A Study of Ethanol Production Cost for Gasoline Substitution in Thailand and Its Competitiveness

Transcription

1 Thmmst Int. J. Sc. Tech., Vol. 12, No. l, Jnury-Mrch 200'l A Study of Ethnol Production Cost for Gsoline Substitution in Thilnd nd Its Competitiveness Suthmm Yoosin nd Chumnong Sorpiptn Energy Division, The Joint Grdute School of Energy nd Environment, King Mongkut's University Technology Thonburi, 9l Prchuthit Rd., Bngmod, Thung Khru, Bngkok 10140, Thilnd. corresponding uthor: suthmm ;z@yhoo.com Abstrct Drmtic increses in imported crude oil prices hve led to the Thi government seeking ltemtive resources to reduce the imported crude oil. One of the possible solutions is to use bioethnol derived from gro-product such s cssv, com, rice, sugrcne nd molsses, etc., which re vilble bundntly in Thilnd. The Thi government ims to use the ethnol to substitute for gsoline octne 95 in the first step, by prtil substitution for the gsoline octne 95, nd expects to substitute more in the ner future, provided tht this first progrm is fully successful. Therefore, the competitiveness of the ethnol price with the gsoline price is the most importnt fctor for the vibility of this progrm. There re four common csh crops cultivted in Thilnd: cssv, corn, rice, nd sugrcne. In this study, these four csh crops nd molsses, one of the by-products, from sugr production, were studied to see if they cn compete with gsoline 95, if they re used s feed stocks to produce nhydrous ethnol for gsohol 95, El0. A multi-feed ethnol plnt with cpciry of 150,000 liters per dy of nhydrous ethnol is ssumed in this study. The totl cpitl investment of this plnt is expected to be bout 1,200 million bht. The verged feed stock costs during yer were estimted for ech type of the csh crops nd molsses. It ws found tht the feedstock cost is the mjor cost burden in ethnol production. It contributes more thn 50% ofthe totl cost. On verge, the ethnol cost t ex-fctory ws found to be in the rnge of bht per liter of gsoline equivlent (excluding ll txes), depending on type of rw mterils. The verge price of gsoline 95 t ex-fctory price ws I 1.50 bht per liter during the yers Thus, we conclude tht the competitiveness of the ethnol is highly dependent on the feed stocks' prices. The voltility of the feed stockprices in the mrket significntly ffects the competitiveness of the bio-ethnol s compred with the gsoline. In generl, the crops, which re most liiely to be ble to compete with gsoline, re sugrcne nd cssv becuse of their low verge prices s feed stocks. Key words: Feedstocks' prices, Ethnol Production, competitiveness, Gsoline price l. Introduction The present trend in depletion of globl fossil fuel energy hs cused n ever-incresing need to seek lterntive sources of energy. Currently, the Thi government hs promoted the use of ethnol from some griculturl csh crops s n ltemtive for fossil fuel substitution in the trnsporttion sector. Ethnol cn be used in vrious methods s trnsporttion fuel. One of the well known fuel pplictions of ethnol is by mixing it into gsoline. Ethnol cn be mixed in gsoline up to l0 percent by volume with only minor modifiction or without modifiction for most conventionl gsoline engines Il]. This blended ethnol is normlly clled "gsohol", nd it is generlly nmed fter n mount of the percentge of ethnol (volumetric) mixer in the gsoline, For exmple, n mount of gsoline 69

2 Thmmst Int. J. Sc. Tech., Vol. 12, No. l, Jnury-Mrch 2007 mixed with 107o ethnol to boost its octne number up to 95, is clled gsohol 95, E 10. ln Thilnd the fuel ethnol progrm ws initited nd officilly nnounced in Since then it hs stimulted much public interest. The Ministry of Energy ims to blend 3 million liters per dy of ethnol into gsoline by Presently, there re 25 ethnol compnies which hve been conceded permission licenses from the Ntionl Ethnol Bord of Thilnd to produce ethnol. It is expected tht these concessionire compnies would hve totl cpcity of production of 4.21 million liters per dy. However, in June 2006, there were only hve ethnol plnts constructed nd operting, by producing ethnol lrom molsses. with totl cpcity of bout million liters per dy. One mjor problem with ethnol production is the vilbility of rw mterils for the production. The vilbility