Enhanced Adsorption Desulfurization Performance ... - ACS Publications

4 downloads 0 Views 6MB Size Report
Mar 23, 2017 - ABSTRACT: The adsorption sulfur capacity and selectivity are two key challenges for adsorption desulfurization technology of. FCC gasoline.
Article pubs.acs.org/IECR

Enhanced Adsorption Desulfurization Performance over Mesoporous ZSM‑5 by Alkali Treatment Haizheng Li,† Lixia Dong,† Liang Zhao,* Liyuan Cao, Jinsen Gao, and Chunming Xu State Key Laboratory of Heavy Oil Processing, China University of Petroleum (Beijing), 18 Fuxue Road, Beijing 102249, P. R. China S Supporting Information *

ABSTRACT: The adsorption sulfur capacity and selectivity are two key challenges for adsorption desulfurization technology of FCC gasoline. In order to improve the above issues, the mesopores were first introduced into ZSM-5 zeolites by a NaOH alkali treatment method to reduce the diffusion limitations and increase the sulfur capacity, in which the effects of the alkali concentration, treatment time, and temperature on the pore properties of zeolites were well discussed. The desulfurization performance was tested in a fixed-bed reactor using model fuel. The results demonstrated that the alkali treatment time was the most influential factor for the sulfur capacity. The mesoporous ZSM-5 adsorbents prepared under the conditions of 0.5 M NaOH, 150 min, and 75 °C showed the maximal sulfur capacity, the breakthrough and saturation sulfur capacities of which were 14.00 and 17.46 mg of S/g of adsorbent, respectively. In addition, Ni2+ ions were exchanged into mesoporous ZSM-5 zeolites to enhance the sulfur selectivity. The results proved that the selectivity of sulfur could be improved by an introduced metal ion because the Brönsted acid sites were decreased while the Lewis acid sites were increased. limitation.16 In the 1960s, Rosback and Neuzil17 claimed that the benzene adsorption capacity of mordenite significantly increased by alkali treatment because of better access to the micropores, which is the first reported in a U.S. patent. After that, Le Van Mao et al.18 conducted the first N2 adsorption− desorption isotherm of mesoporous ZSM-5 treated by alkali. However, the effects of mesopores in improving the diffusion and access to micropores were not discussed in his paper. Recently, Groen et al.19−22 designed a series of experiments to explore mesoporous ZSM-5 obtained by alkali treatment. They also gave the first direct experimental evidence to prove the improved diffusion in mesoporous ZSM-5 zeolites. Our group23−25 also did some research on alkali-treated ZSM-5 and found that the adsorption amount and adsorption rate of the probe molecule increased on the mesoporous-structured ZSM-5. Therefore, it was implied that mesoporous ZSM-5 should be a potential good supporter for ADS because of the improved diffusion and access to micropores. However, it is rarely reported that the mesoporous ZSM-5 zeolites were used as adsorbents for the ADS process. In addition, as for the selectivity, it was a common method that the metal or metal ions were exchanged or loaded onto the zeolites to improve the selectivity of sulfur compounds. Yang et

1. INTRODUCTION Nowadays, the sulfur-containing compounds in transportation fuels, such as gasoline and diesel, are severely limited by various district organizations and governments because of the environmental problems.1,2 Among the desulfurization technologies, the hydrodesulfurization (HDS) process, which is the main body of the desulfurization technologies, can effectively remove the sulfur-containing compounds. However, the loss of octane number will occur during HDS processes, and moreover it is also difficult for this process to realize ultradeep desulfurization of transportation fuels.3,4 Thus, some new processes are proposed, such as oxidative desulfurization, extractive desulfurization, and adsorption desulfurization (ADS). Among these technologies, the ADS process is regarded as a promising “zerosulfur” technology because of its mild conditions, good desulfurization performance, low cost, and minimal octane loss.5 In the research of ADS technology, development of the adsorbent occupies the highest priority. At present, various types of adsorbents, including zeolites,6−8 activated alumina,9 boron nitride,10−12 and activated carbon (AC),13−15 have been well studied. Because of the advantages of porous structure and modified surface, zeolites are considered to be suitable adsorbents for the ADS process. However, the sulfur capacity and selectivity are two obstacles for the ADS technology to achieve industrialization. To increase the sulfur capacity, it is an ongoing effort that a hierarchical architecture of porosity could be introduced in zeolites using the alkali treatment to overcome the diffusion © 2017 American Chemical Society

Received: Revised: Accepted: Published: 3813

December 29, 2016 February 27, 2017 March 23, 2017 March 23, 2017 DOI: 10.1021/acs.iecr.6b05053 Ind. Eng. Chem. Res. 2017, 56, 3813−3821

