Bioresource Technology 136 (2013) 431–436

Contents lists available at SciVerse ScienceDi rect

Biore source Technology

journal homepage: www.elsevier .com/locate /bior tech

Catalytic pyrolysis of mandarin residue from the mandarin juice processing industry

0960-8524/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.biortech.2013.03.062

⇑ Corresponding author at: Graduate School of Energy and Environmental System Engineering, University of Seoul, Seoul 130-743, Republic of Korea. Tel.: +82 2 6490 2870; fax: +82 2 6490 2859.

E-mail address: catalica@uos.ac.kr (Y.-K. Park).

Jeong Wook Kim a, Sung Hoon Park b, Jinho Jung c, Jong-Ki Jeon d, Chang Hyun Ko e, Kwang-Eun Jeong f,Young-Kwon Park a,g,⇑a Graduate School of Energy and Environmental System Engineering, University of Seoul, Seoul 130-743, Republic of Korea b Department of Environmental Engineering, Sunchon National University, Suncheon 540-950, Republic of Korea c Division of Environmental Science & Ecological Engineering, Korea University, Seoul 136-713, Republic of Korea d Department of Chemical Engineering, Kongju National University, Cheonan 331-717, Republic of Korea e School of Applied Chemical Engineering, Chonnam National University, Gwangju 500-757, Republic of Korea f Green Chemistry Research Divison, Korea Research Institute of Chemical Technology, Daejeon 305-600, Republic of Korea g School of Environmental Engineering, University of Seoul, Seoul 130-743, Republic of Korea

h i g h l i g h t s

� Catalytic pyrolysis of mandarin residue was performed using zeolites.� HZSM-5 produced more aromatics from mandarin residue than from woody biomass.� Ga/HZSM-5 increased further the production of valuable aromatic s.

a r t i c l e i n f o

Article history:Received 20 October 2012 Received in revised form 7 March 2013 Accepted 9 March 2013 Available online 18 March 2013

Keywords:Mandarin residue Catalytic pyrolysis AromaticsBio-oilMetal

a b s t r a c t

In this study, the catalytic pyrolys is of mandarin residue from the mandarin juice processing industry was carried out using pyrolys is gas chromatog raphy/mass spectroscopy and employing microporous zeolite catalysts, HZSM-5 (SiO2/Al2O3 = 23 and 80) and HBeta (SiO2/Al2O3 = 25). The effect of acidity of the cat- alyst was investigated by comparing the activity of two HZSM-5 catalysts with different SiO 2/Al2O3 ratios.The effect of catalyst structure was explored by comp aring the results obtained using HZSM-5 (23) and HBeta. Most oxygenates produced from non-catalytic pyrolysis were removed by catalytic upgrading,whereas the yields of mono-aromatics, which are important feedstock materials for the chemical indus- try, increased considerably, improving the quality of the bio-oil produc ed. HZSM-5 (23), having the high- est acidity among the catalysts used in this study, showed superior catalytic activity to those of HZSM-5 (80) and HBeta. Pt/HZSM-5 (23) and Ga/HZSM-5 (23) resulted in an even higher yield of aromatics.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

With the growing demand for alternative energy sources that can replace fossil fuels, biomass, including organic wastes, is gain- ing considerable attention as a sustainable energy source (Subra-manian, 2009 ). Biomass is inherently carbon–neutral: its net greenhouse gas emission is zero.

Rapid growth of the world population and industrial develop- ment are causing an increase in the production of various organic wastes. Secondary air pollution problems, such as emissions ofodorous substances and methane, can arise when organic wastes

are landfilled without appropriate pretreatmen t. Recovering en- ergy from these organic wastes not only produces energy but also reduces waste disposal.

Depending on the compositi on of the waste, biochemical and thermochemi cal processes are usually used to recover energy from organic waste. The biochemical processes include anaerobic diges- tion and ethanol fermentation of sugar-based biomass, whereas thermochemi cal processes include pyrolysis, gasification, and com- bustion (McKendry , 2002 ). Each process has advantages and disad- vantages . Combustion is a simple and easy process to operate but itrequires facilities to remove the various air pollutants emitted,such as SOx, NOx, and dioxins. High operation al costs to maintain the high combusti on temperature s are another drawback of com- bustion. Gasification can be operated at a lower temperature and emits less air pollutants, but large amounts of tar produced during the gasification reaction cause difficulties in the operation, calling

432 J.W. Kim et al. / Bioresource Technology 136 (2013) 431–436

for an additional tar removal process. Pyrolysis, the aim of which isto produce bio-oil, can also be operated at a lower temperature than combustion. The high moisture content, high viscosity, and acidity of the product bio-oil, however, impede its direct applica- tion as a fuel. Additional catalytic upgrading is usually employed to improve the quality of the bio-oil produced from pyrolysis pro- cesses (Park et al., 2011, 2012b; Bu et al., 2012; Ko et al., 2012 ).

