- Email: [email protected]
Catalytic fast pyrolysis of mushroom waste to upgraded bio-oil products via pre-coked modified HZSM-5 catalyst
Bioresource Technology 212 (2016) 6–10
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Catalytic fast pyrolysis of mushroom waste to upgraded bio-oil products via pre-coked modified HZSM-5 catalyst Jia Wang, Zhaoping Zhong ⇑, Kuan Ding, Zeyu Xue Key Laboratory of Energy Thermal Conversion and Control of Ministry of Education, Southeast University, No. 2 Sipailou, Xuanwu District, Nanjing, Jiangsu 210096, China
h i g h l i g h t s HZSM-5 catalyst was modified with pre-coked by methanol to olefins reactions. Modified HZSM-5 catalyst contributed to the carbon yields of hydrocarbons. Coke yields decreased gradually with increasing of pre-coked percentage. Obtained oxygenates decreased at first and then increased at studied pre-coked percentage ranges.
a r t i c l e
i n f o
Article history: Received 18 February 2016 Received in revised form 31 March 2016 Accepted 2 April 2016 Available online 6 April 2016 Keywords: Catalytic fast pyrolysis Pre-coked Modified HZSM-5 catalyst Mushroom waste
a b s t r a c t In this paper, HZSM-5 catalyst was modified by pre-coked to cover the strong external acid sites by methanol to olefins reaction, and the modified catalysts were then applied to conduct the catalyst fast pyrolysis of mushroom waste for upgraded bio-fuel production. Experiment results showed that the strong external acid sites and specific surface area decreased with pre-coked percentage increasing from 0% to 5.4%. Carbon yields of hydrocarbons increased at first and then decreased with a maximum value of 53.47%. While the obtained oxygenates presented an opposite variation tendency, and the minimum values could be reached when pre-coked percentage was 2.7%. Among the achieved hydrocarbons, toluene and p-xylene were found to be the main products, and the selectivity of p-xylene increased at first and then decreased with a maximum value of 34.22% when the pre-coked percentage was 1.3%, and the selectivity of toluene showed the opposite tendency with a minimum value of 25.47%. Ó 2016 Published by Elsevier Ltd.
1. Introduction Concerning the quite heavy dependence on fossil fuels and environmental impacts of greenhouse gas emissions associated with massive consumption of fossil energy, some efforts are currently made to find or invent renewable and alternative resources which are environmentally friendly and with a high reliability and stability (Abnisa and Wan, 2014; Wang et al., 2013b; Zhang et al., 2015b). In this regard, bioenergy is now accepted as a promising alternative that can attribute to meeting the fossil energy needs and enhancing the environmental problems (Akhtar and Saidina, 2012). Among the utilizations of bioenergy, catalytic fast pyrolysis (CFP) is particularly interesting and technically reliable, as well as economically feasible for converting bioenergy into bio-fuel products (Liu et al., 2013; Lu et al., 2012). This technology has attracted increasing attentions as its role in producing the highest bio-fuel
⇑ Corresponding author. E-mail address: [email protected] (Z. Zhong). http://dx.doi.org/10.1016/j.biortech.2016.04.005 0960-8524/Ó 2016 Published by Elsevier Ltd.
yields (75 wt.%) under conditions of moderate temperature (500–600 °C) and short pyrolytic vapor residence time (1 s) (Borges et al., 2014). In this sense, CFP can be selected as a recommend approach to seek the stated objective of biomass application. Among various researched catalysts used in CFP process, HZSM5 zeolite dominated the pronounced performance for shape selectivity and liquid products deoxygenation (Tamiyakul et al., 2015; Brebu et al., 2010; Zhang et al., 2014). The 3-dimensional intersecting pore structure of HZSM-5 is characterized by well-defined 10membered rings with straight channels of 0.53 0.56 nm and sinusoidal channels of 0.51 0.55 nm, in which oxygenated organic compounds will be removed and hydrocarbons will be formed when primary pyrolytic volatile matters passed through the catalyst layer (Zhang et al., 2015a). However, it is important to note that there are several critical problems associated with HZSM-5 catalyst, such as the high coke production and short lifecycle. Coke is a kind of large molecule aromatic compound generated and accumulated on the surface of HZSM-5 catalyst, which leads to the blockage of the pore opening mouth and thus a
J. Wang et al. / Bioresource Technology 212 (2016) 6–10
deactivation of catalyst will be caused (Bertero and Sedran, 2016). Note that the strong external acid sites of HZSM-5 are responsible for coke formation, and a promising measure for decreasing the coke yield is to lower the strong external acid sites but meanwhile retain the internal acid sites (Zhang et al., 2014). In this sense, precoked treatment showed great promise for eliminating the strong external acid sites and improving the stability and activity of HZSM-5 zeolites (Chia and Trimm, 2005; Kim et al., 2011). Al-Khattaf (2007) found that the controlled pre-coking process, which could be obtained by exposing the zeolite to organic compounds at moderate temperature, showed significant effect for the enhancement of p-xylene selectivity during toluene disproportionation. Bauer et al. (2007) also revealed that in terms of inactivation of external surface acidity, specimens modified by pre-coking were proved to be more effective compared to those by one-cycle silica deposition. However, at the best of our knowledge, the effect of pre-coked percentage of HZSM-5 on products distribution achieved from CFP of MW has not been investigated and analyzed yet. Mushroom waste (MW) is a representative kind of biomass energy, which consists of wood chip, cottonseed hull and corn cob. A pronounced way to utilized it is catalytic fast pyrolysis, which can convert MW into fuels or other high value-add chemicals. In this contribution, experiments on selectively eliminating the external strong acid sites were conducted by pre-coked of HZSM-5 through methanol to olefins (MTO) reaction. In contrast with fresh HZSM-5 catalyst, the pre-coked catalyst already has some cokes deposited on the external surface, and thus the strong external acidity and activity will be reduced. Subsequently, the modified catalysts with different pre-coked percentages were used to conduct the CFP of MW using Py-GC/MS. The effects of precoked percentages on products distribution of CFP of MW were investigated. The rest of this paper can be organized as follows: in Section 2 the experimental materials and set up are introduced; in Section 3 the effects of pre-coked percentages on products distribution, such as hydrocarbons, phenols, alcohols and so forth, as well as the selectivities of aromatics, were analyzed and discussed in detail; and in Section 4, the conclusions are drawn. 2. Materials and methods
7
