Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Production Costs for Alternative Liquid Fuels Sources In this chapter the costs of several fuel production technologies are esti- mated to help compare the economic attractiveness of different feedstock, process, and fuel combinations. This chapter also considers the conversion of tar sands, oil shale, coal, natural gas, and wood into transportation fuels as well as the production of ethanol from corn and the use of compressed natural gas (CNG) in vehicles. Cost estimates for electric and hydrogen- powered vehicles are not included. Economic estimates are based only on U.S. resources, but, since there is currently so much interest in methanol, some analyses address methanol produced abroad using natural gas supplies less costly than U.S. natural gas. Using economic assumptions specified by the committee, literature-esti- mated economic parameters of the processes were initially compiled by Bernard Schulman and Frank Biasca of SPA Pacific, Inc. (Schulman and Biasca, 1989~. These estimates were further refined and updated by various committee members. The final results were used to estimate production costs, on a consistent basis across all feedstocks and technologies, based on current technical understanding. However, many of the technologies have not been commercially implemented, and there remains a high degree of uncertainty about many of the cost elements as well as the total costs. STRUCTURE OF THE ANALYSIS Cost analysis of transportation fuels manufactured from alternative en- ergy resources relies on estimates of resource costs, technological assess- ments of the specific production processes, and assumptions about the eco- nomic environment. Together these considerations form the basis for esti- mating the total cost of producing and using the alternatives. Costs are 40
PRODUCTION COSIS 41 expressed in per barrel crude oil equivalent that would make spark ignition vehicle fuel from the alternative energy resource just as expensive to the end user as gasoline from crude oil (Appendix D). Technological assessments for each process take account of the types and quantities of feedstocks utilized, the capacity and investment cost of a typical facility, nonenergy and energy operating and maintenance (O&M) costs, and the value of by-products produced. For fuels characterized by an energy density different from that of gasoline, a gasoline equivalency factor is used. Also, any additional capital costs for alternative-fueled automo- biles are included in the analysis. Economic assumptions include prices of the various energy and non- energy feedstocks, prices of energy used in plant operation, the real dis- count rate and the corresponding annual capital charge factor, costs of crude oil refining, and costs of fuel distribution and marketing. Where possible, energy costs are expressed as functions of the crude oil price. Nonenergy costs are estimated independently of oil prices, even though the committee recognizes that real (inflation-adjusted) construction costs may vary with crude oil prices and with the development rates of alternative energy facili- ties. The analysis assumes a gradual growth of the industry, not a crash program under which construction costs could rise rapidly. Two different real (inflation-adjusted) after-tax annual discount rates within a discounted cash flow (DCF) framework are considered 10 and IS per- cent with corresponding annual capital charge factors of 16 and 24 per- cent. The 10 percent discount rate is based on average historical returns required for equity capital in financial markets and on average historical returns earned by physical capital in U.S. industry. The 15 percent discount rate is based on estimates of typical hurdle rates (minimum estimated rates of return) required by corporations for investments or on costs of capital for risky projects, although hurdle rates might be even higher for particularly . . risky projects. For each technology the cost per equivalent oil barrel, in 1988 dollars, is calculated as the summary cost measure, based on the assumed oil price environment. This calculation begins with the crude oil price, here defined as the average price of crude oil imported into the United States. Prices of natural gas, electricity, and corn (as a feedstock) are calculated based on the crude oil price (see Appendix D). It is assumed that conversion facilities for coal are located on the Gulf Coast and that facilities for natural gas conversion, underground coal gasifi- cation, oil shale mining, ethanol production, and tar sands pyrolysis or extraction are located at the resource site. Natural gas compression would occur at the point of end use. Thus, natural gas used as a feedstock is priced at an estimated wellhead cost, and natural gas used in plant opera- tions or compressed into CNG is priced at a delivered cost to the industrial
