energy of 30 eV. The C 1s hydrocarbon peak was fixed at a binding energy of 284.66 eV to subtract the surface charging effect. Images of transmission electron microscopy (TEM), high-resolution transmission electron microscopy (HRTEM) and high angle annular dark field scanning transmission electron microscopy (HAADF-STEM) were recorded on a Tecnai G2F20 microscope. The samples were suspended in methanol and dispersed on standard Cu TEM-grids covered with lacey carbon.

  1.3   Desulfurization experiments

  CN-II diesel, CN-III diesel, and CN-IV diesel were collected from Sinopec Esso petrol station (492 Xihong Road, Fuzhou, Fujian). The model diesel was prepared by adding 4, 6-dimethyldibenzothiophene into n-dodecane solvent. The initial sulfur contents of CN-II diesel, CN-III diesel, CN-IV diesel and model diesel were 952.9, 292.1, 39.0 and 43.8 mg/kg, respectively, measured by ultraviolet fluorescence using a total sulfur analyzer (TS-3000, Jiangsu Jiangfen Electroanalysis Instrument Co., Ltd.). The lower detection limit of TS-3000 was 0.5 mg/kg.

  Both static batch tests and dynamic breakthrough tests were performed in order to assess the desulfurization performances of the adsorbents. In static equilibrium tests, 1-3 g adsorbent was mixed with 30 mL fuel and the mixture was mechanically shaken at 333 K for 48 h. The equilibrated fuel was analyzed to measure the sulfur content, which was used to calculate equilibrium sulfur adsorption capacity. In adsorption rate curve tests, 3 g adsorbent was mixed with 90 mL CN-IV diesel fuel in a flask at 333 K. The mixture was stirred vigorously and sampled at set intervals.

  The dynamic breakthrough tests were performed using CN-IV diesel at 298 K and atmospheric pressure. In each experiment, 0.7 g adsorbent (270-425 μm size) was packed in a quartz tube with inner diameter of 5 mm using quartz wool to cover both ends and imbedded in a vertical thermostatic oven. Fuels were pumped by a plunger pump and flowed upward vertically with a liquid hourly space velocity (LHSV) of 0.3-0.9 h^-1 to ensure complete wetting of all adsorbent particles. Samples were collected and analyzed at outlet at set intervals. The sulfur content of fuels was determined by ultraviolet fluorescence using a total sulfur analyzer (TS-3000, Jiangsu Jiangfen Electroanalysis Instrument Co., Ltd.). The regeneration experiment of Ag-CeO_x/TiO₂-SiO₂ was performed by heating the adsorbent bed in flowing air. The temperature was ramped to 473 K for 2 h and then ramped to 723 K for 2 h with the ramp heating rates of 2 K/min for 2 h.

  2   Results and discussion

  2.1   Pore structural characterization of the adsorbents

  Compared with incipient wetness impregnation reported in the reference^[5], where the pore volume of Ag-TiO₂-Al₂O₃ was only half of support Al₂O₃, the deposition-precipitation method employed here can effectively reduce the blockage of the pore structure (Figure 1 and Table 1).

  Figure 1.  N₂-physisorption curves of different adsorbents and supports

  图选项

  下载全尺寸图像

  下载幻灯片

  Table 1.  Porous parameters of different adsorbents and supports

  表选项

  导出Excel

  下载全尺寸表图像

  Sample IDSurface area A/(m²·g^-1)Pore volume v/(cm³·g^-1)Pore diameter d/nmSiO₂341.31.0011.7Ag-CeO_x/TiO₂-SiO₂278.80.7610.8

  TiO2-SiO2	335.9	0.91	10.8
  Ag-TiO2-SiO2	297.2	0.83	11.1
  CeOx/TiO2-SiO2	318.0	0.82	10.3

  Table 1.  Porous parameters of different adsorbents and supports

  The N₂-physisorption isotherms of different adsorbents are similar in shape (Figure 1). The obtained porous parameters are shown in Table 1. The differences in the pore structure parameters of the studied adsorbents indicate the decline of specific surface area and pore volume after the introduction of active species. But the decline is slight and the average pore size of the adsorbent does not change significantly. It can be speculated that the active species do not form large clusters which would block the pores, and thus the pore structure of the adsorbent remains well.

  2.2   Determination of valence states of various species using XPS

  The valence state of titanium and cerium were studied by XPS, as shown in Figure 2 and Figure 3, respectively. The binding energy (BE) of Ti 2p_3/2 and Ti 2p_1/2 in CeO_x/TiO₂-SiO₂ is 458.9 eV and 464.5 eV, respectively. These are close to the standard BE of TiO₂ (458.7 and 464.4 eV) ^[32], which suggests that the valence state of titanium is Ti⁴⁺ and titanium species mainly exist as TiO₂. In the Ag-CeO_x/TiO₂-SiO₂, TiO₂ is also detected as the main form of titanium species.

