In-situ and ex-situ catalytic upgrading of vapors from microwave-assisted pyrolysis of lignin

In-situ and ex-situ catalytic upgrading of vapors from microwave-assisted pyrolysis of lignin

Accepted Manuscript In-situ and ex-situ catalytic upgrading of vapors from microwave-assisted pyrolysis of lignin Liangliang Fan, Paul Chen, Nan Zhou,...

691KB Sizes 0 Downloads 29 Views

Accepted Manuscript In-situ and ex-situ catalytic upgrading of vapors from microwave-assisted pyrolysis of lignin Liangliang Fan, Paul Chen, Nan Zhou, Shiyu Liu, Yaning Zhang, Yuhuan Liu, Yunpu Wang, Muhammad Mubashar Omar, Peng Peng, Min Addy, Yanling Cheng, Roger Ruan PII: DOI: Reference:

S0960-8524(17)31776-5 https://doi.org/10.1016/j.biortech.2017.09.200 BITE 19028

To appear in:

Bioresource Technology

Received Date: Revised Date: Accepted Date:

9 August 2017 26 September 2017 28 September 2017

Please cite this article as: Fan, L., Chen, P., Zhou, N., Liu, S., Zhang, Y., Liu, Y., Wang, Y., Omar, M.M., Peng, P., Addy, M., Cheng, Y., Ruan, R., In-situ and ex-situ catalytic upgrading of vapors from microwave-assisted pyrolysis of lignin, Bioresource Technology (2017), doi: https://doi.org/10.1016/j.biortech.2017.09.200

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

In-situ and ex-situ catalytic upgrading of vapors from microwave-assisted pyrolysis of lignin

Liangliang Fana,b, Paul Chenc, Nan Zhouc, Shiyu Liuc, Yaning Zhangc,d, Yuhuan Liua,b, Yunpu Wangb, Muhammad Mubashar Omarc,e, Peng Pengc, Min Addyc, Yanling Chengc, Roger Ruanc,*

a

Nanchang University, State Key Laboratory of Food Science and Technology, Nanchang

330047, China; b

Nanchang University, Engineering Research Center for Biomass Conversion, Ministry of

Education, Nanchang 330047, China; c

Center for Biorefining and Department of Bioproducts and Biosystems Engineering,

University of Minnesota, 1390 Eckles Ave., St. Paul, MN 55108, United States; d

School of Energy Science and Engineering, Harbin Institute of Technology, Harbin

150001, China. e

Department of Farm Machinery and Power, University of Agriculture, Faisalabad,

Pakistan.

* Corresponding author: Distinguished Guest Professor, Nanchang University, and

Professor and Director, Center for Biorefining and Department of Bioproducts and Biosystems Engineering, University of Minnesota, 1390 Eckles Ave., St. Paul, MN 55108, USA. Tel.: +1 612 625 1710; fax: +1 612 624 3005. E-mail address: [email protected] (R. Ruan)

Abstract In-situ and ex-situ catalytic upgrading with HZSM-5 of vapors from microwave-assisted pyrolysis of lignin were studied. The in-situ process produced higher bio-oil and less char than ex-situ process. The gas yield was similar for both processes. The ex-situ process had higher selectivity to aromatics and produced more syngas and less CO2 than the in-situ process. Additional experiments on ex-situ process found that the bio-oil yield and coke deposition decreased while the gas yield increased at higher catalyst-to-lignin ratios and catalytic upgrading temperatures. The increased catalyst-to-lignin ratio from 0 to 0.3 reduced the selectivity of methoxy phenols from 73.7% to 22.6% while increased that of aromatics from 1.1% to 41.4%. The highest selectivity of alkyl phenols (31.9%) was obtained at 0.2 of catalyst-to-lignin ratio. Higher catalytic temperatures favored greater conversion of methoxy phenols to alkyl phenols and aromatics. Appropriate catalyst-tolignin ratio (0.3) together with higher catalytic temperatures favored syngas formation. Keywords: In-situ; ex-situ; catalytic upgrading; microwave-assisted pyrolysis; lignin.

1. Introduction Lignin is one of the main constitutes of plant cell wall and accounts for 15-30% by mass and up to 40% by energy in biomass (Mukkamala et al., 2012). It is a complex phenolic polymer containing three types of phenyl propane units connected through ether and carbon-carbon bonds (Calvo‐Flores & Dobado, 2010). Pulping in the Kraft process of paper making is the primary unit operation producing lignin containing black liquor, which was released as waste to waterway until the invention of recovery boilers in the early 20th century. Fifty million tons of lignin was estimated to be produced every year (Gosselink et

al., 2004). Additional lignin is produced from the bio-ethanol production using lignocellulosic feedstock (Toledano et al., 2014). The recalcitrant structure of lignin limits its biodegradation; direct burning produces low-value fuel; and a very small amount (2%) is used for making value-added materials (Gosselink et al., 2004). The high demand for renewable energy and products makes lignin an attractive feedstock for valuable chemicals and bio-fuels.

Among the various lignin conversion methods, pyrolysis is one of the most promising and economic technologies (Anex et al., 2010). Pyrolysis is a thermal process which converts biomass to recoverable energy and chemicals in an oxygen-deprived environment. Recently, microwave radiation heating has been increasingly used in pyrolysis of biomass (Borges et al., 2014; Du et al., 2011; Xie et al., 2014b; Yin, 2012) due to its uniform volumetric heating, energy saving and efficiency, and control flexibility. Nevertheless, there are few attempts at converting lignin to valuable chemicals and fuels with microwave-assisted pyrolysis (Bu et al., 2014). The liquid product obtained from pyrolysis, which is known as bio-oil, generally presents some undesirable properties including high acid content, low calorific value, and instability (Liu et al., 2014). These undesirable properties may be improved by using catalytic upgrading, which can be operated as in-situ or ex-situ processes. In the in-situ catalytic upgrading process, the feedstock is mixed with the catalyst prior to being introduced into the same reactor. In the ex-situ catalytic upgrading process, the catalyst is separated from the feedstock and put in the secondary reactor (catalytic bed), through which the primary pyrolytic vapors pass. For in-situ catalytic upgrading, a large proportion of catalyst is needed to ensure the sufficient contact between primary pyrolytic vapors and catalyst. However, consumption relatively smaller of catalyst

in ex-situ catalytic upgrading can provide sufficient pyrolytic vapors reforming. Furthermore, separation of pyrolytic bed and catalytic bed allows independent control of the pyrolysis temperature and catalytic temperature and the char produced from pyrolysis will not contaminate the catalysts and can be recovered without the need to separate it from the catalysts.

