Theses Doctoral

Reducing the Production Cost of Hydrogen from Polymer Electrolyte Membrane Electrolyzers through Dynamic Current Density Operation

Ginsberg, Michael J.

A worldwide shift from fossil fuels to zero carbon energy sources is imperative to limit global warming to 1.5°C. While integrating high penetrations of VRE into the grid may introduce the need for upgrading an aging electrical system, renewable energy represents a new opportunity to decarbonize multiple sectors. Otherwise curtailed solar and wind energy can accelerate deep decarbonization in hard-to-reach sectors - transportation, industrial, residential, and commercial buildings, all of which must be decarbonized to limit global warming. With renewable energy as its input, electrolytic H₂ represents a solution to the supply-demand mismatch created by the proliferation of VREs on a grid designed for on-demand power. Electrolytic H₂ can stabilize the grid since the H2 created can be stored and transferred. Thus, Chapter 1 introduces the opportunity of green H2 in the context of low-cost VREs as a means of deep decarbonization through sector coupling, and an overview of the techno-economics, key technologies, and life cycle assessment versus the incumbent steam methane reformation.

The growing imbalances between electricity demand and supply from VREs create increasingly large swings in electricity prices. Capable of operating with variable input power and high current densities without prohibitively large ohmic losses, polymer electrolyte membrane (PEM) electrolyzers are well suited to VREs. In Chapter 2, polymer electrolyte membrane (PEM) electrolyzers are shown to help buffer against these supply demand imbalances, while simultaneously minimizing the levelized cost of hydrogen (LCOH) by ramping up production of H2 through high-current-density operation when low-cost electricity is abundant, and ramping down current density to operate efficiently when electricity prices are high. A techno-economic model is introduced that optimizes current density profiles for dynamically operated electrolyzers, while accounting for the potential of increased degradation rates, to minimize LCOH for any given time-of-use (TOU) electricity pricing. This model is used to predict LCOH from different methods of operating a PEM electrolyzer for historical and projected electricity prices in California and Texas, which were chosen due to their high penetration of VREs. Results reveal that dynamic operation could enable reductions in LCOH ranging from 2% to 63% for historical 2020 pricing and 1% to 53% for projected 2030 pricing. Moreover, high-current-density operation above 2.5 A cm−2 is shown to be increasingly justified at electricity prices below $0.03 kWh−1. These findings suggest an actionable means of lowering LCOH and guide PEM electrolyzer development toward devices that can operate efficiently at a range of current densities.

Chapter 3 uses techno-economic modeling to analyze the benefits of producing green (zero carbon) hydrogen through dynamically operated PEM electrolyzers connected to off-grid VREs. Dynamic electrolyzer operation is considered for current densities between 0 to 6 A cm-2 and compared to operating a PEM electrolyzer at a constant current density of 2 A cm-2. The analysis was carried out for different combinations of VRE to electrolysis (VRE:E) capacity ratios and compositions of wind and solar electricity in 4 locations – Ludlow, California, Dalhart, Texas, Calvin, North Dakota, and Maple Falls, Washington. For optimal VRE:E and wind:PV capacity ratios, dynamic operation of the PEM electrolyzer was found to reduce the LCOH by 5% to 9%, while increasing H₂ production by 134% to 173%, and decreasing excess (i.e. curtailed) electrical power by 82% to 95% compared to constant current density operation. Under dynamic electrolyzer operation, the minimum LCOH is achieved at higher VRE:E capacity ratios than constant current density operation and a VRE mix that was more skewed to whichever VRE source with the higher capacity factor at a given location. In addition, dynamically operated electrolyzers are found to achieve LCOH values within 10% of the minimum LCOH over a significantly wider range of VRE:E capacity ratios and VRE mixes than constant electrolyzers. As demonstrated, the techno-economic framework described herein may be used to determine the optimal VRE:E capacity and VRE mix for dynamically-operated green hydrogen systems that minimize cost and maximize the amount of H2 produced.

Chapter 4 focuses on the production of high-purity water and H₂ from seawater. Current electrolyzers require deionized water so they need to be coupled with desalination units. This study shows that such coupling is cost-effective in H₂ generation, and offers benefits to thermal desalination, which can utilize waste heat from electrolysis. Furthermore, it is shown that such coupling can be optimized when electrolyzers operate at high current density, using low-cost solar and/or wind electricity, as such operation increases both H₂ production and heat generation. Results of techno-economic modeling of PEM electrolyzers define thresholds of electricity pricing, current density, and operating temperature that make clean electrolytic hydrogen cost-competitive with H₂ from steam methane reforming. By using 2020 hourly electricity pricing in California and Texas, H₂ is shown to be produced from seawater in coupled desalination-electrolyzer systems at prices near $2, reaching cost parity with SMR-produced H₂. Chapter 5 concludes the dissertation with an overview of the challenges and research needs for PEM electrolyzers at scale, including projected iridium needs, iridium thrifting, recycling methods, key degradation mechanisms, a failure modes and effects analysis, and LCOH projections.


  • thumnail for Ginsberg_columbia_0054D_18041.pdf Ginsberg_columbia_0054D_18041.pdf application/pdf 5.31 MB Download File

More About This Work

Academic Units
Earth and Environmental Engineering
Thesis Advisors
Fthenakis, Vasilis M.
Esposito, Daniel V.
D.E.S., Columbia University
Published Here
August 9, 2023