  Figure 2.  XPS Ti 2p spectra of CeO_x/TiO₂-SiO₂ and Ag-CeO_x/TiO₂-SiO₂

  图选项

  下载全尺寸图像

  下载幻灯片

  Figure 3.  XPS Ce 3d spectra of CeO_x/TiO₂-SiO₂ and Ag-CeO_x/TiO₂-SiO₂

  图选项

  下载全尺寸图像

  下载幻灯片

  Due to the spin-orbit splitting, the spectra of Ce 3dare roughly separated to 3d_5/2 and 3d_3/2 multiples and exhibits three-lobed envelopes (Figure 3). The standard BE of Ce³⁺ and Ce⁴⁺ are quite distinct, but they will overlap when the two species are present together^{[33, 34]}. As shown in Table 3, v, v₂ and v₃ (u, u₂ and u₃) can be attributed to Ce⁴⁺, while v₀ and v₁ (u₀ and u₁) are due to Ce³⁺. The changes in the peak area and peak position could be used to distinguish between the Ce³⁺ and Ce⁴⁺ ions^{[35, 36]}. The probe peak u₃ (about 918 eV) observed in the XPS Ce 3d spectra of CeO_x/TiO₂-SiO₂ and Ag-CeO_x/TiO₂-SiO₂ (Figure 3) confirms the presence of the CeO₂. Both peaks v and v₂ shift toward v₁, which is the primary peak of Ce³⁺. This indicate that a certain proportion of Ce species are present in the form of Ce₂O₃. The area percentages (the results of curves fitting are not shown here for brevity) of u₃ peak in the Ce 3d region of CeO_x/TiO₂-SiO₂ (13.2%) and Ag-CeO_x/TiO₂-SiO₂ (13.0%) are both less than 14%, which also suggests the co-existence of Ce₂O₃^[36]. However, the proportion of Ce₂O₃ is quite small, because the BE shifts are small and the satellite peak v₀ is almost invisible. Therefore, the Ce should exist primarily in the form of CeO₂ with some oxygen vacancies.

  Table 3.  Static equilibrium test results of different adsorbents and supports

  表选项

  导出Excel

  下载全尺寸表图像

  AdsorbentA/F^b /(g·mL^-1)DieselC_e^{a, b} /(mg·kg^-1)η^{a, b} /%C_S^{a, b} /(mg·g^-1)SiO₂1:30CN-IV30.621.60.21Ag-TiO₂-Al₂O₃1:30CN-IV29.325.00.25

  TiO2-SiO2	1:30	CN-IV	26.5	32.0	0.31
  CeOx/TiO2-SiO2	1:30	CN-IV	25.6	34.4	0.35
  Ag-SiO2	1:30	CN-IV	27.6	29.3	0.29
  Ag-TiO2-SiO2	1:30	CN-IV	23.1	40.8	0.40
  Ag-CeOx/TiO2-SiO2	1:30	CN-IV	19.7	49.5	0.49
  Ag-CeOx/TiO2-SiO2	1:20	CN-IV	14.8	62.0	0.41
  Ag-CeOx/TiO2-SiO2	1:10	CN-IV	8.1	79.1	0.26
  Ag-CeOx/TiO2-SiO2	1:30	CN-III	166.6	43.0	3.16
  Ag-CeOx/TiO2-SiO2	1:30	CN-II	738.5	22.4	5.38
  a all the desulfurization were processed at 333 K and atmospheric pressure, the adsorbents were mixed with the diesels under mechanical shaking for 48 h, respectively; b A/F represent adsorbent to diesel fuel ratios; Ce represent the equilibrium concentration; η is the desulphurization efficiencies; CS represent the sulfur adsorption capacity

  Table 3.  Static equilibrium test results of different adsorbents and supports

  The obvious BE shift of Si 2p (Table 2) reveals the strong interaction between Ce, Ti species and SiO₂ supports. Meanwhile, it is also found that the loading of Ag species caused the BE shift of SiO₂, TiO₂ and CeO_x by comparing XPS data for CeO_x/TiO₂-SiO₂and Ag-CeO_x/TiO₂-SiO₂ supports. These shifts (ranged in 0.2-0.6 eV) should be the result of interactions between Ag species and the other supports, suggesting a high dispersion of Ag species among CeO_x, TiO₂ and SiO₂. All these results suggest that our method is effective for co-loadings of highly dispersed multicomponents. Furthermore, the BE shift of O 1s (Table 2) shows that the metallic Ag is at least partly oxidized and exists as Ag oxides in the Ag-CeO_x/TiO₂-SiO₂ adsorbent. However, the BE of Ag is not a good way to determine Ag oxidation state in the Ag-CeO_x/TiO₂-SiO₂ due to the insensitive BE and the relatively small amount of Ag^[37]. The oxidation state of Ag would be analyzed in following determination.

  Table 2.  XPS binding energy of O 1s, Si 2p and Ce 3d peaks

  表选项

  导出Excel

  下载全尺寸表图像

  LevelsPeakBinding energy E_B/eVreferenceCeO_x/TiO₂-SiO₂Ag-CeO_x/TiO₂-SiO₂Ce 3d_5/2^av₀880.6--O 1s^b533.2532.9533.1

  	v	882.6	883.6	883.8
  	v1	885.45	-	-
  	v2	888.85	887.2	887.5
  	v3	898.4	899.9	900.5
  Ce 3d3/2a	u0	898.9	-	-
  	u	901.05	902	902.3
  	u1	904.05	-	-
  	u2	907.45	905.8	906.1
  	u3	916.7	918.2	918.8
  Si 2pb		103.7	102.9	103.0
  a: the reference binding energy of the ten Ce 3d peaks were from reference[33]; b: the reference binding energy of O 1s and Si 2p in silica gel were from reference[34]

  Table 2.  XPS binding energy of O 1s, Si 2p and Ce 3d peaks

  2.3   Determination of crystall stuctures of various species using XRD

  XRD patterns are shown in Figure 4. The patterns of TiO₂-SiO₂ and CeO_x/TiO₂-SiO₂ supports, and Ag-TiO₂-SiO₂ and Ag-CeO_x/TiO₂-SiO₂ adsorbents do not show any remarkable peaks, except at 2θ of about 25°, implying that the silver, titanium oxide and cerium oxide are in an amorphous and disordered phase or the crystal particles are too small to be detected by XRD. It is also possible that the introduction of TiO₂ and CeO_x provides more surface defects and enhances the interaction between the active Ag species and the carrier, thus facilitating the dispersion of the active Ag species and even changing the valence state of the active Ag species. Therefore, silver, cerium, titanium should have been evenly dispersed on the supports.