42 FUELS TO DRIVE OUR FUTURE sector. Coal price includes a transportation charge to the Gulf Coast in addition to the mine-mouth price. No royalties for oil shale or tar sands are included in the costs, because (1) the magnitude of royalties would vary with the profitability of the resource and conversion technology combina- tion and (2) royalties would be driven toward zero as the overall profitabil- ity were itself driven toward zero. Per-gallon product costs are calculated by adding together feedstock costs, energy and nonenergy O&M costs, and annual capital cost and subtracting by-product credits, all on a per-gallon basis. For processes that produce a product similar to crude oil, cost per equivalent oil barrel is reported as equal to the product cost per barrel. No further calculations are required. For processes that produce gasoline directly, a gasoline substitute, or products intermediate between gasoline and crude oil, gasoline equivalent costs are first calculated. Cost per equivalent oil barrel is then estimated by adjusting gasoline equivalent cost. To calculate gasoline equivalent costs, product cost per barrel is multiplied by the gasoline equivalency factor. The gasoline equivalency factor has a value of 1.8 for methanol, reflect- ing a 10 to 18 percent efficiency gain for automobiles. A value of 1.5 was assumed for ethanol, but its octane advantage and potential for enhanced efficiency could lead to lower ratios. For CNG the analysis is conducted in terms of output per equivalent gallon of gasoline, so that the gasoline equiva- lency factor is 1.0 (although optimized CNG vehicles may have efficiencies greater than gasoline engines). For the Shell middle distillate synthesis (MDS) process, the tar sands solvent extraction process, and direct coal liquefaction, a gasoline equivalency factor of 1.0 is used, reflecting the opportunity to use the outputs from these processes in the production of gasoline. Three adjustments to the gasoline equivalent cost are made to obtain cost per equivalent oil barrel. The distribution and marketing costs of the prod- uct, net of corresponding costs for gasoline, are added to the gasoline equiva- lent cost. A refining credit, based on the historic relation between prices of gasoline and crude oil, is subtracted for those products that produce gaso- line or a direct substitute for gasoline (methanol, ethanol, and gasoline through the methanol-to-gasoline [MTG] process). A smaller spread is subtracted for those products intermediate between crude oil and gasoline. Finally, when appropriate, the additional annualized incremental costs for methanol- and CNG-fueled vehicles (above the costs of gasoline-fueled vehicles) are added. The procedure outlined above provides an estimate of the cost per equiva- lent crude oil barrel based on a given crude oil price. Two different modes of analysis are used to select the crude oil price, referred to as endogenous price determination and exogenous price determination. For exogenous price determination a particular crude oil price is assumed
PRODUCTION COSTS 43 as part of the scenario specification (see Chapter 1~. Two of these scenar- ios, assuming a $20/barrel and a $40Ibarrel crude oil price, have been the basis for complete cost calculations. Using this procedure, if the cost per equivalent crude oil barrel exceeds the crude oil price, the difference be- tween these figures can be interpreted as either the amount of subsidy or the degree of economic improvement required to make that process and fuel competitive with gasoline from crude oil. If the difference is negative, it can be interpreted as the rents or profits (above the discount rate or cost of capital) that could be obtained by using the alternative energy resources to supply fuel. For endogenous price determination, crude oil price is not an input to the analysis but is calculated as the price that would make the specific process and fuel just competitive. More precisely, the crude oil price is chosen such that the calculated cost per equivalent oil barrel is exactly equal to the crude oil price: No profits would be earned above the normal cost of capital or discount rate. Using this procedure, electricity, natural gas, and corn prices are also chosen to be consistent with the crude oil price and the cost of the specific product. The cost per equivalent crude oil barrel can then be interpreted as the crude oil price at which the fuel (based on a given resource and process) would be just competitive (without a subsidy) with gasoline from crude oil. If crude oil prices are in fact higher than the cost per equivalent barrel, rents or profits (above the discount rate or cost of capital) would be available, while for lower prices the alternative process would not be commercially viable without a subsidy. COST ESTIMATES FOR THE VARIOUS TECHNOLOGIES Based on the procedure outlined above, cost estimates have been devel- oped for each technology using domestic feedstocks (see Appendix D). This section presents the basic results. Figure 3-1 shows the costs of various alternative fuels, based on a 10 percent discount rate and endogenous determination of energy prices. Cur- rent commercial technologies include natural gas used as feedstock to pro- duce methanol (NG > Methanol), natural gas converted through a methanol- to-gasoline process (NG, MTG), compressed natural gas (Compressed NG) used directly in automobiles, and corn distilled into ethanol (Corn > Etha- nol). Technologies successfully demonstrated on a commercial scale but not yet commercialized include coal used as a methanol feedstock (Coal ~ Methanol), wood used as a methanol feedstock (Wood > Methanol), and indirect coal liquefaction in a methanol-to-gasoline process (Coal, MTG). There is still considerable uncertainty regarding the economically acces- sible biomass resource base, and research is on-going to overcome prob- lems in conversion processes and improve the economics of producing liq-
