A multi-scale framework for the design and deployment of reversible solid oxide cells in photovoltaic-based energy systems

Reversible solid oxide cell (rSOC) systems are a promising bidirectional conversion technology for bridging the temporal mismatch between photovoltaic (PV) generation and demand, enabling long-duration hydrogen storage, combined heat and power, and sector coupling. Existing rSOC studies, however, treat stack-level electrochemical performance and system-level deployment largely in isolation, so stack design choices are rarely traced through to energy-system outcomes. This paper develops a multi-scale framework that couples a zero-dimensional electrochemical rSOC model with whole-system mixed-integer linear programming (MILP) optimization. Stack-level sweeps across 650–800 °C, 1–10 bar, and 2,500–10,000 A/m 2 map a design space in which electrical round-trip efficiency spans ∼ 36–91 % and the specific active area, normalized by electrolysis-rated input capacity, spans ∼ 0.07–0.40 m 2 /kW, revealing a clear performance–size trade-off. Then, part-load characterization is performed for two selected designs and embedded in the MILP framework through affine inverse-efficiency maps calibrated at minimum- and full-load operation. A capacity–cost sweep (0–16 MW, 500–7,000 EUR/kW) applied to a university campus under three flexibility portfolios shows that rSOC value is strongly context-dependent. Without additional PV systems, operational benefit is limited. With PV system expansion, annualized system cost falls by more than 30 % at mid-range capacities and self-sufficiency approaches 95 %, accompanied by substantial CO 2 reductions. When batteries are also available, the economically attractive rSOC window shifts to smaller capacities and lower cost thresholds, indicating a complementary long-duration-balancing role. Robust rSOC deployment decisions thus require co-analysis of stack design, part-load behavior, and flexibility-portfolio interactions.