methanol is added (Wang et al., 2013a). The fresh HZSM-5 catalyst (SiO2/Al2O3 = 50) was pre-coked by MTO reaction, and the treatment process was illustrated as Fig. 1. An injection pump and heat tape were used to generate methanol vapors and nitrogen was applied as carrier gas with a constant flow rate of 300 ml/min. Meanwhile, in order to obtain a high contact efficiency of methanol vapors and HZSM-5 catalyst, a vertical tube furnace was utilized to conduct the pre-coked experiments. The reaction temperature was controlled by temperature controller system, and in this experiment, the temperature was kept at 450 °C. When the methanol vapors were introduced into the reactor, they passed through the HZSM-5 catalyst layer, resulting the MTO reactions and thus, some cokes would be formed on the catalyst. Posterior to each experiment, some pre-coked HZSM-5 catalysts were combusted at 800 °C for 2 h in a muffle furnace, and the coke yield was calculated by the weight difference. Here, a parameter called pre-coked percentage was used to evaluate the pre-coked degree, and it was defined as the total mass of coke to HZSM-5 catalyst. In this paper, HZSM-5 catalysts with different pre-coked percentages (1.3%, 2.7%, 3.8%, 4.5% and 5.4%) were analyzed and studied, and they were achieved by adjusting and varying the MTO reaction time. Besides, the HZSM-5 catalysts used in the experiments were bought from the Catalyst Plant of Nankai University, Tianjin, China. 2.3. Experimental methods of CFP A CDS Pyroprobe 5200 pyrolyzer was used to conduct the CFP experiments. The probe is a quartz tube, open at both ends, and heated by platinum resistance. Before the experiment, some packed quartz wool, 0.50 mg HZSM-5 catalysts and 0.50 mg feedstock specimens (MW) were put in the quartz in sequence. Besides, the pyrolysis temperature was kept for 20 s with a heating rate of 20 °C/ms, the helium with a high-purity (99.999%) purchased from Nanjing Maikesi Nanfen Special Gas Co., Ltd., which was utilized as carrier gas at a stable flow of 1.0 mL/min. The pyrolysis vapors passed through the quartz tube into a gas chromatography/mass spectrometry (7890A/5975C, Agilent) via the helium sweeper gas stream. The interface and injector temperature of GC/MS was 300 °C and 275 °C, respectively. A capillary called HP-5MS with a size of 0.25 mm 0.25um 30 m was used to conduct GC separation. The split ratio was set at 1:80, and the
2.1. Materials Mushroom waste (MW) was obtained from a farm in Xuzhou, Jiangsu province, China. Before the experiment, some pretreatments of mushroom waste were carried out, which could be shown as follows: (1) Dried at 105 °C for 24 h in a drying oven; (2) Pulverized and grinded mechanically; (3) Sifted through a 60-mesh sieve and then kept in a sealed bottle. The ultimate analysis of dried MW was conducted and the results were as follows: 41.8 wt.% carbon, 6.4 wt.% hydrogen, 41.9 wt.% oxygen, and 4.7 wt.% nitrogen. Simultaneously, the proximate analysis of air-dried MW was also carried out, and the results were: 8.32 wt.% moisture, 76.59 wt.% volatile, 5.08 wt.% ash, and 10.01 wt.% fixed carbon. 2.2. Pre-coked treatment process of HZSM-5 It should be noted that more than 20 reaction mechanisms of methanol to olefins (MTO) have been proposed by different researchers, and among them, a hydrocarbon pool mechanism, which was firstly presented by Kolboe and co-workers, have attracted much more attention (Li et al., 2011). The hydrocarbon pool mechanism reveals that some certain organic active centers known as hydrocarbon pool intermediates act as co-catalysts within the cages of zeolites to which olefins are eliminated and
Fig. 1. Pre-coked treatment process of HZSM-5 catalyst by MTO reaction: 1 – N2 gas cylinder; 2 – Flowmeter; 3 – Injection pump (methanol); 4 – Heat tape; 5 – Vertical tube furnace; 6 – HZSM-5 catalyst; 7 – Temperature controller and thermocouple; 8 – Heat exchanger; 9 – Gas-washing bottle.
J. Wang et al. / Bioresource Technology 212 (2016) 6–10
2.4. Product analysis Note that during the Py-GC/MS experiments, organic vapor pyrolytic products could not be collected, thus the yields of total condensable organic products seemed uncertain. However, the total mass of specimens and HZSM-5 catalysts were fixed in each experiment, thus the changes of organic products could be discriminated with analysis and identification of overall peak area. Besides, in order to attain the relative contents of pyrolytic volatile matters, a semi-quantitative method via the calculation of chromatographic area was utilized. 2.5. Evaluation indicator In this paper, the selectivity of aromatics was defined, which can be shown as follows:
Paro Saro ¼ P Paro
ð1Þ
where Saro and Paro are the selectivity and peak area of a certain kind of aromatics, respectively. RParo is the total peak area of aromatics. 3. Results and discussion 3.1. Physicochemical characteristics of catalysts
0.9
345
0.8
330
0.7
315
0.6
300
285
Strong acid sites Weak acid sites
0.5
270
0.4
BET specific surface area (m2/g)
Acid sites (mmol NH3/g)
The porosity and acidity performances of fresh and pre-coked modified HZSM-5 catalysts were given in Fig. 2. The porosity char-
acteristics were conducted by BET method with N2 porosimetry, and the acidity performances were tested and analyzed by ammonia TPD. As presented in Fig. 2, both the acid sites (strong acid sites and weak acid sites) decreased gradually as pre-coked percentage increased from 0% to 5.4%, and the specific surface area tended to present the similar variation trend. 3.2. Pyrolytic products distribution Prior to the experiment, in order to analyze the effect of catalyst on fast pyrolysis of biomass, some additional experiments on noncatalytic fast pyrolysis (non-CFP) of MW at the temperature of 600 °C were carried out. Experiment results indicated that the total peak area from CFP of MW with fresh HZSM-5 was 2.28 109, while the peak area of non-CFP of MW was 3.7 109. Furthermore, the pyrolytic volatile matters for non-CFP of MW consisted of various oxygenates, and the production of hydrocarbons and aromatics were limited. However, the hydrocarbons and aromatics (e.g., benzene, toluene, xylene, naphthalene, ethylbenzene, propane, indene, 1 h-indene, 3-methyl-) achieved from CFP of MW were very abundant. It could be concluded that an addition of HZSM-5 catalyst to fast pyrolysis of MW would result in a remarkable increase in carbon yields of aromatics. The underlying reason was that, as previously mentioned, because of the unique molecular structure of HZSM-5 catalyst, some target aromatics could be generated by multiple reactions (such as decarboxylation, aromatization, oligomerization, isomerization.) when primary pyrolytic vapor products passed through the internal pores (Du et al., 2013; Valle et al., 2010). Fig. 3 proposed the total peak area of pyrolytic organic products obtained from CFP of MW over HZSM-5 catalysts with different pre-coked percentages. It could be observed that the total peak areas decreased at first and then increased as pre-coked percentages of HZSM-5 changed from 0% to 5.4%, and the minimum value would be obtained when pre-coked percentage was 2.7%. This was due to that, for parent HZSM-5, the strong external acidity limited the catalytic pyrolysis reactions for coke could be easily generated on the surface, and thus led to the pore blockage and catalyst inactivation. However, when the appropriate treatment process of precoked was conducted, some of the external strong acid sites would be covered and the optimization of external acidity and activity could be promoted, which resulted in a decrease in total peak area and an increasing quality in obtained organic desired products. Meanwhile, it was necessary to note that the pre-coked treatment of HZSM-5 catalyst might cause both positive and negative effects
3.3
5.0
3.0
4.5
2.7
4.0
2.4 3.5 2.1 3.0 1.8 2.5 1.5
0
1
2
3
4
5
6
0
1
2
3
4
Pre-coked percentage / %
Pre-coked percentage / %
Fig. 2. Characterization results of fresh and pre-coked modified HZSM-5 catalysts.
