^V#VC\ fB 6Z n8 55") property of E PA-420-S-81 -100 UBSoSt,BRAr FUEL EMISS'0NS 2000 traverwood drive 'Wffi%^i^OftN^LTERMIIVE_TRANSPORTATION._FUEL by: Richard Rykowski, Dwight Atkinson, Daniel Heiser, John McGuckin, David Fletcher, Jeff Alson , and Murray Rosenfeld Emission Control Technology Division U.S. Environmental Protection Agency 2565 Plymouth Road Ann Arbor, MI 48105 ABSTRACT Over the remaining years of this century synthetic fuels will play a key role in the nation's drive for energy independence. Although self-reliance is indeed a desirable goal, many people believe it cannot be achieved without significant compromises m environmental quality. This may not be the case. One synfuel, methanol, could be used to replace both gasoline and diesel fuel and yield environmental benefits. This paper compares methanol with synthetic fuels from other coal liquefaction processes m terms of the environmental and economic consequences of their use. INTRODUCTION Several factors must be addre of an alternative motor fuel. These categories, environmental and eco would include the production, distri fuel in question. In the reDort address these issues for several alt cially methanol, which could be methanol from indirect liquefaction, nologies were examined: the Mobil liquefaction process, and the Exxon Solvent Refined Coal (SRC-II) direct ssed when considering the viability can broadly be grouped into two nomic. Each of these categories bution, and in-use aspects of the that follows, we have attempted to ernative automotive fuels, espe- oroduced from coal. In addition to fuels from the following tech- Methancl to Gasoline (MTG) indirect Donor Solvent (EDS), H-Coal, and liquefaction processes. Of the subjects examined below, the environmental analyses of production and distribution are the most general since the least amount of information was available in these areas. Although more detail is provided in other sections, the preliminary nature of the entire report should be emphasized. More work is needed before final conclusions can ue stated with confidence. ------- -2- ENVIRONMENTAL It shou Id f irst be r ecog nized tha 3 lver se eleme nts a nd co mpou nd s in addition such as org anic nitr o-co mpound s, organi trace metaIs, such as 1 ead, a rsen ie, etc. other f uels of f e rs a num ber of opportuni reach the env ironm ent i n ha rmf ul ways, reg conve rsion pr ocess used • PRODUCTION AND DISTRIBUTION coal itself contains many to hydrogen and carbon, and inorganic sulfur, and ie conversion of coal to .es for these pollutants to rdless of the particular One potential advantage of processes which gasify coal, such as those leading to methanol or gasoline (via methanol), is that the gasification itself places most of the potentially harmful elements and compounds into forms which can be removed relatively easily. For example, minerals and heavy metals are removed from the gasifier as slag which cools to a solid. While the high concentration of metals, etc. requires careful disposal, this disposal may not be as difficult as that connected with coal liquefaction. With direct coal liquefac- tion, these compounds are entrained in the heavy organic liguid and must be separated from the liquid phase later in the process. This solid-liquid separation is very difficult (basic research is still underway in this area[1]) and the separation from a solid cannot be made as completely as the separation from a gas. Inevitably, some liquid will end up with the solid waste and some heavy metals will be left in the crude fuels. Thus, not only may the solid waste disposal problem be worsened by the addition of complex, polycyclic organic material to the waste, but the fuel itself still contains more minerals and heavy metals. One factor which may mitigate or eliminate this problem for most direct liquefaction processes is the high probability that most of the heaviest liquid fraction will be gasified to produce hydrogen.[2,3] If this is done, most of the minerals and heavy metals can be removed from the gas fairly early, since tnis heavy liquid fraction should contain most of the coal's impurities. Thus, the full extent of this disadvantage may depend primarily on the fraction of the impurities which can be removed via gasification and the fraction which must be removed directly from the liquid itself. Another potential advantage of gasification over direct liquefac- tion is the fact that all of the organic nitrogen and sulfur is broken down to simple compounds like ammonia and hydrogen sulfide. These are relatively easy to separate from the carbon monoxide and hydrogen which make up the major part of the synthesis gas. Also, since the carbon monoxide and hydrogen must be essentially free of nitrogen and sulfur before reacting over the catalyst to form methanol, there is an economic incentive to remove these two elements. Although the nitro- gen which is not removed prior to the catalyst will be removed by the ------- -3- catalyst itself, slowly deactivating it, any unremoved sulfur would rapidly deactivate the catalyst. Coal liquefaction, on the other hand, inherently leaves most of the sulfur and nitrogen in the liquid phase, bound with the organics. The most effective technique to remove these compounds is hydrogena- tion, which also is used to upgrade the fuel. However, hydrogenation is expensive, because of the large amounts of hydrogen consumed, and will likely ba limited to only the degree that is necessary to market the fuel. [4] If the fuel is upgraded to gasoline or high quality No. 2 fuel oil, most of the sulfur and nitrogen will be removed and there should not be any significant problems. However, that portion of the synthetic crude which may be burned with little or no refinement could contain relatively high levels of these elements and represents more of an environmental hazard than gasification products. The remaining distinct difference between the environmental effects of coal gasification and coal liquefaction processes (prior to end-use) is in exposure to the fuel itself, after production and in distribution. While coal liquids are for the most part hydrocarbons and, as such, are similar to petroleum, they are more aromatic and contain significant quantities of polycyclic and heterocyclic organic compounds. Some of these compounds are definitely mutagenic in bio- assays and many have produced tumors in animals. Thus, while the non- carcinogenic health effects of these materials would be more similar to those of crude petroleum, they would definitely have the potential to be more carcinogenic. There is also some evidence that much of this bioactivity can be removed by moderate to severe levels of hydro- genation which would occur if high grade