Published: 24 January 2019

Under construction...

Published: 24 January 2019

Thanks to the continuous increase of computationnal power available at the national and international high-performance-computating centers, atomistic simulations in material science are nowaday able to consider sub-micrometric sample with a reasonable accuracy. This give birth to cross-scale simulations that, instead of applying sophisticated multi-scale approach, address the scientific problem directly at multi-scale (atom, dislocation, microstructure). Complex on-the-fly and post-process analyses are involved to process the tremendous amount of data generated by such large-scale approaches.

Here, we perform the experimentally-informed cross-scale atomistic simulation of the deformation of nanoporous gold. The systems containing 470 Mio atoms has required 10 Mio CPUh of highly parallelized molecular dynamics in order to generates approximately 165 TB and 35 TB of raw and post-processed data, respectively.

Fig. 1: Experimentally informed (ExIn, top) and geometrically constructed (GeCo, bottom) nanoporous gold atomistic configuration before (a,c) and (b,d) after 0.30 and 0.26 compressive strain. e) Comparison of the stress-strain curves of MD simulations on ExIn and GeCo and experiment. [SuperMuc Results Report, 2018]

References:
- https://www.lrz.de/services/compute/supermuc/magazinesbooks/2018_SuperMUC-Results-Reports.pdf
- http://dc.engconfintl.org/nanomechtest_vi/105/

Published: 24 January 2019

As Plasticity involves phenomena at multiple scales, there has been many approaches explored to transfer qualitative and quantitative informations through scales.

We propose a novel approach to the transfer of slip accross interfaces to bridge the atomic-scale with the contimuum-scale. In particular, we relie on the determination of the properties of the dislocations and interfaces that can be transfer to higher-scale/continuum models. As an example, Fig. 1 shows the determination of the equivalent Burgers vector for an dislocation absorbed at an interface (here, a pure twist grain boundray). By applying an appropriate laod, the dislocation will initiate a slip transfer to the neigboring grain, which will depend on the properties of system "dislocation/interface".

Fig. 1: Excess in strain energy of a dislocation (edge or screw) absorbed at a high angle symetric Σ19 twist grain boundary. Determination of the associated Burgers vector. [unpublished work]

Within the context of damage induced by focused ion beam (FIB) irradiation in silicon, we propose a scale-bidging approach to transfer strain information from the atomic-scale to the meso-scale.

In particular, the concept of eigenstrain offers a versatile generic framework for the description of inelastic deformation that acts as the source of residual stresses. FIB milling used for nanoscale machining is accompanied by target material modification by ion beam damage having residual stress consequences that can be described in terms of eigenstrain. Due to the lack of direct means of experimental determination of residual stress or eigenstrain at the nanoscale, we adopt a hybrid approach that consists of eigenstrain abstraction from molecular dynamics simulation, its application within a finite element simulation of a flexible silicon cantilever, and satisfactory comparison of the prediction with experimental observation.

Fig. 2: The illustration of (a) molecular dynamics model setup for the simulation of grazing ion irradiation, and (b) depth profile of irradiation induced strain as determined by atomistic calculation.
[From Korsunky, Guénolé et al., Materials Letters 185 (2016) 47-49]

Published: 24 January 2019

Impact of temperature on FIB irradiation induce damage: Ti and its alloys as examples of cryogenic focused ion beam milling of environmentally-sensitive materials

Hydrogen pick-up leading to hydride formation is often observed in commercially pure Ti (CP-Ti) and Ti-based alloys prepared for microscopic observation by conventional methods, such as electro-polishing and room temperature focused ion beam (FIB) milling. Here, we demonstrate that cryogenic FIB milling can effectively prevent undesired hydrogen pick-up. Specimens of CP-Ti and a Ti dual-phase alloy (Ti-6Al-2Sn-4Zr-6Mo, Ti6246, in wt.%) were prepared using a xenon-plasma FIB microscope equipped with a cryogenic stage reaching −135°C. Transmission electron microscopy (TEM), selected area electron diffraction, and scanning TEM indicated no hydride formation in cryo-milled CP-Ti lamellae. Atom probe tomography further demonstrated that cryo-FIB significantly reduces hydrogen levels within the Ti6246 matrix compared with conventional methods. Supported by molecular dynamics simulations (See Fig. 1), we show that significantly lowering the thermal activation for H diffusion inhibits undesired environmental hydrogen pick-up during preparation and prevents pre-charged hydrogen from diffusing out of the sample, allowing for hydrogen embrittlement mechanisms of Ti-based alloys to be investigated at the nanoscale. [See Chang et al. Nature Communications, 2019]

Fig. 1: Molecular dynamics (MD) simulations. a MD snapshots at different stages of the irradiation process of Xe at 30 kV in (0001)-oriented pure Ti surface at normal incidence. Atoms initially within 1 nm of the surface are indicated by the colour scale. The bulk atoms are shown in light purple. The scale bar is 10 nm. b Position of Ti atoms after irradiation (final distance to surface) as a function of their position before irradiation (initial distance to surface). c Similar graph as in (b) following Ga implantation at acceleration voltages of 5 kV (pink) and 30 kV (blue). d Similar graph as in (b) following Xe and Ga implantation at room temperature (RT) (dark red and orange respectively) and at cryogenic temperature, i.e., 138 K (light and dark blue respectively) [From Chang et al. Nature Communications, 2019]

