Peter Keliona Wani Likun ,Huiyan Zhang,Yuyang Fan

1 Key Laboratory of Thermal Energy Conversion and Control Ministry of Education,Southeast University,Nanjing 210096,China

2 Department of Mechanical Engineering,Juba University,Juba,P.O.Box 82,South Sudan

Keywords:Torrefaction Biomass Plastics Co-Pyrolysis Dual-catalyst Aromatics Selectivity

ABSTRACT To increase the low yield and selectivity of aromatic hydrocarbons during the biomass pyrolysis process,we torrefied the biomass and then co-pyrolyzing with plastics such as high-density polyethylene(HDPE),polystyrene (PS),ethylene-vinyl acetate (EVA) and polypropylene (PP) and also single and dual catalyst layouts were investigated by Py-GC/MS.The results showed that non-catalytic fast pyrolysis(CFP)of raw bagasse (RBG) generated no aromatics.After torrefaction non-CFP of torrefied bagasse (TBG) generated low aromatic yield.Indicating that torrefaction would enhance the proportion of aromatics during the pyrolysis process.The CFP of TBG200°C and TBG240°C over ZSM-5 produced the total aromatic yield of 1.96 and 1.88 times higher,respectively,compared to non-CFP of TBG.Furthermore,the addition of plastic could increase H/Ceff ratio of the mixture,consequently,increase the yield of aromatic compounds.Among the various torrefied-bagasse/plastic mixtures,the CFP of TBG/EVA(7:3 ratio)mixture generated the highest the total aromatic yield of 7.7 times more than the CFP of TBG alone.The dual catalyst layout could enhance the yield of aromatics hydrocarbons.The dual-catalytic co-pyrolysis of TBG200°C/plastic(1:1)ratio over USY(ultra-stable Y zeolite)/ZSM-5,improved the total aromatics yield by 4.33 times more than the catalytic pyrolysis of TBG200oC alone over ZSM-5 catalyst.The above results showed that the yield and selectivities of light aromatic hydrocarbons can be improved via catalytic co-pyrolysis and dual catalytic co-pyrolysis of torrefied-biomass with plastics.

1.Introduction

The management and energy recovery from plastic wastes and biomass is a subject of growing interest due to increasing volumes of plastic wastes and the associated environmental concerns and demand for clean energy worldwide.Plastics materials consist of different types of polymers,such as polyvinyl chloride(PVC),polyethylene (PE),polypropylene (PP),polystyrene (PS) and polyethylene terephthalate (PET),which are extensively used in various forms on daily basis [1],and their disposal may lead to serious environmental problems.Incineration technique is the common approach to rid the accumulation of plastics garbage,which is often accompanied by emissions such as NOx,N2O,SO2,HCl,Cl,and dioxins [2,3].However,plastics are essentially petroleum derivatives and represents untapped energy resource;it can be easily converted back to petroleum products with proper emission control.Lignocellulose biomass,on the other hand,is a huge energy reserve and widely distributed across the globe;and is expected to play a significant role in the production of liquid biofuels in the near future.

The catalytic fast pyrolysis (CFP) is the most preferred technique among other several thermochemical and biochemical approaches for converting lignocellulosic biomass directly to a liquid product called pyrolysis oil or bio-oil[4,5].The bio-oil obtained from this process contained high oxygen and water contents,acidic,poor rheological properties,immiscible,unstable during storage,and low heating value[3,6-10],making it less competitive with the premium fossil fuels and limiting its use in most important applications.The oxygen content(mass fraction) of the pyrolysis oil is high(about 35%-60%)[11-15]and has been identified to exist in several forms of oxygenated compounds including,acids,alcohols,phenols,ketones,aldehydes,esters,lignin-derived oligomers,etc.,and mostly found as water[12].Thus,it has to be deoxygenated before becoming a premium fuel.

Catalytic fast pyrolysis (CFP) is an approach that deoxygenates the pyrolysis vapors by a certainly suitable catalyst during the biomass pyrolysis process before the vapors condense to form aromatic and olefins hydrocarbons in a single reactor [16-20].It is simple,can be a cost-effective way to produce hydrocarbons in a single reactor,at atmospheric pressures.However,there are challenges to obtaining higher yields of aromatics hydrocarbon from the CFP of biomass.It has been reported in the literature that the low aromatics yield and rapid catalyst coking are attributed to hydrogen deficiencies and the high oxygen content of biomass materials[21,22].Moreover,other researchers found the coke yield depends on an effective hydrogen index (also called hydrogen to carbon ratio)(H/Ceff)of the biomass feedstocks.Other,researchers[19,23,24],stated that the H/Ceffratio which reflects the relative abundance of hydrogen in feedstocks is calculated according to Eq.(1)[25].For biomass and its derivatives,the value of H/Ceffratio is only between 0 and 0.3[26,27],thus,exhibits an acute deficiency of hydrogen content.According to Chen and several others [16],feedstocks with the H/Ceffratio ＜1.0 cannot be economically converted to premium petrochemicals over ZSM-5 zeolite.Moreover,Zhanget al.[23] observed a strong correlation between the H/Ceffratio of the feedstock and the content of hydrocarbons in the bio-oil during the conversion of different feedstocks over zeolite catalysts.