products were produced. Thus, again the potental hazard is dependent upon the degree of hydro- genation given the products. Indirect liquefaction products, on the other hand, do not appear to exhibit mutagenicity or carcinogenicity. Methanol is neither muta- genic nor carcinogenic and early tests run on M-gasoline have shown it to be nonmutagenic, similar to petroleum-derived gasoline. Therefore, either of these two products offers some degree of benefit over direct liquefaction products. It is possible, however, that methanol pro- duced from coal may contain impurities and that such impurities may affect exhaust products when used. Research needs to be done in this area, also. Methanol, of course, is highly toxic in heavy exposures, leading to blindness or death. Much of its notoriety in this area is due to people confusing it with ethanol and drinking it in large quantities. Hydrocarbon fuels, while also toxic, do not suffer from this confusion and are not often taken internally. With proper education of the public, confusion between methanol and ethanol should be minimized. However, more work is still needed in this area also. ------- -4- The final point which deserves mention here is the difference between the effect of an oil spill and a methanol spill. The effects of oil spills are well known; oil films stretching for miles, ruined beaches, surface fires, etc. The effects of a methanol spill are expected to be quite different, primarily because methanol is soluble in water. While high levels of methanol are toxic to fish and fauna, a methanol spill would quickly disperse to nontoxic concentrations and, particularly in water, leave little trace of its presence after- ward. [5] Sea life should be able to migrate back quickly and plant life should begin to grow back quickly, though complete renewal would take the time necessary for new plants to grow back. Also, if a methanol fire does start, it can be effectively dispersed with water, which is not possible with an oil fire. However, methanol flames can be invisible, making them more difficult to avoid. The various relative environmental aspects of synthetic fuels production and use mentioned above are those which appear to stand out at this time. More work, however, is still needed in most areas. Although natural gas to methanol plants exist and have led to much experience in handling methanol, questions related to methanol produc- tion from coal are not known with absolute certainty since such large scale facilities do not currently exist. Similarly, no real life experience of the effects of the production of synthetic crudes exists, nor of their use. Given these caveats and the need for fur- ther research, however, the indirect liquefaction route to yield methanol or gasoline (from methanol) appears to have some potential environmental advantages over direct liquefaction processes. VEHICLE USE The dat a pr metha nol engi nes • taken using en gi anol; fully opt im impro vements in f u presented below were obtained from tests of actual However, it should be noted that these data were es which were only roughly converted to use of meth- zed engines would be expected to show further The worst problem concerning methanol's actual use centers around its low vapor pressure and high heat of vaporization. These proper- ties make it difficult to start a neat methanol engine in cold wea- ther. [6] Also, methanol has a very low cetane number of approximately 3, which means that it is very difficult to ignite in a compres- sion-ignition engine (e.g., a diesel). Problems associated with materials compatibility and lubrication also exist, but these problems already appear to be solvable with existing technology, requiring only that the auto designer know that methanol is going to be the engine f uel. Various techniques are already being tested which will improve the cold-starting capability of gasoline engines operating on meth- ------- anol, such as better mechanical fuel atomization, electrical fuel pre- heating, and the blending of volatile, low boiling point components into the methanol. Methanol's ignition problems are more serious in diesel engines, but several possible solutions are being investigated, such as glow plugs and spark ignition. Brazil already has an experi- mental methanol-fueled diesel running on the road which uses rela- tively inexpensive glow plugs as ignition aids and H.A.N, in Germany has designed a diesel bus engine with spark ignition which runs on me thanol.[7,7a1 As will be seen later in the section on fuel consumption in the economics section, the fuel properties of methanol which lead to these difficulties also lead to many advantages, such as increased thermal efficiency relative to gasoline engines. Past experience with both gasoline and diesel engines has shown that the disad vantages of a fuel can usually be overcome to allow exploitation of the advantages, particularly when the advantages are as large as they appear to be for methanol. Methanol engines promise improved emission characteristics over gasoline and diesel engines in a number of areas. Especially impor- tant are low emissions of nitrogen oxides (NOx) and an absence of emissions of particulate matter, heavy organics and sulfur-bearing compounds. One possible side benefit of methanol use could be that precious metal catalysts might not be needed for emissions control. Because methanol fuel will contain no sulfur, phosphorus, lead, or other metals, base metal catalysts (e.g., nickel, copper, etc.) may suffice. One likely negative impact of methanol engines would be an increase in engine-out aldehyde emissions, particularly formaldehyde. Catalytic converters, however, would be expected to reduce aldehyde emissions to acceptable levels. The available data supporting these effects are discussed below. A search of the literature shows a general consensus that meth- anol engines produce approximately one-half of the NOx emissions of gasoline engines at similar operating conditions, with individual studies showing reductions between 30 percent and 65 per- cent. [8,9,10, 11,12] One of the major engine design changes expected with methanol engines is the use of higher compression ratios to increase engine efficiency. Experiments have confirmed the theoreti- cal expectation that these higher compression ratios, with no other design changes, will increase NOx emissions considerably due to the higher combustion temperatures.[13,14] However, with high compression ratios, less spark timing advance is needed. Retarding spark timing is known to reduce both NOx emissions and engine efficiency. Fortuna- tely, it has been shown that the combination of a much larger compres- sion ratio with a few degrees of spark timing retard can both increase thermal efficiency and decrease NOx emissions.