of feedstocks for ethnol cn vry considerbly from seson to seson nd depend on geogrphic loctions. The price of the rw mterils is lso highly voltile, which cn highly ffect the production costs of the ethnol. In order to mke sure tht production cpcity of n ethnol plnt is fully utilized, it is necessry to ssure tht feedstocks of the rw mterils re sufficient for the production throughout yer, nd with competitive price. One possible solution to this problem is to use mny different types of feedstocks for the ethnol production. The min purpose of this study is to investigte the fesibility of ethnol production of multi- type feedstock plnt, from different mjor griculturl csh crops, which re commonly cultivted in Thilnd, nd nlyze the competitiveness of the ethnol price with the gsoline price. The selective griculturl csh crops in this study re cssv, com, nce' sugrcne nd molsses ( by product from sugr production). 2. The ethnol production Process Ethnol, which is lso known s ethyl lcohol, is cler, colorless, flmmble, oxygented hydrocrbon with chemicl formul C2H5OH. Currently, there re two commercilized ethnol production technologies deployed in businesses. The first one is bsed on sugr bsed feedstock, nd the second is bsed on strch bsed feed stock. For the sugr bsed feedstock, ethnol cn be produced by microbil conversion through fermenttion. The fermenttion process relies on yests tht convert sugr to ethnol. Thus, those griculture crops which re sugr bsed feedstocks such s sugrcne, sugr beets, molsses, etc. cn be used for ethnol production. Another type of feedstock, which cn be used for ethnol production, is strch bsed mterils. Strch consists of long chins of glucose molecules nd cn lso be converted to "the fermentble sugr by method clled hydrolysis technique". Hydrolysis is rection of strch with wter, which is normlly used to brek down the strch into fermentble sugr. There re two techniques for hydrolysis: enzymtic hydrolysis nd cid hydrolysis. Yest used in the ethnol fermenttion is generlly bkery yest lscchromyces cerevisie). It is dopted s seeding for the fermenttion. The lcohol concentrtion from fermented sugr produced by the yest initilly is only bout 5-15 % by weight [2]. The lcohol is then upgrded to higher concentrtion by seprting it from wter nd other non-fermentble mterils. The concentrtion of the lcohol is then finlly upgrded to be 95oh-96 % by weight, by distilltion method' The lcohol with 950/o-96%o concentrtion is normlly clled "hydrous lcohol", which cn fuel only specilly designed vehicles such s flex fuel crs. However, ethnol, with purpose to be blended with gsoline, for uses in generl vehicles, must be nhydrous ethnol (99.5% lcohol concentrtion). Thus, the residul remining wter in hydrous lcohol must be removed by dehydrtion process. Currently, moleculr sieve technique is common technique to seprte the wter out from the ethnol to produce nhydrous lcohol. Normlly, the ethnol production of sugr bse crops yields three min by-products: stillge, fusel oil nd crbon dioxide. Stillge is residul beer remining in the distilltion wste. fter the lcohol hs completely been removed from the distilltion columns. During the distilltion process, there re two types of distilltes obtined from the distilltion columns: fusel oil, nd lcohol. Fusel oil is higher order of one type of lcohol with more 70

3 Thmmst Int. J. Sc. Tech., Vol. 12, No. l, Jnury-Mrch 200'l thn two crbons in its moleculr structure, nd it is formed during the fermenttion, in conjunction with the ordinry lcohol with two crbons in its moleculr structure. CO2 is generted during the nerobic sugr fermenttion. The CO2 produced from the fermenttion is lmost eqully to the mount of the ethnol produced in the fermenttion process, by weight. Ethnol production from strch bsed crops gives one more by-product, i.e., Distiller's Dried Grin (DDG). The steps of the ethnol production s mentioned bove re summrized nd shown in Figure 1. }:ERMENTATION DISTILLATION DDG (for strch bse crops only) CONVERSION OF FEEDSTOCK TO FERMENTABLE SUGAR FERMENTATION TO PROTJUCETHANOL ETHANOL RECOVERY Figure I nhydrous ethnol production processes 3. Estimtion of ethnol production costs The processing costs of the ethnol cn be ctegorized into four mjor groups: