Modification and regeneration of HZSM-5 catalyst in microwave assisted catalytic fast pyrolysis of mushroom waste

Energy Conversion and Management 123 (2016) 29–34

Contents lists available at ScienceDirect

Energy Conversion and Management journal homepage: www.elsevier.com/locate/enconman

Modification and regeneration of HZSM-5 catalyst in microwave assisted catalytic fast pyrolysis of mushroom waste Jia Wang a, Zhaoping Zhong a,⇑, Zuwei Song a, Kuan Ding a, Aidong Deng b a b

Key Laboratory of Energy Thermal Conversion and Control of Ministry of Education, Southeast University, No. 2 Sipailou, Xuanwu District, Nanjing, Jiangsu 210096, China National Engineering Research Center of Turbo-generator Vibration, Southeast University, Nanjing, Jiangsu 210096, China

a r t i c l e

i n f o

Article history: Received 8 April 2016 Received in revised form 7 June 2016 Accepted 8 June 2016

Keywords: HZSM-5 catalyst modification Microwave assisted catalytic fast pyrolysis Regeneration Mushroom waste

a b s t r a c t In this contribution, HZSM-5 zeolites were modified by ethylene diamine tetraacetic acid (EDTA) to selectively eliminate the strong external acid sites, and the modified catalysts were then used to conduct the microwave assisted catalytic fast pyrolysis (MACFP) of mushroom waste (MW). Experiment results showed that the modification of HZSM-5 with EDTA had no significant effect on topological structure, and the surface area and total acid sites decreased while the pore volume increased within the modification time regions. Among the modified catalysts, an EDTA treatment for 2 h (labeled as 2H-Z5) performed prominent promise for removing oxygenated chemicals and promoting the aromatic species as well as inhibiting the formation of coke. Simultaneously, the effects of various regeneration steps for deactivated 2H-Z5 catalysts on products distribution were studied, and the highest relative content of hydrocarbons (19.9%) and the lowest coke yield (3.48%) could be obtained under the third regeneration cycle condition. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction Bio-fuel, achieved from thermal conversion or liquefaction of bioenergy as an abundant and renewable energy source, has attracted more and more attention in the past few years [1,2]. Despite some high value-added chemicals are found in the biooil, which includes hydrocarbons, alcohols, furans, phenols and so forth, its application is still limited as a consequence of the high oxygen content, low caloric value, high corrosion behavior and chemical instability [3,4]. Upgrading of biomass derived bio-oil provides an effective and efficient process to attain an energy benefit with a minimum effect on environment [5]. In this regard, catalytic fast pyrolysis (CFP) of biomass shows great promise for increasing the hydrocarbons and reducing the generation of oxygenated chemical compounds [6,7]. Among various catalysts which have been investigated and tested for the conversion of bio-energy into bio-fuels, HZSM-5 has been proved to be the most efficient for considerably improving the components of bio-oil by both decreasing the oxygenated chemicals and promoting gasoline like constitutes through various reactions, such as dehydration, decarbonylation, decarboxylation and aromatization [8,9]. However, the high production of coke during the CFP process leads to catalyst deactivation and short recycle life. Due to the high price and large amount of HZSM-5 catalysts consumed in the industry, it is worth noting that efficient modifi⇑ Corresponding author. E-mail address: [email protected] (Z. Zhong). http://dx.doi.org/10.1016/j.enconman.2016.06.024 0196-8904/Ó 2016 Elsevier Ltd. All rights reserved.