[14] This raises the possibility of methanol vehicles being able to meet the current 1.0 gram per mile NOx emission standard without the need for a NOx reduc- tion catalyst. ------- -5- Use of methanol in a diesel engine should also reduce NOx emis- sions by the same degree as that described above. Diesel engines have higher peak combustion temperatures and the effect of a cooler-burning fuel should actually be even more apparent in a diesel than in a gaso- line engine. Unfortunately, no data to confirm this is yet available from a diesel engine running on pure methanol. However, emission tests have been performed on a dual-fuel diesel, where a small amount of diesel fuel is injected to initiate combustion of the methanol. These tests have shown NOx emission reductions as high as 50 per- cent. [15,16] These lower NOx emissions would aid many areas of the country in attaining the ambient standard for NO in the future. (Most areas 2 are currently under compliance with the NO ambient air quality 2 standard, but many are projected to exceed it in the future as NOx emissions continue to rise.) Lower NOx emissions would also help alleviate the acid rain problem, though the majority of this problem appears to be due to stationary source emissions. Finally, the use of methanol would also provide a method for heavy-duty engines to reduce NOx emissions closer to the congressionally-mandated level without giving up any of the fuel economy advantage of diesels, as will be seen later . The lack of hard data on diesels operating on pure methanol indi- cated above will also be evident below as other aspects of meth- anol-fueled diesel engines are discussed. The basic reason for this lack of data is that until recently methanol has not been seriously considered to be an acceptable fuel for a diesel engine because of its very low cetane number. For many years, studies examining methanol as an engine fuel concentrated on gasoline-type engines (fuel inducted with combustion air). However, as the more recent studies are indi- cating, it appears possible to burn methanol in a diesel accompanied with some kind of ignition assist and, therefore, utilize the effi- ciency of the diesel concept. In addition to the positive effect on NOx emissions, use of meth- anol engines should provide even greater benefits with respect to emissions of particulate matter and heavy organics from diesels. Gaso- line engines operated on unleaded fuel emit only small quantities of particulate matter, composed primarily of sulfate particles. Thus, any improvement in particulate emissions from switching to methanol from gasoline would be small. However, diesel engines emit large quantities of particulate mat- ter consisting of solid carbonaceous particles (soot) and liquid aerosols. The former are generally formed when the injected fuel droplets are incompletely combusted, leaving carbon particles. These solid particles can then serve as nuclei for more harmful organic species to adsorb onto and as "vehicles" for such compounds to reach (and possibly lodge in) the deep regions of the lung. Although reduc- tions in diesel engine particulate have been reported, particulate ------- -7- matter seems to be an inherent pollutant in diesel-fueled compression ignition engines. Methanol, on the other hand, is a "light" fuel relative to diesel fuel and should produce far less carbonaceous particles, as do other hydrocarbon fuels "lighter" than diesel fuel. In addition, since methanol does not contain inorganic materials like sulfur or lead, there should not be any other types of solid particulate formed. Accordingly, with pure methanol there would be no nuclei for liquid aerosols to adsorb onto and total particulate emissions would be expected to be near zero.[17] This is certain to be the case with a well designed methanol-fueled spark-ignition engine.[18] Unfortuna- tely, however, we know of no studies which have measured particulate from compression ignition engines burning neat methanol. Several studies (all of which used a small amount of diesel pilot fuel) have reported much lower smoke levels, both in single-cylinder tests and in a 6-cylinder, turbocharged , direct-injected engine.[7,15,19] There seems to be little question, however, that neat methanol combustion in compression ignition engines would result in very low (and possibly zero) particulate emissions. This would result in a very important environmental advantage compared to diesel fuel combustion. As mentioned earlier, formaldehyde emissions from methanol engines are of some concern since formaldehyde is carcinogenic. For- maldehyde is an intermediate specie in methanol oxidation and would be expected to be emitted from methanol engines in greater quantities than either diesel or gasoline engines. Many studies have shown total aldehyde emissions (mostly formaldehyde) from methanol engines to be two to ten times greater than aldehyde emissions from gasoline engines. [20,21,22,23] At the same time, catalytic converters have been shown to be effective in removing approximately 90 percent of exhaust alde- hydes .[ 9 , 1 0 , 2 3 , 24 ] Much research has been performed regarding the parameters which influence aldehyde formation in gasoline engines, with low exhaust temperatures and high oxygen concentrations identi- fied as leading to higher formaldehyde formation rates, and this know- ledge should facilitate aldehyde control in future engine designs.[22,25] Aldehyde emissions from methanol combustion in diesel engines are also expected to be greater than from diesel fuel combus- tion . The last benefit of methanol engines to be discussed concerns sulfur emissions. Because of the way methanol is produced it contains essentially no sulfur. And, if there is no sulfur in the fuel, no emissions of suIfur-bearing compounds, such as sulfur dioxide, sul- furic acid, or hydrogen sulfide, can occur. This is a slight improve- ment over gasoline emissions, since gasoline does have a small amount of sulfur in it. Catalyst-equipped gasoline engines currently emit between 0.005 and 0.03 grams per mile of sulfate and this would dis- appear with the use of methanol, even if catalysts were still used. ------- -8 The improvement over the di nounced. Diesel fuel curre nt lY con weight. This translates i nto a bo sulfur fro m diesel trucks ( 0. 