Fig. 3. Total peak area and coke yield.
5
6
Coke yield / %
ionization energy of MS was 70 eV. In order to reduce experimental errors, experiments were carried out in triplicate, and the average values were reported. It should be noted that in order to determine the moderate temperature for CFP of MW over pre-coked modified catalysts, we conducted several experiments on CFP of MW over fresh HZSM-5 catalyst with reaction temperature increasing from 450 to 700 °C. The results showed that and the relative contents of hydrocarbons tended to increase at first and then decrease with a maximum yield of 44.52% occurring at 600 °C (total relative contents of hydrocarbons: 34.83% at 450 °C, 38.53% at 500 °C, 41.67% at 550 °C, 44.52% at 600 °C, 41.03% at 650 °C and 39.64% at 700 °C). Simultaneously, in this contribution, we focus on aromatic hydrocarbons more than other organics due to the unique effect of HZSM-5 for aromatization reactions during the CFP of MW. For this reason, we selected 600 °C as an optimal reaction condition.
Total peak area (x10 9)
8
9
J. Wang et al. / Bioresource Technology 212 (2016) 6–10
Non-catalyst Fresh HZSM-5 1.3%-HZSM-5 2.7%-HZSM-5 3.8%-HZSM-5 4.5%-HZSM-5 5.4%-HZSM-5
Carbon yield (%)
40
30
10
0 s ls rna eno Fu Ph
Fig. 4. Products distribution.
s gar Su
Relative content / %
12
9
6
3
0 Non-catalyst
0
4 1 3 2 Pre-coked percentage / %
5
6
5
6
Fig. 5. Relative contents of hydrocarbons.
40
Benzene Naphthalene
Toluene p-Xylene Other aromatics
35 30 25 20 15 10
Non-catalyst
0
4 1 2 3 Pre-coked percentage / %
Fig. 6. Selectivities of aromatics.
3.3. Carbon yields of hydrocarbons and selectivities of aromatics
20
s e ls xid bon oho dio car Alc n o r o d rb Hy Ca
15
Benzene Toluene p-Xylene Naphthalene Other aromatics Butane Other hydrocarbons
products and coke, and the strong external acidity was attributed to the high yield of coke. In this regard, in order to reduce the strong external acidity of HZSM-5, it seems that appropriate precoked of HZSM-5 can be a good process to decrease the coke yield and provide a significant contribution to the further shapeselective reactions. However, it should be noted that too high pre-coked percentages (from 2.7% to 5.4%) were responsible for the pore opening blockage, and thus an obstruction of catalytic pyrolysis reaction could be caused.
60
50
18
Selectivity of aromatics / %
on products distribution during CFP of MW, and the negative effect might lie with pre-coked process could block or inactive the internal effective weak acid sites, which might confine the molecules of similar sizes to diffuse in the catalyst pores, and thus resulted in a decrease in catalytic performance of HZSM-5 during the CFP of MW. That was the reason for the increase of total peak area as pre-coked percentage varied from 2.7% to 5.4%. After each experiment, the spent HZSM-5 catalysts with different pre-coked percentages were collected, and then put into an oven for drying (120 °C, 2 h). Subsequently, the dried catalysts were combusted at 800 °C for 2 h in a muffle furnace, and then coke yield was calculated by weight difference. The coke yields were also illustrated in Fig. 3. As could be observed, the coke yields decreased gradually with an increasing of pre-coked percentage, and this experiment result was in good agreement with the above analysis. Note that the pyrolytic products of CFP of MW can be classified into several categories, including hydrocarbons, carbon dioxides, alcohols, furans, phenols, sugars and acids, and the carbon yields of these products with different pre-coked percentages of HZSM5 were shown in Fig. 4. Simultaneously, it was worth noting that due to the high percentage of nitrogen (4.7%) was measured in the ultimate analysis of MW, a negligible number of nitrogen containing species (such as indolizine) in the pyrolytic products were found. As indicated in Fig. 4, the carbon yields of hydrocarbons tended to increase at first and then decrease at the studied precoked percentage regions, and a maximum value of 53.47% occurred at 2.7%. Meanwhile, the yield of carbon dioxide showed the similar trend to hydrocarbons with a maximum yield of 29.90%. Furthermore, the obtained oxygenates presented an opposite variation tendency, and the minimum values could be reached when pre-coked percentage was 2.7%. It seemed that moderate pre-coked percentages (from 0% to 2.7%) of HZSM-5 catalyst could give a rise in production of hydrocarbons and carbon dioxides, as well as promote the decrease of oxygenates, the underlying reason was that appropriate pre-coked of HZSM-5 reduced the formation of oxygenates and removed the oxygen as CO2, water, and CO through deoxygenation reactions (including dehydration, decarboxylation, decarbonylation and so forth) (Chen et al., 2014), and thus an upgraded bio-oil production with a higher content of hydrocarbons and lower fraction of oxygenates could be obtained. Generally, as discussed before, the acidity performance of HZSM-5 catalyst contributed to the formation of pyrolytic organic
ids Ac
Fig. 5 performed the carbon yields of aromatics and some other hydrocarbons obtained from CFP of MW with different pre-coked percentages of HZSM-5, mainly including benzene, toluene, p-xylene, naphthalene and butane. As mentioned above, the unique structure of HZSM-5 could lead to an increasing carbon yield of aromatics due to that the pore diameter of HZSM-5 was similar to the dynamics diameters of toluene, xylene, and benzene. Therefore, when the primary pyrolytic organic vapors passed through the catalyst, they would diffuse into the internal pores of HZSM-5 catalyst and some aromatics could be formed by an
10
J. Wang et al. / Bioresource Technology 212 (2016) 6–10
aromatic carbon pool in the HZSM-5 framework. As performed in Fig. 5, the carbon yields of target products tended to increase at first and then decrease within the researched pre-coked percentage ranges. Among the desired aromatic products obtained from CFP of MW, toluene and p-xylene components were dominant, and both the maximum values of carbon yields (11.01% for toluene and 14.32% for p-xylene) could be conceived as pre-coked percentage was 2.7%. Meanwhile, Fig. 6 presented the selectivities of aromatics, it could be indicated that the selectivity of p-xylene increased at first and then decreased with a maximum value of 34.22% when precoked percentage was 1.3%. However, the selectivity of toluene showed the opposite tendency with a minimum value of 25.47% occurring at 2.7%. Furthermore, the selectivity of benzene tended to present the similar trend to that of toluene, and a minimum value of 13.2% occurred under the same pre-coked percentage condition.