cation, recycle and reuse of catalysts in a continuous operating system is necessary [10]. In this sense, researches on modification, regeneration and reuse of HZSM-5 zeolites are of fundamental interest to obtain their practical implementation. Previous studies related to the modification and regeneration of catalysts by various techniques were reported intensively [10–14]. For the modification of HZSM-5, Huang et al. [15] modified HZSM-5 zeolites by CuO and MoO3 for the thermal conversion of pine sawdust, and they indicated that Mo(3%)-Cu(3%)/HZSM-5 treatment promoted the highest yield of C6
C12 hydrocarbons. Sheng et al. [16] analyzed the effect of stream treatment on HZSM-5 during ethanol dehydration to ethylene, and they detected that the total amount of acid sites of HZSM-5, especially the strong acid sites, decreased after stream treatment, which led to the improvement of catalytic stability. Li et al. [17] conducted a comparative research on catalytic conversion of methanol and propanal, and it was observed that the post-modification step by Ga(NO3)3 and NH4HF2 had significant effects on structure and acidity of HZSM-5 zeolites, and the catalyst lifetime was dramatically improved. In the sense of catalyst regeneration, Ma and van Bokhoven [12] studied the deactivation and regeneration of H-USY catalysts during the catalytic fast pyrolysis of lignin, and they concluded that the regeneration of H-USY was possible to a large extent, and the products distribution were affected by the regeneration cycles. Lopez et al. [10] used TG/FTIR to research the regeneration of H-USY and HZSM-5 catalysts through the catalytic cracking of polyethylene, and it was found that the volatile matters showed slight variation during the regeneration process.

30

J. Wang et al. / Energy Conversion and Management 123 (2016) 29–34

Note that the strong external acid sites are responsible for the coke formation due to the shape selectivity of HZSM-5 zeolites makes it difficult to form coke in the internal pores, a promising process to lower the strong external acid sites but retain the internal weak sites is to selectively eliminate the strong acid sites by dealumination [18]. For this reason, ethylene diamine tetraacetic acid (EDTA) showed prevalent ability to remove the framework aluminums [19]. Simultaneously, compared to some other pyrolysis heating reactors, microwave assisted catalytic fast pyrolysis (MACFP) dominates a series superiorities for well-distributed of heat, easy operation and energy conservation, which has been extensively used and studied [20–22]. However, at the best of our knowledge, the MACFP of mushroom waste (MW) over EDTA modified HZSM-5 catalysts, and the subsequent regeneration of modified HZSM-5 catalysts, have not been studied and reported yet. In this regard, our objective here is to investigate the effects of various EDTA modifications and regeneration steps of HZSM-5 catalysts on products distribution through MACFP of MW, and thus the optimum EDTA treatment condition and regeneration cycle process will be determined and an upgraded bio-oil production will be obtained. 2. Materials and methods 2.1. Materials Mushroom waste (MW) was collected from a farm in Xuzhou, Jiangsu province, China. Prior to experiment, the MW samples were dried at 105 °C for 24 h in a drying oven, and then the specimens were smashed and grinded. Finally, a 60-mesh sieve was used to screen the feedstock specimens, and after the pretreatment, the MW experimental samples were deposited in a sealed bottle. The component analysis of the MW samples (dry basis) were carried out and the results were shown as follows: 34.5% cellulose, 38.1% hemicellulose, 23.1% lignin and 4.3% extractives. The ultimate analysis of dried MW was conducted and the C, H, O and N were 41.8 wt%, 6.4 wt%, 41.9 wt%, and 4.7 wt%, respectively. Note that the proximate analysis results for air-dried MW could be shown as follows: 5.08 wt% ash, 76.59 wt% volatile, 10.01 wt% fixed carbon, and 8.32 wt% moisture. 2.2. EDTA modification of HZSM-5 EDTA, which was used to conduct the chemical modification of HZSM-5, was achieved from Nanjing Chemical Reagent Co. Ltd. Simultaneously, the HZSM-5 zeolites (SiO2/Al2O3 = 50) were purchased from the Catalyst Plant of Nankai University, Tianjin, China. The EDTA modification procedures could be organized as follows: (i) Deionized water was used to prepare 0.33 mol/L EDTA solution, and a magnetic stirrer at 80 °C water bath was applied to intensively mix the solution and HZSM-5 (the zeolite to solution mass ration was 1:10) for 1 h, 2 h, 3 h and 4 h, respectively. (ii) A Buchner funnel was utilized to conduct vacuum filtration steps of mixture, and a Ph-meter was used to measure the filtrate (until neutral). (iii) The EDTA modified HZSM-5 catalysts were dried in an air-circulating oven at 120 °C for 2 h, and then a muffle was used to calcine the modified HZSM-5 samples for 4 h at 550 °C. Note that in this contribution, the studied modified HZSM-5 catalysts were designated as 1H-Z5, 2H-Z5, 3H-Z5 and 4H-Z5 according to the EDTA modification time. 2.3. Characterization of catalysts XRD patterns were carried out by Bruker D8 Focus using CuKa radiation with a step of 0.02 at a current of 40 mA and a voltage of

