Optimizing diamond-like carbon coatings - From experimental era to artificial intelligence

Diamond-like carbon (DLC) coatings are widely used for numerous engineering applications due to their superior multi-functional properties. Deposition of good quality DLC is governed by energy per unit carbon atom or ion and plasma kinetics, which are independent parameters. Translating independent parameters to dependent parameters to produce a best DLC is subjected to deposition method, technology, and system configurations which may involve above 50 combinations of bias voltage, chamber pressure, deposition temperature, gas flow rate, etc. Hence DLC coatings are optimized to identify the best parameters which yield superior properties. This article covers DLC introduction, the role of independent parameters, translation of independent parameters to dependent parameters, and discussion of four generations of DLC optimization. The first-generation of DLC optimization experimentally optimizes the parameter-to-property relationship, and the second-generation describes multi-parameter optimization with a hybrid of experimental and statistical-based analytical methods. The third generation covers the optimization of DLC deposition parameters with a hybrid of statistical methods and artificial intelligence (AI) tools. The ongoing fourth generation not only performs multi-parameter and multi-property optimization but also use AI tools to predict DLC properties and performance with higher accuracy. It is expected that AI-driven DLC optimizations and progress in virtual synthesis of DLC will not only assist in resolving DLC challenges specific to emerging markets and complex environments, but will also become a pathway for DLC to enter a digital-twin era. © 2022 The Authors