  Figure 4.  XRD patterns of different adsorbents

  图选项

  下载全尺寸图像

  下载幻灯片

  2.4   Determination of Ag state on CeO_x/TiO₂-SiO₂ using TEM

  In previous XPS and XRD experiments, it was speculated that at least a part of Ag should be presented as Ag oxides and well dispersed on the CeO_x/TiO₂-SiO₂ supports. TEM study was performed here to further clarify how silver was incorporated in the Ag-TiO₂-SiO₂ and Ag-CeO_x/TiO₂-SiO₂ as shown in Figure 5 and Figure 6.

  Figure 5.  TEM images of Ag-TiO₂-SiO₂

  图选项

  下载全尺寸图像

  下载幻灯片

  Figure 6.  TEM images of Ag-CeO_x/TiO₂-SiO₂

  图选项

  下载全尺寸图像

  下载幻灯片

  Bright field TEM images of Ag-TiO₂-SiO₂ and Ag-CeO_x/TiO₂-SiO₂ are displayed in Figure 5(a) and Figure 6(a). It is obvious that there are granular clusters dispersed on the surface of Ag-TiO₂-SiO₂, whereas it is not observed in Figure 6(a), suggesting a high dispersion of Ag component on CeO_x/TiO₂-SiO₂ support.

  HAADF-STEM (High-Angle Annular Dark Field-Scanning Transmission Electron microscope) tests are also performed as depicted in Figure 5(b) and Figure 6(b). In Figure 5(b), the Ag particles, which have higher Z compared to the Ti and Si, are imaged as brighter dots. The diameters of these dots are about 35 nm. In contrast, no dot is observed in the images of Ag-CeO_x/TiO₂-SiO₂ (Figure 6(b)). This indicates that Ag is more uniformly dispersed on CeO_x modified support as found on the bright field image.

  In addition, HRTEM images were recorded to analyze the component of active species. The distance of crystal plane (d, nm) was measured directly or analyzed through Fast Fourier Transform (FFT). In Figure 5(c), the distance of crystal plane of 0.141, 0.203 and 0.237 nm correspond to the crystallographic planes of Ag in metallic state (JCPDS file: 00-004-0783). The d of 0.168 nm might be attributed to Ag₂O (JCPDS file: 00-041-1104), which has been detected on the Ag/TiO₂ adsorbents in the reference^[37]. These reveal that Ag species particles exist mainly in metallic state together with a small amount in the form of oxides (e.g. Ag₂O).

  The d of 0.234 and 0.244 nm are found to match the crystallographic planes of anatase TiO₂ (JCPDS file: 00-021-1272), which is highly dispersed on the surface of the support with a particle size of about 2-5 nm (Figure 5(d)).

  Likewise, in Figure 6(c) and Figure 6(d), the d of 0.324 and 0.321 nm are found to be close to the (110) crystallographic planes of rutile TiO₂ (JCPDS file: 00-021-1276), the d of 0.201 and 0.308 nm might be attributed to CeO_x (JCPDS file of Ce₂O₃: 00-023-1048, JCPDS file of CeO₂: 00-034-0394), and the d of 0.277, 0.280 and 0.285 nm are attributed to Ag₂O₂ (JCPDS file: 00-051-0945). It is obvious that the crystal structure of TiO₂ and CeO_x is incomplete and highly distorted, and thus a significant amount of coordinative unsaturated sites are formed. These provides extra sites for the incorporation of Ag species and even changes the valence state of Ag active centers. Ag active centers on CeO_x/TiO₂-SiO₂ are in a state of highly electron-deficient Ag²⁺, which could enhance the affinity between adsorbents and sulfur compounds via π interaction. Furthermore, the sizes of Ag₂O₂ are about 5 nm instead of 35 nm, which is the average size of the Ag particles in Ag-TiO₂-SiO₂. The dispersion of Ag species is significantly improved as compared with Ag-TiO₂-SiO₂, which could directly improve the adsorptive desulfurization performance.

  2.5   Effect of Ce on organosulfur adsorption performance

  The static equilibrium tests of the supports and adsorbents were processed at 333 K and atmospheric pressure. The comparison between the capacities of the supports is illustrated in Table 3, which indicates significant improvements of sulfur adsorption capacities of the supports after being modified by titanium oxide and cerium oxide. The improvements in capacity for desulfurization after the loading of Ag species are significantly different among different supports. A following order is found: SiO₂ (0.076 mg/g) ＜ TiO₂-SiO₂ (0.086 mg/g) ＜ CeO_x/TiO₂-SiO₂ (0.149 mg/g).

  An excellent improvement of sulfur capacity (about 22.5%) is observed in static equilibrium tests after the Ag-TiO₂-SiO₂ absorbent is doped with CeO_x. The desulfurization rate of CN-IV diesel could reach 49.5% with 1:30 as the ratio of adsorbent Ag-CeO_x/TiO₂-SiO₂ to diesel. Ag-CeO_x/TiO₂-SiO₂ exhibits a more favorable desulfurization performance than Ag-TiO₂-Al₂O₃ which has been considered to be an better effective adsorbent for diesel desulfurization among the reported adsorbent formulations^[5]. A certain number of higher ratios of adsorbent to diesel oil are also performed in this study. It is found that the sulfur content could be reduced to 8.1 mg/kg (less than 10 mg/kg of the CN-V standard) as the ratio was 1:10 g/mL. On the other hand, the CN-II diesel and CN-III diesel, which have higher sulfur content, are also treated with Ag-CeO_x/TiO₂-SiO₂. The sulfur capacity is found to reach 5.38 mgS/g, when 1 g Ag-CeO_x/TiO₂-SiO₂ is mixed with 30 mL CN-II diesel under mechanical shaking.

