- Email: [email protected]
Synthesis and application of different phthalocyanine molecular sieve catalyst for oxidative desulfurization
Journal of Solid State Chemistry 225 (2015) 347–353
Contents lists available at ScienceDirect
Journal of Solid State Chemistry journal homepage: www.elsevier.com/locate/jssc
Synthesis and application of different phthalocyanine molecular sieve catalyst for oxidative desulfurization Na Zhao a,c,1, Siwen Li a,c,1, Jinyi Wang a,c, Ronglan Zhang a,c, Ruimin Gao b,c, Jianshe Zhao a,c,n, Junlong Wang b,c a Key Laboratory of Synthetic and Natural Functional Molecule Chemistry of Ministry of Education, Shaanxi Key Laboratory of Physico-Inorganic Chemistry, College of Chemistry & Material Science, Northwest University, Xi’an 710069, Shaanxi, China b Research Institute of Shaanxi Yanchang Petroleum Group Corp. Ltd., Xi’an 710075, China c Composites Research Institute, Weinan Normal University, Weinan 714000, China
art ic l e i nf o
a b s t r a c t
Article history: Received 29 October 2014 Received in revised form 4 January 2015 Accepted 11 January 2015 Available online 23 January 2015
M2(PcAN)2 (M¼ Fe, Co, Ni, Cu, Zn and Mn) anchored onto W-HZSM-5 (M2(PcAN)2–W-HZSM-5) or the M2(PcTN)2 doping W-HZSM-5 (M2(PcTN)2/W-HZSM-5) were prepared and their catalytic performances were tested for oxidative desulfurization in the presence of oxygen. Thiophene (T), benzothiophene (BT), and dibenzothiophene (DBT) were considered as sulfur compounds. Among zeolite-based catalysts, the Cu2(PcAN)2–W-HZSM-5 and Cu2(PcTN)2/W-HZSM-5 showed superior desulfurization performance and the activity of selectivity followed the order: T4BT 4DBT. The effects of phthalocyanine concentration were studied by UV–Vis and calcination temperature was obtained by TG-DSC for Cu2(PcTN)2/W-HZSM5. Catalysts were characterized by EA, IR, XRD, SEM, TEM, ICP, and N2 adsorption. Reaction time, temperature and the amount of catalyst were investigated as the important parameters for optimization of the reaction. Furthermore, a possible process of oxidative desulfurization and the reaction products were proposed. & 2015 Elsevier Inc. All rights reserved.
Keywords: Oxidative desulfurization Desulfurization performance Sulfur compounds
1. Introduction Sulfur in transportation fuels is a major source of notorious and undesirable air pollution [1], which leads directly to the emission of SO2 and sulfate particulates [2]. Namely, the combustion of sulfur-containing fuels has negative health and environmental effects [3]. Although increasingly stringent regulations have been established worldwide to limit the sulfur level, a near-zero sulfur content [4,5] is still a challenging problem not only from a fundamental but also from an applied point of view. Thus, finding new processes for ultra-deep desulfurization of fuel oils has been an urgent need [6,7]. Recently, many efforts have been implemented to develop novel catalysts for desulfurization. A wide variety supports materials such as amphoteric carbon, mixed oxides and mesoporous materials have been studied. Soni et al. [8] prepared the 3-D
n Corresponding author at: Key Laboratory of Synthetic and Natural Functional Molecule Chemistry of Ministry of Education, Shaanxi Key Laboratory of PhysicoInorganic Chemistry, College of Chemistry & Material Science, Northwest University, Xi’an 710069, Shaanxi, China. Tel.: þ86 29 88302604; fax: þ 86 29 88303798. E-mail address: [email protected] (J. Zhao). 1 These authors contributed equally to this work.
http://dx.doi.org/10.1016/j.jssc.2015.01.009 0022-4596/& 2015 Elsevier Inc. All rights reserved.
ordered mesoporous KIT-6 support for effective hydrodesulfurization. Klimova et al. [9] reported the novel bifunctional NiMo/AlSBA-15 catalysts for deep hydrodesulfurization. Sarda et al. [4] showed that deep desulfurization of diesel fuel by selective adsorption over Ni/Al2O3 and Ni/ZSM-5 extrudates. Zhang et al. [10] prepared the bifunctional NiPb/ZnO-diatomite-ZSM-5 catalyst for adsorption desulfurization performance. Zhang et al. [11] reported that highly stable and regenerable Mn-based/SBA-15 sorbents for desulfurization. Jabbarnezhad et al. [12] reported that the sonochemical synthesis of NiMo/Al2O3–ZrO2 nanocatalyst in hydrodesulfurization reaction. Duan et al. [13] synthesized the porous L-KIT-6 silica–alumina material for hydrodesulfurization of benzothiophene. Razavian et al. [14] reported that the synthesis and application of ZSM-5/SAPO-34 and SAPO-34/ZSM-5 composite systems for propylene yield enhancement. Among different catalysts, HZSM-5 molecular sieve seems to be more suited to be modified because of its uniform and ordered micropores [15], high extensive surface area, as well as charged framework [2,16]. However, a variety of modification techniques are the key factor to improve the shape-selectivity of molecular sieve and catalytic activity [17,18]. Tungsten has received considerable attention due to its unique physical–chemical properties [19]. Thus, tungsten-containing HZSM-5 molecular sieve is a kind
348
N. Zhao et al. / Journal of Solid State Chemistry 225 (2015) 347–353
of potential catalyst which has highly reactivity for desulfurization [20,21]. Besides, metal phthalocyanines and their symmetrically substituted derivatives possess high sensitivity, fast response, and ease of processability [22–24]. Currently, modified phthalocyanines are of much interest for oxidative desulfurization processes in the oil industry and many scientists have explored their gripping feature [25,26]. In the previous study, Gedeon et al. [17] have reported that the ZSM-5 zeolite was synthesized in the presence of a copper-phthalocyanine complex by hydrothermal treatment. Zhang et al. [27] have prepared the ZSM-5-Ln (Pc)2 by chemical bond for catalytic oxidation of thiophene. However, HZSM-5 zeolite catalysts promoted both by metal ions and phthalocyanine are not yet available. Herein, we report the oxidation of model sulfur compounds with O2 as the oxidant using the W/HZSM-5 modified with metal phthalocyanines as catalyst to get a superior performance for desulfurization. Three model sulfur compounds were used: thiophene (T), benzothiophene (BT), and dibenzothiophene (DBT).
Anal. Calc. for C58H20O12N22Fe2 C, 52.41; H, 1.51; N, 23.19. Found: C, 52.34; H, 1.50; N, 22.20. Yield: 2.88 g, 54.18% m. p4 3001 IR (KBr) νmax/cm 1: 1608(νC ¼ N); 1518,1334(νNO2 ); 1134, 1087, 845, 726, 660 (νPC). Anal. Calc. for C58H20O12N22Co2 C, 52.17; H, 1.50; N, 23.09. Found: C, 52.01; H, 1.50; N, 23.15. Yield: 1.90 g, 35.54% m.p. 43001 IR (KBr) νmax/cm 1: 1607(νC ¼ N); 1522, 1335(νNO2 ); 1137, 1072, 848, 728, 666 (νPC). Anal. Calc. for C58H20O12N22Ni2 C, 51.20; H, 1.50; N, 23.10. Found: C, 51.08; H, 1.52; N, 23.01. Yield: 3.87 g, 72.57% m. p.4 3001 IR (KBr) νmax/cm 1 1606(νC ¼ N); 1517, 1333(νNO2 ); 1138, 1092, 847, 729, 676 (νPC). Anal. Calc. for C58H20O12N22Zn2 C, 51.71; H, 1.49; N, 22.88. Found: C, 51.68; H, 1.50; N, 22.95. Yield: 3.07 g, 57.07% m.p. 43001 IR (KBr) νmax/cm 1 1603(νC ¼ N); 1519, 1330(νNO2 ); 1137, 1090, 844, 726, 675 (νPC). Anal. Calc. for C58H20O12N22Mn2 C, 51.49; H, 1.51; N, 23.23. Found: C, 51.48; H, 1.51; N, 23.23. Yield: 2.66 g, 50.10% m.p. 43001 IR (KBr) νmax/cm 1 1607(νC ¼ N); 1521, 1334(νNO2 ); 1138, 1090, 847, 725, 675 (νPC).