5 gr ams 0.75 gram s per mile of s ul fate , e one-fifth this amount. Sin ce th e expected to rise in the future, th in the future. With the use of meth appear altogether. esel, however, would be more pro- tains 0.2-0.5 percent sulfur by ut 0.25 grams per mile of elemental per mile of sulfur dioxide, or guivalent). Diesel cars emit about sulfur level in diesel fuel is ese emission levels would also rise anol these emissions would dis- ENVIRONMENTAL SUMMARY Although coal contains many substances which could be environmen- tally damaging, it appears that indirect liquefaction processes, meth- anol and Mobil MTG, can facilitate their removal easier than is pos- sible through direct liquefaction routes such as EDS, SRC-II and H-Coal. Further, since indirect liquefaction necessitates the removal of all sulfur before the fuel is synthesized, the use of relatively cheap base metal catalysts (as opposed to noble metals currently in use) on automobiles is a possibility. Neither methanol nor Mobil M-gasoline appear to exhibit mutageni- city or carcinogenicity. It should be remembered, however, that com- mercial coal-to-methanol plants are not yet available so the influence of possible impurities is not yet known. Direct coal liquefaction products are more aromatic and contain significant quantities of poly- cyclic and heterocyclic organic compounds, some of which are muta- genic. There is some evidence, however, that much of this bioactivity can be removed by moderate to severe levels of hydrogenation. More work needs to be done in these areas before definitive conclusions can be reached . The effects of a methanol spill are expected to be quite dif- ferent from that of the classical oil spill since methanol is soluble in water. Although high levels of methanol are toxic to fish, a meth- anol spill should quicKly disperse to nontoxic levels. Methanol engines promise emission benefits over both gasoline and diesel engines. Lower emissions of nitrogen oxides, and the virtual absence of particulate matter, heavy organics and sulfur bearing com- pounds from vehicle exhaust are promising. A possible detriment of methanol engines is that they emit higher amounts of aldehydes, prin- cipally formaldehyde which is carcinogenic. Catalytic converters, however, have been shown to be effective in removing 90 percent of exhaust aldehydes. As was the case with the environmental conse- quences of synfuel production, more work needs to be done in the vehi- cle-use area as well. ------- _g _ ECONOMIC We have analyzed a large number of studies in order to estimate the costs associated with the production and use of synthetic fuels. A superficial review of their conclusions quickly revealed a wide variety of conclusions and recommendations. One reason for this is that the economic bases used by the various studies often differ, affecting costs by as much as 100 percent. Another reason is that each study uses the best information available at the time of the study. Since the product mixes, efficiencies and costs of many of these processes, especially the direct liquefaction processes, change frequently as more is understood about the process, studies performed even 2 or 3 years ago cannot be compared to the latest studies. Thus, we have attempted to go back in each instance to the ori- ginal engineering studies to assess the viability of the cost esti- mates. We also have compared the available designs of each process to ascertain which are out-dated or based on now inaccurate assumptions. After doing this, the projects were placed on the same economic basis and adjusted for plant size. While the difficulties and apparent discrepancies described above primarily involve the costs of producing synthetic fuels, the overall economic picture involves more. The entire process of producing syn- thetic fuels and using them in motor vehicles will be broken down into three areas. The first area consists of the production of a usable liquid fuel from raw materials. The second area consists of distri- bution of this fuel. Finally, the third area includes the use of these fuels in motor vehicles. All costs will be presented in 1981 dol- lars. It should be noted that the general approach followed in this section is from a long-term perspective. That is, we have not identi- fied any detailed costs associated with the implementation of methanol as a "new" transportation fuel. PRODUCTION COSTS Determining the economics of the production of usable synthetic liquid fuels is probably the most difficult of the three areas to be examined. The engineering and financial bases that have been chosen are shown in Tables 1 and 2. As shown in Table 1, two different sets of financial parameters were chosen. These were selected from a sur- vey of recent studies [ 26 ,27 , 28 ,29, 30,31] done on coal liquefaction processes and represent two extreme cases for capital charge. The low capital charge rate and accompanying parameters were chosen from the ESCOE report[26] while the high capital charge data were taken from the Chevron study. [28] The important factors yielding these two CCRs are also shown in Table 1. Table 2 shows the remaining input factors. All plants were nor- malized to 50,000 fuel oil equivalent barrels per calendar day ------- -10- (FOEB/CD)(one FOB equals 5.9 mBtu, higher heating value). The costs selected for bituminous/ subbituminous and lignite coals are respec- tively $27.50, $17.00, and $10.00 per ton. Because capital costs do not usually vary in direct proportion to plant size, a scaling factor (an exponent) is normally used to modify the ratio of plant sizes (by yield). The scaling factor used here was 0.75, which is an average of factors found from various studies.[29,31, 32,33] To adjust labor and supervision costs a scaling factor of 0.2 was used.[ 26 ,32] The rest of the operating costs were assumed to vary directly with plant size. The inflation rate for adjusting the costs of studies to $1981 was based on the Chemical Engineering plant cost index. The product mix expected from each of the various synfuel pro- cesses being investigated can be found in Table 3. In order to put the discussion on costs into a more meaningful perspective, several points should be kept in mind. First, indirect liguefaction processes can yield a product mix which is either essentially 100 percent trans- portation fuel or a 50-50 mix of transportation fuel and SNG. The latter appears to be more efficient and economical for either methanol or MTG-gasoline production, but the cost of producing essentially 100 percent transportation fuel will be used here since the nation's energy shortfalls are primarily in the transportation area. Second, the product mix from direct liquefaction processes depends largely on the degree of refining applied. Each of the direct liquefaction pro- cedures yields some SNG or LPG which can be sold without further pro- cessing, while the remainder of the products in most cases must be refined before marketing. This refining adds to the product's cost. Third, the mixes reported in Table 3 were taken from available