Data-driven sensitivity analysis in surface structure determination using total-reflection high-energy positron diffraction (TRHEPD)

The present article proposes a data analysis method for experimentally-derived measurements, which consists of an auto-optimization procedure and sensitivity analysis. was applied to the results total-reflection high-energy positron diffraction (TRHEPD) experiment, novel technique determining surface structures or position atoms near material surface. This solves numerically partial differential equation in fully-dynamical quantum theory with many trial structures. In analysis, we focused on experimental uncertainties variation over individual fitting parameters, analyzed by solving eigenvalue problem variance-covariance matrix. A modern massively parallel supercomputer used complete within moderate computational time. provides basis choice variables practical reliability. effectiveness demonstrated structure determination Si$_4$O$_5$N$_3$ / 6H-SiC(0001)-($\sqrt{3} \times \sqrt{3}$) R30$^\circ$ Furthermore, this is applicable experiments other than TRHEPD.