Abstract
In this thesis, PMIRRAS and TDS systems were used to measure the amount of water molecules adsorbed on aluminum surface, attempting to clarify the differences between the two methods in the measurement of H2O signal. In the TDS experiment, the optical path of PMIRRAS was blocked to avoid the interference from the heating infrared source when heating the sample. The reliability of the PMIRRAS measurements was significantly improved. In the TDS and PMIRRAS experiments, no matter high or low water coverage conditions, the PMIRRAS method was found only able to detect the water molecules with lower binding energy (i.e. desorbed at lower temperature at the TDS process), but unable to detect the water molecules adsorbed in high temperature region. It was suspected that the electric dipole of H2O molecules adsorbed on the sample surface were horizontally lied, so that the interaction of the incident p-polarized infrared with the adsorbed water molecules was greatly reduced. Another result of the experiments showed that the total amount of water molecules desorbed during the TDS was far less than the calculated quantities of the water molecules adsorbed on the sample surface based on the gas dynamic model with isotropic distribution. The difference might attributed to the following reasons: the large-size sample holder of the experiment system re-adsorb a great deal amount of water vapor that desorbed from the sample, or the isotropic gas dynamic model is insufficient for this experiment, or that the condensation coefficient of water molecules might change with different coverages in the experiment. It is estimated from the experimental results that the ratio of the interference from the large-size sample holder can be reduced to be <10 % of the H2O desorbed from the sample if the surface area of the holder is reduce by a factor of 250, and the holder temperature is controlled between 172~177 K.