feedstocks costs, cpitl costs, operting & mintennce costs. nd by-product gins. - Feedstock costs Feedstock prices cn vry by loction, sesons, locl conditions of the supply-demnd, nd trnsporttion. These mrket price vribles cn ffect decision in selecting feedstock type for ethnol production - Operting nd mintennce costs The operting nd mintennce costs re: lbor. energy, electricity, ingredients (e.g. enzymes, yests, etc.), repirs nd mintennce, txes, insurnce cost, nd dministrtive expenses - Cpitl costs Cpitl investment represents the first costs of ll necessry equipment in production nd their instlltions. Expenses for piping, instrumenttion, insultion, foundtions nd site preprtion re included in the cpitl costs. In ddition, these costs lso include lnd, buildings nd wste tretment fcilities. - By-product gins The fermenttion process of the ethnol production lso yields severl byproducts, including crbon dioxide, fusel oil, yests nd stillge. The by-products of the ethnol production cn generte dditionl incomes. In other words, they ssist to reduce the ethnol production cost, significntly, provided tht they re economiclly recovered, effectively. The byproducts re dependent on the types of feedstocks nd processing methods used in the ethnol production. The unit cost ofthe ethnol production cn be expressed by: Cntou : CF + Co&M + Cr - Cn (l) where Crtog:Cost of ethnol production (bht per liter) Cp : Feedstock cost (bht per liter) Co*rr,r : Operting nd mintennce cost (bht per liter) C1 : Investment cost (bht per liter) Cs : Byproducts gins (bht per liter) 3. Technicl ssumptions of the ethnol production in this study In this study, multi feedstocks: cssv, corn, rice, sugrcne nd molsses, re ssumed 7l

4 ThmmstInt. J. Sc. Tech., Vol. 12, No. l, Jnury-Mrch 20Oj to be processed by n ethnol plnt for fuel grde ethnol production, with cpcity of 150,000 litters per dy, nd operting 330 dys per yer. The project life is ssumed to be 20 yers, nd the interest is 6% per nnum. Vlues in the yer 2005 re chosen s the bse yer for cost clcultions. All vlues in previous yers before 2005 were converted to the bse yer, by the Consumer Index (CPI) of ech yer. The yields of ethnol conversion from feedstock to ethnol were clculted bsed on the figures obtined from Thilnd Institute of Scientific nd Technologicl Reserch (TISTR)'s Pilot Plnt. The mounts of other rw mterils supply needed for processing such s enzymes, yests, molsses, wter nd chemicls were lso bsed on the figures given by TISTR [3]. Tble I shows the feedstocks yield nd generl technicl ssumptions used in this nlysis. 4. Estimtion of the feedstock costs The costs of rw mterils highly fluctute in this study.therefore, the prices of cssv' corn, rice, sugrcne nd molsses, during yers 2002 to 2005 [4], were referenced nd they were djusted to their equivlent prices in the yer 2005, s the bse yer. The verge prices during this period were found to be 1,096, 4.668, 4,728, 494 nd 3,800 bht Per ton, respectively. By using the conversion rtes of the ethnol yields from the feedstocks, s shown in Tble l, the mounts of feedstocks required per one liter of ethnol produced could be estimted. It ws found tht the feedstock costs of cssv for the ethnol production vried from 5.75 to 8.56 bht per liter of ethnol. In the cse of com, rice nd molsses, their feedstock costs re quite high, rnging from l 1.92 to bht per liter for corn, to bht per liter for rice nd 12.5 to 20 bht per liter for molsses, respectively. However, for the cse of sugrcne, if the ethnol is produced directly from the juice of sugrcne, its feedstock cost is much cheper, nd vried between 6.29 to 8.24 bht per liter. Tble 2 shows the mximum' minimum, nd verge prices of ech feedstock mentioned bove. The rw mteril cost is the most significnt expense tht ffects the cost of the ethnol production. Unfortuntely, the prices of the feedstocks re generlly highly voltile s shown in Tble 2. We will discuss this mtter in detil gin in the section of discussion nd conclusion. Tble I Feedstocks yield nd technicl ssumptions in this study Feedstock Cssv Com Rice I Susrcne Molsses Ethnol Yield (l/ton) r Plnt's Feedstock InPut Rtes (tons/dy) Plnt Tvpe Multi-Feed nd Stnd Alone Loction Nkorn Rtchsim, Thilnd Annul Production (million liters) 49.5 C)oertion Dvs 330 Proiect Life (yers) 20 Interest Rte 60 per nnum Yer for Cost Bsis 2005 Tbte. 2 The mximum, minimum nd verge of the feedstock costs in bht per ton nd bht liter of ethnol from 2002 to 2005 Fe Mximum Minimum Averse edstock bht/ton r..",n"r bht/l n'oh bht/ton 1""6.1o"1 bht/l nos bht/tons""6"1o"1 bht/l FtoH Cssv r, , Corn 4, ,472 tl.92 4,668 t2.45 Rice 5, , )3, t2.61 Sugrcne Molsses

