New modelling capability for nano-confined phase change materials

Lead Research Organisation: University of Warwick
Department Name: Physics


The global demand for smaller and more energy efficient devices has been sustained by a steady decrease in the scale on which silicon microelectronics can be manufactured, from 65nm processes in the mid 2000s to 14nm in the very latest Intel processors. To continue this trend beyond the mid 2020s devices with dimensions of just 1-2nm will be required, likely using alternatives to silicon.

In this regime, the cross section of a wire might be no more than 2x2 or 3x3 atoms across, where the relevant materials physics is dominated by surface and confinement effects leading to dramatically different structural and electronic properties to the corresponding bulk material. Such wires can be formed by crystallisation of a molten salt within carbon nanotubes (CNTs) of "Buckytubes", leading to the smallest cross section nano crystals possible, sometimes referred to as Feynman crystals.

Research into the fundamental materials physics of these CNT-encapsulated structures is still in its infancy, with UK experimentalists leading the way. Particularly exciting recent work by one of the applicants (Sloan) has demonstrated the possibility of these wires undergoing transitions between nano-crystalline structures with markedly different properties, in response to bending strain in the CNT. These "phase change" properties open the way for nanoscale electromechanical switches and non-volatile memory, as well as providing a playground for fundamental studies of phase changes at the smallest length scale possible in a material.

Our aim with the current project, inspired by these results, is to develop a computational modelling capability to aid in interpretation of experiments, understand the origin of the phase change behaviour, and guide our experimental colleagues toward compounds with potentially advantageous properties. Counterintuitively, due to a reduction in symmetry, the computational expense of simulating nanowires can be more demanding when compared to bulk crystals. We will address the limitations of currently available modelling tools when applied to these systems. This will involve significant modifications to existing software and a rigorous study of the various approximations one might employ to increase the tractability of simulations.

We will apply cutting-edge methods in structure prediction to these systems, a non-trivial exercise due to the possibility wires with non-crystalline (e.g. helical) symmetry, and connect directly to relevant experiments by computing spectra related to the encapsulated wire's electronic and vibrational properties. Finally, we will study the thermodynamics and kinetics of nano-crystalline phase change, developing an understanding of when and how rapidly structural changes are affected to assess the utility of this mechanism for device applications.

Planned Impact

Economy : Phase change properties of CNT-encapsulated nanowires have only recently been discovered, representing a technology readiness level (TRL) of 1-2. The prospect of attracting future industry investment depends critically on scientific investigation and validation of their utility, exactly the focus of current experimental work and this modelling proposal. There is potential for generation of new IP through identification of materials with improved phase change properties, new methods for inducing phase change in nanoscale systems and improved synthesis of encapsulated nanowires through understanding of defect energetics and dynamics. The economic impact of our research will lie in establishing the UK as the international leader in theory and modelling of these systems, ensuring that any resulting IP is retained, licensed and exploited by UK research groups.

Society : An potential application of encapsulated nanowire technology is as the basis of non-volatile phase change memory (PCM) with storage densities comparable to or even exceeding devices. Existing PCM memories have failed to gain traction, and have even been withdrawn from market due their low storage density in comparison to existing flash-based storage solutions. Successful development of CNT-encapsulated phase change technology could address this issue and become commonplace within the next few decades, reducing charging requirements of consumer devices and the 1% of world energy consumption currently accounted for by data centres.

Public : Previous research by the Warwick PI on biological control of amorphous to crystalline phase transitions in mineral nanoparticles attracted significant media attention with features on Sky News, CNN, and the BBC plus numerous newspaper articles and radio segments. This demonstrated that even rather specialist and fundamental work on nanomaterials can attract media interest and inform the public of how far computer simulation has advanced. We will work with the communications office at Warwick to promote the findings of our research, seeking to exploit public interest in nanotechnology and offset unfounded concerns of 'grey goo' against continuation of current trends in miniaturisation and storage density. The Cambridge PI has considerable experience with public engagement having presented his research to such diverse groups as schoolchildren, science journalists and members of parliament.

People : The project will result in two PDRAs and 1-2 associated DTA/DTC students trained in theory and modelling of nanostructures, and familiar with best practices in computational materials physics. Very little of the proposed research can be accomplished without significant modification to existing codes/packages and hence an understanding of their internal operation will be required. This will go some small way to addressing the dangerous "black box" mentality increasingly prevalent amongst junior scientists in materials modelling. These skills will be disseminated into academia/industry as these researchers continue on their careers.
Description We have successfully developed structure prediction methods for one-dimensional crystals confined inside carbon nanotubes and these have been applied in three publications plus another 2 under review. Our research has also encompassed calculation of functional properties of nanowires, demonstrating that (1) filling nanotubes with a 1D crystal can enhance their electrical properties and (2) wires just a few atoms in diameter grown inside carbon nanotubes can exhibit thermoelectric performance close to that long-predicted from idealised theoretical models - an order or magnitude better than the equivalent bulk material. In the case of (2) our work was the first to use experimentally verified structures for nanowires at this diameter. Both findings are useful steps toward increasing the power efficiency of next-generation devices based on sub-5nm components.
Exploitation Route Our findings open up the possibility of more efficient nano-electronics based on carbon nanotubes and 1D materials grown inside them than would have otherwise been possible. Other avenues for taking this research forward are still emerging as manuscripts based on work at the end of the project are still to be published.
Sectors Digital/Communication/Information Technologies (including Software),Electronics,Energy

Title PyGeTe 
Description This is a Fortran90 implementation of a semi-empirical potential for Germanium Telluride. The model itself was taken from the literature (10.1088/1367-2630/15/12/123006 ). It was created for use in Monte Carlo simulations of phase change under confinement as part of this project. However this particular model was found to be unsuitable for simulating GeTe interfaces and hence the code has been open-sourced in case other groups find it useful for Monte Carlo simulations of bulk phase change. 
Type Of Technology Software 
Year Produced 2020 
Open Source License? Yes  
Impact None as yet.