44 NG > Methanol Coal > Methanol UCG > Methanol Wood > Methanol NG, MTG Coal, MTG NG, Shell MDS Compressed NG Corn > Ethanol Oil Shale Tar Sands, Pyrolysis Tar Sands, Extraction Direct Liquefaction FUELS TO DRIVE OUR FUTURE <//////1///////1///////1///////// 1 1 ~ 1 1 -, W: ~ a ~/~/~i: ~ ~i: ~ ~ . ~ ~ ~~ ~ ~ ; :' ~ ~ ~~ ; 'W ' ~ = _ ~ ?:? . ,?~ ?: .:?:.?:::. ::.:?: :?~:?~:??:.?:.~ Hi:::.:? 62:.:.:: I:. ?:::: I?:?:?-:?:? ?. ~ ? ::. ??:?:?? ?~:?.:~ ?:~ ?~ ?~ . :: ~ //////J/~/~//J/~/J/~/~/A I 7~///1///////1///////1///////~ 1 ~//////I///////~///////l/////i ~ ~////~/1/////~/1////6 ~ No Demonstration C: Demonstrated ~ Commercial $0 $10 $20 $30 $40 $50 $60 $70 COST: CRUDE OIL EQUIVALENT ($ per barrel) FIGURE 3-1 Estimated costs of alternative fuels at 10 percent DCF with en- dogenous calculation of energy prices (see text for abbreviations and further detail). New estimated reduced capital and operating expenses for en~ained-flow coal gasi- fication could lead to coal-to-methanol costs of about $40/barrel. uid fuels. Technologies that have not been successfully demonstrated on a commercial scale include underground coal gasification producing metha- nol (UCG > Methanol), the Shell Middle Distillate Synthesis process (Shell MDS), pyrolysis of oil shale (Oil Shale), pyrolysis of tar sands (Tar Sands, Pyrolysis), solvent extraction of tar sands (Tar Sands, Extraction), and di- rect liquefaction of coal (Direct Liquefaction) (see Appendix D, Table D-9. At a 10 percent discount rate, estimated costs range from $25/barrel to above $70/barrel, with most exceeding $40/barrel. Only compressed natu- ral gas, direct coal liquefaction, and tar sands (solvent extraction) have estimated costs below $40/barrel. Methanol produced from domestic natu- ral gas or underground coal gasification, hydrocarbon fuels from oil shale conversion, and pyrolysis of tar sands have estimated costs between $40 and $50/barrel. These processes could become economically viable with significant technological advances and a high world oil price. All other technologies have oil equivalent costs exceeding $50/barrel and will be less economical unless significant cost reductions are realized. Production of methanol from coal and wood is of special interest since
PRODUCTION COSI S 45 both raw materials are potentially major domestic methanol sources. The endogenous energy prices for coal and wood were estimated to be $53/ barrel and $70/barrel, respectively, with corresponding feedstock costs of $16/barrel and $24/barrel (Table Deb. Much of the difference in cost is attributable to higher investment and operating costs related to the small assumed scale of the wood-based process. Increasing the scale of the wood- based process would narrow or eliminate the difference between nonfeed- stock costs for the two processes. Technological advances can generally be applied to plants using either feed. The synthesis costs are independent of source of the carbon monoxide-hydrogen mixture used and cost reductions would also be applicable to natural gas-based plants. Advances in the gasification section are generally applicable to both coal and wood. While advances in technology can reduce methanol costs from wood and coal materials, the substantially lower projected cost of methanol from low-cost natural gas indicates that as long as overseas natural gas prices are low, im- portation will be the lowest cost methanol source. Synthesis gas-based methanol and hydrocarbons, because of energy losses in both gasification and synthesis, are expected to remain a more expensive route to transporta- tion fuels than direct liquefaction or manufacture from high-grade oil shale. As discussed later (see Figure 3-5), the relative cost of methanol from natural gas is highly sensitive to the cost of natural gas (see Appendix D). The lower estimated cost for direct coal liquefaction versus oil shale re- flects the progress made over the past decade in reducing coal liquefaction costs and the lack of published advances in oil shale conversion technolo- g~es. Major costs incurred in producing fuel from alternative energy resources include those for capital, feedstock, and O&M (Figure 3-2~. The incre- mental cost of purchasing a methanol- or CNG-fueled vehicle, spread over the life of the vehicle, is also shown as a positive component of the oil equivalent cost. For most of the technologies, feedstock and capital costs are the two largest cost categories. By-product credits are small except for the production of ethanol from corn. Ethanol production results in large quantities of organic by-products that the present analysis assumes can be sold for one-half the value of the initial feedstock. However, the market and price at which they can be sold is highly uncertain. Laboratory results for producing ethanol from other forms of biomass such as wood, through advances in biotechnology, show promise but are not included in the pres- ent analysis. Several negative components appear in Figure 3-2. The sum of positive components minus the sum of the negative components is consistent with the cost estimates in Figure 3-1. By-product credits are shown as negative components, since they would subtract from the net cost of producing fuel.
