Thermal conductivity and boundary resistance measurements of GeSbTe and electrode materials using nanosecond thermoreflectance

Abstract

Phase change memory (PCM) uses rapid heating and cooling to induce switching in sub-micron memory cells. The rapid rates of heating and nanoscale dimensions require accurate modeling of thermal transport phenomena in the constituent materials. This motivates improved understanding of the thermal properties of Ge 2 Sb 2 Te 5 (GST) thin films and PCM electrode materials. We report measurements of thermal conductivity and interface resistance of GST and electrode materials by applying nanosecond pump-probe thermoreflectance to multilayer structures of GST-C, GSTTiN, and GST-Ti. We measure the total thermal resistance of the stack from the transient thermal response, separating the intrinsic and boundary resistance terms using a 1-D resistor model of the stack. The intrinsic conductivities for GST are 0.20 W/(m K) for GST-C, 0.33 W/(m K) for GST-TiN, 0.27 W/(m K) for low temperature deposited GST-Ti, and 0.69 for high temperature deposited GST-Ti. The thermal boundary resistances are 27.5 m 2 K/GW for GST-C, 5.2 m 2 K/GW for GST-TiN, 49.8 m 2 K/GW for low temperature GST-Ti, and 11.4 m 2 K/GW for high temperature GST-Ti.

Temperature-Dependent Thermal Properties of Phase-Change Memory Electrode Materials

“…Previous work shows data for GST-electrode TBR in the range of 5-50 m 2 K/GW [6], [7]. However, as shown in [6], electrodes with low thermal conductivity may not have high TBR. For electrode materials such as C, Ti, and TiN, TBR becomes comparable to intrinsic thermal resistance when their thicknesses approach ∼30 nm [6].…”

“…The TBR is a function of several properties, including phonon spectra and interface quality [4], [5]. Previous work shows data for GST-electrode TBR in the range of 5-50 m 2 K/GW [6], [7]. However, as shown in [6], electrodes with low thermal conductivity may not have high TBR.…”

Nanothermal characterization of amorphous and crystalline phases in chalcogenide thin films with scanning thermal microscopy

“…The TBC is determined using the acoustic mismatch model (AMM). 27 It is then considered as an equivalent cylinder representing the tip/sample contact area (see Methods), and the modeled thermal drops are calibrated to match the experimental values. Finally, the influence of the TBC is removed to determine the thermal conductivity of each phase and material.…”

“…the reciprocal of 1/R ts ). [26][27][28][29] The SThM response will strongly depend on the tipsample junction PCM as well as the tip-heater thermal conductance. 30 the true sample thermal conductivity may be extracted from the experimental SThM data.…”

Superlatticelike dielectric as a thermal insulator for phase-change random access memory

“…In the PCRAM, the electrode-phase-change-material interface plays an important role in the heat generation process. 12,13 To simplify the calculations without affecting the final results, the effects of the electrode-phase-change-material interface were not considered in the study to compare the effects of the dielectric materials on the cell performance. In the simulation, the thermal conductivities of the materials employed were important parameters, as shown in Table I.…”