  Adsorption rate curves of Ag-TiO₂-SiO₂ and Ag-CeO_x/TiO₂-SiO₂ are also recorded in batch desulfurization tests at 333 K under vigorous stirring. The changes of relative sulfur concentration (C/C₀) over time are depicted in Figure 7. It is obvious that Ag-CeO_x/TiO₂-SiO₂ exhibits a more favorable desulfurization performance than Ag-TiO₂-SiO₂. As shown in Figure 7, total sulfur content of CN-IV diesel fuel decreases as the time extends. The rate is rather high at the initial stages of the contact period, and then decreased in the rest of time. This phenomenon is due to the fact that a large number of vacant adsorptive sites are available for adsorption during the initial stage, and the remaining vacant adsorptive sites are difficult to be occupied due to competitive adsorption of polycyclic aromatic hydrocarbons through π-interaction and the slow internal diffusion.

  Figure 7.  Relative sulfur concentration (C/C₀) varies with adsorption time 3 g adsorbent was mixed with 90 mL CN-IV diesel fuel in a flask at 333 K and atmospheric pressure under vigorous stirring

  图选项

  下载全尺寸图像

  下载幻灯片

  When the adsorption time is 48 h, the desulfurization efficiencies of Ag-TiO₂-SiO₂ and Ag-CeO_x/TiO₂-SiO₂ are about 51.8% and 64.6%, which are quite higher than that obtained in the static equilibrium test (40.8% and 49.5%). This is due to the fact that vigorous stirring, which was employed in the static equilibrium test, is more favorable than that of shaking. The effect of initial concentration (Table 3) also reveals that the mass transfer driving force becomes larger when the initial concentration increases, and thus resulting in higher sulfur adsorption capacity. All these illustrate that the mass diffusion, including external diffusion and internal diffusion, would have a serious impact on the desulfurization efficiency of Ag-CeO_x/TiO₂-SiO₂. It will take further work to tap the desulfurization potential of Ag-CeO_x/TiO₂-SiO₂.

  On the other hand, the effect of CeO_x doping in supporter was also examined in dynamic breakthrough tests. The breakthrough characteristics of the adsorbents with CN-IV are illustrated in Figure 8. It is obvious that the curve of Ag-TiO₂-SiO₂ breaks more early and rapidly than that of Ag-CeO_x/TiO₂-SiO₂. Additionally, the variation of LHSV in the range of 0.3-0.9 h^-1 shows some slight effect on desulfurization performance (Figure 8). It shows higher performance under a lower LHSV. The breakthrough curves of Ag-CeO_x/TiO₂-SiO₂ with a LHSV of 0.3 h^-1 showed in the inset graph in Figure 8 indicate that the trend toward saturation is quite slow, which is associated with the too slow mass diffusion rate of sulfur compounds or the competitive adsorption of polycyclic aromatic hydrocarbons. When the elution volume reaches 200 mL/g adsorbent, the relative sulfur concentration (C/C₀) is 0.83 and the sulfur capacity of adsorbent is found to be up to 2.0 mg/g.

  Figure 8.  Breakthrough curves of adsorbents obtained with CN-IV diesel at 298 K, atmospheric pressure and different LHSV

  图选项

  下载全尺寸图像

  下载幻灯片

  To further investigate the adsorption performance of Ag-CeO_x/TiO₂-SiO₂, adsorptive selectivity and regenerability of Ag-CeO_x/TiO₂-SiO₂ are also studied. The selectivity of adsorbents is revealed by contrasting the adsorption rate curves of CN-IV diesel and model diesel as shown in Figure 9. It is found that the competitive adsorption components, including polycyclic aromatic hydrocarbon, trace amounts of organonitrogen, fuel additives and moisture in commercial diesel have an impact on adsorption rate, but have no significant effect on equilibrium adsorption value. Ag-CeO_x/TiO₂-SiO₂ has a good adsorptive selectivity property for sulphur compounds. The results of the regeneration experiment (Figure 10) indicate that Ag-CeO_x/TiO₂-SiO₂ also has a favorable regenerability.

  Figure 9.  Contrasting the adsorption rate curves of CN-IV diesel and model diesel 3 g Ag-CeO_x/TiO₂-SiO₂ was mixed with 90 mL diesel fuel in a flask at 333 K and atmospheric pressure under vigorous stirring

  图选项

  下载全尺寸图像

  下载幻灯片

  Figure 10.  Regeneration experiment of Ag-CeO_x/TiO₂-SiO₂ obtained with CN-IV diesel at 298 K, atmospheric pressure and 0.3 h^-1 of LHSV

  图选项

  下载全尺寸图像

  下载幻灯片

  Taken together, all the adsorption performance tests indicate that doping CeO_x in the TiO₂-SiO₂ supporter could improve the desulfurization performance of Ag-based adsorbent effectively and dramatically. Ag-CeO_x/TiO₂-SiO₂ is considered to be one of efficient and potential adsorbents for desulfurization of diesel fuel.

