The transition to a sustainable energy future necessitates efficient conversion of solar energy into usable fuels and chemicals. Photoelectrochemical (PEC) and integrated photovoltaic-electrolyzer (PV+EC) devices have shown promise in lab-scale hydrogen production, but larger-scale demonstrations exceeding 100 W are limited. Scaling up these devices presents significant challenges related to efficiency, production rates, long-term stability, cost, and sustainability. These challenges vary depending on the specific device design, materials, and operating conditions. For example, while photoparticulate systems offer cost advantages, they often suffer from lower efficiency and selectivity, unlike III-V semiconductor and noble metal catalyst-based systems which can be more expensive but highly efficient. Solar concentration presents a viable approach to addressing these challenges, offering higher power density, improved efficiency, and the potential for co-generating useful heat. Thermally integrated PV+EC systems using concentrated solar energy have shown improved performance due to optimized thermal management, but most previous demonstrations have been limited to <32 W hydrogen output power. This research aims to overcome these limitations by developing and testing a kilowatt-scale solar hydrogen and heat co-generation system using concentrated sunlight, focusing on achieving high hydrogen production rates and efficiency.

Existing literature demonstrates successful lab-scale integrated PEC devices, achieving efficiencies exceeding 15% under simulated conditions. However, real-world on-sun demonstrations at larger scales remain scarce. Studies on various solar fuel technologies, including integrated PV+EC, PEC cells, photoparticulate systems, and thermochemical redox cycles, highlight the challenges of scaling-up. Several studies have explored thermally integrated PV+EC designs, demonstrating the synergistic effect of combined photoabsorber cooling, reduced electron-hole recombination, and reduced overpotentials. However, these earlier systems generally lacked sufficient power output (often below 32W) or testing under actual solar conditions. The literature emphasizes the need for careful thermal management and optimal component selection and operating conditions to leverage the benefits of thermal integration. Moreover, previous modeling studies have explored the complex coupled effects in these systems, particularly the role of flow-rate control in mitigating degradation and adapting to variations in solar irradiance.

This study presents a scaled prototype of a solar hydrogen and heat co-generation system using a 7m-diameter dual-axis tracking parabolic dish concentrator. The system integrates a triple-junction III-V PV module and a polymer electrolyte membrane (PEM) electrolyzer stack. A two-pump design, with a global pump and a PV recycle pump, decouples the water flow-rate requirements for optimal heat transfer in the CPV heat exchanger and stoichiometric control in the electrolyzer. A water-cooled shield absorbs excess concentrated light not used by the reactor, a design consideration to be addressed in future iterations. The system was operated for over 13 days under varying environmental and meteorological conditions (ambient temperature ranging from 8 °C to 20 °C, varying cloud cover). Key parameters, such as hydrogen production rate, water flow rates, temperatures and pressures, were continuously monitored. Hydrogen production rate was calculated from the electrolyzer current, assuming a Faradaic efficiency of unity. Gas chromatography confirmed the gas composition. System efficiency was calculated using both Gibbs free energy and higher heating value (HHV) definitions to provide comprehensive performance metrics. A zero-dimensional model was developed to simulate system performance, incorporating energy and mass balance equations for each component (solar dish, homogenizer, PV module, shield, electrolyzer, and piping). Model parameters were obtained from literature or fitted to experimental data.

The kW-scale system successfully produced hydrogen and heat for over 13 days. The instantaneous hydrogen production rate reached 0.9 Nm³/h (49.7 g/h) with a mean of 0.59 Nm³/h over the whole operational period. The peak hydrogen production rate (over a 5-minute window) reached 14.0 Nl/min (1.26 g/min), and a total of over 3.2 kg of hydrogen was produced during the campaign. The system generated an average of 10.6 kW of thermal heat at an outlet temperature of 45.1 °C, with a peak output of 14.9 kW and a total of 679 kWh over 13 days. Thermal integration significantly reduced the need for auxiliary electrical heating, mitigating both energy consumption and capital costs. The average device-level solar-to-hydrogen (STH) efficiency was 20.3% (Gibbs free energy) and 24.4% (HHV), representing a two-order-of-magnitude increase in power output compared to previous work. The overall system fuel efficiency was 5.5% (Gibbs free energy) and 6.6% (HHV), with a thermal efficiency of 35.3%. A strong correlation was observed between operating parameters, such as current, voltage, and outlet water temperature, validating the model's accuracy. The model accurately predicted the linear relationship between fuel and heat power output versus direct normal irradiance (DNI). Deviations from the linear fit at low DNI values were attributed to increased circumsolar radiation during cloudy conditions. A parameter study using the model identified key optimization pathways, including increasing the water flow rate, improving PV module power matching, and enhancing light homogeneity. The model predicts that with realistic improvements in these areas, the system-level STH efficiency could be tripled, approaching the experimentally achieved device-level efficiency.

The findings demonstrate the successful scaling of a thermally integrated PV+EC system for hydrogen production to a kW-scale, achieving significant improvements in both hydrogen production rate and efficiency compared to previous lab-scale demonstrations. The high device-level STH efficiency validates the synergistic benefits of thermal integration. The validated model allows for targeted optimization strategies focusing on improvements in optical efficiency, PV power matching, and light homogeneity. These optimization strategies, including improving the homogeneity of solar irradiation within the reactor, could be achieved through system design modifications such as scaling up of the PV module to match the solar dish power input and improved homogenizer designs. The model suggests that these improvements could significantly enhance the overall system efficiency. The study highlights the importance of considering both energy efficiency and operational aspects in the design and optimization of large-scale solar fuel systems. The efficient co-generation of hydrogen and heat offers further potential for enhanced system value and wider applicability.

This research successfully demonstrated a kW-scale solar hydrogen and heat co-generation system, achieving high hydrogen production rates and device-level STH efficiency under real-world on-sun conditions. Key operational challenges were overcome, and a validated model provided insights for further optimization. Future research directions include investigating advanced control strategies for flexible hydrogen, electricity, and heat co-generation, exploring integration with energy storage, and evaluating long-term stability under intermittent operation. Integrating this technology with residential heating or low-temperature industrial processes holds substantial promise for accelerating the transition to a sustainable energy system.

The current system’s efficiency is limited by suboptimal component selection at the pilot scale. Specifically, the oversized solar dish results in significant light losses at the shield. While the model suggests significant potential for improvement through optimization, achieving these improvements requires further research and development. Additionally, the study's duration was limited to 13 days, necessitating longer-term studies to fully assess system stability and durability. The reliance on a zero-dimensional model simplifies the system’s complex thermal and fluid dynamics, potentially affecting the accuracy of some predictions.