46 NG > Methanol Coal > Methanol UCG > Methanol Wood > Methanol NG, MTG Coal, MTG NG, Shell ADS Compressed NG Corn > Ethanol Oil Shale Tar Sands, Pyrolysis Tar Sands, Extraction Direct Liquefaction FUELS; TO DRIVE OUR FU7VRE _ ~-il ~ i _- . JR~ , ,. ED ~;(,;:4 1 2 ~- ~ ~.'.'.''.."i.i~ .' ~ , ~\\\\~-.o...l I 1 IlIl.l,,,,,,,,,,'`. .. p .~ I J Auto Costs L' . ° ~ O&M ~\\\'1 Feedstock ~ -; ~ Capital 1lilililil Byproducts I.... :'1 Mar/Dis Incr -$60 -$40 -$20 $0 $20 $40 $60 $8Q $1 00 $1 20 COST: CRUDE OIL EQUIVALENT ($ per barrel) FIGURE 3-2 Components of total cost for alternative fuels at 10 percent discounted cash flow wide endogenous price calculation. (O&M, operation and maintenance costs; Mar/Dis Incr, increment associated with marketing, refining, and distribu- tion.) Two componentsone negative and one positive have been aggregated into a component covering refining, marketing, and distribution. The nega- tive component is the historical spread between refined gasoline price and the crude oil price. The positive cost component is the cost of marketing and distributing ethanol and methanol, net of the equivalent cost of market- ing and distributing gasoline. The sum of these two components is negative for most of the processes and thus appears in Figure 3-2 as negative cost components. Figure 3-3 provides cost data similar to the data in Figure 3-1 for both the 10 and 15 percent real discount rates. This diagram shows that the costs of these capital-intensive technologies are very sensitive to the cost of capi- tal. The greatest sensitivity occurs for those technologies that require the largest per-barrel investment cost (see Appendix D, Figure D-1~. Cost estimates for two crude oil prices, $20/barrel and $40/barrel, indi- cate that crude oil equivalent costs of the various technologies do depend on crude oil price but that the sensitivity is relatively small (Figure 3-4~. Two factors explain the impact of crude oil price: the cost of energy inputs and the marketing and refining credit for fuels that directly lead to gasoline
PRODUCTION COST S NG > Methanol Coal ~ Methanol UCG > Methanol Wood > Methanol NG, MTG Coal, MTG NG, Shell MDS Compressed NG Corn > Ethanol Oil Shale Tar Sands, Pyrolysis Tar Sands, Extraction Direct Liquefaction 47 l . , . ~ ~ \\~\~\~\\1\~\\\1 it\ 1\\\\\~ ~\\~\\\\1\~\\\1 it\ 1\\\\\\1\~\\\~ ~1 ~~.:~ 1 \\\\\~\\\\1\\\\\\1 \\\\ 1\\~\\1\\\\> : : i . it- JU~' \ \ \ \ \ ~ \ \ \ \ 1 \ ~ ; \ \ \ 1 \ ~ \ \ \ 1 \ \ \ \ \ \ \ \ \ \ \ \ ~ \ \ \ \ \ \ \ \ \ I .- ~ .~ W~ ~~ ~~S I.. \ \ \ \ \ \ ~ \ \ \ \ 1 \ \ ~ \ \ \ \ 1 ~ ; \ \ 1 \ \ ~ \ \ \ 1 \ \ \ \ \ ~ \ \ \ \ \ \ \ \ \ \ \ ~~ To. ~ ~ an. - a.. ~~ a\\\\ \~\1\\\\\~1 \~ ;.... an:, ....,... \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ N\\y \ ~1 1 \\\\\' ~1 .... ~~ \~\\\~\\~14~\~\1~ ~\\\\Y \~33~- 74~3~3~ R\\\\~\~\\\~\N \\\\] l ~ _ 3= 10% DCF 15% DCF $0 $10 $20 $30 $40 $50 $60 $70 $80 $90 COST: CRUDE OIL EQUIVALENT ($ per barrel) FIGURE 3-3 Role of discount rate on cost for endogenous energy prices. NG > Methanol Coal > Methanol UCG > Methanol Wood > Methanol NG, MTG Coal, MTG NG, Shell MDS Compressed NG Corn ~ Ethanol Oil Shale Tar Sands, Pyrolysis Tar Sands, Extraction Direct Liquefaction ~~ ~ ;3s ~~s,s~ s a. v. ~~ . s3 3~3 -:3 I, of,: ~,3 .~ . ~ -.,. . /////~/~/~////////////////1 \\\\\\1 \~1 \\\\\ 1 N\\~\1 1 , i, V.~ 2~ ..:.:3 . ~ = it, s s ~ s s s 3 Is T.~. ~ 9'. . . ,, 2~ 3 . s s 3 ,., .:s.~ .s s s ~ . -~,:s :: ///////////////~///////////~K,^471 \~\\\\1 \\\\\ I \\\\\ I \\\\\\\\\\\ \\\\\ _ 33::~ i~ ~ ~~ \\\\\\1 \\\\\ I \\~\\\1 \\\\\1 ~ .34 a ~ so s ~3~ .;, , . ,s, . : s .~3~3~. 3 AT,:. i: s . :. s s3~33~ -3 3 3 . ~~ _ ~ <<<<<<1<<<<<<<<<<<<<<<<<<<<<<<<~<<<<<<1<<<<<<<<<<~\\' - , /~/~/A~~/ j/~//~/// j///////~/~/~/~/) \\\\\\1 \\\\\ I \\\\\ I \\\\\ I \\\\\~\\1 1 . ~~s~3'3S7~/~ \\\\\\l \\\\\\\\\\\\~\~$'1''3\\\\~\1\\\\\\ 3 ~ :: 3 so: ~ ~ 3 3 s ~ . ~ - I: ~ - . ssss3 : : ~ :3 3 3 : so ~ 3 3~: 3 3~ . . ~ 1 1 . <<<<<<<<<<<<<<<<~<<<<<<<<<<<<<~////g 1 1 ..:~: 33~33 - 3 -3 a;: 33 a, 3333': "" 3.:~ 3''.;333' 3','': .3'.'2-~. ,.3~ 1 1 1 $<<<<<<Z<<<<<<<<<<<<A ~ I I ~ 1\\\\\\\\\N ^~\\\~\~\~\\\\\\~\~\\\\\\~ 1. 1 1 1 1 1 $'333'w''3 ~ ·'t¢~<~ 1 1 _~- ~ ~ /'^'~ / / / / / //A >^ \\\1 '''''''''''''''''~1 L7 - ,,'~ - 1 ~\\~1~\\~\ 1\~\~ 1 ~ '"" 3i"~"' 3'"~ ; sash.: a:"'. . "is a ~ ~ -,,T: ~ ,~, ~ sol Aft, ,~ ~ .s ~ Vi/i/~/r'-ffirr7~/~7i'1 1 1\~\\\\\\\\\\\\~51 N^~9 1 i i i 1 1 1 Endogenous Price $40 Oil Price $20 Oil Price $0 $10 $20 $30 $40 $50 $60 $70 $80 COST: CRUDE OIL EQUIVALENT ($ per barrel) FIGURE 3-4 Impact of crude oil price on cost.