In South Korea, an average of 650,000 tons of mandarins are produced annually, 120,000 tons of which are processed to pro- duce mandarin juice. About half of the processed mandarins (60,000 tons) are left as waste residue. Conversion of this waste residue into biofuel would reduce the waste disposal costs as well as generate renewable energy. Pyrolysis of waste residue from grape, tomato, and orange juice production processes has been em- ployed for the production of bio-oil (Aguiar et al., 2008; Encinar et al., 2008; Miranda et al., 2009; Xu et al., 2009; Demiral and Ayan,2011). Pyrolysis of mandarin waste and mandarin peel was studied previously by the present authors (Kim et al., 2011a,b ). Catalytic upgrading of the bio-oil produced from mandarin residue using microporous zeolites, however , has not previousl y been reported.

In this study, the catalytic pyrolysis of mandarin residue was carried out, for the first time, employing microporous zeolite cata- lysts. Fast pyrolysis experime nts were performed using pyrolysis gas chromatograp hy/mass spectroscopy (Py-GC/MS) for rapid anal- ysis of the compositi on of the bio-oil produced. The effects of var- ious characteristics of the catalyst on the composition of the bio-oil were investigated.

2. Experimental

2.1. Mandarin residue

To determine the contents of C, H, N and S in the waste manda- rin, ultimate analysis was performed using an elemental analyzer (TruSpec, LECO Co., USA) and a sulfur analyzer (SC-432DR, LECO Co., USA). Proximate analysis was performed using a thermo-gra vi- metric analyzer (pyris1 TGA, Perkin Elmer). The ASTM D2016-74,ASTM E897-82, and ASTM D1102-84 standards were used to deter- mine the moisture, volatile matter, and ash contents. First, 4–8 mgof the waste mandarin sample were heated in the TGA from room temperature to 110 �C under an N2 atmosph ere and then this tem- perature was maintained for 1 h to determine the moisture con- tent. The volatile matter content was determined by raising the temperature to 900 �C. After changing to an oxygen atmosphere ,the temperat ure was maintained at 800 �C until the sample mass reached an asymptotic value. The total mass change observed atthis point was regarded as the fixed carbon content, whereas the remainder was regarded as the ash content.

The mandarin residue used in this study contained 70.5 wt.%volatiles and 14.3 wt.% fixed carbon, which is similar to typical woody biomass. Also, the contents of moisture and ash were 1.2 and 14.0 wt.%, respectively . The contents of C, H, O and N were 36.54, 7.33, 41.64, and 0.49 wt.%, respectively . The nitrogen con- tent was low and sulfur was not detected, indicating good potential to produce a low-emission fuel. The heating value of the mandarin residue was also relatively high (3284.5 kcal/kg).

2.2. Preparation and characterization of catalysts

Two micropor ous zeolite catalysts, HZSM-5 (SiO2/Al2O3 = 23,80) (CBV 2314, 8014) and HBeta (SiO2/Al2O3 = 25) (CP814E⁄) were purchased from Zeolyst Internationa l. Hereafter the number inparentheses means the SiO 2/Al2O3 ratio. Using the incipient wet- ness method, HZSM-5 (23) was impregnate d with 1 wt.% Ga and 0.5 wt.% Pt using gallium nitrate and Pt(NH3)4(NO3)2, respectively.

After the impregnate d catalysts were dried in an oven at 110 �C,the temperature was increased to 550 �C at a rate of 3 �C/min un- der oxygen-free conditions. Calcination was conducted for 4 h un- der an N2 atmosphere (Araujo and Schmal, 2000 ).

The surface area was determined using the Brunauer–Emmett–Teller (BET) equation. The acidity was measured using a BEL-CAT TPD/TPR analyzer (BEL JAPAN INC.) based on the NH3-tempera-ture-progr ammed desorptio n (TPD) method. Prior to each mea- suremen t, samples were treated in a He stream at 450 �C. After cooling to 100 �C, NH3 adsorptio n was performed. After purging samples in the He stream for 30 min to complete ly remove the physically adsorbed NH3, the catalysts were heated to 100–550 �C at a rate of 10 �C/min. The desorbed NH3 was detected using a thermal conductivity detector.