refin- ing reports. The SRC-II study was based on maximizing gasoline pro- duction while the EDS and H-Coal studies also considered No. 2 fuel oil production. Fourth, none of the synfuel processes being examined produce residual oil or diesel fuel. Residual oil could of course be obtained by the direct liquefaction routes simply by applying less refining. However, products from direct liquefaction plants appear to be too high in aromatics to allow economical production of diesel fuel. Turning once again to Table 3, it can be seen that capital costs range from $2.04 billion to $3.3 billion. The methanol plants tend to have the lowest capital costs ($2.0-2.5 billion), while that of the EDS process is in the same range near the high end. Using the incre- mental cost of the MTG process, a gasoline-from-coal plant would cost between $2.6 billion and $3.1 billion. The H-Coal and SRC-II pro- cesses are next at $3.3 billion. (The capital costs do not include refinery costs since it is unlikely that new refineries would be bulit. ) A product value approach was utilized to estimate costs for indi- vidual products. This technique assumes that the future prices of particular fuels will maintain a certain relationship, based on rela- tive demand. All prices are normalized relative to a reference pro- ------- -11- duct, which here was chosen to be gasoline. In this report, a rela- tionship between various fuels similar to that reported in the ICF report was used and-is as follows: 1. If the cost of unleaded regular gasoline is $G/mBtu, 2. The cost of No. 2 fuel oil is (0 .82)(G)/mBtu, and 3. The cost of LPG is (0.77)(G)/mBtu.[29] Since unleaded premium gasoline is produced in some cases (EDS and H-Coal), a relationship between this fuel and regular gasoline is also necessary. Unfortunately, a history of the relationship between these two fuels was not readily available. The cost ratio of leaded premium to leaded regular gasoline was used instead. This relation- ship indicated a cost ratio of 1.075.[34] This product cost relation- ship was then applied to premium and regular unleaded gasoline. The cost for SNG was assumed to be (0.8)(G). This value was obtained by averaging those for No. 2 fuel oil and LPG since SNG should share markets with each, especially No. 2 fuel oil. The product costs, along with capital costs discussed earlier, are shown in Table 3. As can be seen, they follow a similar pattern as capital costs, though not exactly. Speaking first of the low cost scenario, methanol is the cheapest product, ranging from $5.25-$6.97 per million Btu (mBtu) for fully commercial gasifiers and $5.90-$6.16 per mBtu for the near commercial Texaco gasifier. Gasoline via the Mobil MTG process would be $1.72 per mBtu more, or $6.97-$7.69 per mBtu using fully commercial gasifiers and $7.62-$7.84 per mBtu with the Texaco gasifier. H-Coal gasoline costs slightly more at $8.41 per mBtu, while SRC-II gasoline is projected to cost $9.87 per mBtu. Finally, EDS gasoline is projected to cost the most of the automotive products at $10.11 per mBtu. A similar order holds for the higher cost scenario. In this case, SRC-II has replaced EDS as the process yielding the highest cost product. This is primarily due to the higher capital costs involved for SRC-II. It should also be noted that the absolute difference between methanol costs and the cost of gasoline from the other pro- cesses increases because the capital cost of the methanol plant is lower. The same is true for MTG gasoline in most cases. A large change occurs in the difference between EDS and H-Coal process costs. While the EDS costs were 20 percent higher using the low CCR, they are less than 3 percent higher using the high CCR. Using all the studies which are publicly available, it would generally appear that the indirect coal liquefaction processes can produce usable fuel cheaper than the direct liquefaction technologies. ------- DISTRIBUTION COSTS -12- Sxnce distribution systems already exist for gasoline, the econo- mics in this area would, of course, favor the continued use of this fuel over the introduction of methanol. In addition, gasoline also has the advantage of possessing a higher energy density: 115,400 Btu/gal for gasoline compared with 56,560 Btu/gal for methanol. Thus, because transportation costs depend primarily on volume, gasoline would neces- sarily be less expensive to transport per Btu. The costs of distributing a fuel can most easily be divided into three areas; 1) distribution from refinery or plantgate (if no refin- ing is required) to the regional distributor, 2) distribution from the regional distributor to the retailer, and 3) distribution by the retailer (i.e., the gas station). These three aspects of distribution will be discussed below. More detail could of course be added to this analysis to improve the resulting estimates but such information has not yet been assi- milated. However, the general conclusions reached below should not change substantially. To simplify the presentation here, long-range distribution is approximated by that of pipeline transport to a distance of roughly 650 miles. [29] It should be noted that if pipelines are needed to connect coal fields (where synfuel plants are likely to be located) with major markets, then the total costs will be roughly the same whether the plant produces methanol or synthetic gasoline. This is evident since the pipeline must be built in either case and the con- struction and operating costs increase only slightly with a doubling in size. Further, right-of-way and engineering costs should not change at all with capacity in this range. In the case of distributing methanol, the total amount of energy distributed would only be about 80 percent that of gasoline due to vehicle efficiency improvements which will be discussed later. How- ever, a gallon of methanol only contains half the energy contained in a gallon of gasoline, so 60 percent more volume of methanol would need to be transported than that of gasoline. To determine the potential range of the cost of transporting methanol, two bracketing assumptions can be made. One, the cost of transport per volume of fuel can be assumed to remain constant. Two, total distribution costs can be assumed to remain constant. With the first assumption, the estimated cost for gasoline transportation is $0.22 per mBtu.