2. Experimental 2.1. Materials HZSM-5 powder (SiO2/Al2O3 Z500, H-type) was purchased from Shanghai Novel Chemical Technology Co. Ltd. (China). 3-Chloropropyltriethoxysilane and 4-NitroPhthalimide were obtained from Aladdin Reagent Co. Ltd. (Shanghai, China). Model compounds including T, BT, and DBT were bought from J&K technology Co. Ltd. (Beijing, China). All solvents and reagents were used without any further purification. 2.2. The synthesis of catalyst 2.2.1. Preparation of W-HZSM-5 catalyst Incorporation of W into the HZSM-5 zeolite was prepared by impregnation method. It was performed by adding the HZSM-5 powder to the certain amounts of sodium tungstate solution, which was prepared by dissolving Na2WO4 in distilled water. Afterwards, the required amount of HCl was added to deposit tungsten acid on the surface of HZSM-5. The mixture was thoroughly stirred and left for 5 h at ambient temperature. The solid, after filtration, was washed with distilled water until no white precipitate was generated by adding silver nitrate solution. The sample was dried at 120 1C for 24 h and calcined in air at 500 1C for 4 h with a heating rate of 5 1C/min. The content of tungsten component was 1.056 ppm by ICP analysis. IR (KBr) νmax/cm 1:790 (νsSi–O); 1099 (νas Si–O); 450; 550. 2.2.2. Synthesis of M2(PcTN)2 (M ¼Fe, Co, Ni, Cu, Zn and Mn) 4-NitroPhthalimide, urea, NH4Cl, (NH4)6Mo7O24 4H2O, pyromellitic dianhydride and CuCl2 2H2O were ground in a mortar. Then the mixture was placed in a three-neck flask in which about 150 mL of aviation kerosene was added. Afterwards, the mixture solution was heated in a thermostatic oil bath at 180 1C for 5 h and violently stirred until the reaction was completed. The crude product was purified successively with150 mL deionized water, 2% hydrochloric acid, methyl alcohol, acetone, and trichloromethane, respectively. After filtration, the precipitate was dried in the oven. Similarly, M2(PcTN)2 (M¼ Fe, Co, Ni, Zn and Mn) were synthesized (Scheme 1). Anal. Calc. for C58H20O12N22Cu2 C, 51.82; H, 1.49; N, 22.93. Found: C, 51.90; H, 1.50; N, 22.83.Yield: 4.05 g, 75.41% m.p 43001 IR (KBr) νmax/cm 1: 1606(νC ¼ N); 1523, 1334(νNO2 ); 1140, 1087, 846, 726, 657 (νPC).
2.2.3. Synthesis of M2(PcAN)2–W-HZSM-5 M2(PcAN)2 (M ¼Fe, Co, Ni, Cu, Zn, and Mn) and functionalization of W-HZSM-5 were obtained according to the procedure reported by Zhang et al. [20] (Scheme 2). IR (KBr) for functionalization of W-HZSM-5: νmax/cm 1: 2928, 2850(νCH2 ). IR (KBr) νmax/cm 1 C58H32N22Cu2:1606(νC ¼ N); 3254, 3327 (νNH2 ); 1140, 1082, 822, 737, 644 (νPC). IR (KBr) νmax/cm 1 C58H32N22Fe2:1607(νC ¼ N); 3296, 3301 (νNH2 ); 1140, 1086, 823, 736, 648 (νPC). IR (KBr) νmax/cm 1 C58H32N22Co2:1604 (νC ¼ N); 3257, 3325 (νNH2 ); 1140, 1089, 823, 744, 642 (νPC). IR (KBr) νmax/cm 1 C58H32N22Ni2:1602(νC ¼ N); 3258, 3331 (νNH2 ); 1140, 1082, 823, 737, 648 (νPC). IR (KBr) νmax/cm 1 C58H32N22Zn2:1606(νC ¼ N); 3258, 3400 (νNH2 ); 1140, 1086, 824, 736, 647 (νPC). IR (KBr) νmax/cm 1 C58H32N22Mn2:1603(νC ¼ N); 3274, 3323 (νNH2 ); 1140, 1085, 823, 737, 644 (νPC). 2.2.4. Synthesis of M2(PcTN)2/W-HZSM-5 The M2(PcTN)2 (M ¼Fe, Co, Ni, Cu, Zn, and Mn) loaded on WHZSM-5 complexes were prepared by first impregnating a certain amount of the W-HZSM-5 zeolite were stirred with a calculated amount of M2(PcTN)2 in 20 mL DMF solution. The sample was dried at 90 1C and calcined at 300 1C for 3 h, and subsequently impregnated with the same amount phthalocyanine solution. Finally, the sample was dried for 24 h and calcined at 300 1C for 3 h. The sample was designated as xPc/W-HZSM-5(x ¼load of phthalocyanine). A series of nominal load of phthalocyanine had been explored. 2.3. Characterization of catalysts Concentration of W on the W-HZSM-5 catalyst was analyzed using an inductively coupled plasma atomic emission spectrometry (IRIS Advantage) after dissolution of the W-HZSM-5 in the aqua regia and heat treatment. IR spectra were recorded on the EQUINOX-55 FTIR spectrometer from German Bruker Company. Elemental analysis was performed by Vario EL-III CHNOS instrument and the obtained values agreed with the calculated ones. Powder X-ray diffraction patterns were characterized by a D8advance Advance (Bruker, Germany) using CuKα radiation in the 2θ range of 5–801 with a 2θ step size of 0.02. The surface morphology of catalyst was analyzed using SEM images (TM3000)
N. Zhao et al. / Journal of Solid State Chemistry 225 (2015) 347–353
349
Scheme 1. Synthesis of M2(PcTN)2 (M ¼ Fe, Co, Ni, Cu, Zn and Mn).
Scheme 2. Synthesis of M2(PcAN)2–W-HZSM-5.
and samples were coated with gold to make them conductive before analysis. Nitrogen adsorption and desorption measurements were measured on TriStar II 3020 for samples degassed at 423 K for 6 h. Transmission electron microscopic (TEM) images were obtained using Philips TecnaiG2 F20 instrument at an accelerating voltage of 200 kV. Thermostabilies of the samples were determined on NETZSCH STA 449C device from 33 1C to 1000 1C in nitrogen flow of 30 mL min 1. The UV–Vis spectra were recorded with UV-1700 using 1 cm path length cuvette.
magnetic stirrer and a reflux condenser. In this typical reaction run, the appropriate amount of catalysts was added to model oil and molecular oxygen was bubbled through the reaction solution. The mixture was heated at 80 1C in an air bath with stirring at a speed of 300 rpm. The samples were collected and centrifuged after every 30 min. Sulfur concentration in oil phase was determined using GC 6890. The conversion rate of S-compounds was calculated based on the following equation, in which η is the conversion rate, and C0 and Ct stand for the initial and final concentration of S-compounds, respectively.