2.3. Pyrolysis experimen t

The vapor product produced from the catalytic pyrolysis pro- cess in a single-sh ot pyrolyzer (Frontier-Lab Co., Py-2020i D) was introduce d into a GC/MS. The temperature of the pyrolyze r was set to 500 �C. After putting 1.5 mg of the mandarin residue onto the bottom of a sample cup and covering it with quartz wool,1.5 mg of catalyst was laid over the quartz wool layer. In this sys- tem, the pyrolyzed vapor product is upgraded when it passes through the catalyst layer. The metal capillary column used for the analysis was an Ultra ALLOY-5 (MS/HT) (5% diphenyl and 95%dimethyl polysiloxane, length 30 m, i.d. 0.25 mm, film thickness 0.5 lm, Frontier Laboratories Ltd. Japan). The product composition was calculated as the peak area%. More detailed information on the experime ntal procedure can be found in the literature (Heo et al.,2011; Park et al., 2012b; Jeon et al., 2013 ). The detailed procedure for quantitative analysis of the yields of solid (char + coke), gas, and liquid can be seen in the literature (Ma et al., 2012 ). The yield ofaromatic s was quantified using the calibration curves of standards of known concentratio n.

3. Results and discussion

3.1. Characteri stics of the catalysts

Table 1 shows the BET surface areas of the catalysts used in this study. The specific surface area of HZSM-5 was 382 m2/g. The addi- tion of Pt or Ga decreased the specific surface area of HZSM-5,which might be attributed to partial blockage of the pores of the catalyst. Of the two catalyst types used in this study, HBeta had alarger pore size and specific surface area.

The NH3-TPD patterns of each catalyst are shown in Fig. 1 andTable 1 to compare the acid properties of the catalysts. The number of acid sites of HZSM-5 increased as the SiO 2/Al2O3 ratio decrease d(1.47 mmol NH3/gcat for HZSM-5 (23) and 0.46 mmol NH3/gcat forHZSM-5 (80)). HZSM-5 showed strong Brønsted acid sites atapproximat ely 410 �C. The number of acid sites of HBeta (1.03 mmol NH3/gcat) was smaller than that of HZSM-5 (23), but larger than that of HZSM-5 (80). Fig. 1 also shows that the acid strength of HBeta, with a peak at approximately 310 �C, was weak- er than that of HZSM-5. Pt/HZSM-5 (23) (1.39 mmol NH3/gcat) and Ga/HZSM- 5 (23) (1.11 mmol NH3/gcat) catalysts had fewer acid sites than HZSM-5 (23) because Pt and Ga replaced some of the acid sites (Park et al., 2010, 2012a ).

3.2. Catalytic pyrolysis

3.2.1. Effect of the SiO 2/Al2O3 ratio on the product distribution A detailed list of the pyrolysis products is provided in the Sup-

porting material (Table S1). The products were divided into several categories (Fig. 2). The composition of the bio-oil is shown to

Table 1Physical characteristics of catalysts.

SiO2/Al2O3 ratio SBET (m2/g) Number of acid sites (mmol NH3/gcat)

Lower temperature acid sites High temperature acid sties Total acid sites

HZSM-5 23 382 0.81 0.66 1.47 HZSM-5 80 372 0.22 0.24 0.46 Ga(1 wt.%)/HZSM-5 23 303 0.67 0.43 1.11 Pt(0.5 wt.%)/HZSM-5 23 317 0.74 0.65 1.39 HBeta 25 617 0.64 0.40 1.03

Temperature (oC)100 150 200 250 300 350 400 450 500

Inte

nsity

(a.u

.)

HBETA (25)HZSM-5 (23)HZSM-5 (80)Pt/HZSM-5 (23)Ga/HZSM-5 (23)

Fig. 1. NH3-TPD curves of various catalysts.

J.W. Kim et al. / Bioresource Technology 136 (2013) 431–436 433

change considerably as a result of catalytic upgrading. The content of oxygenates, such as acids, ketones, aldehydes, and esters, was reduced by catalytic upgrading, while the aromatic content was enhanced.

Oxygen contained in bio-oil lessens the heating value and causes corrosion and instability. Therefore, oxygen must be re- moved to produce a high-quality bio-oil that can replace conven- tional hydrocarbo n-based fuels.

The acid content, which was 19% before catalytic upgrading, de- creased to 13.1% and 6.0% when the bio-oil was upgraded over HZSM-5 (80) and HZSM-5 (23), respectivel y. Acids contained in abio-oil must be neutralized before use to prevent corrosion and instability resulting from them, and therefore, the decrease in acids by catalytic upgrading is encouraging .

