The possibility of increasing the control coefficient of a nonlinear dielectric is studied. Composite structures are designed for which the control coefficient of the composite is considerably (5-20 times) higher than the control coefficients of its components are developed.
Control Coefficient of a Nonlinear Composite Dielectric.In the electrotechnical industry, use is made of materials with a nonlinear dependence of the dielectric constant on the intensity of the applied electric field [1][2][3][4]. A characteristic of nonlinearity is the quantity T , which is called the relative control of a nonlinear dielectric [1, 2]:depending on the dimension of the problem]. This quantity characterizes the change in the dielectric properties of a material under the action of a constant electric field, which is used in the design of control devices [1, 3, 4]. The function ε(E) is usually adopted in the formHere ε 0 is the dielectric constant at zero (more precisely, at E → 0). Although experimental curves for large values of E do not coincide with the curve described by dependence (2), the latter can be used for weak fields as an approximation [in many studies, (2) is used for arbitrary field magnitudes]. If the function ε(E) is chosen in the form of (2), formula (1) becomesMaterials with high control coefficients (for the currently used materials, a control coefficient of 10-20% is considered high) are characterized by a large dielectric constant (ε 0 = 500-1000) and fairly large losses [tan δ = (50-100)·10 −4 ].To reduce the dielectric constant and dielectric loss, it has been proposed [1, 3, 4] to use composite materials. However, the effectiveness of using these materials has not been determined because of the absence of information on the possible values of the average control of composites. Tagantsev et al. give the following formulas for calculating the dielectric constant ε mix and control T mix for a medium filled with spheres:(q is the volume fraction of spheres). Formulas (3) are based on the solution of the electrostatic problem for one sphere [5] and were obtained under the assumption of smallness of q. According to (3), the possible increase in the