International Colloquium Tribology
ict
expert verlag Tübingen
125
2022
231
Mechanics of tribo-oxidation in high-purity copper
125
2022
Julia S. Rau
Shanoob Balachandran
Reinhard Schneider
Peter Gumbsch
Baptiste Gault
Christian Greiner
ict2310309
23rd International Colloquium Tribology - January 2022 309 Mechanisms of tribo-oxidation in high-purity copper Julia S. Rau Institute for Applied Materials (IAM), Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany IAM-CMS MicroTribology Center µTC, Karlsruhe, Germany Shanoob Balachandran Max-Planck-Institut für Eisenforschung GmbH, Department of Microstructure Physics and Alloy Design, Düsseldorf, Germany Reinhard Schneider Laboratory for Electron Microscopy (LEM), Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany Peter Gumbsch Institute for Applied Materials (IAM), Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany Fraunhofer IWM, Freiburg, Germany Baptiste Gault Max-Planck-Institut für Eisenforschung GmbH, Department of Microstructure Physics and Alloy Design, Düsseldorf, Germany Department of Materials, Royal School of Mines, Imperial College London, London, UK Christian Greiner Institute for Applied Materials (IAM), Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany IAM-CMS MicroTribology Center µTC, Karlsruhe, Germany Corresponding author: greiner@kit.edu 1. Introduction Metals’ surfaces subjected to tribological loading often show a limited lifetime due to accelerated oxidation a critical issue in many applications such as wind turbines, hip implants or micro-mechanical systems. Tribo-oxidation takes place by chemical reactions of the sliding partners or the surrounding medium. Such reactions may alter the friction and wear behaviour drastically. Yet, the mechanisms and pathways for tribo-oxidation, particularly in the very early stages of sliding, are insufficiently understood. Our goal is to identify the elementary mechanisms by paring sapphire spheres with polyand single crystalline high-purity copper plates. In earlier investigations the formation of copper oxides (Cu 2 O) during tribological loading was reported in this setup[1]. Here, tribo-oxidation took place at rates order of magnitudes faster than the native oxidation of copper under the same environmental conditions. Although tribo-oxidation was studied for many years, it stays overall elusive why and how oxides grow so quickly under tribological loading. 2. Materials and methods Using copper as a model system, tribologically-induced oxidation is systematically investigated by varying the sliding speed (0.1 - 50 mm/ s, 12 mm stroke) and test duration (13.5 - 67 h, 0.1 mm/ s) under mild tribological loading. Under theses low loads of 1.5 N and slow sliding speeds, a significant temperature increase in the contact can safely be excluded. In addition, loaded samples were exposed to the ambient environment for different exposure times (0 - 48 h). Experiments were performed in a strictly controlled atmosphere with a reciprocating linear sliding setup at room temperature and a constant number of sliding cycles (1,000). Sample preparation including heat treatment, grinding and (electro-) polishing are described in detail in [2], [3]. The final electro-polishing step was performed right before the tribological tests to achieve a sample surface with minimal native oxidation and plastic deformation. State of the art scanning and transmission electron microscopy (SEM and TEM) techniques as well as atom probe tomography (APT) were used to investigate and characterize the resulting oxides within the plastically deformed subsurface. To protect the loaded surfaces from ion beam damage when milling, two protective platinum 310 23rd International Colloquium Tribology - January 2022 Mechanisms of tribo-oxidation in high-purity copper layers (electron and ion beam) were deposited. Except the samples for the exposure time tests, all samples were constantly evacuated to a pressure below 1 mbar immediately after the end of each experiment. 3. Results and Discussion Figure 1 depicts the oxide thickness after 1,000 cycles for different sliding speeds, test durations and exposure times, measured in cross sectional scanning electron microscopy images. Figure 1b shows that the oxide thickness is controlled by the test duration rather than the sliding speed. After sliding stopped, almost no further oxide growth took place (Figure 1c). Hence, the oxide formation and growth is almost entirely associated with the tribological loading. Chemical interactions of sphere and plate are not detectable: XPS measurements of the sapphire sphere did not detect any Cu on the sphere and APT measurements of the copper plate did not find any Al. The atom probe tomography dataset in Figure 2 shows a pipe-like oxygen-rich feature within the tribologically deformed subsurface. Such features are associated with dislocations, decorated with oxygen along the dislocation line. Plastic deformation from sliding creates defects, such as these dislocations as well as grain and later also phase boundaries between the oxide and the copper matrix that act as high diffusivity pathways. Oxygen diffusion into the bulk as well as of copper towards the free surface along these defects control the oxide formation kinetics. We speculate that the origin of the linear oxide growth law with test duration (Figure 1b) lies in the trapping-effect of the dislocation for the diffusing species: Once a pipe is clogged, the diffusion along its core will decrease. However, the ongoing movement of the sphere ‘frees’ the dislocation from the trapped atoms and provides a fresh pathway, while the subsurface is supersaturated with oxygen. Figure 1: Quantified oxide thickness as a function of increasing (a) sliding speed, (b) test duration and (c) exposure time to the ambient environment after the tribological loading of copper polycrystals. Each data point represents one experiment. The standard deviation stems from the evaluation of multiple SEM images within one cross-section. [4] 23rd International Colloquium Tribology - January 2022 311 Mechanisms of tribo-oxidation in high-purity copper Figure 2: Atom probe tomography (APT) analysis from the middle of the wear track after 5,000 sliding cycles on a high purity copper single crystal with (111) outof-plane orientation and ±<0-11> sliding direction. (a) Three-dimensional reconstruction within the tribologically deformed subsurface shows an oxygen pipe feature with 11 at% of oxygen in the Cu matrix. [4] 4. Conclusion Investigating the fundamental mechanisms of tribo-oxidation with high purity copper as a model system systematically, allowed to draw the following conclusion: The oxidation process is governed by diffusion processes along defects generated while sliding such as dislocations, phase and grain boundaries. Diffusion along dislocations together with the sliding sphere results in a linear oxide growth law with time which is different from the parabolic law observed in native oxidation. Understanding the fundamental mechanisms of tribo-oxidation will eventually help for tailoring materials properties for little wear and low friction. References [1] C. Greiner, Z. Liu, R. Schneider, L. Pastewka, and P. Gumbsch, “The origin of surface microstructure evolution in sliding friction,” Scr. Mater., vol. 153, pp. 63-67, 2018. [2] C. Greiner, Z. Liu, L. Strassberger, and P. Gumbsch, “Sequence of Stages in the Microstructure Evolution in Copper under Mild Reciprocating Tribological Loading,” ACS Appl. Mater. Interfaces, vol. 8, no. 24, pp. 15809-15819, 2016. [3] Z. Liu, T. Höche, P. Gumbsch, and C. Greiner, “Stages in the tribologically-induced oxidation of high-purity copper,” Scr. Mater., vol. 153, pp. 114- 117, 2018. [4] J. S. Rau, S. Balachandran, R. Schneider, P. Gumbsch, B. Gault, and C. Greiner, “High diffusivity pathways govern massively enhanced oxidation during tribological sliding,” Acta Mater., accepted Sept 2021.