To date, a few works on comparative study of in-situ and ex-situ reforming of biomass pyrolytic vapors have been reported (Gamliel et al., 2015; Luo & Resende, 2016; Wang et al., 2014a). However, most of the studies were performed in a micro-pyrolyzer as the catalytic pyrolysis reactor, which is of low relevance to industrial scale applications because the catalytic reaction in the micro-pyrolyzer is very different from that in an industrial large-scale reactor. Furthermore, in a miro-pyrolyzer, a large ratio of catalyst to biomass is generally required to ensure the sufficient catalytic effect for both in-situ and exsitu catalytic upgrading processes (Gamliel et al., 2015; Luo & Resende, 2016; Wang et al., 2014a). Compared with micro-pyrolyzer, a catalytic upgrading study in a bench-scale reactor would provide more information relevant to industrial production. Up to now, limited works have been reported on the comparative study of in-situ and ex-situ catalytic upgrading of biomass pyrolytic vapors in a bench-scale reactor (Iisa et al., 2016). Nevertheless, the comparative study on the catalytic upgrading of lignin with in-situ and exsitu processes in a bench-scale reactor has not been reported.

Zeolite HZSM-5 was used as the catalyst in this study because this low-cost catalyst is most effective for the production of deoxygenated aromatics. Jackson et al. (2009) screened various catalysts, including HZSM-5, KZSM-5, AlMCM-41, solid phosphoric acid, and a

commercial hydrodesulfurization catalyst Co/Mo/Al2O3 in the pyrolysis of Asian lignin, they concluded that the HZSM-5 zeolite was the best catalyst for producing deoxygenated organics because of its strong acidity. Mihalcik et al. (2011) studied catalytic pyrolysis of Asian lignin over various zeolites such as H-mordenite, HZSM-5, HY, H-Beta and HFerrierite and also got the result that HZSM-5 was the most effective zeolite for producing aromatics. Mullen and Boateng (2010) compared the catalytic effect of HZSM-5 and CoO/MoO3 on the catalytic fast pyrolysis of lignin and found that HZSM-5 showed higher efficiency than CoO/MoO3 for the production of aromatics regardless of the type of lignin. The catalytic upgrading of organosolv lignin by HZSM-5 and HY was investigated by Zhang et al. (2014). It is indicated that HZSM-5 produces 2.5 to 40 times more aromatics than HY. In the present work, in-situ and ex-situ catalytic reforming of lignin-derived vapors was studied. The effects of various catalyst-to-lignin ratios and catalytic temperatures on the ex-situ catalytic upgrading process were also investigated.

2. Materials and methods 2.1 Materials The alkali lignin (obtained from Sigma-Aldrich, USA) ZSM-5 (Si/Al = 80, surface area = 425 m2/g, obtained from Zeolyst International, USA) were used as the feedstock and catalyst, respectively. Prior to use, the lignin sample was air dried at 105 °C and ZSM-5 was calcined at 500 °C for 5 h in order to obtain its hydrogen form, HZSM-5. The proximate analysis shows that the lignin is composed of 63.2 wt.% of volatiles, 8.8 wt.% of ash, and 28.0 wt.% of fixed carbon; while the ultimate analysis shows that the lignin contains 51.9 wt.% C, 5.0 wt.% H, 38.2 wt.% O, and 4.9 wt.% S (Fan et al., 2017a).

2.2 Experimental setup Fig. 1 shows schematically the experimental setup, which has been described in the previous study (Fan et al., 2017c). For ex-situ catalytic upgrading, lignin mixed with SiC particles was loaded into the quartz reactor for pyrolysis while the catalyst was placed in a quartz column (secondary reactor) coupled with a quartz fritted disc, which contains quantities of homogeneous pores (90-150 µm) and helps secure the catalyst bed and ensures primary vapors passing through evenly. The other side of the catalyst bed was filled with quartz wool. Effects of various catalyst-to-lignin ratios (0.1, 0.2, 0.3, and 0.4) and catalytic temperatures (250, 350, 450 and 550 °C) on the ex-situ catalytic upgrading were investigated. For in-situ catalytic upgrading, lignin and catalyst were mixed uniformly with SiC particles and introduced into the quartz reactor. The pyrolysis temperature was set at 550 °C for both in-situ and ex-situ catalytic upgrading processes. Prior to pyrolysis, the catalytic upgrading systems were vacuumed for 10 min to ensure the oxygen deprived atmosphere and the secondary reactor was heated to the desired temperature. The pyrolysis bed was heated to the desired temperature under microwave radiation and then maintained by turning on and off the microwave oven manually until no pyrolytic vapors were produced. The vapors after catalytic upgrading were quenched quickly through condensing system and condensed to a liquid, which was collected by the flask. The non-condensable gas product was collected with a gas bag. After the experiment, the SiC particles were recycled by removing the char. The weights of bio-oil and char were calculated by weight difference of the flask with and without bio-oil and the quartz reactor with and without char, respectively. The weight of coke deposited on the catalyst was measured by the weight difference of catalyst bed before and after the reaction. Each run of the experiment was

performed at least in triplicates. The yields of the products (bio-oil, char, coke, and gas) were determined according to the following equations: ௠್೔೚ష೚೔೗

ܻ௕௜௢ି௢௜௟ =

௠೗೔೒೙೔೙

× 100%



ܻ௖௛௔௥ = ௠ ೎೓ೌೝ × 100%

(1) (2)

೗೔೒೙೔೙



ܻ௖௢௞௘ି௟௜௚௡௜௡ = ௠ ೎೚ೖ೐ × 100%

(3)

೗೔೒೙೔೙

௠೎೚ೖ೐

ܻ௖௢௞௘ି௖௔௧௔௟௬௦௧ = ௠

× 100%

(4)

೎ೌ೟ೌ೗೤ೞ೟

ܻ௚௔௦ = 100% − ܻ௕௜௢ି௢௜௟ − ܻ௖௛௔௥ − ܻ௖௢௞௘ି௟௜௚௡௜௡

(5)

Where Yboi-oil, Ychar, and Ygas are the yields of bio-oil, char, and gas based on dry lignin, respectively; Ycoke-lignin and Ycoke-catalyst are the yields of coke based on dry lignin and catalyst, respectively; mbio-oil, mchar, mcoke, mgas, mlignin, and mcatalyst are the weights of bio-oil, char, coke, gas, lignin, and catalyst, respectively.