Article

Industrial & Engineering Chemistry Research al.26−29 prepared a series of nickel(II)-, copper(I)-, and silver(I)-modified zeolites, and the prepared adsorbents were utilized for desulfurization of fuel. The results showed that ultraclean transportation fuel was produced using the modified zeolites and nickel(II), copper(I), and silver(I) interacted with the sulfur-containing compounds by π complexation, which led to the improved selectivity of adsorbent for sulfur-containing compounds. Song et al.30,31 also studied the zeolites modified by nickel metal and cerium(IV), and they found that the adsorbents also had an excellent desulfurization performance. However, different from π complexation, they thought that a direct S−M (S−Ni or S−CeIV) interaction was formed between nickel or cerium(IV) and the sulfur atom, which showed higher sulfur selectivity than π complexation. Furthermore, a dualmetal-modified adsorbent was an approach to further improve the desulfurization performance. CuII−CeIV- and AgI−CeIVmodified zeolites were prepared by Song and co-workers32−35 and utilized for desulfurization of fuel. The prepared adsorbents not only showed higher selectivity like CeIV-modified zeolites but also had a greater sulfur capacity like CuII or AgI-modified zeolites. The function of nickel species in improving the selectivity for thiophene was also reported in our previous research.36,37 Thus, improving the sulfur capacity and selectivity is the main task for this paper. The sulfur capacity was enhanced by introducing mesopores into ZSM-5 by alkali treatment. The selectivity of the adsorbent was improved by introducing nickel species. The relationship between the alkali-treated conditions and textural features was discussed, and the sulfur capacities of the prepared mesoporous ZSM-5 zeolites under different alkali treatment conditions were evaluated in the fixed-bed reactor. In addition, the selectivity of the adsorbent for thiophene was evaluated using the model fuel containing 2 vol % toluene. The role of nickel species to improve the selectivity for thiophene was discussed in this paper.

Table 1. Synthesis Conditions of Mesoporous ZSM-5 Samples

2. EXPERIMENT 2.1. Adsorbent Preparation. 2.1.1. Alkali Treatment of HZSM-5. HZSM-5 zeolites with a SiO2/Al2O3 rate of 25 were purchased from the Catalyst Plant of Nankai University. The parent HZSM-5 zeolites were treated with a NaOH solution (about 10 mL/g of zeolites). After the alkali treatment, the samples were washed with deionized water until the washing water was neutral. Then the samples were placed in an oven at 120 °C overnight followed by calcining at 540 °C in air for 4 h. Thus, mesoporous ZSM-5 samples were obtained. The effects of the alkali concentration, alkali treatment time, and temperature were discussed in this paper. The synthesized samples using HZSM-5 zeolites were denoted as ATZ5-1, -2, ..., -11. The synthesis conditions of each sample are shown in Table 1. 2.1.2. Ion Exchange of the Sample. The alkali-treated samples that showed the largest sulfur capacity were selected as the parent zeolites for ion exchange. The parent zeolites were mixed with a 1 M NH4NO3 solution (about 10 mL/g of zeolites). Then the mixture was stirred at 90 °C for 4 h. After ion exchange, the samples were washed with deionized water and then placed in an oven at 120 °C overnight followed by calcining at 540 °C in air for 4 h. To achieve a higher exchange, the ion exchange was repeated twice. Thus, the obtained samples were denoted as HATZ5. The preparation method of NiATZ5 was similar to that of HATZ5 except for the ionexchange conditions of 0.1 M Ni(NO3)2·6H2O at 75 °C.

2.3. Characterization of the Sample. Power X-ray diffraction (XRD) patterns of the samples were collected at a 2θ range of 5−50° on a Bruker D8 advance-X-ray diffractometer using Cu Kα radiation under the setting conditions of 40 kV and 30 mA. The surface morphology of the samples was obtained on a Quanta 200F scanning electron microscope. The N2 adsorption−desorption isotherms of the samples were obtained on a Micromeritics ASAP 2020. Prior to the adsorption of N2, the samples were degassed at 300 °C for 3 h. The total surface areas of the samples were determined by the Brunauer−Emmett−Teller (BET) method. The micropore volumes and surface areas of the samples were obtained by the t-plot method. The Harkins−Jura thickness curve equation was used in the t-plot method. The pore-size distributions of the samples were obtained by the Barrett−Joyner−Halenda (BJH) model applied to the adsorption branch of the isotherm. The acidic property of the sample was measured by Py-FTIR [Fourier transform infrared (FTIR) analysis of adsorbed pyridine] characterization. The sample was placed in a quartz IR cell with CaF2 windows. It was purged at 400 °C for 1 h and cooled to 30 °C in a vacuum. Then a spectrum was recorded as the background. After the valve between the pump system and cell was closed, pyridine vapor was added for adsorption. Then the sample was vacuumed at 200 °C for 20 min and cooled to 30 °C, and the spectrum was recorded to obtain the amount of total acid sites. Finally, the sample was vacuumed at 350 °C for

synthesis conditions sample

concentration (M)

time (min)

temperature (°C)

ATZ5-1 ATZ5-2 ATZ5-3 ATZ5-4 ATZ5-5 ATZ5-6 ATZ5-7 ATZ5-8 ATZ5-9 ATZ5-10 ATZ5-11

0.05 0.1 0.2 0.5 1 0.5 0.5 0.5 0.5 0.5 0.5

90 90 90 90 90 60 120 150 300 150 150

75 75 75 75 75 75 75 75 75 65 85

2.2. Reagents. The desulfurization performance of the adsorbent was evaluated by a model fuel. One type of model fuel with a sulfur concentration of 100 mg/L was prepared by adding thiophene into a cyclohexane solvent and denoted as model fuel MG-1. To investigate the effect of aromatics for ADS, another type of model fuel containing 60 mg of S/L and 2.0 vol % toluene was prepared and denoted as model fuel MG2. The detailed compositions of the model fuels are listed in Table 2. Table 2. Compositions of the Model Fuels no. MG-1 MG-2

3814

component thiophene/cyclohexane thiophene + toluene/ cyclohexane

toluene content (vol %)

sulfur content (mg/L)