Here,H,O,N,S,and C,are respectively the mole percentages of hydrogen,oxygen,nitrogen,sulfur,and carbon in the feedstock.

Besides,rapid deactivation of the catalyst caused by catalyst coke can reduce the lifetime of the catalyst and hence need frequent catalyst regeneration could make the process daunting[16].Thus,these problems represent a practical challenge to the development of the catalytic pyrolysis process,which needs to be overcome.

It became obvious that to improve the aromatic yield,is by improving the overall H/Ceffratio of biomass materials and to reduce its oxygen content.The former can be improved by co-pyrolyzing biomass with a hydrogen-rich reactant with higher H/Ceffratio [28,29].Meanwhile the latter can be achieved through a torrefaction process.For example,hydrocarbon-based plastics such as polyethylene,polypropylene,and polystyrene are rich in hydrogen and haveH/Ceff.ratio around 2.0 and contain less or no oxygen [16,30].Attempt to minimize coke formation and improve hydrocarbons production from CFP of lignocellulose,researchers have tried co-feeding of biomass (pine wood) with alcohols such as methanol and butanol,and plastics such as polyethylene and polypropylene which have high H/Ceffratio [27,31].Besides improving the yield and quality of bio-oil,the co-pyrolysis of biomass with plastics have additional benefits of promoting a cleaner environment and energy recovery from waste plastics.The relevance of the co-pyrolysis process is its simplicity and economical;do not require any solvent,hydrogen,and less coke formation[8,32,33].

The inherent oxygen in the biomass can be thermally reducedviathe torrefaction process.Given that most of pyrolysis oil problems are fundamentally related to feedstock composition,catalyst characteristics,and probably reactor configuration which needs further in-depth investigations into how to eliminate them.

Recently,pretreatment of biomass before the fast pyrolysis has been identified as one of the promising techniques to improve the quality of bio-oil [34].Several pretreatment methods were employed in the literature,including water leaching,pretreatment with dilute acid and alkali,and hot water compressed water treatment.A kind of mild pyrolysis in the temperature between 200-300 °C,called torrefaction has been identified and can selectively adjust the physical and chemical properties of biomass;is considered as a potential pretreatment approach before pyrolysis of biomass[35,36].It can reduce the bio-oil oxygen content,significantly by losing CO,CO2,H2O,and the removal of acidic components such as acetic acid,which in turn will increase the calorific value for the pretreated biomass[37].The co-pyrolysis of torrefied-biomass and plastics has been uncommon in the literature;hence,it is not well understood.Therefore,combining feedstock pretreatment and catalytic co-pyrolysis with plastics can be a viable approach to increase the yield of aromatic hydrocarbons.Besides,a dualcatalyst layout system has been recently used to improve the quality of bio-oil obtain from catalytic fast pyrolysis process [38-40].

The purpose of this paper is to improve the yield and selectivity hydrocarbonsviaco-pyrolysis of torrefied-biomass with different plastics and the use of single and dual catalyst layouts.We first adjusted the composition of the biomassviatorrefaction process,at temperatures range of 180,200,240,270,and 300 °C and flow rate of 400 ml·min-1,and evaluate the treatment severity on the aromatic hydrocarbons yield.Then,the co-pyrolysis of the torrefied-biomass with plastics(such as high-density polyethylene(HDPE),polystyrene (PS),ethylene-vinyl acetate EVA) and polypropylene (PP) were investigated.Finally,the performances of single and dual catalyst layouts were examined.

2.Experimental

2.1.Materials

Sugarcane bagasse was collected from local fruits store in Nanjing City,China.While the plastics powder such as polypropylene(PP),polystyrene(PS),high-density polyethylene(HDPE)and ethylene-vinyl acetate (EVA)were purchase from a petrochemical factory in Shanghai City,China,in addition to tire powder was also obtained.The bagasse was first sun-dried followed by ovendrying overnight;before performing the torrefaction process.

2.2.Catalysts

Two types of zeolite catalysts,ZSM-5 and USY (ultra-stable Y zeolite) in the hydrogen form purchased from the catalyst plant of Nankai University,Tain,China were used in this study.The single catalytic fast pyrolysis process was when USY or ZSM-5 was used,whereas dual catalytic fast pyrolysis was when two catalysts were placed side by side,herein referred to as dual catalyst layout or dual catalyst design.