2.3 Characterization of bio-oil The organic fraction of bio-oil was extracted with chloroform and injected into GC/MS (7890-5975C, Agilent Corporation, USA) coupled with NIST data library for the chemical selectivity characterization. The parameters set for GC/MS testing was described in the previous study (Fan et al., 2017a). The chemical selectivity of bio-oil was identified by calculating the chromatographic area percentage of each compound.

2.4 Characterization of gas product Micro-GC (CP-4900) purchased from Varian Inc. was used to determine the constitutes of the gas product. TCD was loaded in the GC system as a detector. Gas components were

separated by two separate columns, MS5A and PoraPLOT (PPQ), which were coupled with helium and argon as the carrier gas, respectively. Temperatures set for both columns and injector were 80 °C. The calibration gases include H2, CO, CO2 and C1-C3 hydrocarbon gases (CH4, C2H4, C2H6, C3H6, and C3H8).

3 Results and discussion 3.1 Overall yield from ex-situ catalytic upgrading process Product yield under various reaction conditions is illustrated in Table 1. To investigated the effects of different catalyst-to-lignin ratios (0.1, 0.2, 0.3, and 0.4) on the product yield, constant catalytic temperature of 450

was used in the section. Pyrolysis of lignin without

catalyst was studied as a blank control group and the temperature of the secondary empty reactor was also set at 450

. As shown in Table 1, compared with non-catalytic pyrolysis,

the ex-situ catalytic upgrading process reduced bio-oil yield and improved gas yield. Devolatilization reaction occurred during the pyrolysis of lignin and a lot of vapors was produced. For ex-situ catalytic upgrading, however, the vapors had to pass through the secondary reactor coupled with a certain amount of catalyst and underwent the secondary decomposition reaction. Thus, more condensable volatiles were converted to noncondensable gaseous product. Besides, the oxygen element in the obtained oxygenated compounds was rejected as CO and CO2 because of the deoxygenation reaction induced by the zeolite HZSM-5 (Adjaye & Bakhshi, 1995; Corma et al., 2007), which also contributed to decrease bio-oil yield and increase gas yield. Moreover, an increase in the catalyst-tolignin ratio hindered the bio-oil production while promoted the gas production because more catalyst loading in the secondary reactor represented higher catalytic effect as well as a longer residence time of primary pyrolytic vapors, leading to enhanced decomposition

and deoxygenation reactions (Atutxa et al., 2005). Effects of various catalytic temperatures (250, 350, 450, and 550

) on the product yield were also investigated and the catalyst-to-

lignin ratio was set at 0.3. It was observed that no significant change in the yields of bio-oil and gas at low catalytic temperatures (250-350

), indicating the weak secondary

decomposition at lower temperatures. Furthermore, Williams and Nugranad (2000) suggested that HZSM-5 favored the removal of oxygen as water at lower temperatures and as carbon oxides at higher temperatures. Thus, lower catalytic temperatures did not change bio-oil and gas yields dramatically. The continued increasing catalytic temperature from 350 to 550

reduced the bio-oil yield and increased the gas yield because higher catalytic

temperatures indicate more severe conditions for cracking of primary pyrolytic vapors, resulting in more gas product formation. During the non-catalytic or ex-situ catalytic upgrading process, a relatively large amount of char (around 44 wt.%) was produced, because the reactive groups of lignin including methoxyl and hydroxyl groups are easy to be polymerized to form char (Mu et al., 2013). It was observed that changes in char yield were insignificant under various catalyst-to-lignin ratios and catalytic temperatures because the pyrolytic reactions are mainly affected by pyrolysis temperature (Wang et al., 2014a).

The coke yield under different conditions was also presented in Table 1. Only a very small amount of coke based on lignin (0.41-1.30 wt.%) was obtained during the ex-situ upgrading process. This is very different from other studies (Luo & Resende, 2016; Wang et al., 2014a), in which the catalytic process was performed in a micro pyrolysis system and large quantity of coke was deposited on the catalyst because the residence time of the vapors was relatively short in the micro reactor and large catalyst-to-reactant ratio was needed to ensure the enough contact between pyrolysis vapors and catalyst, which could, in turn, give

rise to the high coke yield. However, in the present work, a microwave-assisted bench-scale pyrolysis reactor coupled with a secondary catalyst reactor was used for ex-situ upgrading and it allowed much lower catalyst-to-lignin ratio to maintain good catalytic effect and reduced the possibility of direct contact between the catalyst and raw reactant. Besides, the longer catalytic reaction time gives more chances for the contact between primary pyrolytic vapors and catalyst as well as the conversion of coke to gas product. By further observation, the coke yield based on reactant was favored at higher catalyst-to-lignin ratios and lower catalytic temperatures. This is because higher catalyst-to-lignin ratios gave more opportunities for the deposition of pyrolytic vapors on the catalyst, while higher catalytic temperatures were favorable for the decomposition of macromolecule volatiles to smaller molecules, which were not easy to be deposited on the catalyst. In addition, higher catalytic temperatures were also conducive to the decomposition of coke. For a better analysis of the coke deposition extent, the coke yield was also determined based on catalyst weight, which decreased with the increase in catalyst loading amount. This is because higher catalytic effect obtained at higher catalyst-to-lignin ratios reduced the coke deposition on the catalyst. Also, according to the aforementioned reasons, higher catalytic temperatures resulted in lower coke yield based on catalyst weight.

3.2 Analysis of bio-oil product from ex-situ catalytic upgrading process The chemical compounds produced from lignin pyrolysis can be classified into phenol, methoxy phenols, alkyl phenols, other oxygenates (such as acids, aldehydes, and ketones), polycyclic aromatic hydrocarbons (PAHs), and monocyclic aromatic hydrocarbons (MAHs). Fig. 2 shows the chemical selectivity of the bio-oil product as a response to various catalyst-to-lignin ratios and catalytic temperatures. As a control group, microwave-

