Dynamic thermomechanical deformation map of FCC microparticles using additive micromanufacturing and machine learning

Project Information

DyThM-FCC

Grant agreement ID: 101064660

DOI

10.3030/101064660

Project closed

EC signature date 24 June 2022

Start date 1 October 2022

End date 30 September 2024

Funded under

• Marie Skłodowska-Curie Actions (MSCA)

Total cost

No data

EU contribution

€ 173 847,36

Coordinated by

MAX-PLANCK-INSTITUT FUR NACHHALTIGE MATERIALIEN GMBH
Germany

Periodic Reporting for period 1 - DyThM-FCC (Dynamic thermomechanical deformation map of FCC microparticles using additive micromanufacturing and machine learning)

Reporting period: 2022-10-01 to 2024-09-30

Summary of the context and overall objectives of the project

The core aim of DyThM-FCC is to push the envelope of additive micro-manufacturing (AµM) and micromechanical testing under extreme conditions and identify a constitutive strengthening law by using machine learning (ML). A novel printing method, based on localized electrodeposition in a voxel-by-voxel manner, will be used to print metallic microparticles of Ni, Cu and Co. Subsequently, these microparticles will be subjected to thermal treatments to control the internal microstructure, i.e. change the dislocation content and grain size. These metallic microparticles will be tested under a combination of unprecedented strain rates (SR) up to 1000/s and temperatures from -150°C up to 600°C for assessing their suitability and reliability in extreme applications (e.g. sensors, where are subjected to high frequencies, or catalysis, where high temperatures can be expected). Depending on the applied SR and temperature several deformation mechanisms can contribute towards the deformation of face centered cubic (FCC) microcrystals as a function of the stacking fault energy (SFE). Ni, Cu and Co has been selected for this study as their SFE is very different, i.e. 128, 75 and 15 mJ/m2, respectively. As such, a complete deformation map of pristine microscale pure FCC metals will be obtained as a function of temperature, SR, SFE and defect density, based on the stress-strain signatures, extracted thermal activation parameters such as activation volume/energy and microstructural characterization from scanning and transmission electron microscopy (SEM/TEM). Finally, all that information will be used to feed a ML strategy in which the parameters of a constitutive law will be extracted. DyThM-FCC will test the limits of cutting-edge technology in EU research arena by exceptional cooperation between complementing scientific fields.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

-> Using localized deposition in liquid, deposition of polycrystalline copper microscale dots was successfully achieved. But serious complications were encountered during the annealing of these dots into the envisioned single crystalline copper particles. This was owing to a combination of electrodeposition based chemical contaminants and their segregation during annealing, and copper oxidation (even under careful high vacuum conditions).
-> As mitigation strategy, thin film based solid state annealing was chosen as the fabrication method to manufacture pristine microparticles.
-> A variety of technical hurdles were overcome to successfully dewet the metal thin films (such as Ni) into microparticles including: i) contamination from the oven (due to the tungsten source), ii) oxidation even under high vacuum conditions resulting in rough surfaces on particles or uneven dewetting on the substrate surface, iii) influence/contamination from the substrate (for example a silicon substrate at high temperatures started reacting with nickel) and iv) oxidation layer formation immediately after the thin film deposition.
-> Following the successful protocol identification for microparticle dewetting, further optimization was required to identify the appropriate film thickness and the corresponding annealing time (for the film to break into particles) to produce a sample surface with appropriately sized/shaped microparticles (, with specific gaps between particles that are appropriate for micromechanical testing (the gaps between the particles should be larger than the flat punch indenter tip diameter). A segmentation algorithm based on Delaunay triangulation based python code was used to analyze the gaps between the particles.
-> Micromechanical testing of the nickel microparticles required further instrumentation development as the typical piezostack actuation methodologies resulted in too low signal-to-noise, especially under high speeds of testing required to achieve the unprecedented strain rates envisioned.
-> Post-processing of the obtained load and displacement data from the high-speed and cryogenic micromechanical testing required the development of custom Python and R analysis.
-> FEM simulations were conducted to interpret potential effect of the particle shape on the mechanical properties identified.
-> Molecular dynamics based simulations of the metal microparticle compressions were employed to further interpret and complement the deformation behaviors seen during experiments.
Achievements:
i) Protocol for finely tunable dewetting of thin films into appropriately sized and spaced metal microparticles were identified
ii) Protocol required for compression testing of the metal microparticles under extreme strain rates and temperatures were identified
iii) Protocols to post-process, analyse and extract the appropriate mechanical markers from the micromechanical tests under these extreme conditions were determined

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

Previous quantitative small scale mechanical studies on metal microparticles have predominantly focused on low or quasi-static strain rates (<0.1 s−1), leaving a critical gap in knowledge regarding the deformation behavior of micro/nanoscale metals including ones with FCC crystal structure like nickel at intermediate(<10/s) and high strain rates(>1000/s). This research is the first to fill this gap by employing a custom-modified piezo-based micromechanical testing platform, which allows for precise quantitative testing with stress-strain signatures across a wide range of conditions and facilitates better comparisons with molecular dynamics (MD) simulations.
For the first time, a comprehensive investigation into the mechanical behavior of nickel microparticles across an unprecedented range of strain rates and temperatures. To our knowledge, no prior research has explored such an extensive spectrum of microparticle mechanics, with strain rates ranging from 0.001 to 1000 s−1 at room temperature and from 0.1 to 10 s-1 at 128 K. This work addresses a significant gap in the understanding of deformation mechanisms of pristine microscale FCC metals, particularly under extreme loading conditions.
Based on these setbacks and numerous following experiments, the following conditions were successfully identified to dewet nickel thin films into pristine nickel microparticles: i) Single crystal sapphire was chosen as the substrate, ii) appropriate thin film thickness of ~100nm to provide appropriate sized/spaced particles were ii) the thin films on substrate were encapsulated in fused silica capsules with a combination of dry Argon and reducing hydrogen, iii) annealing of these capsules were done inside a high vacuum oven.
Distinct stress-strain behaviors: Nickel microparticles exhibit two unique stress-strain behaviors under compression at strain rates up to 1 s−1, influenced by the ability of pre-existing dislocations to escape via mechanical annealing during elastic loading (similar to metallic nanowhiskers) or interact within the particle. Regardless of the stress-strain signature type, surface dislocation nucleation is identified as the dominant deformation mechanism at yield, supported by a small activation volume (<3b³).
For the first time, time scales associated with mechanical annealing of the pre-existing defects at the microscale were identified. This led to a unique rate-dependent transition in deformation behavior of the nickel microparticles from abrupt to controllable deformation at yield.
Rate-dependent strength increase: The strength of nickel microparticles increases monotonically with strain rates upto 1000 s−1, attributed to the increased dislocation flux required to maintain the strain rate.
Consistent deformation mechanisms at low temperatures: At 128 K, the deformation mechanisms of nickel microparticles remain consistent with those at room temperature, showing only a marginal increase in strength due to reduced thermal activation energy.
Simulation-experiment alignment: Our results show a close alignment between experimental data and MD simulations, reducing the gap between them to one order of magnitude in size and three orders of magnitude in strain rates.