2.3.Torrefaction process

A vertical quartz tube fixed bed reactor of 76 mm inner diameter,80 mm outer diameter and 120 mm length was employed in this test.Usually,the torrefaction process is carried out in the temperature range between 200-300 °C [41,42].The products of the torrefaction process include solid as the targeted product and eliminated permanent gases such as hydrogen (H2),carbon dioxide(CO2),carbon dioxide (CO),and hydrocarbons such as methane(CH4),condensable mixtures containing mostly water,organic compounds and lipids [43,44].In each case,a certain amount of bagasse sample was loaded into the reactor and was heated from room temperature to final temperatures of 180,200,240,270 and 300 °C and then held at respective temperatures for 30 minutes before heating was terminated.Thus,five samples with different characteristics were obtained.Nitrogen (N2) at 400 ml·min-1flow rate was used as a purge gas.After each test,the solid product was cooled to ambient temperature,then collected and ground to 0.2-0.35 mm particle size then characterized.

2.4.ICP analysis

To determine whether the torrefaction process effect on the mineral content of biomass,an ICP Optical Emission Spectrometer Varian 720-ES was used to measure the mineral content of the raw and torrefied bagasse.

2.5.Py-GC/MS experiments

A sample size of 0.2 mg and a zeolite catalyst size of 0.5 mg placed on both sides of the sample separated by quartz wool were put into an open-ended quartz tube as shown in Fig.1.In the case of the dual catalyst layout,two different catalysts were placed on both ends of the quartz tube as designed in our previous work[38].The samples were pyrolyzed at 600 °C with a heating rate of 10 °C·ms-1for 20 s using a CDS Pyroprobe 6200 Pyrolyzer(CDS Analytical,USA) in an inert atmosphere.During pyrolysis,the pyrolyzer interface,and both the GC transfer line and oven were kept at 300°C and 280°C respectively.The released pyrolysis volatiles products were transferred online to a gas chromatograph(GC)/mass spectrometer/Flame ionization detector,GC-MS/FID system (Agilent Technologies 7890B/Agilent Technologies 5977B MSD,USA) which performed the separation,identification,and quantification respectively.The released gasses were separated by a RESTEK VMS capillary column,(30 m× 0.25 mm inner diameter of 0.25 μm,film thickness),and a split ratio of 1:50.The compounds were identified in MS and then quantified using FID.The GC oven temperature began at 40 °C and heated at 5 °C·min-1and then increased to 230 °C and held for 10 min.The front inlet was maintained at 280 °C.The mass spectra were obtained in the range of m/z ratio between 33 and 350.One disadvantage of GC/MS is that,no liquid products can be collected.Therefore,it is usually assumed that the chromatograph peak area percentage of a compound is considered proportional to its concentration.The resulting chromatographs were identified by comparing the mass spectrum of each sample with those in the NIST 4.1 library and other literature information [45].

3.Results and Discussion

3.1.Feedstock characterization

Fig.1.Samples designs for the non-catalytic and catalytic experiments.

The proximate and ultimate analyses of the raw,torrefiedbagasse,and different plastics are presented in Table 1.The moisture and volatile content decreased with increasing torrefaction temperature.In contrast,ash and fixed carbon content tend to increase.This is due to the increase in lignin content which high fixed carbon.The oxygen content also decreased with increased in torrefaction temperature,which is important for the efficient conversion of biomass to hydrocarbons.Similarly,hydrogen is also removed,which affects the H/C ratio of the biomass,which is a crucial element for forming hydrocarbons during pyrolysis process.The plastics contained little to no moisture and ash content,and but relatively high volatile matter.The tire,on the other hand,contained low amounts of volatile matter;and high ash and fixed carbon content.Table 2 shows the chemical composition of biomass.As anticipated,the content of hemicellulose and cellulose decreased with an increase in torrefaction temperature.While lignin increases as the torrefaction temperature is increased.This is because,the hemicellulose and cellulose are quickly to decompose with mass loss of hemicellulose occurred at 220-315°C and that of cellulose happened at 315-400 °C.However,lignin decomposed slowly in a wide temperature range between 160 and 900 °C[46].The ICP-OES results shown in Table 3 indicate that torrefaction has an insignificant effect on the mineral content in the biomass.That is say the minerals were not removed during biomass torrefaction process.

3.2.Thermogravimetric analysis

The pyrolysis behaviors of raw,torrefied bagasse and different plastics were shown in Fig.2;when heated from room temperature to the maximum temperature of 800°C at the ramping rate of 10°-C·min-1and nitrogen flow rate of 60 ml·min-1.Fig.2(a)shows the thermogravimetric (TG) profiles of raw and torrefied-bagasse.It can be seen that the mass of residues increased as the torrefaction temperature increases.While Fig.2(b)displays derivative thermogravimetric(DTG)curves.It can be observed,that the raw bagasse decomposed in three stages.It is regarded that,the first stage belonged to hemicellulose decomposition,while the second and the third for cellulose and lignin decomposition respectively.However,after torrefaction,the peaks that belong to hemicellulose and cellulose began to disappear depending on the severity of the torrefaction temperature.The starting reaction temperatures shift to higher values after pretreatment of bagasse,while the intermediate and the final reaction temperatures were not affected by the treatment process.DTGmax.values for the first decomposition stage were slightly influenced by the torrefaction process,whereas the second and the third stage remained unchanged as shown in Fig.2(b).

