Effect of Pd, Ru, Ni and ceramic supports on selective deoxygenation and hydrogenation of fast pyrolysis Jatropha residue vapors

Renewable Energy xxx (2013) 1e10

Contents lists available at ScienceDirect

Renewable Energy journal homepage: www.elsevier.com/locate/renene

Effect of Pd, Ru, Ni and ceramic supports on selective deoxygenation and hydrogenation of fast pyrolysis Jatropha residue vapors Prangtip Kaewpengkrow a, Duangduen Atong b, Viboon Sricharoenchaikul a, c, * a

Department of Environmental Engineering, Chulalongkorn University, Bangkok, Thailand National Metal and Materials Technology Center, Thailand Science Park, Pathumthani, Thailand c Center of Excellence on Hazardous Substance Management (HSM), Chulalongkorn University, Bangkok 10330, Thailand b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 22 February 2013 Accepted 10 July 2013 Available online xxx

Catalytic upgrading of the pyrolytic vapors after fast pyrolysis of Jatropha residue was performed using analytical pyrolysisegas chromatography/mass spectrometry (PyeGC/MS) at 673e873 K. The PyeGC/MS analyses for pyrolysis vapors show a range of aromatic hydrocarbons, hydrocarbon compounds, phenols, alcohols, aldehydes, ketones, acids and esters, furan and N-containing compounds. The result showed that high temperature had positive influence on the yields of pyrolytic products. Catalytic testing was performed by using Al2O3, ZrO2 based catalysts and their modified ones with impregnation of Pd, Ru, and Ni, respectively. The Al2O3 and ZrO2 were impregnated with CeO2 to promote metal dispersion prior to deposition of Pd, Ru, or Ni. From the experiment, these catalysts showed some potential to convert the highly oxygenated compounds to aromatic and hydrocarbons. The hydrocarbon yields increased with increasing catalyst to Jatropha ratio in all catalysts. The hydrocarbon selectivity was NieCe/ Al2O3 > Al2O3 > Ce/Al2O3 > PdeCe/Al2O3 > RueCe/Al2O3 in Jatropha to catalyst ratio of 1:5. PdeCe/Al2O3 was the most effective in terms of increased aromatic and hydrocarbon compounds, decreased oxygenated and N-compounds. Though Al2O3 increased favorable aromatic and hydrocarbon compounds with completely elimination of acid but it also promoted adverse N-containing compounds. Activities of RueCe/Al2O3 were similar to Ce/Al2O3 except for the selectivity toward acid compounds. Comparing to ZrO2 catalysts, all Al2O3 based catalysts displayed relatively higher activity toward deoxygenation reactions which resulted in low quantity of carboxylic acids and other oxygenated compounds while enhanced yields of aliphatic and aromatic hydrocarbons. Al2O3 had surface area of 103.38 m2/g while surface area of ZrO2 was 12 m2/g, indicating that catalyst with high surface area had better catalytic activity. However, Al2O3 catalysts seemed to promote N-compounds suggesting that further denitrogenation is required while pyrolysis with ZrO2 had disadvantage on high yield of acid which could cause the corrosion problem. Nevertheless, overall performances of these two support catalysts are acceptable and can be considered as good candidates for bio-oil upgrading catalysts. Ó 2013 Elsevier Ltd. All rights reserved.

Keywords: Bio-oil Fast pyrolysis Jatropha residue PyeGC/MS

1. Introduction Biomass residues can be converted to more valuable energy forms using thermal, biological, and mechanical or physical processes. Thermal conversions such as gasification and pyrolysis often give multiple and complex products in very short reaction times. Pyrolysis is thermal decomposition occurring in the absence of oxygen [1]. The main product of fast pyrolysis found in a liquid phase is bio-oil with yields of up to 75 wt.%. Liquid yield depends on * Corresponding author. Department of Environmental Engineering, Chulalongkorn University, Bangkok 10330, Thailand. Tel.: þ66 2 218 6689; fax: þ62 2 218 6666. E-mail address: [email protected] (V. Sricharoenchaikul).

biomass type, temperature, hot vapor residence time, char separation, and biomass ash content, the last two may have a catalytic effect on additional vapor cracking [1]. In recent year, biodiesel is an alternative fuel produced from renewable vegetable oils which are widely receiving attention. In particular, Jatropha which is a nonedible plant is one of the most promising renewable energy crops. Jatropha oil may be used directly with slow speed diesel engine or upgraded via transesterification to conventional biodiesel [2]. However, Jatropha residues are needed to be collected and disposed which may create environmental problem. Thus pyrolysis can be a promising residue management that is possible to convert these wastes to liquid fuel products. Generally bio-oil is high in acidity, which would cause instability problem. They also compose of large number of components

0960-1481/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.renene.2013.07.026

Please cite this article in press as: Kaewpengkrow P, et al., Effect of Pd, Ru, Ni and ceramic supports on selective deoxygenation and hydrogenation of fast pyrolysis Jatropha residue vapors, Renewable Energy (2013), http://dx.doi.org/10.1016/j.renene.2013.07.026