5 Thmmst Int. J. Sc. Tech., Vol. 12, No. 1, Jnury-Mrch 2O0j 5. Estimtion of the cpitl investment cost ofthe ethnol plnt The cpitl investment cost of n ethnol plnt cn be rrnged into two mjor prts: direct costs nd indirect costs. The direct costs include the costs of ll mchines nd equipment, s well s their instlltion costs. lnd. min plnt building, lbortories, offices nd wrehouses, instrumenttion nd piping works, etc. The indirect costs include engineering consulting costs nd their contingency llownces. The costs of the mchines nd the equipment were estimted by two methods in this study. In the first method, the equipment costs were obtined from vendor quottions directly, whenever this is possible, prticulrly for specil processing equipment used only in this industry, such s moleculr sieve etc. The second method is to obtin the equipment costs by reviewing from their historicl prices nd then djust them to the present prices. This method is pplied only when the first method is not vilble. We found tht the totl equipment costs were bout million bht (vlues in yer 2005). For the indirect costs, they were estimted in terms of certin verge percentges of the totl direct cost of the plnt, s recommended by Grrett D.E. (1989) [5]. In this study, we found tht the totl cpitl investment is 1,208 million bht, for project life of 20 yers. Figure 3 summrizes ll costs of these importnt components in the cpitl investment. 5.1 The Cpitl Investment Cost per Liter of Ethnol (Cy) The contribution of the first cpitl investment cost per liter of the ethnol is clculted by mortiztion, recovered over the life time of the plnt. A common pproch is to nnulize the cpitl cost by the eqution [6]: -l t i(r+i)" -l_l A : Pl (21 l(l+i)" where A : nnul pyments (bht per yer) P : present worth of the first investment cost (bht) i : nnul interest rte inoh n : project life in yers The nnulized cpitl investment cost is divided by the nnul mounts of the ethnol production of the plnt to estimte the cpitl cost per liter of the ethnol. It ws found tht the cpitl investment cost per liter of the ethnol in this study is 2.13 bht, if the nnul interest rte is 6%o, for the project life of 20 yers. 6, Operting Cost (Conr) As discussed erlier, the operting expenses include wter, enzymes, chemicls, opertions of utilities, repir nd mintennce, lbor, dministrtion, insurnce, nd miscellneous expenses. In this nlysis, the costs of wter, enzymes, chemicls nd the opertions of utilities were clculted nd djusted to the prices in the bse yer The dministrtion expenses were ssumed s fixed percentge of the totl lbor cost. For operting nd mintennce, nd plnt's insurnce cost, ech of them could lso be ssumed s fixed percentge of the nnulized cpitl cost [7]. The operting cost per one liter of the ethnol cn be clculted by dividing the totl nnul operting costs by the nnul cpcity of the ethnol production. The detils of the clcultion re shown in Tble Al in the ppendix. It should be noticed tht while we use strchy crop s rw mteril, the totl operting cost for ech type offeedstocks is not very much different. Nevertheless, when the sugrcne nd the molsses re used s the rw mterils, their operting costs re substntilly lower thn other strchy crops. By nture, sugrcne nd molsses generlly contin sugr which cn directly be fermented by yests nd need no ny pretretment of the feedstocks for the fermenttion. The unit operting cost of the ethnol from cssv, com, rice, sugrcne nd molsses re found to be nd 1.88 bht per liter, respectively, in this study (see the detils in item I to item IV of Tble Al in the ppendix). 7. Gins of By-products (Cs) As discussed erlier, there re some byproducts tht cn be vilble from the production processes nd wstes. These byproducts re considered to be of benefit to investors nd they cn reduce cost burdens of the production. In this study, CO2 nd fusel oil were ssumed to be wstes nd they hve no vlues. DDG is by-product of the ethnol production from corn nd rice. For cssv nd 73