Fig.2(c)and(d)show the TG and DTG curves of HDPE,PS,EVA,and PP plastics thermal decomposition during the pyrolysis process.It can be seen that the plastics decomposed at a relatively higher and narrow temperature range than the biomass counterpart.The HDPE and PP decomposition characteristics were comparable,both decomposed between the temperature range of 393-494 °C,with the maximum mass loss occurring at 460 °C and 440.74 °C for HDPE and PP,respectively.While the PS degraded between temperatures of 364-449 °C.The EVA decomposed in two steps,the first step occurring at 311-400 °C with about mass loss of 20%,and step at 412-498 °C with maximum mass loss occurring at 351 and 460 °C,respectively.

3.3.BET analysis

The surface areas,pore-volume,and the porosity of the USY and ZSM-5 zeolite catalysts were analyzed by the BET surface and porosity analyzer (Micromeritics -ASAP 2020 PLUS HD88),and the results are shown in Table 4 and the nitrogen adsorption-desorption isotherms in Fig.3.All the catalysts showed type III isotherms,which are the characteristics of mesoporous materialsand nonuniform size and/or shape[47].Besides,the ZSM-5 catalyst had pore of 3.4 while USY had 2.5 indicating that they are mesoporous catalysts.In such a situation,adsorption occurs at very low relative pressures due to the interaction of the pore walls with the absorbate.Furthermore,the presence of hysteresis suggests the presence of mesoporous [47].Also,mesoporous catalysts are expected to minimize the diffusion resistance hindering the transport of large pyrolysis products such as anhydrosugars,phenolics,and lignin oligomers[48].The reactivity is expected to be boosted,as more sites become available on the surface and in the mesoporous [49].

Table 1Proximate and elemental analyses of feedstocks

Table 2Composition analysis of raw and torrefied-bagasse

Table 3Analysis of major and minor ash elements in raw and torrefied-bagasse (mg·L-1)

Table 4Catalyst properties.

3.4.The effect of biomass torrefaction on pyrolysis products

To evaluate the effect of biomass torrefaction process temperature on the pyrolysis products,both raw and five bagasse samples torrefied at 180,200,240,270,and 300 °C were non-catalytically pyrolyzed by the Pyroprobe-GC/MS system.The various chromatograms and the main pyrolysis products were shown in Fig.4(a)-(f).It can be noticed in Fig.4(a) that,the raw biomass thermal decomposition mainly produced oxygenated compounds such as carbon dioxide,aldehydes acids,ketone,alcohol,phenols,furans carbonyl,and carboxyl groups.No aromatic compounds were obtained.This explains that no conversion into hydrocarbons in the absence of a catalyst.Similar results were reported elsewhere in the literature [48].The pyrolysis of bagasse torrefied at the temperature of 180 °C,generated compounds similar to those obtained from untreated biomass as shown in Fig.4(b);however,aromatics were obtained during the pyrolysis of samples torrefied at 200,240,270,and 300°C as shown in Fig.4(c)-(f)respectively.It is worth mentioning the proportion of hydrocarbons slightly increased when the samples torrefied at between 200 and 270 °C were pyrolyzed;compared to the other samples.These results indicate that the torrefaction process can be used as an approach to editing the composition of biomass to enhance the production of hydrocarbons even in the absence of a catalyst during the pyrolysis process.Xinet al.[50]found that the pyrolysis of torrefied biomass would enhance the proportion of aromatics compared to raw biomass.It is worth mentioning that the non-CFP of the plastics generate mostly alkanes and alkenes and traces of aromatics as shown in Fig.S1-S4 (in Supplementary Material).

3.5.The effects of torrefaction and catalyst type on hydrocarbon production

Fig.2.Pyrolysis of raw,torrefied-bagasse,and different plastics at 10 °C·min-1,(a and c) TG,and (b and d) DTG curves.

Fig.3.N2 adsorption-desorption isotherms for USY and ZSM-5 zeolite catalysts.