2

P. Kaewpengkrow et al. / Renewable Energy xxx (2013) 1e10

(over 300 chemical species), which cause wide distillation temperature range, making crude bio-oil not suitable for direct combustion usage. There are several methods of upgrading and improving bio-oil quality. One of them is catalytic cracking which can be performed either on liquid bio-oil or on pyrolysis vapor to enhance the cracking reactions of heavy molecules as well as removal of the oxygenated species, leading to the production of bio-oil of high heating values and better stability [3]. However, the important challenge for the success of catalytic cracking process is the selection of proper catalysts. There are several researches on catalytic upgrading of pyrolysis by hydrotreatment using Pd, Ru, and Ni on Al2O3 support leading to deoxygenation [1]. Ying et al. [4] used the 0.5Ru/c-Al2O3 catalyst for upgrading liquid fuel from biomass, resulting in the highest conversion of acetic acid and improved properties of bio-oil. Catalytic cracking is similar to that employed in petroleum refining which is accomplished by hydrotreating with conventional catalysts containing metal (Co, Mo, and Ni) on Al2O3 and hydrocracking (zeolites). However the practicality and economic possibility of such methods are still being debated [5]. Therefore the catalytic upgrading of pyrolytic vapor was to be studied in this work. In early researches, zeolites have been widely studied, such as H-ZSM-5 and Al/SBA-15 [2e6]. These catalysts are effective to convert the highly oxygenated compounds to aromatic hydrocarbons, so that the fuel properties of the catalytic bio-oil could be significantly improved. There are fewer studies about upgrading of bio-oil from fast pyrolysis vapor by PyeGC/MS. Pattiya et al. [7] tested the catalytic effects of four catalysts (ZSM-5, Al-MCM-41, Al-MSU-F and MI-575) by PyeGC/MS. The results showed that all tested catalysts produced aromatic hydrocarbons and reduced oxygenated lignin derivatives. Murata et al. [2] studied fast pyrolysis of Jatropha wastes using mainly GC/MS technique with pyrolyzer in both the absence and presence of catalysts. The result showed that aromatics are largely formed above 90% of GC peak area percentages by using catalysts (H-ZSM-5 and b-zeolite). French and Czernik [5] compared several synthesized and commercially available zeolite catalysts in a pyrolysisegas chromatography mass spectrometry (PyeGC/MS) study. Maximum hydrocarbon yields of 16 wt.% were reported with Ni, Co, Fe and Gasubstituted ZSM-5 catalysts with the latter performing better than other zeolite catalysts. However zeolites are difficult to synthesize and also inherit high production cost. They are difficult to be industrially utilized for upgrading biomass pyrolysis vapors. Modification of the metal catalysts was performed by the impregnation method which has not only low production cost, but also fine hydrothermal stability which makes them possible for treating biomass pyrolysis vapor. These were reported by Lu et al. [8] who studied catalytic upgrading of biomass fast pyrolysis vapors by 5% Ce, 0.5%Ru or Pd with titania and zirconia/anatase based catalysts. The catalyst was prepared via the dry impregnation method. The highest hydrocarbon content of 13.1% was obtained by the zirconia/ anatase based catalysts. There have been several works using ceria (CeO2) based catalysts, mostly for promotion of hydrogen production and prevention of coke deposition during the gasification and pyrolysis process [9e13]. In addition, metals such as Ni, Pd, Pt and Ru have been added to the ceria-based catalyst system to improve the catalytic performance [14]. Therefore Ce-based modified catalysts were prepared for this study, one modified with metal proved their good activity and stability in further fast pyrolysis sequential process. The aims of this study are to investigate the effect of pyrolysis temperature and the activity of catalysts on the distribution of pyrolytic product. The effect of noble metal catalytic activity on pyrolytic products would be examined. The modifications of catalysts were performed by the impregnation of three noble metals

(Pd, Ru and Ni) on Al2O3 and ZrO2 catalyst supporters with ceria (CeO2) as a promoter. Catalytic pyrolysis using the PyeGC/MS instrument would allow direct analysis of the evolved pyrolytic products. The results would suggest the suitability of using these catalysts for upgrading of bio-oil from fast pyrolysis of Jatropha residue. 2. Materials and method 2.1. Raw materials Jatropha residue used as a raw material for fast pyrolysis was obtained from local plants in Chiangrai province in northern Thailand (Jatropha-CR). The residue was left after extraction of oil from biodiesel production process. The residue was first separated from physical impurities and dried at 333 K to reduce moisture contents. The dried sample was crushed with a grinder equipped with 3 mm-mesh. The sample was then passed through sieves to particle size of less than 125 mm. The residue feedstock was then dried for 24 h at 383 K in the oven and its main characteristics are shown in Table 1. The proximate analysis was performed following ASTM to classify the sample in terms of moisture (ASTM E871), volatile matter (ASTM E872), fixed carbon (ASTM E872) and ash (ASTM D1102). Generally, the volatile matter of 73.80 wt.% mainly consists of various organic compounds. Relatively low ash content of 5.80% is a good indicator that this waste should be a good candidate for the production of fuel via pyrolysis process. The ultimate analysis carried out using Leco TruSpecÒ CHNS (micro) analyzer gives the composition of Jatropha in weight percentage of carbon, hydrogen, sulfur, nitrogen and oxygen. Similarly to typical biomass, carbon (49.03 wt.%) is the main element in the residue while hydrogen, nitrogen, and oxygen contents are 6.35, 4.47, and 39.05 wt.%, respectively. Nitrogen plays a major role as precursor for the formation of inorganic compounds from the fuel which may be corrosive substances, not suitable for the engines. The oxygen content is considerably high which could yield unfavorable oxygenated pyrolysis liquid products. Low sulfur content suggests that this waste is suitable for the production of liquid fuel with potentially high energy content and low sulfur by-products. Pyrolysis of biomass is significantly dependent on the main components of cellulose, hemicelluloses and lignin. The chemical structures of these three main components were also identified using TAPPI T203 and T222 standard methods. Jatropha residue contains large amount of hemicellulose (59.20 wt.%) with little cellulose and lignin (18.00 and 22.80 wt.%), suggesting that this waste can be effectively converted into liquid fuel by pyrolysis process at low temperature. 2.2. Catalyst preparation Ten catalyst samples were used to investigate their influence on pyrolysis product distribution and then to predict the change in

Table 1 Proximate, ultimate and component analyses of the Jatropha residue. Proximate analysis

Content (wt.%)

Ultimate analysis

Content (wt.%)

Component analysis

Content (wt.%)

Moisture Ash Volatile organic Fixed Carbon

6.80 5.80 73.80 13.60

Carbon Hydrogen Nitrogen Sulfur Oxygena

49.03 6.35 4.47 1.10 39.05

Cellulose Hemicellulose Lignin

18.00 59.20 22.80

a

By difference.