assisted pyrolysis of lignin without catalyst resulted in main phenolics in the organic compounds because of the highly polymerized phenyl propane structure of lignin. High selectivity of methoxy phenols (73.7%) in the organic compounds indicated the lignin sample was comprised of a high proportion of guaiacyl (G-type) and syringyl (S-type) structure. A few of other oxygenates were produced due to the cracking of side chains in lignin. A negligible amount of PAHs and MAHs were observed, suggesting that aromatization through deoxygenation of phenols or Diels-Alder reaction of side chains occurred weakly during the microwave-assisted non-catalytic pyrolysis process. As shown in Fig. 2(a), the selectivity of methoxy phenols decreased dramatically with the increase of catalyst-to-lignin ratio. This could ascribe to the strong deoxygenation capacity of HZSM-5, resulting in the significant removal of methoxyl groups bonded to lignin structure through cracking reaction (Fan et al., 2017a). In guaiacol, CAr-OCH3 (356 kJ/mol) showed lower bond energy than CAr-OH (414 kJ/mol), making it easier to be cracked as a result of the formation of methoxyl-free phenols. Ben and Ragauskas (2011) also suggested that HZSM5 improved the cracking of methoxyl groups but showed less effect on the removal of phenolic hydroxyl groups. The proportion of phenol should have increased because of the conversion of methoxy phenols through demethoxylation reaction. However, during the demethoxylation of phenols, the alkylation reaction also occurred simultaneously induced by HZSM-5 (Liu et al., 2014), converting more phenol to alkyl phenols. This explains the phenomenon that the phenol proportion remained unchanged while the alkyl proportion increased from 14.6% at non-catalytic pyrolysis to 31.9% at 0.2 of catalyst-to-lignin ratio. On the other hand, a steady upward trend was observed in the chemical selectivity of PAHs and MAHs as the catalyst-to-lignin ratio increased from 0 to 0.3 due to the aromatization induced by HZSM-5. To and Resasco (2014) indicated the phenolic compounds could

convert to aromatics through passing through the “phenolic pool” in the catalyst framework and undergoing a series of reactions. Thilakaratne et al. (2016) also proposed the possible mechanism for converting phenol to aromatics, which was related to the formation and repolymerization of a series of active radicals including aryl, phenoxy, hydroxyl and hydrogen radicals. However, it is generally agreed that the formation of aromatics through direct cracking of CAr-OH bond in the phenol is difficult (Mullen & Boateng, 2010), which could be attributed to the high bond energy of CAr-OH. A plausible mechanism of aromatization of phenols was also proposed in the present study. The guaiacols were converted to alkyl phenols firstly through demethoxylation and alkylation reactions induced by HZSM-5. Meanwhile, the alkyl phenols passed through the aromatics shape-selective pores of HZSM-5 and underwent much easier removal of phenolic hydroxyl groups because the alkylation of phenol lowered the bond energy of CAr-OH. As a result, more aromatics were produced during the catalytic process and the increased ratio of catalyst to reactant enhanced the production of aromatics (PAHs and MAHs) due to the higher catalytic effect. That also interprets that the proportion of alkyl phenols decreased from 31.9% at 0.2 of catalyst-to-lignin ratio to 24.7% at 0.4 of catalyst-to-lignin ratio. It should be noted that no significant change was observed in the selectivity of aromatics as the catalyst-to-lignin increased from 0.3 to 0.4. It suggests that 0.3 was the optimal catalyst-tolignin ratio for the conversion of lignin to aromatics and it is not necessary to study the effects of higher catalyst-to-lignin ratios (>0.4) on the ex-situ catalytic upgrading process. It needs to be noted that the slight decreasing selectivity in other oxygenates because of the deoxygenation and Diels-Alder reactions (Adjaye & Bakhshi, 1995) also contributed to the increasing production of aromatics.

Fig. 2(b) shows the chemical selectivity of bio-oil at various catalytic temperature. Compared with non-catalytic pyrolysis process, during which up to 73.7% of methoxy phenols together with 8.3% of phenol and 14.6% of alkyl phenols were produced, the exsitu catalytic upgrading process at 250

of catalytic temperature decreased the selectivity

of methoxyl phenols to 64.6% and increased the selectivity of phenol and alkyl phenols to 14.0% and 18.1%. This indicates that HZSM-5 showed a certain extent of catalytic effect even at relatively low temperature and promoted the demethoxylation of phenols, leading to the conversion of methoxy phenols to phenol and alkyl phenols. As the catalytic temperature increased from 250

to 550

, the selectivity of methoxy phenols decreased

significantly from 64.6% to 15.5% while that of alkyl phenols increased steadily from 18.1% to 37.6%. This suggests that HZSM-5 at higher catalytic temperatures were conducive to the demethoxylation and alkylation of phenols, converting more methoxy phenols to alkyl phenols. Several reports also indicated that demethoxylation and alkylation reactions were favored at higher reaction temperatures (Jiang et al., 2010; Patwardhan et al., 2011). Besides, the enhanced alkylation at higher temperatures also contributed to the reduction of phenol selectivity from 14.0% at 250

to 7.7% at 450

. However, a slight increasing

selectivity of phenol was observed as the catalytic temperature increased from 450 550

to

. This is because demethoxylation reaction outweighed alkylation reaction at a higher

temperature, more phenol remained as the final product before it was converted to alkyl phenols. The proportion of other oxygenates decreased slightly with the increasing of catalytic temperature until all of the oxygenates were converted at 550

of catalytic

temperature because of the improved aromatization and decomposition at higher temperatures, resulting in higher production of aromatics as well as some gas product. Just a small amount of PAHs and MAHs were produced at relatively low catalytic temperatures

(250-350

), indicating that aromatization of phenols with HZSM-5 was not favored at low

temperatures. Besides, HZSM-5 with low temperature showed little effect on the aromatization, resulting in the unchanged aromatic selectivity under 350

of catalytic

temperature. Interestingly, as the catalytic temperature kept increasing, a significant upward trend in the selectivity of PAHs and MAHs was observed. This indicates that aromatization of phenols occurred increasingly at higher temperatures. It needs to be noted that no dramatic change was observed in the total aromatic selectivity as the catalytic temperature increased from 450 to 550

. Considering energy conservation, 450

was selected as the

optimal temperature for lignin conversion to aromatics and effects of higher catalytic temperatures (>550

) on ex-situ catalytic upgrading process were not further studied.

However, the selectivity of MAHs decreased from 11.4% to 8.9% as the catalytic temperature continued to increase from 450

to 550

. This is attributed to the flexibility

of the catalyst, HZSM-5, whose “effective pore size” could increase at higher temperatures (Yu et al., 2012), improving the selectivity of PAHs (mainly naphthalene and alkylated naphthalenes) while reducing the selectivity of MAHs.

