3. Energy and CO2 Emissions Impacts of Fuel Cell Vehicles

This chapter examines the potential impacts of FCVs on energy demand and full fuel cycle CO2 emissions under a variety of scenarios for: (1) new vehicle market penetration, (2) vehicle fuel economy improvement, (3) sources of hydrogen supply, and (4) transition from distributed to centralized production. The analysis and results presented in this chapter reflect the assumptions made to illustrate the impacts of scenarios where the challenges facing a hydrogen economy are overcome. They are not intended to endorse, support, or imply plausibility or likelihood. This analysis serves to demonstrate the relative time frame and significance of energy and CO2 impacts, given assumptions regarding FCV market penetration, FCV fuel economy, hydrogen feedstocks, and hydrogen production methods.

The VISION model was selected to examine the various fuel cell cases, because the time frame necessary to observe impacts extends beyond the NEMS time frame.⁶³ In addition, use of NEMS would have required the development of very specific assumptions about the timing and success of FCV research and development, hydrogen production and infrastructure development, and the companion Federal and State policies that are likely to be needed to ensure the successful development of hydrogen-powered FCVs within the next 10 to 20 years. To generate the reference case used in this analysis, the VISION 2007 AEO Base Case Expanded Model was updated to reflect the projections of LDV sales, stocks, travel, and fuel economy in the AEO2008 reference case.

Fuel Cell Vehicle Market Penetration Scenarios

Three FCV market penetration scenarios were examined, based on studies completed by DOE, Oak Ridge National Laboratory (ORNL), and the National Research Council (NRC). The market penetration scenarios represent shares of new vehicle sales through 2050 and are taken from studies and reports that have assumed different levels of success in meeting FCV research, development, and cost goals, as well as capital investments needed to produce the vehicles and provide the necessary hydrogen fueling infrastructure. Those reports have determined that a successful transition to hydrogen-powered light-duty FCVs is likely to require some type of policy incentive to stimulate initial investments in the technology, as well as Federal and/or State financial incentives or mandates that significantly reduce the financial risk of investments in vehicle production and infrastructure development.

The first, and least aggressive, market penetration scenario examined in this report is derived from DOE’s Office of Energy Efficiency and Renewable Energy (EERE) fiscal year 2008 budget (Figure 3.1).⁶⁴ In this scenario, FCV penetration of the market for new LDVs begins in 2015 and increases slowly through 2020 to 1 percent, after which it increases rapidly to 22 percent in 2030 and approximately 50 percent in 2045. Although not specified, EERE assumed that Federal and State policies would be in place in the early stages of FCV development to foster vehicle production and sales as well as the development of a companion hydrogen infrastructure.⁶⁵

The third and most aggressive market penetration scenario examined was taken from a scenario put forth by the NRC in an examination of the potential impacts of a rapidly developed hydrogen economy (Figure 3.3).⁶⁷ This scenario assumes that all hydrogen FCV technology and cost goals are met, that the infrastructure is developed in tandem, and that there are no impediments to success. This is the most aggressive market penetration scenario, with market penetration beginning in 2015 and growing by 1 percentage point a year to 10 percent in 2024. After 2024, the rate of market penetration increases to 5 percentage points per year through 2034, when FCVs make up 60 percent of new vehicle sales. In 2038, FCVs account for 100 percent of new LDV sales.

For each of the three FCV market penetration scenarios, projected market shares for other advanced technology and alternative fuel vehicles reflect the projections in the AEO2008 reference case. In each of the scenarios, it is assumed that, as FCV market share increases, the market shares for other vehicle types are reduced in proportion to their AEO2008 reference case market shares in that year.

