Computer Simulations Lead to Production of Material Exhibiting Superlubricity

Scientists at the Argonne Leadership Computing Facility (ALCF) have identified a material that demonstrates superlubricity at the macroscale. Computer simulations using the Mira supercomputer have helped advance the design for a near-frictionless material. The ALCF is a DOE Office of Science User Facility.

Top: This large-scale simulation depicts a phenomena called superlubricity, or a condition of extremely low friction. The simulation reveals that this condition originates at the nanoscale when graphene atoms self-assemble into a tube-like scaffolding that reduces contact area, and friction. Bottom: In this schematic of the superlubricity system, the gold represents nanodiamond particles; the blue is a graphene nanoscroll; green shows underlying graphene on silicon dioxide; and the black structures are the diamond-like carbon interface. Credit: Sanket Deshmukh, Joseph Insley, and Subramanian Sankaranarayanan, Argonne National Laboratory

When Sanket Deshmukh, an Argonne researcher, was evaluating the simulation results of a potential new lubricant material, he observed a new phenomenon that had not been seen previously.

“I remember Sanket calling me and saying ‘you have got to come over here and see this. I want to show you something really cool',” said Subramanian Sankaranarayanan, Argonne computational nanoscientist, who led the simulation work at the ALCF.

The computer simulations revealed astounding information. When two lubricant materials, graphene and diamond-like carbon (DLC), slid against one other, the graphene rolled up and formed hollow cylindrical 'scrolls' that helped to dramatically reduce friction. These nanoscrolls symbolize a novel mechanism of superlubricity, which is a state at which practically no friction exists.

“The nanoscrolls combat friction in very much the same way that ball bearings do by creating separation between surfaces,” said Deshmukh, who finished his postdoctoral appointment at Argonne in January.

Approximately one-third of the fuel consumption of an automobile is devoted to overcoming friction. Hence, superlubricity is very much desired in this industry. Incessant friction wears down mechanical components and reduces their lifetime, a problem which superlubricity could overcome.

Before this computational work was done, Ali Erdemir, Diana Berman and Anirudha Sumant, scientists at Argonne, were conducting experiments on the hybrid material at the Center for Nanoscale Materials and the Tribology Laboratory at Argonne. For their experiment, small graphene patches were slid against a steel ball that was coated with DLC.

A very low friction coefficient was observed with the graphene-DLC combination. However, they observed that the friction levels fluctuated for reasons unknown. Furthermore, they observed that in humid conditions, the friction coefficient increased to a level that was approximately 100 times that of the friction coefficient measured in dry environments.

The team collaborated with Deshmukh and Sankaranarayanan to perform computational work. The experimental conditions were replicated on the ALCF’s 10-petaflops IBM Blue Gene/Q supercomputer, Mira. Large-scale molecular dynamics simulations were performed to find out the mechanisms involved in superlubricity.

The research team discovered these graphene nanoscrolls, and their instability helped explain the fluctuating friction levels of the material. A repeating pattern was observed in which the hollow nanoscrolls were formed but then collapsed due to the load’s pressure.

“The friction was dipping to very low values at the moment the scroll formation took place and then it would jump back up to higher values when the graphene patches were in an unscrolled state,” Deshmukh said.

In order to overcome this problem, the team included nanodiamond particles in their simulations to determine if the nanoscrolls could be stabilized and made more permanent. The simulations showed that the graphene patches rolled around the nanodiamonds, and the scrolls were held in their place. This provided sustained superlubricity. In new experiments using nanodiamonds, the simulation results confirmed the findings.

Sankaranarayanan commented:

The beauty of this particular discovery is that we were able to see sustained superlubricity at the macroscale for the first time, proving this mechanism can be used at engineering scales for real-world applications. This collaborative effort is a perfect example of how computation can help in the design and discovery of new materials.

However, the effects of water on this material were not addressed with the addition of nanodiamonds. The scroll formation was suppressed by the water, and the graphene’s adhesion to the surface was increased. Though this drastically reduces the various possible applications of this material, its ability to maintain its superlubricity in dry environments is still an important breakthrough.

This hybrid material holds promise for various applications in dry environments, which include wind turbine gears, computer hard drives, and mechanical rotating seals for nanoelectromechanical and microelectromechanical systems. Furthermore, this material can be applied by drop casting, which is a cost-effective deposition method. The materials in solution are sprayed on to moving mechanical parts, and after evaporation, diamond-like carbon is left on one side, while nanodiamonds and graphene are left on the other side.

The nanoscroll mechanism is expected to induce future efforts for developing materials required for superlubricity applications. The Argonne team are continuing their computational studies to find out ways to overcome the problems faced due to water.

“We are exploring different surface functionalizations to see if we can incorporate something hydrophobic that would keep water out,” Sankaranarayanan said. “As long as you can repel water, the graphene nanoscrolls could potentially work in humid environments as well.”

The Mira supercomputer played a critical role in the discovery of the nanoscroll. For humid environments, the researchers had to simulate up to 10 million atoms, and for dry environments they had to simulate up to 1.2 million atoms.

In order to perform the computationally demanding reactive molecular dynamics simulations, the Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS) code was utilized. Computational scientists used ALCF catalysts to address a performance bottleneck that they faced with the ReaxFF module of the code. This module is an add-on package that is necessary for modeling the chemical reactions that take place in the system.

LAMMPS and ReaxFF implementation was optimized by adding OpenMP threading, and MPI collectives were used in place of MPI point-to-point communication, with MPI I/O leveraged. This optimization was done by collaboration between ALCF catalysts, IBM, Sandia National Laboratories and the Lawrence Berkeley National Laboratory. The code was able to perform with double the speed  before.

“With the code optimizations in place, we were able to model the phenomena in real experimental systems more accurately,” Deshmukh said. “The simulations on Mira showed us some amazing things that could not be seen in laboratory tests.”

Sankaranarayanan greatly anticipates the future research that may be possible with the ALCF’s next-generation supercomputer, Aurora.

“Given the advent of computing resources like Aurora and the wide gamut of the available two-dimensional materials and nanoparticle types, we envision the creation of a lubricant genome at some point in the future,” he said. “Having a materials database like this would allow us to pick and choose lubricant materials for specific operational conditions.”

The findings from this study have been published as a paper entitled “Macroscale superlubricity enabled by graphene nanoscroll formation,” in Science Express.

The DOE’s “Innovative and Novel Computational Impact on Theory and Experiment” (INCITE) program allocated the computing time required for the project at the ALCF.

Other researchers who contributed to optimizing the code include Chris Knight, Nichols A. Romero and Wei Jiang from the ALCF; Tzu-Ray Shan from Sandia National Laboratories; Hasan Metin Aktulga from Lawrence Berkeley National Laboratory (now at Michigan State University); and Paul Coffman from IBM.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback