AnsweredAssumed Answered

Creating liquids of known permittivy by mixing other liquids

Question asked by drkirkby on Nov 18, 2016
Latest reply on Nov 28, 2016 by hognala

I suspect Dara might be the best person to answer this, but I'd welcome answers from anyone.


Does anyone know a way of computing the complex permittivity


Er = Er' - j Er''


of a liquid formed from a mixture of two or more miscible liquids? (E.g. water and ethanol, but not oil and water.)  In other words, what would the permittivity of 10% ethanol and 90% water be? I realize its going to be temperature and frequency dependent, but that's the sort of information I'm looking for. 


There's a fairly interesting Ph.D. thesis by Philip Bartley (attached), which attempted to find the permittivity of an unknown material using a VNA and a structure similar to a coaxial probe. The method does not rely on the standard coaxial probe shape, but is much more flexible in terms of the shape (e.g. hypodermic needed, or some other odd shape.)


The method works by training fuzzy logic and neural networks  on a number of known liquids, in an attempt to compute the properties of an unknown liquid. The author mixes various liquids together, but there's no formula given for what the resulting permittivity is - they were measured with an  Hewlett Packard 85070B system that uses an open ended
coaxial probe to determine the dielectric properties of unknown materials.


I'm wondering if there's been any work done one what mixtures of A and B will give a known Er.