[29] Methanol transportation would cost twice this amount or $0.44 per mBtu. Using the second assumption, where total costs remain constant, the cost for methanol would be SO.27 per mBtu, since only 80 percent as much energy is being transported. Thus, the cost of long-range distribution of methanol is $0.27-0.44 per mBtu. ------- -13- The costs involved with a switch to methanol will be more related to the increase in volumetric capacity than differences in chemical properties. Pipelines and pumps are almost entirely made from steel or brass, with which methanol is compatible. Rubber seals on pumps may need to be replaced with more durable rubber compounds, but this should be a minor cost. As mentioned earlier, the next sists of storing fuel at the regio to the retailers. This distribution and is estimated to cost just ov $0.46 per mBtu. If one conservative ume remains constant, the $0.46 translate into a $0.92 per mBtu cost step of local distribution con- nal distributor and transporting it is primarily done by tanker truck, er $0.05 per gallon of gasoline, or ly assumes that the cost per vol- per mBtu cost for gasoline would for methanol. Here the cost of conversion to even negligible. The only change and hoses, if they were not already with methanol. methanol should be very small, required should be new rubber seals made from a material compatible The costs of retailing fuel (the last step) are more like that of long-range distribution than local distribution. The costs of retail- ing are primarily fixed costs, such as land or rent. Retailing dif- fers from both long-range and local distribution, however, in that fuel energy is the critical marketing factor, not volume. Typical retailer mark-ups are estimated to be in the range of SO.05-0.18 per gallon of gasoline.[35] However, since the lower mark-ups are usually associated with the high-volume stations, the average mark-up per gallon of gasoline sold in the U.S. should be somewhere between $0.09-0.11, or $0.76-0.95 per mBtu. For methanol, the cost would lie between this range and 25 percent more since the total amount of energy distributed would be 20 percent less. Thus, the cost of retailing methanol would be $0.76-1.19 per mBtu. In deriving these retail costs, for any additional costs the re first introduced. For example, he w allowance for the initial small vol some instances will also incur costs tanks if the existing ones are large demands for the specific fuels retailing costs should therefore after the methanol market stabilizes no attempt was made to account tailer would bear when methanol is ill have to make some monetary ume of customers. The retailers in associated with installing new incompatible or unavailable due to they contain. The abovementioned be considered as long-terra costs. The total cost of distributing methanol and gasoline can now be calculated by simply combining the costs presented m the last three sections. Nethanol would cost $1.95-2.55 per mBtu to distribute; gasoline would cost $1.44-1.53 per mBtu. Gasoline has a significant advantage over methanol in terms of percentage (26-36 percent lower), but the absolute difference is only $0.51-0.92 per mBtu. ------- -14- IN-USE COSTS In order to determine in-use costs associated with methanol, it is necessary to know its fuel efficiency characteristics. There is general agreement among researchers that methanol is a more energy efficient vehicle fuel than gasoline. There are at least two theore- tical reasons why this is so. One, methanol's lower flame temperature reduces the amount of heat transfer from the combustion chamber to the vehicle coolant system. Two, its high heat of vaporization acts as an internal coolant and reduces the mixture temperature during the com- pression stroke. These characteristics are realized in experiments without having to make any major design changes in current gasoline engines. Studies have shown these inherent properties of methanol to increase the energy efficiency of a passenger vehicle by 3 to 10 per- cent with a middle range of about 5 percent.[9 ,12,13] Other properties of neat methanol combustion allow even greater efficiency improvements. Its wider flammability limits and higher flame speeds relative to gasoline allow methanol to be combusted at leaner conditions while still providing good engine performance. This lean burning capability decreases the peak flame temperature even further and allows more complete combustion, improving energy effi- ciency. Early testing on a single-cylinder engine yielded estimated energy efficiency improvements of 10 percent due to leaning of the methanol mixture as compared to gasoline tests.[36] Subsequent vehi- cle testing has shown relative efficiency improvements of lean meth- anol combustion of 6 to 14 percent.[8,9] Given these results, it would appear that methanol 's lean burning capability yields approxi- mately a 10 percent efficiency improvement over and above the 3-10 percent improvement mentioned above. Of course, stratified charge engines have been developed to allow leaner combustion of gasoline as well, and this efficiency advantage of methanol would be lessened with respect to a stratified charge engine. Methanol's higher octane number also allows the usage of higher compression ratios with correspondingly higher thermal efficiencies. Early single-cylinder testing have estimated the thermal energy effi- ciency improvements of the higher compression ratios to be in the range of 16 to 20 percent.[14,36] Unfortunately, little vehicle data exist to confirm these figures, but it must be expected that improve- ments of at least 10 to 15 percent are likely. Adding up the possible improvements indicates that methanol engines may well be 25 to 30 percent more energy efficient than their gasoline counterparts when the methanol engine is designed specifi- cally for methanol. However, since such methanol engines are not available for mass distribution today, this section will use a more conservative fuel efficiency advantage for methanol engines over their gasoline counter- ------- -15- parts of 20 percent. Using a fuel economy of 30 miles per gallon for the average gasoline-fueled vehicle, this average vehicle would require about 0.0038 mBtu per mile to operate. A methanol-fueled vehicle would be expected to use at least 20 percent less energy or about 0.0030 mBtu per mile. Using 12,000 miles per year and the average delivered fuel costs, calculated by combining production and distribution costs, the annual fuel savings relative to gasoline produced via indirect liquefaction (Mobil MTG process) were determined (see Table 4). These savings include two separate effects. One, they include the effect of dif- ferences in at-the-pump fuel costs. Two, they also include the effect of methanol engines being more fuel efficient than gasoline engines. For consistency, all fuels were assumed to be derived from bituminous coal. Following this procedure and using the lowest fuel cost (based on the low CCR) and the highest fuel cost (based on a 30 percent CCR), methanol would produce a savings of $131-240 per year compared