I have manufactured some capacitors with silica as dielectric and I have measured them with an impedance analyzer, the Nyquists I was able to model them with the radle circuit. By plotting the normalized active and reactive powers for the complex power as a function of the frequency I have observed that they intersect at one point, even at one of the capacitors it crosses at two points. I have read in supercapacitor studies that the crossing of normalized powers is attributed to dielectric relaxation, but I do not completely understand what it means and why this happens with my capacitors in which I do not use any electrolyte.