Special Report Analysis of the Economic and Environmental Impacts of Liquefied Petroleum Gas (Propane) as a Vehicle Fuel United States Environmental Protection Agency Office of Mobile Sources April 1995 ------- U.S. ENVIRONMENTAL PROTECTION AGENCY NATIONAL VEHICLE AND FUEL EMISSIONS LABORATORY 2565 Plymouth Road, Ann Arbor, Michigan 48105 Announcement - Publication of LPG Special Report Enclosed is a copy of an Office of Mobile Sources Special Report: "Analysis of the Economic and Environmental Impacts of Liquefied Petroleum Gas (Propane) as a Vehicle Fuel." This is part of a series of reports which describe the impacts of various alternative fuels, such as methanol, ethanol, and compressed natural gas. If you would like additional copies of this report, or any of the previous special reports, please contact Mary Seminitis at (313) 668-4516 (FAX:668-4573). ------- Table of Contents Chapter 1 Introduction Nature of the Report 1 Background 1 Sources of Additional Information 3 References 4 Chapter 2 LPG-Fu»l«d Vehicle Technology 5 Introduction 5 Fuel Properties of LPG 5 Current Converted Vehicle Technology 6 Optimized OEM LPG-Fueled Vehicles 9 Safety 11 Durability 12 Fuel Economy 12 Exhaust Emissions 13 Low Temperature Emissions 20 Evaporative and Other Emissions 20 Vehicle Scenarios 21 Heavy-Duty Engines and Vehicles 22 References 24 Chapter 3 Economic Impacts 26 Introduction 26 LPG Supply 26 Fuel Prices 28 Vehicle Fuel Economy and Cost 30 Summary 32 References 33 Chapter 4 Air Quality Impact* 34 Introduction 34 Urban Ozone Levels 34 Carbon Monoxide 37 Air Toxics 37 Global Warming 40 Summary 41 References 42 Chapter 5 Conclusion* 43 Use of LPG as a Vehicle Fuel 43 Benefits 44 Need for Additional Studies 45 ------- Table of Contents (Appendices) Appendix I Data from Sears Fleet Appendix II LPG Emissions Data from Late Model Vehicles Appendix III Dual-Fuel Gasoline Emissions Data Appendix IV Low Temperature LPG Emissions Data Appendix V Heavy-Duty LPG Certification Data Appendix VI Heavy-Duty LPG Emissions - Bus Chassis Data Appendix VII Material Safety Data Sheet ------- EPA Office of Mobile Sources Special Report Chapter 1 Introduction Nature of the Report This report is part of a series of special reports on the economic and environmental impacts of alternative automotive fuels; previous reports have addressed the impacts related to methanol, compressed natural gas (CNG) and ethanol. These reports are not intended to represent the final word on these alternative fuels, but rather to present the Agency's best estimate of the extent to which each of these fuels can contribute to efforts aimed at improving our nation's urban air quality. Given the rate at which vehicle technology is advancing, the reader is cautioned that the quantitative estimates presented in these reports may become outdated after publication, though the directional trends should remain valid. This report consists of three parts: 1) a technology assessment analysis, 2) an economic impact analysis, and 3) an environmental impact analysis. The technology assessment is described in Chapter 2, and is used as a basis for the analyses of the economic and environmental impacts, which are described in Chapters 3 and 4, respectively. Each of these chapters include discussions which qualify the available data and identify the assumptions made in the analysis. The conclusions of the report are summarized and qualified in Chapter 5. Issues related to the implementation of alternative fuel programs are not considered here. Background Liquefied petroleum gas (LPG) is broadly defined as a gas mixture comprised of propane, butane, ethane, propylene and/or butylene. The fuel is liquefied to reduce its volume for transportation and storage by compressing the gas to pressures greater than 100 pounds per square inch (psi). Because of concerns about fuel volatility for heating use, LPG sold commercially is comprised mostly of propane. Thus, the term "propane" is often used interchangeably with "LPG." The term LPG is used in this report, and means propane-based LPG fuel, unless otherwise noted. The reader is cautioned that when used in other reports, especially when related to LPG supply and/or production, the term LPG is often used in the broader sense, and includes butane, ethane, propylene and/or butylene, in addition to propane. It should also be emphasized that the fact that the current commercial market for LPG requires that it be mostly propane is related to operational requirements for its use as a heating fuel, and property designed vehicles could use LPG fuels with higher concentrations of heavier hydrocarbons such as butane. However, the current market for LPG as a vehicle fuel is too small to justify the existence of a separate ------- EPA Office of MOM* Soura* Sp*c*l Report grade of LPG, since it would require segregated distribution (although it could use many of the same pipelines). Nevertheless, with a sufficient number of LPQ-fueled vehicles, it could become economically feasible to create another grade of fuel, and thus it is possible that future LPG-fueled vehicles will be designed to use LPG containing substantial amounts of heavier hydrocarbons. LPG currently is sold and distributed extensively throughout the U.S. Residential and commercial heating, which is the largest commercial usage segment, accounts for 35 percent of the U.S. market, while automotive use accounts for only 3 percent of LPG sales.[1] Much of the automotive market involves centrally-fueled fleets such as utility trucks and buses. LPG is also commonly used in forklifts, because LPG engines can be calibrated for very low CO emissions, which enables them to be used indoors. Significant numbers of LPG-fueled vehicles are also being used in Canada, Japan and Europe. This experience with LPG-fueled vehicles has demonstrated that such vehicles are quite reliable. However, since the interest in LPG has historically been motivated by economic considerations (primarily lower fuel costs), the vehicles have only recently begun to be optimized with respect to emissions. The potential for the optimization of the emissions control systems of LPG-fueled vehicles is discussed in the technology analysis in Chapter 2. ------- EPA Office of Mobito SOUICM Special Report Sources of Additional Information Given the somewhat limited scope of this report, and the likelihood that there will be unforeseen advancements for LPG-fueled vehicles after this report is published, the reader may wish to obtain additional or updated information regarding LPG-fueled vehicles. Several sources are listed below. It should be emphasized that there may be other reliable sources of information that are not listed here, and that their exclusion from this list should not be considered to be indicative of the Agency's disapproval of their merit. Gas Processors Association 6526 East 60th Street Tulsa, OK 74145 (918) 493-3872 National Propane Gas Association 4301 North Fairfax Drive Suite 340 Arlington, VA 22203 (703) 351-7500 Propane Consumers Coalition 1110 North Glebe Road Suite 610 Arlington, VA 22201 (703) 522-1324 Propane Gas Association of Canada 1800-300 Fifth Avenue S.W. Calgary, Alberta T2P 3C4 Canada (403) 263-0450 Propane Vehicle Council 901 15th Street, NW Suite 470 Washington, DC 20005 (202) 371-6262 ------- EPA Offlc* of Mobil* SOUKCM Special Ftoport References 1. "An Assessment of Propane as an Alternative Transportation Fuel in the United States,' Prepared by R.F. Webb Corporation, for the Natural Propane Gas Association, June 1989. ------- EPA Office of Mobile Sources Special Report Chapter 2 LPG-FueiecJ Vehicle Technology Introduction This chapter addresses issues related to engine performance, including fuel properties, emissions, fuel economy, and safety. These discussions will form the basis for the remainder of the report. Because LPG-fueled vehicles are still in a developmental stage in many respects, the following discussions cannot be definitive; rather they are intended to identify the range of potential in-use engine characteristics and to make reasonable projections of future fuel economy and emissions. Wherever possible, the discussions highlight significant areas of uncertainty in the analyses. This chapter, and the remainder of this report, focus on light-duty vehicles and light-duty trucks; heavy-duty engines and vehicles will be discussed only briefly at the end of this chapter. For simplicity, this chapter considers only two types of vehicles: 1) current generation gasoline-fueled vehicles that were converted to use LPG, and 2) advanced vehicles that would be designed and built in the future specifically to use only LPG. While other possible types of vehicles, such as advanced dual-fuel1 vehicles, are not specifically considered here, much of the following discussion would still be relevant. It is important to note that EPA is not predicting that the "optimized" vehicles discussed in this chapter will ultimately be produced, but rather EPA is presenting its best estimate of what could be achieved given sufficient market and regulatory incentives. Fuel Properties of LPG The properties of LPG as a heating fuel are well studied. The majority of current commercial LPG meets the requirements of ASTM specification 01835 for "Special Duty Propane" (which is essentially the same as "Propane HD-5 of Gas Processors Association Standard 2140") for vehicle use.[1] As the name implies, the specification requires that most of the LPG be propane; the maximum amount of butane and heavier compounds is 2.5 percent, and the maximum amount of propylene is 5.0 percent. Slight compositional deviations from this specification may occur with in-use "propane", but they are not considered here. As noted in the first chapter, other grades of LPG could be used as a vehicle fuel, but the focus of this report is on vehicles using "Special Duty Propane." 1 In this report, the term "dual-fuel" is used to describe vehicles that are capable of operating on two different fuels; the reader is advised, however, that the term 'dual-fuel* can have other meanings in other contexts. ------- EPA Offla <* Mobil* SourcM Special Report When compared to gasoline as a motor fuel, LPG has both advantages and disadvantages. Typical fuel properties of commercial propane are summarized in Table 2-1. The energy content of LPG is about 19,700 BTU/lb (lower heating value2), which is slightly higher than standard gasoline (18,700 BTU/lb). However, due to its lower density (approximately 4.1 Ib/gal for LPG vs. 6.2 Ib/gal for gasoline), it has a lower energy content on a volumetric basis, and thus, LPG-fueled vehicles require larger fuel capacities to obtain driving ranges equivalent to gasoline-fueled vehicles. LPG has an octane number of about* 104 ([R+M]/2), while the octane number of commercial gasoline ranges up to 94 ([R+MJ/2); thus spark ignition engines running on only LPG can use higher compression ratios than those running on gasoline. LPG also tends to bum with a lower peak flame temperature, which can lead to lower NOx. The greatest difference between LPG and gasoline involves fuel volatility. LPG is much more volatile than gasoline; thus the fuel enters the intake system and the cylinders fully vaporized (not partially vaporized, as is the case with gasoline). This difference affects the engine in several ways. First, the higher volatility contributes to better mixing with air prior to ignition. The gaseous nature of the fuel also limits its interaction with the oil used to lubricate the cylinder walls (in gasoline-fueled engines, especially during cold starting, the fuel tends to dilute the oil on the walls, which reduces the lubrication, and thereby increases engine wear). On the other hand, since the gaseous fuel occupies a greater volume than the liquid, less air can enter the engine, which reduces the power output of the engine. This effect can be offset by the ability of LPG-fueled vehicles to use higher compression ratios, as was noted above. Current Converted Vehicle Technology Most current LPG-fueled vehicles are gasoline-fueled vehicles which have been converted to run on LPG. In the past, the conversion process has generally involved the modification of a carburetted engine using a relatively simple kit. The purpose of most conversions was not related to air quality, but rather, the conversions were done to take advantage of economic benefits of LPG. From an emissions perspective, these vehicles represent a starting point for LPG-fueled vehicle technology, and, as such, are the focus of this section. A large number of the vehicles already converted allow the driver to switch fuels from LPG to gasoline and back. These dual-fuel vehicles avoid problems associated with fuel availability; however, since they must be capable of operating on either fuel, they cannot be fully optimized for LPG. 2 Energy content can be expressed as higher or lower heating values. The difference is that the lower heating value assumes that the water that is produced from the oxidation reaction is a vapor, not a liquid. 6 ------- EPA OIK* ol MoOto SOUICM Special Report Table 2-1 Fuel Properties Properties Energy Content* Density* Octane Rating" (R+M)/2 Peak Flame Temperature** Auto-ignition Temperature** Stoichiometric Air/Fuel Ratio** (by weight) Boiling Range** (1 atm.) Special Duty Propane 19,700 BTU/lb 4.1 Ib/gal 104 3600eF 855°F 15.7 -44 to +31 °F Unleaded Gasoline 18,700 BTU/lb 6.2 Ib/gal 87-94 3900-41 00°F 440-880°F 14.5-15.5 80 to 420°F EPA [2] EA Mueller [3] The conversion of a gasoline-fueled vehicle to one that is capable of operating on LPG requires a relatively small number of modifications, including the addition of an LPQ fuel system. Adding an LPG fuel system requires installing a special fuel tank and fuel lines for the pressurized fuel, and a means of controlling the flow of fuel. The pressure associated with LPG is generally 100 to 200 psi under normal storage conditions, which is not an extremely high pressure; thus, the fuel system can be relatively simple. (For comparison, tire pressures are typically 35 psi.) The addition of a normal LPG fuel system does, however, increase the vehicle weight by approximately 100 pounds. It is also necessary to add a pressure regulator and to add a gas mixer, which mixes the LPG with the air before it enters the engine. For carburetted engines, the mixer generally sits on top of the carburetor, and the existing throttle controls the amount of the fuel/air mixture that enters the engine. For fuel- injected engines, the mixer can be located upstream of the throttle. It is generally ------- EPA Office of Mobile Source* Special Report necessary to route engine coolant through the regulator system to prevent localized freezing of the fuel as it expands. For dual-fuel vehicles, it is necessary to stop the flow of one fuel into the engine when operating on the other, so solenoid valves are also needed for the fuel lines. Although it is not necessary to make further modifications, it is often worthwhile to make changes to the ignition timing of the engine to account for the higher auto- ignition temperature and octane rating of LPG. Optimum power and fuel economy for LPG-fueled engines requires more spark advance at low speeds and less at high speeds than is required for gasoline.