Prediction of Nanoscale Friction for Two-Dimensional Materials Using a Machine Learning Approach

Baboukani, Behnoosh Sattari; Ye, Zhijiang; Reyes, Kristofer G.; Nalam, Prathima C.

doi:10.1007/s11249-020-01294-w

Cited by 49 publications

(38 citation statements)

References 119 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This trend contrasts with the predictions of DFT calculations and machine learning models for these materials. [24][25][26] However, those calculations were for sliding between two TMD layers, as opposed to a tip sliding on a TMD sample as in our experiments and simulations. Therefore, the mechanisms proposed by previous calculations for intrinsic interlayer sliding of these materials do not necessarily apply to our case.…”

Section: Resultsmentioning

confidence: 96%

See 1 more Smart Citation

Nanoscale Friction Behavior of Transition-Metal Dichalcogenides: Role of the Chalcogenide

et al. 2020

View full text Add to dashboard Cite

Despite extensive research on the tribological properties of MoS2, the frictional characteristics of other members of the transition metal dichalcogenide (TMD) family have remained relatively unexplored. To understand the effect of the chalcogen on the tribological behavior of these materials and gain broader general insights into factors controlling friction at the nanoscale, we compared the friction force behavior for a nanoscale single asperity sliding on MoS2, MoSe2, and MoTe2 in both bulk and monolayer forms through a combination of atomic force microscopy (AFM) experiments and molecular dynamics (MD) simulations. Experiments and simulations showed that, under otherwise identical conditions, MoS2 has the highest friction among these materials and MoTe2 the lowest. Simulations complemented by theoretical analysis based on the Prandtl-Tomlinson model revealed that the observed friction contrast between theTMDs was attributable to their lattice constants, which differed depending on the chalcogen. While the corrugation amplitudes of the energy landscapes are similar for all three materials, larger lattice constants permit the tip to slide more easily across correspondingly wider saddle points in the potential energy landscape. These results emphasize the critical role of the lattice constant, which can be the determining factor for frictional behavior at the nanoscale.

show abstract

Section: Resultsmentioning

confidence: 96%

“…25 An increase of the energy barrier to sliding with increasing chalcogen size was also predicted using machine learning techniques for Mo-and W-based TMDs. 26 There has been no experimental validation of these predictions so far. ACS Nano 2020 https://doi.org/10.1021/acsnano.0c07558…”

mentioning

confidence: 99%

Nanoscale Friction Behavior of Transition-Metal Dichalcogenides: Role of the Chalcogenide

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Dong et al developed deep learning algorithms to predict the bandgaps of hybrids of graphene and h-BN with arbitrary supercell configurations [42]. Baboukani et al presented an ML method for predicting nanoscale friction in 2D materials [43]. Moreover, ML interatomic potentials have shown outstanding efficiency in predicting novel materials [44,45], lattice dynamics [46], estimating the thermal conductivity [47,48], and exploring the phononic properties of 2D materials [49].…”

Section: Introductionmentioning

confidence: 99%

Bandgap prediction of two-dimensional materials using machine learning

Liu

Zhang

et al. 2021

PLoS ONE

View full text Add to dashboard Cite

The bandgap of two-dimensional (2D) materials plays an important role in their applications to various devices. For instance, the gapless nature of graphene limits the use of this material to semiconductor device applications, whereas the indirect bandgap of molybdenum disulfide is suitable for electrical and photo-device applications. Therefore, predicting the bandgap rapidly and accurately for a given 2D material structure has great scientific significance in the manufacturing of semiconductor devices. Compared to the extremely high computation cost of conventional first-principles calculations, machine learning (ML) based on statistics may be a promising alternative to predicting bandgaps. Although ML algorithms have been used to predict the properties of materials, they have rarely been used to predict the properties of 2D materials. In this study, we apply four ML algorithms to predict the bandgaps of 2D materials based on the computational 2D materials database (C2DB). Gradient boosted decision trees and random forests are more effective in predicting bandgaps of 2D materials with an R2 >90% and root-mean-square error (RMSE) of ~0.24 eV and 0.27 eV, respectively. By contrast, support vector regression and multi-layer perceptron show that R2 is >70% with RMSE of ~0.41 eV and 0.43 eV, respectively. Finally, when the bandgap calculated without spin-orbit coupling (SOC) is used as a feature, the RMSEs of the four ML models decrease greatly to 0.09 eV, 0.10 eV, 0.17 eV, and 0.12 eV, respectively. The R2 of all the models is >94%. These results show that the properties of 2D materials can be rapidly obtained by ML prediction with high precision.

show abstract

“…In addition to these more macro-tribological approaches, some studies can also be found that tend to target even smaller scales [125]. For example, Sattari Baboukani et al [126] employed a Bayesian modeling and transfer learning approach to predict maximum energy barriers of the potential surface energy, which corresponds to intrinsic friction, of various 2D materials from the graphene and the transition metal dichalcogenide (TMDC) families when sliding against a similar material with the aim of application as lubricant additives. The input variables for the model in the form of different descriptors (structural, electronic, thermal, electron-phonon coupling, mechanical and chemical effects) were extracted from density function theory (DFT) and molecular dynamics (MD) simulation studies in literature.…”

Section: Surface Texturingmentioning

confidence: 99%

Current Trends and Applications of Machine Learning in Tribology—A Review

2021

View full text Add to dashboard Cite

Machine learning (ML) and artificial intelligence (AI) are rising stars in many scientific disciplines and industries, and high hopes are being pinned upon them. Likewise, ML and AI approaches have also found their way into tribology, where they can support sorting through the complexity of patterns and identifying trends within the multiple interacting features and processes. Published research extends across many fields of tribology from composite materials and drive technology to manufacturing, surface engineering, and lubricants. Accordingly, the intended usages and numerical algorithms are manifold, ranging from artificial neural networks (ANN), decision trees over random forest and rule-based learners to support vector machines. Therefore, this review is aimed to introduce and discuss the current trends and applications of ML and AI in tribology. Thus, researchers and R&D engineers shall be inspired and supported in the identification and selection of suitable and promising ML approaches and strategies.

show abstract

Prediction of Nanoscale Friction for Two-Dimensional Materials Using a Machine Learning Approach

Cited by 49 publications

References 119 publications

Nanoscale Friction Behavior of Transition-Metal Dichalcogenides: Role of the Chalcogenide

Nanoscale Friction Behavior of Transition-Metal Dichalcogenides: Role of the Chalcogenide

Bandgap prediction of two-dimensional materials using machine learning

Current Trends and Applications of Machine Learning in Tribology—A Review

Contact Info

Product

Resources

About