  2.6   Effect of Ce on Ag-based adsorbent

  Based on the characteristics of the adsorbents, no peak corresponding to Ce species and Ti species could be observed in the XRD patterns (Figure 4). The pore structure of adsorbents remains well during the co-impregnation. Additionally, the sizes of Ce species and Ti species with a particle size about 2-5 nm are observed in the TEM images (Figure 5 and Figure 6). All these indicate that Ce species and Ti species are adequately dispersed on the supports and suggest that our method is effective for co-loadings of highly dispersed multi-components.

  The doped Ce species are demonstrated to be existed as CeO₂ containing a certain amount of Ce₂O₃ through the XPS spectra of Ce 3d (Figure 3). The Ti species is detected as rutile TiO₂ on Ag-CeO_x/TiO₂-SiO₂. It is different from anatase TiO₂ detected on Ag-TiO₂-SiO₂. The doping of CeO_x might promote the phase transition from anatase TiO₂ to rutile TiO₂. Furthermore, the interaction between the doped CeO_x and TiO₂ on the surface of SiO₂ would form a surface containing a large number of defect sites (as reflected in Figure 6), which could provide extra sites for sulfur compound adsorption directly. Thus, the sulfur capacity per unit surface area of TiO₂-SiO₂ and CeO_x/TiO₂-SiO₂is 9.2 mg/cm² and 11.0 mg/cm², respectively.

  Furthermore, the increase of defect sites also leads to the increase of active sites for the Ag loading and enhances the interaction between Ag species and supports, thus promoting the dispersion of Ag species. The binding energy shifts of Si 2p, Ti 2d and Ce 3d after the Ag loading also support this point (Figure 4 and Figure 5).

  Moreover, the reversible CeO₂ and Ce₂O₃ reduction transition might also facilitate the oxidation of metallic Ag to form Ag oxides, which would further enhance the affinity between the Ag active centers and sulfur compounds via π-interactions. Thus, the improvement due to Ag species loading on CeO_x/TiO₂-SiO₂ (0.149 mg/g) is significantly higher than that on TiO₂-SiO₂ (0.086 mg/g). Ag-CeO_x/TiO₂-SiO₂ exhibits a more favorable desulfurization performance compared with Ag-TiO₂-SiO₂.

  3   Conclusions

  CeO_x was doped in the TiO₂-SiO₂ supporter of Ag-based desulfurization adsorbent for Chinese standard diesel fuel by a novel co-impregnation method. Ce species and Ti species were observed to be adequately dispersed on the supports, and the pore structure of adsorbents maintained well. The experimental results demonstrated that Ce species existed as CeO₂ which contained a certain amount of oxygen vacancies. The introduction of CeO_x could promote the phase transition from anatase TiO₂ to rutile TiO_2.The interaction between the doped CeO_x and TiO₂ on the supports could not only provide extra sites for sulfur adsorption, but also offer extra sites for incorporation of Ag species. The doping of CeO_x could dramatically facilitate the dispersion of Ag species and change the valence state of Ag species. Ag species supported on CeO_x/TiO₂-SiO₂ were detected as silver (II) oxides with a particle size of about 5 nm instead of 35 nm, the size of the metallic silver particles. The adequately dispersed Ag active centers in highly electron-deficient state enhanced the affinity between adsorbents and sulfur compounds via π-interactions. The doping of Ce enhanced the desulfurization performances of Ag-TiO₂-SiO₂ adsorbents for CN standard diesel fuels dramatically in static equilibrium test. Ag-Ce_xO/TiO₂-SiO₂ can reduce the sulfur content of CN-IV diesel fuel from 39.0 mg/kg initially down to less than 10 mg/kg in the static equilibrium test. The equilibrium sulfur capacity also reached 5.38 mg/g for CN-II (952.9 mg/kg initially). Compared with Ag-TiO₂-SiO₂, Ag-CeO_x/TiO₂-SiO₂ was proved to be a more efficient adsorbent in both adsorption curve tests and dynamic breakthrough tests. The desulfurization efficiency of Ag-CeO_x/TiO₂-SiO₂ was constrained by mass diffusion. All these findings demonstrated that the adsorptive desulfurization presented in this study should be a promising adsorbent for deep desulfurization of CN-IV diesel to meet CN-V mandatory request for sulfur component.

• 1. [1]

     WU L, XIAO J, WU Y, XIAN S, MIAO G, WANG H, LI Z. A combined experimental/computational study on the adsorption of organosulfur compounds over metal-organic frameworks from fuels[J]. Langmuir, 2014, 30(4):  1080-1088. doi: 10.1021/la404540j

  2. [2]

     PALOMINO J M, TRAN D T, HAUSER J L, DONG H, OLIVER S R. Mesoporous silica nanoparticles for high capacity adsorptive desulfurization[J]. J Mater Chem, 2014, 2(36):  14890-14895. doi: 10.1039/C4TA02570A

  3. [3]

     XU X, ZHANG S, LI P, SHEN Y. Adsorptive desulfurization of liquid Jet-A fuel at ambient conditions with an improved adsorbent for on-board fuel treatment for SOFC applications[J]. Fuel Process Technol, 2014, 124:  140-146. doi: 10.1016/j.fuproc.2014.03.001

  4. [4]

     XIAO J, WANG X, CHEN Y, FUJⅡ M, SONG C. Ultra-deep adsorptive desulfurization of light-irradiated diesel fuel over supported TiO₂-CeO₂ adsorbents[J]. Ind Eng Chem Res, 2013, 52(45):  15746-15755. doi: 10.1021/ie402724q

  5. [5]