0.0 2.0

100 60

DOI: 10.1021/acs.iecr.6b05053 Ind. Eng. Chem. Res. 2017, 56, 3813−3821

Article

Industrial & Engineering Chemistry Research 20 min and cooled to 30 °C, and the spectrum was recorded to assess the amount of the strong acid sites. 2.4. Fixed-Bed Adsorption Experiment. The ADS performance of the samples was carried out in a fixed-bed flow reactor with a diameter of 1 cm and a length of 40 cm. Before desulfurization evaluation, the packed adsorbent was heated at 360 °C under a N2 flow of 50 mL/min for 2 h in order to remove the water in the pore and then cooled to 40 °C. The model fuel was pumped into the fixed-bed flow reactor at 40 °C with a flow rate of 0.4 mL/min. The effluent was collected at regular intervals, and the sulfur content of the effluent was measured by a TCS-2000S ultraviolet fluorescence sulfur analyzer. When the sulfur content of the effluent was 10 mg/L, breakthrough was achieved. It was considered to have reached saturation when the sulfur content of the effluent was 90% of the model fuels. The breakthrough and saturation sulfur capacities were calculated, and the detailed calculation formula is sulfur capacity (mg of S/g of adsorbent) =

(C 0 − C )V m

where C0 and C are the initial and effluent sulfur concentrations (mg/L), respectively, V is the effluent volume (L), and m is the adsorbent mass (g).

3. RESULTS AND DISCUSSION 3.1. Relationship between the Textural Features and Alkali-Treated Conditions. The crystal structure of alkalitreated HZSM-5 was studied by an XRD characterization method, and the results are shown in Figure 1. Except for the ATZ5-5 sample, the characteristic diffraction patterns of other alkali-treated HZSM-5 samples were similar to that of the parent HZSM-5 zeolite, which indicated that the MFI framework of alkali-treated HZSM-5 samples was retained.38 In addition, it was found that the peak intensities of alkalitreated HZSM-5 samples decreased with increasing alkali concentration, alkali treatment time, or temperature, especially in the alkali-treated conditions of 1 M NaOH, 90 min, and 75 °C (ATZ5-5), suggesting that the crystal structures of alkalitreated HZSM-5 samples were more deeply damaged with more rigorous treatment. Scanning electron microscopy (SEM) pictures of alkalitreated HZSM-5 are shown in Figure 2. The parent HZSM-5 zeolite particle size was in the range of 1−5 μm, and the zeolite showed a smooth surface. With an increase of the alkali concentration, alkali treatment time, or temperature, a rougher surface and deeper destruction were present in the SEM pictures of alkali-treated HZSM-5 samples, especially the ATZ5-5 sample. The structure of ATZ5-5 was under severe damage. However, the MFI structure of the other alkali-treated HZSM-5 was retained. These results were in accordance with XRD characterizations. The N2 adsorption−desorption isotherms were carried out to obtain the changes of the textural properties after the alkali treatment. As shown in Figure 3 (left), the N2 adsorption− desorption isotherm of the parent HZSM-5 sample exhibited type I with a rapid uptake increase at low relative pressure, followed by a horizontal adsorption, which is typical of microporous materials.39 However, the N2 adsorption− desorption isotherms of alkali-treated HZSM-5 samples showed both types I and IV with an enhanced uptake at intermediate pressure, indicating that hierarchical porous systems combining

Figure 1. XRD patterns of alkali-treated HZSM-5 samples.

micropores and mesopores were formed after the alkali treatment.40,41 With an increase of the alkali concentration, the N2-saturated adsorption amount increase from 126 cm3/g for the parent HZSM-5 to 216 cm3/g for ATZ5-5, accompanied by a more and more obvious hysteresis loop, indicating that the total pore volume of the sample was enhanced and more mesopores were introduced into the zeolites after the alkali treatment. As for the effects of the treatment time and temperature, similar conclusions could be drawn. The pore-size distributions of alkali-treated HZSM-5 samples are shown in Figure 3 (right). Obviously, the pores of the parent HZSM-5 were mainly micropores of less than 2 nm. As for alkali-treated HZSM-5 samples, in addition to the micropores, a wide distribution in the range of 2−40 nm can be found, which also indicated that mesopores were formed after the alkali treatment. The effect of the alkali concentration in the range of 0.05−1 M was studied. No remarkable mesopores were discovered in the pore-size distributions of ATZ5-1, indicating that the parent HZSM-5 sample cannot effectively desiliconize to introduce mesopores under the alkali treatment of a 0.05 M NaOH solution. However, different results can be found for the ATZ5-4 and ATZ5-5 samples with 3815

