Advanced construction materials for thermo-chemical hydrogen production from VHTR process heat

Full text: The (very) high temperature reactor concept ((V)HTR) is characterized by its potential for process heat applications. The production of hydrogen by means of thermo-chemical cycles is an appealing example, since it is more efficient than electrolysis due to the direct use of process heat. The sulfur-iodine cycle is one of the best studied processes for the production of hydrogen, and solar or nuclear energy can be used as a heating source for the high temperature reaction of this process. The chemical reactions involved in the cycle are: I2 (l) + SO2 (g) +2 H2O (l) → 2HI (l) + H2SO4 (l) (70-120 deg. C); H2SO4 (l) → H2O (l) + SO2 (g) + 1/2 O2 (g) (800-900 deg. C); 2HI (l) → I2 (g) + H2 (g) (300-450 deg. C) The high temperature decomposition of sulphuric acid, which is the most endothermic reaction, results in a very aggressive chemical environment which is why suitable materials for the decomposer heat exchanger have to be identified. The class of candidate materials for the decomposer is based on SiC. In the current study, SiC based materials were tested in order to determine the residual mechanical properties (flexural strength and bending modulus, interfacial strength of brazed joints), after exposure to an SO2 rich environment, simulating the conditions in the hydrogen production plant. Brazed SiC specimens were tested after 20, 100, 500 and 1000 hrs exposure to SO2 rich environment at 850oC under atmospheric pressure. The gas composition in the corrosion rig was: 9.9 H2O, 12.25 SO2, 6.13 O2, balance N2 (% mol). The characterization involved: weight change monitoring, SEM microstructural analysis and four-point bending tests after exposure. Most of the specimens gained weight due to the formation of a corrosion layer as observed in the SEM. The corrosion treatment also showed an effect on the mechanical properties. In the four-point bending tests performed at room temperature and at 850 deg. C, a decrease in bending modulus with exposure time was observed. While at room temperature differences were small, they became significant at 850 deg. C. The bending modulus of the exposed samples also exhibited a stronger decrease with temperature than that of the untreated reference samples. Flexural strength did not exhibit a specific pattern, since an increase of strength in some samples and a decrease in one batch of samples resulted from the exposure. (author)