Numerical and experimental crossed analysis of coated nanostructures through nanoindentation

Abstract

Nanostructuration can bring unique functional properties to optical window surfaces, such as superhydrophobic and antireflective capacities. However, their sustainability is conditioned by the mechanical resistance of the nanostructures, which can exhibit high aspect ratios to meet the military industry requirements in terms of optical transmission. Thus, improving the mechanical strength of such surfaces without affecting their functional properties is a key challenge. In that respect, this work investigates the protective impact of an annealed alumina thin film on a nanostructured silicon surface with conical shape. First, the elasto-plastic properties (Young's modulus, yield stress and hardening modulus) of the untreated and heat-treated coating are extracted from nanoindentation experiments on a plane sample using a numerical approach. The latter relies on the finite element model updating method from a 2D axisymmetric finite element model of a dual nanoindentation test, combing Berkovich and cube corner geometries, designed by a methodology based on an a priori identifiability analysis using an indicator (I-index) to ensure a good conditioning of the inverse problem. Identification results reveal that the heat-treated coating is stiffer and harder, which is in accordance with the crystallisation phenomena highlighted by X-ray diffraction measurements. Thereafter, single nanostructure microcompression tests are implemented, and the obtained mechanical responses clearly illustrate the protective effect of the coating and emphasise different solicitation regimes. Simulations of microcompression tests using 2D axisymmetric and 3D finite element models which integrate the previously identified parameters on plane sample allow to corroborate some of the experimental observations. Lastly, two uncertain and yet essential nanostructure geometric parameters for accurate simulations are retrieved using the numerical methodology applied on plane sample and validated by comparing identified values with post-mortem microscopic observation of a tested nanocone. It is thus shown that well-designed nanoindentation experiments, using a priori identifiability analysis, allow to identify with confidence reliable constitutive material parameters which can be used to describe the mechanical behaviour of a coated nanostructure. This methodology undeniably simplifies the design and optimization of coated nanostructures by avoiding too many unnecessary cleanroom manufacturing steps.