Please cite this article in press as: Kaewpengkrow P, et al., Effect of Pd, Ru, Ni and ceramic supports on selective deoxygenation and hydrogenation of fast pyrolysis Jatropha residue vapors, Renewable Energy (2013), http://dx.doi.org/10.1016/j.renene.2013.07.026

P. Kaewpengkrow et al. / Renewable Energy xxx (2013) 1e10

Al2O3

ZrO2

CeO2

PdO

RuO2

NiO (f)

Intensity

(e)

(d) (c) (b) (a) 20

30

40

50 Two-theta (deg)

60

70

80

Fig. 1. XRD patterns of alumina based catalysts (a) PdeCe/Al2O3, (b) RueCe/Al2O3, (c) NieCe/Al2O3 and zirconia based catalysts (d) PdeCe/ZrO2, (e) RueCe/ZrO2, (f) NieCe/ ZrO2.

bio-oil properties. All catalyst samples were ground to particle size of less than 125 mm. Catalytic testing was performed by using Al2O3, ZrO2 based catalysts and their modified ones with impregnation of Pd, Ru, and Ni, respectively. The catalyst supports were employed in this research such as alumina oxide (SigmaeAldrich) and zirconium (Alfa-Aesar). Catalyst preparation was commenced by stirring mixture of cerium (III) nitrate solution and aluminum oxide (Al2O3) for 3 h at 353 K and 300 rpm, then dried at 283 K for 24 h. The dried Ce (NO3)2/Al2O3 was calcined in air at 973 K for 4 h to obtain 6% CeO2/Al2O3 which acts as a promoter to achieve well dispersion of the noble metal. After that, palladium nitrate (or ruthenium chloride hydrate, nickel nitrate) was then impregnated in CeO2eAl2O3 catalyst by stirring at 353 K and 300 rpm for 3 h, dried at 383 K for 24 h, and calcined in air at 973 K for 4 h. The catalysts contained 1.0 wt.% Pd on 6%CeO2/Al2O3catalyst (and Ru or Ni-CeO2/Al2O3 catalyst). By the same method, the Pd, Ru and Ni were impregnated in CeO2eZrO2 catalyst. Finally, the synthesized catalysts were 1% Pde6%CeO2/ZrO2, 1%Rue6%CeO2/ZrO2 and 1%Nie6%CeO2/ZrO2, a total of eight catalysts were prepared and tested in this study. 2.3. Characterization of catalysts After impregnation with promoter (Ce) and active phase on Al2O3 support, the particle sizes of prepared catalysts increased significantly from 11 mm to 27e65 mm. Specific surface areas (SBET) were determined from adsorption isotherms using the BET equation. SBET of Al2O3 based catalyst were measured as 85e90 m2/g which were slightly lower than that of Al2O3 support (103 m2/g).

3

Among two catalyst groups, alumina based catalyst has the highest surface area but appeared in large agglomerated sizes (10e65 mm). The particle sizes (9e11 mm) and surface area (10e12 m2/g) of ZrO2 supported catalysts did not change compared to that of the support. X-ray elemental analysis was also used to estimate the amount of elemental deposition. The X-ray diffraction (XRD) patterns of the catalysts were obtained using X’Pert Pro MPD diffract meter with a Cu Ka radiation operated at 40 kV and 40 mA. The X-ray diffraction patterns of alumina based catalysts calcined at 973 K are displayed in Fig. 1(aec). The highest peak intensity at 2q of 45.32 and 67.32 matched to the tetragonal structure of Al2O3 according to JCPDS: 46-1131. After impregnating with Ce(NO3)2 promoter, the minor peaks, 2q of 28.55 and 56.44 matched to the CeO2 cubic phase (JCPDS: 34-0394) were also observed. When further doping with various active species, Al2O3 and CeO2 main phases were found accompanied by a small amount of active metal phase. The small peaks of Pd, Ru and Ni doped catalyst were indexed as tetragonal of PdO phase (JCPDS: 34-1101), and RuO2 phase (JCPDS: 40-1290), and hexagonal of NiO phase (JCPDS: 44-1159), respectively. The X-ray diffraction patterns of zirconia based catalysts are displayed in Fig. 1(def). It was found the highest peak intensity at 2q of 28.174 and 31.467 matched to the monoclinic structure of ZrO2 according to JCPDS: 37-1484. Morphologies of active metal on Al2O3 and ZrO2 based catalyst were investigated by SEM analysis, conducted after calcination at 973 K, and are shown in Fig. 2(aed). Energy dispersive X-ray spectrometry (EDS) was used to investigate the elemental distributions of the synthesized catalysts and approximate amount of the element. The metal peaks of Pd, Ru as well as Ce were detected from the EDS analysis of the prepared catalysts. The Pd and Ru contents on alumina and zirconia of 1.07e1.19% and 1.64e2.01%, respectively, were consistent with the intended metal loadings. 2.4. Experimental apparatus and method Pyrolysis experiments were carried out in a pyroprobe (multifunctional pyrolyzer, PY-2020iD, Frontier Lab) with auto-shot sampler AS-1020E interfaced to a gas chromatograph coupled to a mass selective detector (GC/MS-QP2010, Shimadzu) to separate and identify the evolved products. The sample was placed in a sample cup and the catalyst was placed above Jatropha residue. The Jatropha residue powder used in each experiment was 0.4 mg and subjected to pyrolysis in a pyroprobe at three different temperatures, 673, 773 and 873 K, respectively, in helium atmosphere. During the catalytic experiment, the amount of catalysts used were 0.4 and 2.0 mg to obtain Jatropha:catalyst ratio of 1:1 and 1:5, respectively. During each experiment, sample was placed in an auto-shot sampler using a quartz wool plugs and then dropped into the heated zone of pyroprobe and hold for 0.50 min. Vapor from

Fig. 2. SEM coupled to EDS of catalysts calcined at 973 K (a) PdeCe/Al2O3, (b) RueCe/Al2O3, (c) PdeCe/ZrO2, (d) RueCe/ZrO2.

