State of Michigan Hydrogen Demand Analysis: Current (2022), Near-term (2030), and Long-term (2050)

Abstract

Climate Change & Hydrogen as a Decarbonization Strategy Climate change will have significant consequences should global warming exceed a 1.5 °C increase over pre-industrial levels. Limiting warming to 1.5°C requires significant greenhouse gas (GHG) emission reductions, including reaching net zero emissions by 2050. Key actions to achieve global net zero emissions by 2050 include decarbonizing GHG intensive sectors by electrifying or deploying low-carbon alternatives in the place of fossil fuels. Hydrogen can play a vital role in achieving decarbonization goals by reducing GHG emissions in both the industrial and transportation sector, particularly where electrification is challenging to implement. Hydrogen can be produced through various pathways, each with its own associated GHG emissions; while electrolysis is generally considered to be the leading candidate for decarbonization when powered from renewable energy or nuclear generation, natural gas steam methane reforming (SMR) with carbon capture and storage (CCS) has the potential to be another low-carbon alternative. Industrial pilots have shown the potential for hydrogen to be used in high-temperature process heat applications as well as to displace incumbent fossil fuels (e.g., natural gas, coal, coke) in various end-uses including glassmaking, steelmaking, and cement production. In the transportation sector, hydrogen-powered fuel cell electric vehicles (FCEVs) have been both piloted and deployed, with advantages including zero tank-to-wheel emissions and higher powertrain efficiencies than diesel-powered internal combustion engine vehicles (ICEVs) in medium- and heavy-duty vehicle (MHDV) applications. Hydrogen Deployment in Michigan Significant federal support for hydrogen programs, including the $8 billion designated in Bipartisan Infrastructure Law (BIL) has enabled the Department of Energy (DOE) to focus on creating “hydrogen ecosystems'' throughout the United States. These “hydrogen hubs'' serve to accelerate the use of hydrogen as a clean energy carrier, while diversifying end-users and the pathways to produce hydrogen. Michigan’s interest in hydrogen deployment is indicated in the “MI Healthy Climate Plan” and the state’s involvement in the Midwest Alliance for Clean Hydrogen (MachH2), which was selected to receive $0.9 billion of H2Hubs funding. Given its robust manufacturing economy and strategic transportation corridors, Michigan also stands out as a pivotal arena for hydrogen deployment. The University of Michigan's Center for Sustainable Systems (CSS) conducted a workshop to identify hydrogen deployment opportunities within the state, resulting in the "Michigan Hydrogen Roadmap Report" and the creation of the MI Hydrogen Initiative (MI Hydrogen). This initiative brings together UM research expertise to create hydrogen solutions that accelerate clean and just energy transitions. MI Hydrogen developed four initial projects, including a Michigan-specific hydrogen demand analysis as its first priority. Analysis Objectives & Scope Determination The present project targets Michigan's industrial and transportation sectors. Its core objectives are to analyze current hydrogen demand, project future demand in 2030 and 2050, and quantify the potential to reduce GHG and nitrogen oxide (NOx) emissions through hydrogen deployment. The findings from the present study will contribute to the planning and execution of a regional hydrogen ecosystem such as the MachH2 hub. Based on prior work from the CSS as well as a literature review and informational interviews, this analysis focused on eight uses including petroleum refining, chemicals, pulp and paper, steelmaking, cement, glass, semiconductor manufacturing, and medium- and heavy-duty vehicles (MHDVs). These end-uses were selected due to current hydrogen usage, future hydrogen opportunities, and decarbonization potential. The analysis excludes light-duty vehicles, non-road transportation, power generation, and commercial and residential heating as other decarbonization pathways such as electrification may be more efficient. State-specific data was also difficult to occur for some of these end-uses which also led to their omission. Demand Analysis Model (2022, 2030, 2050) MHDVs were the focus of the transportation analysis, as they contributed 11.1 million metric tons of CO2eq in 2019, 21% of Michigan’s entire transportation sector. The analysis specifically focused on seven MHDV classes that are difficult to electrify, so that hydrogen could be explored as a potential decarbonization strategy. For the industry analysis, the latest Environmental Protection Agency’s (EPA) Greenhouse Gas Reporting Program (GHGRP) dataset was utilized to identify in-scope industrial facilities and to provide emissions data for indirect fuel demand estimates. Using the GHGRP dataset, 25 Michigan facilities were selected for analysis, each with substantial fossil fuel use and GHG emissions. The 25 analyzed facilities reported 7.75 million metric tons of CO2eq to GHGRP in 2022 including the combustion of fuels, process emissions, and merchant hydrogen production emissions. In comparison, the “MI Healthy Climate Plan” reported 28.05 million metric tons of CO2eq were emitted by Michigan’s “energy intensive” industries (oil, gas, and industry) in 2019. As a result, the 25 facilities analyzed account for about 28% of state-wide industrial emissions and their decarbonization would contribute substantially to state-wide efforts to reduce GHG emissions. To assess current hydrogen demand and forecast future demand, the study required facility-level metrics such as annual hydrogen consumption or hydrogen intensity, incumbent fossil fuel use, and production capacities. Data acquired directly from stakeholders included facility-specific fuel mixes for the cement industry, hydrogen demand for Flint MTA’s public transit operations, and statewide vehicle mileage for MHDVs. For sectors where data was unavailable or proprietary, the demand model was informed through decarbonization and hydrogen roadmaps, industry and transportation pilots, and federal datasets and tools. The demand model indirectly estimated current (2022) hydrogen production and energy demand from incumbent fossil fuels using facility CO2 emissions. For the transportation sector, state MHDV miles traveled were utilized to estimate annual energy demand. These estimates were converted to hydrogen demand and incumbent fossil fuel use by utilizing the physical properties of the fuels and feedstocks and assuming process parameters. To evaluate future demand, different deployment scenarios were designed for the near-term (2030) and long-term (2050) to reflect feasible hydrogen applications for each sector, in each year. These scenarios

