Transferable, Deep-Learning-Driven Fast Prediction and Design of Thermal Transport in Mechanically Stretched Graphene Flakes

ACS Nano ◽  
2021 ◽  
Author(s):  
Qingchang Liu ◽  
Yuan Gao ◽  
Baoxing Xu
Author(s):  
Pamela M. Norris ◽  
Justin L. Smoyer ◽  
John C. Duda ◽  
Patrick E. Hopkins

Due to the high intrinsic thermal conductivity of carbon allotropes, there have been many attempts to incorporate such structures into existing thermal abatement technologies. In particular, carbon nanotubes (CNTs) and graphitic materials (i.e., graphite and graphene flakes or stacks) have garnered much interest due to the combination of both their thermal and mechanical properties. However, the introduction of these carbon-based nanostructures into thermal abatement technologies greatly increases the number of interfaces per unit length within the resulting composite systems. Consequently, thermal transport in these systems is governed as much by the interfaces between the constituent materials as it is by the materials themselves. This paper reports the behavior of phononic thermal transport across interfaces between isotropic thin films and graphite substrates. Elastic and inelastic diffusive transport models are formulated to aid in the prediction of conductance at a metal-graphite interface. The temperature dependence of the thermal conductance at Au-graphite interfaces is measured via transient thermoreflectance from 78 to 400 K. It is found that different substrate surface preparations prior to thin film deposition have a significant effect on the conductance of the interface between film and substrate.


2011 ◽  
Vol 134 (2) ◽  
Author(s):  
Pamela M. Norris ◽  
Justin L. Smoyer ◽  
John C. Duda ◽  
Patrick E. Hopkins

Due to the high intrinsic thermal conductivity of carbon allotropes, there have been many attempts to incorporate such structures into existing thermal abatement technologies. In particular, carbon nanotubes (CNTs) and graphitic materials (i.e., graphite and graphene flakes or stacks) have garnered much interest due to the combination of both their thermal and mechanical properties. However, the introduction of these carbon-based nanostructures into thermal abatement technologies greatly increases the number of interfaces per unit length within the resulting composite systems. Consequently, thermal transport in these systems is governed as much by the interfaces between the constituent materials as it is by the materials themselves. This paper reports the behavior of phononic thermal transport across interfaces between isotropic thin films and graphite substrates. Elastic and inelastic diffusive transport models are formulated to aid in the prediction of conductance at a metal-graphite interface. The temperature dependence of the thermal conductance at Au-graphite interfaces is measured via transient thermoreflectance from 78 to 400 K. It is found that different substrate surface preparations prior to thin film deposition have a significant effect on the conductance of the interface between film and substrate.


2021 ◽  
Author(s):  
Samuel Genheden ◽  
Ola Engkvist ◽  
Esben Jannik Bjerrum

We expand our recent work on clustering of synthesis routes and train a deep learning model to predict the distances between arbitrary routes. The model is based on an long short-term memory (LSTM) representation of a synthesis route and is trained as a twin network to reproduce the tree edit distance (TED) between two routes. The ML approach is approximately two orders of magnitude faster than the TED approach and enables clustering many more routes from a retrosynthesis route prediction. The clusters have a high degree of similarity to the clusters given by the TED-based approach and are accordingly intuitive and explainable. We provide the developed model as open-source (https://github.com/MolecularAI/route-distances).


2021 ◽  
Author(s):  
Samuel Genheden ◽  
Ola Engkvist ◽  
Esben Jannik Bjerrum

We expand our recent work on clustering of synthesis routes and train a deep learning model to predict the distances between arbitrary routes. The model is based on an long short-term memory (LSTM) representation of a synthesis route and is trained as a twin network to reproduce the tree edit distance (TED) between two routes. The ML approach is approximately two orders of magnitude faster than the TED approach and enables clustering many more routes from a retrosynthesis route prediction. The clusters have a high degree of similarity to the clusters given by the TED-based approach and are accordingly intuitive and explainable. We provide the developed model as open-source (https://github.com/MolecularAI/route-distances).


Author(s):  
Samuel Genheden ◽  
Ola Engkvist ◽  
Esben Jannik Bjerrum

Abstract We expand the recent work on clustering of synthetic routes and train a deep learning model to predict the distances between arbitrary routes. The model is based on an long short-term memory (LSTM) representation of a synthetic route and is trained as a twin network to reproduce the tree edit distance (TED) between two routes. The ML approach is approximately two orders of magnitude faster than the TED approach and enables clustering many more routes from a retrosynthesis route prediction. The clusters have a high degree of similarity to the clusters given by the TED-based approach and are accordingly intuitive and explainable. We provide the developed model as open-source.


Author(s):  
Stellan Ohlsson
Keyword(s):  

Sign in / Sign up

Export Citation Format

Share Document