2.4. Analysis The oxidative desulfurization experiments were carried out in a 100 mL three-neck flask connected with an oxygen cylinder, a
η¼
ðC 0 C t Þ 100%: C0
350
N. Zhao et al. / Journal of Solid State Chemistry 225 (2015) 347–353
Table 1 Desulfurization performances of different catalysts. Entry
Catalyst
M2(PcAN)2–W-HZSM-5 Cu2(PcAN)2–W-HZSM-5 Fe2(PcAN)2–W-HZSM-5 Co2(PcAN)2–W-HZSM-5 Ni2(PcAN)2–W-HZSM-5 Zn2(PcAN)2–W-HZSM-5 Mn2(PcAN)2–W-HZSM-5 M2(PcTN)2/W-HZSM-5 Cu2(PcTN)2/W-HZSM-5 Fe2(PcTN)2/W-HZSM-5 Co2(PcTN)2/W-HZSM-5 Ni2(PcTN)2/W-HZSM-5 Zn2(PcTN)2/W-HZSM-5 Mn2(PcTN)2/W-HZSM-5
Model sulfur compounds conversion (%) T
BT
DBT
93.82 92.12 93.23 93.12 91.00 90.89 91.99 89.99 90.33 90.68 90.45 90.46
91.23 88.33 90.88 90.18 89.03 88.36 88.12 86.98 87.32 88.23 84.35 83.88
87.32 86.02 86.32 87.23 85.88 85.45 85.78 84.23 84.99 85.76 82.13 81.08
3. Results and discussion 3.1. The selection of the catalyst for oxidative desulfurization Oxidative desulfurization of three sulfur compounds (T, BT, and DBT) on M2(PcAN)2–W-HZSM-5 and M2(PcTN)2/W-HZSM-5 (M ¼Fe, Co, Ni, Cu, Zn and Mn) were studied and summarized in Table 1. Among metal phthalocyanines based on zeolite catalysts, the Cu2(PcAN)2–W-HZSM-5 and Cu2(PcTN)2/W-HZSM-5 catalysts have been paid much attention due to its superior performance and the Cu2(PcAN)2–W-HZSM-5 have exhibited higher desulfurization rate compared to the Cu2(PcTN)2/W-HZSM-5. The results clearly point out that Cu2(PcAN)2–W-HZSM-5 could remove 93.82% for T, 91.23% for BT, and 87.32% for DBT, respectively. It is attributed to the different synthesis methods and copper phthalocyanine based on W-HZSM-5 is stronger interaction than those catalysts. Besides, the copper is easy to disperse. From Table 1, it can be also seen that conversion rate of model sulfur compounds on the same catalyst follows the order: T 4BT 4DBT because of its aromaticity and low electron density on its sulfur atom. 3.2. Characterization of Cu2(PcAN)2–W-HZSM-5- and Cu2(PcTN)2/ W-HZSM-5 3.2.1. XRD patterns The XRD patterns of (a) HZSM-5, (b) Cu2(PcAN)2–W-HZSM-5 and (c) Cu2(PcTN)2/W-HZSM-5 samples were illustrated in Fig. 1. All catalysts exhibit the typical peaks at the ranges of 2θ ¼ 7–91 and 2θ ¼23–251, which implies the MFI structure remained intact after the treatment. No diffraction peak associated with tungsten species and copper phthalocyanine are observed and the modified HZSM-5 has a negligible effect on the original structure. The possible reason is that the tungsten species and copper phthalocyanine are well dispersed on the surface of the HZSM-5 membrane and the crystalline size is too small to be detected by XRD. It is also indicated that the HZSM-5 membrane possesses unique pores properties and high specific surface area. 3.2.2. SEM analysis Fig. 2 shows the images of the three samples. From the SEM top view in Fig. 2(a), the needle-like ZSM-5 is presented and grown well together. However, the different morphologies between the two synthetic routes attribute to the pretreatment methods. It can be seen that the load catalysts have an obvious cubic structure and Cu2(PcAN)2–W-HZSM-5 is evenly dispersed compared to Cu2(PcTN)2/ W-HZSM-5 in crystal sizes and shapes. These cubic structures appear
Fig. 1. X-ray diffraction patterns: (a) HZSM-5, (b) Cu2(PcAN)2–W-HZSM-5, (c) Cu2(PcTN)2/W-HZSM-5.
more regular than the morphology of pure HZSM-5 zeolite, which indicates that copper phthalocyanine has been successfully combined onto the surface of the W-HZSM-5. 3.2.3. N2 adsorption/desorption isotherm analysis Low temperature nitrogen physisorption isotherms of (a) HZSM-5, (b) Cu2(PcAN)2–W-HZSM-5 and (c) Cu2(PcTN)2/W-HZSM-5 are shown in Fig. 3. For all the HZSM-5 samples, the most striking features are the sharp step over the narrow range of relative pressure, p/p0. As shown in Fig. 3, the amount of N2 adsorbed by HZSM-5 is higher than Cu2(PcAN)2–W-HZSM-5 and Cu2(PcTN)2/W-HZSM-5. The shape of nitrogen isotherms for all samples presents a type of IV isotherms with a H3 type hysteresis loop, which attributed to the existence of micro and mesoporosity in all samples. At the beginning, due to the volume filling of HZSM-5 micropores, all the N2 adsorption/desorption isotherms show the volume adsorbed increases obviously and then with the relative pressures increase, the volume adsorbed increases continually caused by the multilayer adsorption. However, the N2 adsorption/desorption isotherms of Cu2(PcAN)2–W-HZSM-5 and Cu2(PcTN)2/W-HZSM-5 exhibit a much smaller hysteresis loop, which may be attributed to the copper phthalocyanine incorporation in zeolite. Values of the BET surface area as well as the pore properties are summarized in Table 2. Compared to the HZSM-5 zeolite, modified samples show a significant decrease in both the BET surface area and the micropore volume, which mainly attributed to the coverage of the external surface of tungsten species and copper phthalocyanine. The pore diameters of Cu2(PcAN)2–W-HZSM-5 and Cu2(PcTN)2/W-HZSM-5 are 2.36 and 2.47 nm, respectively. In contrast, the HZSM-5 offers a relative lower pore diameter (2.20 nm), which can be caused by the thin molecular sieve pore wall due to the tungsten species and the copper phthalocyanine are supported on HZSM-5. 3.2.4. Dispersion morphology Fig. 4 displayed the HRTEM images of H-ZSM-5 (a), Cu2(PcAN)2– W-HZSM-5 (b) and Cu2(PcTN)2/W-HZSM-5 (c). Fig. 4(a) clearly showed some nanopores and HZSM-5 substrate was uniform and regular. Typical shapes for Cu2(PcAN)2–W-HZSM-5 and Cu2(PcTN)2/ W-HZSM-5 were found, which correlated with the different synthetic methods. Furthermore, the basic structure of HZSM-5 was shown not damaged after loading the phthalocyanines. Among these figures, Cu2(PcAN)2–W-HZSM-5 revealed an obvious highly
N. Zhao et al. / Journal of Solid State Chemistry 225 (2015) 347–353
351
Fig. 2. SEM images of (a) HZSM-5, (b) Cu2(PcAN)2–W-HZSM-5, (c) Cu2(PcTN)2/W-HZSM-5.
and we choose 300 1C temperature as the calcination conditions to keep the complete structure. 3.4. Optimization of Cu2(PcTN)2 loading in Cu2(PcTN)2/W-HZSM-5 for oxidative desulfurization
Fig. 3. N2 adsorption/desorption isotherms of different samples: (a) HZSM-5, (b) Cu2(PcAN)2–W-HZSM-5, (c) Cu2(PcTN)2/W-HZSM-5.
Table 2 Pore structure features of samples. Samples
SBET Volumn (cm3/g) (cm3/g) Total Micro Meso
HZSM-5 341.32 Cu2(PcAN)2–W-HZSM-5 159.08 Cu2(PcTN)2/W-HZSM-5 178.49
0.19 0.09 0.11
0.12 0.07 0.06
0.05 0.02 0.04
In order to investigate the optimization of Cu2(PcTN)2 loading, UV–Vis adsorption spectra of Cu2(PcTN)2 is carried out with phthalocyanine solution after dipping molecular sieve. Generally, the Q band appears at the range of 650–700 nm resulted from the HOMO-LUMO (π-πn) transitions and the B band absorptions at 300–350 nm are observed due to the transitions from deeper πlevels to LUMO. The two characteristic absorbance bands at 340 and 650 nm (Fig. 6) correspond to the phthalocyanine. Moreover, with the increase of concentration, a red shift or blue shift does not occur in the band, but absorption intensity of bands is enhanced slightly. On the other hand, Beer–Lambert law is obeyed. Through the standard curve, we get the concentration of a solution for phthalocyanine after impregnation and calculate the actual Cu2(PcTN)2 load by the concentration of a solution of phthalocyanine. We had a series of theoretical quality of phthalocyanine into molecular sieve (0.2 g, 0.4 g, 0.6 g, 0.8 g) and continuous dipping 2 times. We find that the catalytic performance of 0.4 g is better than those of 0.2 g, 0.6 g, 0.8 g, and after two times dipping, the actual load is 0.563 g.
Pore diameter (nm)
3.5. The optimal reaction conditions for oxidation desulfurization 2.20 2.36 2.47
ordered two-dimensional and well-developed mesoporous between adjacent nanorods. Moreover, it can be seen that the high crystallintity from its lattice fringes corresponding to the MFI structure. Mesoporous were clearly exhibited by TEM images in combination with the nitrogen isotherm and pore-size distribution results.