40 kV. The scanning range of 2h was from 5° to 50° with a 10 min1 scanning rate. The pore characteristics and surface features of the parent and modified HZSM-5 specimens were conducted by a Belsorp-Max adsorption instrument. BET and Langmuir methods were used to calculate the specific surface area. NH3-TPD was conducted by the FineSorb 3010 automatic temperature programmed chemical adsorption equipment to investigate the acidity performances of various specimens. For a typical run, 100 mg specimens were introduced in a U-shaped quartz cell and the temperature was fixed at 873 K with a He steam for 1 h, and then cooled down to 423 K. Finally, the samples were saturated by some NH3 gas, then the NH3-TPD was conducted with a fixed heating rate of 15 °C/min from 150 to 650 °C. 2.4. Experimental procedure Fig. 1 presented the schematic diagram of MACFP process. Obviously, the MACFP experiments were conducted in a bench scale with a microwave oven (power: 750 W; frequency: 2.45 GHz). A self-designed quartz fixed bed reactor was placed into the microwave oven cavity, and the stream inlet tube was connected to the N2 gas system, and the stream outlet was connected to the condensation system. Furthermore, a K-type thermocouple was used to measure the reaction temperature. It should be noted that due to the strong absorptive ability for microwave, some SiC samples were utilized as microwave absorbent [21,23]. For a typical run, 10 g MW feedstock specimens were introduced onto the surface of SiC bed, and some SiC particles and 10 g HZSM-5 catalysts were mixed uniformly at first, and then the mixtures were placed on the mesh sieve to form a fixed bed for upgrading the primary volatile matters. Prior to each experiment, N2 was purged into the reactor to keep an inert atmosphere with a flow rate of 50 mL/min. After sufficient purging, the power supply was started and the pyrolysis process was proceeded. After reaching the desired reaction temperature of 550 °C, the temperature control system was used to maintain the temperature, and for each experiment, the reaction time was fixed at 20 min. The pyrolytic volatile matters passed through the condensation system, and then the condensable products were collected as bio-oil. The incondensable gas were gathered in a gas collection bag. After each experiment, in order to determine the coke yield, the spent HZSM5 catalysts were collected and then were calcined in a muffle for 4 h at 800 °C, and thus the coke yield could be determined. Furthermore, the carbon yield of bio-oil was calculated by the weight difference of condensation collector before and after each MACFP experiment, and the water content in the liquid products was identified by Karl Fischer Titration. Simultaneously, the char yield was determined by the weight difference of the reactor before and after each experiment. Besides, the carbon yield of gas was calculated by the difference based on the law of mass conservation. 8

9

3 4

10

2

1

5 6 7

Fig. 1. Schematic diagram of MACFP system. (1) N2 gas; (2) pressure regulator; (3) flow meter; (4) microwave oven; (5) quartz glass reactor; (6) HZSM-5 catalysts; (7) pyrolysis feedstock samples; (8) thermocouple (K-type); (9) condensation system; (10) gas collection bag.

31

J. Wang et al. / Energy Conversion and Management 123 (2016) 29–34

2.5. Product analysis

0.24 345 0.22 335 330 0.20 325 320 0.18 315 310

Total pore volume (cm 3/g)

340

BET surface area (m2 /g)

A gas chromatography/mass spectrometry (GC/MS: 7890A/5975C, Agilent) was utilized to investigate the obtained chemical constitutes of bio-fuel. The interface temperature of GC/MS was set at 300 °C, and the injector temperature was controlled at 275 °C. Note that the GC separation was carried out by a capillary column called HP-5MS, and the split ratio was 1:80. The National Institute of Standards of Technology (NIST) was applied to analyze and identify the chromatographic peaks. It should be noted that during the GC/MS experiments, the pyrolytic organic volatile matters could not be collected, and thus the total organic products seemed uncertain. Nevertheless, the overall mass of feedstock specimens (MW and HZSM-5 zeolites) were fixed at the same value (20 g) in each experiment, and therefore, the variations of pyrolytic products could be identified by the comparison of peak areas and the relative content of organic products could be attained by calculating the chromatographic area.

