Life cycle assessment shows that retrofitting coal-fired power plants with fuel cells will substantially reduce greenhouse gas emissions

Abstract

•Wind-electrolysis hydrogen fuel cells have the largest CO2 mitigation potential•CO2, PM2.5, and SO2 mitigation potentials of different trajectories are estimated•Fuel cells reduce over 70% of CO2 emissions from coal-fired power plants•Deploying fuel cells in northern China enables the largest environmental benefits China has committed to reaching carbon neutrality by 2060. One of the largest challenges the country faces is a transition away from emission-intensive coal-fired energy. To help meet this challenge, China plans to retrofit its fleet of coal-fired power plants (CFPPs) that produce 240 TWh worth of electricity with fuel cell (FC) technology by 2050. FCs can generate zero-emission electricity and provide grid-scale energy storage. However, FC-associated greenhouse gas emissions can be generated beyond the electricity production at other life cycle stages, such as during the mining of raw materials, manufacturing, and the recycling of retired FCs. Therefore, the extent to which FCs will reduce CFPP emissions from a life cycle aspect remains unclear. To fill this knowledge gap, we consider emissions generated throughout the entire FC life cycle for four different types of FC and compared this with the emissions produced by the existing CFPPs. We found that retrofitting China’s 240 TWh CFPPs with these FCs can substantially reduce carbon dioxide, particulate matter, and sulfur dioxide, with benefits for both climate change mitigation and air quality. Addressing emissions released from coal-fired power plants (CFPPs) is vital to mitigate climate change. China aims to replace 240 TWh CFPPs with fuel cell (FC) technologies by 2050 to achieve carbon-neutrality goals. However, FCs are not emission-free throughout their technology life cycle, and FC effectiveness will vary depending on the CFPP configuration. Despite these uncertainties, a comprehensive evaluation of on-site CFPP-to-FC mitigation potential throughout the entire life cycle remains underexplored. Here, we use a prospective life cycle assessment to evaluate the inclusive mitigation potential of retrofitting 240 TWh CFPPs via four FCs that use wind power/natural gas as feedstocks. We find CO2, PM2.5, and SO2 emissions decrease by 72.0%–97.0%, 55.5%–92.6%, and 23.1%–86.1%, respectively, by 2050. Wind-electrolysis hydrogen FCs enable the largest life cycle CO2 reduction, but mining metals for wind turbines reduces PM2.5 and SO2 savings. Prioritizing FC deployment in northern China could double the mitigation potential. Our study provides insights for designing carbon-neutrality CFPP-to-FC roadmaps in China. Addressing emissions released from coal-fired power plants (CFPPs) is vital to mitigate climate change. China aims to replace 240 TWh CFPPs with fuel cell (FC) technologies by 2050 to achieve carbon-neutrality goals. However, FCs are not emission-free throughout their technology life cycle, and FC effectiveness will vary depending on the CFPP configuration. Despite these uncertainties, a comprehensive evaluation of on-site CFPP-to-FC mitigation potential throughout the entire life cycle remains underexplored. Here, we use a prospective life cycle assessment to evaluate the inclusive mitigation potential of retrofitting 240 TWh CFPPs via four FCs that use wind power/natural gas as feedstocks. We find CO2, PM2.5, and SO2 emissions decrease by 72.0%–97.0%, 55.5%–92.6%, and 23.1%–86.1%, respectively, by 2050. Wind-electrolysis hydrogen FCs enable the largest life cycle CO2 reduction, but mining metals for wind turbines reduces PM2.5 and SO2 savings. Prioritizing FC deployment in northern China could double the mitigation potential. Our study provides insights for designing carbon-neutrality CFPP-to-FC roadmaps in China. Graphical AbstractView Large Image Figure ViewerDownload Hi-res image Download (PPT) The coal power sector in China accounts for more than 50% of global coal power capacity1Evans S. Pearce R. Mapped: The World’s Coal Power Plants.2020https://www.carbonbrief.org/mapped-worlds-coal-power-plantsGoogle Scholar and was responsible for approximately 40%, 25%, and 5% of China’s total CO2, SO2, and PM2.5 emissions in 2020,2Ministry of Ecology and Environment of the People’s Republic of ChinaEco-environmental Statistics Annual Report.2022Google Scholar,3Guan Y. Shan Y. Huang Q. Chen H. Wang D. Hubacek K. Assessment to China's recent emission pattern shifts.Earths Future. 