Scholarlink

<p>High-throughput automated testing offers accelerated ways to discover and characterize novel tribological materials. Here, we describe the design of a custom, fully automated, parallelized ball-on-flat reciprocating tribometer capable of performing upward of 1000 friction experiments a day depending on contact conditions. Combinatorial physical vapor deposition was utilized to develop 448 Pt–Au alloy coatings spanning the full binary compositional range. Tribological performance was evaluated in both lab air and dry nitrogen environments, revealing multiparametric dependencies of friction on composition, hardness, reduced modulus, surface roughness, and sputtered atom kinetic energy. In dry nitrogen, ultralow friction coefficients (<i>µ</i><sub>ss</sub> &lt; 0.1) were observed for Pt-rich coatings, with friction strongly influenced by both composition and the kinetic energy of Pt atoms during deposition. The low friction behavior in dry <i>N</i><sub>2</sub>, attributed to the formation of tribofilms on Pt–Au, highlights the role of deposition conditions on surface tribochemical processes. Lab air sliding experiments showed large variations in friction coefficients (~ 0.2 to 1) across the entire compositional range and no trends with other modalities or modeled deposition atomistics. Benefits of adapting automation and/or parallelization to reduce operator and testing time were explored by calculating the total times for the presented dataset. Systems with parallelized friction probes as well as automated systems are shown to reduce operator time by 99% and testing time by 90% compared to conventional serial testing. This work demonstrates the power of high-throughput tribological methods to generate large, multimodal datasets, paving the way toward self-driving laboratories in tribology that combine mechanistic insights and machine learning-driven materials optimization.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Novel Materials Discovery via High-Throughput Automated Tribological Testing of Thin Films

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月5日

查看全文
  • Tomas F. Babuska,
  • Kyle Dorman,
  • Justin M. Hall,
  • Manish Jain,
  • Michael T. Dugger,
  • Brandon A. Krick,
  • Filippo Mangolini,
  • Frank W. DelRio,
  • David P. Adams,
  • Brad L. Boyce,
  • John F. Curry

摘要

High-throughput automated testing offers accelerated ways to discover and characterize novel tribological materials. Here, we describe the design of a custom, fully automated, parallelized ball-on-flat reciprocating tribometer capable of performing upward of 1000 friction experiments a day depending on contact conditions. Combinatorial physical vapor deposition was utilized to develop 448 Pt–Au alloy coatings spanning the full binary compositional range. Tribological performance was evaluated in both lab air and dry nitrogen environments, revealing multiparametric dependencies of friction on composition, hardness, reduced modulus, surface roughness, and sputtered atom kinetic energy. In dry nitrogen, ultralow friction coefficients (µss < 0.1) were observed for Pt-rich coatings, with friction strongly influenced by both composition and the kinetic energy of Pt atoms during deposition. The low friction behavior in dry N2, attributed to the formation of tribofilms on Pt–Au, highlights the role of deposition conditions on surface tribochemical processes. Lab air sliding experiments showed large variations in friction coefficients (~ 0.2 to 1) across the entire compositional range and no trends with other modalities or modeled deposition atomistics. Benefits of adapting automation and/or parallelization to reduce operator and testing time were explored by calculating the total times for the presented dataset. Systems with parallelized friction probes as well as automated systems are shown to reduce operator time by 99% and testing time by 90% compared to conventional serial testing. This work demonstrates the power of high-throughput tribological methods to generate large, multimodal datasets, paving the way toward self-driving laboratories in tribology that combine mechanistic insights and machine learning-driven materials optimization.