[3] For engines which are not computer- controlled, changes to timing can be implemented using relatively simple ignition devices. For dual-fuel vehicles, such devices allow the timing to be optimized somewhat for both LPG and gasoline, but not as well as can be achieved with computer controls. However, optimizing the timing of computer-controlled engines for both LPG and gasoline is not trivial since it requires installation of a modified microprocessor. Because of this, as well as the challenges associated with fuel- injection, current LPG-fueled vehicles are disproportionately converted carburetted gasoline-fueled vehicles that did not have computer controls. More recently, higher technology conversion kits have become available that allow computer-controlled, fuel-injected gasoline-fueled vehicles to be converted to also run on LPG. Since such vehicles are today's conversions, the portion of the analysis that deals with 'current* vehicle technology will be based primarily upon these more recent technology vehicles. The reader is cautioned, therefore, that some of the discussions that follow are not generally applicable to older LPG-fueled vehicles. For the analyses contained in this report, it will be assumed that the primary goals of manufacturers of current conversion kits are to comply with EPA emissions standards for conventional vehicles,[4] and to minimize kit costs and fuel consumption. This assumption is reasonable since: 1) EPA prohibits conversions which result in vehicles not complying with the standards to which they were originally certified, 2) there is currently little incentive for manufacturers to lower emissions significantly below these levels, especially if such reductions would result in higher kit costs or increased fuel consumption, and 3) the purpose of conversion to LPG use has generally been to lower fuel costs. The effects that this assumed optimization strategy would have on the costs and performance of "current* LPG-fueled vehicles are described in later sections of this report. It should be noted that this assumption applies only to 'current converted* vehicles, and that a different strategy is assumed for "optimized OEM" vehicles (see following section). The Agency does recognize that some noncompliance may occur in use, and is most concerned about improper calibration of the air/fuel ratio (see "Exhaust Emissions") and slowly leaking fuel lines (see "Evaporative and Other Emissions") from vehicles that are converted incorrectly. 8 ------- EPA Office at Mobile Sources Special Report Optimized OEM3 LPG-Fueled Vehicles Few current LPG-fueled vehicles have been built specifically for use with LPG fuel, though several automobile manufacturers are currently involved in LPG-fueled vehicle research. This situation may change, however, if the demand for LPG-fueled vehicles increases. An increase in demand would have two benefits. First, it would make it more cost-effective (on a per-vehicle basis) for manufacturers to invest in the research and development of LPG-fueled vehicle technology. Second, it would lead to greater availability of refueling facilities, and thus, greater acceptance of dedicated LPG-fueled vehicles. This latter point is significant because many of the potential improvements are not possible for vehicles that must also be able to operate on gasoline. Therefore, it is reasonable to expect that if LPG-fueled vehicles were produced in significant numbers, the vehicles would be more advanced than the vehicles discussed above. Among the improvements that would be expected are fully optimized ignition timing and air/fuel ratio calibration, increased compression ratio, improved fuel injection, and optimization of catalyst formulations for LPG. Most gasoline-fueled vehicles being produced now have computerized controls for ignition timing and air/fuel ratio control. With "closed-loop" systems, the optimum settings are continuously determined by the computer based on the inputs received from various engine sensors, which monitor parameters such as engine temperature, the oxygen content of the exhaust, and engine speed. For each set of inputs, there is a set of engine parameters that optimize engine performance, including emissions control. Most of these systems have been further improved by the addition of adaptive learning capability. Adaptive learning allows the system to modify its control behavior, based on patterns observed from previous operation. This is very advantageous since it allows the controller to correct for deterioration and defects in the engine and emission control systems, or changes in the air or fuel. For example, systems with adaptive learning will respond to reductions in the concentration of oxygen in the exhaust that result from operation at higher altitudes, where the volumetric concentration (mass per unit volume) of oxygen in air is lower, by increasing the volumetric air/fuel ratio. Such systems are now becoming available for LPG-fueled vehicles as well. It is expected that, with the appropriate research, these systems will soon be able to perform virtually all of the functions performed by the systems currently used with gasoline-fueled vehicles, and thus will provide a means of further optimization of LPG-fueled vehicle emissions and fuel economy. s The compression ratio of an internal combustion engine greatly affects the efficiency of the engine. Higher compression ratios lead to greater efficiency, but also lead to increased combustion temperatures and thus increased NOx formation. With 3 OEM (Original Equipment Manufacturer) LPG-fueled vehicles (or engines) are vehicles (or engines) that were designed and produced specifically to use LPG fuel, as opposed to vehicles (or engines) that were originally intended to use gasoline but were modified after production to use LPG. ------- EPA Office of Mobil* Source* Special Report spark ignition engines, the upper limit on the compression ratio is determined by the octane rating of the fuel. Since LPG has a significantly higher octane rating than even "premium" gasoline, higher compression ratios can be used in dedicated LPG-fueled vehicles. Also, the lower peak flame temperature and potentially lower air/fuel intake temperature of LPG can offset the increase in temperatures that is associated with higher compression ratios, so that the net effect on NOx emissions would be small. Converted vehicles generally do not take full advantage of this benefit, since compression ratios are difficult to modify for an engine that has already been built. Similarly, compression ratios for dual-fuel vehicles are limited to about 9:1 by the octane rating of gasoline. Engines that were originally designed to use only LPG could take advantage of the octane benefits of LPG, although practical considerations, such as production cost, may limit the compression ratio of even dedicated vehicles to about 12:1. The proper delivery of fuel and air to the engine plays an important role in the operation of a vehicle. Earlier gasoline-fueled vehicles relied upon carburetors to combine the fuel and air together into a mixture which was drawn into the cylinder by the piston movement. Most engines today, however, rely upon fuel-injection to deliver the fuel to the engine, since it results in more precise control of the air/fuel ratio. Similarly, new LPG-fueled engines are also expected to rely upon fuel-injection to deliver fuel to the engine. Fuel-injection systems have been developed for current conversions to gaseous fuels, but these systems have generally not been able to provide the level of control that is provided by gasoline fuel-injection systems. However, improved fuel injection systems (both gaseous fuel-injection systems and liquid fuel-injection systems) are continually being developed for LPG-fueled vehicles,[5,6,7] and it appears very likely that these newer systems will ultimately have performance that is essentially as good as fuel-injection systems for gasoline-fueled vehicles. Virtually all current LPG-fueled vehicles use catalytic converters that were optimized to treat exhaust from gasoline-fueled vehicles. Since hydrocarbon emissions from LPG-fueled vehicles have a different composition than hydrocarbon emissions from gasoline-fueled vehicles, the efficiency of the catalytic converters on current LPG-fueled vehicles is less than optimal. Catalyst manufacturers are currently working to develop catalysts specifically for LPG-fueled vehicles, and it is expected these newer catalysts will lead to further reductions in emissions.[6,8] / It is beyond the scope of this report to list all of the potential and expected improvements for LPG-fueled vehicles. The improvements discussed above are considered to be the most important, but are certainly not the only ones. In addition to the LPG-specific improvements described above, it is expected that future optimized LPG-fueled vehicles will also incorporate many of the technological advancements that are currently being developed for other fuels. For this report it is assumed that future optimized LPG-fueled vehicles will have fully optimized computer 10 ------- EPA Offic* of Mobito SOUICM Special Report controls for the air/fuel ratio and spark timing, optimized fuel injection, and optimized catalytic converters, and that these systems will perform as well as gasoline-fueled vehicle systems. Moreover, it is assumed that they will have higher compression ratios than gasoline-fueled vehicles. It is expected that most of these vehicles will be used in fleet applications, especially in fleets that will be affected by the federal "Clean-Fuel Fleets" program, which requires the purchase of low-emitting vehicles.[9] Since fleet operators are also generally concerned with keeping fuel costs low, manufacturers are expected to design future LPG-fueled vehicles for both low fuel consumption and low emissions. In this report, it is assumed that manufacturers of future "optimized OEM" LPG-fueled vehicles will follow this strategy. Safety Some concerns have been raised regarding the safety of LPG-fueled vehicles. These concerns result from the gaseous nature of the fuel at normal atmospheric conditions, which presents some additional challenges regarding safety. Nevertheless, the current body of evidence suggests that LPG use in properly designed vehicles is as safe as gasoline.[10,11,12] It is important to emphasize that all fuels are dangerous when they are not handled properly, and in this respect, LPG is no different. The most important aspects of LPG-fueled vehicle safety are described in more detail below. There currently are no federal safety requirements for the fuel system integrity of LPG-fueled vehicles. The National Highway Traffic Safety Administration (NHTSA), in a recent Notice of Proposed Rulemaking,[13] suggested that such requirements may not be necessary since every state has already adopted some form of the National Fire Protection Association (NFPA) Standard 58, "Standard for the Storage and Handling of Liquefied Petroleum Gases."[14] The NFPA standard addresses such issues as installation of LPG systems, transport and transfer of LPG, and storage. There is currently no evidence that short term exposure to LPG, such as inhalation, causes any adverse health effects. In this respect it is unlike gasoline and diesel fuel, which are both reasonably toxic. The primary concern with respect to LPG exposure is the potential for frostbite-like bums resulting from direct contact with the liquid as it evaporates and expands.[11,12] The liquid itself is not cold, but the evaporation of the liquid and the expansion of the gas draw heat from surroundings, lowering the temperature well below 0°C. Overall, however, the risks associated with LPG exposure are small when compared to those of conventional fuels. The greatest LPG safety concern involves the potential for explosions in enclosed spaces. If a leak occurs in an LPG fuel system, the leaking fuel vaporizes as it escapes. In an open area, this is advantageous, since it results in a fairly rapid 11 ------- EPA Office of Mobile Sources Special Report dispersal of the fuel, which greatly reduces the risk of fire or explosion. In an enclosed area, however, the fuel can mix with the room air to form a potentially explosive mixture. This hazard can be avoided by maintaining proper ventilation in all areas where LPG vapors could accumulate. Proper ventilation must account for the fact that LPG vapors are heavier than air, and thus will disperse downward. For a more thorough discussion of LPG fire safety, the reader is referred to NFPA Standard 58. Durability As was noted above, lubrication of the engine cylinder during cold-starting is generally better with LPG than with gasoline. With gasoline-fueled vehicles, when the engine is cold, the fuel can remain in liquid form and dilute the oil on the cylinder walls. The dilution of the oil reduces its ability to provide proper lubrication for the piston, which leads to higher rates of engine wear, and decreased engine life. LPG is much more volatile than gasoline, and interacts very little with the oil on the cylinder walls; thus, the engine cylinders in LPG-fueied vehicles can have less deterioration in use than gasoline-fueled vehicles. This benefit is difficult to quantify, however, since it is very dependent upon driving and maintenance practices. Nevertheless, there is some evidence to suggest that, in some cases, the life of an LPG-fueled engine may be as much as twice as long as that of a comparable gasoline-fueled engine.[15] It should be noted, however, that the significance of this benefit has decreased, and will probably continue to decrease in the future, due to improvements in gasoline-fueled engine designs that reduce this type of wear. On the other hand, LPG-fueled engines without hardened valve seats can have greater valve seat wear, presumably because the fuel lacks the lubricity of liquid fuels. When the problem occurs, it can have a very negative impact on both fuel economy and emissions. However, this problem should not affect properly designed OEM vehicles, and can be prevented in conversions by installing hardened seats if they were not originally present in the gasoline-fueled vehicle. The LPG fuel system, while different, is also expected to be designed to be as durable as the fuel systems of other vehicle types. Durability of other LPG-fueled vehicle components, including the emissions control system, is expected to be similar to that observed in gasoline-fueled vehicles, since most of these components are essentially the same for both fuels, and are used under similar conditions. Fuel Economy Fuel economy for LPG-fueled vehicles is often expressed as miles per gallon of LPG fuel. However, since it takes 1.4 gallons of LPG to provide the same amount of energy as one gallon of gasoline, the fuel economy is also commonly expressed on an energy-equivalent basis with gasoline (i.e., miles per equivalent gallon of 12 ------- EPA Office of Mobile Sources Special Report gasoline, where one equivalent gallon of gasoline equals 1.4 gallons of LPG). Generally, as can be determined from the C02 data in the appendices, current dual- fueled vehicles have comparable fuel economy ratings on an energy equivalent basis when using LPQ and gasoline; however, the volumetric fuel economy ratings are much lower for LPG than for gasoline (30 percent less with equivalent thermal efficiency). By optimizing the engine for LPG, however, it is possible to obtain a higher thermal efficiency than is possible with gasoline. In a report for the Congressional Research Service, the R.F. Webb Corporation estimated that converted vehicles could achieve 77 percent of the volumetric fuel economy of similar vehicles running on gasoline, and that vehicles that were originally designed for LPG could achieve 87 percent of the volumetric fuel economy of similar vehicles running on gasoline.