2021; 9e2021EF002241https://doi.org/10.1029/2021EF002241Crossref Scopus (10) Google Scholar respectively, posing great challenges to global decarbonization and local citizens’ health.4Li X. César Alejandro Hernández AlvaPower Sector Reform in China: An International Perspective.2018Google Scholar To tackle these challenges, it is urgent for China to promote clean and low-carbon power transition. According to the International Energy Agency,5International Energy AgencyAn Energy Sector Roadmap to Carbon Neutrality in China.https://www.iea.org/reports/an-energy-sector-roadmap-to-carbon-neutrality-in-chinaDate: 2021Google Scholar to achieve carbon neutrality, the share of coal-fired power plants (CFPPs) in China’s power mix needs to substantially drop from 60% in 2020 to 5% in 2060, while the percentage of renewable power needs to rapidly increase from 25% to 80% during the same period. Fuel cells are an integral part of this ambitious transition strategy, as they are a key solution to the inherent inflexibility of renewable power.6de Chalendar J.A. Benson S.M. Why 100% renewable energy is not enough.Joule. 2019; 3: 1389-1393https://doi.org/10.1016/j.joule.2019.05.002Abstract Full Text Full Text PDF Scopus (20) Google Scholar,7Fuhrman J. Clarens A.F. McJeon H. Patel P. Doney S.C. Shobe W.M. Pradhan S. China's 2060 carbon neutrality goal will require up to 2.5 GtCO2/year of negative emissions technology deployment.arXiv. 2020; (Preprint at) (2010.06723)Google Scholar An additional 2,000 GW of flexible power sources from fuel cells is required by China’s carbon-neutral power system to substitute CFPPs.5International Energy AgencyAn Energy Sector Roadmap to Carbon Neutrality in China.https://www.iea.org/reports/an-energy-sector-roadmap-to-carbon-neutrality-in-chinaDate: 2021Google Scholar Moreover, it is versatile with regard to a variety of energy feedstocks, ranging from coal and natural gas to solar- and wind-electrolysis hydrogen, making it a competitive option for the transition from coal-based to renewable energy-based power production.8Staffell I. Scamman D. Velazquez Abad A. Balcombe P. Dodds P.E. Ekins P. Shah N. Ward K.R. The role of hydrogen and fuel cells in the global energy system.Energy Environ. Sci. 2019; 12: 463-491https://doi.org/10.1039/c8ee01157eCrossref Scopus (1172) Google Scholar, 9Jensterle M. Narita J. Piria R. Samadi S. Prantner M. Crone K. Siegemund S. Kan S. Matsumoto T. Shibata Y. The Role of Clean Hydrogen in the Future Energy Systems of Japan and Germany.2019Google Scholar, 10Mi Shuhua Y.Z. Zhang W. Xu W. A White Paper on China's Hydrogen and Fuel Cell Industry (In Chinese). China Hydrogen Alliance, 2019Google Scholar, 11Staffell I. Zero carbon infinite COP heat from fuel cell CHP.Appl. Energy. 2015; 147: 373-385https://doi.org/10.1016/j.apenergy.2015.02.089Crossref Scopus (38) Google Scholar China, therefore, has set ambitious fuel cell expansion goals, i.e., deploying 5,000 fuel cell power units by 2035 and 20,000 by 2050, which is projected to substitute ∼60 and ∼240 TWh of coal power, respectively.10Mi Shuhua Y.Z. Zhang W. Xu W. A White Paper on China's Hydrogen and Fuel Cell Industry (In Chinese). China Hydrogen Alliance, 2019Google Scholar,12Lu Y. Cai Y. Souamy L. Song X. Zhang L. Wang J. Solid oxide fuel cell technology for sustainable development in China: an over-view.Int. J. Hydrog. Energy. 2018; 43: 12870-12891https://doi.org/10.1016/j.ijhydene.2018.05.008Crossref Scopus (33) Google Scholar Understanding emission mitigation potentials of replacing CFPPs with fuel cells is essential to a proper roadmap of fuel cell development, as this can identify an environmentally optimal trajectory to deploy fuel cells. Although fuel cell technology is considered to be zero emission from the end-of-pipe perspective,11Staffell I. Zero carbon infinite COP heat from fuel cell CHP.Appl. Energy. 