DOI: 10.1021/acs.iecr.6b05053 Ind. Eng. Chem. Res. 2017, 56, 3813−3821

Article

Industrial & Engineering Chemistry Research

with harsher alkali treatment conditions. The alkali concentration was the most important factor for textural features. Compared with the parent HZSM-5 zeolites, the total specific surface area of ATZ5-8 only decreased by 7% and the MFI structure was retained, while the mesoporous volume increased by 333%. Therefore, the structure of the ATZ5-8 sample was considered optimal, and the alkali treatment conditions of that sample (0.5 M, 150 min, and 75 °C) were the most effective. 3.2. Desulfurization Performance of Alkali-Treated HZSM-5. In order to examine the desulfurization performance of alkali-treated HZSM-5, several experiments were carried out with model fuel MG-1 (100 mg of S/L), which was prepared by adding thiophene into a cyclohexane solvent. When the effluent sulfur contents were 10 and 90 mg of S/L, it was considered to reach breakthrough and saturation, respectively. Figure 4 showed the breakthrough curves over alkali-treated HZSM-5, γ-Al2O3, and AC samples. The breakthrough and saturation sulfur capacities of alkali-treated HZSM-5 samples were calculated and summarized in Table S1. For the effect of the alkali concentration, a series of alkalitreated samples were prepared by different alkali concentrations in the range of 0.05−1 M, and the sulfur capacities of the samples were studied. From Figure 4, it can be intuitively observed that the desulfurization performances of the samples were first improved and reduced with an increase of the alkali concentration. When the alkali concentration was at 0.5 M, the ATZ5-4 sample showed the best desulfurization performance. Detailed data of the breakthrough and saturation sulfur capacities are shown in Table S1. At the range of 0.05−1 M alkali concentrations, ATZ5-4 showed the largest breakthrough and saturation sulfur capacities, which were 5.76 and 7.12 mg of S/g of adsorbent, respectively. The breakthrough sulfur capacity of ATZ5-1 and the saturation sulfur capacity of ATZ5-5 were the smallest (3.35 and 4.51 mg of S/g of adsorbent, respectively). Compared with the ATZ5-1 and ATZ5-5 samples, the breakthrough and saturation sulfur capacities of ATZ5-4 were separately improved by 72% and 58%, respectively. For the effect of the alkali treatment time, the samples with different treatment times in the range of 60−300 min were evaluated in the fixed-bed reactor. As shown in Figure 4, it can be obviously observed that the ATZ5-8 sample showed the best performance and the samples with treatment times in the range of 60−120 min showed poor desulfurization capacity. Table S1 also showed the breakthrough and saturation sulfur capacities of the samples. It can be found that the largest and smallest breakthrough sulfur capacities of the samples were 14.00 and 4.92 mg of S/g of adsorbent, respectively. At the same time, the largest and smallest saturation sulfur capacities of the samples were 17.46 and 7.12 mg of S/g of adsorbent, respectively. Compared with the sample with poor desulfurization capacity, the breakthrough and saturation sulfur capacities of ATZ5-8 were improved by 184% and 145%, respectively. For the effect of the treatment temperature, three temperature points (65, 75, and 85 °C) were discussed in the study. From Figure 4, ATZ5-8 showed the best desulfurization capacity and the other two samples had similar performances. The breakthrough and saturation sulfur capacities of ATZ5-10 were 11.68 and 14.56 mg of S/g of adsorbent, respectively (Table S1), which was the worst desulfurization performance among the three samples. Compared with ATZ5-10, both the breakthrough and saturation sulfur capacities of ATZ5-8 were improved by 20%.

Figure 2. SEM pictures of alkali-treated HZSM-5 samples.

mesopore-size distributions. The mesopore-size distribution of ATZ5-4 was narrow and two peaks were centered at about 8 and 23 nm, while a wide mesopore-size distribution and a single peak centered near 20 nm were present in the ATZ5-5 sample. In terms of the alkali treatment time and temperature, the poresize distributions of alkali-treated samples were similar to that of ATZ5-4, with narrow mesopore-size distribution and two peaks at about 8 and 23 nm. The detailed textural property information on the parent HZSM-5 and alkali-treated samples is shown in Table 3. The parent HZSM-5 showed a higher total specific surface area of 326 m2/g and a lower total pore volume of 0.18 cm3/g. After the alkali treatment, the total specific surface area was reduced but the total pore volume was developed, especially for the micropore specific surface area and mesopore volume. The effects of the alkali concentration were discussed by the detailed textural properties of ATZ5-1, -4, and -5 zeolites. Compared to the parent HZSM-5 zeolites, the total specific surface areas of ATZ5-1, -4, and -5 separately decreased by about 6%, 12%, and 39%, while the mesopore volumes of those increased by 0%, 367%, and 800%, respectively. In addition, the ATZ5-4, -8, -9, -10, and -11 samples were selected to study the effects of the treatment time and temperature. With respect to the parent HZSM-5, the total specific surface areas of ATZ5-4, -8, -9, -10, and -11 reduced by 12%, 7%, 13%, 11%, and 12%, respectively. Meanwhile, the mesopore volumes of those samples separately enlarged by 367%, 333%, 367%, 267%, and 400%, respectively. As shown above, the main effect rules of the alkali concentration, alkali treatment time, and temperature on the textural features were consistent with those of the previous literature.24,42 More mesopores were introduced into the zeolites, and the micropore specific surface area decreased 3816

DOI: 10.1021/acs.iecr.6b05053 Ind. Eng. Chem. Res. 2017, 56, 3813−3821

Article

Industrial & Engineering Chemistry Research

Figure 3. N2 adsorption−desorption isotherms (left) and BJH adsorption pore-size distributions (right) of alkali-treated HZSM-5 samples.

Table 3. Textural Properties of Alkali-Treated Samples surface area (m2/g)

a

a

sample

total

HZSM-5 ATZ5-1 ATZ5-4 ATZ5-5 ATZ5-8 ATZ5-9 ATZ5-10 ATZ5-11

326 308 286 198 302 283 291 288

micropore

b

pore volume (cm3/g) mesopore

total

microporeb

mesopore

10 13 65 76 66 65 48 58

0.18 0.17 0.25 0.33 0.24 0.25 0.23 0.26

0.15 0.14 0.11 0.06 0.11 0.11 0.12 0.11

0.03 0.03 0.14 0.27 0.13 0.14 0.11 0.15

316 295 221 122 236 218 243 230

BET method. bt-plot method.