Fuel Cell Vehicle Fuel Economy Scenarios

Two FCV fuel economy scenarios are examined, based on projected improvement relative to a 2005 base year conventional gasoline vehicle.⁶⁸ The first scenario assumes that FCV fuel economy improvements mirror those projected in AEO2008 through 2030 and remain constant at 2030 levels through 2050 (Figure 3.4). In this scenario, the fuel economy of FCV cars is approximately twice that of conventional gasoline cars in 2005. After 2005, the FCV fuel economy ratio for cars decreases, as power output in conventional gasoline vehicles changes over time.⁶⁹

It is assumed that FCVs are introduced in the large car size class in 2013, which further reduces the average FCV fuel economy ratio to approximately 1.8, where it remains relatively constant for the remainder of the projection. For FCV light trucks, the fuel economy ratio varies in response to power output in conventional gasoline vehicles between 2005 and 2013, when it peaks at an improvement ratio of 1.8. The FCV fuel economy ratio decreases to 1.7 in 2014, when it is assumed that fuel cells are introduced into the large light truck classes, and remains relatively constant through the remainder of the projection. Scenarios using this assumption are designated as “2X.”

The second scenario assumes that FCV efficiency improves from twice the fuel economy of the 2005 base year vehicle in 2005 to three times the fuel economy of the base year vehicle in 2025 (Figure 3.4). The fuel economy improvements are assumed to be linear, although it is highly unlikely that improvement would occur in such a uniform fashion. After 2025, FCV fuel economy is assumed to remain constant. Scenarios using this assumption are designated “3X.”

Hydrogen Feedstock and Production Scenarios

To examine the potential impacts on full fuel cycle CO2 emissions associated with the market transition to hydrogen FCVs, five sources of hydrogen production were considered: (1) distributed natural gas SMR, (2) centralized natural gas SMR, (3) centralized coal gasification with CCS, (4) centralized biomass gasification, and (5) centralized nuclear power high-temperature electrolysis (HTE) of water. Table 3.1 provides the scenario descriptors and definitions used in the analysis.

To examine the relative CO2 impacts of moving from distributed natural gas SMR to one of the other four centralized production methods, the production sources were combined into four production pathways. The hydrogen production methods chosen are not intended to provide an exhaustive list of possibilities but were selected to demonstrate a range of outcomes, given current expectations of CO2 emissions for the fuel delivered to the vehicle. The “wells to tank” CO2 emissions associated with each of the sources of production are provided in Figure 3.5.⁷⁰

For each of the FCV market penetration scenarios, a companion hydrogen production transition scenario was developed to examine the range of potential full fuel cycle CO2 emission impacts. It is difficult to say with any certainty how and when the transition from distributed to central hydrogen production for vehicle refueling will occur and what actions will spur those developments. The scenarios envisioned for this analysis were constructed to reflect infrastructure development commitments that are correlated with the FCV market penetration scenarios. Figure 3.6 illustrates the share of total centralized hydrogen production for each of the FCV market penetration scenarios.

The hydrogen production pathways examined for this analysis illustrate the potential CO2 emissions associated with each production scenario when transitioning from distributed natural gas SMR to one of the other central production methods (i.e., coal gasification with CCS or nuclear power HTE of water). In all likelihood, hydrogen feedstock and production methods will vary by region to optimize production economically, based on available resources, infrastructure availability or limitations, and levels of demand.

Impacts on Light-Duty Vehicle Direct Energy Use

Projections of LDV energy demand are made for each of the FCV market penetration scenarios using the FCV fuel economy projections reflected in the AEO2008 reference case and the assumed 3X FCV fuel economy improvement. Projections of LDV energy demand are presented for 2030 and 2050 to demonstrate the relative energy impacts across market penetration scenarios and assumed levels of FCV fuel economy. There are two issues to consider when interpreting these results: (1) The energy consumption numbers reported in this analysis are at the point of use—i.e., at the LDV fleet level—and do not reflect primary energy use, which includes energy losses associated with the production, compression, and transportation of hydrogen. (2) The FCV market penetration rate will affect the total LDV stock.