The second set of experiments was performed to evaluate the combined effects of the biomass torrefaction temperatures and the performance of USY and ZSM-5 zeolite catalysts.The targeted aromatic hydrocarbons include benzene,toluene,ethylbenzene,indene,and naphthalene (BTEXIN).The BTEXIN content and their selectivities are shown in Fig.5.Five different samples of biomass,torrefied at 180,200,240,270,and 300 °C were investigated.It is worth stating here that,a liquid product cannot be collected during Py-GC/MS process,therefore,the total and relative peak areas of the compounds can be used to evaluate the yield and selectivities of the targeted aromatic hydrocarbons,respectively.In the previous section,the non-catalytic fast pyrolysis of raw bagasse (RBG)generated no aromatics,while the TBG generated traces of aromatics compounds.However,when the USY catalyst was employed an insignificant improvement in the aromatics hydrocarbons content was observed between the CFP of both RBG and the TBG samples.A slight increase in the compounds such as benzene,toluene,and xylenes by 1.2,1.13,and 1.4 times for TBG200°C;1.3,1.24,and 1.2 folds for TBG240°Csample,respectively,compared to CFP of RBG has been witnessed.Comparable improvements were also achieved for TBG300°Ccatalytic fast pyrolysis.In terms of total BTEXIN hydrocarbons yield,the improvement was regarded as insignificant.The USY catalyst seems not to favor the production of ethylbenzene and indene during CFP of TBG samples.The low content aromatic hydrocarbons explain that the USY catalyst was not effective in the conversion of pyrolytic to hydrocarbons due to limited confinement of compounds in the cavities,which may lead to accumulation in the interior cavities of the catalyst.Consequently,may lead to coke formation and subsequent deactivation of the catalyst.It can be reasonably said that the content of aromatic hydrocarbon generated was mainly by the exterior sites of the USY catalyst.The selectivities towards BTEXIN over the USY catalyst are presented in Fig.5(b).The selectivities towards toluene were highest in all samples with a maximum of 40.9%,42.59%,and 47.2% were obtained during CFP of TBG200°Cand TBG270°CTBG300°Csamples over USY catalyst,respectively.Moderate selectivities towards benzene and xylenes were attained for all samples.USY catalyst showed no selectivities towards ethylbenzene and indene during the CFP of TBG.The naphthalene selectivities gradually decrease with an increase in torrefaction temperature.

Fig.4.Compounds produced from the non-catalytic fast pyrolysis of raw and torrefied-bagasse.

Fig.4 (continued)

Fig.4 (continued)

As anticipated the content of total aromatic hydrocarbons generated increased drastically and then decreased correspondingly when the ZSM-5 catalyst was employed as shown in Fig.5(c).The most dominant aromatics species were toluene and xylenes,with a maximum of 2.0 and 2.1 folds increase for TBG200°C;and 2.0 and 1.96 folds higher for TBG240°C,respectively,compared against the CFP of untreated bagasse(RBG).This suggests that optimum treatment condition that favored aromatic production is between 200 °C and 240 °C.Moreover,there was an aromatics yield reduction during CFP of TBG300°C,the underlying reasons could be that at higher torrefaction temperatures much of the biomass component particularly hemicellulose had been decomposed and removed,hence reducing the quantities of vapors in a torrefied sample.Secondly,hydrogen was also removed along with the oxygen,depending on the severity of torrefaction temperatures as shown in Table 1.Moderate quantities of benzene,naphthalene,were also generated for both TBG200°Cand TBG240°CCFP over ZSM-5 catalyst.Traces or no production of ethylbenzene and indene were observed during CFP of all torrefied-bagasse (TBG)over USY and ZSM-5 catalysts.

Fig.5(d)shows the changes in selectivities towards BTEXIN during CFP of RBG and TBG over the ZSM-5 catalyst.It can be seen that significant improvement in toluene and xylenes selectivities were achieved for all samples,with the highest of 37.7% and 35.28%for TBG270°C;and 39.43%and 34.33%for TBG300°C.Meanwhile benzene and naphthalene showed moderate selectivities.In contrast,ethylbenzene and indene showed a similar trend as in the case of the USY catalyst.These results revealed that the ZSM-5 catalyst is more suitable for the conversion of pyrolytic vapor to aromatics hydrocarbons than the USY counterpart and the optimum biomass torrefaction temperature for the production of aromatic hydrocarbon is between 200-240 °C.

3.6.The effect of co-pyrolysis of torrefied-bagasse with plastics on aromatics yield

Fig.6(a)-(h) show the yield of BTEXIN aromatics and their selectivities produced from the catalytic co-pyrolysis of torrefiedbagasse with different plastics(include HDPE,PS,EVA,and PP)over ZSM-5 catalyst at the mixing ratios of 90:10,70:30,50:50 and 30:70.Based on the results obtained above bagasse torrefied at the temperature of 200 °C produced higher yields of aromatic hydrocarbons,hence it was chosen for further investigations.Biomass is hydrogen-deficient fuel,and hence it has a low hydrogen-to-carbon (H/C) ratio compared to fossils counterpart.Although the torrefaction process reduced the inherent oxygen in the biomass material,it also removed along with hydrogen during the process as shown in Table 1.Thus,further,deteriorate the already low H/C ratio of the biomass.Therefore,to remedy the hydrogen-deficiency in the torrefied-biomass is by co-processing with plastics to donate the hydrogen required for the formation of aromatic hydrocarbons during the co-pyrolysis process.

Fig.6(a) depicts the yield of BTEXIN aromatics during catalytic co-pyrolysis of TBG200°C/HDPE.The addition of plastic proportion boosted the yield and selectivities of light aromatics hydrocarbons,while that of larger aromatics decreased.It can be seen that the pyrolysis of TBG200°Calone generated low yields of BTEXIN aromatics.At 10% percent HDPE addition,the yield of total BTEXIN increased by 1.61 fold more than obtained from CFP of TBG200°Calone.A similar BTEXIN yield of 1.61 times was generated at 30%HDPE and then increased to a maximum of 3.23 fold at 70% HDPE addition.The most dominant aromatic species were benzene,toluene,and xylenes in all mixtures,with the highest enhancing of factors of 5.0,2.8,and 3.2 folds with selectivities of 18.14%,30.43% and 36.7% for 70% TBG200°C/HDPE,respectively.