On the basis of the above discussions, it could be drawn that the desulfurization capacities of the samples all first increase and then decrease with the harsher treatment conditions and the treatment time was the most influential factor for desulfurization of alkali-treated HZSM-5 samples among the three factors. Alkali-treated samples under mild alkali-treated conditions (such as ATZ5-1) could retain the higher specific surface, while mesopores were not effectively introduced to reduce the diffusion limitation. Thus, these samples showed a poor desulfurization performance. On the other hand, although

more mesopores were introduced into the samples with rough alkali-treated conditions (such as ATZ5-5), the MFI structure of the samples was deeply damaged and the specific surface decreased significantly. Thus, the desulfurization performance of the samples was also poor. In summary, the sample could have a good desulfurization performance under suitable alkalitreated conditions. Among the alkali-treated samples, the ATZ5-8 sample with a higher specific surface and more mesopores showed the largest sulfur capacity. Lee and Valla16 mixed 100 ppmw of thiophene in octane and utilized it to 3817

DOI: 10.1021/acs.iecr.6b05053 Ind. Eng. Chem. Res. 2017, 56, 3813−3821

Article

Industrial & Engineering Chemistry Research

Figure 4. Breakthrough of thiophene from MG-1 over alkali-treated HZSM-5, γ-Al2O3, and AC samples, at 40 °C with a flow rate of 0.4 mL/min, based on 1 g of adsorbents.

evaluate the desulfurization performance of mesoporous Y zeolites. Compared with the research of Lee and Valla, the desulfurization performance of ATZ5-8 was also improved. In addition, conventional materials (γ-Al2O3 and AC) were chosen as comparisons to show the good desulfurization of the ATZ5-8 sample. From Figure 4, the γ-Al2O3 sample showed a poor desulfurization capacity and the sulfur content of the first effluent sample was 73 mg of S/L. Compared with the γ-Al2O3 sample, the desulfurization performance of AC was enhanced to some degree. However, the AC could not still obtain effluent samples with sulfur contents of less than 10 mg of S/L. These results indicated that ATZ5-8 had a good advantage in ADS. 3.3. Improved Selectivity over the NiATZ5 Adsorbent. It is well-known that coexisting aromatics in gasoline leads to serious competition adsorption with thiophene compounds and results in a great decrease in the selectivity of sulfides. In order to improve the sulfur selectivity of adsorbents, the NiATZ5 adsorbent was prepared by exchanging Ni2+ over the ATZ5-8 zeolite, which showed the best hierarchical architecture of porosity. The XRD patterns of parent HZSM-5 and ion-exchanged samples (HATZ5 and NiATZ5) are shown in Figure 5. The XRD patterns of ion-exchanged samples were similar to those of parent HZSM-5, which indicated that the original structure of ion-exchanged samples was retained. The characteristic peak intensity of ion-exchanged samples decreased slightly. Those results showed that the modified process had little effect on the structure of HZSM-5 except a slightly decreased crystallinity. In addition, no diffraction peaks related to Ni2+ or NiO can be observed in the XRD patterns of NiATZ5, indicating that the nickel species were well dispersed on the surfaces of the zeolites. The micromorphology of ion-exchanged samples was characterized by the SEM method. SEM pictures are shown in Figure 6. The results were similar to the SEM character-

Figure 5. XRD patterns of ion-exchanged samples.

Figure 6. SEM pictures of ion-exchanged samples.

izations above in Figure 2. Although the crystal structure of ionexchanged samples was damaged to a certain degree, the MFI structure of the samples was retained. The acid amounts and strengths of ion-exchanged samples were obtained by Py-FTIR analysis according to a previous literature method.43 Generally, the bands at about 1540 and 1450 cm−1 can be assigned to the Brönsted and Lewis acid sites, respectively. The acid amounts recorded at 200 and 350 °C were representative of weak and strong acid sites, respectively. 3818

DOI: 10.1021/acs.iecr.6b05053 Ind. Eng. Chem. Res. 2017, 56, 3813−3821

Article

Industrial & Engineering Chemistry Research

and the effluent sulfur content of the first sample (22 mg of S/ L) exceeded the breakthrough point. As for the NiATZ5 adsorbents, the desulfurization capacity was improved, four effluent samples (less than 10 mg of S/L) were obtained, and the breakthrough capacity of this adsorbent was 0.46 mg of S/g of adsorbent. When these results were combined with our previous research,36,37 it could be drawn that HATZ5 with a hierarchical architecture of porosity could have a large sulfur capacity. However, its desulfurization performance would obviously reduce because of the competitive adsorption of toluene. Compared with HATZ5 zeolites, although the pore volume and specific surface area of NiATZ5 declined because of the introduction of Ni2+, the sulfur selectivity was improved. Moreover, more Brönsted acid sites existed on the surface of the HATZ5 adsorbent, which was harmful for desulfurization due to protonation and oligomerization reactions of thiophene on the Brönsted acid sites, while more Lewis acid sites (especially strong Lewis acid sites) were found on the surface of the NiATZ5 adsorbent, which was beneficial for desulfurization due to complexation between thiophene and Lewis acid sites. As shown above, although the sulfur capacity of the adsorbent was increased by introducing mesopores into the zeolites, the sulfur selectivity was not improved and the desulfurization performance was seriously affected by competitive adsorption. The sulfur selectivity of the adsorbent could be enhanced by modification of Ni2+. However, a lot of work was needed to further improve the selectivity of the adsorbent for sulfur compounds.

