Advancing Solid Interfaces and Lubricants by First Principles Material Design

Informacje na temat projektu

SLIDE

Identyfikator umowy o grant: 865633

DOI

10.3030/865633

Projekt został zamknięty

Data podpisania przez KE 14 Maja 2020

Data rozpoczęcia 1 Listopada 2020

Data zakończenia 31 Października 2025

Finansowanie w ramach

• EXCELLENT SCIENCE - European Research Council (ERC)

Koszt całkowity

€ 1 403 662,00

Wkład UE

€ 1 403 662,00

Koordynowany przez

ALMA MATER STUDIORUM - UNIVERSITA DI BOLOGNA
Italy

Periodic Reporting for period 4 - SLIDE (Advancing Solid Interfaces and Lubricants by First Principles Material Design)

Okres sprawozdawczy: 2025-05-01 do 2025-10-31

Podsumowanie kontekstu i ogólnych celów projektu

Friction and wear are ubiquitous phenomena that result in massive energy and environmental costs. By improving tribology technologies, largely based on advanced materials, a huge amount of energy could be saved, with consistent reduction in fuel consumption and carbon dioxide emissions. However, this is not an easy task as friction is governed by atomistic processes that occur at the sliding buried interface, which is very difficult to monitor in real time by experiments. Simulations can play a crucial role here, in particular those based on a quantum mechanical approach, which is essential for an accurate description of stress-induced chemical reactivity.
SLIDE aimed at achieving two main goals by porting the most advanced paradigms for materials modeling and design into the field of tribology:
i) harness tribochemical reactions to reduce interface friction. SLIDE focused on developing environmental-friendly alternatives to commercial additives used in engine oils;
ii) developing a high-throughput workflow for screening solid interfaces and creating public databases of adhesion and shear strength for many material pairs. The database analysis will highlight relationships and general trends.

SLIDE successfully achieved both objectives. i) It introduced machine-learning–accelerated molecular dynamics to computational tribology, greatly improving the efficiency of ab initio simulations while preserving accuracy. This made realistic in-silico tribology experiments possible and, together with laboratory experiments, led to key discoveries — including operando formation of solid lubricants and carbon coatings, as well as superlubricity from plant-based additives for greener aqueous lubricants. Real-time tribofilm formation by ZDDP was also captured with large-scale ML-MD.
ii) High-throughput first-principles calculations were ported into tribology, developing software to automatically screen adhesion and shear strength across many interfaces. Database analysis revealed links between adhesion and surface energies and identified strategies to systematically increase or decrease interfacial adhesion.

Prace wykonane od początku projektu do końca okresu sprawozdawczego oraz najważniejsze dotychczasowe rezultaty

The project activity has been organized in different research lines, that are here summarized together with the main results obtained for each line.

Adsorption and dissociation of (additive) molecules on solid surfaces
The program XSORB for automated adsorption studies was developed and released. Adsorption/dissociation of conventional lubricant additives, such as MoDTC and ZDDP, and carbon-based compounds were studied taking into account the effects of substrate oxidation. Mechanisms of passivation by the dissociative adsorption of H2O/O2/H2 molecules were also identified on diamond, phosphorene, and MXenes. XSORB was also successfully applied to predict the performance of catalysts, paints, and anti-wear additives. Rationally designed doping of substates enhanced the adsorption of environmental molecules and lubricant additives on carbon-based substrates.

Tribofilm formation
The role of interfacial confinement in promoting molecular dissociation was identified and explained by analyzing load-induced changes in the electronic structure of simple molecules; a correlation between the electron redistribution in solids and their stress–strain curves was discovered; in-plane compression was also found to reduce diamond wear resistance.
The formation mechanisms of MoS2 from MoDTC, graphene/carbon tribofilms from aromatic molecules/hydrocarbons, 2D selenides from Se nanopowder, and polyphosphate tribofilms from ZDDP were unraveled by AIMD and ML-MD. Substrate oxidation effects were clarified. Strategies to accelerate tribofilm synthesis were suggested.

Frictional performance
A multiscale ab initio–Green’s-function MD method was developed and applied to calculate the friction coefficient of diamond and quantify the effects of passivation and sliding velocity.
A software package, SCS, for active-learning MLPs was released; ML-MD assessed the tribological performance of carbon tribofilms and organophosphorus compounds. Molecules of plant origin were demonstrated to be effective additives for water lubrication. The performance of different solid lubricants, including MXenes, black phosphorus, transition-metal carbides, and electrides, was evaluated by combining simulations and experiments.

High-throughput screening of solid interfaces
TRIBCHEM, an advanced software for surface matching and high-throughput calculation of interface adhesion and shear strength, was developed and released. Databases were populated for metallic heterostructures and FAIR principles applied. ML revealed a relationship linking interfacial adhesion to the energies of the two surfaces in contact; The adhesion of graphene, MoS2, MoSe2, MXenes, phosphorene, PTFE, and carbon films on different substrates was screened and general prescriptions for tuning layer adhesion on substrates were identified to improve the functionality of solid lubricants. The use of Bayesian algorithms to calculate interface PESs was explored.
Surface chemical modification to tune adhesion
A database for adatom chemisorption was generated. The tribological effects of surface chemical modification of metals, MXenes and diamond/DLC were examined. The surface energy in high-entropy alloys was shown to be controlled by the surface composition. Databases analysis revealed universal trends: boron as an adhesion enhancer and fluorine as an adhesion reducer; the latter property was linked to PFAS functionality.

The above results were published in more than 60 papers and presented in about 30 invited talks and several other oral and poster presentations at international conferences.

Innowacyjność oraz oczekiwany potencjalny wpływ (w tym dotychczasowe znaczenie społeczno-gospodarcze i szersze implikacje społeczne projektu)

The pioneering use within SLIDE of machine-learning molecular dynamics (ML-MD), as well as first-principles high-throughput calculations, has significantly advanced the field of computational tribology.
The central idea of SLIDE was to move beyond purely atomistic descriptions of tribological processes and explicitly account for the role of electrons at nano-asperity contacts. This was already a major innovation compared with previous simulations of tribochemical processes based on empirical interaction models, which are often inadequate because stress-induced reactivity is fundamentally quantum-mechanical in origin.
The pioneering adoption of ML-MD further advanced SLIDE simulations, dramatically improving computational efficiency while preserving accuracy. This made it possible, for example, to reveal the atomistic mechanisms behind the operando formation of slipper/anti-wear films from commercial additives and hydrocarbons, and to computationally design more environmentally friendly additives for water-based lubricants. These achievements highlight the power of ML-MD to complement experiments in the rational design of friction-reducing materials, with societal impact given that friction accounts for nearly 30% of global energy consumption.
The use of high-throughput calculations was also transformative for computational tribology. Building databases according to FAIR principles brought materials informatics into the field, allowing us to establish direct links between tribological properties and electronic structure, discover relationships useful to predict the adhesion of an interface from the surface energies of its constituents, and rationally design chemical modification to tune it.

fig.jpg