0.16 305 0

1

2

3

4

Modification time (h)

(a) BET surface area and total pore volume

3. Results and discussion 3.1. Physicochemical characteristics of fresh and EDTA modified HZSM-5 catalysts The porosity properties of fresh and EDTA modified HZSM-5 catalysts were investigated by BET method (N2 adsorption), and the results were performed in Fig. 2a. As could be seen, the BET surface area decreased gradually with the modification time increasing, while the total pore volume tended to increase within the modification time regions. The underlying reason for this phenomenon was that during the EDTA modification process, firstly, the aluminiumoxygen tetrahedron performed hydrolysis reaction to generate aluminium hydroxide through reaction (1) as shown below, and then the aluminium hydroxide reacted with the proton acid centers to form Al(OH)+2 and H2O by reaction (2). Finally, the Al(OH)+2 cation was exchanged by Na+ and then conducted chelation reaction with EDTA through reaction (3) to remove the external framework aluminums, which contributed to the formation of hydroxy holes and increment of pore size, and thus resulted in the increase in total pore volume. Simultaneously, the formation of hydroxyl holes and increment of pore size led to the loss of original micropores, which were responsible for the decrease of total specific surface area.

Acid sites (mmol NH3 /g)

1.6

1.4

Strong acid sites Weak acid sites Total acid sites

1.2

1.0

0.8

0.6 0

1

2

3

4

Modification time (h)

(b) Acid sites

4H-Z5

Intensity

3H-Z5 2H-Z5 1H-Z5

ð1Þ Fresh HZSM-5

0

10

20

30

40

50

2 theta (°)

ð2Þ

(c) XRD patterns of modified and fresh HZSM-5 catalysts Fig. 2. Characterization results of fresh and EDTA-modified HZSM-5 catalyst.

ð3Þ

32

J. Wang et al. / Energy Conversion and Management 123 (2016) 29–34

Furthermore, Fig. 2b gave the acidity properties of parent and modified HZSM-5 catalysts, which were analyzed by ammonia TPD. Obviously, both the strong acid sites (>300 °C) and weak acid sites (<300 °C) of modified HZSM-5 samples decreased with the increment of modification time, and the decrease rate of strong acid sites was higher than that of weak acid sites, which reflected that the dealumination process had significantly higher selectivities toward strong acid sites and lower selectivities toward weak acid sites. In order to investigate whether the EDTA modification had significant effects on topological structure of HZSM-5, the X-ray diffraction (XRD) patterns of fresh and EDTA modified catalysts were conducted and the results were presented in Fig. 2c. As could be indicated, the dealumination process did not break the crystalline structure because all the samples performed typical peaks and kept the MFI topological shape. Besides, the relative crystallinity was analyzed and the results could be shown as follows: 100% at HZSM-5; 102.4% at 1H-Z5; 105.3% at 2H-Z5; 104.2% at 3H-Z5; and 103.5% at 4H-Z5. This was due to that the proton ions (H+) inside HZSM-5 catalyst would be exchanged with sodium ions during the EDTA modification process, and thus the free proton ions dissolved extra-framework aluminums to cause the subtle variation of relative crystallinity. 3.2. Effect of EDTA modification on products distribution In this section, experiments for MACFP of MW over parent and modified HZSM-5 catalysts were conducted to study the effects of different modification conditions on products distribution, and the results were performed in Fig. 3. Note that some additional experiments for microwave assisted non-catalytic fast pyrolysis of MW at the reaction temperature of 550 °C were carried out to investigate the effect of catalyst on products distribution, and the results were also shown in Fig. 3. It could be observed that compared to the non-catalytic pyrolysis, the use of HZSM-5 catalyst produced a decrease in liquid products and a significant increase in gases and water. This is as a function of both the acid sites and textural properties of HZSM-5. In practical terms, the primary organic vapors first underwent dehydration, decarboxylation, and decarbonylation chemical reactions to generate CO2, CO, and H2O during the MACFP process, which were reflected in both the yields of gas and water [24]. For the EDTA modified HZSM-5 catalysts, as indicated in Fig. 3, the oil fraction decreased at first and then increased with modifica-