4. Conclusions Experiments for CFP of MW with different modified HZSM-5 catalysts were conducted using Py-GC/MS. The external acid sites and specific surface area of HZSM-5 decreased when pre-coked percentages increased from 0% to 5.4%. The yields of hydrocarbons and carbon dioxides tended to increase at first and then decrease with the increasing of pre-coked percentage. While the carbon yields of oxygenated compounds occurred the opposite tendency. The selectivity of p-xylene increased at first and then decreased with a maximum value of 34.22%, however, the selectivity of toluene showed the opposite tendency with a minimum value of 25.47%.
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).
Appendix A. Supplementary data The X-ray diffraction (XRD) patterns of pre-coked modified HZSM-5 zeolites associated with this article could be found. Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biortech.2016.04.005.
References Abnisa, F., Wan, D.W., 2014. A review on co-pyrolysis of biomass: an optional technique to obtain a high-grade pyrolysis oil. Energy Convers. Manage. 87, 71– 85. Akhtar, J., Saidina, A.N., 2012. A review on operating parameters for optimum liquid oil yield in biomass pyrolysis. Renewable Sustainable Energy Rev. 16 (7), 5101– 5109. Al-Khattaf, S., 2007. Enhancing p-xylene selectivity during m-xylene transformation using mildly pre-coked ZSM-5 catalyst. Chem. Eng. Process: Process Intensification 46 (10), 964–974. Bauer, F., Chen, W., Bilz, E., Freyer, A., Sauerland, V., Liu, S., 2007. Surface modification of nano-sized HZSM-5 and HFER by pre-coking and silanization. J. Catal. 251 (2), 258–270. Bertero, M., Sedran, U., 2016. Immediate catalytic upgrading of soybean shell biooil. Energy 94, 171–179. Borges, F.C., Xie, Q., Min, M., Muniz, L.A., Farenzena, M., Trierweiler, J.O., 2014. Fast microwave-assisted pyrolysis of microalgae using microwave absorbent and HZSM-5 catalyst. Bioresour. Technol. 166, 518–526. Brebu, M., Ucar, S., Vasile, C., Yanik, J., 2010. Co-pyrolysis of pine cone with synthetic polymers. Fuel 89 (8), 1911–1918. Chen, G., Liu, C., Ma, W., Zhang, X., Li, Y., Yan, B., 2014. Co-pyrolysis of corn cob and waste cooking oil in a fixed bed. Bioresour. Technol. 166, 500–509. Chia, D.A., Trimm, D.L., 2005. The effect of pre-coking on the activity and selectivity of the catalytic cracking of squalane. J. Chem. Technol. Biotechnol. 80 (3), 353– 355. Du, Z., Ma, X., Li, Y., Chen, P., Liu, Y., Lin, X., 2013. Production of aromatic hydrocarbons by catalytic pyrolysis of microalgae with zeolites: catalyst screening in a pyroprobe. Bioresour. Technol. 139, 397–401. Kim, Y.H., Lee, K.H., Lee, J.S., 2011. 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 178 (1), 72–78. Li, J., Wei, Y., Liu, G., Qi, Y., Tian, P., Li, B., 2011. Comparative study of MTO conversion over SAPO-34, H-ZSM-5 and H-ZSM-22: correlating catalytic performance and reaction mechanism to zeolite topology. Catal. Today 171 (1), 221–228. Liu, W.J., Tian, K., Jiang, H., Zhang, X.S., Yang, G.X., 2013. Preparation of liquid chemical feedstocks by co-pyrolysis of electronic waste and biomass without formation of polybrominated dibenzo-p-dioxins. Bioresour. Technol. 128, 1–7. Lu, Y., Wei, X.Y., Cao, J.P., Li, P., Liu, F.J., Zhao, Y.P., 2012. Characterization of a bio-oil from pyrolysis of rice husk by detailed compositional analysis and structural investigation of lignin. Bioresour. Technol. 116, 114–119. Tamiyakul, S., Ubolcharoen, W., Tungasmita, D.N., Jongpatiwut, S., 2015. Conversion of glycerol to aromatic hydrocarbons over Zn-promoted HZSM-5 catalysts. Catal. Today 256, 325–335. Valle, B., Gayubo, A.G., Aguayo, A.T., Olazar, M., Bilbao, J., 2010. Selective production of aromatics by crude bio-oil valorization with a nickel-modified HZSM-5 zeolite catalyst. Energy Fuels 24 (3), 2060–2070. Wang, C.M., Wang, Y.D., Xie, Z.K., 2013a. Insights into the reaction mechanism of methanol-to-olefins conversion in HSAPO-34 from first principles: are olefins themselves the dominating hydrocarbon pool species? J. Catal. 301, 8–19. Wang, S., Ru, B., Lin, H., Luo, Z., 2013b. Degradation mechanism of monosaccharides and xylan under pyrolytic conditions with theoretic modeling on the energy profiles. Bioresour. Technol. 143, 378–383. Zhang, C., Xing, J., Song, L., Xin, H., Lin, S., Xing, L., 2014. Aqueous-phase hydrodeoxygenation of lignin monomer eugenol: influence of Si/Al ratio of HZSM-5 on catalytic performances. Catal. Today 234, 145–152. Zhang, H.Y., Xiao, R., Nie, J., Jin, B., Shao, S.S., Xiao, G., 2015a. Catalytic pyrolysis of black-liquor lignin by co-feeding with different plastics in a fluidized bed reactor. Bioresour. Technol. 192, 68–74. Zhang, B., Zhong, Z.P., Wang, X.B., Ding, K., Song, Z.W., 2015b. Catalytic upgrading of fast pyrolysis biomass vapors over fresh, spent and regenerated ZSM-5 zeolites. Fuel Process. Technol. 138, 430–434.
