Project: Dielectric Hole Burning |
|
In a dielectric hole burning experiment, the material under study is subject to a large sinusoidal electric field (burn) prior to measuring the dielectric response [91, 98, 103, 110, 112, 115, 135]. The effect of burning is to alter the dielectric properties only in a selected time range, which depends upon the frequency of the burn field. Observing such a selective modification is a clear indication of dynamic heterogeneity, which can be understood qualitatively by the picture that only domains associated with a specific time scale are being heated. In this poject, we are exploring the effects of 'selective heating', the persistence time of such modifications, their dependence upon burn frequency and amplitude, and the possibility to modify the response at frequencies far away from the main loss peak. We model the local increases of the effective temperatures in terms of a network [110, 115] (see schematic picture below), which allows us to calculate the dispersive polarization and the 'heating' for the heterogenous system. The thermal relaxation is modelled accordingly, assuming that thermal and dielectric relaxation times are locally correlated quantities. In this manner, quantitative agreement between model and experiment is achieved. |
|
Schematic diagram of the electric and thermal circuit analog of a system subject
to a distribution of relaxation times. The probability density g(lnτ) indicates
that both the capacitance values Ci and the specific heat capacities
cp,i are determined by the common g(lnτ)dlnτ,
whereas τ itself determines the resistance values via τ = RC
and the thermal coupling of the domain to the phonon bath.
We assume dynamic as well as thermodynamic heterogeneity. [115] |
|
The experimental basis for testing the above model is hole-burning data obtained for the high frequency wing of glycerol, covering a range of approx. 4.5 decades in frequency. The figure below compares an observed with a calculated hole signal. The model is capable of reproducing the shape of these modifications, their dependence on burn frequency, amplitude, and the variation with the waiting time between pump and probe. |
|
Vertical and horizontal representation of a dielectric hole measured in glycerol
at T = 187.30 K (symbols). The experimental conditions for this measurement are
E0 = 90 V/6.4 μm, fb = 0.2 Hz,
n = 6, and tw = 1 s. The lines are calculated
results from the present model. Blue symbols and line refer to the
vertical hole, ΔM(t), while the
red symbols and line reflect the
horizontal hole, ΔH(t), calculated via division by the
derivative of M(t). Note that the
feature of the curves approaching zero at short and long times is not forced by
normalization. [115] |
|
Processes studied in this way are the ionic diffusivities in vitreous CKN [91], the dielectric β-relaxation in glassy D-sorbitol [98], and the structural relaxation of glycerol near its glass transition [112, 115]. Subsequently, these studies are being continued as high-field non-linear impedance measurements [147, 151, 157, 193, 215], or as their time-resolved variant [157, 162, 168, 172, 174, 185, 215], see Selected Project: non-linear dielectrics. |
|
Reference numbers refer to the list of publications |
|
Experimental techniques:
|
Selected projects:
|