represent a range of hydrogen demand for 2030 and 2050, reflecting different hydrogen applications for each sector, in each year. For 2030, two hydrogen deployment scenarios were defined to encompass the continuation of current uses of hydrogen as well as new uses in steelmaking and increased use in the transportation sector. For 2050, four hydrogen deployment scenarios were outlined. As in 2030, one scenario reflects status-quo hydrogen use in the petroleum refining, semiconductor, glass and transportation sector. The “Low Use” scenario also accounts for 20% hydrogen blending for process heat, partial thermal replacement in cement kilns, 4% MHDV fleet penetration with FCEVs, and 30% coke replacement in steelmaking. The “High Use” scenario maintains the same hydrogen blending percentage for process heat but explores increased MHDV fleet penetration, increased thermal replacement in cement kilns, and the addition of hydrogen-enhanced electric arc furnaces in steelmaking. A theoretical upper limit for hydrogen demand is modeled in the “Complete Hydrogen Substitution” scenario, where hydrogen use is projected for 100% of industrial process heat demands, complete MHDV fleet adoption, and remaining feedstock applications. Since these future demand estimates utilize a 2022 baseline, they were scaled to reflect future demand using projections of economic growth for industry and vehicle miles traveled (VMT) for transportation, for analysis years, 2030 and 2050. GHG & NOx Emissions Analysis Model The emissions analysis model assessed the GHG and NOx reduction potential associated with displacing fossil fuels when used as both feedstocks and fuels with hydrogen. The production and combustion emissions associated with current hydrogen use and incumbent fossil fuels (coal, coke, natural gas, etc.) were compared to the emissions from hydrogen deployment opportunities in 2030 and 2050. For each demand scenario in each target analysis year (2022, 2030, and 2050) different hydrogen production pathways were modeled including: natural gas steam methane reforming (SMR) with and without CCS, PEM electrolysis with renewables, solid oxide electrolyzer cell (SOEC) electrolysis with nuclear, and PEM electrolysis with RFC grid mix. The emissions analysis also accounted for changes in feedstock or fuel mix in industries such as steelmaking, glassmaking and cement production, where emission sources may change as a result of hydrogen deployment. For the transportation sector, the seven MHDV classes were modeled to compare the emissions of FCEVs with those of incumbent, diesel-powered ICEVs. The emissions analysis model also presumes that hydrogen is produced on-site for industrial sectors, thus excluding transport-related emissions. While for the transportation sector, the emissions associated with the transport of both hydrogen and diesel were included in order to compare total resulting emissions for both fuels. Hydrogen Demand Results (2022, 2030, 2050) The results of the demand analysis are summarized in Table ES1 – Table ES3, with key findings separated into current (2022), near-term (2030), and long-term (2050) time horizons. Michigan’s current (2022) annual hydrogen demand was estimated to be 39,100 metric tons with sources of demand comprising petroleum refining, semiconductor, glass, and transportation. As seen in Table ES1, the two largest consumers of hydrogen are the petroleum refining sector (93.4%) and the semiconductor sector (6.3%). Both of these sectors meet their