6 Thmmst Int. J. Sc. Tech., Vol. 12' No. l, Jnury-Mrch 2007 Purchsed Equipment o U PiPtng Engineering Consulttion Equipment lnstlltion Environmentl Mngement I I1 38 (e 2%) 1'71.6 (t4.'7%) (13.7%) Totl - 1,208 million bht Instrumenttion Buildings Contingency Lnd l 00 r Million Bht Figure 2 The investment cost of n ethnol plnt, with cpcity of 150,000 liters per dy sugrcne, their by-products re cssv cke nd bgsse. They cn be sold for niml feeds, nd for fuel burning, respectively. Similrly, the biogs from wste wter tretment cn be used s fuel for heting. Thus, it cn lso be considered s benefit s well. The vlue of the by-product gins is 1.09 bht per liter of the ethnol from cssv, 1'27 bht per liter ofthe ethnol from corn, 1.17 bht per liter of the ethnol from rice, 1.22 bht per liter of the ethnol from sugrcne nd 0.44 bhr per liter of the ethnol from molsses, respectively. All by-products mentioned bove re generlly considered to be dditionl incomes (or gins) to ethnol producers (see the detils of the clcultion in item V of Tble Al in the ppendix). The benefits from ll by-products were subtrcted from the operting costs. Finlly, the net operting costs per liter of the ethnol, from cssv, corn, rice, sugrcne nd molsses s feedstocks, were found to be 2'58, , 0.78 nd 1.44 bht per liter of ethnol, respectively. 8. Summry of the totl cost per liter of the ethnol production All costs of the ethnol production s mentioned in section 4 to section 7 hve been summrized, s shown in Figure 4 nd Tble 3. As shown in Figure 3, it is clerly obvious tht the costs of rw mterils re the highest proportion to the totl unit cost of the ethno production. From the sttisticl pst records of the verged prices of the rw mterils, during yer , it ws found tht the feedstock costs were 53oh for cssv, 70o/o for rice,72o/o for corn, 650/o, for sugrcne nd '79%o for molsses, respectively, s percentge of the totl unit cost (per liter) of the ethnol production, The net operting costs were found to be 14.1"/o for cssv, 13.4o/o for rice, 14.87o for corn, 8.2% for sugrcne, nd 7 '2% for molsses, s percentge of the totl unit cosl (per liter) of the ethnol production, respectively. It ws found tht the proportion of the investment cost contributed to be 16.5%o for cssv, l7.4oh for rice, 18.3% for corn, 28.8%o for sugrcne nd 13.7o/o for molsses' s percentge ofthe totl unit cost (per liter) ofthe ethnol production, respectively. Since the unit cost structure (per liter) of the ethnol production, s shown in Figure 3, is clculted bsed on volumetric bsis, the ethnol's energy intensity is different from tht of gsoline. The heting vlue of ethnol is generlly less thn tht of gsoline for the sme mount of volume. A liter of ethnol contins only 20.5 MJ, while liter of gsoline contins 32.1 MJ. In order to compre the prices of both on the sme energy content bsis, the compred volume of the ethnol must be converted to be equivlent to the sme vlue of the energy intensity per one liter of gsoline. This cn be done by multiplying by fctor of 1'57 (32.1M1 divided by 20.5MJ), the cost per liter of the ethnol, for the cost comprison between the ethnol nd the gsoline. 1A

7 Thmmst Int. J. Sc. Tech.. Vol. 12. No. l. Jnurv-Mrch 2007 Tble 3 Cost structures of the unit costs of the ethnol production for the selective crops nd molsses. Feedstock Feedstock Cost (Bht/l) Net Operting Cost (Bht/l) Investment Cost(BhVl) Totl Ethmol price per liter (Bht) tthnol pnces per liter of gsoline equivlent (Bht) Cssv Corn Rice Sugrcne Molsses Mx Min Ave Mx Min Ave Mx Min Ave Mx Min Ave Mx Min Ave t s 0.78 t.44 l.lj 2.13 z.tj I ll l I 8.1s z).t3 zb-jb t z-) Mx I Min Averge I Mx Min Averge Mx Min lavergel Mx Min Averge I Mx Min Averge Cssv Corn Ri@ Sugr ne trfeedstock Cost tr Net Operting Cost trlnvestment Cost Molsses Figure 3 cost structures of the ethnol production from the selective crops The price comprison between gsoline octne 95 nd ethnol, s per liter ofgsoline's energy equivlent, for ech of the selective crops re shown in Tble 3 nd Figure Discussion nd conclusions Presently, the Thi government hs n intention to substitute for the imported crude oil by indigenous resources of bio-fuel. The primry purpose of the Thi government is to promote ethnol s substitute fuel for gsoline octne 95. Hence, the price competitiveness between them is the most importnt fctor tht the government is concerned bout. Ethnol cn compete with gsoline, without ny intervention, if its totl production cost is lower thn the gsoline's price. The pst sttisticl price records ofgsoline octne 95, during the yers [8], were used s the reference for the comprisons in this study. The verge price of gsoline octne 95 75