3.3. The calcination temperature of Cu2(PcTN)2/W-HZSM-5 TG-DSC curves of Cu2(PcTN)2 in Fig. 5 show the decomposed temperature of phthalocyanine when Cu2(PcTN)2/W-HZSM-5 is calcined, the total weight loss of Cu2(PcTN)2 in the temperature range of 410 1C to 470 1C is 90% and the corresponding exothermic peak emerges at 460 1C, which attributed to the collapse of the phthalocyanine skeleton. The result of TG curve shows that the Cu2(PcTN)2 structures remain stable before calcination at 400 1C. Therefore, the calcination temperature should be below this value
The sulfur removal depends relatively on the amount of catalyst, reaction time, and reaction temperature. Three design variables list in Table 3, where the interactions among the design variables are disregarded. It is shown that the reaction temperature played a significant role in the oxidation desulfurization process. As the reaction temperature increasing, the desulfurization rate increases to a maximum at 60 1C and then decreases. And the reaction time of 3 h is suitable for the reaction process. It can be explained that O2 could not contact sulfur-containing substance effectively at the beginning of reaction and is likely to aggregate in a local part. With the homogenizer continuous high speed stirring, O2 has been fully utilized and additional sulfur-containing substances are oxidized. The effect of catalyst amount from 0.08 to 0.12 g is investigated under the same reaction condition and the superior performance is shown when 0.1 g catalyst is added. 3.6. The analysis of the reaction products and reaction process In order to explore the reaction products, barium nitrate was added to the tail gas absorption equipment, while no white precipitate was generated, which indicates that the corresponding sulfones materials are not be oxidized to sulfur dioxide. Moreover, the oxidative reaction process of Cu2(PcAN)2–W-HZSM-5 and
352
N. Zhao et al. / Journal of Solid State Chemistry 225 (2015) 347–353
Fig. 4. HRTEM results for (a) HZSM-5, (b) Cu2(PcAN)2–W-HZSM-5, (c) Cu2(PcTN)2/W-HZSM-5.
Table 3 Results of orthogonal experiment. No. Factors
1 2 3 4 5 6 7 8 9
Desulfurization rate (%)
The amount Reaction of catalyst (g) temperature (1C)
Reaction time (h)
Cu2(PcAN)2–W- Cu2(PcTN)2/ HZSM-5 (DBT) W-HZSM-5 (DBT)
0.08 0.08 0.08 0.10 0.10 0.10 0.12 0.12 0.12
2 2.5 3 2.5 3 2 3 2 2.5
69.54 85.12 79.23 73.32 87.32 80.23 71.08 83.89 82.14
40 60 80 40 60 80 40 60 80
70.05 82.78 78.99 72.89 85.78 78.98 69.03 81.23 81.09
Fig. 5. TG-DSC curves of Cu2(PcTN)2.
Fig. 7. Probable reaction schematic for the oxidative desulfurization by O2.
4. Conclusion
Fig. 6. UV–Vis adsorption spectra of Cu2(PcTN)2 in DMF under the different concentrations.
Cu2(PcTN)2/W-HZSM-5 might proceed via the following steps (Fig. 7): the oxidative process includes two consecutive stages: the first stage includes sulfur compounds obtain one oxygen free radical ion yielding sulfoxides; second, the sulfoxides are oxidized to corresponding sulfones. Oxygen free radicals are produced under the continuous agitation and the catalysts could be dispersed well into the substrate phase during the oxidation reaction.
In summary, copper phthalocyanine has been successfully immobilized on the surface of HZSM-5 modified by tungsten and we found that 0.563 g Cu2(PcTN)2 was the suitable load and 300 1C was the best calcination temperature. The high catalytic activity for oxidative desulfurization is under the optimum conditions: the reaction temperature is 60 1C, the reaction time is 3 h and the amount of catalyst is 0.1 g. The removal ratio of T, BT, DBT reaches 93.82%, 91.23% and 87.32%, respectively.
Acknowledgment This work was supported by the National Natural Science Foundations of China (nos. 21171139 and 21371143) and the National Basic Research Program (973 Program) (2013CB934001).
N. Zhao et al. / Journal of Solid State Chemistry 225 (2015) 347–353
References [1] [2] [3] [4] [5]
[6] [7] [8] [9] [10] [11] [12] [13] [14]
Z. Long, C. Yang, G. Zeng, L. Peng, C. Dai, H. He, Fuel 130 (2014) 19–24. L. Wang, H. Cai, S. Li, N. Mominou, Fuel 105 (2013) 752–756. K.S. Triantafyllidis, E.A. Deliyanni, Chem. Eng. J. 236 (2014) 406–414. K.K. Sarda, A. Bhandari, K.K. Pant, S. Jain, Fuel 93 (2012) 86–91. D. Stratiev, V. Yankov, l. Petrov, I. Shishkova, A. Pavlova, P. Ivanova, A. Surleva, K. Hristov, E. Todorova, A. Obryvalina, R. Telyashev, Fuel Process. Technol. 126 (2014) 332–342. A. Stanislaus, A. Marafi, M.S. Rana, Catal. Today 153 (2010) 1–68. X. Pang, L. Zhang, S. Sun, T. Liu, X. Gao, Catal. Today 125 (2007) 173–177. K. Soni, B.S. Rana, A.K. Sinha, A. Bhaumik, M. Nandi, M. Kumar, G.M. Dhar, Appl. Catal., B: Environ. 90 (2009) 55–63. T. Klimova, J. Reyes, O. Gutiérrez, L. Lizama, Appl. Catal., A: Gen. 335 (2008) 159–171. K. Zhang, Y. Liu, S. Tian, E. Zhao, J. Zhang, C. Liu, Fuel 104 (2013) 201–207. F.M. Zhang, B.S. Liu, Y. Zhang, Y.H. Guo, Z.Y. Wan, F. Subhan, J. Hazard. Mater. 233-234 (2012) 219–227. P. Jabbarnezhad, M. Haghighi, P. Taghavinezhad, Fuel Process. Technol. 126 (2014) 392–401. A. Duan, T. Li, Z. Zhao, B. Liu, X. Zhou, G. Jiang, J. Liu, Y. Wei, H. Pan, Appl. Catal., B: Environ. 165 (2015) 763–773. M. Razavian, S. Fatemi, Microporous Mesoporous Mater. 201 (2015) 176–189.
353
[15] T. Zhang, X. Zhang, X. Yan, L. Lin, H. Liu, J. Qiu, K.L. Yeung, Catal. Today 236 (2014) 41–48. [16] D.D. Anggoro, I. stadi, J. Nat. Gas Chem. 17 (2008) 39–44. [17] A. Gedeon, M. Valeux, M. Gruia, G. Minghua, J. Fraissard, Solid State Nucl. Magn. Reson. 9 (1997) 269–276. [18] H.L. Janardhan, G.V. Shanbhag, A.B. Halgeri, Appl. Catal., A: Gen. 471 (2014) 12–18. [19] A. Satter, G. Parkin, Nature 463 (2010) 523–526. [20] G. Rodriguez-Gattorno, A. Galano, E. Torres-García, Appl. Catal., B: Environ. 92 (2009) 1–8. [21] C. Jiang, J. Wang, S. Wang, H.Y. Guan, X. Wang, M. Huo, Appl. Catal., B: Environ. 106 (2011) 343–349. [22] E.N. Kaya, S. Tuncel, T.V. Basova, H. Banimuslem, A. Hassan, A.G. Gürek, V. Ahsen, M. Durmus, Sens. Actuators, B: Chem. 199 (2014) 277–283. [23] B.N. Achar, P.K. Jayasree, Synth. Met. 114 (2000) 219–224. [24] J. Wei, M. Xu, J. Zhang, R. Zhao, X. Liu, J. Magn. Magn. Mater. 324 (2012) 2696–2700. [25] R. Bayrak, O. Bekircan, M. Durmus, I. Degirmencioglu, J. Organomet. Chem. 767 (2014) 101–107. [26] X. Zhou, J. Li, X. Wang, K. Jin, W. Ma, Fuel Process. Technol. 90 (2009) 317–323. [27] Y. Zhang, D. Wang, R. Zhang, J. Zhao, Y. Zheng, Catal. Commun. 29 (2012) 21–23.
