Abstract
Microscopic objects in fluid are subject to thermal fluctuation. The thermal fluctuation of the surrounding fluid molecule leads to Brownian motion, and diffusion coefficient indicates the overall speed of diffusive motion in the long time limit. In our previous work, we have shown that the diffusion anisotropy of a prolate particle can be extracted from the time-series trajectory data even if there is no explicit information of particle orientation at each time step, as far as the precision, resolution, and total amount of the data is sufficient (Phys. Rev. E, vol.85, 051134 (2012)). The analysis was based on the large deviation principle. In this study, we apply our methodology to the thermal motion of a polymer molecule in aqueous solution. In particular, we analyze the time series data of end-to-end distance. We show that the numerically-obtained approximate values of large deviation quantities indicate the clear difference of intra-molecular interactions while the time-averaged value is close.
