The DiMarzio-Bishop model is thoroughly analyzed and reformulated in such a way that different unphysical simplifications that have been used earlier, are avoided. New testable qualitative predictions of the DiMarzio-Bishop model are formulated and the model is put in a form where quantitative tests can be made by using only one macroscopic parameter.
The DiMarzio-Bishop model is tested by extensive dielectric and shear mechanical measurements on various molecular liquids. The shear mechanical measurements are performed using a transducer that has been developed at IMFUFA by Christensen and Olsen [Rev. Sci. Instrum. 66 (1995) 5019]. This transducer allows measurements to be made in an exceptionally large frequency range (1mHz-50kHz). The systematic errors and uncertainties of the two measuring methods are analyzed in detail. Furthermore these errors and uncertainties are taken into account in the reformulation and tests of the DiMarzio-Bishop model.
It is found that the DiMarzio-Bishop model to a large extend has qualitative agreement with our data and data from the literature, whereas the quantitative agreement is moderate or poor depending on the liquid tested. This suggests that the model is too coarse grained to capture details of the relaxation processes, but that it does in fact capture the fundamentals of the physics involved, and consequently that there is a direct relation between shear mechanical relaxation and dielectric relaxation.
Abstract of a posterpresented at the Danish Physical Society Annual Meeting 2003.
A viscoelastic liquid behaves as a liquid at long time scales whereas it behaves as a solid on short time scales, and the characteristic time scale of the corresponding relaxation is strongly temperature dependent. The temperature dependence and the nature of this relaxation are the most fundamental questions in this area of research.
We study the relaxation process by measuring and comparing the frequency dependency of the shear mechanical modulus and of the dielectric constant. The primary motivation for this is that any relation between shear- and dielectric relaxation, whether quantitative or qualitative, would be a stepping stone in the direction of understanding the microscopic dynamics behind the relaxation processes.
The study is based on measurements on various molecular liquids and small polymers. The dielectric measurements are performed using standard methods (1mHz-1MHz), whereas the shear mechanical measurements are performed using a transducer which has been developed at IMFUFA [Rev. Sci. Instrum. 66 (1995) 5019] to obtain a large frequency range (1mHz-50kHz).
The temperature dependency of the two relaxations are compared and a model [J. Chem. Phys. 60 (1974) 3802], [J. Non-Cryst. Solids, 172-174 (1994) 357] of the connection between the two relaxation processes is tested.
The starting point has been a generalized Debey model which has been proposed in two slightly different versions in [J.Chem.Phys. 60 (1974) 3802] and [J.Non-Cryst. Solids, 172-174 (1994) 357]. In the master thesis this model has been thoroughly analyzed and reformulated such that different erroneous simplifications which have been used earlier are avoided.
The model is tested by extensive dielectric and shear mechanical measurements on various molecular liquids. The dielectric measurements are standard whereas the shear mechanical measurements are performed using a transducer which has been developed at IMFUFA [Rev. Sci. Instrum. 66 (1995) 5019] allowing measurements to be made in large frequency range (1mHz-50kHz), and hereby obtaining the best possible test of the model. The liquids are chosen in two catagories: 1) Liquids that exhibit no secondary relaxation and obey time-temperature superposition are examined in order to test the model for these ``well behaved'' liquids. 2) Liquids with secondary relaxations are chosen such that the loss peak of the secondary relaxation lies in the narrow interval where it can be separated from the primary relaxation and still be access by the shear mechanical measurements (100Hz-1kHz). One of the reasons to chose liquids with a secondary relaxation is that it gives us a relaxation spectrum with more features which the model has to capture in order to fit. It also opens the possibility of examining whether the model holds better for the primary than the secondary relaxation, as it has been suggested in [J. Chem. Phys.107 (1997) 3645].
The multilayered capacitor, the PSG and a daisy (a flower with radius of approximately 1cm).