Contents lists available at ScienceDirect
Journal of Solid State Chemistry journal homepage: www.elsevier.com/locate/jssc
Synthesis and application of different phthalocyanine molecular sieve catalyst for oxidative desulfurization Na Zhao a,c,1, Siwen Li a,c,1, Jinyi Wang a,c, Ronglan Zhang a,c, Ruimin Gao b,c, Jianshe Zhao a,c,n, Junlong Wang b,c a Key Laboratory of Synthetic and Natural Functional Molecule Chemistry of Ministry of Education, Shaanxi Key Laboratory of Physico-Inorganic Chemistry, College of Chemistry & Material Science, Northwest University, Xi’an 710069, Shaanxi, China b Research Institute of Shaanxi Yanchang Petroleum Group Corp. Ltd., Xi’an 710075, China c Composites Research Institute, Weinan Normal University, Weinan 714000, China
art ic l e i nf o
a b s t r a c t
Article history: Received 29 October 2014 Received in revised form 4 January 2015 Accepted 11 January 2015 Available online 23 January 2015
M2(PcAN)2 (M¼ Fe, Co, Ni, Cu, Zn and Mn) anchored onto W-HZSM-5 (M2(PcAN)2–W-HZSM-5) or the M2(PcTN)2 doping W-HZSM-5 (M2(PcTN)2/W-HZSM-5) were prepared and their catalytic performances were tested for oxidative desulfurization in the presence of oxygen. Thiophene (T), benzothiophene (BT), and dibenzothiophene (DBT) were considered as sulfur compounds. Among zeolite-based catalysts, the Cu2(PcAN)2–W-HZSM-5 and Cu2(PcTN)2/W-HZSM-5 showed superior desulfurization performance and the activity of selectivity followed the order: T4BT 4DBT. The effects of phthalocyanine concentration were studied by UV–Vis and calcination temperature was obtained by TG-DSC for Cu2(PcTN)2/W-HZSM5. Catalysts were characterized by EA, IR, XRD, SEM, TEM, ICP, and N2 adsorption. Reaction time, temperature and the amount of catalyst were investigated as the important parameters for optimization of the reaction. Furthermore, a possible process of oxidative desulfurization and the reaction products were proposed. & 2015 Elsevier Inc. All rights reserved.
Keywords: Oxidative desulfurization Desulfurization performance Sulfur compounds
1. Introduction Sulfur in transportation fuels is a major source of notorious and undesirable air pollution [1], which leads directly to the emission of SO2 and sulfate particulates [2]. Namely, the combustion of sulfur-containing fuels has negative health and environmental effects [3]. Although increasingly stringent regulations have been established worldwide to limit the sulfur level, a near-zero sulfur content [4,5] is still a challenging problem not only from a fundamental but also from an applied point of view. Thus, finding new processes for ultra-deep desulfurization of fuel oils has been an urgent need [6,7]. Recently, many efforts have been implemented to develop novel catalysts for desulfurization. A wide variety supports materials such as amphoteric carbon, mixed oxides and mesoporous materials have been studied. Soni et al. [8] prepared the 3-D
n Corresponding author at: Key Laboratory of Synthetic and Natural Functional Molecule Chemistry of Ministry of Education, Shaanxi Key Laboratory of PhysicoInorganic Chemistry, College of Chemistry & Material Science, Northwest University, Xi’an 710069, Shaanxi, China. Tel.: þ86 29 88302604; fax: þ 86 29 88303798. E-mail address: [email protected] (J. Zhao). 1 These authors contributed equally to this work.
http://dx.doi.org/10.1016/j.jssc.2015.01.009 0022-4596/& 2015 Elsevier Inc. All rights reserved.
ordered mesoporous KIT-6 support for effective hydrodesulfurization. Klimova et al. [9] reported the novel bifunctional NiMo/AlSBA-15 catalysts for deep hydrodesulfurization. Sarda et al. [4] showed that deep desulfurization of diesel fuel by selective adsorption over Ni/Al2O3 and Ni/ZSM-5 extrudates. Zhang et al. [10] prepared the bifunctional NiPb/ZnO-diatomite-ZSM-5 catalyst for adsorption desulfurization performance. Zhang et al. [11] reported that highly stable and regenerable Mn-based/SBA-15 sorbents for desulfurization. Jabbarnezhad et al. [12] reported that the sonochemical synthesis of NiMo/Al2O3–ZrO2 nanocatalyst in hydrodesulfurization reaction. Duan et al. [13] synthesized the porous L-KIT-6 silica–alumina material for hydrodesulfurization of benzothiophene. Razavian et al. [14] reported that the synthesis and application of ZSM-5/SAPO-34 and SAPO-34/ZSM-5 composite systems for propylene yield enhancement. Among different catalysts, HZSM-5 molecular sieve seems to be more suited to be modified because of its uniform and ordered micropores [15], high extensive surface area, as well as charged framework [2,16]. However, a variety of modification techniques are the key factor to improve the shape-selectivity of molecular sieve and catalytic activity [17,18]. Tungsten has received considerable attention due to its unique physical–chemical properties [19]. Thus, tungsten-containing HZSM-5 molecular sieve is a kind
348
N. Zhao et al. / Journal of Solid State Chemistry 225 (2015) 347–353
of potential catalyst which has highly reactivity for desulfurization [20,21]. Besides, metal phthalocyanines and their symmetrically substituted derivatives possess high sensitivity, fast response, and ease of processability [22–24]. Currently, modified phthalocyanines are of much interest for oxidative desulfurization processes in the oil industry and many scientists have explored their gripping feature [25,26]. In the previous study, Gedeon et al. [17] have reported that the ZSM-5 zeolite was synthesized in the presence of a copper-phthalocyanine complex by hydrothermal treatment. Zhang et al. [27] have prepared the ZSM-5-Ln (Pc)2 by chemical bond for catalytic oxidation of thiophene. However, HZSM-5 zeolite catalysts promoted both by metal ions and phthalocyanine are not yet available. Herein, we report the oxidation of model sulfur compounds with O2 as the oxidant using the W/HZSM-5 modified with metal phthalocyanines as catalyst to get a superior performance for desulfurization. Three model sulfur compounds were used: thiophene (T), benzothiophene (BT), and dibenzothiophene (DBT).
Anal. Calc. for C58H20O12N22Fe2 C, 52.41; H, 1.51; N, 23.19. Found: C, 52.34; H, 1.50; N, 22.20. Yield: 2.88 g, 54.18% m. p4 3001 IR (KBr) νmax/cm 1: 1608(νC ¼ N); 1518,1334(νNO2 ); 1134, 1087, 845, 726, 660 (νPC). Anal. Calc. for C58H20O12N22Co2 C, 52.17; H, 1.50; N, 23.09. Found: C, 52.01; H, 1.50; N, 23.15. Yield: 1.90 g, 35.54% m.p. 43001 IR (KBr) νmax/cm 1: 1607(νC ¼ N); 1522, 1335(νNO2 ); 1137, 1072, 848, 728, 666 (νPC). Anal. Calc. for C58H20O12N22Ni2 C, 51.20; H, 1.50; N, 23.10. Found: C, 51.08; H, 1.52; N, 23.01. Yield: 3.87 g, 72.57% m. p.4 3001 IR (KBr) νmax/cm 1 1606(νC ¼ N); 1517, 1333(νNO2 ); 1138, 1092, 847, 729, 676 (νPC). Anal. Calc. for C58H20O12N22Zn2 C, 51.71; H, 1.49; N, 22.88. Found: C, 51.68; H, 1.50; N, 22.95. Yield: 3.07 g, 57.07% m.p. 43001 IR (KBr) νmax/cm 1 1603(νC ¼ N); 1519, 1330(νNO2 ); 1137, 1090, 844, 726, 675 (νPC). Anal. Calc. for C58H20O12N22Mn2 C, 51.49; H, 1.51; N, 23.23. Found: C, 51.48; H, 1.51; N, 23.23. Yield: 2.66 g, 50.10% m.p. 43001 IR (KBr) νmax/cm 1 1607(νC ¼ N); 1521, 1334(νNO2 ); 1138, 1090, 847, 725, 675 (νPC).