AcidsOxygenates

AromaticsNaphthalene

Prod

uct d

istr

ibut

ion

(Pea

k ar

ea %

)

0

10

20

30

40

50

Fig. 2. Product distribution of bio-oil from

Oxygena tes in bio-oil, including ketones, esters, and ethers, are known to reduce stability and miscibility with hydrocarbo ns(Bridgwat er, 2012 ). Generally, the oxygenate content of a bio-oil is determined by the oxygen content of the biomass from which the bio-oil is derived. The oxygenate content of the bio-oil pro- duced from the non-catalytic pyrolysis of mandarin residue was 42%, whereas after catalytic upgrading over HZSM-5(23) and HZSM-5(80), it decreased to 6.1% and 11.5%, respectively, demon- strating that the catalysts used in this study were very effective in removing oxygenates . Fig. 3 shows changes in the contents ofvarious oxygenate species in detail. The contents of ketones, esters,and ethers were reduced dramatically by the catalytic upgrading.Esters, whose content was 16% before upgrading, were completely removed by the upgrading over HZSM-5 (23). Upgrading over HZSM-5 (80), whose acid strength and acid amount are weaker and lower than those of HZSM-5 (23), reduced the ester content to 2.1%. In particular, long-chain fatty acid esters (e.g., hexadeca- noic acid, ethyl ester, 15-octadece noic acid, and methyl ester),which accounted for a considerable fraction of the total esters,were removed completely by upgrading over HZSM-5 (23) (Sup-porting material Table 1). These long-chain fatty acid esters are reportedl y converted to aromatic s and hydrocarbo ns over the cat- alysts (Ooi et al., 2004, 2005; Heo et al., 2011 ). Removal of ketones was also more effective over HZSM-5 (23) than over HZSM-5(80).

The phenolic content, which was small before upgrading, be- came even smaller when the bio-oil was upgraded over HZSM-5 (23), whereas upgrading over HZSM-5 (80) did not significantly al- ter the phenolic content. The higher phenolic conversion perfor- mance of HZSM-5 (23) is attributed to its larger number of acid sites.

Nitrogen compounds were also removed complete ly by upgrad- ing over HZSM-5 (23). HZSM-5 (80) showed little ability to remove nitrogen compounds. This indicates that the use of an appropriate catalyst can reduce the nitrogen content of a bio-oil, and thus, the

PhenolicsN-Compounds

Hydrocarbons

Non-catalystHBETA (25)HZSM-5 (23)HZSM-5 (80)Pt/HZSM-5(23)Ga/HZSM-5(23)

non-catalytic and catalytic pyrolyses.

Aldehydes Ketones Esters Ethers Alcohols Furans

Prod

uct d

istr

ibut

ion

(Pea

k ar

ea %

)

0

5

10

15

20

25

30Non-catalystHBETA (25)HZSM-5 (23)HZSM-5 (80)Pt/HZSM-5(23)Ga/HZSM-5(23)

Fig. 3. Product distribution of oxygenates.

AcidsOxygenates

AromaticsNaphthalene

PhenolicsN-Compounds

Hydrocarbons

Prod

uct d

istr

ibut

ion

(Pea

k ar

ea%

)

0

10

20

30

40

50Larch - Non catalystLarch - HZSM-5(23)Mandarin Residue - HZSM-5(23)

Fig. 4. Comparison of bio-oil derived from Larch with that of mandarin residue.

434 J.W. Kim et al. / Bioresource Technology 136 (2013) 431–436

correspondi ng NOx emissions upon combustion of the bio-oil. Choiet al. (2013) also reported that the catalytic upgrading of the bio- oil produced from the pyrolysis of particle board reduced the nitro- gen content.

The hydrocarbo n content, which was 16.1% before the catalytic upgrading, was reduced to 10.1% and 12.5% when the bio-oil was upgraded over HZSM-5 (23) and HZSM-5 (80), respectively . This implies that hydrocarbo ns were converte d to other compounds ,such as aromatics, via cracking, isomeriz ation, and aromatiz ation over the acid catalysts. It was reported in previous studies (Kitag-awa et al., 1986; Guisnet et al., 1992; Gayubo et al., 2004 ) that var- ious hydrocarbo ns including propane, butane, hexane, and heptane were converted to aromatics in the presence of HZSM-5.

The most remarkabl e feature of the catalytic pyrolysis of man- darin residue found in this study is the large yield of monoaromat- ics. The monoaromati c content, which was as low as 3.4% before catalytic upgrading, increased dramatically to 36.0% and 41.0% byupgrading over HZSM-5 (80) and HZSM-5 (23), respectively . The capability of HZSM-5 to produce aromatics is well known. HZSM- 5 has a shape selectivity to the synthesis of aromatics, which is en- hanced by the presence of strong acid sites (Carlson et al., 2009;Park et al., 2010, 2012a ). High aromatic yields have been reported from catalytic pyrolysis over HZSM-5 or meso-MFI of various bio- mass materials (miscanthus, wood, rice husk, micro algae, macro algae, etc.) (Park et al., 2010, 2012a; Lee et al., 2011; Jeon et al.,2012; Pan et al., 2010; Heo et al., 2011 ). Williams and Horne (1995) proposed the following mechanism for aromatic synthesis from the pyrolysis of wood: (1) the formatio n of low-molecular- weight hydrocarbons on the catalyst, which subsequent ly undergo aromatizati on reactions to produce aromatic hydrocarbo ns and PAHs; and (2) the deoxygenati on reaction of oxygenated com- pounds in the non-phenol ic fraction of the bio-oil, which can di- rectly yield aromatic compounds .

