During a H2-TPR measurement of a Fe-beta zeolite in an H2/Ar gas mixture, the H2 consumption, measured by a TCD, suddendly drops below the baseline at around 700 °C and does not return to the end of the measurement (1000 °C).

If only H2 and Ar is present in the gas flow, this would represent a generation of H2. Is it possible, that the zeolite is releasing signigicant amount of hydrogen at around 700°C? If not, may there be any other volatile products, which affect the TCD signal?

Thank you!