next up previous
Next: References

Research profile
From daily experience, liquids are well-known to everybody. Besides water, which certainly represents the most prominent example, there is a huge variety of liquids with distinctive properties employed for special purposes such as cooling engines or tarring roads, dilution, lubrification, hydraulics, automatic transmissions, or glass fiber optics etc.

In the search for new materials, liquids of an extremely high viscosity --ie glasses-- have attracted much attention, recently. It turns out that our knowledge is still poor of how an amorphous solid (glass) forms from a high-density fluid (liquid). Although people have been able to produce silicon-oxide glass bowls for more than 5000 years, they failed to understand the mechanism behind the liquid-to-glass transition. Why does a liquid show this dramatic increase of viscosity upon supercooling? Thorough understanding of those dynamical processes which govern the transport properties --eg the viscosity-- of a liquid is, however, indispensable for developing new amorphous materials tailored to the needs of their prospective applications.

This implies the general aim of our current research: we intend to contribute to a better understanding of the relaxation mechanism(s) responsible for the dynamical properties of supercooled liquids and the glass transition. In the following you will find a selection of results obtained lately.

  1. We discussed the motion of a particle moving through the voids of a glassy matrix and calculated the spatial correlation function of random-potential fluctuations to which the particle is exposed. Usually, in a disordered system, lacking more accurate information, one simply assumes a Gaussian correlation of potential fluctuations. In our work we showed that the potential fluctuations as a function of wavenumber are strongly non-Gaussian and definitely reflect the static structure of the glassy matrix.[BGK94]
  2. For a binary mixture of disparate-size neutral particles we calculated the partial-density relaxation functions within a mode-coupling approximation, which, we believe, gives a good description of the nonlinear dynamical feedback processes dominating the relaxation of density fluctuations in the supercooled phase. We find a glass transition at a critical packing fraction tex2html_wrap_inline28 in connection with the localization of the large particles, only. The small spheres enter a delocalized phase. They move in an almost static random potential produced by the large particles. At a critical packing fraction tex2html_wrap_inline30 , the random-potential fluctuations are sufficiently large to localize the small particles within the glass. [KB95]
  3. Frequency-dependent self-diffusion coefficients tex2html_wrap_inline32 for both particle species of a supercooled binary hard-sphere liquid have been calculated within a mode-coupling approximation. It is found that a small bump appears in tex2html_wrap_inline32 of the small particles at low frequencies when the glass transition is approached. This is interpreted as the onset of a change in the diffusion mechanism of the small particles. When the glass is formed, the small-particle diffusion changes from fast liquid-like diffusion to a slower motion (``anomalous diffusion'') in a random potential.[BK95]
  4. The long-time limits of the density-relaxation functions (non-ergodicity parameters) of a supercooled ionic melt, a symmetric molten salt (SMS), have been calculated within a mode-coupling approach. From the non-ergodicity parameters the wavenumber-dependent static dielectric function tex2html_wrap_inline36 for an ionic glass was deduced.[Sed95]
Please, refer to section Recent Publications for more references.


next up previous
Next: References

Juergen Bosse
Sun Feb 2 18:51:37 MET 1997