[16] Webb also reports dedicated LPG-fueled vehicles that are already achieving 95 percent of the volumetric fuel economy of similar gasoline-fueled vehicles, which indicates an increase in thermal efficiency of more than 30 percent.[11] However, it should be noted that no emissions rates were given for this vehicle. EPA recognizes that the available data are not sufficient to allow definitive estimates of in-use fuel economy. The Agency believes that it would be unreasonable to expect that a 30 percent increase in engine efficiency will be applicable to all LPG- fueled vehicles. In fact, given the challenges of optimizing emissions and fuel economy simultaneously, it is doubtful that any LPG-fueled vehicles would have 30 percent greater thermal efficiency than gasoline-fueled vehicles with comparable emissions. Nevertheless, the Agency is reasonably confident that some increase in efficiency will occur, primarily as the result of increased compression ratios. For this report, it is estimated that the thermal efficiency of current LPG-fueled vehicles is the same as gasoline-fueled vehicles, and that the thermal efficiency of future optimized LPG-fueled vehicles will be 10 percent higher than gasoline-fueled vehicles. These assumptions mean that LPG-fueled vehicles will have 70 to 77 percent of the volumetric fuel economy of comparable gasoline-fueled vehicles. The estimate for current vehicles is consistent with the data discussed below. The estimated increase in thermal efficiency for future vehicles is comparable to the theoretical increase in thermal efficiency that would result from increasing the compression ratio of an engine from 9:1 to 12:1 .[17] Such an increase is possible with LPG, given its higher octane. Exhauat Emissions The primary emissions advantage of LPG is that, as a gaseous fuel, LPG is easier to mix with air than gasoline. This better mixing results in a more homogeneous charge in the cylinder which tends to lower both hydrocarbon (HC) and carbon monoxide (CO) emissions. Another advantage is that the combustion of LPG can have a lower peak flame temperature than gasoline, which tends to lower emissions of oxides of nitrogen (NOx). The lower peak flame temperature can, however, result in higher HC emissions. As is the case with gasoline-fueled vehicles, 13 ------- EPA Office of Mobil* Sources Spec*) Report the actual emissions from an LPG-fueled vehicle are highly dependent upon how the vehicle is calibrated. One of the primary parameters of emissions-related engine calibration is the ratio of air to fuel that is mixed for combustion. The air/fuel ratio of a vehicle can be varied from rich, where the air/fuel ratio is low and there is not enough air to fully combust the fuel, to lean, where the air/fuel ratio is high and there is more air than is necessary to fully combust the fuel. When air/fuel ratio is such that there is just enough air to fully combust the fuel, then the engine is said to be calibrated at stoichiometry. Often, especially when comparing different fuels, the air/fuel ratio is normalized by dividing the actual air/fuel ratio by the stoichiometric air/fuel ratio; this is the excess-air factor (X). The value of the air/fuel ratio affects the emissions of an engine in several ways, as shown in Figure 2-1. Obviously, when the engine is calibrated rich (i.e., low air/fuel ratio), the fuel cannot be completely oxidized to CO2 due to the lack of sufficient oxygen for reaction. The incomplete reactions leave unbumed fuel (HC) and partially burned products (HC and CO). Increasing the air/fuel ratio (i.e., leaning the mixture) tends to decrease emissions of HC and CO; but if the mixture is leaned too much, misfiring or failure of full flame front propagation can occur, which leads to increased HC emissions. The air/fuel ratio affects NOx emissions in two ways: by affecting the combustion temperature, and by changing the concentrations of oxygen and nitrogen in the cylinder. Enriching a lean mixture can lead to increased combustion temperatures, while leaning the mixture increases the concentrations of oxygen and nitrogen in the cylinder; either of these effects can lead to increased NOx emissions. Because of these competing effects, NOx formation tends to be greatest when the air/fuel ratio is set slightly lean of stoichiometry. Clearly then, it is not possible to control the air/fuel ratio to simultaneously minimize engine-out emissions4 of HC, CO, and NOx. The manufacturer's optimal calibration depends on the relative importance of reductions in engine-out emissions of HC, CO, and NOx, based on applicable emission and fuel economy standards, and the availability of exhaust aftertreatment that can further reduce emissions. Gasoline-fueled vehicles are typically calibrated rich for start-up, but calibrated to operate with a stoichiometric air/fuel ratio in a closed-loop mode under normal operation.[18] During start-up, it is necessary to run rich because the gasoline is not fully vaporized until the engine becomes hot. This is not the case for LPG-fueled vehicles, and properly calibrated LPG-fueled vehicles should be calibrated at 4 Engine-out emissions are the emissions before (without) any aftertreatment such as a catalytic converter. 14 ------- EPA Offlc* of MoM« Sourctt Special Report stoichiometry or leaner during start-up. Two of the most important reasons why gasoline-fueled vehicles are calibrated to operate with a stoichiometric air/fuel ratio under normal operation are that current oxygen sensors only work for vehicles calibrated with stoichiometric air/fuel ratios, and that stoichiometric calibration maximizes the efficiency of the catalytic converters. Both of these factors apply to current LPG-fueled vehicles as well; thus, it is reasonable to expect that typical "current" LPG-fueled vehicles will also be calibrated to operate with a stoichiometric air/fuel ratio under normal operation. Existing LPG-fueled vehicles were generally not designed to take advantage of the potential emissions benefits of LPG. In theory, some minor emissions reductions might be expected from improved mixing, or from LPG combustion properties; however, it is very difficult to quantify the reductions because of the limited amount of emissions data that is available. Older LPG-fueled vehicles can have extremely high emissions, similar to older gasoline-fueled vehicles. A report by the California Air Resources Board (GARB) showed CO emissions from a 1979 LPG- fueled van to be 201 g/mile.[19] For comparison, the federal CO emission standard for light-duty trucks in 1979 was 18 g/mile. The CARB report also included data from testing of other LPG-fueled vehicles, all of which are shown in detail in Appendix I. These data are included to demonstrate that LPG-fueled vehicles can have fairly high emissions. Thus, it is important to remember that simply switching to LPG will not provide emissions benefits if the vehicles are not properly calibrated. There are data available from testing of later model LPG-fueled vehicles; these data are summarized in Appendix II. Many of these were dual-fuel vehicles, anjd testing was performed using both LPG and gasoline (see Appendix III). While comparison of such data from testing using the two fuels with the same vehicle would appear to provide an ideal means to estimate the benefits of LPG, there are several reasons why this is not the case. First, it is not necessarily appropriate to compare the emissions of a dual-fuel vehicle using LPG to the emissions from the same vehicle using gasoline after the conversion since it is likely that the conversion process affected the gasoline-fueled performance of the vehicle somewhat. Second, the conversions generally are not made at zero miles, so that the gasoline systems can have significantly higher mileages than the LPG systems have when tested. Finally, since engines can have slightly lower power output using LPG than when using gasoline,[15] it is possible that a customer planning to convert a gasoline-fueled vehicle to an LPG-fueled vehicle would purchase a larger engine than would have been necessary with dedicated gasoline use. 15 ------- EPA Offic* at Mobte SOUICM Spatial Report Figure 2-1 ppml OJ Nm g/kWh •O 5001 E -400 | o o 300*5 3 200 100 .0 a W 0.8 1 12 Excess-air factor A From 'Bosch Automotive Handbook,' 2nd Edition, 1986. 16 ------- EPA Office of Mobile Sources Special Report It is also not possible to determine the fleet-wide exhaust benefits of LPG- fueled vehicles when compared to representative gasoline-fueled vehicles by simply calculating the average emission rates from the available data. The emissions data that are available vary a great deal from vehicle to vehicle, and the distribution of vehicles for which emissions data are available is not comparable to a cross section of current in-use gasoline-fueled vehicles; the number of trucks and large cars is disproportionately high. Further, the testing was performed at different mileages, and is difficult to compare to average in-use levels since these LPQ-fueled vehicles do not have well characterized deterioration rates. Also, as noted above, there is some reason to believe that LPG-fueled vehicles may have less engine deterioration in-use than gasoline-fueled vehicles, but the significance of this is unclear. This uncertainty makes it difficult to determine just how LPQ-fueled vehicles will perform in use. There are data from four LPG-fueled vehicles tested by the State of Colorado, for which there are also baseline gasoline data available.[20] This test program measured emissions from gasoline-fueled vehicles before they were converted for use of LPG, then measured emissions from the same vehicles, using LPG. These data are shown in Table 2-2. While many of the same concerns raised above remain valid with respect to these data, and the testing was performed at high altitude, the gasoline-fueled emissions were measured before conversion, and can be considered as truly baseline emissions. As such, these data are a somewhat reasonable basis for comparison of emissions. The only clear trends from these data are that the CO and CO2 emissions from the LPG-fueled vehicles are consistently lower. For CO, the reductions range from 10 to 82 percent, and the average reduction is nearly 50 percent. The average COa reduction is 13 percent which indicates a nearly 5 percent increase in the thermal efficiency of the engine. The HC and NOx emissions can be higher or lower with LPG, but the average emissions from the four vehicles are similar for both LPG and gasoline. Recently, emissions results from two other studies, that may be even more useful, have become available. In both studies, gasoline- and LPG-fueled vehicles were operated side-by-side for several thousand miles. The first is a study by Phillips Petroleum Company that compared the emissions from converted LPG-fueled vehicles and identical unconverted gasoline-fueled vehicles.[21] The testing showed CO emissions to be 21 percent lower, HC emissions to be 5 percent lower, and NOx emissions to be 30 higher on average, with essentially identical thermal efficiency. The second study involves an ongoing demonstration of the use of alternate fuels in delivery vehicles.[22] The most useful data from this study are the emissions results from three converted LPG-fueled vans and three unconverted gasoline-fueled vans. These data are shown in Table 2-3. The LPG-fueled vans averaged 45 percent lower CO emissions, 56 percent higher NMHC emissions, 47 percent higher NOx emissions, and 10 percent lower CO2 emissions. 17 ------- EPA Offlc* at MobM Source* Special Report Table 2-2 Comparison of Emissions from Identical Vehicles Before and After Conversion VEHICLE 1988 CROWN VICTORIA 1988 CROWN VICTORIA 1989 CHEVY PICKUP 1989 CHEVY PICKUP 1990 DAKOTA 1990 DAKOTA 1990 FORD 1-TON TRUCK 1990 FORD 1-TON TRUCK FUEL GASOLINE LPG GASOLINE LPG GASOLINE LPG GASOLINE LPG HC (G/MI) 0.28 027 0.47 0.37 0.31 0.45 0.91 0.93 CO (G/MI) 1.20 0.28 4.00 3.30 2.00 0.37 6.5 5.8 NOx (G/MI) 0.80 0.60 0.70 0.70 1.10 1.70 5.9 4.7 CO2 (G/MI) 458 398 525 485 449 379 1065 907 Source: Colorado Department of Health [20] Additional information is shown in Appendix II Table 2-3 Comparison of Emissions from Converted LPG-Fueled and Unconverted Gasoline-Fueled Delivery Vans VEHICLE 1992 FORD VAN 1992 FORD VAN FUEL GASOLINE LPG HC (G/MI) 0.264 0.412 CO (G/MI) 1.8 1.1 NOx (G/MI) 0.75 1.1 CO, (G/MI) 662 594 Source: Clean Fleet [22] 18 ------- EPA Office of Mobile Source* Special Report Given these data, and the previous assumption that the manufacturers' optimization strategy would be aimed at minimizing fuel consumption and complying with federal emission standards, (and that they would do this by recalibrating the engine for LPG to run leaner than the previous calibration for gasoline during start-up), it is most reasonable to assume that fleet-wide exhaust emissions of HC and NOx from current in-use LPG-fueled vehicles would be similar to the average values for gasoline, but that CO emissions would be lower. However, if the air/fuel ratio is not property calibrated, emissions would be higher HC and CO would be higher if the vehicle were calibrated too rich, while NOx would be higher if the vehicle were calibrated too lean. Optimized LPG-fueled vehicles could have more significant emissions benefits compared to conventional gasoline-fueled vehicles than current LPG-fueled vehicles have. However, it is not clear at this time just how such vehicles will be calibrated, and thus it is not possible to precisely predict the future emissions levels of LPG- fueled vehicles. Nevertheless, it is possible to estimate emissions for the assumed emissions optimization strategy of optimizing fuel economy while meeting the federal Ultra-Low Emission Vehicle (ULEV) emission standards: - 0.040 g/mile NMOG5 -1.7 g/mile CO - 0.2 g/mile NOx - 0.008 g/mile Formaldehyde For this analysis, it is assumed that optimized LPG-fueled vehicles will comply with these standards in use. The emissions from prototype gasoline-fueled vehicles with advanced emission control systems such as electrically heated catalysts are beginning to approach these levels. Thus, given the cold-start advantage of LPG over gasoline, it is very reasonable to expect that a fully optimized LPG-fueled vehicle could ultimately be able to meet these standards. These are reasonable goals for alternative fueled vehicles since CARB's LEV program provides strong encouragement for the production of a significant number of vehicles meeting these standards. In fact, IMPCO Technologies is currently working to develop a passenger vehicle that meets these standards.[6] The Agency recognizes, however, that production of LPG- fueled ULEVs would require a significant developmental effort and may ultimately be prohibitively expensive. 9 For LPG, NMOG (Non-methane organic gas) is approximately equal to NMHC multiplied by a reactivity adjustment factor (RAF), and is intended to provide a better indication of the effect of the emissions on ambient ozone levels. If, for example, the RAF for LPG-fueled vehicles were 0.5, then a 0.04 g/mi NMOG standard would be roughly equivalent to a 0.08 g/mi NMHC standard. (See Chapter 4 for additional discussion.) 