Phonon engineering pulls heat from stacked 2D materials
- 著者:Ella Cai
- 公開：:2018-09-14
US researchers have found a way to double heat extraction from a transistor made from 2D materials.
“In the field of nano-electronics, the poor heat dissipation of 2D materials has been a bottleneck to fully realising their potential in enabling the manufacture of ever-smaller electronics while maintaining functionality,” said University of Illinois at Chicago engineer Amin Salehi-Khojin.
It transpires that Van der Waals interactions allow the 2D materials to be stacked, but don’t transfer heat between the layers very well.
For the project, the 2D material Ti3C2 was used – this is a ‘MXene’, which is a class of conductive 2D material.
Being electrically conductive, it could be both heater and temperature sensor – allowing heat transfer between the MXene layer and the silicon supporting wafer – through an intervening SiO2 layer – to be studied.
Measurements of thermal boundary conductivity were made like this, and with an aluminium oxide encapsulating layer added (see diagram).
Without encapsulation, conductance was 10.8MW/m2/K, according to the Advanced Materials paper ‘Enhanced thermal boundary conductance in few‐layer Ti3C2 MXene with encapsulation‘, which doubled to 19.5 MW/m2/K with encapsulation.
“While our transistor is an experimental model, it proves that by adding an additional, encapsulating layer to these 2D nanoelectronics, we can significantly increase heat transfer to the silicon base, which will go a long way towards preserving functionality of these components by reducing the likelihood that they burn out,” said Salehi-Khojin. “Our next steps will include testing out different encapsulating layers to see if we can further improve heat transfer.”
How does it work?
According to the paper:
Boltzmann transport modelling reveals that the thermal boundary conductance can be understood as a series combination of an external resistance between the MXene and the substrate, due to the coupling of low‐frequency flexural acoustic phonons to substrate modes, and an internal resistance between flexural acoustic and in‐plane phonon modes. It is revealed that internal resistance is a bottle‐neck to heat removal and that encapsulation speeds up the heat transfer into low‐frequency flexural acoustic modes and reduces their depopulation, thus increasing the effective thermal boundary conductance.
The University of Illinois at Chicago worked with the University of Massachusetts Amherst and Drexel University.
“In the field of nano-electronics, the poor heat dissipation of 2D materials has been a bottleneck to fully realising their potential in enabling the manufacture of ever-smaller electronics while maintaining functionality,” said University of Illinois at Chicago engineer Amin Salehi-Khojin.
It transpires that Van der Waals interactions allow the 2D materials to be stacked, but don’t transfer heat between the layers very well.
For the project, the 2D material Ti3C2 was used – this is a ‘MXene’, which is a class of conductive 2D material.
Being electrically conductive, it could be both heater and temperature sensor – allowing heat transfer between the MXene layer and the silicon supporting wafer – through an intervening SiO2 layer – to be studied.
Measurements of thermal boundary conductivity were made like this, and with an aluminium oxide encapsulating layer added (see diagram).
Without encapsulation, conductance was 10.8MW/m2/K, according to the Advanced Materials paper ‘Enhanced thermal boundary conductance in few‐layer Ti3C2 MXene with encapsulation‘, which doubled to 19.5 MW/m2/K with encapsulation.
“While our transistor is an experimental model, it proves that by adding an additional, encapsulating layer to these 2D nanoelectronics, we can significantly increase heat transfer to the silicon base, which will go a long way towards preserving functionality of these components by reducing the likelihood that they burn out,” said Salehi-Khojin. “Our next steps will include testing out different encapsulating layers to see if we can further improve heat transfer.”
How does it work?
According to the paper:
Boltzmann transport modelling reveals that the thermal boundary conductance can be understood as a series combination of an external resistance between the MXene and the substrate, due to the coupling of low‐frequency flexural acoustic phonons to substrate modes, and an internal resistance between flexural acoustic and in‐plane phonon modes. It is revealed that internal resistance is a bottle‐neck to heat removal and that encapsulation speeds up the heat transfer into low‐frequency flexural acoustic modes and reduces their depopulation, thus increasing the effective thermal boundary conductance.
The University of Illinois at Chicago worked with the University of Massachusetts Amherst and Drexel University.