48 NG > Methanol Coal ~ Methanol UCG > Methanol Wood > Methanol NG, MTG Coal, MTG NG, Shell MDS Compressed NG Corn > Ethanol Oil Shale Tar Sands, Pyrolysis Tar Sands, Extraction Direct Liquefaction FUELS TO DRIVE OUR FUTURE it' '~'''''"='1"'''''=''''''"'''''1 . . . At. . my. ~ ///////~/////////////~///1 ~1 /// :. I...... Base Case $2/Mcf Drop in NG Price Function $0 $10 $20 $30 $40 $50 $60 $70 COST: CRUDE OIL EQUIVALENT ($ per barrel) FIGURE 3-5 Impact of natural gas price on cost. NG Price ($/hicf) $4.89 $1 .78 $5.00 $3.00 $4.83 $2.83 $5.00 $3.00 $5.00 $2.77 $5.00 $3.00 $5.00 $2.36 $4.22 $~.57 $5.00 $3.00 $4.83 $2.67 $4.65 $2.53 $3.88 $1.88 $4.49 $2.47 or a gasoline substitute. Higher oil prices result in a greater credit. The net effect of these two factors will determine the overall oil equivalent cost. Sensitivity of costs to natural gas prices were calculated by reducing the natural gas price by $2/million British thermal units at each oil price (Fig- ure 3-5~. While this variation in the natural gas price function reduced the estimated cost for most technologies, it led to far less cost variation than did the cost of capital variations, except for technologies using natural gas as feedstock. The importance of capital and feedstock costs is indicated in Figure 3-6. Capital plus feedstock costs exceed $50 per oil equivalent barrel for all MTG processes, for ethanol from corn, for methanol from wood, and for the Shell MDS process. Two methanol-producing technologies using natural gas and coal as feedstocks have feedstock plus capital costs exceeding $40/barrel. Unless R&D advances lead to great reductions in investment costs or to great increases in feedstock conversion efficiency, such tech- nologies are unlikely to become economical. Some current research efforts suggest that capital costs for wood to methanol conversion could be reduced . . A. s~gn~cant y. Sensitivity tests for domestically produced methanol (using natural gas as a feedstock) illustrate major uncertainties about the costs of this gasoline
PRODUCTION COSIS 49 substitute. Ranges of crude oil equivalent cost estimates are based on sensitivity tests for five parameters, varied one at a time, using the en- dogenous price determination method. Results are presented in Figure 3-7 and in Table D-10 (Appendix D). The base case assumes that automobiles fueled with methanol will enjoy a 10 to 1X percent efficiency gain over gasoline-powered vehicles. This assumption translates to a gasoline equivalency factor of 1.8. Methanol vehicle efficiencies that are 15 percent greater than in the base case (bar labeled "15% More") and 15 percent smaller (bar labeled "None") are illus- trated by the pair of bars in Figure 3-7 denoted by "Automobile Efficiency Gain." The former is based on a gasoline equivalency factor of 1.57, the latter on a gasoline equivalency factor of 2.06. This range of automobile efficiency variations leads to a $14/barrel variation in crude oil equivalent cost of methanol. The investment cost of a methanol production facility is increased by 25 percent and decreased by 20 percent for the third set of sensitivity tests. This range of investment costs leads to a $5/barrel variation in the oil equivalent cost of methanol. $40 - _ $eo Q - a' Q - cn $20 oh o ~ $10 c: $0 Coal, MTG . ~ Wood ~ MeOH Underground Coal Gasification · ~ Coal ~ MeOH )|(Shale Coal Liquefaction < . ~ Tar Sands Pyrolysis ' Tar Sa,nds, Solvent Extraction Compressed NG Corn ~ EtOH NG, MrG ~ Shell MDS LONG > MeOH 1 "1 1 " 1 1 " 1 " 0 10 20 30 40 50 60 70 80 FEEDSTOCK COSTS ($ per barrel) FIGURE 3-6 Capital costs versus feedstock costs (dollars per equivalent barrel). Dotted diagonal lines show combinations of these two costs totaling to $20/barrel, $40/barrel, $60/barrel, and $80/barrel. For shale, mining costs are included in the feedstock costs, whereas for tar sands they are not.