8 Thmmst Int. J. Sc. Tech., Vol. 12, No. 1, Jnury-Mrch 2007 CASSAVA CORN RICE SUGARCANE i I MOLASSESi I I I tr >20 Mx. gsolinc price j- zo.r: tr o," { bio Mx. Min. Cssv Averge Mx. t\,4in. Pflce Corn Averge Nlx. l\rin. Rice Averge Mx. Min. Sugrcne Averge Mx. Min. Molsses Figure 4 Comprisons between mximum, minimum nd verge prices of the ethnol per litjr of gsolini equivlent nd gsoline octne 95 prices (t ex-refinery, excluding ll txes), during yers (t ex-refinery plnts) in the pst, s well s the mximum nd the minimum price of it were determined s shown in Figure 4. All prices of them did not include ny txes (the excise txes, the oil fund nd other levied tx were pproximtely bout 8 to l0 bht per liter of gsoline). By comprison between the ethnol's (s per liter of gsoline equivlent) nd the gsoline's prices in Figure 4, we cn conclude their competitiveness s follows: () lt is obvious tht com, rice nd molsses, cnnot compete with gsoline. We cn see tht even when their feedstocks prices re lowest, (i.e. their production cost for the ethnol re minimum) while the gsoline price is t the mximum, none of them re cheper thn the gsoline's price. For the best competitive condition of rice, i.e. when the rice's price is t the minimum, it is ble to be competitive with gsoline only when the ex-reftnery price of the refined gsoline 95 should be, t lest, higher thn bht per liter (crude oil price round 58 USD per brrel). For the best competitive condition of com nd molsses, i.e. their feedstock prices re t the minimum, the ex refinery prices of the refined gsoline 95 must be even higher thn nd bht per liter (crude oil price round 65 nd 72 USD per brrel), respectively' Molsses is the most unlikely feedstock to be competitive with gsoline mong ll types of feedstocks in this study. This is becuse the price of molsses is very high. This rises from the fct tht it is lso used for other purposes. The most serious competitiveness in uses of it is to produce potble lcohol drinks in Thilnd, which cn generlly be sold t much higher price thn the fuel ethnol. As result, under sifution when the supply of molsses is very tight, beverge producers cn lwys fford to offer much higher prices to procure molsses thn bio-ethnol fuel producers. Hence, producers in Thilnd, who solely use molsses s the feedstock nd do not own ny equity in sugr production plnts, would be unlikely to survive in the bio-fuel business for long time. From interntionl experience' ethnol producers in Brzil cn be highly competitive with gsoline becuse they integrte both of their fuel ethnol production nd the sugr production processes together in order to 16

9 Thmmst Int. J. Sc. Tech., Vol. 12, No. I. Jnury-Mrch 2007 minimize their totl production costs, which is one of the key importnt fctors to mke them successful in this business [9]. (b) Sugrcne nd cssv in Thilnd re the only two types of feedstocks in this study, which re likely to hve high potentil to be competitive with gsoline in the mrketplce. However their competitiveness could be possible only when they re under specil lvorble conditions s follows: It is clerly evident tht, in the long term, s determined from the verge fuel prices from the pst historicl records in this sfudy, the verge prices of ethnol produced from sugrcne nd cssv were not yet ble to compete with the verge gsoline price during the period from 2002 to It should be remrked here tht, from 2002 to 2005, the gsoline price in the world mrket ws on n upwrd trend. This implies tht when the gsoline price is in slump, the ethnol produced from sugrcne nd cssv re unlikely to be competitive with gsoline in the long term. From our study, they could be competitive with gsoline only when the verge gsoline price ws over nd bht per