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Catalytic fast pyrolysis of mushroom waste to upgraded bio-oil products via pre-coked modified HZSM-5 catalyst Jia Wang, Zhaoping Zhong ⇑, Kuan Ding, Zeyu Xue Key Laboratory of Energy Thermal Conversion and Control of Ministry of Education, Southeast University, No. 2 Sipailou, Xuanwu District, Nanjing, Jiangsu 210096, China
h i g h l i g h t s HZSM-5 catalyst was modified with pre-coked by methanol to olefins reactions. Modified HZSM-5 catalyst contributed to the carbon yields of hydrocarbons. Coke yields decreased gradually with increasing of pre-coked percentage. Obtained oxygenates decreased at first and then increased at studied pre-coked percentage ranges.
a r t i c l e
i n f o
Article history: Received 18 February 2016 Received in revised form 31 March 2016 Accepted 2 April 2016 Available online 6 April 2016 Keywords: Catalytic fast pyrolysis Pre-coked Modified HZSM-5 catalyst Mushroom waste
a b s t r a c t In this paper, HZSM-5 catalyst was modified by pre-coked to cover the strong external acid sites by methanol to olefins reaction, and the modified catalysts were then applied to conduct the catalyst fast pyrolysis of mushroom waste for upgraded bio-fuel production. Experiment results showed that the strong external acid sites and specific surface area decreased with pre-coked percentage increasing from 0% to 5.4%. Carbon yields of hydrocarbons increased at first and then decreased with a maximum value of 53.47%. While the obtained oxygenates presented an opposite variation tendency, and the minimum values could be reached when pre-coked percentage was 2.7%. Among the achieved hydrocarbons, toluene and p-xylene were found to be the main products, and the selectivity of p-xylene increased at first and then decreased with a maximum value of 34.22% when the pre-coked percentage was 1.3%, and the selectivity of toluene showed the opposite tendency with a minimum value of 25.47%. Ó 2016 Published by Elsevier Ltd.
1. Introduction Concerning the quite heavy dependence on fossil fuels and environmental impacts of greenhouse gas emissions associated with massive consumption of fossil energy, some efforts are currently made to find or invent renewable and alternative resources which are environmentally friendly and with a high reliability and stability (Abnisa and Wan, 2014; Wang et al., 2013b; Zhang et al., 2015b). In this regard, bioenergy is now accepted as a promising alternative that can attribute to meeting the fossil energy needs and enhancing the environmental problems (Akhtar and Saidina, 2012). Among the utilizations of bioenergy, catalytic fast pyrolysis (CFP) is particularly interesting and technically reliable, as well as economically feasible for converting bioenergy into bio-fuel products (Liu et al., 2013; Lu et al., 2012). This technology has attracted increasing attentions as its role in producing the highest bio-fuel
⇑ Corresponding author. E-mail address: [email protected] (Z. Zhong). http://dx.doi.org/10.1016/j.biortech.2016.04.005 0960-8524/Ó 2016 Published by Elsevier Ltd.
yields (75 wt.%) under conditions of moderate temperature (500–600 °C) and short pyrolytic vapor residence time (1 s) (Borges et al., 2014). In this sense, CFP can be selected as a recommend approach to seek the stated objective of biomass application. Among various researched catalysts used in CFP process, HZSM5 zeolite dominated the pronounced performance for shape selectivity and liquid products deoxygenation (Tamiyakul et al., 2015; Brebu et al., 2010; Zhang et al., 2014). The 3-dimensional intersecting pore structure of HZSM-5 is characterized by well-defined 10membered rings with straight channels of 0.53 0.56 nm and sinusoidal channels of 0.51 0.55 nm, in which oxygenated organic compounds will be removed and hydrocarbons will be formed when primary pyrolytic volatile matters passed through the catalyst layer (Zhang et al., 2015a). However, it is important to note that there are several critical problems associated with HZSM-5 catalyst, such as the high coke production and short lifecycle. Coke is a kind of large molecule aromatic compound generated and accumulated on the surface of HZSM-5 catalyst, which leads to the blockage of the pore opening mouth and thus a
J. Wang et al. / Bioresource Technology 212 (2016) 6–10
deactivation of catalyst will be caused (Bertero and Sedran, 2016). Note that the strong external acid sites of HZSM-5 are responsible for coke formation, and a promising measure for decreasing the coke yield is to lower the strong external acid sites but meanwhile retain the internal acid sites (Zhang et al., 2014). In this sense, precoked treatment showed great promise for eliminating the strong external acid sites and improving the stability and activity of HZSM-5 zeolites (Chia and Trimm, 2005; Kim et al., 2011). Al-Khattaf (2007) found that the controlled pre-coking process, which could be obtained by exposing the zeolite to organic compounds at moderate temperature, showed significant effect for the enhancement of p-xylene selectivity during toluene disproportionation. Bauer et al. (2007) also revealed that in terms of inactivation of external surface acidity, specimens modified by pre-coking were proved to be more effective compared to those by one-cycle silica deposition. However, at the best of our knowledge, the effect of pre-coked percentage of HZSM-5 on products distribution achieved from CFP of MW has not been investigated and analyzed yet. Mushroom waste (MW) is a representative kind of biomass energy, which consists of wood chip, cottonseed hull and corn cob. A pronounced way to utilized it is catalytic fast pyrolysis, which can convert MW into fuels or other high value-add chemicals. In this contribution, experiments on selectively eliminating the external strong acid sites were conducted by pre-coked of HZSM-5 through methanol to olefins (MTO) reaction. In contrast with fresh HZSM-5 catalyst, the pre-coked catalyst already has some cokes deposited on the external surface, and thus the strong external acidity and activity will be reduced. Subsequently, the modified catalysts with different pre-coked percentages were used to conduct the CFP of MW using Py-GC/MS. The effects of precoked percentages on products distribution of CFP of MW were investigated. The rest of this paper can be organized as follows: in Section 2 the experimental materials and set up are introduced; in Section 3 the effects of pre-coked percentages on products distribution, such as hydrocarbons, phenols, alcohols and so forth, as well as the selectivities of aromatics, were analyzed and discussed in detail; and in Section 4, the conclusions are drawn. 2. Materials and methods
7