Especially, the aromatic content obtained from the catalytic pyrolysis of mandarin residue in this study is higher than those re- ported for the catalytic pyrolysis of woody biomass materials.Fig. 4 compares the pyrolysis results of a woody biomass (larch)and mandarin residue. The aromatic content obtained from the catalytic pyrolysis of larch over HZSM-5 (23) was 20%, which isabout half that obtained from the catalytic pyrolysis of the manda- rin residue in this study. This indicates that mandarin residue is auseful biomass material for the production of aromatics. The com- position of the bio-oil was also very different. For example, the phenol content was very large when larch was pyrolyzed, whereas it was small in the case of mandarin residue. This is attributed tothe different constituent components of larch and mandarin resi-

due. The content of lignin, from which phenolics are derived upon pyrolysis, of larch is 20.1% (Park et al., 2008 ), whereas that of the mandarin residue is 9% (private data), leading to a smaller phenolic content. Phenolics are known to be compara tively difficult to con- vert to aromatics. Mullen and Boateng (2010) reported that cata- lytic fast pyrolysis of lignins produced phenols, which could not be easily further deoxygenated to aromatics. In addition, Jaeet al. (2010) demonstrat ed that the yield of aromatics of catalytic fast pyrolysis of maple wood decreased with increasing content of the lignin in the maple wood. The lignin in the wood primarily produced coke on the catalyst surface. Therefore, the low lignin content of the mandarin residue and resulting low phenolic con- tent of mandarin residue-deri ved bio-oil appears to be a reason for the high aromatic yield. Ketones and esters are reported to beeffectively converte d to aromatic s over HZSM-5 via cracking, isom- erization , and aromatization (Gayubo et al., 2004 ). The high con- tent of these species in the bio-oil obtained from the non- catalytic pyrolysis of mandarin residue is believed to be another reason for the high aromatic yield.

Among the aromatic s, the contents of benzene, toluene, ethyl- benzene, and xylene (BTEX), which are important feedstock mate- rials in the petroleum chemical industry, were shown to beparticular ly high (Fig. 5). HZSM-5 (23) led to a BTEX content of30%. In particular, the content of p-xylene, which has the highest economic value, was the largest (14%). Therefore, mandarin residue can be used as an important resource for the production of BTEX.

The contents of toxic compounds , such as naphthalene and other polycyclic aromatic hydrocarbo ns (PAHs), were also signifi-cantly increased by catalytic upgrading (data not shown). Naph- thalene, which accounted for the largest fraction of PAHs, is used as a raw material for industrial production of phthalic anhydrid e.It is also used for the production of surfactants and plasticizers,such as naphthalene sulfonate and naphthalene sulfonate–formal- dehyde condensate. Therefore, separation of naphthalene from the bio-oil is recomme nded before use (Carlson et al., 2009 ). More naphthalene was produced over HZSM-5 (23) because the larger number of acid sites promotes the reaction between aromatics toform PAHs.

3.2.2. Effect of the structure of the zeolite on the product distribution The removal efficiencies of acids and oxygenates of HBeta (25)

were lower than those of HZSM-5 (23) but higher than those ofHZSM-5 (80) (Fig. 2). As shown in the ammonia TPD result (Fig. 1and Table 1), the number of acid sites of HBeta (25) is smaller than that of HZSM-5 (23) but larger than that of HZSM-5 (80). Therefore,

Benzene Toluene Ethylbenzene p-Xylene total BTEX

Prod

uct d

istr

ibut

ion

(Pea

k ar

ea %

)

0

10

20

30

40Non-catalystHBETA (25)HZSM-5 (23)HZSM-5 (80)Pt/HZSM-5(23)Ga/HZSM-5(23)

Fig. 5. Product distribution of benzene, toluene, ethylbenzene and xylenes.

J.W. Kim et al. / Bioresource Technology 136 (2013) 431–436 435

it appears that the removal efficiencies of acids and oxygenates are mainly determined by the number of acid sites that promote dehy- dration, decarbony lation, and decarboxylati on.

The aromatic content obtained with HBeta (25) was lower than those of the HZSM-5 catalysts. The production of aromatics isknown to be promoted by strong Brønsted acid sites. Therefore,the stronger Brønsted acid sites of HZSM-5 are believed to be a rea- son for the higher aromatic content obtained with HZSM-5. The structure of HZSM-5 with a high shape selectivity toward the syn- thesis of aromatics (Heo et al., 2011; Pan et al., 2010; Park et al.,2012a; Williams and Horne, 1995 ) is another reason for the higher aromatic content obtained with HZSM-5 than that with HBeta.