hydrogen demand by producing hydrogen on-site via a natural gas SMR facility. Guardian Glass, the only in-scope glass facility, has minimal annual hydrogen demand (0.21%) and has their hydrogen delivered via liquefied tanker truck and stored on-site. The transportation sector has the lowest estimated hydrogen demand in 2022 as the only user is Flint Mass Transportation Authority (MTA), which currently operates one hydrogen fuel cell bus. Flint MTA produces hydrogen on-site through a PEM electrolyzer that is powered by the grid electricity. Near-term (2030) annual hydrogen demand was estimated to range from 40,100 metric tons (“Incumbent Technology” scenario) to 63,400 metric tons (“Near-term Hydrogen Opportunities” scenario). While the “Incumbent Technology” scenario assumes that no advancements are made in the deployment of hydrogen, the near-term scenarios involve numerous new deployment opportunities. Though it is important to note the variance in near-term estimates among different sectors; estimates vary based on the cost of hydrogen, technology readiness, and adoption. Similar to current (2022) demand, Table ES2 shows that petroleum refining accounts for the majority of demand at 93.3% with the semiconductor sector following in second with 6.4% of demand. The glass sector is projected to decrease in demand since the facility does not expect to undergo major furnace modifications needed to generate new hydrogen demand. Conversely, more significant growth is projected in the near-term scenarios for the steel and transportation sectors. Steel generates demand from replacing 30% of the coke used in the blast furnace with hydrogen, and transportation estimates the conversion of 1% of MHDVs to hydrogen FCEVs, along with the addition of two new fuel cell buses to Flint MTA’s fleet. Long-term (2050) annual hydrogen demand was estimated through four different scenarios, as characterized in Table ES3, ranging from 36,700 metric tons (“Incumbent Technology” scenario) through 108,000 metric tons (“Low Use” scenario) to 206,000 metric tons (“High Use” scenario); the fourth scenario then represents the theoretical upper limit of 1,096,800 metric tons (“Complete Substitution” scenario). Numerous scenarios are included to account for the increased uncertainty of the extended timeline, and it should be noted that the sector with the highest relative demand differs based on the scenario. While refining remains the highest-demand sector in both the “Incumbent Technology” (91.1%) and “Low Use” (31.6%) scenarios, transportation becomes the highest-demand sector in the “High Use” (35.3%) scenario with 20% penetration among all MHDV classes, and remains the highest in the “Complete Substitution” (33.2%) scenario as well. Also noteworthy is that the “Incumbent Technology” scenario has a lower hydrogen demand in 2050 relative to 2030; this is because demand for petroleum refining products is projected to decrease in 2050 relative to 2022, which therefore decreases the estimated hydrogen demand. Total GHG & NOx Emission Reduction Results Results from the GHG and NOx emissions analysis are highlighted in Table ES4. Like the estimates of future hydrogen demand, the potential to reduce emissions from hydrogen deployment ranges, as it is especially dependent on the hydrogen production pathway. From the analysis, it is apparent that the deployment of hydrogen has the potential to reduce emissions regardless of production pathway. However, it is evident that low-carbon pathways (PEM

electrolysis via renewables, nuclear) result in the greatest emission reduction across years and demand scenarios. Other pathways have tradeoffs, as seen with introducing CCS to natural gas SMR, resulting in greater GHG emission reductions but lower NOx reduction due to the CCS technology. For example, the 2030 “Near-Term Hydrogen Opportunities” scenario, has a GHG emission reduction potential of 3.0 million metric tons with PEM electrolysis with renewables when compared to emissions from the “Incumbent Technology” scenario. In comparison, for the same scenario, PEM electrolysis with the RFC grid mix, would only result in a GHG emission reduction of 1.7 million metric tons. Similar trends exist with NOx emissions, with PEM electrolysis via renewables leading to 2.2 thousand metric tons of reduction and PEM electrolysis via RFC grid mix having 1.3 thousand metric tons for the “Near-term Hydrogen Opportunities” scenario. In 2050, GHG and NOx emission reduction potential varies considerably across scenarios and production pathways. The GHG reductions range from 5.3 million metric tons in the “Low Hydrogen Use” scenario with the PEM electrolysis via RFC grid mix to 7.5 million metric tons in the “High Hydrogen Use” scenario with PEM electrolysis via renewables. NOx emission reductions also exhibit a large range with the same scenarios and production pathways as mentioned prior yielding a reduction range of 6.5 thousand metric tons to 14 thousand metric tons. The “Complete Hydrogen Substitution” scenario with PEM electrolysis via renewables leads to the greatest GHG and NOx emission reduction, 20 million metric tons, and 14 thousand metric tons, respectively, due to the scale of hydrogen deployed in this scenario. While this analysis finds that hydrogen deployment has the potential to reduce GHG and NOx emissions, it is important to note that there is still ongoing research regarding other climatic impacts from increasing hydrogen use such as atmospheric methane and hydrogen leaks.