Analysis of the hybrid copper oxide–copper sulfate cycle for the thermochemical splitting of water for hydrogen production

Autor: John C. Prindle, Victor J. Law, Ross B. Gonzales
Rok vydání: 2009
Předmět:
Zdroj: International Journal of Hydrogen Energy. 34:4179-4188
ISSN: 0360-3199
Popis: The hybrid copper oxide–copper sulfate water-splitting thermochemical cycle involves two principal steps: (1) hydrogen production from the electrolysis of water, SO2(g) and CuO(s) at room temperature and (2) the thermal decomposition of the CuSO4 product to form oxygen and SO2, which is recycled to the first step. A four-reaction version of the cycle (known in the literature as Cycle H-5) was used as the basis of the present work. For several of the four reactions, a rotating batch reactor sequence is proposed in order to overcome equilibrium limitations. Pinch technology was used to optimize heat integration. Sensitivity analyses revealed it to be economically more attractive to use a 10 °C approach to minimize heat loss (rather than 20 °C). Using standard Aspen Plus features and the Peng–Robinson equation of state for separations involving oxygen and sulfur oxides, a proposed flowsheet for the cycle was generated to yield “Level 3” results. A cost analysis of the designed plant (producing 100 million kmol/yr hydrogen) indicates a total major equipment cost of approximately $45 million. This translates to a turnkey plant price (excluding the cost of the high-temperature heat source or electrolyzer internals) of approximately $360 million. Based on a $2.50/kg selling price for hydrogen, gross annual revenue could be on the order of $500 million, resulting in a reasonable payback period when all capital and operating costs are considered. Previous efficiency estimates using Level 1 and Level 2 methods gave the process efficiency in the neighborhood of 47–48%. The Level 3 efficiency computation was 24–25% depending on the approach temperature used for recuperation. If the low quality heat rejected by the process can be recovered and used elsewhere, the Level 3 analysis could be as high as 51–53%.
Databáze: OpenAIRE