tion time increasing from 1 h to 4 h. However, the gas fraction and water showed the opposite trend, and the gas fraction reached a maximum yield of 22.7 wt% as the modification time was selected at 1 h (i.e., 1H-Z5), while the maximum yield of water (14.5 wt%) emerged at 2H-Z5. Besides, the char fraction tended to be practically equivalent (almost 15 wt%) in all the experiments, which was in accordance with char formation during pyrolysis. Simultaneously, the coke yield decreased at first and then slightly increased with a minimum yield of 3.36% at 2H-Z5 (as shown in Table 1). It could be concluded that the EDTA modified HZSM-5 showed significant promise for upgrading the quality of bio-oil and inhibiting the formation of coke. During the EDTA modification process, the external framework aluminums would be removed, which led to the decrease of the external strong acid sites. Note that in MACFP process, the internal acid sites in the pores dominated the catalytic properties to generate target products like olefins and aromatics, while the strong external acid sites were responsible for coke formation. The accumulated coke on the external surface of catalyst caused blockage of pore opening, which resulted in rapid deactivation during the biomass thermal conversion. In this regard, the EDTA modification method could give a decrease in external strong acid sites, which made a significant contribution to enhancing the stability of catalysts, retarding the catalyst deactivation, and improving the catalyst life to facilitate the subsequent cracking conversion reactions.

3.3. Chemical constitutes in oil fraction Note that chemical constituents in the liquid products detected and analyzed by GC/MS could be summarized as follows: acids, ketones, hydrocarbons, sugars, alcohols, esters, furans and phenols. Fig. 4 presented the specific products distribution of oil fraction. As described in Fig. 4, acids were found to be the main products in the liquid fraction in all the cases, and followed by hydrocarbons. The relative contents of acids tended to decrease at first and then

Table 1 Coke yield of fresh and modified HZSM-5 catalysts. Catalysts

Fresh HZSM-5

1H-Z5

2H-Z5

3H-Z5

4H-Z5

Coke yield (wt%)

4.02

3.83

3.36

3.45

3.56

50

60

Yield (%)

40

30

Relative content (%)

No catalyst Fresh HZSM-5 1H-Z5 2H-Z5 3H-Z5 4H-Z5

50

40

30

No catalyst Fresh HZSM-5 1H-Z5 2H-Z5 3H-Z5 4H-Z5

20

20 10 10 0 0 Oil

Char

Gas

Water

Fig. 3. Pyrolytic products obtained from thermal and catalytic experiments.

s ls ns ds es gar lcoho Aci ton carbo Su Ke A o r d Hy

ls ers urans eno Est F Ph

Fig. 4. Products distribution of oil fraction.

33

J. Wang et al. / Energy Conversion and Management 123 (2016) 29–34

increase as the modification time increased from 1 h to 4 h. Besides, the relative content of furans showed the similar trend with a minimum relative content of 4.7% at 3H-Z5. Simultaneously, the fresh and modified HZSM-5 catalysts had comparative remarkable higher selectivities toward sugars and alcohols as the relative contents of them decreased or increased conspicuously, and lower selectivities toward low molecular oxygenated chemical compounds such as ketones, esters and phenols for their slight variations in relative contents. It was important to point out that the relative contents of hydrocarbons were rarely low during the microwave assisted non-catalytic fast pyrolysis of MW, however, a positive increase in hydrocarbons could be realized by the application of parent and different EDTA modified HZSM-5 catalysts, and the modification time of 2 h (i.e., 2H-Z5) was found to be the optimal condition as the most prominent growth of hydrocarbons could be attained. The underlying reason was that, as mentioned before, the unique molar structure and shape selectivity of HZSM-5 played a determining role in increasing the selectivity to fuel range hydrocarbons through various chemical reactions including isomerization, oligomerization, and aromatization [25], which could be reflected in Fig. 5 as discussed below. Fig. 5 proposed the relative contents of aromatics with different modification times. As presented in Fig. 5, the relative contents of BTX aromatics (i.e., benzene, toluene, and xylene) performed deviation as the modification time exceeded the certain limitation. For example, the relative content of toluene was 6.2% as the modification time was fixed at 0 (i.e., parent HZSM-5) but increased to 9.9% when the modification time was raised to 2 h. This is as a consequence of the elimination of strong external acid sites by EDTA modification method. However, further increased in modification time to 3 h and 4 h tended to decrease the yield to 8.1% and 7.1%, respectively. The underlying reason for this phenomenon might lie in the removal of some necessary acid sites as modification time beyond 2 h, which suppressed the catalytic conversion [26].