methanol is added (Wang et al., 2013a). The fresh HZSM-5 catalyst (SiO2/Al2O3 = 50) was pre-coked by MTO reaction, and the treatment process was illustrated as Fig. 1. An injection pump and heat tape were used to generate methanol vapors and nitrogen was applied as carrier gas with a constant flow rate of 300 ml/min. Meanwhile, in order to obtain a high contact efficiency of methanol vapors and HZSM-5 catalyst, a vertical tube furnace was utilized to conduct the pre-coked experiments. The reaction temperature was controlled by temperature controller system, and in this experiment, the temperature was kept at 450 °C. When the methanol vapors were introduced into the reactor, they passed through the HZSM-5 catalyst layer, resulting the MTO reactions and thus, some cokes would be formed on the catalyst. Posterior to each experiment, some pre-coked HZSM-5 catalysts were combusted at 800 °C for 2 h in a muffle furnace, and the coke yield was calculated by the weight difference. Here, a parameter called pre-coked percentage was used to evaluate the pre-coked degree, and it was defined as the total mass of coke to HZSM-5 catalyst. In this paper, HZSM-5 catalysts with different pre-coked percentages (1.3%, 2.7%, 3.8%, 4.5% and 5.4%) were analyzed and studied, and they were achieved by adjusting and varying the MTO reaction time. Besides, the HZSM-5 catalysts used in the experiments were bought from the Catalyst Plant of Nankai University, Tianjin, China. 2.3. Experimental methods of CFP A CDS Pyroprobe 5200 pyrolyzer was used to conduct the CFP experiments. The probe is a quartz tube, open at both ends, and heated by platinum resistance. Before the experiment, some packed quartz wool, 0.50 mg HZSM-5 catalysts and 0.50 mg feedstock specimens (MW) were put in the quartz in sequence. Besides, the pyrolysis temperature was kept for 20 s with a heating rate of 20 °C/ms, the helium with a high-purity (99.999%) purchased from Nanjing Maikesi Nanfen Special Gas Co., Ltd., which was utilized as carrier gas at a stable flow of 1.0 mL/min. The pyrolysis vapors passed through the quartz tube into a gas chromatography/mass spectrometry (7890A/5975C, Agilent) via the helium sweeper gas stream. The interface and injector temperature of GC/MS was 300 °C and 275 °C, respectively. A capillary called HP-5MS with a size of 0.25 mm 0.25um 30 m was used to conduct GC separation. The split ratio was set at 1:80, and the
2.1. Materials Mushroom waste (MW) was obtained from a farm in Xuzhou, Jiangsu province, China. Before the experiment, some pretreatments of mushroom waste were carried out, which could be shown as follows: (1) Dried at 105 °C for 24 h in a drying oven; (2) Pulverized and grinded mechanically; (3) Sifted through a 60-mesh sieve and then kept in a sealed bottle. The ultimate analysis of dried MW was conducted and the results were as follows: 41.8 wt.% carbon, 6.4 wt.% hydrogen, 41.9 wt.% oxygen, and 4.7 wt.% nitrogen. Simultaneously, the proximate analysis of air-dried MW was also carried out, and the results were: 8.32 wt.% moisture, 76.59 wt.% volatile, 5.08 wt.% ash, and 10.01 wt.% fixed carbon. 2.2. Pre-coked treatment process of HZSM-5 It should be noted that more than 20 reaction mechanisms of methanol to olefins (MTO) have been proposed by different researchers, and among them, a hydrocarbon pool mechanism, which was firstly presented by Kolboe and co-workers, have attracted much more attention (Li et al., 2011). The hydrocarbon pool mechanism reveals that some certain organic active centers known as hydrocarbon pool intermediates act as co-catalysts within the cages of zeolites to which olefins are eliminated and
Fig. 1. Pre-coked treatment process of HZSM-5 catalyst by MTO reaction: 1 – N2 gas cylinder; 2 – Flowmeter; 3 – Injection pump (methanol); 4 – Heat tape; 5 – Vertical tube furnace; 6 – HZSM-5 catalyst; 7 – Temperature controller and thermocouple; 8 – Heat exchanger; 9 – Gas-washing bottle.
J. Wang et al. / Bioresource Technology 212 (2016) 6–10
2.4. Product analysis Note that during the Py-GC/MS experiments, organic vapor pyrolytic products could not be collected, thus the yields of total condensable organic products seemed uncertain. However, the total mass of specimens and HZSM-5 catalysts were fixed in each experiment, thus the changes of organic products could be discriminated with analysis and identification of overall peak area. Besides, in order to attain the relative contents of pyrolytic volatile matters, a semi-quantitative method via the calculation of chromatographic area was utilized. 2.5. Evaluation indicator In this paper, the selectivity of aromatics was defined, which can be shown as follows:
Paro Saro ¼ P Paro
ð1Þ
where Saro and Paro are the selectivity and peak area of a certain kind of aromatics, respectively. RParo is the total peak area of aromatics. 3. Results and discussion 3.1. Physicochemical characteristics of catalysts
0.9
345
0.8
330
0.7
315
0.6
300
285
Strong acid sites Weak acid sites
0.5
270
0.4
BET specific surface area (m2/g)
Acid sites (mmol NH3/g)
The porosity and acidity performances of fresh and pre-coked modified HZSM-5 catalysts were given in Fig. 2. The porosity char-
acteristics were conducted by BET method with N2 porosimetry, and the acidity performances were tested and analyzed by ammonia TPD. As presented in Fig. 2, both the acid sites (strong acid sites and weak acid sites) decreased gradually as pre-coked percentage increased from 0% to 5.4%, and the specific surface area tended to present the similar variation trend. 3.2. Pyrolytic products distribution Prior to the experiment, in order to analyze the effect of catalyst on fast pyrolysis of biomass, some additional experiments on noncatalytic fast pyrolysis (non-CFP) of MW at the temperature of 600 °C were carried out. Experiment results indicated that the total peak area from CFP of MW with fresh HZSM-5 was 2.28 109, while the peak area of non-CFP of MW was 3.7 109. Furthermore, the pyrolytic volatile matters for non-CFP of MW consisted of various oxygenates, and the production of hydrocarbons and aromatics were limited. However, the hydrocarbons and aromatics (e.g., benzene, toluene, xylene, naphthalene, ethylbenzene, propane, indene, 1 h-indene, 3-methyl-) achieved from CFP of MW were very abundant. It could be concluded that an addition of HZSM-5 catalyst to fast pyrolysis of MW would result in a remarkable increase in carbon yields of aromatics. The underlying reason was that, as previously mentioned, because of the unique molecular structure of HZSM-5 catalyst, some target aromatics could be generated by multiple reactions (such as decarboxylation, aromatization, oligomerization, isomerization.) when primary pyrolytic vapor products passed through the internal pores (Du et al., 2013; Valle et al., 2010). Fig. 3 proposed the total peak area of pyrolytic organic products obtained from CFP of MW over HZSM-5 catalysts with different pre-coked percentages. It could be observed that the total peak areas decreased at first and then increased as pre-coked percentages of HZSM-5 changed from 0% to 5.4%, and the minimum value would be obtained when pre-coked percentage was 2.7%. This was due to that, for parent HZSM-5, the strong external acidity limited the catalytic pyrolysis reactions for coke could be easily generated on the surface, and thus led to the pore blockage and catalyst inactivation. However, when the appropriate treatment process of precoked was conducted, some of the external strong acid sites would be covered and the optimization of external acidity and activity could be promoted, which resulted in a decrease in total peak area and an increasing quality in obtained organic desired products. Meanwhile, it was necessary to note that the pre-coked treatment of HZSM-5 catalyst might cause both positive and negative effects
3.3
5.0
3.0
4.5
2.7
4.0
2.4 3.5 2.1 3.0 1.8 2.5 1.5
0
1
2
3
4
5
6
0
1
2
3
4
Pre-coked percentage / %
Pre-coked percentage / %
Fig. 2. Characterization results of fresh and pre-coked modified HZSM-5 catalysts.