Please cite this article in press as: Kaewpengkrow P, et al., Effect of Pd, Ru, Ni and ceramic supports on selective deoxygenation and hydrogenation of fast pyrolysis Jatropha residue vapors, Renewable Energy (2013), http://dx.doi.org/10.1016/j.renene.2013.07.026

4

P. Kaewpengkrow et al. / Renewable Energy xxx (2013) 1e10

rapid pyrolysis was led through the quartz tube for an instant analysis by gas chromatograph. The GC separation was carried out by an ultra-alloy capillary column (UA5-30M-0.25F, 30 m, 0.25 mm i.d., 0.25 mm film thickness). Helium (99.999%) was used as a carrier gas with a gas flow rate of 1.1 mL/min, and the split injector ratio was 1:50. During the analysis of pyrolytic products, the oven temperature was started from 323 K (3 min) to 473 K (heating rate of 5 K/min) then to 623 K (heating rate of 10 K/min and hold for 10 min). The injector and detector temperature was held at 553 K. The separated compounds were then analyzed using a mass selective detector operating in electron impact mode (EI) at 70 eV from mass ions (m/z) of 20e800. 3. Results and discussions 3.1. Effect of temperature on yields of pyrolytic products The effect of pyrolysis temperature on yields of pyrolytic products from PyeGC/MS analysis of Jatropha residue (<125 mm) is shown in Fig. 3. The Jatropha residue was pyrolyzed at 673e873 K. The yields of pyrolytic products were represented by percentages of peak area. Fast pyrolysis initiates decomposition of the Jatropha residue to produce a complex mixture of products. The main compounds in the vapors of the pyrolysis of Jatropha residue were classified into twelve groups, such as alcohols, aldehydes, aromatic HC, carboxylic acids, ether, esters, furans, hydrocarbons, ketones, N-compound (N-containing hydrocarbon), phenol, and sugar. Biomass pyrolysis products were generally known as a complex combination of products from pyrolysis of cellulose, hemicellulose, lignin, and extractives, each of which has its own kinetic characteristics. Firstly, the decomposition of lignin led to formation of phenolic compounds. Secondly, the depolymerization of holocellulose (cellulose and hemicellulose) generated anhydroglucose (levoglucosan), related derivatives and furan. The structures of sugar (levoglucosan and a-D-glucopyranose) from pyrolysis of Jatropha residue are shown in Fig. 4. Whereas, the pyrolytic ring scission of holocellulose produced various light products such as aldehydes, ketones, alcohols, and esters [2]. As shown in Fig. 3, the yields of sugar gradually reduced with increasing temperature from 673 to 873 K. It was inferred that high temperature showed high efficiency on cracking heavy compounds from cellulose and hemicelluloses. This result agreed with Lu et al. [15] who studied the effects of pyrolysis temperature and time on the distribution of pyrolytic products. They found that the cellulose

Fig. 4. Structure of sugars from fast pyrolysis of Jatropha residue.

started decomposition to form organic volatile products at the set pyrolysis temperature of 673 K and the LG (levoglucosan) was favorable to be produced at low temperatures. The yield of aromatic hydrocarbon and hydrocarbon compounds (aliphatic and cyclic) tended to increase when heating temperature up to 873 K. Biomass liquids are known to have an acidic structure. From Fig. 3, the main products of pyrolysis vapors of Jatropha residue were fatty acid (palmitic acid, oleic acid and acetic acid) which contained carboxylic acid of 41.11e63.86%. The highest yield was obtained at 673 K. The carboxylic acids are also good indicators of the quality of bio-oils in term of pH values. Large amount of acids can cause corrosion problem in the engines [2e5]. Some of the complex pyrolytic products of Jatropha residue are shown in Fig. 5, which were grouped under oxygenated compounds. Fig. 5(aec) showed structure of free fatty acids (FFA) which are usually found in Jatropha oil. These results are similar to Murata et al. [2] who found high content of FFA from Jatropha waste in non-catalytic pyrolysis. However their products would be completely deoxygenated to form linear hydrocarbon compounds using catalyst cracking in another experiment. Oxygenated compounds presented in slight amount were ketone, alcohol, aldehyde, and furan at lower temperature. The increase in temperature increased their presence to 18.93% and 8.44% for pyrolysis temperature of 773 and 873 K, respectively. High yield of 10.73% ester was obtained at pyrolysis temperature of 773 K. It is possibly due to carboxylic acid conversion into ester form, the

70 673K

773K

873K

60

% Peak area

50

40

30

20

10

0

Fig. 3. The effect of pyrolysis temperature on pyrolytic products.

Please cite this article in press as: Kaewpengkrow P, et al., Effect of Pd, Ru, Ni and ceramic supports on selective deoxygenation and hydrogenation of fast pyrolysis Jatropha residue vapors, Renewable Energy (2013), http://dx.doi.org/10.1016/j.renene.2013.07.026

P. Kaewpengkrow et al. / Renewable Energy xxx (2013) 1e10

5

Fig. 5. Typical of carboxylic acid (aec) and oxygenated compounds of pyrolytic products (deh).