ducted in a muffle furnace and the condition was controlled at 550 °C for 4 h. Secondly, the regenerated 2H-Z5 modified catalysts were applied to investigate the effects of various regeneration steps of catalysts on products distribution through MACFP of MW at 550 °C, and the products distribution of bio-oil were proposed in Fig. 6. Obviously, the products distribution were dominated by acids, which tended to decrease at first and then increase with a minimum yield of 38.7% at the third regeneration cycle. However, the carbon yield of alcohols performed an opposite trend with an optimal value appearing at the same condition. Furthermore, the variations for the yields of sugars and esters were negligible. For the yield of hydrocarbons, the relative content slightly increased from 17.3% to 19.9% as the regeneration steps increased from 1 to 3, however, the yields of furans and phenols were inclined to present an adverse tendency. This was attributed to both the decreases of strong and weak acid sites as shown in Table 2. Compared to the characterization results of the original 2H-Z5, the BET surface area, pore volume and the total number of acid sites decreased as the regeneration steps increased to 3, which might favor the formation of hydrocarbons derived from furans and phenols through depolymerization. It was worth noting that further increased in regeneration cycles led to a dramatic decrease in hydrocarbons and a slight increase in furans and phenols, this might due to some necessary acid sites, especially the weak acid sites, would be eliminated by further regeneration, which led to a decrease in the carbon yield of hydrocarbons and an increase in the carbon yields of phenols and furans. Therefore, it could be concluded that a moderate regeneration step would

56

Acids Sugars Furans

Ketones Alcohols Phenols

Hydrocarbons Esters

As previously mentioned, the optimum yield of target products could be achieved at 2H-Z5, and in this section, first of all, the regeneration of deactivated 2H-Z5 modified catalysts was con-

Relative content (%)

42

3.4. Effects of regeneration steps on products distribution

28

14 12

Benzene Xylene

Toluene Other aromatics 0 1

Relative contents (%)

9

2

3

4

5

Regeneration steps Fig. 6. Products distribution of bio-oil with different regeneration steps.

6 Table 2 Physical and chemical characteristics of the catalysts after regeneration at 550 °C.

3

0 0

1

2

3

Modification time (h) Fig. 5. Relative contents of aromatics.

4

Regen. Step

BET surface area (m2/g)

Total pore volume (m3/g)

Weak acid sites (mmol/g)

Strong acid sites (mmol/g)

Total acid sites (mmol/g)

1 2 3 4 5

315 301 283 272 259

0.201 0.186 0.175 0.168 0.159

0.41 0.40 0.39 0.33 0.31

0.58 0.54 0.49 0.46 0.45

0.99 0.94 0.88 0.79 0.76

34

J. Wang et al. / Energy Conversion and Management 123 (2016) 29–34

Table 3 Desired products and coke yields. Regen. Step

1 2 3 4 5

Appendix A. Supplementary material

Desired products (%)

Coke yields (%)