19 ------- EPA Office of Mobil* Source* Special Report Low Temperature Emissions The primary emission concern regarding gasoline-fueled vehicles during winter months is CO. At lower temperatures, gasoline does not vaporize well when the engine is cold, and this leads to poor combustion and higher emissions. To account for the poor vaporization, gasoline-fueled vehicles are calibrated to run rich during start-up. LPG, on the other hand, vaporizes fully and mixes more completely with the air, even at extremely low ambient temperatures, and thus does not require a rich calibration during start-up. Data from cold temperature testing, which are shown in Appendix IV, show that the difference in emissions can be dramatic.[23] These data suggest that an LPG-fueled vehicle emitting the same amount of CO at 25*C (77T) as a gasoline-fueled vehicle would emit only 7 to 17 percent as much CO as the gasoline-fueled vehicle at O'C (32'F), or as little as 5 percent at -10'C (14'F). If the LPG-fueled vehicle emitted less CO at 25°C, then the reductions would be even greater. Based on these data, it is estimated that LPG-fueled vehicles will average 80 percent lower emissions at very low temperatures than gasoline-fueled vehicles. This estimate is intended to be somewhat conservative; however, it is important to remember that since the amount of low temperature data available is very limited, it is possible that actual emissions benefits may be less than this. Evaporative and Other Emissions Since LPG fuel systems are designed to contain pressurized fuel, it is possible to virtually eliminate evaporative emissions and greatly reduce refueling emissions. Some older LPG-fueled vehicles can have significant evaporative emissions, but such emissions can be eliminated for properly designed vehicles. For this report, it is being assumed that LPG-fueled vehicles will have negligible, evaporative emissions. Nevertheless, the Agency recognizes that there is a potential for some LPG to escape from the fuel system, especially with improperly converted vehicles. In such cases it would be possible for significant evaporative emissions to occur, even if the leaks did not significantly affect fuel economy. There are two sources of refueling emissions from LPG-fueled vehicles. First, many older vehicles rely on outage valves (or spit valves) to indicate when the fuel tank has been filled to its maximum level. When these valves are used, LPG vapors are continuously emitted through the open valve during refueling. However, more recent vehicles have automatic shut-off valves, which eliminate the need to use an outage valve during refueling. EPA has recently finalized regulations for LPG-fueled vehicles, that include provisions for refueling emissions.[4] These regulations will require automatic shut-off valves, thus this source of emissions is being eliminated. The second source of emissions is the small amount of liquid fuel that remains in the connection between the vehicle fuel inlet and the fuel dispensing nozzle. When this connecting seal is broken, the liquid fuel evaporates into the atmosphere. The 20 ------- EPA Office of Mobile Sources Special Report mass emitted can be reduced by reducing the dead volume in the connection. This volume can be very large for current systems which, until recently, have not been regulated.[4] It has been estimated that dead volumes for natural gas systems can be more than 100 cm3.[24] Although a similar analysis is not available for LPG systems, it is reasonable to assume that dead volumes could also be that large in some LPG systems. In optimized LPG refueling systems, the dead volumes range from 1 to 12 cm3.[25] For this report, it will be assumed that the average dead volume of current in-use refueling systems would be 5 cm3, which leads to an emission rate of 2.5 grams per refueling event. For future vehicles, it is assumed that they will comply with the recent EPA refueling regulations, and have average dead volumes of 3 cm3. [4] As can be seen in the environmental analysis of Chapter 4, the impact of refueling emissions is small, and the validity of these assumptions is not critical. Vehicle Scenarios The following chapters will analyze and discuss the economic and environmental impacts of LPG-fueled vehicles. These analyses will be performed using two sets of assumptions to bracket possible benefits. "Current converted" LPG- fueled vehicle emissions and fuel economy are based on a very limited amount of data from late model vehicles tested at various mileages and vehicle conditions. It is assumed that these vehicles are not optimized for low emissions, but rather for fuel economy. Exhaust HC and NOx emission rates are estimated to be the same as from gasoline-fueled vehicles, and CO emissions are estimated to be 80 percent lower at low temperatures. No corrections are made to account for potential differences in in- use deterioration between LPG- and gasoline-fueled vehicles. "Optimized OEM" LPG-fueled vehicle emissions and fuel economy ratings are projections of what could be achieved with a significant developmental effort, and are based upon engineering judgment. It is assumed that these vehicles will be optimized for fuel economy, but also will be designed to comply with the federal ULEV standards. The ULEV standards are an appropriate goal for alternative fueled vehicles since there are emissions related credit programs available for vehicles that meet these standards. The reader is cautioned that these performance estimates are assumptions that are based upon very limited data. Nevertheless, they are sufficient for the purposes of this report, which is to estimate the potential for economic and environmental impacts of LPG-fueled vehicles when compared to conventionally-fueled and other alternately-fueled vehicles. 21 ------- EPA Office of Mobile Sources Special Report The conventionally-fueled vehicles to which the LPQ-fueled vehicles will be compared in Chapters 3 and 4 are: "1990 Base" An average vehicle in the 1990 (calendar year) light-duty gasoline-fueled vehicle fleet; "2010 Base* An average vehicle in the 2010 (calendar year) light-duty gasoline-fueled vehicle fleet; "Gasoline ULEV" A vehicle certified to the ULEV standards using conventional gasoline fuel. It should be noted that, at this time, it is not certain that any future vehicles actually produced will be able to comply with the ULEV requirements using conventional gasoline. Nevertheless, as is the case with LPG, such a vehicle is theoretically possible, though it may not be economically feasible; thus it is an appropriate vehicle for comparison with advanced LPG-fueled ULEVs. Heavy-Duty Engines and Vehicles There is little emissions data available from LPG-fueled heavy-duty engines and vehicles. In general, however, emissions from heavy-duty spark-ignited engines should be similar to emissions from light-duty vehicles; fleet average exhaust HC and NOx emissions should be about the same as those from gasoline-fueled engines, low temperature CO emissions should be much lower than those from gasoline-fueled engines, there should be little or no evaporative emissions, and very small refueling emissions. It is important to note that, as was the case with light-duty vehicles, actual emissions from individual engines will vary greatly, depending on the calibration of the engines. Emissions data from an LPG-fueled truck which was recently certified by the State of California are summarized in Appendix V, and emissions data from a bus demonstration project are summarized in Appendix VI. It is possible, if not probable, that heavy-duty LPG-fueled engines will have higher compression ratios than light-duty engines, and thus also have better thermal efficiencies. This is because many of the physical and economic limitations that apply to light-duty engines, are less important in heavy-duty applications. In general, operators of heavy-duty vehicles are more willing to pay the extra cost for an engine that has better fuel economy. Thus, it is very possible that many heavy-duty LPG- fueled engines would have compression ratios significantly higher than 12:1, although not as high as those of diesel engines (-20:1). 22 ------- EPA Otfka of Mobil* Sourctt Special R«port It is also worth noting that, despite its high octane/low cetane rating, LPG can be used with Diesel-cycle engines. This, however, requires a pilot injection of conventional diesel fuel to ignite the LPG without a spark plug. Such systems are referred to as "fumigation11 systems. While this technology is promising, at this time, it is not very advanced. 23 ------- EPA Office of MoM« Sources Special Report References 1. "1990 Annual Book of ASTM Standards," Vol. 05.05. 2. "Environmental and Economic Study of Alternate Motor Fuel Use: Report to Congress," EPA Office of Mobile Sources, Draft, April 1991. 3. "Gaseous Fuel Vehicle Technology State of the Art Report," EA Mueller Inc., for U.S. Department of Energy, December 1988. 4. "Standards for Emissions From Natural Gas-Fueled, and Liquefied Petroleum Gas-Fueled Motor Vehicles and Motor Vehicle Engines, and Certification Procedures for Aftermarket Conversion Hardware," Final Rule, 59 FR 48472, September 21,1994. 5. Engine, Fuel, and Emissions Engineering, Inc., for Natural Gas Vehicle Coalition, comments on draft Special Report, July 19, 1994. 6. Personal communication with David Smith, IMPCO Technologies Inc., July 11, 1994 7. "Vialle Liquid Propane Autogas-lnjection System," Vialle Autogas Systems, from LP Gas Clean Fuels Coalition. 8. Manufacturers of Emissions Controls Association meeting with EPA, May 11, 1993. 9. "Emission Standards for Clean-Fuel Vehicles and Engines, Requirements for Clean-Fuel Vehicle Conversions, and California Pilot Test Program," Final Rule, 59 FR 50042, September 30, 1994. 10. Krupka, M.C., A.T. Peaslee, and H.L. Laquer, "Gaseous Fuel Safety Assessment for Light-Duty Automotive Vehicles," Los Alamos National Laboratory Report 0LA-9829-MS, UC-96, November 1983. 11. "An Assessment of Propane as an Alternative Transportation Fuel in the United States," R.F. Webb Corp., for National Propane Gas Association, June 1989. 12. "Propane and Environment in Ontario - A Clean Transportation Fuel Alternative," Propane Gas Association of Canada, September, 1992. 13. "Federal Motor Vehicle Safety Standards; Compressed Natural Gas Fuel System and Fuel Tank Integrity: Notice of Proposed Rulemaking," NHTSA, 58 FR 5323, January 21, 1993. 14. "Standard for the Storage and Handling of Liquefied Petroleum Gases," NFPA 58, 1989. 15. "Vehicle Reet Survey: Opportunities for Reducing Air Pollution Through the Use of Compressed Natural Gas and Propane," Denver Metropolitan Air Quality Council, October 1987. x 16. "Investigation Regarding Federal Policy Actions for Encouraging Use of Liquified Petroleum Gas as a Motor Vehicle Fuel,' R.F. Webb Corporation, for Congressional Research Service, April 1992. 17. Muranaka, S., Y. Takagi, T. Ishida, "Factors Limiting the Improvement in Thermal Efficiency of S.I. Engine at Higher Compression Ratio," SAE #870548, 1987. 18. Grouse, W., and D. Anglin, "Automotive Engines," McGraw Hill, 1986. 24 ------- EPA Office of MOM* Source* Special Report 19. "Summary Report for Emission Testing of Liquified Petroleum Gas Fueled Vehicles Provided by Sears," California Air Resources Board. 20. Data from Ron Ragazzi, Colorado Department of Health, February 1991. 21. O'Connor, J., "Report on Propane Vehicle Fleet Tailpipe Emissions, Drivability, and Durability Study," Phillips Petroleum Company, November, 1993. 22. "Clean Fleet: Vehicle Exhaust Emissions Early Mileage Results," Batelle Memorial Institute, Statistical Analysis Report No. 4, July 1994. 23. Rideout, G., "Effect of Ambient (Cold) Temperatures on Exhaust Emissions and Fuel Consumption of LPG Fueled Vehicles,* Environment Canada, May 1991. 24. Comments from Sherex Industries in response to the proposed emissions standards for gaseous-fueled engines and vehicles, December 1,1992, (Docket itemA-92-14-IV-D-13). 25. Information from Bob Myers, LP Gas Clean Fuels Coalition, Docket A-92-14. 25 ------- EPA Office erf Mobile Sources Special Report Chapter 3 Economic Impacts Introduction This chapter describes the potential economic impacts of LPG-fueled vehicles on a per-vehicle basis, and discusses the current and expected future supply of LPG. The analysis of costs presented here is based upon the current energy market; thus, since the prices of fossil fuels can change significantly within a very short period of time, the reader is cautioned to consider this analysis to be primarily for the purpose of identifying trends, and providing reasonable estimates of the costs associated with LPG-fueled vehicles at this time. The analysis is described in a step-by-step manner, which would allow the reader to revise it to include changes in either LPG or crude oil prices. This analysis does not make quantitative projections of how the price of LPG would change in response to the increase in demand that would result from a significant increase in the number of vehicles using LPG. LPG Supply As noted in Chapter 1, LPG is widely distributed throughout the United States, for residential and commercial heating as "propane," and this market is already well defined. Most commercial LPG pipelines have requirements that the LPG meet the composition specifications for "Special Duty Propane," which is the grade of LPG currently used for automotive engines.[1] This specification has a minimum requirement of 90 percent propane and a maximum 2.5 percent butane and higher hydrocarbons. The United States' domestic supply of LPG comes from two sources: 1) wet gases, or natural gas liquids, stripped during the processing of natural gas; and 2) byproduct gases removed from crude oil during refining. (About 80 percent of U.S. LPG currently comes from natural gas production, while the remaining 20 percent comes from refineries.[2]) In 1993, the average rate of domestic LPG production (propane and propylene only) was 0.96 million barrels per day6, and the rate of LPG imports (propane and propylene only), much of which comes from Canada, was 0.10 million barrels per day.[3] The Department of Energy estimated U.S. natural gas liquids reserves7 to be 7,464 million barrels as of December 31, 1991 .[4] (The propane content of natural gas liquids is generally less than 50 percent.) These reserves are located primarily in the southwest. 6 A standard petroleum barrel equals 42 U.S. gallons. 7 The known amount of natural gas liquids in the ground that can be extracted economically. 26 ------- EPA Offie* of Mobil* SourcM Special Report If LPG gains widespread use as a transportation fuel, there are several potential sources of additional supply to meet the demand, although at a higher price. The domestic supply for vehicular LPQ could be increased by a combination of increased production and shifts in demand resulting from higher prices for LPQ. Even more LPG could be available through imports. These potential supplies are described below. Domestic vehicular LPG supply could be increased by a several changes in the production and use of LPG, especially if vehicles begin to use LPG fuels with higher butane content. Such changes would result to some extent if the price of LPG rose significantly, due to a higher demand for vehicular LPG. A higher price for LPG would lead to some increases in the recovery of LPG at natural gas processing plants and refineries. Changes in other industries, such as switching the chemical industry back to using ethane as a feedstock for many of its processes, instead of naphtha, could make even more LPG available. The recent regulations requiring reductions in Reid vapor pressure (RVP) of reformulated gasoline will decrease the demand for butane by refineries. One estimate is that such changes could increase the U.S. production of LPG by as much as 1.59 million barrels per day.