composition of biodiesel, which is advantageous for operation at this temperature. This is a significant influence of pyrolysis temperature; the oxygenated compounds from pyrolytic product are a marking for fuel quality and the abatement amount of oxygenated compounds in the liquid fuel to obtain the higher energy content [16]. The result showed that high temperature (873 K) had positive influence on the yields of pyrolytic products. However, the yield of aromatic and hydrocarbon compounds was low compared with other pyrolytic products. This result agreed with several researches [2,3,16,17] who found the low yield of aromatic in non-catalytic fast pyrolysis of biomass. 3.2. Catalytic pyrolysis of Jatropha residue with alumina based catalysts (Jatropha to catalyst ratio of 1:1 and 1:5) The effect of catalytic pyrolysis and Jatropha to catalyst ratio on yields of pyrolytic products from Jatropha residue is shown in Fig. 6. The Jatropha residue was pyrolyzed at 873 K. The main compounds in the vapors of the pyrolysis of Jatropha residue were classified into ten groups, alcohols, aldehydes, aromatic HC, carboxylic acids, esters, furans, hydrocarbons, ketones, N-compound (N-containing hydrocarbon), and phenol. It can be clearly seen that every catalyst

has an effect on yields of pyrolytic compounds. In terms of levoglucosan and D-glucose (sugar), the yield of sugar was not found with any catalysts. From previous experiment, the yields of sugar gradually reduced with increasing temperature from 673 to 873 K. Therefore it could be completely eliminated by every catalyst at the highest temperature (873 K). The results agree well with Lu et al. [6,8] that levoglucosan is easy to be converted by catalyst and it is significantly reduced or even completely eliminated. Whereas phenol compounds (lignin degradation products) were detected in the presence of catalyst which are regarded as value-added chemicals because they can be used for production of resins and other products [3,6]. The highest yield of phenol (4.52%) was found after catalyzed by the RueCe/Al2O3, while Lu et al. [8] found that the phenols were increasing after catalyzed by the Pd/CeTiO2. Pyrolytic product from holocellulose produced various light products of oxygenated compounds such as aldehydes, ketones, alcohols, and esters. The catalytic effects on major oxygenated compounds indicated that the presence of catalyst did not significantly reduce aldehyde and ketones. Aldehyde yields increased noticeably from 0.36% to 4.99e22.05% after catalysis process. The highest yield was obtained after catalyzed by Ce/Al2O3 which formed (z)-9-hexadecenal as shown in Fig. 5(h). This catalyst

Fig. 6. Effects of catalysts on the pyrolytic products.

Please cite this article in press as: Kaewpengkrow P, et al., Effect of Pd, Ru, Ni and ceramic supports on selective deoxygenation and hydrogenation of fast pyrolysis Jatropha residue vapors, Renewable Energy (2013), http://dx.doi.org/10.1016/j.renene.2013.07.026

6

P. Kaewpengkrow et al. / Renewable Energy xxx (2013) 1e10

30

(1:1)

(1:5)

C5-C10 C12-C20 > C20

25

% Peak area

20 15 10 5 0

Fig. 7. Hydrocarbon selectivity for fast pyrolysis of Jatropha residue with Al2O3 based catalysts.

promoted aldehyde formation which is mainly responsible for the aging reactions and instability of bio-oil. Other oxygenated compounds presented in slight amount were ketone, alcohol, and furan. It was noticed that yield of furan which is toxic and may be carcinogenic was not found with the presence of RueCe/Al2O3 catalyst. The main organic acid presents in large amount in non-catalytic trials are oleic acid and palmitic acid. Fig. 6 shows that all catalysts led to the decrease in acid from 60.74% to 30.00%. In addition, the yield of ester increased with the presence of catalyst. Since long chain fatty acids composed of mono alkyl esters [2], therefore it was found that ester increased after catalyzed cracking of fatty acid. This is also the case for bio-oil derived from vegetable oil. Generally, bio-oil contains some heavy aromatic compounds while gasoline and diesel include mixtures of hydrocarbons with relatively low chain lengths of 5e10 and 12e20, respectively [17]. The most important point at investigating the similarity of bio-oil as a fuel is the aliphatic compounds, i.e., alkanes and alkenes [16]. From the result, the major compounds were aliphatic hydrocarbons which were presented by the carbon range of C5eC10 and C12eC20

as shown in Fig. 7. The main hydrocarbons are 1-heptene, 3-tetradecene, and tetracosane. For Jatropha residue to catalyst ratio of 1:1, the hydrocarbon compounds’ yields increased for all catalysts. Yield of hydrocarbon compounds increased from 5.96% to 12.58e23.38% with the highest yield obtained by Al2O3 catalyst. It might be due to larger surface area of Al2O3 than other catalysts. High carbon chain lengths of 12e20 were the main hydrocarbon products (4.41e8.89%) after catalyzed, which are suitable for diesel fuel. The result indicated that these catalysts were effective for deoxygenation of the pyrolysis vapors to form hydrocarbons. Fig. 8 shows the aromatic distribution from catalytic fast pyrolysis with different catalysts, which includes monoaromatics such as ethylbenzene, toluene, and styrene. The structures of aromatic and hydrocarbon are shown in Fig. 9. The aromatic selectivity obviously increased with the presence of all catalysts. The selectivity of aromatic hydrocarbon increased from 3.58% to 5.22e7.31% of which the highest yield obtained by PdeCe/Al2O3 catalyst. Aromatic hydrocarbons are considered as valuable products because of their contribution to the high heating value [18].

10.00 (1:1)

(1:5)

Benzene, ethylStyrene Toluene

8.00

% Peak area

6.00

4.00

2.00

0.00

Fig. 8. Aromatic hydrocarbon selectivity for fast pyrolysis of Jatropha residue with Al2O3 based catalysts.

Please cite this article in press as: Kaewpengkrow P, et al., Effect of Pd, Ru, Ni and ceramic supports on selective deoxygenation and hydrogenation of fast pyrolysis Jatropha residue vapors, Renewable Energy (2013), http://dx.doi.org/10.1016/j.renene.2013.07.026

P. Kaewpengkrow et al. / Renewable Energy xxx (2013) 1e10

7

Fig. 9. Products of catalytic fast pyrolysis using catalyst are typically aromatic hydrocarbons (aed) and linear hydrocarbon (eeg).