so FUEL; TO DRIVE OUR FUTURE The real discount rate is varied from a low of 5 percent to a high of 15 percent for the fourth set of sensitivity tests. This range leads to a $9/barrel variation in the cost of methanol. Finally, the differential between the cost of a methanol-fueled automo- bile and that of an equivalent gasoline-powered vehicle is reduced to zero from the base case of $200/car and increased to $500/car. This range leads to an $8/barrel variation in the crude oil equivalent cost of methanol. Sensitivity tests for methanol production from natural gas show why methanol is more likely to be produced abroad than in the United States. These tests varied a hypothetical natural gas feedstock price from a high of $3/Mcf to a low of $1/Mcf. Three different investment cost conditions were also compared: estimated investment cost in the United States, that cost increased by 25 percent, and that cost increased by 75 percent, to evaluate possibly higher investment costs in remote foreign locations. O&M costs were increased along with investment costs. Finally, costs were added for transportation of methanol to the United States. Methanol costs for dif- ferent assumptions are shown in Figures 3-8 and 3-9. Endogenous crude oil price is assumed, but natural gas price is varied independently to assess sensitivity specifically to the price of gas, even though natural gas and oil prices are likely to be linked. The cost estimates at the far left in Figures 3-S and 3-9 are based on U.S. - $50 - Q ~ $40 is LLI Is > a AL $30 o LL ~ $20 . . o $10 $0 None $60 1 $2/Mcf Higher 15% More // ¢~ Base Automobile Natural Gas Case Efficiency Price Gain v/~L $2/Mcf Lower 15% DCF $500 20% Lower ~ t j,:~ <a/' 5% ~ $0 Investment Real Automobile Cost Discount Cost Rate Differential FIGURE 3-7 Sensitivity of costs for domestically produced methanol.
PRODUCTION COST S $50 U.S. Casts c <~, $40 Q a) Q z $30 C, $20 IIJ o ~ $10 C' o 51 U.S. ~ . ~ Investment cost up 25% _ ~ Investment cost up 75% ~ Levels 10% DCF ~ ~ 1 5°/a DCF _ ~ 1 n°/^ DCF _ _ S°/^ DC,F _ $o - I -T- I -1- $4.8 $3 $1 $3 . ~ $2 $1 $3 $2 NATU RAL GAS PRICE ($ per Mcf) 1 1 1 1 1 $1 $3 $2 $1 $3 $2 $1 FIGURE 3-8 Total cost for methanol production (in dollars per barrel at 1988 prices). See Figure 3-9 for cost components. conditions, including a natural gas price of $4.80/Mcf (this cost was deter- mined as the endogenous natural gas price). The next two estimates are also based on U.S. investment and O&M costs, but natural gas price is set at $3/Mcf and $1/Mcf, and a $4/barrel cost is added for transportation of the methanol to the United States. These figures show the sensitivity of the total cost to natural gas prices. The remaining cost estimates represent conditions that might characterize natural gas-to-methanol production in remote locations. These 12 estimates are separated into four groups of three cases. Cost changes are based on estimates developed by Bechtel, Inc. (California Fuel Methanol Study, 1989~. Two groups represent investment costs 25 percent above U.S. levels and corresponding increases in O&M costs. These values might characterize methanol production in the Middle East. Discount rates of 10 and 15 per- cent are presented for these two groups. For these two groups crude oil equivalent costs range from $26/barrel, were natural gas priced as low as $1/Mcf and the real discount rate at 10 percent, to $43/barrel for a natural gas price of $3/Mcf and discount rate of capital of 15 percent. The final two groups represent investment costs 75 percent above U.S. levels and 64 percent increases in O&M costs. These values might charac- terize methanol production in northwestern Australia. A very low discount
52 FUEL; TO DRIVE OUR FUTURE U.S. Costs ~ ~ $50- a' - ~ $40 - a' - z $30- at: $20 a LL O $10 1 ~ $0- . c' ~ LUevsis_ Investment Cost Up 25% _ _ Investment Cost Up 75% 10% DCF ~ 15% DCF __ 10% DCF .. 5% DCF k. ~ I' ::::: ,J.. ...2. ....... ~ I... ~ .. K/ // a/ a/ ~ __ :::~:::~ T:ST: i: : ::: : Ti: :: :. ~:~ ~ :., a,' :' :~:':~':~: ~~::~? -a i: ".': I've ~~: :~ ~ ~~ . it.'.. .... .. :::: .~ . . . ,~ :: ma: . : ::~ ~~ : :~ ~~: I..:: ..~. ~' :.:.: 'I ~:~,~ ~~ I,.. :: T: ::'T:~ :::::::::::: ~~ ~~'~ ~ :: :~ T ::::: :::: ?~:S~ TO ": .L '''. :~ 'I .'~"'~ ~2~"~ :~ ,~ ~~: :~ Hi: ~ '::::: ::: T~ ~ 'TiS,:: :: ::::: :: :::: : C S,:~: "2:.:.~',~ "'"' : :: OTT: T~'~ ,. ':. ,:~.~S,~'~ ~ _ . I... .... J....~.... . J.~.. : ~~ ~~: S::~T:~::T: ,,. : ~~ :'? ' ' '...,: SO :: S:::'S:E,~:~ a,': :~:~2:~' ~~.~'~:'~: ~'~'~ :~ :S,::: TS'T~ ~~: 6:::: :: :: : Hi:::::: : :: :::::: Hi: ~ ::'S:': T~ ~~: A- ~~: ~~:,:~ :"'~2~: ::: :~:: i::. a.... ~:''~ ,:,,, ..~.. by.. A:: ~ ~ TO Am.' ''''if _'s:. _ O -$10 - ~ i i ~ ~ ~ i ~ ~ ~ ~ <) $4.8 $3 $1 $3 $2 $1 $3 $2 $1 $3 $2 $1 $3 $2 $1 NATURAL GAS PRICE ($ per Mcf) E: ~ Transportn Auto Costs O&M Feedstock Capital IT i~ Mar/Dis Incr FIGURE 3-9 Components of total cost for methanol production (in dollars per barrel at 1988 prices). (The refining, marketing, and distribution increment Mar/ Dis Incr should be subracted from the sum of the other costs to obtain the crude oil equivalent costs.) rate of 5 percent is shown in addition to that of a value of 10 percent to illustrate values that might be obtained if, for example, the Australian gov- ernment promoted development through low-interest loans or loan guaran- tees. For these two groups crude oil equivalent costs range from $27/barrel, were natural gas priced as low as $1/Mcf and the cost of capital subsidized (5 percent), to $44/barrel for a natural gas price of $3/Mcf and discount rate of 10 percent. Estimates for methanol production in remote locations vary greatly from estimates for the continental United States. Investment cost, natural gas feedstock cost, and the cost of capital can each greatly influence cost esti- mates. The cost of transportation for the finished methanol is a smaller source of variation. Production at different locations is likely to be charac- terized by quite different values of the major cost determinants. U.S. production costs for methanol would be significantly higher than in foreign locations characterized by low natural gas prices and modestly in- creased capital costs. This result strongly suggests that methanol produced in the United States is unlikely to be economical compared with methanol from remote foreign locations. To the extent that methanol is used in the United States, then, it will likely come from remote foreign locations. The committee's cost estimates, based on continental U.S. conditions, should
