Characterization Testing of H20-SO2 Electrolyzer at Ambient Pressure

Abstract

This document reports work performed at the Savannah River National Laboratory (SRNL) that resulted in a major accomplishment by demonstrating the proof-of-concept of the use of a proton exchange membrane or PEM-type electrochemical cell to produce hydrogen via SO{sub 2}-depolarized water electrolysis. For the first time sulfur dioxide dissolved in liquid sulfuric acid was used to depolarize water electrolysis in a modern PEM cell. The use of such a cell represents a major step in achieving the ultimate goal of an economical hydrogen production process based on the Hybrid Sulfur (HyS) Cycle. The HyS Process is a hybrid thermochemical cycle that may be used in conjunction with advanced nuclear reactors or centralized solar receivers to produce hydrogen by water-splitting. Like all other sulfur-based cycles, HyS utilizes the high temperature thermal decomposition of sulfuric acid to produce oxygen. The unique aspect of HyS is the generation of hydrogen in a water electrolyzer that is operated under conditions where dissolved sulfur dioxide depolarizes the anodic reaction, resulting in substantial voltage reduction. Sulfur dioxide is oxidized at the anode, producing sulfuric acid, that is sent to the acid decomposition portion of the cycle. The focus of this work was to conduct single cell electrolyzer tests in order to prove the concept of SO{sub 2}-depolarization and to determine how the results can be used to evaluate the performance of key components of the HyS Process. A test facility for conducting SO{sub 2}-depolarized electrolyzer (SDE) testing was designed, constructed and commissioned. The maximum cell current is 50 amperes, which is equivalent to a hydrogen production rate of approximately 20 liters per hour. The test facility was designed for operation at room temperature with pressures up to 2 bar. Feed to the anode of the electrolyzer can be water, sulfuric acid of various concentrations, or sulfuric acid containing dissolved sulfur dioxide. Provisions are included to allow variation of the operating pressure in the range of 1 to 2 bar. Hydrogen generated at the cathode of the cell can be collected for the purpose of flow measurement and composition analysis. The test facility proved to be easy to operate, versatile, and reliable. Two slightly different SDE's were designed, procured and tested. The first electrolyzer was based on a commercially available PEM water electrolyzer manufactured by Proton Energy Systems, Inc. (PES). The PES electrolyzer was built with Hastelloy B and Teflon wetted parts, a PEM electrolyte, and porous titanium electrodes. The second electrolyzer was assembled for SRNL by the University of South Carolina (USC). It was constructed with platinized carbon cloth electrodes, a Nafion 115 PEM electrolyte, carbon paper flow fields, and solid graphite back plates. Proof-of-concept testing was performed on each electrolyzer at near-ambient pressure and room temperature under various feed conditions. SDE operation was evidenced by hydrogen production at the cathode and sulfuric acid production at the anode (witnessed by the absence of oxygen generation) and with cell voltages substantially less than the theoretical reversible voltage for simple water electrolysis (1.23 V). Cell performance at low currents equaled or exceeded that achieved in the two-compartment cells built by Westinghouse Electric Corporation during the original development of the HyS Process. Performance at higher currents was less efficient due to mass transfer and hydraulic issues associated with the use of cells not optimized for liquid feed. Test results were analyzed to determine performance trends, improvement needs, and long-term SDE potential. The PES cell failed after several days of operation due to internal corrosion of the titanium electrodes in the presence of sulfuric acid. Although it was anticipated that the titanium would react in the presence of acid, the rapid deterioration of the electrodes was unexpected. The USC cell was constructed of carbon-based components and had excellent corrosion resistance. However, it was a modified design originally based on gaseous reactants, and it had poor mass transfer characteristics when using liquid sulfuric acid feed with dissolved sulfur dioxide. This resulted in substantially increased polarization losses at higher current densities. Future work will focus on operation at higher temperature and pressure, as well as improved cell designs specifically considering the unique flow conditions for SDE operation.