     HUSSAIN A S, TATARCHUK B J. Adsorptive desulfurization of jet and diesel fuels using Ag/TiO_x-Al₂O₃ and Ag/TiO_x-SiO₂ adsorbents[J]. Fuel, 2013, 107:  465-473. doi: 10.1016/j.fuel.2012.11.030

  6. [6]

     QIN Y, MO Z, YU W, DONG S, DUAN L, GAO X, SONG L. Adsorption behaviors of thiophene, benzene, and cyclohexene on FAU zeolites:Comparison of CeY obtained by liquid-, and solid-state ion exchange[J]. Appl Surf Sci, 2014, 292:  5-15. doi: 10.1016/j.apsusc.2013.11.036

  7. [7]

     KHAN N A, HASAN Z, JHUNG S H. Ionic liquids supported on metal-organic frameworks:Remarkable adsorbents for adsorptive desulfurization[J]. Chem Eur J, 2014, 20(2):  376-380. doi: 10.1002/chem.v20.2

  8. [8]

     HUSSAIN A S, TATARCHUK B J. Mechanism of hydrocarbon fuel desulfurization using Ag/TiO₂-Al₂O₃ adsorbent[J]. Fuel Process Technol, 2014, 126:  233-242. doi: 10.1016/j.fuproc.2014.05.006

  9. [9]

     HUSSAIN A S, MCKEE M L, HEINZEL J M, SUN X, TATARCHUK B J. Density functional theory study of organosulfur selective adsorption on Ag-TiO₂ adsorbents[J]. J Phys Chem C, 2014, 118(27):  14938-14947. doi: 10.1021/jp503097y

  10. [10]

      LIU B, ZHU Y, LIU S, MAO J. Adsorption equilibrium of thiophenic sulfur compounds on the Cu-BTC metal-organic framework[J]. J Chem Eng Data, 2012, 57(4):  1326-1330. doi: 10.1021/je300130s

  11. [11]

      YANG R T, HERNANDEZ-MALDONADO A J, YANG F H. Desulfurization of transportation fuels with zeolites under ambient conditions[J]. Science, 2003, 301(5629):  79-81. doi: 10.1126/science.1085088

  12. [12]

      HE G, SUN L, SONG X, LIU X, YIN Y, WANG Y. Adjusting host properties to promote cuprous chloride dispersion and adsorptive desulfurization sites formation on SBA-15[J]. Energy Fuels, 2011, 25(8):  3506-3513. doi: 10.1021/ef200723m

  13. [13]

      XU X, ZHANG S, LI P, SHEN Y. Equilibrium and kinetics of Jet-A fuel desulfurization by selective adsorption at room temperatures[J]. Fuel, 2013, 111:  172-179. doi: 10.1016/j.fuel.2013.04.068

  14. [14]

      MA X, SPRAGUE M, SONG C. Deep desulfurization of gasoline by selective adsorption over nickel-based adsorbent for fuel cell applications[J]. Ind Eng Chem Res, 2005, 44(15):  5768-5775. doi: 10.1021/ie0492810

  15. [15]

      VELU S, MA X, SONG C, NAMAZIAN M, SETHURAMAN S, VENKATARAMAN G. Desulfurization of JP-8 jet fuel by selective adsorption over a Ni-based adsorbent for micro solid oxide fuel cells[J]. Energy Fuels, 2005, 19(3):  1116-1125. doi: 10.1021/ef049800b

  16. [16]

      SHEN Y, XU X, LI P. A novel potential adsorbent for ultra deep desulfurization of jet fuels at room temperature[J]. RSC Adv, 2012, 2(15):  6155-6160. doi: 10.1039/c2ra20224g

  17. [17]

      TRIANTAFYLLIDIS K S, DELIYANNI E A. Desulfurization of diesel fuels:Adsorption of 4, 6-DMDBT on different origin and surface chemistry nanoporous activated carbons[J]. Chem Eng J, 2014, 236:  406-414. doi: 10.1016/j.cej.2013.09.099

  18. [18]

      WANG L, YANG R T, SUN C L. Graphene and other carbon sorbents for selective adsorption of thiophene from liquid fuel[J]. AlChE J, 2013, 59(1):  29-32. doi: 10.1002/aic.v59.1

  19. [19]

      BALTZOPOULOU P, KALLIS K X, KARAGIANNAKIS G, KONSTANDOPOULOS A G. Diesel fuel desulfurization via adsorption with the aid of activated carbon:Laboratory-and pilot-scale studies[J]. Energy Fuels, 2015, 29(9):  5640-5648. doi: 10.1021/acs.energyfuels.5b01133

  20. [20]

      XU X, ZHANG S, LI P, SHEN Y. Desulfurization of Jet-A fuel in a fixed-bed reactor at room temperature and ambient pressure using a novel selective adsorbent[J]. Fuel, 2014, 117:  499-508. doi: 10.1016/j.fuel.2013.09.074

  21. [21]

      GUO J, JANIK M J, SONG C. Density functional theory study on the role of ceria addition in Ti_xCe_1-xO₂ adsorbents for thiophene adsorption[J]. J Phys Chem C, 2012, 116(5):  3457-3466. doi: 10.1021/jp2063996

  22. [22]

      TIAN F, SHEN Q, FU Z, WU Y, JIA C. Enhanced adsorption desulfurization performance over hierarchically structured zeolite Y[J]. Fuel Process Technol, 2014, 128:  176-182. doi: 10.1016/j.fuproc.2014.07.018

  23. [23]

      WANG Y, YANG R T, HEINZEL J M. Desulfurization of jet fuel byπ-complexation adsorption with metal halides supported on MCM-41 and SBA-15 mesoporous materials[J]. Chem Eng Sci, 2008, 63(2):  356-365. doi: 10.1016/j.ces.2007.09.002

