Results of the numerical experiment are presented, which allow considerably extending the range of the waveguide method capabilities in determination of the complex permittivity of materials. Two main methodological errors are estimated, which affect the measurement accuracy and are related to errors in determination of the sample boundary positions in the measuring section and to presence of an air gap between the sample and the wide wall of the waveguide. Calculations are performed in the centimeter and millimeter ranges for materials with a large real component of the permittivity (e1 ⩾ 25) and tangents of the dielectric loss angle tan δe varying between 0.1 and 0.2. Ways of decreasing the errors are proposed. To decrease the measurement error caused by the inaccurate positioning of the sample boundaries, it is proposed to use measuring section-flanges with the length equal to the length of the sample. To decrease the error caused by the air gap between the sample and the wide wall of the waveguide, it is proposed to fill the gap with a nonconducting paste composed of finely dispersed copper powder mixed with petrolatum. The centimeter-range experiment on determination of the complex permittivity of high-ohmic silicon shows an appreciable decrease in the measurement error when the air gap is filled with the above paste. The presented method also allows the complex permeability of a material to be determined. The real component μ1 of the complex permeability of silicon as a dielectric is unity. This is experimentally confirmed.
Read full abstract