[5] The plant cost associated with this increase was estimated to be $18.8 billion (35 cents per gallon). Other estimates suggest that the domestic supply of LPG for vehicles in 2010 could be 0.7 to 0.9 million barrels per day.[6,7] The reader is cautioned that these estimates involve dramatic changes to the LPG market, and should be viewed as upper bounds on the amount of LPG that could be available domestically. Such changes would likely lead to very substantial increases in price. Any significant increase in the demand for LPG would also lead to more LPG being imported. It has been estimated that more than 0.6 million gallons of LPG per day could be available through imports.[8] As was the case with domestic supply, although to a lesser extent, increasing imports by this much would be accompanied by an increase in the price of LPG. In summary, the actual availability of LPG will be highly dependent on market demand. It appears that, with very high demand, and some changes in fuel composition (e.g., higher butane content), the supply of LPG for vehicle use could be on the order of one million barrels per day. However, there would probably be a very significant increase in price associated with increasing the supply by this much. One million barrels per day would be enough for a fleet of vehicles averaging 15 miles per gallon (LPG) to travel more than 600 million miles per day. This amount would represent more than 13 percent of the total vehicle miles traveled (VMT) for gasoline vehicles the U.S. in 1990.[9] 27 ------- EPA Office of Mobito Sources Special Report Fuel Prices This analysis of LPG fuel prices addresses the costs that would be associated with a scenario involving a moderate number of LPQ-fueled vehicles. One such scenario would be the use of a relatively large numbers of vehicles in a limited geographical range, such as a single metropolitan area. Under such a scenario, the number of vehicles would be large enough to provide significant economies of scale for the region, but small enough to not have an impact on the national LPG market price. For the purpose of consistency, the methodology used to estimate the resale price of LPG will be similar to the methodology used for methanol in the Special Report on methanol,[10] and will use the same estimated price for gasoline (which was based upon the assumption that crude oil was 20 dollars per barrel). This will allow the reader to compare the costs of LPG, with those of other fuels for which there are QMS Special Reports, such as CNG or methanol. This analysis could be revised by the reader to account for other crude oil prices by assuming that the wholesale price of gasoline is roughly proportional to the price of crude oil. The reader could also revise this analysis to be based upon the current wholesale price of gasoline, which would be available from EIA. Based on the analysis described below, it is estimated that LPG sold at service stations for automotive use would cost 79 to 85 cents per gallon if LPG-fueled vehicles were widely used within a given region. The price on an energy-equivalent basis with gasoline is 1.13 to 1.22 dollars per gallon, which is comparable to the estimated price of gasoline (1.20 dollars per gallon based on an assumed wholesale gasoline price of 69 cents per gallon). This analysis of the costs of LPG and gasoline is summarized in Table 3-1. It must be emphasized that this estimate is valid only for moderate numbers of vehicles; if the demand for LPG increased substantially due to an large increase in the number of LPG-fueled vehicles, the price of LPG could be also substantially higher. According to EIA, the 1994 average refiner sales price for resale was 32.5 cents per gallon for consumer grade propane.[11] This price was used for this analysis, but the Agency recognizes that the future price is likely to be different, especially if the demand for LPG changes. Moreover, the price of propane is subject to substantial seasonal variations, as can be seen in Figure 3-1; and the yearly average price is highly dependent upon the severity of the preceding winter. 28 ------- EPA Office of Mobile Source* Special Report Table 3-1 Summary of Fuel Costs Wholesale Price Distribution Costs Service Station Markup Federal Taxes State Taxes Total Pump Price $/gallon LPG 0.325 0.060-120 0.096 0.183 0.125 0.79-85 $/gallon Gasoline 0.690' 0.060 0.086 0.184 0.183 1.20 Based on a crude oil price of $20/bbl Since LPG is often located in the regions where gasoline is produced, the long range/local distribution cost of LPG may be similar to that of gasoline. However, it has also been estimated that distribution cost could be twice as high as that of gasoline, assuming that new pipelines had to be buitt.[12] In this analysis it is assumed that the LPG distribution cost for vehicle use would range from 6 to 12 cents per gallon, based on a gasoline distribution cost of 6 cents per gallon.[13] Service station mark-up has been estimated to be 8.6 cents per gallon for gasoline and 9.6 cents per gallon for LPG (for stations selling 69,000 gallons of LPG per month).[14] Higher costs would result if a gasoline service station sold only a small volume of LPG, and could not take advantage of the economies of scale of larger tank systems. However, since this report is attempting to quantify the impacts of a significant number of LPG-fueled vehicles, it is assumed that the LPG vehicle fuel market will be large enough, at least locally, so that the stations would be able to avoid such problems. Federal taxes (1993) for LPG and gasoline are 18.3 and 18.4 cents per gallon, respectively.[15] The state taxes (1993) used in this analysis, 12.5 cents per gallon for LPG and 18.3 cents per gallon for gasoline, are weighted national average values from the Department of Transportation.[15] 29 ------- EPA Office of Mobile Sources Special Report Figure 3-1 Wholesale LPG Prices January 1990 through December 1994 100 80 -2 60 £ 40 20 Jan Jul 90 Jen Jul 91 Jan 92 Jan Jul Jan 93 94 - Avw»g»PrmtorPrMadng12Monf« Vehicle Fuel Economy and Coet As was noted in Chapter 2, the fuel economy of an LPG-fueled vehicle is very dependent upon how the vehicle is calibrated. For this analysis, it is assumed (see Chapter 2) that current LPG-fueled vehicles have a volumetric fuel economy that is 70 percent of that of a similar gasoline-fueled vehicle. For advanced LPG-fueled vehicles it is assumed that they will have a volumetric fuel economy that is 77 percent of that of a similar gasoline-fueled vehicle. The fleet-average fuel economy for light- duty gasoline-fueled vehicles is 20.5 miles per gallon (MPG).[9] Average fuel economy for a similar fleet of LPG-fueled vehicles would be 14.3 MPG for current vehicles and 15.7 MPG for advanced vehicles, on a volumetric basis. Using these 30 ------- EPA Offlc* of Mobil* Sourm Special Roport estimates, fuel costs would be 5.9 cents per mile for a gasoline-fueled vehicle, 5.5 - 5.9 cents per mile for a current LPQ-fueled vehicle, and 5.0 - 5.4 cents per mile for an advanced LPG-fueled vehicle. Of course, these estimates are very sensitive to the relative wholesale prices of LPG and gasoline. It may be more reasonable, however, to consider fuel costs for a fleet of larger vehicles with lower fuel economy. For example, the fuel costs for gasoline-fueled vehicles with an average fuel economy of 10 MPG would be 12.0 cents per mile, while the fuel costs of a comparable fleet of optimized LPG-fueled vehicles (7.7 MPG) be 10.3 -11.0 cents per mile. Thus, the savings for fuel cost would be twice as large as for the average fleet described above (1.0 -1.7 cents per gallon). There are several reasons why LPG is more likely to be used in these vehicles. First, due to refueling convenience, LPG is easier to use in fleet applications, and many of these fleet vehicles (e.g., utility trucks, delivery vans, etc.) are fairly large. Second, the inherent emission advantages of LPG are more valuable to manufacturers for vehicles with large engines, since it is more difficult for these engines to comply with emissions standards. Finally, owners of large vehicles with low fuel economy are more likely to be interested in the potentially lower fuel costs of an optimized LPG-fueled vehicle. The cost of converting a gasoline-fueled vehicle to a LPG-fueled vehicle varies by vehicle type. For this analysis, it is assumed that the current cost of conversion is $1500. This is consistent with other reported estimates.[2,16] Amortizing this cost over 5 years and 100,000 miles, a current LPG-fueled vehicle would cost an additional 1.8 cents per mile over the cost of a conventional vehicle. For future vehicles, it is assumed that the costs associated with producing an LPG-fueled vehicle capable of complying with the ULEV requirements will be roughly comparable to those associated with producing a vehicle capable of complying with the ULEV requirements using conventional gasoline. Either vehicle would be expected to cost significantly more to produce than a gasoline-fueled vehicle designed only to comply with conventional standards. It should also be noted that current literature indicates that LPG-fueled vehicles could have less in-use engine wear, and thus have reduced operation and maintenance costs compared to gasoline-fueled vehicles. As noted earlier, there is some evidence to suggest that this benefit could lead to as much as a doubling of engine life. If this is the case, it would obviously provide a very significant economic benefit by reducing the extra cost of the vehicle. However, these cost reductions have not been rigorously quantified. Therefore, while the Agency recognizes a potential for benefits, in the absence of sufficient information, they cannot be accurately determined for this report. 31 ------- EPA Office of MOM* SourcM Special Report Summary Given the estimated price of LPG used in this analysis, current LPG-fueled vehicles may not provide savings in fuel costs when compared to gasoline-fueled vehicles. Moreover, even with lower LPG prices, savings in fuel costs might still be offset by the costs associated with the vehicle conversion. Thus the potential for economic benefits from LPG-fueled vehicles is highly dependent upon the extent to which maintenance costs are reduced. However, if the differential vehicle costs were reduced, the relative price of LPG decreased, and/or the fuel economy of LPG-fueled vehicles were improved, it is apparent that LPG-fueled vehicles could provide significant economic benefits even without considering reduced maintenance costs. The total market penetration of LPG-fueled vehicles in the future will be limited by the supply of LPG. The increase in demand that would result from a large increase in the number of LPG-fueled vehicles would lead to some increase in the price of LPG, as LPG production and usage rates are shifted to accommodate the new demand. The expected supply and demand of LPG in the future are such that LPG could be used as an automotive fuel in a small but significant number of fleet applications. The supply would be improved somewhat if the composition of the fuel was changed to include other light hydrocarbons such as butane. 32 ------- EPA Office of Mobito Sources Special Report References 1. Personal communication with Bill Butterbaugh, National Propane Gas Association, August 2, 1990. 2. "An Assessment of Propane as an Alternative Transportation Fuel in the United States," R.F. Webb Corp., for National Propane Gas Association, June 1989. 3. "Petroleum Supply Monthly: July 1993," Energy Information Administration, DOE, DOE/EIA-0109(94/07). 4. "U.S. Crude Oil, Natural Gas, and Natural Gas Liquids Reserves: 1991 Annual Report,' Energy Information Administration, DOE/EIA-0216(91). 5. Myers, R.E., for L.P. Gas Clean Fuels Coalition, "The Supply, Infrastructure and Economics of an Expanded Motor Vehicle Fuel Market for LPG in the U.S.," Draft, March 1992. 6. "Second Interim Report of the Interagency Commission on Alternative Motor Fuels," September 1991. 7. "Investigation Regarding Federal Policy Actions for Encouraging Use of Liquified Petroleum Gas as a Motor Vehicle Fuel," R.F. Webb Corporation, for Congressional Research Service. 8. "First Interim Report of the Interagency Commission^on Alternative Motor Fuels," September 1990. 9. MOBILE4 Fuel Consumption Model, Draft Report, April 1991. 10. "Analysis of the Economic and Environmental Effects of Methanol as an Automotive Fuel," U.S. EPA/OAR/OMS Special Report, September 1989.(PB 90-225806) 11. "Monthly Energy Review: March 1995," Energy Information Administration, DOE, DOE/EIA-0035(94/07). 12. "Assessment of LPG Infrastructure for Transportation Use," Draft Report, EA Energy Technologies Group, September 1992. 13. "Preliminary Perspective on Pure Methanol Fuel for Transportation," U.S. EPA/OAR/OMS, EPA 46/3-83-003, September 1982.(PB 83-180232) 14. U.S. Department of Energy, comments on draft Special Report, May, 1994 15. "Highway Statistics 1993," U.S. Department of Transportation, FHWA-PL-94- 023,1994. 16. "Propane and Environment in Ontario - A Clean Transportation Fuel Alternative," Propane Gas Association of Canada, September, 1992. 