liter, respectively. The most fvorble sitution, i.e. when the feed stock prices of sugrcne nd cssv were lowest (440 nd 920 bht per ton, or 41.8 oh nd 55 % ofthe totl ethnol cost, respectively). The gsoline t ex-refinery price should not be cheper thn nd bht per liter, respectively. Otherwise, the ethnol produced from them would be unlikely to be competitive with gsoline. From Figure 4, it is cler tht the production cost of ethnol, produced directly from sugr juice, is the lowest one, s comprec to ll types of the feed stocks in this study. According to the pst records, even when price of feedstock of the sugrcne becme highest, the ethnol produced from it ws still ble to be competitive with gsoline, provided tht the gsoline price ws not lower thn bht per liter. (c) Consequently, bsed on the bove rguments, sugrcne nd cssv re likely the most competitive csh crops, which hve the highest potentil feed stocks for commercilized ethnol production. At present, when the exrefinery price of gsoline 95 is bht per liter, while the crude oil price is bout 56 USD per brrel (price on July, 16'", 2006) [10], no subsidy or intervention from the govemment is needed for ethnol production promotion. However, under the sitution when the gsoline price is either too low or the feedstock price is too high, the bio-ethnol would not be ble to compete with gsoline. If the govemment wnts to promote long term biofuel policy, then the government should help to support the competitiveness of ethnol, by using txtion or other finncil instruments, such s n increse in the excise tx of the gsoline, when the gsoline price is too low, or subsidize the price of the bio-ethnol when its feedstock price is too high, in order to keep it to be competitive with gsoline ll the time. 10. Policy Recommendtion for the Thi Government from This Study () As discussed erlier, the supply of griculturl crops re generlly vilble sesonlly. In order to ensure tht the ethnol producers re ble to hve feed stocks tht re sufficient ll yer-round for production, the government should promote nd encourge investors to invest in ethnol production plnts, by using multi-type feedstocks, rther thn to promote ethnol production plnts, using only single type of rw mteril. This policy hs more dvntges, becuse the plnts cn use ny type of bundnt feedstock crop in ny seson. Thus, it would help to stbilize the production cost of ethnol. (b) Currently, the Thi government regultes the ethnol price by ttempting to relte it to the gsoline price in the world mrket nd tries to set policy to impose ceiling price on the ethnol supply in the locl mrket for fuel blending with gsohol 95, El0. Ethnol producers in Thilnd generlly oppose this policy becuse the imposed prices re frequently too low in their points of view. In order to justify whether this policy is pproprite on not, this study nlyzed the reltionship between the feed stock prices of sugrcne nd crude oil prices in the world mrket (Dubi price) during the yers [ l]. Similrly, the price of cssv ws lso correlted with the crude oil prices in the world mrket during the sme period. We found tht the coreltion coefficients between them re very poor, only 0.34 for sugrcne nd 0.28 for cssv, respectively, s shown in Figure 5. 11

10 Thmmst The results of our study re consistent with the study of the World Bnk under the Energy Sector Mngement Assistnce Progrmme [9]. It reported tht the correltion coefficient between the sugr nd the gsoline prices in the world mrket during yer 1982 to 2005 is only These results re strong evidence to show tht the feedstock prices of feedstocks for bioethnol production nd crude prices re hrdly correlted t ll. Consequently, we conclude tht the current policy of the Thi government to limit the ceiling price of the locl ethnol supply to the crude oil price, would not be pproprite. Our strong rgument is tht the lrgest portion of the totl ethnol production costs hevily depends on feedstock prices, which is generlly highly voltile nd re subject to the demnd nd supply of foodstuffs in the world mrkets, nd the sesonl locl supply vritions. On the other hnd, the vrition of the gsoline prices depends on the crude oil prices in the world mrket, which is unrelted. Consequently, the bove policy would jeoprdize the promotion