2. Experimental 2.1. Materials HZSM-5 powder (SiO2/Al2O3 Z500, H-type) was purchased from Shanghai Novel Chemical Technology Co. Ltd. (China). 3-Chloropropyltriethoxysilane and 4-NitroPhthalimide were obtained from Aladdin Reagent Co. Ltd. (Shanghai, China). Model compounds including T, BT, and DBT were bought from J&K technology Co. Ltd. (Beijing, China). All solvents and reagents were used without any further purification. 2.2. The synthesis of catalyst 2.2.1. Preparation of W-HZSM-5 catalyst Incorporation of W into the HZSM-5 zeolite was prepared by impregnation method. It was performed by adding the HZSM-5 powder to the certain amounts of sodium tungstate solution, which was prepared by dissolving Na2WO4 in distilled water. Afterwards, the required amount of HCl was added to deposit tungsten acid on the surface of HZSM-5. The mixture was thoroughly stirred and left for 5 h at ambient temperature. The solid, after filtration, was washed with distilled water until no white precipitate was generated by adding silver nitrate solution. The sample was dried at 120 1C for 24 h and calcined in air at 500 1C for 4 h with a heating rate of 5 1C/min. The content of tungsten component was 1.056 ppm by ICP analysis. IR (KBr) νmax/cm 1:790 (νsSi–O); 1099 (νas Si–O); 450; 550. 2.2.2. Synthesis of M2(PcTN)2 (M ¼Fe, Co, Ni, Cu, Zn and Mn) 4-NitroPhthalimide, urea, NH4Cl, (NH4)6Mo7O24 4H2O, pyromellitic dianhydride and CuCl2 2H2O were ground in a mortar. Then the mixture was placed in a three-neck flask in which about 150 mL of aviation kerosene was added. Afterwards, the mixture solution was heated in a thermostatic oil bath at 180 1C for 5 h and violently stirred until the reaction was completed. The crude product was purified successively with150 mL deionized water, 2% hydrochloric acid, methyl alcohol, acetone, and trichloromethane, respectively. After filtration, the precipitate was dried in the oven. Similarly, M2(PcTN)2 (M¼ Fe, Co, Ni, Zn and Mn) were synthesized (Scheme 1). Anal. Calc. for C58H20O12N22Cu2 C, 51.82; H, 1.49; N, 22.93. Found: C, 51.90; H, 1.50; N, 22.83.Yield: 4.05 g, 75.41% m.p 43001 IR (KBr) νmax/cm 1: 1606(νC ¼ N); 1523, 1334(νNO2 ); 1140, 1087, 846, 726, 657 (νPC).
2.2.3. Synthesis of M2(PcAN)2–W-HZSM-5 M2(PcAN)2 (M ¼Fe, Co, Ni, Cu, Zn, and Mn) and functionalization of W-HZSM-5 were obtained according to the procedure reported by Zhang et al. [20] (Scheme 2). IR (KBr) for functionalization of W-HZSM-5: νmax/cm 1: 2928, 2850(νCH2 ). IR (KBr) νmax/cm 1 C58H32N22Cu2:1606(νC ¼ N); 3254, 3327 (νNH2 ); 1140, 1082, 822, 737, 644 (νPC). IR (KBr) νmax/cm 1 C58H32N22Fe2:1607(νC ¼ N); 3296, 3301 (νNH2 ); 1140, 1086, 823, 736, 648 (νPC). IR (KBr) νmax/cm 1 C58H32N22Co2:1604 (νC ¼ N); 3257, 3325 (νNH2 ); 1140, 1089, 823, 744, 642 (νPC). IR (KBr) νmax/cm 1 C58H32N22Ni2:1602(νC ¼ N); 3258, 3331 (νNH2 ); 1140, 1082, 823, 737, 648 (νPC). IR (KBr) νmax/cm 1 C58H32N22Zn2:1606(νC ¼ N); 3258, 3400 (νNH2 ); 1140, 1086, 824, 736, 647 (νPC). IR (KBr) νmax/cm 1 C58H32N22Mn2:1603(νC ¼ N); 3274, 3323 (νNH2 ); 1140, 1085, 823, 737, 644 (νPC). 2.2.4. Synthesis of M2(PcTN)2/W-HZSM-5 The M2(PcTN)2 (M ¼Fe, Co, Ni, Cu, Zn, and Mn) loaded on WHZSM-5 complexes were prepared by first impregnating a certain amount of the W-HZSM-5 zeolite were stirred with a calculated amount of M2(PcTN)2 in 20 mL DMF solution. The sample was dried at 90 1C and calcined at 300 1C for 3 h, and subsequently impregnated with the same amount phthalocyanine solution. Finally, the sample was dried for 24 h and calcined at 300 1C for 3 h. The sample was designated as xPc/W-HZSM-5(x ¼load of phthalocyanine). A series of nominal load of phthalocyanine had been explored. 2.3. Characterization of catalysts Concentration of W on the W-HZSM-5 catalyst was analyzed using an inductively coupled plasma atomic emission spectrometry (IRIS Advantage) after dissolution of the W-HZSM-5 in the aqua regia and heat treatment. IR spectra were recorded on the EQUINOX-55 FTIR spectrometer from German Bruker Company. Elemental analysis was performed by Vario EL-III CHNOS instrument and the obtained values agreed with the calculated ones. Powder X-ray diffraction patterns were characterized by a D8advance Advance (Bruker, Germany) using CuKα radiation in the 2θ range of 5–801 with a 2θ step size of 0.02. The surface morphology of catalyst was analyzed using SEM images (TM3000)
N. Zhao et al. / Journal of Solid State Chemistry 225 (2015) 347–353
349
Scheme 1. Synthesis of M2(PcTN)2 (M ¼ Fe, Co, Ni, Cu, Zn and Mn).
Scheme 2. Synthesis of M2(PcAN)2–W-HZSM-5.
and samples were coated with gold to make them conductive before analysis. Nitrogen adsorption and desorption measurements were measured on TriStar II 3020 for samples degassed at 423 K for 6 h. Transmission electron microscopic (TEM) images were obtained using Philips TecnaiG2 F20 instrument at an accelerating voltage of 200 kV. Thermostabilies of the samples were determined on NETZSCH STA 449C device from 33 1C to 1000 1C in nitrogen flow of 30 mL min 1. The UV–Vis spectra were recorded with UV-1700 using 1 cm path length cuvette.
magnetic stirrer and a reflux condenser. In this typical reaction run, the appropriate amount of catalysts was added to model oil and molecular oxygen was bubbled through the reaction solution. The mixture was heated at 80 1C in an air bath with stirring at a speed of 300 rpm. The samples were collected and centrifuged after every 30 min. Sulfur concentration in oil phase was determined using GC 6890. The conversion rate of S-compounds was calculated based on the following equation, in which η is the conversion rate, and C0 and Ct stand for the initial and final concentration of S-compounds, respectively.