HZSM-5

2H-Z5

HZSM-5

2H-Z5

45.65 46.45 44.76 42.80 39.14

46.07 46.96 47.78 43.67 41.36

3.94 3.71 3.91 4.58 4.92

3.85 3.61 3.48 4.35 4.81

encourage the yield of target products, and thus an upgraded biooil production could be obtained. Note that some additional experiments for the regeneration of parent HZSM-5 catalysts were conducted to investigate the effects of repetition steps on different catalysts during the MACFP of MW. The desired products and coke yields were shown in Table 3. Here, the desirables were defined as those organic pyrolytic products which facilitated the heating value, stability, combustion characteristic, and so forth. For example, hydrocarbons and alcohols were desirables as they could upgrade the quality of bio-oil to gasoline fuels. However, acids were undesirables as the high content of acids in bio-oil would cause corrosion and made it practically difficult for being introduced into energies as transportation fuel. It could be observed from Table 3 that, both the desired products obtained from the regeneration of HZSM-5 and 2H-Z5 increased at first and then decreased with the repetition cycles increasing from 1 to 5, and for 2H-Z5, the maximum yield of desirables was 47.78% occurring at the third regeneration step, which was 1.33% higher than that of HZSM-5 occurring at the second regeneration step. Besides, the coke yield tended to decrease at first and then increase within the studied regeneration step regions, and the minimum coke yield for 2H-Z5 catalyst was 3.48%, which was 0.23% lower than that for HZSM-5.

4. Conclusions The modification and regeneration of HZSM-5 catalysts provided an efficient process to remove the oxygenated chemicals and facilitate the production of hydrocarbons. Some experiments for MACFP of MW over EDTA modified HZSM-5 zeolites were conducted. The modification treatment made no difference in topological structure, but led to an increase in pore volume and a decrease in both the total acid sites and surface area. The carbon yields of BTX tended to increase at first and then decrease with modification time varying from 1 h to 4 h, while the coke yield decreased at first and then slightly increased under the same condition. The 2H-Z5 catalyst showed the most excellent performance for the promotion of hydrocarbons and the removal of oxygenated compounds. Simultaneously, the effect of regeneration cycles of deactivated 2H-Z5 catalysts on products distribution through MACFP of MW were also investigated. Note that the regeneration steps had a remarkable impact on products distribution as the third regeneration cycle attributed to the highest yield of hydrocarbons, as well as the lowest coke yield.

Acknowledgements The authors are grateful for the financial support from the National Natural Science Fund Program of China (51276040 & U1361115), the National Key Basic Research Program of China (973 Program) (No. 2013CB228106).