Primary Energy Use Considerations

In discussing the energy use impacts of hydrogen consumed by LDVs, it must be noted that direct energy use is not the same as primary energy use. For impacts on primary energy use, it is important to consider the differences among fuel and technology combinations with regard to the efficiency of conversion from feedstock to product and the delivery of the product in a suitable form to the vehicle’s tank. For example, gasoline in an LDV contains 91 percent of the total primary energy used to supply the finished fuel. For hydrogen, the fuel load in the LDV may represent between 70 and 73 percent of the primary energy if natural gas was the primary feedstock but only 48 to 63 percent if another feedstock and production technology was used, as described in Chapter 2. Adding compression or liquefaction of the hydrogen, if required, and other transportation losses would decrease the primary energy content of the hydrogen fuel delivered to the LDV. For assessing petroleum impacts, however, because the production, transport, distribution, and dispensing of hydrogen use little if any petroleum, the changes in petroleum use described below are reasonably representative of the economy-wide changes in petroleum use.

FCV Market Penetration Considerations

By 2030, the rate of FCV market penetration in each of the three scenarios examined does not reach a level significant enough to have a large impact on LDV energy demand. This is due to the amount of time it takes to turn over the vehicle stock. Currently, the median lifetime of an LDV is approximately 16 years.⁷¹ As a result of slow stock turnover, as market penetration increases for newly introduced technologies or alternative-fuel vehicles, the impact of those vehicles will not be fully realized for well over a decade, when stock accumulations account for a larger percentage of total vehicles in use. For this reason alone, the investments needed to transition from a gasoline-centric market to a hydrogen-fueled market will initially present great economic risk for both industry participants and consumers.

LDV Direct Energy Use Impacts

As indicated in Figure 3.7, 2030 LDV energy use in the 2X FCV fuel economy scenarios is reduced by between 0.15 and 0.52 quadrillion Btu (between 0.8 and 2.9 percent) relative to the reference case, depending on the market penetration and fuel economy scenario chosen. The energy demands associated with the most optimistic FCV scenario, market penetration scenario 3 with the 3X FCV, are also shown in Figure 3.7. In this scenario, LDV energy demand in 2030 is reduced by 1.1 quadrillion Btu (6.1 percent) in comparison with the reference case. The reduction in LDV demand for petroleum products, which unlike the change in LDV demand for all energy would be representative of changes at the economy-wide level, is more dramatic. Across the three FCV market penetration scenarios, demand for gasoline and diesel is reduced by a range of 0.58 to 1.97 quadrillion Btu (3.5 to 11.9 percent) relative to the reference case, indicating a significant level of substitution of hydrogen for petroleum-based fuels.

LDV energy demand is noticeably reduced by 2050 in each of the FCV market penetration scenarios under both assumptions for FCV fuel economy. In comparison with the reference case, LDV energy demand in 2050 is reduced by 1.6 to 3.7 quadrillion Btu (8.0 to 18.1 percent), and petroleum consumption is reduced by 7.0 to 15.8 quadrillion Btu (37.1 to 84.1 percent) across the 2X FCV cases, depending on the market penetration scenario (Figure 3.8). In both scenario 2 and scenario 3, hydrogen becomes the primary fuel for LDVs, accounting for 62.8 percent and 80.4 percent of total demand, respectively. For the reasons outlined above, the change in petroleum use is likely to represent an economy-wide impact, but the change in total energy demand by LDVs does not reflect the increase in primary energy use in other sectors to produce, transport, distribute, and dispense hydrogen.

Assuming that FCVs achieve 3X fuel economy, energy use by LDVs in 2050 is reduced by 3.9 to 8.8 quadrillion Btu, or between 19.1 and 43.3 percent (Figure 3.9). Because FCVs are operating on an alternative fuel and the rate of conventional vehicle displacement determines the amount of petroleum reduction achieved across the market penetration scenarios, petroleum displacement realized across the FCV market penetration scenarios in the 2X FCV and the 3X FCV fuel economy scenarios are the same. However, relative to the 2X FCV scenarios, total hydrogen demand is lower under the 3X FCV fuel economy scenarios. In the 3X FCV scenarios, total demand for hydrogen in 2050 is between 2.2 quadrillion Btu and 5.1 quadrillion Btu lower, reducing total hydrogen demand by 38 percent across the scenarios relative to the 2X FCV scenarios.