  24. [24]

      HERNANDEZ-MALDONADO A J, YANG R T. Desulfurization of commercial liquid fuels by selective adsorption viaπ-complexation with Cu (I)-Y zeolite[J]. Ind Eng Chem Res, 2003, 42(13):  3103-3110. doi: 10.1021/ie0301132

  25. [25]

      PERALTA D, CHAPLAIS G, SIMON-MASSERON A, BARTHELET K, PIRNGRUBER G D. Metal-organic framework materials for desulfurization by adsorption[J]. Energy Fuels, 2012, 26(8):  4953-4960. doi: 10.1021/ef300762z

  26. [26]

      LIU X, WANG J, LI Q, JIANG S, ZHANG T, JI S. Synthesis of rare earth metal-organic frameworks (Ln-MOFs) and their properties of adsorption desulfurization[J]. J Rare Earths, 2014, 32(2):  189-194. doi: 10.1016/S1002-0721(14)60050-8

  27. [27]

      NAIR S, TATARCHUK B J. Supported silver adsorbents for selective removal of sulfur species from hydrocarbon fuels[J]. Fuel, 2010, 89(11):  3218-3225. doi: 10.1016/j.fuel.2010.05.006

  28. [28]

      WATANABE S, MA X, SONG C. Characterization of structural and surface properties of nanocrystalline TiO₂-CeO₂ mixed oxides by XRD, XPS, TPR, and TPD[J]. J Phys Chem C, 2009, 113(32):  14249-14257. doi: 10.1021/jp8110309

  29. [29]

      SCIRE S, MINICO S, CRISAFULLI C, SATRIANO C, PISTONE A. Catalytic combustion of volatile organic compounds on gold/cerium oxide catalysts[J]. Appl Catal B:Environ, 2003, 40(1):  43-49. doi: 10.1016/S0926-3373(02)00127-3

  30. [30]

      ZHANG Y, ANDERSSON S, MUHAMMED M. Nanophase catalytic oxides:I. Synthesis of doped cerium oxides as oxygen storage promoters[J]. Appl Catal B:Environ, 1995, 6(4):  325-337. doi: 10.1016/0926-3373(95)00041-0

  31. [31]

      XIAO J, WANG X, FUJⅡ M, YANG Q, SONG C. A novel approach for ultra-deep adsorptive desulfurization of diesel fuel over TiO₂-CeO₂/MCM-48 under ambient conditions[J]. AlChE J, 2013, 59(5):  1441-1445. doi: 10.1002/aic.14085

  32. [32]

      GONBEAU D, GUIMON C, PFISTER-GUILLOUZO G, LEVASSEUR A, MEUNIER G, DORMOY R. XPS study of thin films of titanium oxysulfides[J]. Surf Sci, 1991, 254(1):  81-89.

  33. [33]

      ROMEO M, BAK K, FALLAH J E, NORMAND F L, HILAIRE L. XPS study of the reduction of cerium dioxide[J]. Surf Interface Anal, 1993, 20(6):  508-512. doi: 10.1002/(ISSN)1096-9918

  34. [34]

      GROSS T, RAMM M, SONNTAG H, UNGER W, WEIJERS H M, ADEM E H. An XPS analysis of different SiO₂ modifications employing a C 1s as well as an Au 4f_7/2 static charge reference[J]. Surf Interface Anal, 1992, 18(1):  59-64. doi: 10.1002/(ISSN)1096-9918

  35. [35]

      NAGPURE I, PITALE S S, TSHABALALA K, KUMAR V, NTWAEABORWA O, TERBLANS J, SWART H. Luminescence response and CL degradation of combustion synthesized spherical SiO₂:Ce nanophosphor[J]. Mater Res Bull, 2011, 46(12):  2359-2366. doi: 10.1016/j.materresbull.2011.08.051

  36. [36]

      LARACHI F, PIERRE J, ADNOT A, BERNIS A. Ce 3d XPS study of composite Ce_xMn_1-xO_2-y wet oxidation catalysts[J]. Appl Surf Sci, 2002, 195(1):  236-250.

  37. [37]

      SAMOKHVALOV A, NAIR S, DUIN E C, TATARCHUK B J. Surface characterization of Ag/titania adsorbents[J]. Appl Surf Sci, 2010, 256(11):  3647-3652. doi: 10.1016/j.apsusc.2010.01.002

• 

  Figure 1  N₂-physisorption curves of different adsorbents and supports

  a: SiO₂; b: TiO₂-SiO₂; c: Ag-TiO₂-SiO₂; d: CeO_x/TiO₂-SiO₂; e: Ag-CeO_x/TiO₂-SiO₂

  下载: 全尺寸图片 幻灯片
  

  Figure 2  XPS Ti 2p spectra of CeO_x/TiO₂-SiO₂ and Ag-CeO_x/TiO₂-SiO₂

  a: CeO_x/TiO₂-SiO₂; b: Ag-CeO_x/TiO₂-SiO₂

  下载: 全尺寸图片 幻灯片
  

  Figure 3  XPS Ce 3d spectra of CeO_x/TiO₂-SiO₂ and Ag-CeO_x/TiO₂-SiO₂

  a: CeO_x/TiO₂-SiO₂; b: Ag-CeO_x/TiO₂-SiO₂

  下载: 全尺寸图片 幻灯片
  

  Figure 4  XRD patterns of different adsorbents

  a: SiO₂; b: TiO₂-SiO₂; c: Ag-TiO₂-SiO₂; d: CeO_x/TiO₂-SiO₂; e: Ag-CeO_x/TiO₂-SiO₂