3.3 Analysis of gas product from ex-situ catalytic upgrading process The gas product from lignin degradation was mainly composed of CO2, CO, H2, CH4, and a trace of C2-C3 hydrocarbons. CO2 and CO were usually originated from decarboxylation and decarbonylation, respectively (Wang et al., 2014b). The formation of hydrogen was attributed to the cracking of phenolic compounds while C1-C3 hydrocarbons were generally produced from the release of aliphatic side groups attached to the lignin phenolic structure. It should be noted that the scission of methoxyl groups also contributed to the formation of methane in addition to methanol (Shen et al., 2010). Fig. 3 shows the gas composition

affected by varied catalytic conditions. As shown in Fig. 3(a), CO2 was dominant in the gas product from non-catalytic pyrolysis process, taking up 54.1% in volume percentage. This suggests that decarboxylation reaction occurred dramatically during the pyrolysis process. Besides, a relatively large concentration of hydrogen (18.7%) was produced. This is different from other studies (Thring et al., 2000; Trinh et al., 2013), in which the production of hydrogen was negligible. It could be attributed to the advantage of the microwaveassisted pyrolysis system, which provided even and volumetric heating for lignin pyrolysis, triggering more cracking reactions as a result of higher hydrogen production. Dominguez et al. (2007) also suggested that microwave-assisted pyrolysis resulted in more hydrogen yield compared with conventional pyrolysis. The ex-situ catalytic upgrading process significantly changed the gas composition. A downward trend in CO2 concentration from 54.1% to 40.5% and an upward trend in CO concentration from 7.8% to 9.5% was observed as the catalystto-lignin ratio increased from 0 to 0.3. This could be the reason that decarboxylation reaction was favored over decarbonylation reaction at higher catalyst-to-lignin ratios (<0.3) (Wang et al., 2014b). However, as the catalyst-to-lignin ratio increased from 0.3 to 0.4, the concentration of CO2 increased. This could be attributed to higher residence time at higher catalyst-to-lignin ratios, which were conducive to decarboxylation to CO2 and decarbonylation to CO (Liu et al., 2014). Thus, the more dominant effect of residence time on those reactions at 0.4 of catalyst-to-lignin ratio resulted in the increasing production of CO2 as well as CO. The hydrogen concentration also increased from 18.7% of the noncatalytic process to 25.9% at 0.3 of catalyst-to-lignin ratio, suggesting that appropriate increase of catalyst-to-lignin ratio favored the cracking and reforming reactions. In addition, the steam reforming reactions could also occur during the catalytic process and resulted in the formation of H2 and CO (Xie et al., 2014a). The decreasing concentration of H2 was

observed as the catalyst-to-lignin ratio increased from 0.3 to 0.4. This could ascribe to the increasing concentration of other three main gas components including CO2, CO, and CH4. The concentration of CH4 in the gas product increased steadily with increasing catalyst-tolignin ratio because of more significant removal of methoxyl groups at higher catalyst-tolignin ratios. This is also consistent with the result that the methoxy phenols decreased significantly with the increase of catalyst-to-lignin ratio (Fig. 2(a)). Furthermore, the higher residence time induced by the increasing catalyst loading in the secondary reactor was favorable to cracking of light hydrocarbons to CH4. That is also the reason for the decreasing volume percentage of light hydrocarbons after the initial increase from 1.3% (non-catalytic process) to 3.1% (0.2 of catalyst-to-lignin ratio) due to the dehydration reaction induced by HZSM-5.

The effect of catalytic temperature on the gas composition at 0.3 of catalyst-to-lignin ratio was also investigated and shown in Fig. 3(b). Compared with non-catalytic pyrolysis process, the ex-situ catalytic upgrading process resulted in lower CO2 but higher H2 and CH4 concentration in the gas product regardless of the catalytic temperature. This suggests that the catalytic process over HZSM-5 was more in favor of cracking and dehydration reactions, which were responsible for the production of H2 and hydrocarbon gases. The increased catalytic temperature from 250 to 550

lowered the concentration of CO2 from

46.4% to 36.3% while increased that of CH4 from 19.0% to 23.3% and CO from 7.6% to 9.9%. This is because higher catalytic temperatures were more favorable to the demethoxylation and dehydration to CH4 and decarbonylation to CO much than decarboxylation to CO2 (Trinh et al., 2013). It seems that the H2 concentration was higher at a higher catalytic temperature (550

) and kept unchanged at lower than 550

of

catalytic temperature. It has been reported that the production of H2 induced by cracking reaction needs to be enhanced by much higher temperatures (Trinh et al., 2013). It also should be noted that higher gas yield was obtained at higher catalytic temperatures (Table 1). Thus, although the H2 concentration seemed unchanged from 250 to 550

, the

formation of H2 was still promoted by the increased catalytic temperatures. A few of light hydrocarbon gases were also produced and no dramatic change was observed in the concentration. Higher catalytic temperatures favored the dehydration of side aliphatic groups, which was beneficial to the formation of C2-C3 hydrocarbon gases. However, the enhanced cracking reaction at higher catalytic temperatures could also result in more conversion of C2-C3 hydrocarbon gases to CH4.

3.4 Comparison of in-situ and ex-situ catalytic upgrading process To compare in-situ and ex-situ catalytic upgrading of lignin pyrolytic vapors, the pyrolysis temperature and catalyst-to-lignin ratio were kept constant, at 550

and 0.3, respectively,

for both two processes. Because the catalytic temperature during the in-situ process was determined by the pyrolysis temperature (550 process was also set at 550

), the catalytic temperature for ex-situ

. To avoid the influence of secondary-reactor temperature, the

temperature of the secondary empty reactor for in-situ process was kept at 550

. The

product yields from two catalytic upgrading processes were summarized in Fig. 4. Compared with 28.95 wt.% of bio-oil and 39.45 wt.% of char for in-situ catalytic upgrading process, ex-situ catalytic upgrading process generated lower bio-oil yield (23.3 wt.%) but higher char yield (43.63 wt.%). This disagrees with those studies on catalytic pyrolysis of biomass, in which higher char yield was produced from in-situ catalytic upgrading process than from ex-situ catalytic upgrading process (Luo & Resende, 2016; Wang et al., 2014a).

This could be attributed to the low melting point of lignin. For ex-situ catalytic upgrading process, in which lignin and catalyst were separately placed in two different reactors, lignin was easy to be agglomerated and difficult to get uniform heating during pyrolysis, resulting in large char formation. While for in-situ catalytic upgrading process, in which lignin and catalyst were pre-mixed uniformly prior to pyrolysis, the agglomeration of lignin could be mitigated by the mixing of catalyst, leading to lower char formation and higher bio-oil and gas conversion rate. In addition, it was reported that HZSM-5 enhances the release of lignin-derived phenols from lignin polymer when lignin and catalyst are in direct contact (Kim et al., 2015), which could increase the yield of bio-oil. Yan et al. (2017) also indicated that the strong acidic sites and uniform pore system of the catalyst are conducive to the conversion of lignin-derived monomers or fine chemicals from lignin. However, it was observed that the gas yield from in-situ catalytic upgrading process was similar to that from ex-situ catalytic upgrading process. This is because ex-situ upgrading resulted in longer residence time of vapors and more contacting chances for primary pyrolytic vapors and catalyst, improving the cracking reaction and converting more condensable volatiles to gas product.