If the assumptions in scenarios 2 and 3 were realized, energy use by LDVs in 2050 would decrease below the demand level realized in 2005 and in scenario 3 would approach a level of LDV energy use last realized in 1980. Again, these estimates of energy use by LDVs do not reflect the increase in primary energy use in other sectors to produce, transport, distribute, and dispense hydrogen.

Similar to the energy impacts realized in 2030 across the scenarios, the full fuel cycle CO2 emission reductions in 2030 are minimal. From the most conservative to the most aggressive scenario analyzed, reductions in CO2 emissions are estimated to be between 0.4 percent and 7.8 percent (Figure 3.10). Again, because FCVs do not account for a significant percentage of the operating vehicle stock in 2030, their impact on overall LDV CO2 emissions is minimal. In addition, the full transition to central hydrogen production has not occurred by 2030. In scenario 1 and scenario 2, hydrogen demand is met primarily by distributed natural gas SMR (82.7 percent and 59.7 percent, respectively), which is the highest CO2 emitter of the hydrogen production methods analyzed. As hydrogen production transitions to the lower CO2 emitting central production methods over the projection period, greater emissions reductions are realized.

As shown in Figure 3.11, CO2 emission reductions are achieved in all FCV scenarios relative to the reference case in 2050. The projections show CO2 emission reductions in 2050 varying from 2.0 percent (in scenario 1 with 2X fuel economy and hydrogen production transitioning to centralized SMR) to 63.8 percent (in scenario 3 with 3X fuel economy and hydrogen production transitioning to centralized nuclear HTE of water). Appendix C provides a description of each of the hydrogen FCV scenarios examined and graphical projections of CO2 emissions through 2050.

Plug-in Hybrid Electric Vehicle Comparison Scenario

To provide a comparative reference of the potential energy and CO2 emission impacts of a similar advanced technology to those of the hydrogen FCV, an alternative case was developed to examine the successful development of a PHEV with a 40-mile range. For purposes of evaluation, PHEVs were assumed to penetrate under market penetration scenario 2. As in the FCV scenarios, the success of PHEVs will require that all technology and cost issues be successfully resolved, that the necessary infrastructure be developed, and that policies be enacted to ensure a successful market transition. This scenario is not offered as an endorsement of PHEVs over FCVs but only as a demonstration of their relative impacts on energy demand and CO2 emissions in 2030 and 2050.

For the PHEV scenario, it is assumed that the PHEV would operate on gasoline and achieve approximately 50 miles per gallon in hybrid mode of operation and approximately 130 miles per gallon of gasoline equivalent in all-electric mode. It is also assumed that approximately 50 percent of annual PHEV travel will be provided by the all-electric mode of operation. Comparatively, the FCV achieves approximately 50 miles per gallon of gasoline equivalent in the AEO2008 reference scenario and 90 miles per gallon of gasoline equivalent in the 3X fuel economy scenario.

Figure 3.12 shows the 2030 LDV energy use under market penetration scenario 2 for the reference case, the FCV with AEO2008 reference case fuel economy, the FCV with 3X fuel economy, and the PHEV-40. As discussed previously, vehicle penetration is not at a level aggressive enough to stimulate significant energy impacts across the different scenarios, with total reductions from the reference case projected to be between 0.5 percent and 1.8 percent.

As shown in Figure 3.13, projections of LDV energy use in 2050 indicate that PHEVs could provide energy reductions commensurate with those projected under similar FCV scenarios. In the PHEV scenario, total LDV energy demand is reduced by 5.4 quadrillion Btu (26.3 percent), as compared with 3.0 quadrillion Btu (14.8 percent) in the fuel cell with AEO2008 reference fuel economy scenario and 7.2 quadrillion Btu (35.3 percent) in the fuel cell with 3X fuel economy scenario. Although reductions in petroleum demand are projected across the scenarios, the PHEV scenario reduces petroleum demand by 38.0 percent (7.1 quadrillion Btu) relative to the reference case, while a 68.5-percent reduction (12.9 quadrillion Btu) is projected in the FCV scenarios.⁷² In the PHEV scenario, electricity demand in 2050 is increased by 2.5 quadrillion Btu compared to the reference case. Although the VISION model does not make projections of total electricity demand for all sectors, the AEO2008 reference case projects total electricity demand in 2030 at 49.2 quadrillion Btu. Assuming that total electricity demand remained constant between 2030 and 2050, PHEVs would increase that demand by 5.1 percent.