33 ------- EPA Office of Mobile Sources Special Report Chapter 4 Air Quality Impacts Introduction Much of the impetus for the development of alternative fuels such as methanol and compressed natural gas has been due to their potential air quality benefits. The use of LPG as a motor fuel, which to date has generally been driven by economic concerns, can also provide significant benefits to urban air quality. This chapter provides an analysis of the overall air quality impact of using LPG in place of gasoline, focusing on its impacts on urban ozone and CO levels, air toxics, and global warming. Urban Ozone Levels High levels of ozone (smog) represent one of the most pressing air quality concerns facing urban areas today. These high ozone levels are caused by photochemical reactions between hydrocarbons (HC) and oxides of nitrogen (NOJ, and the impact dl mobile source controls has traditionally been assessed by examining the percent reduction in HC and NOx emissions. Another approach has been to consider the reduction in non-methane hydrocarbon (NMHC) emissions, since methane emissions do not significantly affect ozone levels due to methane's low reactivity in the atmosphere. This second approach is used in this report. While NOx emissions contribute to the formation of ozone in urban areas, the analysis in Chapter 2 suggests that the use of LPG-fueled vehicles will not lead to significant changes in mobile source NOx emissions. Therefore, NOx emissions are not discussed further here. In recent years, there has also been a significant amount of work to determine the relative rates at which other individual hydrocarbons react to form ozone, in order to better assess the impacts of HC emissions on ozone levels. One approach has been to assign Reactivity Adjustment Factors (RAFs) to each vehicle-type. The RAF is intended to be an estimate of the amount of ozone produced by a vehicle-type, relative to gasoline-fueled vehicles. For example, an RAF of 0.50 for LPG would mean that one gram of organic emissions from an LPG-fueled vehicle would produce only 50 percent as much ozone as one gram from a gasoline-fueled vehicle. While such work has shown that there are significant differences among the reaction rates for various classes of compounds, it is still problematic to attempt to precisely quantify the role of different HC species in the formation of ozone. The California Air Resources Board (CARB) had estimated the RAF for LPG to be 0.50, but this estimate has been withdrawn because of insufficient informational] It is necessary to assume a value for the RAF in order to convert the applicable NMOG standard into an effective NMHC standard for ULEVs, and this report assumes the RAF for LPG to be 0.50. However, this value is not used in the analyses of this chapter to estimate 34 ------- EPA Office of Mobile Sources Special Report the actual relative reactivity of LPG emissions. While the relative reactivity of LPG emissions is discussed qualitatively below, the estimated impact of LPG-fueled vehicles on urban ozone levels is limited to estimates of NMHC emission reductions. Estimated NMHC emissions for light-duty gasoline vehicles are shown in Table 4-1. Emission factors were estimated for three scenarios. The first row represents average in-use emissions from all vehicles in the nationwide 1990 fleet, operating with an 8.7 psi fuel. These emissions were estimated for the Motor Vehicle-Related Air Toxics Study. [2] The second row represents the in-use emissions expected from vehicles in the 2010 fleet. Among the additional requirements that were assumed to apply to these future vehicles are: a 0.25 gram/mile tailpipe NMHC standard, enhanced evaporative emission controls, onboard refueling controls, and improved state inspection and maintenance programs. These estimates are also based on emission rates discussed in the Motor Vehicle-Related Air Toxics Study. The emission rates from that study were adjusted by assuming that refueling emissions were reduced by 90 percent with onboard refueling controls. All of the emissions estimates for that report were obtained from EPA's emission model MOBILE4.[3] Estimated emissions from a gasoline-fueled vehicle meeting the federal ULEV standards are shown in the third row. Also included in Table 4-1 are the estimated NMHC emissions for both current and optimized LPG vehicles which were discussed in Chapter 2. Comparing the NMHC emissions from LPG vehicles to those from gasoline vehicles shows a significant reduction. This reduction results from the virtual elimination of evaporative, running loss, and resting loss emissions, and ve/y substantial reductions in refueling emissions (compared to 1990 baseline) because of the tightly sealed fuel system necessary for the high pressures associated with LPG. When comparing current LPG vehicles to current gasoline vehicles, one would expect to see a 41 percent reduction in NMHC emissions. Optimized LPG vehicles could result in as much as a 91 percent reduction when compared to baseline gasoline fueled vehicles, or a 79 percent reduction when compared to a gasoline-fueled vehicle that was designed to comply with the ULEV standards, but that had non-exhaust emissions similar to the baseline vehicle. It is very important to note that these estimated reductions are very sensitive to the assumptions made in this analysis. This is especially true when considering deterioration. Most of the emissions data for LPG- fueled vehicles are from test programs in which the vehicles received fairly good care. At this point it is not possible to determine how LPG-fueled vehicles will perform in use, under less controlled conditions. Also, while it is possible that evaporative emissions can be essentially eliminated from LPG-fueled vehicles, in-use LPG-fueled vehicles may actually have significant evaporative emissions. Thus, while the directional trends shown above are valid, the exact magnitude of the benefits is less certain. 35 ------- EPA Office of Motxte Sources Special Report Table 4-1 Estimated Emissions from Gasoline- and LPG-Fueled Vehicles Vehicle Type 1990 Base Gasoline 2010 Base Gasoline Gasoline ULEV Current LPG Optimized LPG Exhaust NMHC (g/mile) 1.48 0.61 0.04 1.48 0.08" Hot soak and Diurnal NMHC (g/mile) 0.44 0.18 0.18 0.00 0.00 Other NMHC' (g/mile) 0.58 0.18 0.18 0.01 0.01 • Includes refueling emissions, running losses, and resting losses. "Assumes a Reactivity Adjustment Factor of 0.5. As was mentioned earlier, changes to the fuel used by a vehicle can not only affect the amount of NMHC emitted by the vehicle, but it can also impact the role its emissions will play in the formation of ozone. Unfortunately, there is still much uncertainty related to quantifying these changes, although several trends have already been shown. Paraffinic hydrocarbons, such as propane, tend to react much more slowly in the atmosphere than olefinic species, such as propylene. Since the LPG fuel does not normally contain large amounts of olefins, it may be expected to result in less reactive emissions. CARB's preliminary estimate of the RAF for LPG (which has subsequently been withdrawn) suggests that LPG emissions may be only half as reactive as gasoline emissions. However, olefin formation during combustion can result in LPG vehicles having an olefinic hydrocarbon fraction in their exhaust that is similar to gasoline vehicles,[4] and it is possible that LPG would not offer a large reactivity benefit. Also, the relative reactivity of exhaust emissions is fairly dependent on fuel composition, which may vary substantially in use. 36 ------- EPA Offlc* o* Mobil* SourcM Specol Import In summary, the speciated emissions data that are currently available suggest that there is the potential that using LPG would lead to a reduction in the reactivity of exhaust emissions, but this potential cannot be precisely estimated at this time. More importantly, however, it is estimated that there will be a significant reduction in NMHC emissions on a mass basis. It is expected that the use of LPG would provide significant reductions in ozone on a per-vehicle basis, with or without reductions in reactivity. Carbon Monoxide Carbon monoxide (CO) is a gaseous product of combustion. At sufficiently high levels it can cause breathing difficulties. It has been estimated that approximately two-thirds of CO emissions are produced by transportation sources.[5] As such, any reduction in CO emissions from vehicles can be very significant. The data discussed in Chapter 2 suggest that current LPG-fueled vehicles may have CO emissions that are 40 to 50 percent lower than gasoline-fueled vehicles. Optimized LPG-fueled vehicles are expected to have emissions that are similar to the emissions from gasoline-fueled ULEVs. These estimates, however, are for operation at moderate temperatures; at lower temperatures (where CO emissions tend to be the highest), the reductions should be even greater due to the high volatility of LPG. This is important because CO nonattainment is primarily a problem during cold weather. The analysis in Chapter 2 indicates that CO emissions from LPG-fueled vehicles at low temperatures could be as low as 5 to 10 percent of the amount emitted by gasoline vehicles. Thus, optimized LPG vehicles should provide CO benefits at low temperatures, even when compared to gasoline-fueled vehicles that were also designed to comply with the ULEV standards. Based on the available data, it is estimated that LPG-fueled vehicles will average 80 percent lower emissions at low temperatures than gasoline-fueled vehicles. This estimate is intended to be somewhat conservative; however, it is important to remember that since the amount of low temperature data available is very limited, it is possible that actual emissions benefits may be less than this. Nevertheless, it is clear that the potential for lower CO emissions is very significant, and may even be the most significant air quality benefit provided by LPG-fueled vehicles. Air Toxics LPG is expected to provide significant benefits with respect to air toxics. EPA has presented its estimate of the impact of conventional mobile sources on air toxics in great detail previously.[2] The analysis in this chapter uses the estimates of cancer incidences caused by light-duty vehicle emissions from that work to provide a perspective on the toxics impacts due to light-duty LPG vehicles. It then develops an estimate of the per-vehicle reductions expected to result from LPG use. The following mobile source-related toxic pollutants were examined: benzene (including exhaust, evaporative, running loss, resting loss, and refueling benzene), acetaldehyde, 37 ------- EPA Office of Mobile Source* Special Report 1,3-butadiene, and formaldehyde (direct and indirect8). These pollutants are emitted by gasoline-fueled vehicles and are classified by EPA as either known or probable human carcinogens. The numbers of cancer cases in 1990 and 2010 due to emissions from light-duty gasoline-fueled vehicles were estimated to be 423 and 264 per year, respectively. The per-vehicle air toxic reductions for LPG vehicles were calculated by comparing the toxic emissions of LPG vehicles to those of gasoline-fueled vehicles. This was straightforward for the fuel system emissions (evaporative, refueling, and running loss emissions) of benzene, since LPQ fuel contains little or no benzene and IRQ-fueled vehicles should not have significant fuel system emissions. The exhaust emissions for LPG-fueled vehicles were based on very limited speciated hydrocarbon data found in references 7,8 and 9. The indirect formaldehyde impact was assumed proportional to NMHC reductions (from lower fuel system emissions). Emissions rates for gasoline-fueled vehicles are from reference 2. Exhaust emissions fractions are summarized in Table 4-2. Table 4-2 Toxic Exhaust Emission Rates Relative to NMHC Emissions Toxic Benzene 1 ,3-Butadiene Acetaldehyde Formaldehyde Percent of NMHC (Gasoline) 4.0 1.1 0.8 2.0 Percent of NMHC (LPG) 0.1 0.01 0.5 1.0 Table 4-3 shows the relative reductions in toxic emissions from LPG vehicles compared to future gasoline-fueled vehicles, for both current technology and for future optimized technology. In most cases, the LPG vehicles show major reductions in air toxics, especially for benzene and 1,3-butadiene. The only cases in which LPG-fueled vehicles are not expected to result in reductions are acetaldehyde and direct formaldehyde, but this is only when comparing an LPG-fueled ULEV to a gasoline- fueled ULEV. Also, these two compounds make a relatively small contribution to the 8 Direct formaldehyde is emitted in the exhaust of vehicles, while indirect formaldehyde is formed in the atmosphere from the reactions of various reactive hydrocarbons. Indirect formaldehyde is responsible for the majority of the formaldehyde in ambient air.[6] 38 ------- EPA Offlc* of MoM* Source* Special Report overall risk. When all the emissions are considered, the LPG vehicles are expected to have approximately 92 percent lower toxic emissions (weighted by number of incidences) than gasoline-fueled vehicles. Table 4-3 Air Toxics Impacts of LPG-fueled Vehicles Toxic Exhaust Benzene Other Benzene 1.3-Butadene Acetaldehyde Direct Formaldehyde Indirect Formaldehyde Total/ Weighted Average Vehicle-Related Cancer Incidence 1990 2010 63 7 304 5.3 11 33 423 31.5 3.5 204 3.0 5.5 16.5 264 Per-Vehide Percent Reduction Current Optimized LPG-Fueled LPG-Fueled Vehicles Vehicles' 98 100 99 38 50 41 92 98 100 98 -25** 0 79 92 * Reductions for optimized LPG-fueled vehicles are based on a comparison to gasoline-fueled ULEVs. •• Acetaldehyde toxic impact of an LPG-fueled ULEV is estimated to be 25 percent higher than that of a gasoline-fueled ULEV. The overall impact of these per-vehicle reductions depends on the fraction of the fleet that is eventually replaced by LPG. Since, at this time, it is very difficult to predict overall penetration, this analysis is limited to per-vehicle reductions. Also, it should be reemphasized that these predictions are based on low-mileage emissions from a limited number of vehicles, and the analysis is very sensitive to changes in the emissions fractions of the toxic compounds. The emissions fractions are very dependent on the speciation of the LPG fuel, especially the 1,3-butadiene emission fraction, which would increase with increasing butane content in the fuel. Nevertheless, while the actual in-use impacts could be somewhat different from the estimates given here, the directional trends should remain valid. 39 ------- EPA Office of Mobil* Sourcn Special Report Global Wanning Recently, the greenhouse effect (i.e., the effect of emissions of certain "greenhouse" gases, most notably carbon dioxide (CO2), on global temperatures) has been receiving a great deal of attention. Since combustion of different fossil fuels can result in different CO2 emission levels, it is appropriate that the analysis of the environmental impact of LPG vehicles include its CO2 impact. Different fuels can also result in different emissions of other greenhouse gases, such as methane. In this analysis, however, it is assumed that emissions of all other greenhouse gases from LPG-fueled vehicles will be very similar to those from gasoline-fueled vehicles, with the exception of CO. However, because the mass of CO emitted from modem vehicles is very small compared to the mass of CO2 emitted, and because the impact of CO emissions is less than an order of magnitude higher than that of CO2 per unit mass,[10] the impact of the lower CO emissions of LPG-fueled vehicles is negligible for this global warming analysis. The estimated CO2 emissions of LPG vehicles used in this analysis are based upon the fuel economy estimates described in Chapter 2. The CO2 emissions for gasoline-fueled vehicles is based on MOBILE4 fuel consumption estimates. As was done for the other analyses in this chapter, the global warming analysis is based on emissions from passenger cars. While emissions of NMHC, CO, and NOx are not expected to be very different for cars and trucks because they have similar emissions standards, fuel consumption rates are quite different because trucks generally are larger and have larger, higher output engines. Thus, it would not be appropriate to compare the CO2 emissions from an LPG-fueled truck to those from a gasoline-fueled passenger car, or vice versa. This analysis is structured, however, so that it provides estimates of the reductions in CO2 emissions on a percentage basis, which can be applied to other sizes of vehicles. The total effect of the use of any fuel on global warming also depends on the secondary emissions of CO2 that occur during the production and distribution of the fuel. The energy consumption at all stages of production and distribution can be converted to equivalent CO2 emissions and added to the vehicular emissions of CO2 and methane. These effects were analyzed previously in a draft EPA report.[10] There it was calculated that the ratio of secondary CO2 to vehicular CO2 emissions for both LPG and gasoline vehicles is about 1:2.5. / The equivalent CO2 emissions of both gasoline- and LPG-fueled vehicles are shown in Table 4-4. The results show that LPG-fueled vehicles could have significant greenhouse gas reductions, even at current fuel economy levels. Optimized LPG- fueled vehicles could have 19 percent lower CO2 emissions than gasoline-fueled vehicles. 