The FTIR spectrum of ion-exchanged samples adsorbing pyridine are shown in Figure S1, and the calculated acid amounts are summarized in Table 4. From Figure S1 and Table Table 4. Acid Amounts from FTIR Spectra of Pyridine Adsorption over Ion-Exchanged Samplesa Lewis (μmol/g)

Brönsted (μmol/g)

adsorbent

TL

SL

WL

TB

SB

WB

ATZ5-8 HATZ5 NiATZ5

382 99 364

188 66 267

194 33 97

6 205 67

2 129 25

4 76 42

a

TL = total L acid; SL = strong L acid; WL = weak L acid; TB = total B acid; SB = strong B acid; WB = weak B acid.

4, the primary Lewis and little Brönsted acid sites were found on the surface of ATZ5-8. After ion exchange, it was clear that the Brönsted acid site amounts of ion-exchanged samples were increased, especially that for HATZ5. The calculated Brönsted acid site amounts were about twice as much as the Lewis acid site amounts on the surface of HATZ5. As for the NiATZ5 sample, the number of Lewis acid sites was still much higher than that of Brönsted acid sites on the surface of NiATZ5. It can be concluded that the Lewis acid site amounts of NiATZ5 were maintained mainly because of the introduction of Ni2+, which is consistent with the research of Gong et al.44 They explained that the metal ions were the acceptors of electrons; thus, higher Lewis acid sites were formed on the surface of NiATZ5. It was well-known that gasoline contained a certain quantity of aromatics and the desulfurization performance of the adsorbent was greatly affected by the competitive adsorption of aromatics in previous literature.45,46 Therefore, in order to investigate the effect of aromatics for the desulfurization performance of ion-exchanged samples, some experiments were carried out using model fuel MG-2 containing 2 vol % toluene and 60 mg of S/L. When the effluent sulfur content was 10 mg of S/L, it was considered to reach breakthrough. Figure 7 shows the breakthrough curves over ion-exchanged samples.

4. CONCLUSION To meet the sulfur capacity and selectivity challenges, the alkali treatment of HZSM-5 was used to reduce the micropore diffusion and increase the sulfur capacity. Moreover, Ni2+ was supported over the zeolites to improve the sulfur selectivity. The relationship between the textural properties and alkali treatment conditions was discussed in this paper. The optimal alkali treatment conditions were 0.5 M NaOH, 150 min, and 75 °C. At these conditions, the total specific surface of ATZ5-8 only decreased by about 7%; however, its mesopore volume increased by 333%. Meanwhile, the ATZ5-8 sample showed the largest sulfur capacity among alkali-treated samples, whose breakthrough and saturation sulfur capacities were 14.00 and 17.46 mg of S/g of adsorbent, respectively. In addition, for model fuel MG-2 containing 2 vol % toluene, NiATZ5 had the best desulfurization performance and showed higher selectivity for thiophene. In summary, the sulfur capacity can be improved by introducing mesopores into the zeolites; however, this was the key to further improving the selectivity of the adsorbents for sulfur compounds, which will be the focus of our future research.

Figure 7. Breakthrough of thiophene from MG-2 over ion-exchanged samples, at 40 °C with a flow rate of 0.4 mL/min, based on 1 g of adsorbents.



From Figure 7, when the ATZ5-8 sample was evaluated in the fixed-bed reactor with model fuel MG-2, the sulfur contents of two effluent samples could be less than 10 mg of S/L. Compared to the ATZ5-8 sample, HATZ5 and NiATZ5 showed different desulfurization performances. The desulfurization performance of HATZ5 was worse than that of ATZ5-8,

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.iecr.6b05053. Breakthrough and saturation sulfur capacities from MG-1 over alkali-treated HZSM-5 samples (Table S1) and FTIR spectra of ion-exchanged samples adsorbing pyridine at 200 and 350 °C (Figure S1) (PDF)