However, the highest conversio n of high-molecu lar-mass hydrocarbo ns to low-molecular- mass hydrocarbo ns occurred when HBeta was used. As shown in Fig. 6, catalytic upgrading over HBeta resulted in the complete removal of high-molecula r-mass (PC16) hydrocarbo ns and the largest content of low- molecular-m ass (C1–C4) hydrocarbo ns. This is attributed to the large pore size of HBeta. The pore size of the HBeta catalyst (0.65 � 0.56 nm, 0.75 � 0.57 nm) is larger than that of ZSM-5 (0.51 � 0.55 nm, 0.53 � 0.56 nm), and therefore, using HBeta isadvantageous for the decompositi on of pyrolytic bio-oil intermedi- ates with large molecular size.

3.2.3. Effect of Ga and Pt addition on the product distribution Not only changing the SiO 2/Al2O3 ratio but also introducing

metals into the zeolite is a way to change the acidic properties of

C1~C4 C5~C9 C10~C15 over C16

Prod

uct d

istr

ibut

ion

(Pea

k ar

ea %

)

0

3

6

9

12

15Non-catalystHBETA (25)HZSM-5 (23)HZSM-5 (80)Pt/HZSM-5(23)Ga/HZSM-5(23)

Fig. 6. Distribution of hydrocarbons with carbon number.

zeolite catalysts. In particular, the addition of a metal might change the propertie s of both Brønsted and Lewis acid sites (Heo et al.,2011; Park et al., 2010, 2012a ). In this study, Ga and Pt were impregnate d into HZSM-5 (23) to investigate the effect of metal addition on its catalytic activity for the pyrolysis of mandarin res- idue. The aromatic content increased from 41.0% to 42.5% and 44.77% as a result of the impregnation with 1 wt.% Ga and 0.5 wt.% Pt, respectivel y (Fig. 2). The content of BTEX also increased from 30.0% to 32.1% and 33.4% (Fig. 5). It was reported that the incorporation of Ga or Pt cations into a MFI-type zeolite increased the initial dehydrogen ation rates of hydrocarbo ns, as well as the subsequent dehydrogen ation rates required to more rapidly trans- form alkenes to aromatics, leading to considerably enhanced aro- matic production compared to that obtained with the parent MFI-type zeolite (Park et al., 2010, 2012a ). Cheng et al. (2012) alsosuggested that Ga/ZSM-5 is most likely a bifunctiona l catalyst,where Ga promotes both the desired decarbonylatio n reactions and olefin aromatization, while the ZSM-5 portion of the catalyst catalyzes the remaining reactions for the production of aromatic s(e.g., oligomeri zation and cracking).

The addition of Pt and Ga to HZSM-5 created more monoaro- matics than pure HZSM-5. However, metal-incor porated HZSM-5 produced less PAHs, such as naphthalene , than pure HZSM-5. For example, the naphthalene content obtained with HZSM-5, Pt/ HZSM-5, and Ga/HZSM-5 were 19.8%, 15.5%, and 18.9%, respec- tively. The decreased PAH content, owing to the addition of Ptand Ga, can be attributed to the replacemen t of Brønsted acid sites by Ga and Pt (Fig. 1), the Brønsted acid sites being responsible for promoting the reaction between aromatic molecules to form PAH (Heo et al., 2011; Park et al., 2010, 2012a ).

3.2.4. Absolute yields of gas, oil and solid Besides the product distribution obtained by GC/MS, absolute

yields of the solid (char and coke), gas and liquid products as well as the absolute yield of aromatic s are very important for the eval- uation of catalytic pyrolysis of biomass. As shown in Supportingmaterial Fig. S1, the catalytic pyrolysis increased the yield of gas and decreased the yield of oil compared to non-catalyti c pyrolysis.Also, catalytic pyrolysis increased the yield of solid due to the coke formatio n. Especially, HBeta (25) led to the highest yield of solid.Because the pore size of HBeta is larger than that of HZMS-5, there are a lot of possibilit ies to form high molecular weight coke inside the pores. The yield of aromatics was highest for Ga/HZSM-5 (ca.8.8 wt.%).

Carbon balance analysis was also done for the gas, liquid and solid products (Supporting material Fig. S2). It can be seen that high amount of carbon was concentr ated in solid. Especially, HBeta (25) showed highest carbon content in solid due to the highest coke amount inside its large pores. Meanwhi le, in the cases of Pt/ HZSM-5 (23) and Ga/HZSM-5 (23), carbon content in liquid was higher than in the case of HZSM-5 (23). This implied that qualities of the bio-oils obtained from Pt/HZSM- 5 (23) and Ga/HZSM-5 (23)may be better than that from HZSM-5 (23).

