| Summary: | 博士 === 國立清華大學 === 工程與系統科學系 === 87 === An improved method using the indirect transform method plus hard-sphere model for interparticle effect to reconstruct the free-form size distribution from the small-angle X-ray scattering (SAXS) data is developed. In this model-independent way, hard-sphere model considers the depletion zone around the particles. This method is tested by analyzing simulated SAXS data and shows a better agreement than other similar studies. Then this method is applied to the analysis of a series of measured SAXS data from the coarsening stage of δ''precipitation for AL-7.9%Li and Al-9.7%Li alloys. More accurate parameters involving structural and thermodynamic information are obtained. The average size, size distribution, number density and volume fraction of δ'' precipitates are determined as a function of aging temperature, time and Li-content, respectively. The determined activation energies for Li diffusion are close to the other measured result. The measured growth rate constants have satisfactory agreement with the predictions based on several transmission electronic microscopy (TEM) investigations. Therefore, this work also establishes a semi-empirical expression of rate constant relating aging temperature and Li-content.
In the aging temperatures ranging from 150℃ to 170℃, the prediction of the MLSW model is found to be in the best agreement with the experimental results than other theories. The time evolutions and scaling behaviors of the size growth and the number density also follow the Ostwald ripening kinetics. Moreover, This work gives a more clear interpretation of the coarsening mechanism than other experimental SAXS and TEM results. The result also demonstrates that when the aging temperature is increased up to 180℃ for Al-9.%Li, the shape of recovered size distribution changes from negative skew to positive one due to the domination of grain boundary precipitation (GBP). The dynamic scaling behavior of this phase separation is investigated to check the time dependence of the absolute SAXS data and the related mechanism. The validity of this dynamic scaling is confirmed using the experimentally obtained power growth laws and their relationship
|
|---|