PRODUCTION COSTS 53 therefore not be used to assess the probable costs of methanol to be used in the United States. ISSUES OF FUEL DISTRIBUTION AND USE The previous sections have presented cost estimates of producing and using transportation fuels manufactured from alternative energy resources. This section further addresses issues associated directly with the distribu- tion and end use of such fuels. Changing the fuel infrastructure and vehicle ownership patterns for alternative fuel use is a truly formidable barrier. Environmental issues are addressed in Chapter 5. On the other hand, many of the processes provide a product similar to crude oil. Such products would be refined into gasoline and other petro- leum products that would be used just as fuels refined from conventional crude oil. These products (as well as reformulated gasolines) would not require important changes to the distribution system or to automobiles and are not discussed further in what follows. End-Use Issues Alcohol fuels such as ethanol and methanol would be used in automo- biles in much the same way as conventional fuels. Although alcohols weaken and corrode some automobile materials, automotive redesign would solve any problems. Natural gas can be used directly as fuel in an engine. Because of its low liquefaction temperature, natural gas would normally be stored in the gase- ous state as CNG at about 3000 psi. Although engineering advances might be expected before large-scale introduction of CNG vehicles into the United States, the basic technology is already commercially available: Several countries are already using CNG vehicles to a limited extent. In addition to single-fuel vehicles, multifuel engines that can use any combination of gasoline, methanol, and ethanol are under development by automobile companies. These vehicles would allow the use of either fuel during a potential transition from a gasoline to a methanol fuel system. While such engines have important flexibility advantages, they could not be optimized with respect to both fuels. Therefore? they cannot be expected to take advantage of the special properties of some fuels, such as the high octane of methanol. Multifuel vehicles can also be designed to use either natural gas or gasoline. Both single-fuel and multifuel vehicles using alcohols or CNG will be more costly than vehicles using gasoline engines. Once large production volumes are obtained, it is estimated that multifuel vehicles operating on methanol and gasoline will cost at least $100 to $300 more than conven-
54 F[JEl~ TO DRIVE OUR FUTURE tional gasoline vehicles; these estimates are also a best guess for what a dedicated methanol vehicle might cost. Multifuel autos operating on CNG and gasoline will cost about $700 to $1000 more (DeLuchi et al., 1988b). Additional purchase costs of multifuel engines have been incorporated above in the cost estimates for methanol ($200) and for CNG ($1000~. The additional purchase cost was translated to an equivalent per-mile additional operating cost. This per-mile equivalent operating cost was derived so as to give the same discounted present value of costs to the consumer as would the one-time additional automobile purchase cost. (See Appendix D.) Vehicle performance could vary among the alternative fuels. Consumer acceptance of these fuels can be expected in turn to depend on vehicle performance. The road performance acceleration, hill-climbing ability, and so onof CNG- and alcohol-fueled vehicles is comparable to that of similar gasoline-fueled vehicles. For optimized engines the road perform- ance of alcohol-fueled vehicles can be expected to be slightly better than their gasoline-fueled equivalents. Methanol and CNG have lower volumetric energy densities than gaso- line. Unless methanol and CNG vehicles carry more fuel, a shorter driving range between refuelings would be inevitable. Increasing the size of the fuel tank so as to increase the range may encroach on passenger space, es- pecially in small cars. Compensating increases in automobile size would decrease fuel economy. It remains to be seen how the automobile industry would adjust to this challenge. Methanol's relatively low vapor pressure and high heat of vaporization hinder satisfactory cold start and driveway performance. Fuel vapor pres- sure determines the air-to-fuel ratio delivered to the engine. Vapor pressure decreases as its temperature decreases. Thus, the ratio of air to vaporized fuel of methanol is too high for satisfactory cold-start ignition. However, a small amount of gasoline added to methanol (around 15 percent; this is known as M85 fuel) seems to reduce the cold-starting problem at least in moderate climates without exacerbating water miscibility problems in the fuel tank. Cold starting is not a problem with CNG vehicles. The octane number of natural gas and methanol is higher than for gaso- line, permitting optimized engines for these fuels to have higher compres- sion ratios and better thermodynamic efficiency. Multifuel engines, how- ever, need to be operable on the lower-octane gasoline. These multifuel vehicles would not gain the fuel-economy benefits of these alternate fuels. In the cost analysis above, a range of fuel-economy assumptions in com- parison to gasoline-fueled vehicles was used. Increased wear has been reported in engines using methanol, but it ap- pears that this problem can be solved with appropriate reformulated motor oils. These oils may be more expensive and may require more frequent oil changes. Natural gas engines have durability equal to or greater than gaso-