Comparisons of projected CO2 emissions were also examined for the scenarios. For the PHEVs, two CO2 emission scenarios were developed, based on projected electricity generation mix—one based on the generation sources projected in the AEO2008 reference case, the other on generation sources projected in an analysis of S.2191, the Lieberman-Warner Climate Security Act of 2007, where costs for CCS and nuclear and biomass plants are 50 percent more than in the AEO2008 reference case.^73,74 Figure 3.14 illustrates the shares of electricity production by fuel type in both cases.⁷⁵

The impacts of the PHEV utility mix scenarios on full fuel cycle CO2 emissions from electric power generation are provided in Figure 3.15. In the AEO2008 reference case, CO2 emissions from electricity generation increase by 7.1 percent over the projection period, due to the greater percentage of total generation coming from coal. In the S.2191 high cost case, electric power full fuel cycle CO2 emissions decline significantly over the projection period, by 72.4 percent from 2010 to 2050, as the generation sector transitions to low-CO2 generation to meet the policy-imposed CO2 emission constraints.

Relative to the FCV scenarios that assume AEO2008 reference case fuel economy improvement, the PHEV scenarios project full fuel cycle CO2 emission reductions in 2050 that are similar to those achieved in the hydrogen production scenarios considered. In the PHEV scenario with AEO2008 reference case generation mix, total CO2 emissions are reduced by 165 million metric tons CO2 equivalent (8.5 percent) in comparison with the reference case in 2050, as shown in Figure 3.16. In comparison, the reductions projected in the FCV scenarios that assume the transition of hydrogen production to centralized natural gas SMR or coal with CCS, where CO2 emissions are 3.9 percent and 20.9 percent, respectively. If the generation mix projected in the S.2191 high cost scenario were achieved, CO2 emissions from PHEVs would be reduced by 30.9 percent (601 million metric tons CO2 equivalent) relative to the reference case in 2050, comparable to the reductions projected in the most optimistic fuel cell scenarios with 2X fuel economy improvement.

If fuel cell vehicles achieve 3X fuel economy improvement, as shown in Figure 3.17, then projected full fuel cycle CO2 emission reductions for all the hydrogen production scenarios exceed those projected in the PHEV scenario with the AEO2008 reference case utility mix. The projected emissions reductions for the PHEV scenario with the S.2191 high cost scenario utility mix exceed the reductions projected for the natural gas SMR FCV scenario.

Conclusion

Considerable reductions in LDV energy demand and full fuel cycle CO2 emissions could be achieved if the assumptions for FCVs and hydrogen infrastructure development were to come to fruition. The development of a large market for hydrogen-powered LDVs probably will require a massive financial commitment by industry and government and, ultimately, will hinge on success in fuel cell R&D as described in previous sections of this report. Competition from other promising technologies represents a further market challenge to hydrogen-powered LDVs.

The following are key findings from this analysis:

• It is highly unlikely that hydrogen FCVs will have significant impacts on LDV energy use and CO2 emissions by 2030.

• Depending on fuel economy improvement and rate of market penetration, hydrogen FCVs could reduce petroleum demand in 2050 by 37.1 to 84.1 percent.

• Depending on the method of hydrogen production, full fuel cycle CO2 emissions in 2050 could be reduced by 2.0 to 63.8 percent, depending on the market penetration scenario.

• Under similar market penetration assumptions, successful development of a PHEV-40 could provide significant reductions in petroleum use; however, the maximum reductions in petroleum use would be less than those projected in the most aggressive FCV scenarios. PHEVs can also achieve significant reductions in CO2 emissions, but the full fuel cycle emissions reductions fall short of those projected in some of the hydrogen FCV scenarios. The fuel economy of FCVs and the electricity generation mix are the key determinants of relative emissions outcomes.

Notes