2015; 147: 373-385https://doi.org/10.1016/j.apenergy.2015.02.089Crossref Scopus (38) Google Scholar its upstream processes involve emission-intensive activities, such as mineral mining, fuel cell manufacturing, and fuel transportion.13Rillo E. Gandiglio M. Lanzini A. Bobba S. Santarelli M. Blengini G. Life cycle assessment (LCA) of biogas-fed solid oxide fuel cell (SOFC) plant.Energy. 2017; 126: 585-602https://doi.org/10.1016/j.energy.2017.03.041Crossref Scopus (51) Google Scholar,14Notter D.A. Kouravelou K. Karachalios T. Daletou M.K. Haberland N.T. Life cycle assessment of PEM FC applications: electric mobility and μ-CHP.Energy Environ. Sci. 2015; 8: 1969-1985https://doi.org/10.1039/c5ee01082aCrossref Scopus (52) Google Scholar A few case studies have been carried out on the contribution of upstream activities to overall emissions of fuel cells. Rillo et al.13Rillo E. Gandiglio M. Lanzini A. Bobba S. Santarelli M. Blengini G. Life cycle assessment (LCA) of biogas-fed solid oxide fuel cell (SOFC) plant.Energy. 2017; 126: 585-602https://doi.org/10.1016/j.energy.2017.03.041Crossref Scopus (51) Google Scholar revealed that per kWh electricity generation by natural gas fuel cells induces ∼100 g indirect CO2 emissions and ∼650 mg indirect PM2.5 emissions. Longo et al.15Longo S. Cellura M. Guarino F. Brunaccini G. Ferraro M. Life cycle energy and environmental impacts of a solid oxide fuel cell micro-CHP system for residential application.Sci. Total Environ. 2019; 685: 59-73https://doi.org/10.1016/j.scitotenv.2019.05.368Crossref PubMed Scopus (28) Google Scholar evaluated the manufacturing process for a natural gas fuel cell and showed that the process contributes to ∼40% of total environmental impact. These existing studies have verified that, if we neglect the upstream impacts, emission leakage problems will occur. This will lead to an overestimation of emission mitigation potential of the CFPP-to-fuel cell transition or, even worse, biased decision-making for fuel cell deployment. However, the previous studies usually focus on one of the many types of fuel cells, such as solid oxide fuel cell (SOFC), molten carbonate fuel cell (MCFC), phosphoric acid fuel cell (PAFC), and proton exchange membrane fuel cell (PEMFC). A comprehensive comparison between various fuel cells, in terms of their emission mitigation potentials, remains lacking. Moreover, these studies did not consider the future advancement of fuel cell manufacturing and end-of-life recycling technologies, which might greatly change the environmental impact of fuel cells. Some studies pointed out that technological advances, such as lowering the metal loading or elongating the lifetime of fuel cells,14Notter D.A. Kouravelou K. Karachalios T. Daletou M.K. Haberland N.T. Life cycle assessment of PEM FC applications: electric mobility and μ-CHP.Energy Environ. Sci. 2015; 8: 1969-1985https://doi.org/10.1039/c5ee01082aCrossref Scopus (52) Google Scholar,16Nease J. Adams T.A. Life cycle analyses of bulk-scale solid oxide fuel cell power plants and comparisons to the natural gas combined cycle.Can. J. Chem. Eng. 2015; 93: 1349-1363https://doi.org/10.1002/cjce.22207Crossref Scopus (14) Google Scholar,17Smith L. Ibn-Mohammed T. Yang F. Reaney I.M. Sinclair D.C. Koh S.C.L. Comparative environmental profile assessments of commercial and novel material structures for solid oxide fuel cells.Appl. Energy. 2019; 235: 1300-1313https://doi.org/10.1016/j.apenergy.2018.11.028Crossref Scopus (11) Google Scholar could alleviate life cycle emissions of fuel cells. Therefore, a prospective life cycle assessment (LCA) is needed to assess the impact of technological changes on fuel cells’ life cycle environmental impacts. In addition, the emission intensity of CFPPs is a key factor that determines the mitigation potential of the substitution of CFPPs with fuel cells. Researchers8Staffell I. Scamman D. Velazquez Abad A. Balcombe P. Dodds P.E. Ekins P. Shah N. Ward K.R. The role of hydrogen and fuel cells in the global energy system.Energy Environ. Sci. 2019; 12: 463-491https://doi.org/10.1039/c8ee01157eCrossref Scopus (1172) Google Scholar,18Staffell I. Measuring the progress and impacts of decarbonising British electricity.Energy Policy. 