to gasoline from the Mobil MTG process. Direct liquefaction gasoline would cost an extra $36-410 per year over MTG gasoline, because of its potentially higher at-the-pump cost. To this fuel savings must be added any difference in engine or vehicle cost. While a methanol-fueled diesel engine may be developed with a fuel efficiency advantage comparable to that of a standard diesel, the conservative 20 percent efficiency advantage over the gasoline engine should be attainable with engines similar to the gaso- line engine in terms of both design and cost. While a larger fuel tank and a special cold start system may increase costs, savings should be attained with respect to emission control, particularly if NOx reduction catalysts are no longer needed and if base metal oxida- tion catalysts can be used instead of platinum and paladium. Thus, whether a methanol engine will cost more or less than a gasoline engine in the long run is still an open question at this time. It would be rather safe to project, however, that any potential extra cost would not override the kind of fuel efficiency benefit described earlier. ECONOMICS SUMMARY The results of the past three sections are shown in Table 4. As can be seen when the results are combined , methanol compares favorably to the other fuels. With respect to synthetic gasoline, methanol appears to cost less at the plant gate. This is true whether the low CCR is used or the high CCR. Higher distribution costs lower the dif- ference, but even after distribution, methanol appears to still hold some advantage. This advantage is $1.21- $2.25 per mBtu over MTG gasoline and $2.00-$6.41 per mBtu over direct liquefaction gasoline. For vehicles driven 12,000 miles per year and achieving 30 miles per ------- -16- gallon (gasoline), methanol would save $131-$240 per year over MTG gasoline and $167-$429 per year over direct liquefaction gasoline if allowances are made for the increased efficiency of methanol engines. Without including the improved engine efficiency, annual savings would be $55-5103 relative to MTG gasoline and $91-292 over direct liquefac- tion gasoline. It should b e stated th at no cosipar ISO and diesel fuel si nee none of the coal c onv produces diese 1 fuel of sufficient qu engines. All of t hese econom ic results a re qualifications wh ich have been stated P being that the d et ail of the engineering des across processe s, and that cost estimat es development for di fferent syn fuels. n was made between methanol ersion processes examined ality for today's diesel of course subject to the reviously; the primary ones igns could not be compared reflect different points of CONCLUDING STATE ME NT Looking back over the topics addressed in this paper, it can be concluded that at this point m time methanol appears to have environ- mental and economic advantages over other synthetic transportation fuels derived from coal. The ultimate viability of this conclusion depends on a number of key events or findings. One, a cost-competi- tive methanol engine must be able to meet the driveability needs of most of the U.S. (e.g., cold-starting in nearly all climates). Two, aldehyde emissions must be controllable at low cost. Three, no other unique and uncontrollable environmental problems of methanol use or production are discovered. Four, the production and distribution cost comparisons made here must hold up against future scrutiny. The probability of these events occurring can only be estimated by a review of the support for each presented m this study. At this time, we believe the evidence available suggests that the benefits of methanol outweigh its costs. ------- -17- REFERENCES 1. O'Leary, J.R. and G.C. Rappe, "Scale-Up of an SRC Deashing Process," Chemical Engineering Progress, Vol. 77, No. 5, May 1981, pp. 67-72. 2. "EDS Coal Liquefaction Process Development, Phase V, EDS Commercial Plant Study Design Update/Illinois Coal," FE-2893- 61, Ma rch 1981. 3. Schmid, B.K. and D.H. Jackson, "The SRC-II Process," (Pittsburg and Midway Coal Mining) presented at discussion meeting on New Coal Chemistry, Organized by the Royal Society, London, England, May 21-22, 1980. 4. "Catalogue of Synthetic Fuels Projects in the U.S.," Energy Policy Division, U.S. EPA, April 1981. 5. D'Elisen, Prof. P.N., "Biological Effects of Methanol Spills into Marine, Estuarine, and Freshwater Habitats," Presented at the International Symposium on Alcohol Fuel Technology, Methanol and Ethanol, Wolfsburg, FRG, November 21-23, 1977, CONF - 771 175. 6. "Methanol Fuels in Automobiles — Experiences at Volks- wagenwerk AG and Conclusions for Europe," Dr. Ing. W. Bernhardt, Volkswagenwerk AG, Wolfsburg, Germany. 7. "B-39, Use of Glow-Plugs in Order to Obtain Multifuel Capability of Diesel Engines," Instituto Maua de Tecnologia, Fourth International Symposium on Alcohol Fuels Technology, October 5-8, 1980. 7a. "Results of MAN-FM Diesel Engines Operating on Straight Alcohol Fuels," A. Neitz, MAN AG, Nuremberg, Germany, at Fourth International Symposium on Alcohol Fuels Technology, October 5-8, 1980. 8. "Methanol as a Motor Fuel or a Gasoline Blending Compo- nent," J.C. Ingamells and R.H. Lindquist, SAE 750123. 9. "Vehicle Evaluation of Neat Methanol - Compromises Among Exhaust Emissions, Fuel Economy and Driveability," Norman D. Brinkman, Energy Research, Vol. 3, 243-274, 1979. 10. "The Influence of Engine Parameters on the Aldehyde Emis- sions of a Methanol Operated Four-Stroke Otto Cycle Engine," Franz F. Pischmger and Klaus Kramer, Paper 11-25, Third International Sympo- sium on Alcohol Fuels Technology, May 29-31, 1979, published by DOE in April 1980. ------- -18- 11. "Research and Development - Alcohol Fuel Usage in Auto- mobiles," University of Santa Clara, DOE Automotive Technology Development Contractor Coordination Meeting, November 13, 1980. 12. "A Motor Vehicle Powerplant for Ethanol and Methanol Opera- tion," H. Menrad, Paper 11-26, Third International Symposium on Alcohol Fuels Technology, May 29-31, 1979, published by DOE in April 1980. 13. "Development of a Pure Methanol Fuel Car," Holger Menrad, Wenpo Lee, and Winfried Bernhardt, SAE 770790. 14. "Effect of Compression Ratio on Exhaust Emissions and Per- formance of a Methanol-Fueled Single-Cylinder Engine," Norman D. Brinkman, SAE 770791. 15. "A New day of Direct Injection of Methanol in a Diesel Engine," Franz F. Pischinger and Cornelis Havenith, Paper 11-28, Third International Symposium on Alcohol Fuels Technology, May 29-31, 1979, published by DOE in April 1980. 16. "Alternative Diesel Engine Fuels: An Experimental Investi- gation of Methanol, Ethanol, Methane, and Ammonia m a D.I. Diesel Engine with Pilot Injection," Klaus Bro and Peter Sunn Pedersen, SAE 770794. 