Fig. 5 shows the chemical selectivity of bio-oil from in-situ and ex-situ catalytic upgrading process. Compared with 73.7% of methoxy phenols from non-catalytic pyrolysis process, both two catalytic upgrading processes produced fewer methoxy phenols (19.7% and 15.5%, respectively). This suggests that both in-situ and ex-situ processes were effective in catalytic reforming of lignin pyrolytic vapors catalyzed by HZSM-5. The fact that no other oxygenates was observed in both processes also suggests the relatively high catalytic effect during the two processes. Much higher selectivity of methoxy phenols was obtained from

in-situ process than from ex-situ process, indicating that ex-situ process provided a higher catalytic effect on the cracking of methoxyl groups. This is because the primary pyrolytic vapors had to exposed to the catalyst bed more completely during the ex-situ catalytic upgrading process. Besides, the relatively long catalyst bed in the ex-situ catalytic upgrading process provided a longer residence time for pyrolytic vapors to contact with the catalyst. However, in-situ catalytic upgrading process produced a higher proportion of alkyl phenols than ex-situ catalytic upgrading process. It has been reported that higher temperatures favored the alkylation of phenols (Fan et al., 2017b). During the catalytic pyrolysis process at a relatively high pyrolytic temperature (550

), the alkyl phenols could

be formed largely. Besides, the removal of methoxyl groups enhanced by HZSM-5 could also convert more methoxy phenols to alkyl phenols. However, due to the higher catalytic effect provided by ex-situ catalytic upgrading process, more alkyl phenols could be converted to aromatics, resulting in a lower proportion of alkyl phenols in bio-oil from exsitu catalytic upgrading process accompanied by higher aromatic proportion.

As shown in Fig. 6, compared with the gas product from in-situ catalytic upgrading process, a lower percentage of CO2 and a higher percentage of H2 and CO were obtained from exsitu catalytic upgrading process. This suggests that the cracking reaction to H2 and decarbonylation reaction to CO were favored over the decarboxylation reaction to CO2 during the ex-situ catalytic upgrading process. One probable reason is that the secondary catalyst bed provided more contact time for pyrolytic vapors and catalyst, enhancing the cracking reactions. Besides, the steam reforming reaction, which contributes to the formation of H2 and CO, could also be improved by more contact time between pyrolytic vapors and catalyst. However, a little higher percentage of hydrocarbon gases was observed

in the gas product from ex-situ catalytic upgrading process. This could be attributed to the reason that less residence time provided by in-situ catalytic upgrading process resulted in less conversion of light olefins to aromatics.

4. Conclusions Effects of catalyst-to-lignin ratio and catalytic temperature on the ex-situ catalytic upgrading process were studied. Lower bio-oil yield and higher gas yield was obtained at higher catalyst-to-lignin ratios as well as higher catalytic temperatures. Appropriate catalyst-to-lignin ratio (0.3) together with higher catalytic temperatures were favorable for the conversion of methoxy phenols to alkyl phenols and aromatics and syngas formation. Compared with in-situ upgrading, ex-situ upgrading produced a lower yield of bio-oil with higher aromatic selectivity. The similar gas yield was observed for both processes but more syngas and less CO2 was obtained from ex-situ upgrading process.

Acknowledgements The present work is supported in part by National Natural Science Foundation of China (No. 21466022; No. 21766019), Key Research and Development Program of Jiangxi Province (20161BBF60057; 20171BBF60023; GCXZ2014-124), International Science and Technology Cooperation Program of China (2014DFA61040), the Minnesota Environment and Natural Resources Trust Fund as recommended by the Legislative-Citizen Commission on Minnesota Resources (LCCMR), Xcel Energy, and University of Minnesota Center for Biorefining.

References

1. Adjaye, J., Bakhshi, N. 1995. Production of hydrocarbons by catalytic upgrading of a fast pyrolysis bio-oil. Part II: Comparative catalyst performance and reaction pathways. Fuel Process. Technol. 45(3), 185-202. 2. Anex, R.P., Aden, A., Kazi, F.K., Fortman, J., Swanson, R.M., Wright, M.M., Satrio, J.A., Brown, R.C., Daugaard, D.E., Platon, A. 2010. Techno-economic comparison of biomass-to-transportation fuels via pyrolysis, gasification, and biochemical pathways. Fuel 89, S29-S35. 3. Atutxa, A., Aguado, R., Gayubo, A.G., Olazar, M., Bilbao, J. 2005. Kinetic description of the catalytic pyrolysis of biomass in a conical spouted bed reactor. Energ. Fuel. 19(3), 765-774. 4. Ben, H., Ragauskas, A.J. 2011. Pyrolysis of kraft lignin with additives. Energ. Fuel. 25(10), 4662-4668. 5. Borges, F.C., Du, Z., Xie, Q., Trierweiler, J.O., Cheng, Y., Wan, Y., Liu, Y., Zhu, R., Lin, X., Chen, P., Ruan, R. 2014. Fast microwave assisted pyrolysis of biomass using microwave absorbent. Bioresour. Technol. 156, 267-274. 6. Bu, Q., Lei, H., Wang, L., Wei, Y., Zhu, L., Zhang, X., Liu, Y., Yadavalli, G., Tang, J. 2014. Bio-based phenols and fuel production from catalytic microwave pyrolysis of lignin by activated carbons. Bioresour. Technol. 162, 142-147. 7. Calvo Flores, F.G., Dobado, J.A. 2010. Lignin as renewable raw material. ChemSusChem 3(11), 1227-1235. 8. Corma, A., Huber, G.W., Sauvanaud, L., O'connor, P. 2007. Processing biomass-derived oxygenates in the oil refinery: catalytic cracking (FCC) reaction pathways and role of catalyst. J. Catal. 247(2), 307-327.