Fig. 3. Total peak area and coke yield.
5
6
Coke yield / %
ionization energy of MS was 70 eV. In order to reduce experimental errors, experiments were carried out in triplicate, and the average values were reported. It should be noted that in order to determine the moderate temperature for CFP of MW over pre-coked modified catalysts, we conducted several experiments on CFP of MW over fresh HZSM-5 catalyst with reaction temperature increasing from 450 to 700 °C. The results showed that and the relative contents of hydrocarbons tended to increase at first and then decrease with a maximum yield of 44.52% occurring at 600 °C (total relative contents of hydrocarbons: 34.83% at 450 °C, 38.53% at 500 °C, 41.67% at 550 °C, 44.52% at 600 °C, 41.03% at 650 °C and 39.64% at 700 °C). Simultaneously, in this contribution, we focus on aromatic hydrocarbons more than other organics due to the unique effect of HZSM-5 for aromatization reactions during the CFP of MW. For this reason, we selected 600 °C as an optimal reaction condition.
Total peak area (x10 9)
8
9
J. Wang et al. / Bioresource Technology 212 (2016) 6–10
Non-catalyst Fresh HZSM-5 1.3%-HZSM-5 2.7%-HZSM-5 3.8%-HZSM-5 4.5%-HZSM-5 5.4%-HZSM-5
Carbon yield (%)
40
30
10
0 s ls rna eno Fu Ph
Fig. 4. Products distribution.
s gar Su
Relative content / %
12
9
6
3
0 Non-catalyst
0
4 1 3 2 Pre-coked percentage / %
5
6
5
6
Fig. 5. Relative contents of hydrocarbons.
40
Benzene Naphthalene
Toluene p-Xylene Other aromatics
35 30 25 20 15 10
Non-catalyst
0
4 1 2 3 Pre-coked percentage / %
Fig. 6. Selectivities of aromatics.
3.3. Carbon yields of hydrocarbons and selectivities of aromatics
20
s e ls xid bon oho dio car Alc n o r o d rb Hy Ca
15
Benzene Toluene p-Xylene Naphthalene Other aromatics Butane Other hydrocarbons
products and coke, and the strong external acidity was attributed to the high yield of coke. In this regard, in order to reduce the strong external acidity of HZSM-5, it seems that appropriate precoked of HZSM-5 can be a good process to decrease the coke yield and provide a significant contribution to the further shapeselective reactions. However, it should be noted that too high pre-coked percentages (from 2.7% to 5.4%) were responsible for the pore opening blockage, and thus an obstruction of catalytic pyrolysis reaction could be caused.
60
50
18
Selectivity of aromatics / %
on products distribution during CFP of MW, and the negative effect might lie with pre-coked process could block or inactive the internal effective weak acid sites, which might confine the molecules of similar sizes to diffuse in the catalyst pores, and thus resulted in a decrease in catalytic performance of HZSM-5 during the CFP of MW. That was the reason for the increase of total peak area as pre-coked percentage varied from 2.7% to 5.4%. After each experiment, the spent HZSM-5 catalysts with different pre-coked percentages were collected, and then put into an oven for drying (120 °C, 2 h). Subsequently, the dried catalysts were combusted at 800 °C for 2 h in a muffle furnace, and then coke yield was calculated by weight difference. The coke yields were also illustrated in Fig. 3. As could be observed, the coke yields decreased gradually with an increasing of pre-coked percentage, and this experiment result was in good agreement with the above analysis. Note that the pyrolytic products of CFP of MW can be classified into several categories, including hydrocarbons, carbon dioxides, alcohols, furans, phenols, sugars and acids, and the carbon yields of these products with different pre-coked percentages of HZSM5 were shown in Fig. 4. Simultaneously, it was worth noting that due to the high percentage of nitrogen (4.7%) was measured in the ultimate analysis of MW, a negligible number of nitrogen containing species (such as indolizine) in the pyrolytic products were found. As indicated in Fig. 4, the carbon yields of hydrocarbons tended to increase at first and then decrease at the studied precoked percentage regions, and a maximum value of 53.47% occurred at 2.7%. Meanwhile, the yield of carbon dioxide showed the similar trend to hydrocarbons with a maximum yield of 29.90%. Furthermore, the obtained oxygenates presented an opposite variation tendency, and the minimum values could be reached when pre-coked percentage was 2.7%. It seemed that moderate pre-coked percentages (from 0% to 2.7%) of HZSM-5 catalyst could give a rise in production of hydrocarbons and carbon dioxides, as well as promote the decrease of oxygenates, the underlying reason was that appropriate pre-coked of HZSM-5 reduced the formation of oxygenates and removed the oxygen as CO2, water, and CO through deoxygenation reactions (including dehydration, decarboxylation, decarbonylation and so forth) (Chen et al., 2014), and thus an upgraded bio-oil production with a higher content of hydrocarbons and lower fraction of oxygenates could be obtained. Generally, as discussed before, the acidity performance of HZSM-5 catalyst contributed to the formation of pyrolytic organic
ids Ac
Fig. 5 performed the carbon yields of aromatics and some other hydrocarbons obtained from CFP of MW with different pre-coked percentages of HZSM-5, mainly including benzene, toluene, p-xylene, naphthalene and butane. As mentioned above, the unique structure of HZSM-5 could lead to an increasing carbon yield of aromatics due to that the pore diameter of HZSM-5 was similar to the dynamics diameters of toluene, xylene, and benzene. Therefore, when the primary pyrolytic organic vapors passed through the catalyst, they would diffuse into the internal pores of HZSM-5 catalyst and some aromatics could be formed by an
10
J. Wang et al. / Bioresource Technology 212 (2016) 6–10
aromatic carbon pool in the HZSM-5 framework. As performed in Fig. 5, the carbon yields of target products tended to increase at first and then decrease within the researched pre-coked percentage ranges. Among the desired aromatic products obtained from CFP of MW, toluene and p-xylene components were dominant, and both the maximum values of carbon yields (11.01% for toluene and 14.32% for p-xylene) could be conceived as pre-coked percentage was 2.7%. Meanwhile, Fig. 6 presented the selectivities of aromatics, it could be indicated that the selectivity of p-xylene increased at first and then decreased with a maximum value of 34.22% when precoked percentage was 1.3%. However, the selectivity of toluene showed the opposite tendency with a minimum value of 25.47% occurring at 2.7%. Furthermore, the selectivity of benzene tended to present the similar trend to that of toluene, and a minimum value of 13.2% occurred under the same pre-coked percentage condition.