4. Conclusion s

The catalytic upgrading of the bio-oil obtained from mandarin residue over HZSM-5 catalysts considerably reduced the contents of oxygenates that make bio-oil unstable. Moreover, the yield ofmonoaromati cs increased as a result of catalytic upgrading. In par- ticular, the aromatic content was about twice that obtained from pyrolysis of larch. The aromatic yield showed a high correlation with the acidity of the catalyst used. Catalytic upgrading over HBe- ta, whose acidity is weaker than that of HZSM-5, resulted in a low-

436 J.W. Kim et al. / Bioresource Technology 136 (2013) 431–436

er aromatic yield. The impregnation of Pt and Ga in the catalyst in- creased the yield of monoaromatics even further.

Acknowled gements

This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2012R1A1B3003394).

Appendix A. Supplementar y data

Supplement ary data associated with this article can be found, inthe online version, at http://dx.doi .org/10.1016/j.biortech.2013.03.062.

References

Aguiar, L., Márquez-Montesinos, F., Gonzalo, A., Sánchez, J.L., Arauzo, J., 2008.Influence of temperature and particle size on the fixed bed pyrolysis of orange peel residues. J. Anal. Appl. Pyrolysis 83, 124–130.

Araujo, L.R.R., Schmal, M., 2000. The calcination effects on Pt/HZSM-5 catalysts inthe aromatization of propane. Appl. Catal. A 203, 275–284.

Bridgwater, A.V., 2012. Review of fast pyrolysis of biomass and product upgrading.Biomass Bioenergy 38, 68–94.

Bu, Q., Lei, H., Zacher, A.H., Wang, L., Ren, S., Liang, J., Wei, Y., Liu, Y., Tang, J., Zhang,Q., Ruan, R., 2012. A review of catalytic hydrodeoxygenation of lignin-derived phenols from biomass pyrolysis. Bioresour. Technol. 124, 470–477.

Carlson, T.R., Tompsett, G.A., Conner, W.C., Huber, G.W., 2009. Aromatic production from catalytic fast pyrolysis of biomass-derived feedstocks. Top. Catal. 52, 241–252.

Cheng, Y.T., Jae, J., Shi, J., Fan, W., Huber, G.W., 2012. Production of renewable aromatic compounds by catalytic fast pyrolysis of lignocellulosic biomass with bifunctional Ga/ZSM-5 catalysts. Angew. Chem. 124, 1416–1419.

Choi, S.J., Park, S.H., Jeon, J.K., Lee, I.G., Ryu, C., Suh, D.J., Park, Y.K., 2013. Catalytic conversion of particle board over microporous catalysts. Renew. Energy 54,105–110.

Demiral, _I., Ayan, E.A., 2011. Pyrolysis of grape bagasse: effect of pyrolysis conditions on the product yields and characterization of the liquid product.Bioresour. Technol. 102, 3946–3951.

Encinar, J.M., González, J.F., Martínez, G., 2008. Energetic use of the tomato plant waste. Fuel Process. Technol. 89, 1193–1200.

Gayubo, A.G., Aguayo, A.T., Atutxa, A., Aquado, R., Olazar, M., Bilbao, J., 2004.Transformation of oxygenate components of biomass pyrolysis oil on a HZSM-5 zeolite. II. Aldehydes, ketones, and acids. Ind. Eng. Chem. Res. 43, 2619–2626.

Guisnet, M., Gnep, N.S., Alario, F., 1992. Aromatization of short chain alkanes onzeolite catalysts. Appl. Catal. A 89, 1–30.

Heo, H.S., Kim, S.G., Jeong, K.E., Jeon, J.K., Park, S.H., Kim, J.M., Kim, S.S., Park, Y.K.,2011. Catalytic upgrading of oil fractions separated from food waste leachate.Bioresour. Technol. 102, 3952–3957.

Jae, J.H., Tompsett, G.A., Lin, Y.C., Carlson, T.R., Shen, J.C., Zhang, T.Y., Yang, B.,Wyman, C.E., Conner, W.C., Huber, G.W., 2010. Depolymerization oflignocellulosic biomass to fuel precursors: maximizing carbon efficiency bycombining hydrolysis with pyrolysis. Energy Environ. Sci. 3, 358–365.

Jeon, M.-J., Kim, S.S., Jeon, J.K., Park, S.H., Kim, J.M., Sohn, J.M., Lee, S.H., Park, Y.K.,2012. Catalytic pyrolysis of waste rice husk over mesoporous materials.Nanoscale Res. Lett. 7, 1–5.

Jeon, M.J., Jeon, J.K., Suh, D.J., Park, S.H., Sa, Y.J., Joo, S.H., Park, Y.K., 2013. Catalytic pyrolysis of biomass components over mesoporous catalysts using Py-GC/MS.Catal. Today 204, 170–178.