PRODUCTION COSTS 55 line engines. Thus, it appears that from an engine durability standpoint there is no insurmountable problem with these alternative fuels. Fuel Distribution Issues Bulk distribution of alcohols would create additional problems, but prob- lems that could be overcome. The bulk distribution system for liquid fuels depends primarily on pipelines and secondarily on river and ocean barges from U.S. refineries, with terminals located near major centers of demand. Trucks are used to deliver fuels to retail service stations. Because present fuels are not miscible with water, the current system is not kept water-free. Water-free facilities would be required for alcohol-gasoline blends because of phase separation problems (pure alcohols would be less of a problem), and the use of methanol would require changes in materials. Because of the solvent properties of methanol, pipelines and tankage must be cleaned be- fore methanol is introduced into facilities that have been used for petroleum products. Distribution of methanol by barge, rail car, and tank truck, rather than pipeline, is quite feasible at similar or increased cost per gallon. Natural gas distribution for CNG may not be a problem because there is currently excess natural gas pipeline capacity. However, that capacity might be insufficient in some regions if natural gas became a major transportation fuel. But pipeline capacities could be expanded in adequate time. Since many homes already have gas delivered directly to them, home-fueling sta- tions for natural gas for overnight refilling are possible. However, this option would entail the high cost of a compressor and may involve safety risks. Retail distribution of alternative fuels would require start-up investments to modify the existing refueling system. For example, the cost of a refuel- ing station for CNG vehicles would be about $300,000 (Sperling and DeLu- chi, 1989~. These additional costs, to establish a CNG refueling station, comprise the CNG capital costs underlying the economic analysis presented earlier in this chapter. Methanol refueling stations may require an addi- tional cost of around $40,000, perhaps more depending on the need for a new vapor control system. Potential purchasers of CNG- or methanol-fueled vehicles would be unwilling to buy unless a refueling system were in place. And there would be little incentive to establish a refueling system unless there were custom- ers expected to purchase the fuel. Multifuel vehicles represent a natural transition response to this "chicken and egg" problem. Additional early steps toward problem solution might involve facilities dedicated for vehicle fleets, both public and private. The U.S. light-duty vehicle fleet market is estimated at 0.65 million to 1.3 million bbl/day (10 billion to 20 billion gal/ year). Although not all of the fleet market is suitable for alternative fuels,
56 FUELS TO DRIVE OUR FUTURE the market could well be large enough to provide the beginnings of a CNG or methanol refueling system. Public "filling stations" may remain rare until there are many vehicles, although fuel companies in California are now establishing methanol fuel outlets for demonstration fleets. In summary, with additional vehicle development work and demonstra- tion programs, alternative fuels could be used successfully in current trans- portation engines. With the possible exception of reduced driving range, performance should be acceptable for both CNG and M85. Broad use of either fuel involves significant issues of bulk and retail distribution. How- ever, these are cost, not technical, issues that could be solved if alcohol or CNG were economically competitive with conventional or reformulated gasoline. CONCLUSIONS The results of the preceding analysis for producing liquid fuels from alternative energy resources suggest that none of these alternatives can be expected to provide transportation liquids at as low a cost as is currently possible by refining crude oil. Even if crude oil prices were to increase by 50 percent above current levels, only a few of these options would be economically viable. Economic acceptability of the various processes would come about if there were significant increases in the world crude oil price, the U.S. energy policy environment, the technical characteristics of these processes, or a better understanding of their economics. If world oil prices were to increase above $30/barrel and were to stay at that elevated level, some processes would become economically attractive under current cost estimates. More would become attractive if oil prices were to increase above $40 to $50/barrel range and remain there for ex- tended periods. Likewise, U.S. energy policy could be altered so as to apply a large premium, say $10/barrel, to energy conservation or production activities that reduced U.S. petroleum imports. Under such an aggressive policy some processes would become viable immediately in some situations. Finally, R&D activities might decrease the production costs to below those estimated in this chapter. In particular, technological improvements might reduce investment costs or increase the conversion efficiencies of the various processes (see Figure 3-6 for the magnitudes of these two costs). Such changes could significantly reduce the overall costs such that the various technologies could become economically viable at lower world oil prices or with less aggressive energy policy stances than suggested above.