Focused ion beam machining of strained silicon

The focused ion beam (FIB) technique has established itself as an indispensable tool in the material science community, both to analyze samples and to prepare specimens by FIB milling. In combination with digital image correlation (DIC), FIB milling can, furthermore, be used to evaluate intrinsic stresses by monitoring the strain release during milling. The irradiation damage introduced by such milling, however, results in a change in the stress/strain state and elastic properties of the material; changes in the strain state in turn affect the bonding strength, and are hence expected to implicitly influence irradiation damage formation and sputtering. To elucidate this complex interplay between strain, irradiation damage and sputtering, we perform TRIM calculations and molecular dynamics simulations on silicon irradiated by Ga + ions, with slab and trench-like geometries, whilst simultaneously applying uniaxial tensile and compressive strains up to 4% (Fig. 2). In addition we calculate the threshold displacement energy (TDE) and the surface binding energy (SBE) for various strain states. The sputter rate and amount of damage produced in the MD simulations show a clear influence of the strain state. The SBE shows no significant dependence on strain, but is strongly affected by surface reconstructions. The TDE shows a clear strain-dependence, which, however, cannot explain the influence of strain on the extent of the induced irradiation damage or the sputter rate. [See Guenole et al. Applied Surface Science, 2017]

Fig. 2: Induced damage in collision cascade simulations of 5 kV Ga + ions in Si at 300 K. (a) Perspective view and (b) projected view of a slice with a thickness of 4 nm located at the simulation box center, i.e., with the highest concentration of incoming ions. (c) Perspective view where atoms in a perfect diamond cubic structure environment (until the second neighbor) are removed to highlight the irradiation-induced damage. toms are colored according the diamond structure identification method implemented in Ovito: blue – diamond cubic, cyan – diamond cubic structure with 1st and or 2nd neighbor in non-diamond structure environment, e.g. surfaces, white – non-diamond structure. [From Guenole et al. Applied Surface Science, 2017]

Influence of intrinsic strain on irradiation induced damage: the role of threshold displacement and surface binding energies

Current models for the formation of irradiation damage and the sputter yield are based on two key parameters, the threshold displacement energy (TDE) and surface binding energy (SBE), which are usually determined from unstrained systems with idealized surfaces. Here we use atomistic simulations to determine the TDE and SBE for strained silicon and aluminum and compare the results to full cascade simulations. A clear, material class dependent influence of the strain state on the TDE is observed, and surface amorphisation is shown to significantly increase the SBE of {001} surfaces. [See Guenole et al., Materials and Design, 2016]

Fig. 3: Average threshold displacement energy as function of applied strain mode, for Si. Applied strain modes considered are uniaxial ([010], red), equibiaxial (along [010] and [100] blue) and hydrostatic (green). Error bars correspond to the standard error of the data set (1000 independent values each). Solid lines are guides to the eye.
Induced damage in collision cascade simulations of 5 keV Ga + ions in Si at 300 K. Perspective view and projected view of a slice with a thickness of 4 nm located at the simulation box center, i.e. the highest concentration of incoming ions. [From Guenole et al., Materials and Design, 2016]

Published: 24 January 2019

Under construction...

Published: 24 January 2019

Crystalline nano-objects like nano-pillars, nano-wires, nano-wiskers, etc., are often charaterized by a very low density or a complete abscence of structural defects. In addition, materials that are commonly brittle at the micro-scale, exhibit ductile behavior at the nano-scale. In such a case, the plasticity is initiated from surfaces. The onset of plasticity thus depend on dislocation nucleation instead of dislocation multiplication/interactions/transmission.

I investigate the onset of plasticity in silicon nanstructures, in particular in nano-pillars/wires.

We evidence the importance of considering the perimeter instead of the diameter while comparing nano-object (Fig 1). Also, the detailed strcture of the nano-object surfaces appear to be crucial for dislocation nucleation (Fig. 2).

Supported by abinitio calculations, we evidenced the activation of slip wihtin the {110} planes, that is unexpected for diamond silicon (Fig. 3). We demonstrated this feature to be the consequence of the nano-scale.

Real nano-structure are often coated with an amorphous outer-shell. In this context, we evidenced an onset of plasticity controlled by defects located at the crystalline amorphous interface (Fig. 4).

Fig. 1: perimeter energy as a function of perimeter lenght [DOI: 10.4028/www.scientific.net/KEM.465.89]

Fig. 2: nucleation of a perfect dislocation in a cubic-shaped silicon nano-pillar [DOI: 10.1088/0965-0393/19/7/074003]

Fig. 3: dislocation glide within a {110} plane in a silicon nano-pillar under compression. [DOI: 10.1016/j.actamat.2011.08.039]

Fig. 4: On set of plasticity controlled by interfacial defects in core-shell crystalline-amorphous nano-structured [DOI: 10.1103/PhysRevB.87.045201].