Confocal laser 3D microscopy studies of colloidal dynamics (typical size of particles on the order of a micron) have been producing a lot of results in the literature lately. Here are two PRLs, pulled almost by random out of selection of several other papers on similar topics just this month alone. Both have Eric Weeks of Emory as an author, who also happens to have a nice writeup about confocal microscopy technique.
The first paper Phys. Rev. Lett. 99, 028301 (2007) compares local response of colloidal glass to shear with the global rheology measurements and finds a discrepancy – local relaxation rates scale as shear with exponent of 0.8 (precise meaning of which is still not entirely clear to me at the moment), but rheology (global) measurements show Herschel-Bulkley behavior with a substantially different power-law.
The second paper Phys. Rev. Lett. 99, 025702 (2007) uses 3D confocal microscopy to track down glass transition in confined geometry. Normally a monodisperse particle solution would result in wall-induced crystallization, a problem that is overcome by using a mixture of two different (but similar) sizes of particles. The main finding is that glass transition in enhanced in confinement – perhaps not surprisingly, as the particles have less degrees of freedom. For denser solutions the effect appears to be stronger (larger confinement size is sufficient to induce glass transition). Intuitively it makes good sense, somehow.
It’s interesting that even after one tracks thousands of particles in 4D space (3 spatial + 1 temporal dimensions) using these techniques, such information by itself is not meaningful, and one has to reduce it to some sort of spatio-temporal autocorrelation function, or Fourier transform the particle coordinates to say something meaningful about average local structure/dynamics. With scattering techniques (laser, x-ray, neutron) one gets these correlation functions in momentum space immediately and “for free”.