Kim, J.W., Lee, S.H., Kim, S.-S., Park, S.H., Jeon, J.K., Park, Y.K., 2011a. The pyrolysis ofwaste mandarin residue using thermogravimetric analysis and a batch reactor.Korean J. Chem. Eng. 28, 1867–1872.

Kim, Y.M., Lee, H.W., Lee, S.H., Kim, S.-S., Park, S.H., Jeon, J.-K., Kim, S., Park, Y.K.,2011b. Pyrolysis properties and kinetics of mandarin peel. Korean J. Chem. Eng.28, 2012–2016.

Kitagawa, H., Sendoda, Y., Ono, Y., 1986. Transformation of propane into aromatic hydrocarbons over ZSM-5 zeolites. J. Catal. 101, 12–18.

Ko, C.H., Park, S.H., Jeon, J.K., Suh, D.J., Jeong, K.E., Park, Y.K., 2012. Upgrading ofbiofuel by the catalytic deoxygenation of biomass. Korean J. Chem. Eng. 29,1657–1665.

Lee, H.W., Jeon, J.K., Park, S.H., Jeong, K.E., Chae, H.J., Park, Y.K., 2011. Catalytic pyrolysis of Laminaria japonica over nanoporous catalysts using Py-GC/MS.Nanoscale Res. Lett. 6, 1–7.

Ma, Z., Troussard, E., van Bokhoven, J.A., 2012. Controlling the selectivity tochemicals from lignin via catalytic fast pyrolysis. Appl. Catal. A 423–424, 130–136.

McKendry, P., 2002. Energy production from biomass (part 2): conversion technologies. Bioresour. Technol. 83, 47–54.

Miranda, R., Bustos-Martinez, D., Blanco, C.S., Villarreal, M.H.G., Cantu, M.E., 2009.Pyrolysis of sweet orange (Citrus sinensis ) dry peel. J. Anal. Appl. Pyrolysis 86,245–251.

Mullen, C.A., Boateng, A.A., 2010. Catalytic pyrolysis-GC/MS of lignin from several sources. Fuel Process. Technol. 91, 1446–1458.

Ooi, Y.S., Zakaria, R., Mohamed, A.R., Bhatia, S., 2004. Catalytic conversion of palm oil-based fatty acid mixture to liquid fuel. Biomass Bioenergy 27, 477–484.

Ooi, Y.S., Zakaria, R., Mohamed, A.R., Bhatia, S., 2005. Catalytic conversion of fatty acids mixture to liquid fuel and chemicals over composite microporous/ mesoporous catalysts. Energy Fuels 19, 736–743.

Pan, P., Hu, C., Yang, W., Li, Y., Dong, L., Zhu, L., Tong, D., Qing, R., Fan, Y., 2010. The direct pyrolysis and catalytic pyrolysis of Nannochloropsis sp. residue for renewable bio-oils. Bioresour. Technol. 101, 4593–4599.

Park, H.J., Dong, J.I., Jeon, J.K., Park, Y.K., Yoo, K.S., Kim, S.S., Kim, J., Kim, S., 2008.Effects of the operating parameters on the production of bio-oil in the fast pyrolysis of Japanese larch. Chem. Eng. J. 143, 124–132.

Park, H.J., Heo, H.S., Jeon, J.K., Kim, J., Ryoo, R., Jeong, K.E., Park, Y.K., 2010. Highly valuable chemicals production from catalytic upgrading of radiate pine sawdust-derived pyrolytic vapors over mesoporous MFI zeolites. Appl. Catal.B 95, 365–373.

Park, H.J., Jeon, J.K., Suh, D.J., Suh, Y.W., Heo, H.S., Park, Y.K., 2011. Catalytic vapor cracking for improvement of bio-oil quality. Catal. Surv. Asia 15, 161–180.

Park, H.J., Park, K.H., Jeon, J.K., Kim, J., Ryoo, R., Jeong, K.E., Park, S.H., Park, Y.K.,2012a. Production of phenolics and aromatics by pyrolysis of miscanthus. Fuel 97, 379–384.

Park, Y.K., Choi, S.J., Jeon, J.K., Park, S.H., Ryoo, R., Kim, J., Jeong, K.E., 2012b. Catalytic conversion of waste particle board to bio-oil using nanoporous catalyst. J.Nanosci. Nanotechnol. 12, 5367–5372.

Subramanian, P., 2009. A review on gasification of biomass. Renew. Sustain. Energy Rev. 13, 179–186.

Williams, P.T., Horne, P.A., 1995. The influence of catalyst regeneration on the composition of zeolite-upgraded biomass pyrolysis oils. Fuel 74, 1839–1851.

Xu, R., Ferrante, L., Briens, C., Berruti, F., 2009. Flash pyrolysis of grape residues into biofuel in a bubbling fluid bed. J. Anal. Appl. Pyrolysis 86, 58–65.