Melting and freezing of colloidal crystals

Micrometer-sized colloidal particles can serve as an outstanding model systems for studies of the phase transition between gas, liquid, solid and liquid crystal because their thermal motions can be directly visualized and measured by video microscopy. We use thermosensitive N-isopropylacrylamide (NIPA) microgel colloidal spheres whose diameters can be temperature tuned to drive the phase transitions. Specifically, we study crystallization in colloidal liquid monolayers and homogenous melting of 3D superheated colloidal crystals. In Chapter 1, we report the freezing transitions from 2D liquids to polycrystals. The empirical freezing criteria have proved important for assignment of freezing points since the fundamental theories of freezing are absent. We verified the existing phenomenological freezing criteria in 2D colloidal systems for the first time. We found that these freezing criteria, usually applied in the context of single crystals were also applied to the formation of polycrystals. At the freezing point, we found four new features which can potentially serve as new freezing criteria with some advantages. Crystal melting is of fundamental interest and practical importance to science and technology, but its microscopic mechanism is still not well understood, especially for homogenous melting. Crystals always melt heterogeneously from surfaces or grain boundaries once they are heated above the melting point. Experimentally, it is difficult to create a superheated crystal and study its homogeneous melting at the microscopic scale. In Chapter 2, we report the first direct visualization of homogenous nucleation kinetics with single-particle dynamics. We develop a new technique to uniformly superheated the interior of NIPA colloidal crystals with a focused beam of light and induced the homogenous melting. Based on ~ 200 observations of nucleation processes and two-dimensional and three-dimensional measurements, we clarified the complex kinetics and made several novel discoveries. First, we found that the precursors are particle-exchange loops surrounded by strong vibrating particles rather than any defects. This can be considered as an intermediate step during the nucleation. Newborn nuclei from a large strong vibrating region can be larger than the critical size. Second, we found that the superheat limit of our hard-sphere-like colloidal crystals is at volume fraction Φs = 42% using various methods. This contributes an important point to the famous hard-sphere phase diagram. We also discovered that the superheat limit for colloidal crystals depends on Lindemann's and Born's instabilities rather than those thermodynamic instabilities. In addition, we measured the critical size, incubation time, surface tension, chemical potential difference and free energy barrier and studied the evolutions of nucleus shape and size and their impact on nucleation. The growth of the postcritical nucleus at different degrees of superheating and the nucleation on a single defect are also investigated