Fig.5.The yields and selectivities of aromatic hydrocarbons during CFP of RBG and TBG over USY and ZSM-5 zeolite catalysts.

The selectivities to BTEXIN hydrocarbons during catalytic copyrolysis of TBG200°C/HDPE mixtures were shown in Fig.6(b).Notably,the toluene,and xylenes selectivities were the highest among the rest of the compounds during the pyrolysis of all mixtures.Toluene average selectivity was 36.36% and tends to remain constant at 10%,30%,and 50% HDPE and then decreased to 30.43% at 70% HDPE addition.Similarly,the xylenes selectivity tends to follow a similar trend at 10% and 30% HDPE addition and then decreased from 37.74%at 30%to 32.68%at 50%HDPE and then rose to 36.7% at 70% HDPE.The selectivities towards benzene,toluene,and xylenes tend to increase with an increased in HDPE proportion in the mixture,while that of naphthalene tends to show the opposite trend.The increased in BTEXIN content was a result of an increase in the H/C ratio of the mixture due to plastic addition.In other words,the hydrogen atoms required for hydrocarbons formation have been supplied by the HDPE,leading to increased yield of light aromatic compounds.It is worth stating that reduction in naphthalene content is positive in the sense that it may lead to less coke formation,and hence long catalyst life.

Fig.6(c) exhibits the yields of aromatic species produced from catalytic co-pyrolysis of TBG200°C/PS mixtures.It can be seen that PS addition enhanced the aromatic compounds significantly.For instance,benzene content increased by 8.92 times and 15.71 times when the PS percentages in the mixture were increased from 30%to 70%,respectively.Moreover,the toluene content doubles by enhancing factors of 2.0,2.12,and 2.49 folds corresponding to 30%,50%,and 70% PS,respectively.The ethylbenzene content rose by 14.6 fold at 30% and then slightly decreased at 50% PS and increased to 14.86 fold at 70% PS.The total xylenes tend to decrease with an increasing proportion of PS.It decreased from 0.93 fold at 30% PS to 0.68 times at 70% PS.While,indene significantly increases by 26.0,28.47,and 34.45 factors at the PS proportion of 30%,50%,and 70%,respectively.The larger aromatics such as naphthalene increase by an average factor of 8.63 times at 30,50,and 70% PS.The highest overall increase in BTEXIN content was 5.2 times attained at 70% PS addition.

Fig.6(d) shows how the selectivities for aromatic species changes with the addition of PS plastic.It can be seen that benzene selectivity was most dominant and it linearly increased from 16.0%at 10%PS through 25.22%,25.65%at 30%,50%reaching a maximum of 34.8%70%PS,respectively.This suggested that the selectivity of benzene is a strong function of the H/C ratio.Whereas,the selectivities for toluene continuously decreased from 37.0% at 10% to 17.27% at 30% and remained constant at an average of 16.61%between 50% and 70% PS,respectively.The total xylenes content followed a similar trend.Its selectivity sharply decreased from 36.98%at 10%PS addition to 4.8%at 70%PS addition.The selectivities towards ethylbenzene were low and dropped slightly from 2.84% at TBG200°Cto 1.92% at 10% PS,and swiftly rose to 10.52%at 30% PS,12.82% at 50% PS and then finally decreased to 8.39% at 70% PS.The larger aromatics species selectivities such as indene and naphthalene first increased to a maximum at intermediate concentrations of PS and remained somewhat constant thereafter.For indene,its selectivity was slightly affected by the addition of 10% PS,and then swiftly rose to 17.87% at 30% PS,16.42% at 50%PS and 18.71% at 70% PS addition.Similarly,naphthalene selectivity showed the same tendency,it rapidly increased to 20.73%,21.27% at PS proportions of 30% and 70%,respectively.

Fig.6.The aromatic hydrocarbon content and selectivities from catalytic co-pyrolysis of TBG/plastics mixture.

Fig.6(e)displays the BTEXIN aromatics yields obtained from the catalytic co-pyrolysis of TBG200°Cwith EVA plastic.Here the content of the aromatic compounds generally increased to a high value occurring at the intermediate amounts of EVA plastic and then decreased at higher EVA quantities.The EVA decomposed in two events with the first event occurring at lower temperatures between 311-400 °C and second even happening in the range of 412-498 °C,respectively.According to McGrattan [51],the first event is acetate pyrolysis of the copolymer leaving polyunsaturated linear hydrocarbons and evolving mainly acetic acid,followed by the breakdown of the hydrocarbons backbone to produce a large number of straight-chain hydrocarbons products.The benzene content almost doubles at 10% EVA addition with 1.98 times higher and further increased to a maximum by 9.44 enhancing factor at 30% EVA addition,and then decreased to a 4.1 factor at 70% EVA incorporation into the mixture.The toluene and xylenes yields are comparable.The toluene content improved from 1.67 fold at 10% EVA addition to 2.62 fold,while going through a maximum of 7.9 fold at 30%EVA.While the total xylenes increased by 1.69 times at 10%EVA addition to a maximum of 7.82 times at 30%EVA addition and then finally decreased to 3.12 times at 70%EVA.The quantities of other aromatic species such as ethylbenzene,indene,and naphthalene were relatively low.The highest total boost in the BTEXIN aromatic yield of 7.7 times was achieved at 30% EVA addition.