17. "Alcohols in Diesel Engines - A Review," Henry Adelman, SAE790956. 18. "The Utilization of Alcohol in Light-Duty Diesel Engines," Ricardo Consulting Engineers, Ltd., for EPA, May 28, 1981, EPA-460/3- 8 1-010. 19. "The Utilization of Different Fuels in a Diesel Engine with Two Separate Injection Systems," P.S. Berg, E. Holmer, and B.I. Bertilsson, Paper 11-29, Third Symposium on Alcohol Fuels Technology, May 29-31, 1979, published by DOE in April 1980. 20. Hilden, David L. and Fred B. Parks, " A single-Cylinder Engine Study of Methanol Fuel-Emphasis on Organic Emissions," SAE 760378 . 21. "Driving Cycle Economy, Emissions, and Photochemical Reac- tivity Using Alcohol Fuels and Gasoline," Richard Bechtold and J. Barrett Pullman, SAE 800260. 22. Browning, L. H. and R. K. Pefley, "An Analytical Study of Aldehyde Formation During the Exhaust Smoke of a MethanolFueled SI Engine," Paper B-62, Fourth International Symposium on Alcohol Fuels Technology, Oct. 5-8, 1980. ------- -19- 23. Baisley, W.H. and C.F. Edwards,"Emission and Wear Charac- teristics of an Alcohol Fueled Fleet Using Feedback Carburetion and Three-Way Catalysts/ B-61, Fourth International Symposium on Alcohol Fuels Technology, Oct. 5-8, 1980. 24. "Alcohol Engine Emissions - Emphasis on Unregulated Com- pounds," M. Matsuno et al., Paper 111-64, Third International Sympo- sium on Alcohol Fuels Technology, Kay 29-31, 1979, published by DOE in April 1980. 25. "Methanol and Formaldehyde Kinetics in the Exhaust System of a Methanol Fueled Spark Ignition Engine," Kenichito and Toshiaki Yaro, Paper B-65, Fourth International Symposium on Alcohol Fuels Technology, Oct 5-8, 1980. 26. K.A. Rogers, R.F. Hill, "Coal Conversion Comparison," ESCOE, DOE FE-2468-51, July 1979, pg. 59. 27. "Methanol From Coal, An Adaptation From the Past," E.E. Bailey, (Davy McKee) , presented at The Sixth Annual International Con- ference; Coal Gasification, Liquefaction and Conversion to Electri- city, University of Pittsburgh, 1979. 28. Sullivan and Frankin, "Refining and Upgrading of Synfuels from Coal and Oil Shales by Advanced Catalytic Processes," March 1980, Chevron Research Co., for DOE, FE-2315-47. 29. "Methanol from Coal: Prospects and Performance as a Fuel and a Feedstock," ICF, Inc., for the National Alcohol Fuels Commis- sion, December 1980. 30. "Economic Feasibility Study, Fuel Grade Methanol from Coal for Office of Commercialization of the Energy Research and Development Administration," McGeorge, Arthur, Dupont Company, for U.S. ERDA TID-27606. 31. "Methanol Use Options Study," (Draft) DHR, Inc. for DOE, December, 1980; Contract No. DE-ACOI-79 PE-70027. 32. Peters, Max S. and Timmerhaus, Klaus D., Plant Desig_n and Economics for Chemical Engineers, McGraw-Hill Co., 2nd Ed., 19 68. 33. Kermode, R. I., A. F. Nicholson, D. F. Holmes, and M. E. Jones, Jr., "The Potential for Methanol from Coal: Kentucky's Per- spective on Costs and Markets," Div. of Technology Assessment, Ken- tucky Center for Energy Research, Lexington, KY, March 1979. 34. Monthly Enerai Review, U.S. DOE, DOE/EIA-0035 (8 1/04), April 1981. ------- -20- 35. Ayling, John, personal communication, 2/11/81, Lundberg Survey Inc., North Hollywood, California. 36. Most, W. J., and J. P. Longwell, "Single-Cylinder Engine Evaluation of Methanol Improved Energy Economy and Reduced NOx," SAE 750119. ------- TABLE 1. Financial Parameters Capital Charge Rate, Percent Debt/Equity Ratio Discounted Cash Flow Rate of Return on In- vestment, Percent Project Life, Yrs. Construction Period, Yrs. Investment Schedule, %/Yr. Plant Start Up Ratios Debt Interest, Nominal Rate, Percent Investment Tax Credit, % Depreciation Method Tax Life, Yrs. -21- COMMON FINANCIAL PARAMETERS Low Cost Case!26] 11.5 40/60 Not Available 20 4 9/25/36/30 50, 90, 100... 10 High Cost Case[28] 30 0/100 15 20 4 10/15/25/50 50/100 Sum of Year's Digits 15 10 Sum of Year's Digits 13 ------- -22- TABLE 2. PROCESS COST INPUTS AND OTHER FACTORS COMMON TO ALL STUDIES Cost Inputs and Other Factors Product Yield Coal a) b) c) Bitumin ous Subbitu minous Lignite Operating Costs a) Utilities b) Working Capital c) Fuel Cost Interest Scaling Factors a) Capital Costs b) Labor Costs c) Maintenance, Taxes, Insurance, General d) Coal, Catalysts and Chemicals, Utilities, Fuel, Natural Gas By-Product Credit a) Sulfur b) Ammonia c) Phenol Contingency factor Inflation Rate a) 1976 b) 1977 c) 1978 d) 1979 e) 1980 Real Cost Increases a) Fuel Oil b) Natural Gas c) Coal ( % / ye a r ) Value 50,000 F0EB/CD $27.50/ton S17.00/ton $10.00/ton $0.0 3 5/kw-HR 6% of working capital per year. $35/bbl 0.75 0.20 Same percentage of plant invest- ment as specified by each individ- ual study. Amount varies directly propor- tional to plant size. $50/ton $180/ton $112 .6/bbl 15% 5% 6% 7% 9 % 9% 2% 2% 0% ------- -23- TABLE 3. PRODUCT AND CAPITAL COSTS OF SELECTED COAL LIQUEFACTION PROCESSES (1981 DOLLARS) Process Dicect Liquefaction EDS (Bituminous) Product Mix Refined Product Cost_(_$/mBtu). 11.5% 30% CCR CCR 32.7% Reg. Gasoline 10.11 16.57 14.0% Prem. Gasoline 10.87 17.81 25.6% No. 2 Fuel Oil 8.29 13.*9 9.6% LPG 7.78 12.76 18.1% SNG 8.09 13.26 Capital Cost * (Billions of Dollars) 2.50 H-Coal (Bituminous) 50.7% Reg. Gasoline 8.41 16.13 11.0% Prem. Gasoline 9.04 17.34 20.1% No. 2 Fuel Oil 6.90 13.23 18.2% LPG 6.48 12.42 3.30 SRC-II (Bituminous) 64.7% Gasoline 12.1% LPG 2 3.2% SNG 9.87 7.60 7.90 19.06 14.68 15.24 3.30 Indirect Liauefaction Texaco (Bituminous) Koppers (Bitum.) Lurgi (Subbit.) Modified Winkler (Lignite) Lurgi Mobil MTG (Subbit.) Mobil MTG Incremental Cost 100% MeOH** 100% MeOH** 47.9% MeOH** 49.7% SNG 2.4% Gasoline 100% MeO H * * 41.2% Reg. Gasoline 5 3.3% SNG 5.5% LPG 85-90% Reg. Gasoline 10-15% LPG 5.90- 9.80- 2.06 6.16 10.00 6.97 11.73 2.51 5.82 10.02 2.32 6.03 10.55 7.54 13.19 5.25 9.12 2.04 7.54 13.19 2.92 6.03 10.55 5.80 10.16 1 .72 3. 17 0.6 * Capital costs are instantaneous costs. Capital costs do not include refinery capital costs. ** MeOH = 95-98% methanol, 1-3% water, and the remainder higher alcohols. ------- -24- TABLE 4. SYNTHETIC FUEL COSTS ($ per mBtu)* Production Plantgate Cost Distribution Long-Range Local Retail Cost at Pump ANNUAL FUEL SAVINGS (RELATIVE TO GASOLINE AT $9.06-16.53 per MBtu)** $131-240 $0 S-(36-189) ADDED ENGINE COST OVER GASOLINE ENGINE 0 0 0 Indirect Coal Liquefaction Methanol Gasoline Direct Coal Liquefaction Gasoline 5.90-11.73 7.62-14.90 8.41-19.06 0 .27-0.44 0.92 0^7621^19 7 .85-14.28 0.22 0 . 46 0.76-0.95 9.06-16.53 0.22 0. 46 0i76z0:95 9 .85-20.69 * Range of plantgate cost is the lowest cost using the low CCR and the highest cost using the high CCR for bituminous feed- stocks . ** Includes effect of increased engine efficiences and dif- ferences in at-the-pump fuel costs. ------- |