“…This sensitivity to the sample stoichiometry is also observed in other transport parameters, like the Hall or Seebeck coefficients, and resembles doping effects in semiconductors. In addition, ͑i͒ the temperature dependence of the Seebeck coefficient: [5][6][7] is clearly nonlinear, exhibiting well-defined extrema in most instances, ͑ii͒ small variations in the chemical composition give rise to sign reversals in the S(T) value, ͑iii͒ for a given sample stoichiometry it shows a strong dependence on the heat treatments applied to the sample, 15 and ͑iv͒ S(T) has large values when compared to those of disordered metallic systems. 23 Therefore, the situation is quite different from that observed in amorphous materials, where the S(T) curve is dominated by electron diffusion yielding a linear temperature dependence.…”