Among these compounds, the presence of PAH deserves more attention due to their toxicity in environmental and potential roles in the incidence of cancer [6]. Fortunately the PAH from catalytic fast pyrolysis of Jatropha residue was not detected which is another main advantage for pyrolysis in the presence of these catalysts. Murata et al. [2] found more than 20% of PAHs (such as naphthalene) with ZSM-5. Several researches [2,4,16] showed that catalysts produced aromatic hydrocarbons and reduced oxygenated lignin derivatives. Other important group of compounds that is affecting the quality of the bio-oil is N-compound. From the results, fast pyrolysis of Jatropha generated various N-containing hydrocarbon compounds in high contents (11.56%). After catalyzed by catalysts, the N-containing hydrocarbon compounds increased to 26.19e32.41%. Al2O3 and NieCe/Al2O3 favored the formation of N-containing hydrocarbon compounds. With increasing Jatropha:catalyst ratio to 1:5, it can be clearly seen that yield of overall hydrocarbon increased through the higher aliphatic and aromatic hydrocarbon compounds (Figs. 7 and 8). The selectivity of aromatic hydrocarbon increased to 8.15e16.80% with the highest yield obtained by PdeCe/Al2O3 catalyst. Moreover high yield of hydrocarbon compound was increased to 25.86e40.60% with the highest yield obtained by NieCe/Al2O3 catalyst. It is apparent that short chain hydrocarbon (C6eC12) increased with higher catalyst to Jatropha ratio in all catalysts. This implied that these catalysts were effective for cracking large molecules of the pyrolysis vapors to form smaller hydrocarbon molecules. All catalysts led to the decrease in main carboxylic acids from 60.74% to less than 20.00% with Jatropha to catalyst ratio of 1:5.

Al2O3 was the most effective in which acid could be reduced completely. Moreover greater amount of catalyst could reduce the contents of oxygenated compounds (alcohol, aldehyde, furan, ketone, and phenol) almost completely, especially furan. Since Jatropha contains relatively high quantity of nitrogen, the result showed relatively high percentages of N-containing compounds. N-containing compounds in liquid fuel are undesirable since they are sources of NOx when combusted as well as corrosive nature of these substances [19]. The results indicated that the higher Jatropha to catalysts ratio promoted N-containing compound formation which is not preferable. The high yield of N-containing compound increased from 26.19e32.41% to 31.17e 40.37% with Jatropha to catalysts ratio of 1:5. The highest yield obtained by Al2O3 catalyst while higher PdeCe/Al2O3 and RueCe/ Al2O3 contents did not significantly increase N-compound. The main N-containing compounds contain the heteroatom (oleanitrile, olelic acid amide, hexadecanamide and aniline) and their structures are show in Fig. 10. These products may create problems such as corrosion, poisoning of catalysts during bio-oil upgrading [20]. As a result, these catalysts are not effective on reducing N-containing compounds. Therefore it is desirable to remove the N-containing compounds prior to using these renewable biofuels through denitrogenation reactions. In general the hydroprocessing reactions require high pressure and hydrogen to remove N-heteroatoms. This process is similar to the removal of N-containing compounds out of petroleum feedstocks through a series of hydrodenitrogenation (HDN) as explained by equation (1).

C4 H5 N þ 3H/C4 H8 þ NH3

(1)

Fig. 10. N-containing products from catalytic fast pyrolysis of bio-oil vapor.

Please cite this article in press as: Kaewpengkrow P, et al., Effect of Pd, Ru, Ni and ceramic supports on selective deoxygenation and hydrogenation of fast pyrolysis Jatropha residue vapors, Renewable Energy (2013), http://dx.doi.org/10.1016/j.renene.2013.07.026

8

P. Kaewpengkrow et al. / Renewable Energy xxx (2013) 1e10

Table 2 The effect of catalytic fast pyrolysis vapors of biomass on product selectivity. Type of catalyst

Biomass to catalyst ratio

SBA-15, Al/SBA-15 H-ZSM-5 H-ZSM-5 ZSM-5 Pd, Ru, or Ce/ZrO2 Pd, Ru, or Ni on Ce/Al2O3 or Ce/ZrO2

1:1 1:6.25 1:5 1:1 1:2 1:1 and 1:5

The main HDN products are n-butane and n-pentane while nitrogen was removed as ammonia which can be stored as chemical feedstock [19,20]. 3.3. Catalytic performance of the Al2O3 based catalysts and ZrO2 based catalysts In recent years, mesoporous catalysts (zeolites) with much higher pore sizes and surface area are widely used by several researches. However, due to their difficulties to be synthesized and high production cost, they are hard to be industrially utilized for upgrading of biomass pyrolysis vapors. Several researches have used zeolites as catalysts for pyrolysis of biomass that led to upgrading of bio-oil. Lu et al. [6], compared non-catalytic and catalytic pyrolysis of sawdust in the presence of different zeolites. They observed that the catalytic cracking reduced the yields of light aldehydes and ketones, while increased the formation of hydrocarbons, but their yields were low of only 0.14% peak area. However Lu et al. [8] have used meso- or macroporous catalysts (TiO2 (rutile), TiO2 (anatase) and, ZrO2/TiO2) for catalytic upgrading of biomass fast pyrolysis vapors. The results indicated that these catalysts increased the formation of hydrocarbons and phenols while reduced the yields of sugar similar to zeolite catalysts. Furthermore Pd and Ru-based catalysts were investigated by several researches for improvement of the quality of pyrolysis oils by hydrogenation and deoxygenation. The effects of catalytic pyrolysis on product distribution were compared with other studies as shown in Table 2. Product selectivity depends on biomass type and catalyst used in the process. Therefore, the types of support and noble metal

Pyrolysis product (% peak area)