9. Dominguez, A., Menéndez, J., Fernandez, Y., Pis, J., Nabais, J.V., Carrott, P., Carrott, M.R. 2007. Conventional and microwave induced pyrolysis of coffee hulls for the production of a hydrogen rich fuel gas. J. Anal. Appl. Pyrol. 79(1), 128-135. 10. Du, Z., Li, Y., Wang, X., Wan, Y., Chen, Q., Wang, C., Lin, X., Liu, Y., Chen, P., Ruan, R. 2011. Microwave-assisted pyrolysis of microalgae for biofuel production. Bioresour. Technol. 102(7), 4890-4896. 11. Fan, L., Chen, P., Zhang, Y., Liu, S., Liu, Y., Wang, Y., Dai, L., Ruan, R. 2017a. Fast microwave-assisted catalytic co-pyrolysis of lignin and low-density polyethylene with HZSM-5 and MgO for improved bio-oil yield and quality. Bioresour. Technol. 225, 199-205. 12. Fan, L., Zhang, Y., Liu, S., Zhou, N., Chen, P., Cheng, Y., Addy, M., Lu, Q., Omar, M.M., Liu, Y., Wang, Y., Dai, L., Anderson, E., Peng, P., Lei, H., Ruan, R. 2017b. Bio-oil from fast pyrolysis of lignin: Effects of process and upgrading parameters. Bioresour. Technol. 241, 1118-1126. 13. Fan, L., Zhang, Y., Liu, S., Zhou, N., Chen, P., Liu, Y., Wang, Y., Peng, P., Cheng, Y., Addy, M. 2017c. Ex-situ catalytic upgrading of vapors from microwave-assisted pyrolysis of low-density polyethylene with MgO. Energ. Convers. Manage. 149, 432-441. 14. Gamliel, D.P., Du, S., Bollas, G.M., Valla, J.A. 2015. Investigation of in situ and ex situ catalytic pyrolysis of miscanthus× giganteus using a PyGC–MS microsystem and comparison with a bench-scale spouted-bed reactor. Bioresour. Technol. 191, 187-196.

15. Gosselink, R., De Jong, E., Guran, B., Abächerli, A. 2004. Co-ordination network for lignin-standardisation, production and applications adapted to market requirements (EUROLIGNIN). Ind. Crop. Prod. 20(2), 121-129. 16. Iisa, K., French, R.J., Orton, K.A., Yung, M.M., Johnson, D.K., ten Dam, J., Watson, M.J., Nimlos, M.R. 2016. In-situ and ex-situ catalytic pyrolysis of pine in a benchscale fluidized bed reactor system. Energ. Fuel. 30(3), 2144-2157. 17. Jackson, M.A., Compton, D.L., Boateng, A.A. 2009. Screening heterogeneous catalysts for the pyrolysis of lignin. J. Anal. Appl. Pyrol. 85(1), 226-230. 18. Jiang, G., Nowakowski, D.J., Bridgwater, A.V. 2010. Effect of the temperature on the composition of lignin pyrolysis products. Energ. Fuel. 24(8), 4470-4475. 19. Kim, J.-Y., Lee, J.H., Park, J., Kim, J.K., An, D., Song, I.K., Choi, J.W. 2015. Catalytic pyrolysis of lignin over HZSM-5 catalysts: Effect of various parameters on the production of aromatic hydrocarbon. J. Anal. Appl. Pyrol. 114, 273-280. 20. Liu, C., Wang, H., Karim, A.M., Sun, J., Wang, Y. 2014. Catalytic fast pyrolysis of lignocellulosic biomass. Chem. Soc. Rev. 43(22), 7594-7623. 21. Luo, G., Resende, F.L.P. 2016. In-situ and ex-situ upgrading of pyrolysis vapors from beetle-killed trees. Fuel 166, 367-375. 22. Mihalcik, D.J., Mullen, C.A., Boateng, A.A. 2011. Screening acidic zeolites for catalytic fast pyrolysis of biomass and its components. J. Anal. Appl. Pyrol. 92(1), 224-232. 23. Mu, W., Ben, H., Ragauskas, A., Deng, Y. 2013. Lignin pyrolysis components and upgrading—technology review. BioEnerg. Res. 6(4), 1183-1204. 24. Mukkamala, S., Wheeler, M.C., van Heiningen, A.R., DeSisto, W.J. 2012. Formateassisted fast pyrolysis of lignin. Energ. Fuel. 26(2), 1380-1384.

25. Mullen, C.A., Boateng, A.A. 2010. Catalytic pyrolysis-GC/MS of lignin from several sources. Fuel Process. Technol. 91(11), 1446-1458. 26. Patwardhan, P.R., Brown, R.C., Shanks, B.H. 2011. Understanding the fast pyrolysis of lignin. ChemSusChem 4(11), 1629-36. 27. Shen, D.K., Gu, S., Luo, K.H., Wang, S.R., Fang, M.X. 2010. The pyrolytic degradation of wood-derived lignin from pulping process. Bioresour. Technol. 101(15), 6136-6146. 28. Thilakaratne, R., Tessonnier, J.-P., Brown, R.C. 2016. Conversion of methoxy and hydroxyl functionalities of phenolic monomers over zeolites. Green Chem. 18(7), 2231-2239. 29. Thring, R.W., Katikaneni, S.P., Bakhshi, N.N. 2000. The production of gasoline range hydrocarbons from Alcell® lignin using HZSM-5 catalyst. Fuel Process. Technol. 62(1), 17-30. 30. To, A.T., Resasco, D.E. 2014. Role of a phenolic pool in the conversion of m-cresol to aromatics over HY and HZSM-5 zeolites. Appl. Catal. A-Gen. 487, 62-71. 31. Toledano, A., Serrano, L., Pineda, A., Romero, A.A., Luque, R., Labidi, J. 2014. Microwave-assisted depolymerisation of organosolv lignin via mild hydrogen-free hydrogenolysis: catalyst screening. Appl. Catal. B-Enviro. 145, 43-55. 32. Trinh, T.N., Jensen, P.A., Sárossy, Z., Dam-Johansen, K., Knudsen, N.O., Sørensen, H.R., Egsgaard, H. 2013. Fast pyrolysis of lignin using a pyrolysis centrifuge reactor. Energ. Fuel. 27(7), 3802-3810. 33. Wang, K., Johnston, P.A., Brown, R.C. 2014a. Comparison of in-situ and ex-situ catalytic pyrolysis in a micro-reactor system. Bioresour. Technol. 173, 124-131.