40 ------- EPA Office at MobM Sourest SfMda! R«port Table 4-4 Equivalent CO2 Emissions Vehicle Gasoline LDV Current LPQ LDV Optimized LPGLDV Fuel Economy (MPG)' 20.5 14.3 15.7 Vehicle CO2 (g/mile) 436 390 354 Secondary COj (g/mile) 175 156 142 Total Equivalent CO, (g/mile) 611 546 496 Percent Reduction -— 11 19 Volumetric Fuel Economy (miles per gallon of LPQ for LPG-fueled vehicles) Summary The preceding analyses show that LPG-fueled vehicles have the potential to provide significant environmental benefits in the areas of urban ozone, carbon monoxide, air toxics, and global warming. These potential benefits are summarized in Table 4-5. It must be emphasized, however, that the accuracy of these estimated benefits is limited by the amount of data available, especially in the area of emissions control system durability. Given the magnitude of the estimated benefits, however, it is certain that LPG-fueled vehicles have a tremendous potential to provide significant environmental benefits, especially in the area of CO emissions. Table 4-5 Summary of Environmental Benefits of LPG-Fueled Vehicles Pollutant NMHC NOx CO Air Toxics CO2 Percent Reduction From Current LPG-Fueled Vehicles 41 0 80 92 11 Percent Reduction From Optimized LPG-Fueled Vehicles 85 0 80 92 19 41 ------- EPA Office of Mobil* Sources Special Report References 1. "Proposed Reactivity Adjustment Factors for Transitional Low-Emission Vehicles," Technical Support Document, California Air Resources Board, September 1991. 2. "Mobile Source-Related Air Toxics Study, 1993," EPA-420-R-93-005,1993, (PB 93-182590). 3. MOBILE4, EPA Mobile Source Emission Factor Model. 4. "Definition of a Low-Emission Motor Vehicle in Compliance with the Mandates of Health and Safety Code Section 39037.65 (Assembly Bill 234, Leonard, 1987)', California Air Resources Board, May 1989. 5. "National Air Quality and Emissions Trends Report, 1989," EPA-450/4-91 -003, February 1991, (PB 91 -172247). 6. "Standards for Emissions From Methanol-Fueled Motor Vehicles and Motor Vehicle Engines; Final Rule," 54 FR 14426, April 11, 1989. 7. Data from Bob Myers, LP Gas Clean Fuels Coalition, October 1991. 8. Rideout, G., "Comparison of LPG & Gasoline Fueled Emissions from a Light- Duty Truck," Environment Canada, July 1991. 9. "Clean Fuel Fleet Emission Standards, Conversions, and General Provisions and Amended Heavy-Duty Averaging, Banking, and Trading Credit Accounting Regulations," Notice of Proposed Rulemaking, 58 FR 32474, June 10, 1993. 10. "Environmental and Economic Study of Alternate Motor Fuel Use: Report to Congress," EPA Office of Mobile Sources, Draft, April 1991. 42 ------- EPA Office of Mobil* Source* Special Report Chapter 5 Conclusions Use of LPG as a Vehicle Fuel There are a significant number of LPQ-fueled vehicles in use today in the United States, and there are even more vehicles in use in Canada, Europe and Japan. The experience with these vehicles has shown them to be both reliable and safe. Only recently, however, have LPG-fueled vehicles begun to be optimized with respect to emissions. Thus, it is expected that future LPG-fueled vehicles will be very different from current LPG-fueled vehicles. Current LPG-fueled vehicles are generally converted gasoline-fueled vehicles, and have been used in fleet applications due to the potentially lower fuel costs of LPG. Often, these vehicles retain the ability to operate on gasoline when LPG is not available. Future LPG-fueled vehicles are also expected to be used in fleet applications, but are expected to be dedicated vehicles that are optimized for both low emissions and low fuel consumption. The amount of LPG that will be available for use as a vehicle fuel will limit the number of LPG-fueled vehicles. It appears that, with sufficient demand and some change in the composition of LPG motor fuel, up to one million barrels per day of LPG could be available between 2000 and 2010. This would be enough fuel to allow LPG- fueled vehicles to travel more than 600 million miles per day. (This amount would be about 13 percent of the total number of miles travelled by the entire U.S. vehicle fleet in 1990.) However, actual market penetration of LPG-fueled vehicles will probably not be this high, because there would probably be a very significant increase in price that would be associated with increasing supply by this much. Given the limited supply of LPG, it appears that the most appropriate use of LPG-fueled vehicles would be for fleets in specific geographic areas, especially those areas that have problems with high ambient concentrations of CO during cold weather (see "Benefits' section below). Based on the current fuel market, broader use is possible, but may not be economical. 43 ------- EPA Office ot Mobil* SOUICM Special Report Benefits LPG-fueled vehicles have the potential to provide significant environmental and economic benefits. Estimates of these benefits, which were discussed in the body of this report, are summarized below. Economic - Based on current fuel prices, fuel costs for current LPG-fueled vehicles should be about the same as for gasoline-fueled vehicles, while optimized vehicle fuel costs should be slightly lower than for gasoline-fueled vehicles. The estimated benefit would be up to 0.9 cents per mile for optimized average-sized vehicles, and up to 1.7 cents per mile for larger vehicles (i.e., gasoline-fueled vehicle fuel economy of 10 MPG). These estimates are highly dependent on the relative prices of LPG and crude oil, and significantly dependent on the thermal efficiencies assumed in Chapter 2. Even with slightly lower LPG prices, the cost of a current conversion would more than offset the potential benefit. Perhaps the most significant benefit will be lower engine maintenance costs and/or prolonged engine life, though there currently is not enough information to characterize this potential benefit. Ozone - Exhaust emissions of NMHC and NOx from LPG-fueled vehicles should be essentially the same as those from gasoline-fueled vehicles. Evaporative emissions should be negligible from LPG-fueled vehicles. Overall, LPG-fueled vehicles should have 41 to 79 percent less NMHC emissions than gasoline. This estimate should be fairly accurate, though it is dependent on the manner in which LPG-fueled engines are calibrated, and on the extent to which evaporative emissions are controlled. The impact on ambient ozone levels would vary from city to city, but should be substantial in any area where a significant fraction of the vehicle fleet uses LPG. Carbon Monoxide - CO emissions from LPG-fueled vehicles at low temperatures (most CO problems involve cold weather) should be much lower than those from gasoline-fueled vehicles. The limited amount of low temperature data that is available suggests that CO emissions could be about 80 percent less than from gasoline-fueled vehicles. However, because the amount of low temperature data that is currently available is small, there is a great deal of uncertainty associated with this estimate. Nevertheless, this benefit is probably the most important air quality benefit of LPG-fueled vehicles. 44 ------- EPA Office o« Mottle Sources Special Report Toxics - The analysis presented in this report suggests that LPG-fueled vehicles may have much lower emissions of toxic compounds. This analysis, however, is based on very limited speciated emissions data from LPG-fueled vehicles. Further, the analysis is very sensitive to small changes in emission estimates. Therefore, the numerical estimates of the impact of LPG-fueled vehicles on toxics should be used very cautiously. The analysis is adequate, however, to demonstrate that LPG-fueled vehicles should provide a net benefit with respect to air toxics. Global Wanning - LPG-fueled vehicles are expected to provide small but significant benefits with respect to greenhouse gas emissions. It is estimated that CO2 emissions from LPG-fueled vehicles will be 11 to 19 percent lower than those from gasoline-fueled vehicles. These benefits result from both the lower amount of CO2 produced per BTU consumed, and the expected improvement in fuel efficiency, relative to gasoline. These estimates should be reasonably accurate, though the upper range relies on projections of future technologies. Need for Additional Studies For gasoline-fueled vehicles, there is a very large amount of data available, including speciated data, low temperature data, and high mileage emissions data. This is not true for LPG-fueled vehicles. While this lack of data generally limited the analyses throughout this report, it was critically limiting in several of them. First, the economics of LPG use may be greatly affected by prolonged engine life or lower maintenance costs for LPG-fueled vehicles. At this time, there is no comprehensive and reliable information that would allow the potential economic benefits of prolonged engine life to be quantified. There is a need for a study in which there are large numbers of LPG- and gasoline-fueled vehicles, operating in the same manner, and perhaps receiving identical maintenance. The study would need to provide detailed records of total vehicle costs, including maintenance and vehicle replacement. Second, there is a need for side-by-side emissions data from LPG- and gasoline-fueled vehicles under identical operating conditions. The vehicles should be as similar as possible, considering such factors as model, power output of the engines, and equipment (e.g., air conditioning, transmission). The testing should be conducted at low, intermediate and high mileages. Some studies like this have been initiated, and should provide much needed information about the emissions and fuel economy of LPG-fueled vehicles relative to gasoline-fueled vehicles, as well as about the durability of LPG-fueled vehicle emission control systems. 45 ------- EPA Otfic* o( Motote SOUICM Special Report Finally, there is a need for more speciated and low temperature emissions data. The speciated data would allow a better characterization of ozone reactivities and toxics impacts, while the low temperature data would allow a more reliable estimate of the benefit of LPG-fueled vehicles with respect to CO emissions. 46 ------- EPA Office of Mobile Sources Special Report Appendix I Data from Sears Fleet From: 'Summary Report for Emission Testing of Liquified Petroleum Gas Fueled Vehicles Provided by Sears," California Air Resources Board. ------- TABLE t FEDERAL TEST PROCEDURE RESULTS TEST VEH. NO. Lt kt 1,3 L4 L5 L6 L7 L8 L9 y§ L11 L12 L13 L14 L15 SEARS VEH. NO. NA NA NA NA NA NA NA NA NA 9777 1214 1154 60529 •519 •499 MILEAGE 15949 55549 48666 26115 4405 69948 75547 36688 37596 29727 52791 50817 17946 107145 69258 POST- RETRO. .MILEAGE -15.649 -49 -46,644 7 7 -59.946 -65.547 -28.669 •27.596 -17.727 -32.791 -39.817 -14.946 •70.145 •59.258 VEH. MFR. G.M. Chry*. Nl**. G.M. Chry* . Chry* . Chry* . Chry* . Chry* . Ford Ford Ford Ford Ford Ford MODEL NAME Chov. Oodg* Cho'yonn* Ham 5* Dot*un Pickup Ch*v. Dodg* Oodg* Oodg* Oodg* Dodg* Econe 1 Econol Econol Ecenol Econol Econol Coaoro Dakota Ra« 256 Ro« 259 Ra« 250 ROM 259 In* 159 In* 139 In* 159 In* 159 In* 259 In* 25* MODEL ENG. YEAR DISPL 1988 1964 1980 1979 1989 1966 1988 1989 1989 1967 1984 1964 1988 1979 1979 395 eld 155 eld 119 eld 350 eld 239 eld 318 old 318 eld 316 eld 318 eld 390 eld 300 eld 300 eld 300 eld 302 eld 302 eld FEDERAL TEST PROC. RESULTS* (CVS-7S) (CRAMS/MILE) HC NMHC CO NOx 0.24 0.99 0.26 0.68 0.69 1.62 1.28 0.60 1.19 1.96 V72 • .84 • .95 3.54 3.73 0.19 • .66 0.15 0.72 0.41 0.99 1.28 0.90 0.91 1.95 1.47 • .71 • .86 3.25 3.27 • .89 38.30 7.57 46.67 2.66 57.36 19.33 11.72 2.66 1.11 60.29 1.58 • .98 11.93 291.64 1.05 0.72 1.28 1.72 0.52 0.32 0.67 0.58 1.33 1.30 0.92 2.39 1.75 9.65 0.25 APPLIC. STDS. FOR VEM. (GRAMS/MILE) HC NMHC CO NOx 0.5 0.41 0.41 9.41 9.41 9.5 9.5 0.5 0.5 0.5 0.5 0.5 0.5 0.9 9.9 0.5 0.39 0.39 NA 9.39 0.3 9.5 0.5 0.9 0.5 0.5 0.5 0.5 NA NA 9 9 9 9 9 9 9 9 9 9 9 9 9 17 17 1 .0 1 .0 1 .5 1 .5 1 .0 1.0 ! 1.0 1 t.o » 1.0 { 1.0 £ t.e | 16 f ,., | «.3 I 2.3 HC - Hydrocarbon* NMHC - Non-«*than* hydrocarbon* CO - Carbon Monoxld* NOx - Oxld«« of nitrogen ------- TABLE 2 ORIGINAL EQUIPMENT INSPECTION TEST VEH. NO. NA NA NA NA NA NA NA NA L11 NA NA L12 NA NA NA NA NA NA NA Lit NA L13 NA L14 L15 SEARS LICENSE VEHICLE NUMBER NUMBER 0238 2779 6541 0303 0363 0403 0423 1044 1214 1024 1034 1154 9615 •625 0635 1636 1666 1676 •617 •777 00522 00520 00523 051» 0409 1M62370 1S03405 1Z01136 2037707 2020445 2037706 2037715 2J06S64 2L33004 2J06563 2J06S62 2L33011 2N05576 2N05575 2L04920 2V40775 2V48750 2V40771 3F97429 3F07430 3M01067 3M01050 3M01070 1P51106 1P52106 MODEL YEAR 1078 1079 1981 1963 1983 1983 1983 1984 1984 1984 1984 1984 1985 1965 1985 1986 1986 1966 1987 1987 1988 1966 1968 1979 1979 VEH. MFR. G.M. O.M. Ford Ford Ford Ford Ford Ford Ford Ford Ford Ford Ford Ford Ford Ford Ford Ford Ford Ford Ford Ford Ford Ford Ford MODEL NAME Chevrbl et Chevrolet Eeonol In* Eoonol ino Eoonol Ino Eoonol Ino Eoonol Ino Eoonol Ino Eoonol Ino Eoonol Ino Eeonol In* Eoonol Ino Eeonol Ino Eoonol Ino Eoonol Ino Eeonol Ino Eoonol Ino Eeonol Ino Eeonol Ino Eoonol Ino Eeonol Ino Eeonol Ino Eeonol Ino Eeonol Ino Eeonol Ino Von Von 150 150 150 150 150 150 150 ISO 150 130 150 150 150 150 150 150 150 ISO 150 150 150 250 250 MILEAGE 100311 04636 76013 46503 00001 75450 00065 51273 52701 40007 56388 50806 58522 45607 46221 36023 41647 37163 36201 20734 10516 17056 21060 107156 00258 EMISSION CONTROL EQUIPMENT AIR. AIR. AIR. AIR. AIR. AIR. AIR. AIR. AIR. AIR. AIR. AIR. AIR. AIR. AIR. AIR. AIR. AIR. AIR. AIR. AIR. AIR. AIR. AIR. AIR. ECR. OC ECR. OC ECR. ECS. ECR. OC EGR. OC ECR. OC ECR. OC ECR. ECS. ECR. ECS. ECR. ECS. ECR. ECS. ECR. ECS. EGR. ECS. ECR. ECS. ECR. EOS. EGR. EOS. ECR. EOS. EGR. EOS. EOR. ECS. ECR. EOS. ECR. EOS. ECR. EOS. ECR. EOS. ECR. OC EOR. OC TVC. TVC. TVC. TVC. TVC. TVC. TVC. TNC. TVC. TVC. TVC. TVC. TVC. TVC. TVC. TVC. TVC. CL CL CL CL CL CL CL CL CL CL CL CL Fl. FI. FI. FI. FI. CL CL OBD. CL OBD. CL OBD. CL I j? i Acronyms for emission control components: \ AIR - Secondary air Injection CL - Cloeod loop ECR - Exhouet aoe roe IrculatIon ECS - Oxygen tensor FI - Fuel Injection OBO - On-board diagnostics OC - Oxidation catalyet TVC - Three-way catalyst ------- TABLE 3 RETROFIT EQUIPMENT INSPECTION TEST VEH. NO. NA NA NA NA NA NA NA NA L11 NA NA L12 NA NA NA NA NA NA NA L19 NA L13 NA L14 L15 SEARS VEHICLE NUMBER 9236 2779 9541 9393 •363 •493 •423 1044 1214 1024 1934 1154 9915 9925 9935 1959 1999 1979 9917 9777 \ 86522 68529 99523 9519 9499 RETRO. KIT MFR. lapoo lapoo lapoo lapoo lapeo lapoo lapoo lapoo lapoo lapco lapco lapco lapco lapco lapoo lapco lapco lapco lapco lapco lapoo \ lapco lapco lapco lapco RETROFIT DATE INSTALLER RETROFIT INSTALLED Potrolono Potrolono Potrolono Potrolono Potrolano Potrolano Potrolano Potrolano Potrolano 7 Potrolano Potrolano Potrolano Potrolano Potrolano Potrolano Potrolano Potrolano Potrolano Potrolano Potrolano Patrolano Potrolono 7 Potrolano 7 7/67 4/67 8/67 5/67 12/66 3/67 4/67 11/67 7 5/67 4/67 8/66 1/66 12/86 5/67 3/67 7/67 19/99 5/66 8/86 16/68 7/66 7 12/83 CLOSED FUEL LOOP CONTI CONV.T PROC N N N N N N N Y Y Y Y Y Y V Y Y Y Y Y Y Y Y Y N N N N N N N N ^ N N N N N Y Y V Y Y FCP FUEL FUEL CONTROL MIRING CONTROL ENRICH. VALVE7 SETTING? 