2.4. Analysis The oxidative desulfurization experiments were carried out in a 100 mL three-neck flask connected with an oxygen cylinder, a
η¼
ðC 0 C t Þ 100%: C0
350
N. Zhao et al. / Journal of Solid State Chemistry 225 (2015) 347–353
Table 1 Desulfurization performances of different catalysts. Entry
Catalyst
M2(PcAN)2–W-HZSM-5 Cu2(PcAN)2–W-HZSM-5 Fe2(PcAN)2–W-HZSM-5 Co2(PcAN)2–W-HZSM-5 Ni2(PcAN)2–W-HZSM-5 Zn2(PcAN)2–W-HZSM-5 Mn2(PcAN)2–W-HZSM-5 M2(PcTN)2/W-HZSM-5 Cu2(PcTN)2/W-HZSM-5 Fe2(PcTN)2/W-HZSM-5 Co2(PcTN)2/W-HZSM-5 Ni2(PcTN)2/W-HZSM-5 Zn2(PcTN)2/W-HZSM-5 Mn2(PcTN)2/W-HZSM-5
Model sulfur compounds conversion (%) T
BT
DBT
93.82 92.12 93.23 93.12 91.00 90.89 91.99 89.99 90.33 90.68 90.45 90.46
91.23 88.33 90.88 90.18 89.03 88.36 88.12 86.98 87.32 88.23 84.35 83.88
87.32 86.02 86.32 87.23 85.88 85.45 85.78 84.23 84.99 85.76 82.13 81.08
3. Results and discussion 3.1. The selection of the catalyst for oxidative desulfurization Oxidative desulfurization of three sulfur compounds (T, BT, and DBT) on M2(PcAN)2–W-HZSM-5 and M2(PcTN)2/W-HZSM-5 (M ¼Fe, Co, Ni, Cu, Zn and Mn) were studied and summarized in Table 1. Among metal phthalocyanines based on zeolite catalysts, the Cu2(PcAN)2–W-HZSM-5 and Cu2(PcTN)2/W-HZSM-5 catalysts have been paid much attention due to its superior performance and the Cu2(PcAN)2–W-HZSM-5 have exhibited higher desulfurization rate compared to the Cu2(PcTN)2/W-HZSM-5. The results clearly point out that Cu2(PcAN)2–W-HZSM-5 could remove 93.82% for T, 91.23% for BT, and 87.32% for DBT, respectively. It is attributed to the different synthesis methods and copper phthalocyanine based on W-HZSM-5 is stronger interaction than those catalysts. Besides, the copper is easy to disperse. From Table 1, it can be also seen that conversion rate of model sulfur compounds on the same catalyst follows the order: T 4BT 4DBT because of its aromaticity and low electron density on its sulfur atom. 3.2. Characterization of Cu2(PcAN)2–W-HZSM-5- and Cu2(PcTN)2/ W-HZSM-5 3.2.1. XRD patterns The XRD patterns of (a) HZSM-5, (b) Cu2(PcAN)2–W-HZSM-5 and (c) Cu2(PcTN)2/W-HZSM-5 samples were illustrated in Fig. 1. All catalysts exhibit the typical peaks at the ranges of 2θ ¼ 7–91 and 2θ ¼23–251, which implies the MFI structure remained intact after the treatment. No diffraction peak associated with tungsten species and copper phthalocyanine are observed and the modified HZSM-5 has a negligible effect on the original structure. The possible reason is that the tungsten species and copper phthalocyanine are well dispersed on the surface of the HZSM-5 membrane and the crystalline size is too small to be detected by XRD. It is also indicated that the HZSM-5 membrane possesses unique pores properties and high specific surface area. 3.2.2. SEM analysis Fig. 2 shows the images of the three samples. From the SEM top view in Fig. 2(a), the needle-like ZSM-5 is presented and grown well together. However, the different morphologies between the two synthetic routes attribute to the pretreatment methods. It can be seen that the load catalysts have an obvious cubic structure and Cu2(PcAN)2–W-HZSM-5 is evenly dispersed compared to Cu2(PcTN)2/ W-HZSM-5 in crystal sizes and shapes. These cubic structures appear
Fig. 1. X-ray diffraction patterns: (a) HZSM-5, (b) Cu2(PcAN)2–W-HZSM-5, (c) Cu2(PcTN)2/W-HZSM-5.
more regular than the morphology of pure HZSM-5 zeolite, which indicates that copper phthalocyanine has been successfully combined onto the surface of the W-HZSM-5. 3.2.3. N2 adsorption/desorption isotherm analysis Low temperature nitrogen physisorption isotherms of (a) HZSM-5, (b) Cu2(PcAN)2–W-HZSM-5 and (c) Cu2(PcTN)2/W-HZSM-5 are shown in Fig. 3. For all the HZSM-5 samples, the most striking features are the sharp step over the narrow range of relative pressure, p/p0. As shown in Fig. 3, the amount of N2 adsorbed by HZSM-5 is higher than Cu2(PcAN)2–W-HZSM-5 and Cu2(PcTN)2/W-HZSM-5. The shape of nitrogen isotherms for all samples presents a type of IV isotherms with a H3 type hysteresis loop, which attributed to the existence of micro and mesoporosity in all samples. At the beginning, due to the volume filling of HZSM-5 micropores, all the N2 adsorption/desorption isotherms show the volume adsorbed increases obviously and then with the relative pressures increase, the volume adsorbed increases continually caused by the multilayer adsorption. However, the N2 adsorption/desorption isotherms of Cu2(PcAN)2–W-HZSM-5 and Cu2(PcTN)2/W-HZSM-5 exhibit a much smaller hysteresis loop, which may be attributed to the copper phthalocyanine incorporation in zeolite. Values of the BET surface area as well as the pore properties are summarized in Table 2. Compared to the HZSM-5 zeolite, modified samples show a significant decrease in both the BET surface area and the micropore volume, which mainly attributed to the coverage of the external surface of tungsten species and copper phthalocyanine. The pore diameters of Cu2(PcAN)2–W-HZSM-5 and Cu2(PcTN)2/W-HZSM-5 are 2.36 and 2.47 nm, respectively. In contrast, the HZSM-5 offers a relative lower pore diameter (2.20 nm), which can be caused by the thin molecular sieve pore wall due to the tungsten species and the copper phthalocyanine are supported on HZSM-5. 3.2.4. Dispersion morphology Fig. 4 displayed the HRTEM images of H-ZSM-5 (a), Cu2(PcAN)2– W-HZSM-5 (b) and Cu2(PcTN)2/W-HZSM-5 (c). Fig. 4(a) clearly showed some nanopores and HZSM-5 substrate was uniform and regular. Typical shapes for Cu2(PcAN)2–W-HZSM-5 and Cu2(PcTN)2/ W-HZSM-5 were found, which correlated with the different synthetic methods. Furthermore, the basic structure of HZSM-5 was shown not damaged after loading the phthalocyanines. Among these figures, Cu2(PcAN)2–W-HZSM-5 revealed an obvious highly
N. Zhao et al. / Journal of Solid State Chemistry 225 (2015) 347–353
351
Fig. 2. SEM images of (a) HZSM-5, (b) Cu2(PcAN)2–W-HZSM-5, (c) Cu2(PcTN)2/W-HZSM-5.
and we choose 300 1C temperature as the calcination conditions to keep the complete structure. 3.4. Optimization of Cu2(PcTN)2 loading in Cu2(PcTN)2/W-HZSM-5 for oxidative desulfurization
Fig. 3. N2 adsorption/desorption isotherms of different samples: (a) HZSM-5, (b) Cu2(PcAN)2–W-HZSM-5, (c) Cu2(PcTN)2/W-HZSM-5.
Table 2 Pore structure features of samples. Samples
SBET Volumn (cm3/g) (cm3/g) Total Micro Meso
HZSM-5 341.32 Cu2(PcAN)2–W-HZSM-5 159.08 Cu2(PcTN)2/W-HZSM-5 178.49
0.19 0.09 0.11
0.12 0.07 0.06
0.05 0.02 0.04
In order to investigate the optimization of Cu2(PcTN)2 loading, UV–Vis adsorption spectra of Cu2(PcTN)2 is carried out with phthalocyanine solution after dipping molecular sieve. Generally, the Q band appears at the range of 650–700 nm resulted from the HOMO-LUMO (π-πn) transitions and the B band absorptions at 300–350 nm are observed due to the transitions from deeper πlevels to LUMO. The two characteristic absorbance bands at 340 and 650 nm (Fig. 6) correspond to the phthalocyanine. Moreover, with the increase of concentration, a red shift or blue shift does not occur in the band, but absorption intensity of bands is enhanced slightly. On the other hand, Beer–Lambert law is obeyed. Through the standard curve, we get the concentration of a solution for phthalocyanine after impregnation and calculate the actual Cu2(PcTN)2 load by the concentration of a solution of phthalocyanine. We had a series of theoretical quality of phthalocyanine into molecular sieve (0.2 g, 0.4 g, 0.6 g, 0.8 g) and continuous dipping 2 times. We find that the catalytic performance of 0.4 g is better than those of 0.2 g, 0.6 g, 0.8 g, and after two times dipping, the actual load is 0.563 g.
Pore diameter (nm)
3.5. The optimal reaction conditions for oxidation desulfurization 2.20 2.36 2.47
ordered two-dimensional and well-developed mesoporous between adjacent nanorods. Moreover, it can be seen that the high crystallintity from its lattice fringes corresponding to the MFI structure. Mesoporous were clearly exhibited by TEM images in combination with the nitrogen isotherm and pore-size distribution results.