2017; 102: 463-475https://doi.org/10.1016/j.enpol.2016.12.037Crossref Scopus (64) Google Scholar have compared the CO2 emission intensity of natural gas fuel cells and other power generation techniques and have concluded that the CO2 mitigation potential of natural gas-based fuel cells is relatively low. However, these studies considered the grid-average emission intensity, which may drastically deviate from the emission intensity of individual power plants.19Cui R.Y. Hultman N. Cui D. McJeon H. Yu S. Edwards M.R. Sen A. Song K. Bowman C. Clarke L. et al.A plant-by-plant strategy for high-ambition coal power phaseout in China.Nat. Commun. 2021; 12https://doi.org/10.1038/s41467-021-21786-0Crossref Scopus (49) Google Scholar For example, Zhou et al.20Zhou S. Wei W. Chen L. Zhang Z. Liu Z. Wang Y. Kong J. Li J. Impact of a coal-fired power plant shutdown campaign on heavy metal emissions in China.Environ. Sci. Technol. 2019; 53: 14063-14069https://doi.org/10.1021/acs.est.9b04683Crossref PubMed Scopus (29) Google Scholar found that the heavy metal emissions of small-sized CFPPs are much higher than those of average-capacity CFPPs. Currently, more than a thousand CFPPs with strikingly different environmental emission intensities remain in operation in China. Ignoring the difference between the grid-average and marginal emission intensities of CFPPs may adversely affect cumulative mitigation potential estimation. Hence, a detailed CFPP database with high spatial resolution may help to reveal the differences in emissions of CFPPs and comprehensively assess the mitigation potentials of replacing CFPPs with fuel cells. Here, we selected four types of fuel cells (SOFC, MCFC, PAFC, and PEMFC), which are widely accepted as the mainstream fuel cell technologies.21E4techThe Fuel Cell Industry Review 2020.2020Google Scholar We first established life cycle inventories of natural gas (NG) and wind-electrolysis hydrogen (WEH) fuel cells. Then, the prospective LCA is used to estimate life cycle CO2-eq, PM2.5-eq, and SO2-eq emissions of different fuel cells, as the three emissions are major derivatives from coal power production and has been listed as China’s Ambient Air Quality Index. In line with the fuel cell deployment target by the China Hydrogen Alliance, we then developed three trajectories to replace 240 TWh electricity from CFPPs with fuel cells based on a high spatial resolution CFPP database, and assessed the cumulative mitigation potentials. We estimated that prospective life cycle CO2-eq, PM2.5-eq, and SO2-eq emission mitigation potential of replacing CFPPs with fuel cells in China will be 235–532 Mt, 115–315 kt, and 77–471 kt, equivalent to 72.0%–97.0%, 55.5%–92.6%, and 23.1%–86.1% of emissions reduction by 2050. WEH fuel cells enable larger CO2 emission mitigation potential than NG fuel cells. Comparison between different trajectories suggests that deploying fuel cells in northern China should be the priority. The findings of this study provide not only a useful tool to evaluate overall benefits of different energy transition strategies but also insights for fossil-energy-dominated countries to formulate appropriate low-carbon energy transition roadmaps. We examined mitigation potentials in three dimensions: full cell technology, prospective technological changes, and full cell deployment trajectories. For fuel cell technology, we selected four types of fuel cells (SOFC, MCFC, PAFC, and PEMFC) with two types of energy feedstocks (NG and WEH) to establish life cycle inventories (see https://github.com/panday1995/2021-FuelCellLCA for detailed life cycle inventories, Table S1 for relevant assumptions, and Figure S1 for system boundary). For prospective technological changes, we considered four types of technological changes in the foreground system of LCA under two scenarios (optimistic and pessimistic scenarios). In the optimistic scenario, end-of-life (EoL) recycling rate will increase to 76% for platinum and 87% for nickel, fuel efficiency will decrease by 5% due to ancillary device consumption, fuel cell lifetime will extend by 50%, and full load hours (FLHs) will be 6,500 h. For the pessimistic scenario, the values of the aforementioned parameters are 