of bio-ethnol production for substitution of fossil fuel in the long term. We suggest tht, insted of limit of the ceiling prices of the ethnol supply, the government should liberte ethnol prices in the mrket, nd let them flot freely, s long s the ethnol production cost is still ble to compete with the gsoline price. However, in order to ssure tht there will be no disruption of the supply of ethnol in the long run, only under very specil circumstnce, such s when the feedstock prices of sugrcne nd cssv re extremely low, the govemment should subsidize the cost of the feedstocks. This is to mke sure tht frmers hve incentive to keep on growing the fuel crops for uses in bio-fuel production without disruption. In contrst, when the feedstock prices re too expensive, the price of ethnol from the locl supply cnnot be competitive with imported ethnol from bord, then the government should intervene. The import txes should be incresed to help the locl ethnol mnufcturers to be competitive in the mrket nd mke sure tht there would be no disruption of the ethnol supply from the locl mrket in the lone term. I D 4 3 q o U Jn 2000-Jn 2001-Jn 2002-Jn 2003-Jn 2004-Jn 2005-Jn Figure 5 Comprisons of sugrcne, cssv nd crude oil prices (exchnge rte is 40 bht per USD) 78

11 Thmmst Int. J. Sc. Tech., Vol. 12, No. l, Jnury-Mrch 200i. 11. Acknowledgements S.Y. is indebt to the Joint Grdute School of Energy nd Environment for grnting scholrship nd providing reserch funds in this study. We specil thnk to Dr. Teerpt Srinorkrutr, of Thilnd Institute of Scientific nd Technologicl Reserch (TISTR), for giving us vluble dt nd informtion of TISIR's ethnol production pilot plnt for nlysis in this study. 12. References [] Koide, S., Brooks, R. B..nd Vichrngsn T' Regionl Study on Production of Fuel Ethnol from Agro-Producl, Economic nd Socil Commission for Asi nd the Pcific, p.23,1982. [2] Goswmi D Y., Alterntive Energy in Agriculture vol. II, CRC press, Inc., Florid, USA, p. 108,2000. t3] Srinorkutr T., Privte Communiction. Thilnd Institute of Scientific nd Technologicl Reserch, Bngkok, Thilnd,2002. l4l The Office of Agriculturl Economics, Monthly Report of Mjor Csh Crops prices, Bngkok,Thilnd,Avilbleonline:http ://w ww.doe. go. thlprice/monthly, [5] Gnett, D. E., Chemicl Engineering Economics, Vn Nostrnd Riendhold, New York, USA, p.29, [6] Blnk, L. T. And Trquin, A. J., Engineering Economy, 3rd edition, McCrw-Hill, New York, USA, p.28, [7] McAloon, A. et l, Determining the Cost of Producing Ethnol from Corn Strch nd Lignocellulosic Feedstocks, Colordo, USA, p.33,2000. t8l Energy nd Policy Plnning Office, Thilnd, of Petroleum Products: Exr ej i n e ry, Av ilbleonline : http ://www. eppo. g o.th/info/t1 5.html, t9l M Kojim nd T. Johnson, Energy Sector Mngement Asistnce Progrmme under Wordbnk, Potentil for Biofuels for Trnsport in Developing Countries, Wshington D.C., USA, [0] Energy nd Policy Plnning Office, Retil of Petroleum Products in Bngkok, Thilnd,Avilbleonline : go.thlinfo/t I 5.html, [11] Energy Informtion Administrtion, USA, Asi Dubi Fteh Spot, Avilble online:

12 Thmmst lnt. J. Sc. Tech., Vol. 12, No' 1, Jnury-Mrch 2007 Soco--3\ EIE =i; z 47. _::"-----d, q o.n F. iq d - - r?9::1 : --Y'1 ; v1 =e?gkig! ESSEFqF9 E3i tr?> 5 7E 3 - o c - o 5 F :.) d. r - e c! El e 9 - i7i i = d +? n re 9f ; ;-;. r = 9! : ---.-l I-. g?853 ===--t='.i Y, oi ;:-;l d- -?3;:;:t5 > t ) ) c r - h q i E!! = 3.i e l 3:::!5'5. l d - o A - i l (),9) C) () 0) q (-) xt l o zl -c t-ll 9 ll 6) <l o () f - EIE c 9 > i;i EZE tr 9> i 7E! : = x = X X - i, ;:=;55; r ++sd.ir {q:ee 6 e _ -?. - i.! _ r -roqr--n.tq.l =:::! s'5 f, - o - o o o ( - = = - o o x+ri^ir 3.2. t..,.itr,2'. q * i Es?qi{9E esrffgle EIE e? > i77 =Es.gR3tE esilfsle S 9 e E 9 - i;+ c9 -i..1 E ts + 6 c c d r, = - j o J c c c o c d r - C,l bo ( tr 9: i7.1! 5.E;.E 4 ;. 6 9 r C d r *-q:!;,q c' r c r 6 - s ==?{gb! 2 e e5*ffefe tr 9i i71 ' t.? 9! -t-c-to\., d d 3 - O 6 - IJ l - ' = 8r. 6 d Z= = -^ l- i.-.'>ee -.3: ll Et _:i? :; " l9 :=;:= - % c = = 1 * ^ 1 =?. -?;;=::!:r I ii::eo l i- 'J-iZ E : C r c) ().7 C) (d F 2i=<t3 A F < = E r E :.=.zii-= i EE *tiiz 523<d.4.<i F A U= EE=3,E ;> <> '; 5 c : v... { : ^. t _ v - E t3 2zv7E =+!EEE7Y.E iie=zhei: i ;5> E F E E q JF,,:LeII ;' I t- L P IE II It &z 80