References

Hydrocarbon

Aromatic

Phenol

0.14 e e e 0.1e13.1 12.71e40.60

e 90 w9e14 w7 e 0.99e16.80

w7.5e16 e 0.087e0.22 0.55 10.6e37.2 1.49e4.51

[6] [2] [3] [7] [8] This paper

catalysts were investigated in order to upgrading the properties of the bio-oil. The catalytic effects on product distribution of pyrolytic vapors catalyzed with Al2O3 and ZrO2 based catalysts with the Jatropha residue to catalyst ratio of 1:5 were investigated and results are shown in Fig. 11. The highest hydrocarbon content of 24.30% was obtained by the RueCe/ZrO2 compared with only 7.74% in the non-catalytic products, while aromatic yields were only 0.70e6.14%. This may be due to the low metal content loading (1% w/w) on the support and low surface area of ZrO2 (12 m2/g). This result is similar to Lu et al. [8] who studied catalytic upgrading of wood by 5 wt.% of Ce, 0.5 wt.% Ru or Pd with TiO2 and ZrO2/TiO2 based catalysts. From the result, it was found that the overall trend of the carboxylic acid group decreased with the presence of catalyst. The major compound in carboxylic group is oleic acid and all catalysts led to the lower content of oleic acid to 23.40e28.61%. In addition, aldehyde increased to 11.10% after catalyzed by NieCe/ZrO2 while ketone increased to 9.11%. The yields of ester increased after catalysis process to 8.42% whereas the contents of phenols and alcohol did not significantly change. Moreover larger amount of N-containing compounds are measured (15.00e25.69%), as compared to the noncatalytic condition of only 10.02%. Hydrocarbon compound increased to 12.71e24.30% with the highest yield obtained by Rue Ce/ZrO2 catalyst. Fig. 12 showed that all Al2O3 based catalysts exhibited relatively higher activity toward deoxygenation reactions than those of ZrO2 catalysts which resulted in high aliphatic hydrocarbons. It is apparent that long chain hydrocarbon (C12eC20) increased with the presence of catalyst in all ZrO2 based catalysts except Ce/ZrO2 implying that these catalysts were not effective for

Fig. 11. Pyrolytic products of Jatropha residue with Al2O3 catalysts and ZrO2 catalysts at 873 K using Jatropha to catalyst ratio of 1:5.

Please cite this article in press as: Kaewpengkrow P, et al., Effect of Pd, Ru, Ni and ceramic supports on selective deoxygenation and hydrogenation of fast pyrolysis Jatropha residue vapors, Renewable Energy (2013), http://dx.doi.org/10.1016/j.renene.2013.07.026

P. Kaewpengkrow et al. / Renewable Energy xxx (2013) 1e10

9

30 C5-C10

C12-C20

> C20

% Peak area

25

20

15

10

5

0

Fig. 12. Hydrocarbon compounds selectivity for fast pyrolysis of Jatropha residue with Al2O3 and ZrO2 based catalysts.

1H-Pyrrole 1H-Pyrrole, 2-methylPyridine Hexadecanamide 9-Octadecenamide, (Z)Oleanitrile

14.00 12.00

% Peak area

10.00 8.00 6.00 4.00 2.00 0.00

Fig. 13. N-containing hydrocarbon compounds selectivity for fast pyrolysis of Jatropha residue with Al2O3 and ZrO2 based catalysts.

cracking large molecules of the pyrolysis vapors. It may be due to high surface area of Al2O3 (103.38 m2/g). These results are similar to several researches [2,8] where the effect of support on the pyrolysis products of Jatropha residue seems to be related to the surface area of catalyst. Among Al2O3 based ones, PdeCe/Al2O3 showed the greatest decrease in oxygenated compounds with a relatively high hydrocarbon yield. While among ZrO2 based, RueCe/ZrO2 showed the highest decrease in N-compounds with a relatively high hydrocarbon yield. Another attention should be paid to the N-containing hydrocarbons, and the results are shown in Fig. 13. Even though the oxygenated compounds’ yields of both catalysts were not much different, it appeared that Al2O3 based catalyst promoted N-compounds formation such as N-heterocyclic compounds (1H-pyrrole, 2-methyl-1H-pyrrole and pyridine) and N-heteroatom (hexadecanamide, 9-octadecanamide and oleanitrile). Total N-containing compounds are more than 30% for Al2O3 catalysts (31.17e40.37%) which are higher than ZrO2 based catalysts (15.00e25.69%). The highest N-compound yield was obtained by Al2O3 based in which the result indicated that the doping of metal catalyst on Al2O3 support

was effective to reduce the production of N-containing compounds. However some unfavorable N-containing components still existed in the products suggesting further denitrogenation is required as previously mentioned. 4. Conclusion Catalytic fast pyrolysis of Jatropha residue for upgrading of the pyrolytic vapors was performed using analytical pyrolysisegas chromatography/mass spectrometry (PyeGC/MS) at 673e873 K. The result suggested that high temperature had positive influence on low yields of oxygenated compounds and high yields of hydrocarbon compounds. The synthesized catalysts were tested for catalytic upgrading of Jatropha fast pyrolysis vapors and their activities on changing pyrolysis products of Jatropha residue. Levoglucosan was significantly reduced or completely eliminated after catalysis process. For Jatropha residue to catalyst ratio of 1:1, the hydrocarbon compounds’ yields increased in all catalysts and the highest yield obtained by Al2O3 catalyst. The PAH from catalytic fast pyrolysis of Jatropha residue was not detected which is the

Please cite this article in press as: Kaewpengkrow P, et al., Effect of Pd, Ru, Ni and ceramic supports on selective deoxygenation and hydrogenation of fast pyrolysis Jatropha residue vapors, Renewable Energy (2013), http://dx.doi.org/10.1016/j.renene.2013.07.026

