Abstrakt:

High Asymmetric Heat Transport in Multilayer Phase Change Materials

Timm Swoboda1, Katja Klinar2, Andrej Kitanovski2 and Miguel Muñoz Rojo1;

1University of Twente, Netherlands; 2University of Ljubljana, Slovenia

In this work, we determine how a multilayer structure made of different types of nanoscale phase change materials results in a highly non-linear and asymmetric heat flow depending on thermal bias directionality. This will set the basis of solid-state thermal diodes that can be scaled up for integration in energy management, conversion or storage applications. Thermal diodes (TD) or rectifiers are capable to manage heat in a similar manner as how electronic diodes control electricity, propagating heat preferably in one direction.[1] The rectification ratio is defined as, RR = (Qfwd–Qrev)/Qrev, where Qfwd and Qrev correspond to the heat flux in the forward and reverse direction, respectively. Unfortunately, current solid-state thermal diodes typically present moderate RRs, complex fabrication designs, and/or lack of operating temperature tuneability.[1]

Phase change materials (PCM) have become popular for the development of TDs due to their thermal conductivity (k) change during the phase transition at a critical temperature (Tcrit), e.g. VO2 (klow = 1.5 W/(m×K) to khigh = 3.5 W/(m×K) at Tcrit~340 K).[1,2] These materials are often combined with phase invariant materials (PIM) to develop TDs.[1] As an example, in a PIM/PCM structure based on VO2, when the heat source is at temperatures T> Tcrit and close to the PCM side, both PIM and PCM conduct the heat well (Qfwd). The situation is reversed when the heat source is applied to the PIM (Qrev). This typically leads to RR larger than those found for individual PCMs.[1]

Here, we used finite element modeling (COMSOL®) to develop a versatile TD based on a novel multilayer PCM/PIM structure. Our design includes an alternating combination of two PCM layers with two PIM layers. The total length of the multilayer structure was set to 1 μm, while the thickness of the individual layers was varied at the nanoscale. We selected carefully different types of PCM and PIM materials, whose experimental thermal properties were extracted from literature, to find the optimum PCM/PIM configuration that led to the highest RR. Then, we applied a temperature difference across the two ends of this structure. RRs larger than 100% were observed when we used Si and SiO2 as PIMs and Ag2Te and Ag2S0.6Se0.4 as PCMs with transition temperatures at 420 K and 360 K, respectively.[3] Compared to the state of the art[1] of PCM/PIM diodes, this configuration represents one of the highest RRs at room temperature as well as offers new possibilities for advanced thermal control. Additionally, from the point of view of applicability, this thermal diode offers numerous advantages over previously reported TDs, including simple design, scalability and operating temperature tunability.

To determine the potential of this thermal device for solid state refrigeration, we analyzed the effect of integrating this TD into a magnetocaloric (MC) device. For that purpose, we used a one-dimensional MC device consisting of gadolinium as MC material with two TDs at their ends, sandwiched between the heat sink and the heat source. The presence of TDs avoids the heat to flow back to the heat source improving the efficiency of the MC device.[4] We selected a PIM/PCM structure that worked optimally for the MC system operating temperature and we considered quasi-steady-state operation in an alternating magnetic field with 1 T change. We observed that the integration of this TD enables higher operating frequencies compared to the conventional active magnetic regeneration process, increasing the cooling power density. Beyond MC refrigeration, this versatile and unique TD can be used in other solid state refrigeration, heat pump and energy harvesting technologies to improve their performances or efficiencies.[1,4]

References

[1] Swoboda et al., Adv. Electron. Mater., 2021,  DOI: 10.1002/aelm.202000625

[2] Oh et al., Appl. Phys. Lett., 96, 2010

[3] Hirata et al., J. Electron. Mater., 49, 2020

[4] Kitanovski, Adv. Energy Mater., 10, 2020