ASSOCIATED CONTENT

S Supporting Information *

3819

DOI: 10.1021/acs.iecr.6b05053 Ind. Eng. Chem. Res. 2017, 56, 3813−3821

Article

Industrial & Engineering Chemistry Research



(16) Lee, K. X.; Valla, J. A. Investigation of metal-exchanged mesoporous Y zeolites for the adsorptive desulfurization of liquid fuels. Appl. Catal., B 2017, 201, 359. (17) Rosback, D. H.; Neuzil, R. W. Method for manufacturing an adsorbent useful for olefin separation. U.S. Patent 4,048,111, Sept 13, 1977. (18) Le Van Mao, R.; Ramsaran, A.; Xiao, S.; Yao, J.; Semmer, V. pH of the sodium carbonate solution used for the desilication of zeolite materials. J. Mater. Chem. 1995, 5, 533. (19) Groen, J. C.; Zhu, W.; Brouwer, S.; Huynink, S. J.; Kapteijn, F.; Moulijn, J. A.; Pérez-Ramírez, J. Direct demonstration of enhanced diffusion in mesoporous ZSM-5 zeolite obtained via controlled desilication. J. Am. Chem. Soc. 2007, 129, 355. (20) Groen, J. C.; Jansen, J. C.; Moulijn, J. A.; Perez-Ramirez, J. Optimal aluminum-assisted mesoporosity development in MFI zeolites by desilication. J. Phys. Chem. B 2004, 108, 13062. (21) Groen, J. C.; Peffer, L.; Moulijn, J. A.; Perez-Ramirez, J. Mesoporosity development in ZSM-5 zeolite upon optimized desilication conditions in alkaline medium. Colloids Surf., A 2004, 241, 53. (22) Groen, J. C.; Peffer, L.; Moulijn, J. A.; Perez-Ramirez, J. On the introduction of intracrystalline mesoporosity in zeolites upon desilication in alkaline medium. Microporous Mesoporous Mater. 2004, 69, 29. (23) Zhao, L.; Gao, J.; Xu, C.; Shen, B. Alkali-treatment of ZSM-5 zeolites with different SiO2/Al2O3 ratios and light olefin production by heavy oil cracking. Fuel Process. Technol. 2011, 92, 414. (24) Zhao, L.; Xu, C.; Gao, S.; Shen, B. Effects of concentration on the alkali-treatment of ZSM-5 zeolite: a study on dividing points. J. Mater. Sci. 2010, 45, 5406. (25) Zhao, L.; Shen, B.; Gao, J.; Xu, C. Investigation on the mechanism of diffusion in mesopore structured ZSM-5 and improved heavy oil conversion. J. Catal. 2008, 258, 228. (26) Hernández-Maldonado, A. J.; Yang, F. H.; Qi, G.; Yang, R. T. Desulfurization of transportation fuels by π-complexation sorbents: Cu(I)-, Ni(II)-, and Zn(II)-zeolites. Appl. Catal., B 2005, 56, 111. (27) Hernández-Maldonado, A. J.; Yang, R. T. Desulfurization of diesel fuels by adsorption via π-complexation with vapor-phase exchanged Cu(I)-Y zeolites. J. Am. Chem. Soc. 2004, 126, 992. (28) Hernández-Maldonado, A. J.; Yang, R. T. Desulfurization of diesel fuels via π-complexation with nickel(II)-exchanged X- and Yzeolites. Ind. Eng. Chem. Res. 2004, 43, 1081. (29) Hernández-Maldonado, A. J.; Yang, R. T. Desulfurization of liquid fuels by adsorption via π complexation with Cu(I)-Y and Ag-Y Zeolites. Ind. Eng. Chem. Res. 2003, 42, 123. (30) Velu, S.; Ma, X.; Song, C. Selective adsorption for removing sulfur from jet fuel over zeolite-based adsorbents. Ind. Eng. Chem. Res. 2003, 42, 5293. (31) Velu, S.; Song, C.; Engelhard, M. H.; Chin, Y. Adsorptive removal of organic sulfur compounds from jet fuel over K-exchanged NiY zeolites prepared by impregnation and ion exchange. Ind. Eng. Chem. Res. 2005, 44, 5740. (32) Song, H.; Chang, Y.; Song, H. Deep adsorptive desulfurization over Cu, Ce bimetal ion-exchanged Y-typed molecule sieve. Adsorption 2016, 22, 139. (33) Song, H.; Gao, H.; Song, H.; Yang, G.; Li, X. Effects of Si/Al ratio on adsorptive removal of thiophene and benzothiophene over ion-exchanged AgCeY zeolites. Ind. Eng. Chem. Res. 2016, 55, 3813. (34) Song, H.; Wan, X.; Dai, M.; Zhang, J.; Li, F.; Song, H. Deep desulfurization of model gasoline by selective adsorption over Cu-Ce bimetal ion-exchanged Y zeolite. Fuel Process. Technol. 2013, 116, 52. (35) Song, H.; Cui, X.; Song, H.; Gao, H.; Li, F. Characteristic and adsorption desulfurization performance of Ag-Ce bimetal ionexchanged Y zeolite. Ind. Eng. Chem. Res. 2014, 53, 14552. (36) Li, H.; Han, X.; Huang, H.; Wang, Y.; Zhao, L.; Cao, L.; Shen, B.; Gao, J.; Xu, C. Competitive adsorption desulfurization performance over K - doped NiY zeolite. J. Colloid Interface Sci. 2016, 483, 102. (37) Han, X.; Li, H.; Huang, H.; Zhao, L.; Cao, L.; Wang, Y.; Gao, J.; Xu, C. Effect of olefin and aromatics on thiophene adsorption

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: 86-10-89739078. Fax: 8610-69724721. ORCID

Liang Zhao: 0000-0003-3585-6020 Author Contributions †

These authors contributed equally.

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was supported by the National Natural Science Foundation of China (Grants 21336011, 21476260, 21236009, and U1162204) and the Science Foundation of China University of Petroleum, Beijing (Grants 2462015YQ0316 and 2462015YQ0311).



