Abstract

In our previous paper [Phys. Rev. B 44, 2046 (1991)], results of high-resolution transmission electron microscopy (HRTEM) at each stage of the complicated hysteretic phase transformation of 1T-${\mathrm{TaS}}_{2}$ are described. Here, on warming from the commensurate (C) phase to the incommensurate (IC) phase, the triclinic (T) phase and the nearly commensurate (NC) phase on warming were found to serve as a discommensuration stage. Although both structures consist of similar discommensuration processes, the temperature dependence of the structure of the T phase can be explained quantitatively by a projected discommensuration model where domains with the commensurate structure are connected by the discommensurate network, but that of the NC phase on warming deviates from the prediction by the same model. In order to understand this discrepancy, a high-resolution (HR) observation of the NC phase on warming was extended to a temperature somewhat above the room temperature. The HR iamge obtained reveals, in the NC phase on warming, a ``melting'' of the commensurate domain into the discommensuration network. Based on this observation, a modified discommensuration model is proposed for the NC phase on warming as an intermediate phase in the commensurate-incommensurate phase transformation. In the Appendix, the difference in the image of the domainlike structure in the discommensurate stage obtained by TEM (projected image) and that of scanning tunneling microscopy (STM) [and of atomic force microscopy (AFM)] is pointed out. Then, the recently reported observation by STM and AFM of the discommensuration process in the NC phase on cooling is critically compared with our observation by TEM in the NC phase on warming.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call