35% for platinum and 30% for nickel, 15% for fuel efficiency loss, 5% for lifetime expansion, and 6,000 h for FLHs. For fuel cell deployment trajectories, we assumed that 240 TWh electricity from CFPPs will be replaced by fuel cells, in line with the 2050 fuel cell target by the China Hydrogen Alliance. We then set and examined cumulative emission mitigation potentials of the following three CFPP-to-fuel cell deployment trajectories. The best-policy trajectory (BPT) assumes that CFPPs with the highest CO2 emission intensity will be replaced by fuel cells; the capacity-order pathway (CaP) assumes that the lowest-capacity CFPPs will be substituted by fuel cells; and the location of consumption (LoC) projects that provinces with the largest NG demand will replace CFPPs to fuel cells first (see detailed rationales for different deployment trajectory in the experimental procedures). Figure 1 shows the life cycle CO2-eq, PM2.5-eq, and SO2-eq emissions and their breakdown into the manufacturing and operational phases for NG and WEH fuel cells. The life cycle CO2-eq emissions of WEH fuel cells range from 49.2 to 55.1 g/kWhe, significantly lower than that of NG fuel cells (283.2–357.2 g/kWhe). However, for PM2.5-eq and SO2-eq emissions, WEH fuel cells exhibit higher emissions compared with NG fuel cells.Figure 1Life cycle emissions of natural gas and wind-electrolysis hydrogen fuel cellsShow full captionSystem boundary of fuel cell technology is shown in Figure S1. Electricity and heat are the main outputs of fuel cell systems. One kWh electricity generation (kWhe) is adopted as the functional unit for the system. The overall impact of the system is allocated between electricity and heat based on exergy allocation scheme (see Method summary for details).(A) The life cycle emissions of natural gas (NG) and wind-electrolysis hydrogen (WEH) fuel cells per kWhe are measured in g CO2-eq, g PM2.5-eq, and g SO2-eq. The error bars indicate uncertainty distributions with a CI of 99% based on 1,000-iteration Monte Carlo simulation.(B) Breakdown of the contributions of fuel cell components to the life cycle emissions of the manufacturing phase, including cell stack, fuel processor, power conditioning, thermal management, assembly, and disposal. For visualization, the PM2.5-eq and SO2-eq values are multiplied by factors of 5 and 3, respectively.(C) Breakdown of emissions in the operational phase, including system operation and upstream activities.View Large Image Figure ViewerDownload Hi-res image Download (PPT) System boundary of fuel cell technology is shown in Figure S1. Electricity and heat are the main outputs of fuel cell systems. One kWh electricity generation (kWhe) is adopted as the functional unit for the system. The overall impact of the system is allocated between electricity and heat based on exergy allocation scheme (see Method summary for details). (A) The life cycle emissions of natural gas (NG) and wind-electrolysis hydrogen (WEH) fuel cells per kWhe are measured in g CO2-eq, g PM2.5-eq, and g SO2-eq. The error bars indicate uncertainty distributions with a CI of 99% based on 1,000-iteration Monte Carlo simulation. (B) Breakdown of the contributions of fuel cell components to the life cycle emissions of the manufacturing phase, including cell stack, fuel processor, power conditioning, thermal management, assembly, and disposal. For visualization, the PM2.5-eq and SO2-eq values are multiplied by factors of 5 and 3, respectively. (C) Breakdown of emissions in the operational phase, including system operation and upstream activities. For different types of fuel cell with the same energy feedstock, i.e., SOFC, MCFC, PAFC, and PEMFC, their life cycle emissions are comparable with each other. Taking NG fuel cells as examples, the PM2.5-eq and SO2-eq emissions of MCFC and PAFC are slightly higher than SOFC and PEMFC due to their higher resource demands, but their CO2-eq emissions are slightly lower compared with those of SOFC and PEMFC owing to their higher system efficiency (Table S2). On average, the life cycle emissions of NG fuel cells encompass 326.3 g CO2-eq, 0.130 g PM2.5-eq, and 0.451 g SO2-eq per kWhe (Figure 1A), which are better than those of combined-cycle gas turbines.8Staffell I. Scamman D. Velazquez Abad A. Balcombe P. Dodds P.E. Ekins P. Shah N. Ward K.R. The role of hydrogen and fuel cells in the global energy system.Energy Environ. Sci. 2019; 12: 463-491https://doi.org/10.1039/c8ee01157eCrossref Scopus (1172) Google Scholar,22Martin-Gamboa M. Iribarren D. Dufour J. Environmental impact efficiency of natural gas combined cycle power plants: a combined life cycle assessment and dynamic data envelopment analysis approach.Sci. Total Environ. 