  下载: 全尺寸图片 幻灯片
  

  Figure 5  TEM images of Ag-TiO₂-SiO₂

  (a): bright field image; (b): HADDF-STEM image; (c), (d): HRTEM image

  下载: 全尺寸图片 幻灯片
  

  Figure 6  TEM images of Ag-CeO_x/TiO₂-SiO₂

  (a): BF image; (b): HADDF-STEM image; (c), (d): HRTEM image; (d) is the enlargement of the part enclosed by dotted lines in (c)

  下载: 全尺寸图片 幻灯片
  

  Figure 7  Relative sulfur concentration (C/C₀) varies with adsorption time 3 g adsorbent was mixed with 90 mL CN-IV diesel fuel in a flask at 333 K and atmospheric pressure under vigorous stirring

  ●: Ag-TiO₂-SiO₂; ■: Ag-CeO_x/TiO₂-SiO₂

  下载: 全尺寸图片 幻灯片
  

  Figure 8  Breakthrough curves of adsorbents obtained with CN-IV diesel at 298 K, atmospheric pressure and different LHSV

  ■: Ag-CeO_x/TiO₂-SiO₂ LHSV=0.3 h^-1;
  ●: Ag-CeO_x/TiO₂-SiO₂ LHSV=0.6 h^-1;
  ▲: Ag-CeO_x/TiO₂-SiO₂ LHSV=0.9 h^-1;
  ▼: Ag-TiO₂-SiO₂ LHSV=0.3 h^-1;
  ◆: Ag-TiO₂-SiO₂ LHSV=0.6 h^-1;
  ◀: Ag-TiO₂-SiO₂ LHSV=0.9 h^-1

  下载: 全尺寸图片 幻灯片
  

  Figure 9  Contrasting the adsorption rate curves of CN-IV diesel and model diesel 3 g Ag-CeO_x/TiO₂-SiO₂ was mixed with 90 mL diesel fuel in a flask at 333 K and atmospheric pressure under vigorous stirring

  ●: CN-IV diesel; ■: model diesel

  下载: 全尺寸图片 幻灯片
  

  Figure 10  Regeneration experiment of Ag-CeO_x/TiO₂-SiO₂ obtained with CN-IV diesel at 298 K, atmospheric pressure and 0.3 h^-1 of LHSV

  ■: fresh;
  ●: 1st recycling;
  ▲: 2nd recycling;
  ▼: 3rd recycling;
  ◆: 4th recycling;
  ◀: 5th recycling

  下载: 全尺寸图片 幻灯片

  Table 1.  Porous parameters of different adsorbents and supports

  Sample IDSurface area A/(m²·g^-1)Pore volume v/(cm³·g^-1)Pore diameter d/nmSiO₂341.31.0011.7Ag-CeO_x/TiO₂-SiO₂278.80.7610.8

  TiO2-SiO2	335.9	0.91	10.8
  Ag-TiO2-SiO2	297.2	0.83	11.1
  CeOx/TiO2-SiO2	318.0	0.82	10.3

  下载: 导出CSV

  Table 3.  Static equilibrium test results of different adsorbents and supports

  AdsorbentA/F^b /(g·mL^-1)DieselC_e^{a, b} /(mg·kg^-1)η^{a, b} /%C_S^{a, b} /(mg·g^-1)SiO₂1:30CN-IV30.621.60.21Ag-TiO₂-Al₂O₃1:30CN-IV29.325.00.25

  TiO2-SiO2	1:30	CN-IV	26.5	32.0	0.31
  CeOx/TiO2-SiO2	1:30	CN-IV	25.6	34.4	0.35
  Ag-SiO2	1:30	CN-IV	27.6	29.3	0.29
  Ag-TiO2-SiO2	1:30	CN-IV	23.1	40.8	0.40
  Ag-CeOx/TiO2-SiO2	1:30	CN-IV	19.7	49.5	0.49
  Ag-CeOx/TiO2-SiO2	1:20	CN-IV	14.8	62.0	0.41
  Ag-CeOx/TiO2-SiO2	1:10	CN-IV	8.1	79.1	0.26
  Ag-CeOx/TiO2-SiO2	1:30	CN-III	166.6	43.0	3.16
  Ag-CeOx/TiO2-SiO2	1:30	CN-II	738.5	22.4	5.38
  a all the desulfurization were processed at 333 K and atmospheric pressure, the adsorbents were mixed with the diesels under mechanical shaking for 48 h, respectively; b A/F represent adsorbent to diesel fuel ratios; Ce represent the equilibrium concentration; η is the desulphurization efficiencies; CS represent the sulfur adsorption capacity

  下载: 导出CSV

  Table 2.  XPS binding energy of O 1s, Si 2p and Ce 3d peaks

  LevelsPeakBinding energy E_B/eVreferenceCeO_x/TiO₂-SiO₂Ag-CeO_x/TiO₂-SiO₂Ce 3d_5/2^av₀880.6--O 1s^b533.2532.9533.1

  	v	882.6	883.6	883.8
  	v1	885.45	-	-
  	v2	888.85	887.2	887.5
  	v3	898.4	899.9	900.5
  Ce 3d3/2a	u0	898.9	-	-
  	u	901.05	902	902.3
  	u1	904.05	-	-
  	u2	907.45	905.8	906.1
  	u3	916.7	918.2	918.8
  Si 2pb		103.7	102.9	103.0
  a: the reference binding energy of the ten Ce 3d peaks were from reference[33]; b: the reference binding energy of O 1s and Si 2p in silica gel were from reference[34]

  下载: 导出CSV
•