Revisiting the problem of crystallisation and melting of Selenium.

The presence of a unique melting point is associated with crystals and not glasses. The boundary between an imperfect crystal ( laden with defects ) which still possess a well defined melting temperature and a glass is not a well defined one. To convert a crystal to a glass, one needs to introduce a large density of defects in a manner that it destroys the long range order in the system but preserves its structural integrity. In this paper, we bridge this gap between a crystal and a glass by studying the process of crystallization and melting of elemental selenium. Because of its two-fold coordination, Se is found both in the form of both polymeric linear chain and closed ring-like structures. Incorporation of rings in an assembly of linear chains destroy long-range order. Moreover, is easy for the long strands of linear chains to slide past the neighboring ones, and hence the system has a large concentration of disinclination like defects in addition to the defects caused by the entanglement of the chains. In this paper, we provide direct visual evidence that apparently faceted crystal like shapes obtained from vapor phase deposition is indeed self-assembly of linear strands that have a persistence length of 10$\mu$m. These strands are held together by weak forces and can easily be separated. These chains occasionally get entangled to form chiral structures and often meander about destroying long range orientation and translation order in a continuous manner. Like organic polymers, the obtained selenium structures also exhibit a spread in the melting temperature. This spread is closely related to the density of the substructures present in the system. This paper provides justification for treating defects as continuous tensorial quantities in one-dimensional Van der Waals solids.