Fig.6(f)illustrates how the selectivity for the BTEXIN aromatics varies with the addition of EVA to TBG200°Cco-pyrolysis.Toluene selectivity (36.16%) tends to constant up to 30% EVA addition,and then gradually decreased to 30.64%at 70%EVA addition.Moreover,xylenes selectivity (37.52%) tends to remain constant for all mixtures.The benzene selectivity slightly increased from 13.75%to 15.43% at 10% and 70% EVA,respectively.Similarly,ethylbenzene selectively improved from 4.61%to 8.72%as the EVA proportion was increased from 10% to 70%.For indene,the selectivities were low throughout.The naphthalene selectivity somewhat decreased from 9.0% to 5.0% with EVA increased from 10% to 70%.

Fig.6(g) demonstrates the changes of BTEXIN aromatic species content with PP plastic addition during co-pyrolysis with TBG200°C.The generated aromatic species were predominantly toluene and xylenes in all scenarios.The toluene content had the enhancing factor of 1.69 fold at 10% PP addition and further improved by 3.13 times at 70% PP addition.While xylenes content was comparable with enhancing factor of 1.64 times at 10%PP and boosted to 3.73 times at 70%PP addition.Benzene on the other hand increased by 1.83 times higher at 10% PP to 4.56 times at 70% PP addition.Moreover,the ethylbenzene yield was rapidly boosted by 2.32 times higher at 10% PP to 11.57 times more at 70% PP addition.Meanwhile,indene and naphthalene species showed an insignificant increase.The highest total BTEXIN aromatics yield of 3.56 times more was reached at 70% PP addition.

Fig.6(h)displays the selectivities toward BTEXIN aromatic species during catalytic co-pyrolysis of TBG200°C/PP mixtures.The toluene selectivity increased with an increase in the PP content and then decreased thereafter at higher PP percentages.While the total xylenes selectivity showed the opposite trend as PP amounts increased.The selectivity towards toluene was 36.41%at 10% PP and rose to a maximum of 57.84% at 50% PP and finally reduced to 31.01%at 70%PP.Furthermore,the total xylenes selectivity decreased from 36.89% at 10% PP to the lowest value of 12.69% at 50% PP and then rose to 38.74% at 70% PP.Meanwhile,the selectivities toward the other aromatic species insignificantly vary with an increase in the PP content in the mixture.The highest aromatic content of 3.56 times more was reached at 70% PP amounts in the mixture.

3.7.Effect of single and dual catalysts layout on aromatic hydrocarbon production

This set of experiments was conducted to examine the performance of single and dual catalysts layouts,in cracking of TBG200°Cpyrolysis vapors.The yields of BTEXIN aromatic species and their selectivities over single and dual-catalyst layout were shown in Fig.7(a)and(b),respectively.0.5 mg of USY and ZSM-5 zeolite catalysts were used.The USY catalyst possesses a large pore volume compared to the ZSM-5 counterpart.Therefore,placing the two catalysts adjacent to each other may form a new pore structure.Thus,ZSM-5/USY design would form divergent-like pores,while the USY/ZSM-5 layout will form convergent-like pores.The amount of feedstock used was 0.2 mg giving a catalyst-to-feed ratio of 0.1:1 for dual layout and 0.1:0.5 for the single catalyst.That is to say,half of the vapor is expected to exit from one end of the quartz tube,and the other half gets out from the opposite end.

In both dual designs,the pyrolysis vapors were cracked by two successive catalysts to form aromatics hydrocarbons.In the ZSM-5/USY set-up,the yield of BTEXIN was very low.The total BTEXIN yield improves by 2.01 times more than that of the single USY catalyst.The benzene,toluene,and xylenes yields had doubled,while ethylbenzene and indene appeared,which is believed to have generated at ZSM-5 catalyst.Moreover,the yield of larger aromatics such as naphthalene tends to be constant.It was observed that the yield of BTEXIN hydrocarbons in the ZSM-5/USY layout was less than that obtained from a single ZSM-5 catalyst.This could be due to several reasons.First,the divergent-cavity would permit diffusion of mostly smaller oxygenates through its internal grooves to form hydrocarbons,meanwhile,conversion of larger oxygenates to BTEXIN hydrocarbons would occur at external acids sites.Secondly,the low BTEXIN yield could be due to over cracking of the pyrolysis vapor by the two catalysts,thus,the aromatics generated by the first catalyst were further cracked into smaller compounds by the second catalyst leading to a reduction in the BTEXIN yield.Besides,coking is likely to occur at the first catalyst due to the buildup of the larger compounds at the pore-entry leading blockage of catalyst pores,and hence,no conversion of pyrolysis to hydrocarbons thereafter.