3.3. The calcination temperature of Cu2(PcTN)2/W-HZSM-5 TG-DSC curves of Cu2(PcTN)2 in Fig. 5 show the decomposed temperature of phthalocyanine when Cu2(PcTN)2/W-HZSM-5 is calcined, the total weight loss of Cu2(PcTN)2 in the temperature range of 410 1C to 470 1C is 90% and the corresponding exothermic peak emerges at 460 1C, which attributed to the collapse of the phthalocyanine skeleton. The result of TG curve shows that the Cu2(PcTN)2 structures remain stable before calcination at 400 1C. Therefore, the calcination temperature should be below this value
The sulfur removal depends relatively on the amount of catalyst, reaction time, and reaction temperature. Three design variables list in Table 3, where the interactions among the design variables are disregarded. It is shown that the reaction temperature played a significant role in the oxidation desulfurization process. As the reaction temperature increasing, the desulfurization rate increases to a maximum at 60 1C and then decreases. And the reaction time of 3 h is suitable for the reaction process. It can be explained that O2 could not contact sulfur-containing substance effectively at the beginning of reaction and is likely to aggregate in a local part. With the homogenizer continuous high speed stirring, O2 has been fully utilized and additional sulfur-containing substances are oxidized. The effect of catalyst amount from 0.08 to 0.12 g is investigated under the same reaction condition and the superior performance is shown when 0.1 g catalyst is added. 3.6. The analysis of the reaction products and reaction process In order to explore the reaction products, barium nitrate was added to the tail gas absorption equipment, while no white precipitate was generated, which indicates that the corresponding sulfones materials are not be oxidized to sulfur dioxide. Moreover, the oxidative reaction process of Cu2(PcAN)2–W-HZSM-5 and
352
N. Zhao et al. / Journal of Solid State Chemistry 225 (2015) 347–353
Fig. 4. HRTEM results for (a) HZSM-5, (b) Cu2(PcAN)2–W-HZSM-5, (c) Cu2(PcTN)2/W-HZSM-5.
Table 3 Results of orthogonal experiment. No. Factors
1 2 3 4 5 6 7 8 9
Desulfurization rate (%)
The amount Reaction of catalyst (g) temperature (1C)
Reaction time (h)
Cu2(PcAN)2–W- Cu2(PcTN)2/ HZSM-5 (DBT) W-HZSM-5 (DBT)
0.08 0.08 0.08 0.10 0.10 0.10 0.12 0.12 0.12
2 2.5 3 2.5 3 2 3 2 2.5
69.54 85.12 79.23 73.32 87.32 80.23 71.08 83.89 82.14
40 60 80 40 60 80 40 60 80
70.05 82.78 78.99 72.89 85.78 78.98 69.03 81.23 81.09
Fig. 5. TG-DSC curves of Cu2(PcTN)2.
Fig. 7. Probable reaction schematic for the oxidative desulfurization by O2.
4. Conclusion
Fig. 6. UV–Vis adsorption spectra of Cu2(PcTN)2 in DMF under the different concentrations.
Cu2(PcTN)2/W-HZSM-5 might proceed via the following steps (Fig. 7): the oxidative process includes two consecutive stages: the first stage includes sulfur compounds obtain one oxygen free radical ion yielding sulfoxides; second, the sulfoxides are oxidized to corresponding sulfones. Oxygen free radicals are produced under the continuous agitation and the catalysts could be dispersed well into the substrate phase during the oxidation reaction.
In summary, copper phthalocyanine has been successfully immobilized on the surface of HZSM-5 modified by tungsten and we found that 0.563 g Cu2(PcTN)2 was the suitable load and 300 1C was the best calcination temperature. The high catalytic activity for oxidative desulfurization is under the optimum conditions: the reaction temperature is 60 1C, the reaction time is 3 h and the amount of catalyst is 0.1 g. The removal ratio of T, BT, DBT reaches 93.82%, 91.23% and 87.32%, respectively.
Acknowledgment This work was supported by the National Natural Science Foundations of China (nos. 21171139 and 21371143) and the National Basic Research Program (973 Program) (2013CB934001).
N. Zhao et al. / Journal of Solid State Chemistry 225 (2015) 347–353
References [1] [2] [3] [4] [5]
[6] [7] [8] [9] [10] [11] [12] [13] [14]
Z. Long, C. Yang, G. Zeng, L. Peng, C. Dai, H. He, Fuel 130 (2014) 19–24. L. Wang, H. Cai, S. Li, N. Mominou, Fuel 105 (2013) 752–756. K.S. Triantafyllidis, E.A. Deliyanni, Chem. Eng. J. 236 (2014) 406–414. K.K. Sarda, A. Bhandari, K.K. Pant, S. Jain, Fuel 93 (2012) 86–91. D. Stratiev, V. Yankov, l. Petrov, I. Shishkova, A. Pavlova, P. Ivanova, A. Surleva, K. Hristov, E. Todorova, A. Obryvalina, R. Telyashev, Fuel Process. Technol. 126 (2014) 332–342. A. Stanislaus, A. Marafi, M.S. Rana, Catal. Today 153 (2010) 1–68. X. Pang, L. Zhang, S. Sun, T. Liu, X. Gao, Catal. Today 125 (2007) 173–177. K. Soni, B.S. Rana, A.K. Sinha, A. Bhaumik, M. Nandi, M. Kumar, G.M. Dhar, Appl. Catal., B: Environ. 90 (2009) 55–63. T. Klimova, J. Reyes, O. Gutiérrez, L. Lizama, Appl. Catal., A: Gen. 335 (2008) 159–171. K. Zhang, Y. Liu, S. Tian, E. Zhao, J. Zhang, C. Liu, Fuel 104 (2013) 201–207. F.M. Zhang, B.S. Liu, Y. Zhang, Y.H. Guo, Z.Y. Wan, F. Subhan, J. Hazard. Mater. 233-234 (2012) 219–227. P. Jabbarnezhad, M. Haghighi, P. Taghavinezhad, Fuel Process. Technol. 126 (2014) 392–401. A. Duan, T. Li, Z. Zhao, B. Liu, X. Zhou, G. Jiang, J. Liu, Y. Wei, H. Pan, Appl. Catal., B: Environ. 165 (2015) 763–773. M. Razavian, S. Fatemi, Microporous Mesoporous Mater. 201 (2015) 176–189.
353
[15] T. Zhang, X. Zhang, X. Yan, L. Lin, H. Liu, J. Qiu, K.L. Yeung, Catal. Today 236 (2014) 41–48. [16] D.D. Anggoro, I. stadi, J. Nat. Gas Chem. 17 (2008) 39–44. [17] A. Gedeon, M. Valeux, M. Gruia, G. Minghua, J. Fraissard, Solid State Nucl. Magn. Reson. 9 (1997) 269–276. [18] H.L. Janardhan, G.V. Shanbhag, A.B. Halgeri, Appl. Catal., A: Gen. 471 (2014) 12–18. [19] A. Satter, G. Parkin, Nature 463 (2010) 523–526. [20] G. Rodriguez-Gattorno, A. Galano, E. Torres-García, Appl. Catal., B: Environ. 92 (2009) 1–8. [21] C. Jiang, J. Wang, S. Wang, H.Y. Guan, X. Wang, M. Huo, Appl. Catal., B: Environ. 106 (2011) 343–349. [22] E.N. Kaya, S. Tuncel, T.V. Basova, H. Banimuslem, A. Hassan, A.G. Gürek, V. Ahsen, M. Durmus, Sens. Actuators, B: Chem. 199 (2014) 277–283. [23] B.N. Achar, P.K. Jayasree, Synth. Met. 114 (2000) 219–224. [24] J. Wei, M. Xu, J. Zhang, R. Zhao, X. Liu, J. Magn. Magn. Mater. 324 (2012) 2696–2700. [25] R. Bayrak, O. Bekircan, M. Durmus, I. Degirmencioglu, J. Organomet. Chem. 767 (2014) 101–107. [26] X. Zhou, J. Li, X. Wang, K. Jin, W. Ma, Fuel Process. Technol. 90 (2009) 317–323. [27] Y. Zhang, D. Wang, R. Zhang, J. Zhao, Y. Zheng, Catal. Commun. 29 (2012) 21–23.