Peter Gumbsch

The mechanisms governing wear evolution in multilayer hard coatings remain a key challenge in tool tribology and dry machining. This study investigates the dry sliding behaviour of TiN/AlTiN multilayer-coated (thickness 1 μm/5 μm) as well as uncoated tungsten carbide (WC) cutting tools in a pin-on-disk configuration. Sliding tests were performed in a cylinder-on-line contact using commercial grooving inserts against AISI 1045 steel disks across a load–velocity matrix (20, 30 N; 10, 40, 80 m/min), with extended duration runs at 40 m/min to capture wear progression. The TiN/AlTiN-coated tools exhibited more stable friction coefficients and significantly reduced wear volumes compared to uncoated tools (73–92% lower specific wear rates, corresponding to 4×–12× improvement). The enhanced performance is attributed to the formation of a dense, oxide-rich tribofilm that effectively suppressed both adhesive and abrasive wear. In contrast, uncoated tools exhibited unstable frictional behaviour and severe material loss due to grooving and transfer. Wear evolution analysis using SEM/EDS, FIB cross-sectioning, WLI, and XPS depth profiling indicated gradual layer-by-layer wear progression in the TiN/AlTiN coating system, while uncoated tools showed accelerated nonlinear wear at higher velocities. The quantified wear coefficients together with the tribofilm chemistry provide mechanistic insight into coating wear and define key experimental inputs for future, physics-based tool-life modelling under dry machining conditions.

Lubricated friction is a multiscale problem where molecular processes dictate the macroscopic response of the system. Traditional lubrication models rely on semiempirical constitutive relations, which become unreliable under extreme conditions. Here, we present a simulation framework that seamlessly couples molecular and continuum models for boundary lubrication without fixed-form constitutive laws. We train Gaussian process regression models as surrogates for predicting interfacial shear and normal stress in molecular dynamics simulations. An active learning algorithm ensures that our model adapts in scenarios where common constitutive laws fail, such as at layering transitions. We demonstrate our approach for nanoscale fluid flow over rough and heterogeneous surfaces, paving the way for accurate boundary lubrication simulations at experimental length and timescales.