34. Wang, K., Kim, K.H., Brown, R.C. 2014b. Catalytic pyrolysis of individual components of lignocellulosic biomass. Green Chem. 16(2), 727-735. 35. Williams, P.T., Nugranad, N. 2000. Comparison of products from the pyrolysis and catalytic pyrolysis of rice husks. Energy 25(6), 493-513. 36. Xie, Q., Borges, F.C., Cheng, Y., Wan, Y., Li, Y., Lin, X., Liu, Y., Hussain, F., Chen, P., Ruan, R. 2014a. Fast microwave-assisted catalytic gasification of biomass for syngas production and tar removal. Bioresour. Technol. 156, 291-296. 37. Xie, Q., Peng, P., Liu, S., Min, M., Cheng, Y., Wan, Y., Li, Y., Lin, X., Liu, Y., Chen, P., Ruan, R. 2014b. Fast microwave-assisted catalytic pyrolysis of sewage sludge for bio-oil production. Bioresour. Technol. 172, 162-168. 38. Yan, K., Liu, Y., Lu, Y., Chai, J., Sun, L. 2017. Catalytic application of layered double hydroxide-derived catalysts for the conversion of biomass-derived molecules. Catal. Sci. Technol. 7(8), 1622-1645. 39. Yin, C. 2012. Microwave-assisted pyrolysis of biomass for liquid biofuels production. Bioresour. Technol. 120, 273-284. 40. Yu, Y., Li, X., Su, L., Zhang, Y., Wang, Y., Zhang, H. 2012. The role of shape selectivity in catalytic fast pyrolysis of lignin with zeolite catalysts. Appl. Catal. AGen. 447, 115-123. 41. Zhang, M., Resende, F.L.P., Moutsoglou, A. 2014. Catalytic fast pyrolysis of aspen lignin via Py-GC/MS. Fuel 116, 358-369.

Figure captions Fig. 1. Schematic diagram of ex-situ catalytic microwave-assisted pyrolysis system. (1) Microwave oven; (2) quartz reactor; (3) K-type thermocouple; (4) quartz wool; (5) catalyst bed; (6) fritted disc; (7) flask; (8) condensing system; (9) connection for the vacuum pump. Fig. 2. Effects of (a) catalyst-to-lignin ratio and (b) catalytic temperature on the chemical selectivity of bio-oil products from ex-situ process. Fig. 3. Effects of (a) catalyst-to-lignin ratio and (b) catalytic temperature on the gas composition from ex-situ process. Fig. 4. Product yield from in-situ and ex-situ catalytic upgrading of lignin pyrolytic vapors. Fig. 5. Chemical selectivity of bio-oil from in-situ and ex-situ catalytic upgrading of lignin pyrolytic vapors. Fig. 6. Gas composition from in-situ and ex-situ catalytic upgrading of lignin pyrolytic vapors.

Fig. 1. Schematic diagram of ex-situ catalytic microwave-assisted pyrolysis system. (1) Microwave oven; (2) quartz reactor; (3) K-type thermocouple; (4) quartz wool; (5) catalyst bed; (6) fritted disc; (7) flask; (8) condensing system; (9) connection for the vacuum pump.

80

0 0.1 0.2 0.3 0.4

Chemical selectivity (%)

70 60 50 40 30 20 10 0 tes ols ols nol Phe y phen l phen oxgena y x Alk Other tho Me

Hs Hs PA MA

(a) 70

250 °C 350 °C 450 °C 550 °C

Chemical selectivity (%)

60 50 40 30 20 10 0 ols ols nol ates Phe y phen l phen oxgen y x Alk Other tho Me

Hs Hs PA MA

(b) Fig. 2. Effects of (a) catalyst-to-lignin ratio and (b) catalytic temperature on the chemical selectivity of bio-oil products from ex-situ process.

60

0 0.1 0.2 0.3 0.4

Volume percentage (vol.%)

50 40 30 20 10 0

e e e n ns ioxid Hydroge Methan monoxid rocarbo d n o on t hyd Carb Carb Ligh

Volume percentage (vol.%)

(a) 50 45 40 35 30 25 20 15 10 5 0

250 °C 350 °C 450 °C 550 °C

ide gen ethane noxide arbons diox Hydro o M oc n o on m ht hydr b r Carb a C Lig

(b) Fig. 3. Effects of (a) catalyst-to-lignin ratio and (b) catalytic temperature on the gas composition from ex-situ process.

50

Product yield (wt.%)

40

In-situ Ex-situ

30 20 10 0 Bio-oil

Gas

Char

Fig. 4. Product yield from in-situ and ex-situ catalytic upgrading of lignin pyrolytic vapors.

60

Chemical selectivity (%)

50

In-situ Ex-situ

40 30 20 10 0

ol ols tes PAHs AHs ols M Phen y phen yl phen oxgena hox Alk Other Met

Fig. 5. Chemical selectivity of bio-oil from in-situ and ex-situ catalytic upgrading of lignin pyrolytic vapors.

Volume percentage (vol.%)

50 In-situ Ex-situ

40 30 20 10 0

io on d Carb

xide

s e e n roge Methan onoxid ocarbon r m Hyd on hyd Carb Light

Fig. 6. Gas composition from in-situ and ex-situ catalytic upgrading of lignin pyrolytic vapors.

Table captions Table 1. Product yield under various catalyst-to-lignin ratios and catalytic temperatures.

Table 1. Product yield under various catalyst-to-lignin ratios and catalytic temperatures. Run Catalyst-toCatalytic Bio-oil yield Gas yield Char yield lignin ratio temperature ( ) (wt.%) (wt.%) (wt.%) R-1 0 450 34.11 ± 0.37 21.99 ± 0.29 43.90 ± 0.66 R-2 0.1 450 29.24 ± 0.62 26.40 ± 1.28 43.95 ± 0.62 R-3 0.2 450 25.89 ± 0.44 29.30 ± 1.16 44.23 ± 0.67 R-4 0.3 450 26.01 ± 0.04 28.36 ± 0.60 44.84 ± 0.50 R-5 0.4 450 21.75 ± 0.79 34.08 ± 1.74 43.42 ± 0.98 R-6 0.3 250 28.11 ± 0.22 26.30 ± 0.01 44.29 ± 0.16 R-7 0.3 350 28.19 ± 0.27 26.91 ± 0.99 43.87 ± 0.76 R-8 0.3 550 23.30 ± 0.26 32.18 ± 0.19 43.63 ± 0.04

34

Coke yield (w based on react 0.41 ± 0.05 0.58 ± 0.06 0.80 ± 0.07 0.76 ± 0.02 1.30 ± 0.06 1.04 ± 0.04 0.89 ± 0.03

Highlights 1. Vapors from microwave pyrolysis of lignin were upgraded in a separated catalyst bed. 2. Ex-situ upgrading decreased methoxy phenols production significantly. 3. Up to 39.1% of syngas together with 24.6% of hydrocarbon gases were obtained. 4. Less char was produced from in-situ catalytic pyrolysis of lignin. 5. Ex-situ upgrading produced more aromatics and fewer phenols than in-situ upgrading.

35