In the USY/ZSM-5 catalyst scenario,the total BTEXIN hydrocarbons yield rose by 12.69 fold higher than that of single USY catalyst and 2.52 times more than that of single ZSM-5 catalyst.Here the end product is believed to be mainly determined by the second catalyst (ZSM-5 catalyst).The advantage of USY/ZSM-5 over ZSM-5/USY layout is that both smaller and larger compounds can be permitted to enter the pores of the USY catalyst,where the first stage of conversion to hydrocarbons commenced at USY and then passed to ZSM-5 for further conversion into BTEXIN hydrocarbons.In this set-up,the likelihood of coking of both catalysts would be minimized,and hence both catalysts would be available for a longer time during the pyrolysis process.

Fig.7(b) shows the selectivities toward BTEXIN hydrocarbons over single (USY or ZSM-5) and dual catalyst layouts (ZSM-5/USY or USY/ZSM-5).It can be seen that both single and dual catalysts designs showed higher selectivities for benzene,toluene,and xylenes.Except the USY/ZSM-5 set-up which mainly favors toluene and xylenes with selectivities of 35.25% and 36.9%,respectively.

3.8.Effect of dual-catalyst co-pyrolysis on aromatic hydrocarbons production

Fig.7.The aromatic yield and selectivities obtained from dual catalytic pyrolysis.

Fig.8.Co-pyrolysis of torrefied-bagasse with different plastics over dual catalyst layout (USY/ZSM-5).

Fig.8(a)and(b)shows the yield and selectivities of BTEXIN during dual-catalytic co-pyrolysis of torrefied-bagasse with plastics such as HDPE,PS,EVA and PP at 50:50 percent.As expected,the aromatic yields have been boosted more than that in the case of single feedstock and dual-catalysts.The obvious reason is due to the presence of plastics which donated hydrogen atoms required by reactions for forming hydrocarbons.The TBG200°C/PS mixture generated the highest total aromatic yield of 4.33 times and then followed by TBG200°C/HDPE which produced 2.22 fold compared to the pyrolysis of TBG200°Calone as shown in Fig.8a.Moreover,TBG200°C/EVA and TBG200°C/PP mixtures showed a slight increase in the total aromatic yields.The most dominant aromatic species were benzene,toluene,and xylenes in all the scenarios.

Fig.8(b) shows the selectivities towards BTEXIN over dualcatalytic co-pyrolysis of torrefied-bagasse with different plastics.It is noticeable that the dual-catalytic co-pyrolysis of all samples showed high selectivities towards benzene,toluene,and xylenes,with the highest of 36.3% (TBG200°C/PS),41.4% (TBG200°C/EVA) and 43.5% (TBG200°C/PP),respectively.

4.Conclusions

In this study,four sets of experiments were conducted in an analytical Py-GC/MS system to investigate the effects of biomass torrefaction,co-pyrolysis of torrefied-bagasse with different plastics (HDPE,PS,EVA,and PP);and single and dual catalyst layouts on aromatic hydrocarbons production.The yield and selectivities of aromatics hydrocarbons were evaluated.The results showed that the non-CFP of torrefied biomass can produce some of the aromatic hydrocarbons.The CFP experiments showed that the zeolite catalysts such as USY and ZSM-5 can provide the desired shape selectivity and acidity to produce aromatic hydrocarbons from biomass.The ZSM-5 catalyst generated higher total aromatic hydrocarbons yield compared to the USY counterpart.The increase in the total aromatics yield was 1.96 fold and 1.88 times were achieved from TBG200°Cand TBG240°C,respectively,compared to non-CFP of TBG.Catalytic co-pyrolysis of TBG/Plastics can further increase the yield and selectivity of total aromatic hydrocarbons,suggesting that the aromatic hydrocarbons yield is a function of the H/C ratio of the mixture.The highest yield of total aromatic hydrocarbons was boosted by 7.7 times more during the catalytic co-pyrolysis of TBG/EVA at a 30:70 percent.While the dualcatalyst design (USY/ZSM-5) generated the total aromatic yield of 4.33 times more during the dual-catalytic co-pyrolysis of TBG200°C/PS at 50:50 ratio compared to pyrolysis of TBG200°Cover ZSM-5 catalyst.The yield of light aromatics hydrocarbons were more dominant in hydrocarbons in all scenarios.This study is beneficial for improving the yield of aromatic hydrocarbons by the combinations of biomass torrefaction,catalytic co-pyrolysis,and dual catalyst design.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was supported by the National Natural Science Foundation for Excellent Young Scholar of China (51822604) and the Nature Science Foundation of Jiangsu Province for Distinguished Young Scholar (BK20180014).

Supplementary Material

Supplementary data to this article can be found online at https://doi.org/10.1016/j.cjche.2020.09.074.