REFERENCES

(1) Song, C. An overview of new approaches to deep desulfurization for ultra-clean gasoline, diesel fuel and jet fuel. Catal. Today 2003, 86, 211. (2) Breysse, M.; Djega-Mariadassou, G.; Pessayre, S.; Geantet, C.; Vrinat, M.; Pérot, G.; Lemaire, M. Deep desulfurization: reactions, catalysts and technological challenges. Catal. Today 2003, 84, 129. (3) Ma, X.; Sun, L.; Song, C. A new approach to deep desulfurization of gasoline, diesel fuel and jet fuel by selective adsorption for ultraclean fuels and for fuel cell applications. Catal. Today 2002, 77, 107. (4) Yang, R. T.; Hernandez-Maldonado, A. J.; Yang, F. H. Desulfurization of transportation fuels with zeolites under ambient conditions. Science 2003, 301, 79. (5) Hernández Maldonado, A. J.; Yang, R. T. Desulfurization of transportation fuels by adsorption. Catal. Rev.: Sci. Eng. 2004, 46, 111. (6) Bakhtiari, G.; Abdouss, M.; Bazmi, M.; Royaee, S. J. Optimization of sulfur adsorption over Ag-zeolite nanoadsorbent by experimental design method. Int. J. Environ. Sci. Technol. 2016, 13, 803. (7) Tian, F.; Shen, Q.; Fu, Z.; Wu, Y.; Jia, C. Enhanced adsorption desulfurization performance over hierarchically structured zeolite Y. Fuel Process. Technol. 2014, 128, 176. (8) Salem, A. B. S. H. Naphtha desulfurization by adsorption. Ind. Eng. Chem. Res. 1994, 33, 336. (9) Srivastav, A.; Srivastava, V. C. Adsorptive desulfurization by activated alumina. J. Hazard. Mater. 2009, 170, 1133. (10) Xiong, J.; Yang, L.; Chao, Y.; Pang, J.; Wu, P.; Zhang, M.; Zhu, W.; Li, H. A large number of low coordinated atoms in boron nitride for outstanding adsorptive desulfurization performance. Green Chem. 2016, 18, 3040. (11) Xiong, J.; Zhu, W.; Li, H.; Ding, W.; Chao, Y.; Wu, P.; Xun, S.; Zhang, M.; Li, H. Few-layered graphene-like boron nitride induced a remarkable adsorption capacity for dibenzothiophene in fuels. Green Chem. 2015, 17, 1647. (12) Wu, P.; Zhu, W.; Chao, Y.; Zhang, J.; Zhang, P.; Zhu, H.; Li, C.; Chen, Z.; Li, H.; Dai, S. A template-free solvent-mediated synthesis of high surface area boron nitride nanosheets for aerobic oxidative desulfurization. Chem. Commun. 2016, 52, 144. (13) Liu, X.; Guo, J.; Chu, Y.; Luo, D.; Yin, H.; Sun, M.; Yavuz, R. Desulfurization performance of iron supported on activated carbon. Fuel 2014, 123, 93. (14) Triantafyllidis, K. S.; Deliyanni, E. A. Desulfurization of diesel fuels: Adsorption of 4,6-DMDBT on different origin and surface chemistry nanoporous activated carbons. Chem. Eng. J. 2014, 236, 406. (15) Farzin Nejad, N.; Shams, E.; Amini, M. K.; Bennett, J. C. Synthesis of magnetic mesoporous carbon and its application for adsorption of dibenzothiophene. Fuel Process. Technol. 2013, 106, 376. 3820

DOI: 10.1021/acs.iecr.6b05053 Ind. Eng. Chem. Res. 2017, 56, 3813−3821

Article

Industrial & Engineering Chemistry Research desulfurization over modified NiY zeolites by metal Pd. RSC Adv. 2016, 6, 75006. (38) Ahmadpour, J.; Taghizadeh, M. Selective production of propylene from methanol over high-silica mesoporous ZSM-5 zeolites treated with NaOH and NaOH/tetrapropylammonium hydroxide. C. R. Chim. 2015, 18, 834. (39) Shi, Y.; Zhang, W.; Zhang, H.; Tian, F.; Jia, C.; Chen, Y. Effect of cyclohexene on thiophene adsorption over NaY and LaNaY zeolites. Fuel Process. Technol. 2013, 110, 24. (40) Wei, X. T.; Smirniotis, P. G. Development and characterization of mesoporosity in ZSM-12 by desilication. Microporous Mesoporous Mater. 2006, 97, 97. (41) Plana-Pallejà, J.; Abelló, S.; Berrueco, C.; Montané, D. Effect of zeolite acidity and mesoporosity on the activity of Fischer−Tropsch Fe/ZSM-5 bifunctional catalysts. Appl. Catal., A 2016, 515, 126. (42) Groen, J. C.; Moulijn, J. A.; Perez-Ramirez, J. Desilication: on the controlled generation of mesoporosity in MFI zeolites. J. Mater. Chem. 2006, 16, 2121. (43) Emeis, C. A. Determination of integrated molar extinction coefficients for infrared absorption bands of pyridine adsorbed on solid acid catalysts. J. Catal. 1993, 141, 347. (44) Gong, Y.; Dou, T.; Kang, S.; Li, Q.; Hu, Y. Deep desulfurization of gasoline using ion-exchange zeolites: Cu(I)- and Ag(I)-beta. Fuel Process. Technol. 2009, 90, 122. (45) Shi, Y.; Yang, X.; Tian, F.; Jia, C.; Chen, Y. Effects of toluene on thiophene adsorption over NaY and Ce(IV)Y zeolites. J. Nat. Gas Chem. 2012, 21, 421. (46) Laborde-Boutet, C.; Joly, G.; Nicolaos, A.; Thomas, M.; Magnoux, P. Selectivity of thiophene/toluene competitive adsorptions onto NaY and NaX zeolites. Ind. Eng. Chem. Res. 2006, 45, 6758.

3821

DOI: 10.1021/acs.iecr.6b05053 Ind. Eng. Chem. Res. 2017, 56, 3813−3821