4. Conclusions Experiments for CFP of MW with different modified HZSM-5 catalysts were conducted using Py-GC/MS. The external acid sites and specific surface area of HZSM-5 decreased when pre-coked percentages increased from 0% to 5.4%. The yields of hydrocarbons and carbon dioxides tended to increase at first and then decrease with the increasing of pre-coked percentage. While the carbon yields of oxygenated compounds occurred the opposite tendency. The selectivity of p-xylene increased at first and then decreased with a maximum value of 34.22%, however, the selectivity of toluene showed the opposite tendency with a minimum value of 25.47%.
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).
Appendix A. Supplementary data The X-ray diffraction (XRD) patterns of pre-coked modified HZSM-5 zeolites associated with this article could be found. Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.biortech.2016.04.005.
References Abnisa, F., Wan, D.W., 2014. A review on co-pyrolysis of biomass: an optional technique to obtain a high-grade pyrolysis oil. Energy Convers. Manage. 87, 71– 85. Akhtar, J., Saidina, A.N., 2012. A review on operating parameters for optimum liquid oil yield in biomass pyrolysis. Renewable Sustainable Energy Rev. 16 (7), 5101– 5109. Al-Khattaf, S., 2007. Enhancing p-xylene selectivity during m-xylene transformation using mildly pre-coked ZSM-5 catalyst. Chem. Eng. Process: Process Intensification 46 (10), 964–974. Bauer, F., Chen, W., Bilz, E., Freyer, A., Sauerland, V., Liu, S., 2007. Surface modification of nano-sized HZSM-5 and HFER by pre-coking and silanization. J. Catal. 251 (2), 258–270. Bertero, M., Sedran, U., 2016. Immediate catalytic upgrading of soybean shell biooil. Energy 94, 171–179. Borges, F.C., Xie, Q., Min, M., Muniz, L.A., Farenzena, M., Trierweiler, J.O., 2014. Fast microwave-assisted pyrolysis of microalgae using microwave absorbent and HZSM-5 catalyst. Bioresour. Technol. 166, 518–526. Brebu, M., Ucar, S., Vasile, C., Yanik, J., 2010. Co-pyrolysis of pine cone with synthetic polymers. Fuel 89 (8), 1911–1918. Chen, G., Liu, C., Ma, W., Zhang, X., Li, Y., Yan, B., 2014. Co-pyrolysis of corn cob and waste cooking oil in a fixed bed. Bioresour. Technol. 166, 500–509. Chia, D.A., Trimm, D.L., 2005. The effect of pre-coking on the activity and selectivity of the catalytic cracking of squalane. J. Chem. Technol. Biotechnol. 80 (3), 353– 355. Du, Z., Ma, X., Li, Y., Chen, P., Liu, Y., Lin, X., 2013. Production of aromatic hydrocarbons by catalytic pyrolysis of microalgae with zeolites: catalyst screening in a pyroprobe. Bioresour. Technol. 139, 397–401. Kim, Y.H., Lee, K.H., Lee, J.S., 2011. 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 178 (1), 72–78. Li, J., Wei, Y., Liu, G., Qi, Y., Tian, P., Li, B., 2011. Comparative study of MTO conversion over SAPO-34, H-ZSM-5 and H-ZSM-22: correlating catalytic performance and reaction mechanism to zeolite topology. Catal. Today 171 (1), 221–228. Liu, W.J., Tian, K., Jiang, H., Zhang, X.S., Yang, G.X., 2013. Preparation of liquid chemical feedstocks by co-pyrolysis of electronic waste and biomass without formation of polybrominated dibenzo-p-dioxins. Bioresour. Technol. 128, 1–7. Lu, Y., Wei, X.Y., Cao, J.P., Li, P., Liu, F.J., Zhao, Y.P., 2012. Characterization of a bio-oil from pyrolysis of rice husk by detailed compositional analysis and structural investigation of lignin. Bioresour. Technol. 116, 114–119. Tamiyakul, S., Ubolcharoen, W., Tungasmita, D.N., Jongpatiwut, S., 2015. Conversion of glycerol to aromatic hydrocarbons over Zn-promoted HZSM-5 catalysts. Catal. Today 256, 325–335. Valle, B., Gayubo, A.G., Aguayo, A.T., Olazar, M., Bilbao, J., 2010. Selective production of aromatics by crude bio-oil valorization with a nickel-modified HZSM-5 zeolite catalyst. Energy Fuels 24 (3), 2060–2070. Wang, C.M., Wang, Y.D., Xie, Z.K., 2013a. Insights into the reaction mechanism of methanol-to-olefins conversion in HSAPO-34 from first principles: are olefins themselves the dominating hydrocarbon pool species? J. Catal. 301, 8–19. Wang, S., Ru, B., Lin, H., Luo, Z., 2013b. Degradation mechanism of monosaccharides and xylan under pyrolytic conditions with theoretic modeling on the energy profiles. Bioresour. Technol. 143, 378–383. Zhang, C., Xing, J., Song, L., Xin, H., Lin, S., Xing, L., 2014. Aqueous-phase hydrodeoxygenation of lignin monomer eugenol: influence of Si/Al ratio of HZSM-5 on catalytic performances. Catal. Today 234, 145–152. Zhang, H.Y., Xiao, R., Nie, J., Jin, B., Shao, S.S., Xiao, G., 2015a. Catalytic pyrolysis of black-liquor lignin by co-feeding with different plastics in a fluidized bed reactor. Bioresour. Technol. 192, 68–74. Zhang, B., Zhong, Z.P., Wang, X.B., Ding, K., Song, Z.W., 2015b. Catalytic upgrading of fast pyrolysis biomass vapors over fresh, spent and regenerated ZSM-5 zeolites. Fuel Process. Technol. 138, 430–434.