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.enconman.2016. 06.024. References [1] Abnisa F, Wan Daud WMA. A review on co-pyrolysis of biomass: an optional technique to obtain a high-grade pyrolysis oil. Energy Convers Manage 2014;87:71–85. [2] Lehto J, Oasmaa A, Solantausta Y, Kytö M, Chiaramonti D. Review of fuel oil quality and combustion of fast pyrolysis bio-oils from lignocellulosic biomass. Appl Energy 2014;116:178–90. [3] Oasmaa A, van de Beld B, Saari P, Elliott DC, Solantausta Y. Norms, standards, and legislation for fast pyrolysis bio-oils from lignocellulosic biomass. Energy Fuels 2015;29:2471–84. [4] Oasmaa A, Sundqvist T, Kuoppala E, Garcia-Perez M, Solantausta Y, Lindfors C, et al. Controlling the phase stability of biomass fast pyrolysis bio-oils. Energy Fuels 2015;29:4373–81. [5] Abnisa F, Daud WMAW, Sahu JN. Pyrolysis of mixtures of palm shell and polystyrene: an optional method to produce a high-grade of pyrolysis oil. Environ Prog Sustain Energy 2014;33:1026–33. [6] Du Z, Hu B, Ma X, Cheng Y, Liu Y, Lin X, et al. Catalytic pyrolysis of microalgae and their three major components: carbohydrates, proteins, and lipids. Bioresour Technol 2013;130:777–82. [7] Mao Y, Dong L, Dong Y, Liu W, Chang J, Yang S, et al. Fast co-pyrolysis of biomass and lignite in a micro fluidized bed reactor analyzer. Bioresour Technol 2015;181:155–62. [8] Zhang H, Luo M, Xiao R, Shao S, Jin B, Xiao G, et al. Catalytic conversion of biomass pyrolysis-derived compounds with chemical liquid deposition (CLD) modified ZSM-5. Bioresour Technol 2014;155:57–62. [9] Zhang W, Yuan C, Xu J, Yang X. Beneficial synergetic effect on gas production during co-pyrolysis of sewage sludge and biomass in a vacuum reactor. Bioresour Technol 2015;183:255–8. [10] Lopez A, de Marco I, Caballero BM, Adrados A, Laresgoiti MF. Deactivation and regeneration of ZSM-5 zeolite in catalytic pyrolysis of plastic wastes. Waste Manage 2011;31:1852–8. [11] Kim YH, Lee KH, Lee JS. The effect of pre-coking and regeneration on the activity and stability of Zn/ZSM-5 in aromatization of 2-methyl-2-butene. Catal Today 2011;178:72–8. [12] Ma Z, van Bokhoven JA. Deactivation and regeneration of H-USY zeolite during lignin catalytic fast pyrolysis. ChemCatChem 2012;4:2036–44. [13] Yildiz G, Lathouwers T, Toraman HE, van Geem KM, Marin GB, Ronsse F, et al. Catalytic fast pyrolysis of pine wood: effect of successive catalyst regeneration. Energy Fuels 2014;28:4560–72. [14] Zhang H, Zheng J, Xiao R, Shen D, Jin B, Xiao G, et al. Co-catalytic pyrolysis of biomass and waste triglyceride seed oil in a novel fluidized bed reactor to produce olefins and aromatics integrated with self-heating and catalyst regeneration processes. RSC Adv 2013;3:5769. [15] Huang Y, Wei L, Crandall Z, Julson J, Gu Z. Combining Mo–Cu/HZSM-5 with a two-stage catalytic pyrolysis system for pine sawdust thermal conversion. Fuel 2015;150:656–63. [16] Sheng Q, Ling K, Li Z, Zhao L. Effect of steam treatment on catalytic performance of HZSM-5 catalyst for ethanol dehydration to ethylene. Fuel Process Technol 2013;110:73–8. [17] Li M, Zhou Y, Oduro IN, Fang Y. Comparative study on the catalytic conversion of methanol and propanal over Ga/ZSM-5. Fuel 2016;168:68–75. [18] Galadima A, Muraza O. Zeolite catalysts in upgrading of bioethanol to fuels range hydrocarbons: a review. J Ind Eng Chem 2015;31:1–14. [19] Zhang B, Zhong Z, Song Z, Ding K, Chen P, Ruan R. Optimizing anti-coking abilities of zeolites by ethylene diamine tetraacetic acid modification on catalytic fast pyrolysis of corn stalk. J Power Sources 2015;300:87–94. [20] Borges FC, Xie Q, Min M, Muniz LA, Farenzena M, Trierweiler JO, et al. Fast microwave-assisted pyrolysis of microalgae using microwave absorbent and HZSM-5 catalyst. Bioresour Technol 2014;166:518–26. [21] Ganesapillai M, Manara P, Zabaniotou A. Effect of microwave pretreatment on pyrolysis of crude glycerol–olive kernel alternative fuels. Energy Convers Manage 2016;110:287–95. [22] Guo X, Ren J, Xie C, Lin J, Li Z. A comparison study on the deoxygenation of coal mine methane over coal gangue and coke under microwave heating conditions. Energy Convers Manage 2015;100:45–55. [23] Mushtaq F, Mat R, Ani FN. Fuel production from microwave assisted pyrolysis of coal with carbon surfaces. Energy Convers Manage 2016;110:142–53. [24] Önal E, Uzun BB, Pütün AE. Bio-oil production via co-pyrolysis of almond shell as biomass and high density polyethylene. Energy Convers Manage 2014;78:704–10. [25] Zhang B, Zhong Z, Xie Q, Chen P, Ruan R. Reducing coke formation in the catalytic fast pyrolysis of bio-derived furan with surface modified HZSM-5 catalysts. RSC Adv 2015;5:56286–92. [26] Shao S, Zhang H, Heng L, Luo M, Xiao R, Shen D. Catalytic conversion of biomass derivates over acid dealuminated ZSM-5. Ind Eng Chem Res 2014;53:15871–8.