2018; 615: 29-37https://doi.org/10.1016/j.scitotenv.2017.09.243Crossref PubMed Scopus (40) Google Scholar We then separately investigated the contributions of the manufacturing (Figure 1B) and operational (Figure 1C) phases to the life cycle emissions of fuel cells. We found that the emission characteristics of the considered fuel cells are different from those of traditional power generation techniques, such as a thermal boiler,23Oberschelp C. Pfister S. Raptis C.E. Hellweg S. Global emission hotspots of coal power generation.Nat. Sustain. 2019; 2: 113-121https://doi.org/10.1038/s41893-019-0221-6Crossref Scopus (88) Google Scholar which primarily emits pollutants on site. Although a major share of CO2-eq (96.7%–99.5% for NG fuel cells and 84.1%–97.4% for WEH fuel cells) still originates from on-site operation, a notable share of PM2.5-eq and SO2-eq emissions is shifted to the fuel cell manufacturing phase. This shift differs across the fuel cell types: for NG PEMFCs, the contributions of the manufacturing phase to PM2.5-eq and SO2-eq are 1.43% and 1.47%, respectively. Regarding NG MCFCs, the contributions to PM2.5-eq and SO2-eq increase to 53.8% and 58.9%. Figure 1B shows the contribution of each system component to the life cycle emissions in the manufacturing phase. Cell stacks occupy a leading share of all the emissions (e.g., 25.6%, 47.9%, and 55.1% of the CO2-eq, PM2.5-eq, and SO2-eq for NG fuel cells, respectively). The use of nickel in SOFCs and MCFCs and platinum in PAFCs and PEMFCs are the main causes of the PM2.5-eq and SO2-eq emissions for both NG fuel cells and WEH fuel cells, due to their associated pollution-heavy extraction processes.14Notter D.A. Kouravelou K. Karachalios T. Daletou M.K. Haberland N.T. Life cycle assessment of PEM FC applications: electric mobility and μ-CHP.Energy Environ. Sci. 2015; 8: 1969-1985https://doi.org/10.1039/c5ee01082aCrossref Scopus (52) Google Scholar Fuel processor constitutes of another notable part of the PM2.5-eq (28.7%) and SO2-eq (29.6%) emissions for NG fuel cells, driven by the use of noble-metal catalysts. WEH fuel cells incur no emissions related to fuel processors, as electrolytic hydrogen can be directly fed into fuel cells. The operational phase is a hotbed of CO2-eq emissions, similar to conventional power generation techniques. The majority of CO2-eq can be ascribed to NG reforming in system operation (72.2%–77.8%) for NG fuel cells (Figure 1C). NG production is the dominant contributor to PM2.5-eq and SO2-eq emissions, as the system operation of fuel cells is a near-zero-emission process. Furthermore, WEH fuel cells have zero emissions at the system operation stage. The upstream activities, such as wind power station installation, the construction of electrolyzers, and corresponding mineral extraction take over all the emissions in this phase. They make WEH fuel cells heavy emitters in PM2.5-eq and SO2-eq compared with NG fuel cells. Nonetheless, its CO2-eq emissions (45.0–48.1 g/kWhe) are drastically lower than that of NG fuel cells (73.7–78.9 g/kWhe). We then performed a prospective LCA on the foreground of fuel cell system using the scenario range approach24Arvidsson R. Tillman A.