10

P. Kaewpengkrow et al. / Renewable Energy xxx (2013) 1e10

advantage of using these catalysts. With increasing Jatropha to catalyst ratio to 1:5, NieCe/Al2O3 appeared to increase the formation of hydrocarbon compounds but also increased unfavorable aldehyde and other oxygenated compounds. PdeCe/Al2O3 has good selectivity toward the reduction of all oxygenated lignin-derived compounds and N-compounds while enhanced the formation of aromatic hydrocarbons and phenols. Among ZrO2 supported catalysts, the highest hydrocarbon was obtained by the Ru/Ce-ZrO2. Al2O3 based catalysts produced more aromatic and hydrocarbon compounds than ZrO2 based catalysts. High yield of carboxylic acid was also detected when using ZrO2 based catalysts. However Al2O3 based catalysts seemed to promote N-compounds so it is necessary to be removed prior to using this renewable biofuel. Results indicate that these catalysts may be applied to improve the properties of pyrolytic products’ vapors from Jatropha residue. Acknowledgments This work was partially supported by Integrated Innovation Academic Center: IIAC Chulalongkorn University Centenary Academic Development Project (CU56-EN12), the Higher Education Research Promotion and National Research University Project of Thailand, Office of the Higher Education Commission (EN1189A56), National Metal and Materials Technology Center (MTEC), and Department of Environmental Engineering, Faculty of Engineering, Chulalongkorn University. P. Kaewpengkrow also appreciates the scholarship provided by the Thailand Graduate Institute of Science and Technology (TGIST, TG-33-09-55-008D). References [1] Bridgwater AV. Review of fast pyrolysis of biomass and product upgrading. Biomass and Bioenergy 2011:1e27. [2] Murata K, Liu Y, Inaba M, Takahara I. Catalytic fast pyrolysis of jatropha wastes. Journal of Analytical and Applied Pyrolysis 2012;94:75e82. [3] Mihalcik DJ, Mullen CA, Boateng AA. Screening acidic zeolites for catalytic fast pyrolysis of biomass and its components. Journal of Analytical and Applied Pyrolysis 2011;92:224e32.

[4] Ying X, Tiejun W, Longlong M, Guanyi C. Upgrading of fast pyrolysis liquid fuel from biomass over Ru/c-Al2O3 catalyst. Energy Conversion and Management 2012;55:172e7. [5] French R, Czernik S. Catalytic pyrolysis of biomass for biofuels production. Fuel Processing Technology 2010;91:25e32. [6] Lu Q, Li W-z, Zhang D, Zhu X-f. Analytical pyrolysisegas chromatography/ mass spectrometry (PyeGC/MS) of sawdust with Al/SBA-15 catalysts. Journal of Analytical and Applied Pyrolysis 2009;84:131e8. [7] Pattiya A, Titiloye JO, Bridgwater AV. Fast pyrolysis of cassava rhizome in the presence of catalysts. Journal of Analytical and Applied Pyrolysis 2008;81:72e9. [8] Lu Q, Zhang Y, Tang Z, Wz Li, Xf Zhu. Catalytic upgrading of biomass fast pyrolysis vapors with titania and zirconia/titania based catalysts. Fuel 2012;89:2096e103. [9] Tomishige K, Asadullah M, Kunimori K. Syngas production by biomass gasification using Rh/CeO2/SiO2 catalysts and fluidized bed reactor. Catalysis Today 2004;89:389e403. [10] Miyazawa T, Kimura T, Nishikawa J, Kado S, Kunimori K, Tomishige K. Catalytic performance of supported Ni catalysts in partial oxidation and steam reforming of tar derived from the pyrolysis of wood biomass. Catalysis Today 2004;89:389e403. [11] Adrados A, Urionabarrenechea AL, Solar J, Requies J, Marco ID, Cambra JF. Upgrading of pyrolysis vapours from biomass carbonization. Journal of Analytical and Applied Pyrolysis 2013 xxx:xxxexxx. [12] Westrich TA, Chen X, Schwank JW. Isooctane decomposition and carbon deposition over ceriaezirconia supported nickel catalysts. Applied Catalysis A: General 2010;386:83e93. [13] Pattiya A, Titiloye JO, Bridgwater AV. Evaluation of catalytic pyrolysis of cassava rhizome by principal component analysis. Fuel 2010;89:244e53. [14] Iojoiu EE, Domine ME, Davidian T, Guilhaume N, Mirodatos C. Hydrogen production by sequential cracking of biomass-derived pyrolysis oil over noble metal catalysts supported on ceria-zirconia. Applied Catalysis A: General 2007;323:147e61. [15] Lu Q. Yang Xiao-chu, Dong Chang-qing, Zhang Zhi-fei, Zhang Xu-ming, Zhu Xifeng. Influence of pyrolysis temperature and time on the cellulose fast pyrolysis products: analytical Py-GC/MS study. Journal of Analytical and Applied Pyrolysis 2011;92:430e8. [16] Ates F, Ikdag M. Evaluation of the role of the pyrolysis temperature in straw biomass samples and characterization of the oils by GC/MS. Energy & Fuels 2008;22:1936e43. [17] Bulushev DA, Ross JRH. Catalysis for conversion of biomass to fuels via pyrolysis and gasification: a review. Catalysis Today 2011;171:1e13. [18] Compton DL, Jackson MA, Mihalcik DJ, Mullen CA, Boateng AA. Catalytic pyrolysis of oak via pyroprobe and bench scale packed bed pyrolysis reactors. Journal of Analytical and Applied Pyrolysis 2011;90:174e81. [19] Choudhary TV, Phillips CB. Renewable fuels via catalytic hydrodeoxygenation. Applied Catalysis A: General 2011;39:1e12. [20] Duan P, Savage PE. Catalytic hydrothermal hydrodenitrogenation of pyridine. Applied Catalysis B: Environmental 2011;108e109:54e60.

Please cite this article in press as: Kaewpengkrow P, et al., Effect of Pd, Ru, Ni and ceramic supports on selective deoxygenation and hydrogenation of fast pyrolysis Jatropha residue vapors, Renewable Energy (2013), http://dx.doi.org/10.1016/j.renene.2013.07.026