7 7 75 X 7 7 7 7 Y Y 7 NY 7 Y Y 7 Y Y 7 NY 7 Y Y 7 Y Y 7 V Y 7 Y Y 7 Y Y 7 Y Y 7 Y Y 48X Y Y 6X Y Y SOX N Y ex Y Y 25X 7 7 *V«hlcl«* with N In Fu«l Control Proc«««or coluan u*«d th« OEM processor for fuel control ------- TABLE 4 VEHICLE OPERATION DIAGNOSIS TEST VEH. MO. NA MA MA MA NA NA •NA NA in NA NA L12 NA MA NA NA NA NA NA Lit NA L13 NA 114 na SCARS STRONG VEHICLE SMELL OF NUMBER PROPANE? 6238 2779 •341 •393 •383 •403 •423 1944 1214 1924 1*34 1154 9613 992S 9935 1959 1989 1979 9917 9777 99322 99329 99323 9319 9499 V N V N N N N N N N V V N M V V M y Y V V V V Y Y •HERE DOES VEHICLE WARN. PROPANE SMELL OPERATING IN LIGHT COME FROMT CLOSED LOOP? ON? R«0wl«t*r Regulator Exhaust. rag. Regulator Rag., cold exhouat Exhaust Rag., air cleaner Rag. . ol r claanar Cxhauet and earb. Regulator Regulator Regulator Regulator Y N ? ? Y N N Y Y Y 7 Y Y 7 N N N N N N N N N N N N N N N N N N N N N N N N N N N FCV IF SOLNO. NOT EGR EGR SOLNO. WORKING. IS IT SYSTEM PRE •ORK? REC. SIGNAL OK? OK? Y N N Y Y Y N Y »S * V Y Y Y V V Y N N N Y Y Y Y Y N Y Y V Y Y Y N Y V Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y N N Y N 7 N N Y Y Y N N N Y Y Y Y N N Y PCV RICH OR SYSTEM LEAN OK? OPERATION? Y N Y N Y Y N Y Y N Y N Y Y V Y N Y Y Y N Y Y Y Y OK OK OK Lean 7 7 OK OK Rich Rich Lean OK OK Leon OK 7 Lean Rich OK OK Rich OK Rich Lean OK <5 I • f I ------- EPA Offlct <* Mobito SOUICM Sp«cfel Report Appendix II LPG Emissions Data From Late Model Vehicles Tested Using the Federal Test Procedure ------- EPA Office of Mobile Sources Special Report Current LPG-fueled Vehicle Emissions 1 2 3 4 5 6 7 8 9 . 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Vehicle 1985VWJETTA 1988 CHEYENNE 1989 DODGE DAKOTA 1988 DODGE RAM 250 1988 DODGE RAM 250 1989 DODGE RAM 250 1989 DODGE RAM 250 1987 FORD ECONOLINE 1988 FORD ECONOLINE 1988 CHEW PICKUP 1990 DODGE VAN 1988 FORD CROWN VICTORIA' 1990 DODGE DAKOTA' 1989 CHEVY PICKUP' 1990 FORD 1 -TON TRUCK* 1989 PLYMOUTH K-CAR 1989 PLYMOUTH K-CAR 1989 PLYMOUTH K-CAR 1989 POND AC 6000 1989 OLDS DELTA 88 1988 FORD CROWN VICTORIA 1989 OLDS DELTA 88 HC 0.60 0.24 0.69 1.62 1.28 0.90 1.19 1.06 0.95 0.239 0.405 0.27 0.45 0.37 0.93 0.19 0.276 0.128 0.168 0.117 0.268 0.156 1990 PONT! AC 6000 I 0.141 1990BUICKLESABRE | 0.19 NMHC 0.19 0.41 0.90 0.91 1.05 0.86 0.179 0.340 0.131 0.061 0.231 0.111 0.101 CO 1.8 0.89 2.88 57.38 19.33 11.72 2.66 1.11 0.98 2.426 0.897 0.28 0.37 3.3 5.8 0.877 2.403 2.103 0.912 2.659 0.495 2.68 0.72 2.56 NOx 0.76 1.05 0.52 0.32 0.67 0.58 1.33 1.30 1.75 0.23 0.55 0.60 1.7 0.7 4.7 0.32 0.299 0.241 0.216 0.102 0.39 / 0.108 0.292 0.34 Reference [1] [2] [2] [2] [2] [2] [2] [2] [2] [3J [4] [5] [5] [5] [5] [6] [6] [6] [6] m [8] [9] [9] [10] DATA INuLUUtU IN IADLC £.-£. ------- EPA Offic* ol Mobito Source* Special Report Current LPG-fueled Vehicle Emissions 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Vehicle 1985 VWJETTA 1988 CHEYENNE 1989 DODGE DAKOTA 1988 DODGE RAM 250 1988 DODGE RAM 250 1989 DODGE RAM 250 1989 DODGE RAM 250 1987 FORD ECONOUNE 1988 FORD ECONOUNE 1988 CHEW PICKUP 1990 DODGE VAN 1988 FORD CROWN VICTORIA 1990 DODGE DAKOTA 1989 CHEVY PICKUP 1990 FORD 1-TON TRUCK 1 989 PLYMOUTH K- CAR 1989 PLYMOUTH K- CAR 1988 PLYMOUTH K- CAR 1989 PONTIAC 6000 1989 OLDS DELTA 88 1988 FORD CROWN VICTORIA 1989 OLDS DELTA 88 1990 PONTIAC 6000 1990BUICKLESABRE C02 522 578 398 379 485 907 377 398 411 431 389 Odometer 15940 4405 69048 75547 38680 37596 20727 17496 13500 3590 7541 4253 47417 4006 14622 31118 22658 Fuel Metering TBI R PR TBI TBI PR SPI R CARB. Engine Displ. 1.8L 305CI 239CI 31 SCI 318CI 31 SCI 31 SCI 300CI 5.7 318CI 5.0L 3.9L 5.7L 7.5L 2.2L 2.2L 2.2L 2JBL 3.8L 5.0L 3.8L Dual- Fueled? NO YES YES YES YES YES YES YES YES YES YES ------- EPA Offle* of Mobil* Sources Special Report References 1. Menrad, H., R. Wegener, and H. Loeck, "An LPG-Optimized Engine-Vehicle Design," SAE #852071, October 1985. 2. "Summary Report for Emission Testing of Liquified Petroleum Gas Fueled Vehicles Provided by Sears," California Air Resources Board. 3. Johnson, T., "Preliminary LPG Emissions Report," IMPCO, July 1988. 4. Data from Bryan Memmott, Cleanfuels, October 1991. 5. Data from Ron Ragazzi, Colorado Department of Health, February 1991. 6. Data from Gordon Lilley, IMPCO, August 1990. 7. Data from Bob Myers, LP Gas Clean Fuels Coalition, January 1991. 8. Bruetsch, R.I., Letter to James Burroughs, April 1989. 9. Data from Bob Myers, LP Gas Clean Fuels Coalition, October 1991. 10. Hilden, D.L., P.A. Mulawa, and S.H. Cadle, "Liquefied Petroleum Gas as an Automotive Fuel: Exhaust Emissions and their Atmospheric Reactivity," General Motors Research Publication GMR-7442, September 9,1991. ------- EPA Office of Mobil* SOUICM Special Report Appendix III Dual Fuel Gasoline Emissions Data Tested Using the Federal Test Procedure ------- EPA Offlc* at MoM« SourcM Special Report Current LPG-fueled Vehicle Emissions Operated Using Gasoline 10 11 16 17 18 20 21 24 Vehicle 1988 CHEVY PICKUP 1990 DODGE VAN 1989 PLYMOUTH K-CAR 1989 PLYMOUTH K-CAR 1989 PLYMOUTH K-CAR 1989 OLDS DELTA 88 1988 FORD CROWN VICTORIA 1990BUICKLESABRE HC 0.38 0.34 0.38 0.33 0.37 0.25 0.21 0.14 NMHC 0.34 0.24 0.15 ~0 3.07 2.39 7.13 8.7 7.75 3.20 0.68 2.65 NOx 0.76 1.02 1.22 0.49 0.48 0.26 0.68 0.65 CO, 594 656 491 446 ------- EPA Office of MOM* Sourcm Special Report Appendix IV Low Temperature LPG Emissions Data Tested Using the Federal Test Procedure From: Q. Rideout, 'Effect of Ambient (Cold) Temperatures on Exhaust Emissions and Fuel Consumption of LPG Fueled Vehicles,' Environment Canada, May 1991. ------- EPA Offlo* at Mot* SouroM SMC* Report 8.0 RESULTS AND DISCUSSION A summary of the composite emission rates of each of the vehicle/fuel configurations is presented in TabU* 5,6,7,8. In some cases multiple tests at the same temperature have been averaged for the purposes of this discussion, while the entire body of results are contained in the appendices. The average standard temperature test result for each configuration is presented for reference to the ambient temperature effect on the emission rate. The Ford F150 results for each of the test years are presented separately since there was a slight shift in the standard temperature emission rates from year 1 to year 2. TABLES. Composite Emission Rates: Chevrolet Silverado Ambieat Temperature T«mp«raUir« C Gaoolia* Fu«led 21.0 10.5 9.0 7.5 6.0 3.8 Propane Fueled 25.0 11.1 7.8 -2.1 Grams per mile THC 0.37 0.58 0.63 0.68 0.73 0.74 0.28 0.28 0.32 0.31 CO 2.78 10.84 9.89 10.37 11.98 12.54 4.33 5.14 4.81 4.12 NOX 0.36 0.94 1.11 1.11 1.10 1.25 0.32 0.31 0.38 0.60 C02 593.77 690.22 581.72 561.44 631.96 606.70 548.25 550.01 565.29 575.47 Fu«l Economy (mpg) 17.88 17.45 17.73 18.32 16.27 16.90 GBFB (mpg) 16.89 16.79 16.37 16.10 e ffKt of CoW T«mp»f«ruf« oo i motor* oftnifC futlfd V*h«/*-»t -OJ ------- EPA OfflM ol MoM* SOUICM Sc*etf Repot TABLE 6. Composite Emission Rates: Ford F150 Ambient Temperature* Year 1 Temperature C Guoline Fueled 25 t24 10 9 8.7 7.5 7.25 4.5 Propane Fueled 24 11.8 7.6 7.0 4.5 3.0 2.0 0.0 -2.7 Grams per mile THC 0.78 0.64 1.66 1.82 1.53 1.66 1.46 NA 1.09 1.02 1.17 1.12 1.19 1.21 1.36 1.20 1,37 CO 1.26 2.34 17.65 19.55 16.20 16.98 14.29 16.57 0.04 0.00 0.06 0.03 0.04 0.07 0.03 0.03 0.02 NOX 1.17 1.20 1.19 1.12 1.15 1.13 1.20 0.99 0.86 0.84 0.77 0.82 0.75 0.75 0.89 0.88 0.76 CO2 563.24 577.00 585.97 606.49 679.65 594.13 548.47 565.48 490.22 532.85 537.15 559.98 538.87 553.26 599.25 624.42 603.97 Fuel Economy (mpg) 18.72 17.65 17.16 16.66 17.42 16.97 17.33 17.92 GEPE (mpg) 19.02 17.54 17.36 16.69 17.33 16.86 15.56 17.78 15.46 T • Year 2 standard temperature result! e H»tl of Cold 7>mp»r«fur#i on f m-u/om of in I PC fu»l»d V«h*/»-»t-07 ------- EPA OfflM of MeM* SOUIVM Sptcid Report TABLE 7. Composite Emission Rates: Ford K150 Ambient Temperature • Year 2 Temperature C Propane Fueled 24 10.30 7.00 6.7 6.7 4 3.7 2.0 0.66 . 0.33 -.170 •2.30 -2.70 -3.0 -3.3 -3.3 -4.3 -4.3 -7.3 -12.3 -18.0 Grams per mile THC 1.16 1.06 1.14 1.19 1.14 1.22 1.12 1.23 1.30 1.27 1.07 1.27 1.26 1.27 1.28 1.19 1.32 1.26 1.06 1.14 1.29 CO 0.47 1.68 1.48 1.69 1.37 0.67 0.60 1.20 0.93 0.77 0.64 1.47 0.60 1.03 0.85 0.90 1.60 0.99 0.60 0.24 0.50 NOX 1.46 1.33 1.31 1.39 1.33 1.30 1.31 1.40 1.42 1.41 1.34 1.42 1.32 1.34 1.42 1.20 1.69 1.25 1.23 1.33 1.39 C02 495.05 465.40 497.80 501.10 525.40 509.90 50420 505.50 519.10 529.10 515.70 539.80 563.70 511.20 529.70 525.60 508.90 495.40 510.20 535.20 577.00 Fuel Economy (GEFE)mpg 18.80 19.92 18.64 18.50 17.68 18.24 18.46 18.37 17.90 17.58 18.06 17.26 16.51 18.17 17.55 17.69 18.21 18.75 18.25 17.42 16.14 £Hect of CoM Tempertturtt o« <• --woni of »n IPG fu*i*d Vehidf 91-02 ------- EPA Offlc* ol MoM* SOUICM SMOM R«pott TABLE 8. Composite Emission Rates: Ford Taurus Ambient Temperature Temperature C 1987 Ford Taurus 24.0 17.7 12.3 11.7 10.0 4.7 1.0 -2.7 Propane Fueled Ford Taurus 24.0 8.7 8.7 8 2.70 1.70 -2.70 -3.50 -4.00 -4.00 -5.70 -7.30 -10.00 -12.70 Grams per mile THC 0.24 0.31 0.38 0.37 0.49 0.47 0.66 1.06 0.16 0.23 0.21 0.21 0.22 0.20 0.26 ND 0.27 0.24 0.27 0.24 0.25 0.24 CO 1.70 1.34 1.62 1.94 2.21 2.83 2.77 6.89 1.71 2.10 2.16 2.06 3.09 2.07 2.77 ND 2.42 2.60 3.65 3.65 3.72 3.81 NOX 0.86 1.48 1.44 1.33 1.33 1.25 1.42 1.09 0.86 0.62 0.76 0.70 0.67 0.68 0.83 ND 1.06 0.86 0.90 1.03 1.06 1.07 C02 444.9 498.0 522.0 523.0 520.0 504.0 541.0 479.0 386.0 389.0 393.0 389.0 409.2 389.8 426.9 ND 438.9 408.0 395.4 391.4 424.8 420.1 Fuel Economy (mpg) 21.24 20.18 21.08 20.32 20.92 19.48 21.73 16.14 GEFE(mpg) 24.12 23.88 23.65 23.89 22.63 23.85 21.75 ND 21.16 22.74 23.37 23.59 21.76 21.99 INL): INO data available tHta of Cold r»mp«ufuf« onemauont ot»n LPG fu«/ed V«h«c/»-9f -02 ------- EPA Office of Mobile Sources Special Report Appendix V Heavy-Duty LPG Emissions Certification Data Below are California certification data for an IRQ-fueled 1993 Ford heavy-duty engine, and federal certification data for a comparable gasoline-fueled engine. LPG Engine size: 7.0 liters Rated power 218 HP @ 3600 RPM Emission controls: EGR, air injection Fuel metering: Carburetted HC (g/BHP-hr): 0.5 CO (g/BHP-hr): 23.2 NOx (g/BHP-hr): 2.8 Gasoline Engine size: 7.0 liters Rated power 235 HP @ 3600 RPM Emission controls: EGR, air injection, closed loop Fuel metering: Fuel injection HC (g/BHP-hr): 1.2 CO (g/BHP-hr): 21.5 NOx (g/BHP-hr): 4.4 ------- EPA Office o< Mobile Sources Special Report Appendix VI Heavy-Duty LPG Emissions Bus Chassis Data Below are gram/mile emissions data for a converted LPG-fueled bus (with and without catalyst) and for a comparable diesel bus. The buses were tested as part of the Orange County Transportation Authority's Cleaner Air through Reduced Emission Systems (OCTA CARES) program (December 1993). LPG Base engine: Maximum power (at wheels): Catalyst: HC (g/mi): CO (g/mi): NOx (g/mi): PM (g/mi): Base engine: Maximum power (at wheels): Catalyst: HC (g/mi): CO (g/mi): NOx (g/mi): PM (g/mi): Diesel Base engine: Maximum power (at wheels): Catalyst: HC (g/mi): CO (g/mi): NOx (g/mi): PM (g/mi): Cummins L10 190 HP Yes 0.67 0.02 3.04 0.03 Cummins L10 190 HP No 10.50 45.1 36.5 0.04 Cummins L10 230 HP No 1.24 13.5 31.7 1.23 ------- EPA Office of Mobile Sources Special Report Appendix VII Material Safety Data Sheet for LPG From: 'Propane and Environment in Ontario - A Clean Transportation Fuel Alternative,' Propane Gas Association of Canada, September, 1992. ------- EPA Offic. at MoM* SOUICM SMC* Report MATERIAL SAFETY DATA SHEET SECTION I — PRODUCT INFORMATION Product Nama: Proeana Trad* Nama: LPO (liquified Pttretaum Qaa) Chemical Formula: C*Ht Supplier: SECTION II —HAZARDOUS INQREDIENT8 COMPONfNTB CAS RKMtTHY *RQP:Oj|ngM' ic» Prapww EtfMIW 74888 74840 781088 78288 74828 3%- 5% 1%- 3% 0.1%'04% ai%-o2% Not Not Not Not Not Not Net SECTION III — CHEMICAL AND PHYSICAL DATA Povne WNto Pott* -42«C • 780 RMHo 10TC pec*ettre*ft;OJ1 (WATEP. -1) Vapour Denatty: 1J« (Mr • 1) ml of WMatfOl DlatiBiMUyn. Not »•» t ta WMOB 8.1% by VWMo • 17*C and —tat any ^. Ccmmartp.l preaoiK. has an ihaaoft «NU OM In natural wntehla SECTION IV - FW8 OR EXPLOSION HAZARD DATA Pomt-iea^c To Moot Or . of 8w aaoapinB QM 8M mad off. Pto OM D» «d dry latamdtolaakto knotonioMtio on to el SECTION V — REACTIVITY DATA Condlttona To Avoid: Kaop of ignMon and m to Wlnot ------- EPA Offlc* at Mow* SOUIQM SpKiat Riport SECTION VI — TOXICOLOGICAL PROPERTIES OF MATERIAL ACUTE EXPOSURE Eyes: As • gas. none. Liquid causes 'cold bum*". SUn: Liquid causes *eoM bums' similar to frostbite. Respiratory System: Little pnysiologicai effect at concentrations below 10.000 PPM. High* concentrations may cauaa dizziness and unconsciousness due to asphyxiation. ChronJe Exposure: There are no reported effects from tone-term towievel exposure. Other Liquid can cause bums and hosttoe • In direct contact with skin. SenslttxatioA Properoee; SUn — Unknown, Respiratory — Unknown C«iUnoo,enlcllr. Not determined. Reproductive Effects: Not determined MEDIAN LETHAL DOSE: Oral: Net appUcabta for gas. Inhalation: Not determined Dermec Not applicable tor gaa. Other. Not determined IRRITATION INDEX SUn: No appreciable effect (gas). Eyes: No appreciable effect (gait Symptoms Of Exposure: Above 100.000 PPM Dtulnees, Stupor; Unconsciousness. American Conference of Governmental Industrial Hygienists (ACQH> dinlfkn propane aa an asphyxiant there la no neam^ed ^hreahoU Umtt Value* (TLV% SECTION VII — OCCUPATIONAL CONTROL PROCEDURES Ey«s:Sa/etyc4a*ftea.QogQle*.<»tacetNetdr»c*«red when Iransf ernng product Skint Insulated gloves t contact with squid or squid I equipment la expected Inhaieoon: in _ mtislsd air. as* required :Exptce) where the concentration of leveibelow ia% to breatMng apparatus OUip SECTION VIII — EMERGENCY AND FIRST AID PROCEDURES FIRST AID: Eyes: Should eye contact with squid occur, flush eye* wtth lukewarm water for 16 minutes. Obtain Immeciate medcaics/*. SUn: In ease of •Cold BumT from contact with squat tmmeoiaterr place affected area miuhewann water and keep at this temperature untt etacunaon returns. • fingers or hands are frostbttan, nave the vtoftn he*) Ms hand nead to his body such as under the amp*. Obaun SPIU.OIILEAX! EiMnsleleakl eMurecyander is upright Disperse vapours wxh hose strsama using fog *><>T**^. watch lor low area* aa propane la heavier than air and can settle Mo low areas. Remain upwind el leak, keep end/or squid worn entering Into sewers* Ing* Into eVfleut or has stopped i to fresh ale. • breathing kj SECTION DC — TRANSPORTATION, HANDLING AND STORAGE anucrtCjMpeeJDonlna • CylndersmeiMnotlRuae not atom ytaecvtndi Tranepoft hanole and store aceon* ln • oxygen or TTOCattaffkaflart2.1 TOO Shipping Name Llqutled Petroleum Oas TOO SpecW ProvWonc Sfl, 90,103 UN7NA;107S SECTION X — PREPARATION INFORMATION Pfepeired toys i Mr* ------- LPG-Fueled Nonroad Equipment Emission Factors Listed below are estimated emissions of hydrocarbons (HC), carbon monoxide (CO), oxides of nitrogen (NOx), participate matter (PM), aldehydes (Aid.), and oxides of sulfur (SOx) from LPG-fueled equipment. These estimates were developed from emission factors from gasoline-fueled equipment, based on the testing of two converted LPG-fueled engines. (Source: EPA 21A-2001, "Nonroad Engine and Vehicle Emission Study - Appendixes") Exhaust Emissions (g/hp-hr) Pumps <50hp Compressors <50hp Aerial Lifts, Forklifts, Sweepers/Scrubbers, Terminal Tractors HC 9.0 6.4 4.5 CO 215 147 83 NOx 2.8 7.0 17.9 PM 0.2 0.1 0.1 Aid. 0.2 0.2 0.2 SOx 0.0 0.0 0.0 Other HC Emissions Pumps and Compressors <50hp Aerial Lifts, Forklifts, Sweepers/Scrubbers Terminal Tractors Crankcase (g/hp-hr 1.4 1.5 1.0 Evaporative (g/day) 2 55 17 Refueling (g/hp-hr) 6.3 0.5 0.5 August 1995 ------- |