-M. Sandén B.A. Janssen M. Nordelöf A. Kushnir D. Molander S. Environmental assessment of emerging technologies: recommendations for prospective LCA.J. Ind. Ecol. 2018; 22: 1286-1294https://doi.org/10.1111/jiec.12690Crossref Scopus (125) Google Scholar to investigate the impact of technological changes (including EoL recycling rate increase, fuel efficiency loss, lifetime expansion, and FLH decrease) on fuel cells’ life cycle emissions. Fuel cells are assumed to have experienced those technological changes at the time of replacing CFPPs for power generation, i.e., fuel cells are at the highest technology readiness level. Our results show that the prospective technological changes moderately alter the life cycle emissions of NG and WEH fuel cells (Figure 2). The PM2.5-eq (−42.0% to 14.6%) and SO2-eq (−50.8% to 14.3%) emissions of all fuel cells change greatly if those technological changes occur, while the CO2-eq emissions exhibit a moderate variation (−4.10% to 14.7%).Figure 2Technological changes and their effects on life cycle CO2-eq, SO2-eq, and PM2.5-eq emissions of NG and WEH fuel cellsShow full caption(A) Solid oxide fuel cell (SOFC).(B) Molten carbonate fuel cell (MCFC).(C) Phosphoric acid fuel cell (PAFC).(D) Proton exchange membrane fuel cell (PEMFC). The left two traces of the waterfall plot in each chart represent the changes in life cycle emissions of NG fuel cells under the optimistic scenario (76% platinum and 87% nickel recycling rates in end-of-life [EoL] recycling, 5% fuel efficiency loss due to ancillary device consumption, 50% lifetime expansion, and 6,500 full load hours [FLHs]), and under the pessimistic scenario (35% platinum and 30% nickel recycling rates in EoL recycling, 15% fuel efficiency loss due to ancillary device consumption, 5% lifetime expansion, and 6,000 FLHs), respectively. The right two traces of the waterfall plot represent the changes in life cycle emissions of WEH fuel cells under the same scenario as NG fuel cells.View Large Image Figure ViewerDownload Hi-res image Download (PPT) (A) Solid oxide fuel cell (SOFC). (B) Molten carbonate fuel cell (MCFC). (C) Phosphoric acid fuel cell (PAFC). (D) Proton exchange membrane fuel cell (PEMFC). The left two traces of the waterfall plot in each chart represent the changes in life cycle emissions of NG fuel cells under the optimistic scenario (76% platinum and 87% nickel recycling rates in end-of-life [EoL] recycling, 5% fuel efficiency loss due to ancillary device consumption, 50% lifetime expansion, and 6,500 full load hours [FLHs]), and under the pessimistic scenario (35% platinum and 30% nickel recycling rates in EoL recycling, 15% fuel efficiency loss due to ancillary device consumption, 5% lifetime expansion, and 6,000 FLHs), respectively. The right two traces of the waterfall plot represent the changes in life cycle emissions of WEH fuel cells under the same scenario as NG fuel cells. The life cycle emissions of the fuel cells are most sensitive to EoL recycling rate and fuel efficiency change. EoL recycling is the main driver of the changes in PM2.5-eq (−29.2% to 1.17% for NG fuel cells and −16.7% to 0.35% for WEH fuel cells) and SO2-eq (−36.2% to 1.57% for NG fuel cells and −26.1% to 0.57% for WEH fuel cells). Fuel efficiency loss is the primary driver of the changes in CO2-eq emissions (1.10%–4.02% for NG fuel cells and 4.21%–14.7% for WEH fuel cells). This leads to 11.04–11.83 g/kWhe life cycle CO2-eq emissions increase for NG fuel cells under the pessimistic scenario (15% efficiency loss), which is approximately twice the emissions in the manufacturing phase. System lifetime expansion exerts a major impact on PM2.5-eq and SO2-eq emissions, but aggressive improvement is required (50% lifetime extension). FLH decrease contributes least to all the emissions, but we caution that it may cause efficiency variation through physio-chemical mechanisms, thus indirectly affecting the environmental impacts of these fuel cells.11Staffell I. Zero carbon infinite COP heat from fuel cell CHP.Appl. Energy. 2015; 147: 373-385https://doi.org/10.1016/j.apenergy.2015.02.089Crossref Scopus (38) Google Scholar In line with China Hydrogen Alliance fuel cell development 2050 target, we set three fuel cell deployment trajectories, where fuel cell power stations are assumed to replace CFPPs by producing the same amount of electricity. The location and capacity of substituted CFPPs as well as cumulative emission mitigation potentials of substituting CFPPs with fuel cells (calculated according to Equations 1 and 2) are shown in Figure 3. It is estimated that 235–532 Mt CO2-eq, 115–315 kt PM2.5-eq, and 77–471 kt SO2-eq will be reduced if China transitions from CFPPs to fuel cells for power generation in 2050, e