The electricity sector has been at the forefront of global efforts to reduce energy-related greenhouse gas (GHG) emissions, through continued adoption of wind and solar photovoltaic (PV) generation, energy storage and fuel switching from coal to natural gas. Yet meaningful climate change mitigation efforts require identifying cost-effective emissions reduction strategies for all sectors, including traditionally difficult-to-decarbonize sectors like industry and transportation. For these sectors, while direct electrification could be expanded, the unique attributes of energy services, such as the need for high temperature heating for some industrial applications and energy density requirements for transport, could make it more cost-effective to use alternative energy carriers beyond a certain level of electrification. In this context, hydrogen (H
2) is an appealing alternative energy vector, if it can be produced at scale in a cost-effective manner and without GHG emissions. In this study, we evaluate the near-term cost-effectiveness of continuous, renewable H
2 production using technologies commercially available today: PV, low-temperature electrolysis, battery and gaseous H
2 energy storage. Our approach goes beyond prior techno-economic assessments of electrolytic H
2 production by explicitly accounting for the variability in hourly PV resource availability and its implications on process design, such as relative sizing of PV and electrolyzer and the magnitude and type of energy storage required for achieving >95% annual plant availability. The integrated design and operations optimization framework used here also considers various operational aspects such as: 1) inter-temporal constraints governing energy storage (H
2, battery) and electrolysis operation and 2) the need for H
2 compression prior to storage.
We evaluate the model for near-term projected costs of PV, low-temperature electrolysis, gaseous H2 storage and 4-hour battery storage, to reveal the possibility of standalone, continuous H2 production at levelized costs around $4/kg for locations in the U.S. southwest (e.g. Southern California, Arizona or New Mexico). The energy storage capacity needed to ensure 95% annual plant availability can vary significantly (at least a factor of 2X) across locations with similar annual average PV capacity factor, due to the differences in temporal variability of the resource. This highlights the importance of considering high temporal resolution of operations in evaluating the cost-effectiveness of renewable H2 production pathways. Since the optimal plant design for most locations uses PV capacity that is larger than electrolyzer capacity, the process economics could be further improved by exporting excess electricity to the grid. As an example, we evaluate the reductions in levelized H2 cost achievable through excess electricity sales for plant locations within the California Independent System Operator (CAISO) territory and historical locational marginal prices for 2010-2017.
