Role of activation energy in resistance drift of amorphous phase change materials

Abstract

The time evolution of the resistance of amorphous thin films of the phase change materials Ge 2 Sb 2 Te 5 , GeTe and AgIn-Sb 2 Te is measured during annealing at T = 80 • C. The annealing process is interrupted by several fast temperature dips to determine the changing temperature dependence of the resistance. This procedure enables us to identify to what extent the resistance increase over time can be traced back to an increase in activation energy E A or to a rise of the prefactor R * . We observe that, depending on the material, the dominating contribution to the increase in resistance during annealing can be either a change in activation energy (Ge 2 Sb 2 Te 5 ) or a change in prefactor (AgIn-Sb 2 Te). In the case of GeTe, both contribute about equally. We conclude that any phenomenological model for the resistance drift in amorphous phase change materials that is based on the increase of one parameter alone (e.g., the activation energy) cannot claim general validity.

Structural change with the resistance drift phenomenon in amorphous GeTe phase change materials’ thin films

“…By fitting the experimental data to the commonly used power-law ρ(t) = ρ 0 (t + t 0 ) ν , where ρ is the resistivity and ρ 0 the initial resistivity, a drift exponent ν ~ 0.145 ± 0.02 at 50 °C (323 K) is deduced. This ν value is in good agreement with the results of [5]. In a second step, 3 distinct a-GeTe samples named D1, D2 and D3 were set at different resistance drift levels by thermal annealing at a constant temperature (50 °C under the crystallization temperature) for respectively 1, 15 and 30 min.…”

“…Different values of ν are observed as a function of the PCM alloy composition (Ge/Sb/Te ratio or dopant incorporation). For example ν ~ 0.1 for GST 225 [5]. Publications in the current literature aiming to understand the resistance drift mechanism agree on the fact that the drift is due to some kind of relaxation of the amorphous structure.…”

“…This binary alloy is the end-point of the wellstudied pseudo-binary line to which GST 225 belongs [12]. Its ν value is around 0.125 ( [5] and references therein). It is a good candidate for applications since its crystallization temperature is higher than that of GST 225.…”

“…Publications in the current literature aiming to understand the resistance drift mechanism agree on the fact that the drift is due to some kind of relaxation of the amorphous structure. However, they propose very different, and sometimes opposing, physical mechanisms, such as increase [6] or decrease [5,7] of disorder, increase [8] or decrease [9] of the defect density of states, drift impacted [10] or not impacted [11] by stress. This confused situation arises from the lack of structural studies of the a-phase as a function of ageing.…”

Morphology-dependent space charge polarization and dielectric relaxation of CdO nanomorphotypes

“…Dielectric semiconductor materials with high activation energy indicate that the material is more resistive against electron transport and also have high dielectric polarization/ dielectric constant value. [40][41][42][43] As the resistance increase, the high dielectric polarization energy in the materials gets dissipated in the form of heat which causes large dielectric loss values in the material. In the present study, the higher activation energy in spherical morphology is due to its large grain resistance and dielectric loss values over flakes and parallelepiped morphotypes.…”