PSI - Issue 61

Enes Günay et al. / Procedia Structural Integrity 61 (2024) 34–41 E. Gu¨nay et al. / Structural Integrity Procedia 00 (2024) 000–000

35

2

At the grain level, metallic materials exhibit size-dependent behavior, as extensively discussed in the literature. (see e.g., Kim et al. (2006), Yalc¸inkaya et al. (2011), Yalc¸inkaya et al. (2018)). Numerous studies have explored the size e ff ects in nanoscratching tests (see e.g. Koch et al. (2009), Lee et al. (2018)). In general, size e ff ects in nanoscratching is described as the phenomenon where the yield stress and hardness of a material increase as the scale of the contact decreases (e.g., when indenter size or grain size is reduced) (see e.g., Beake et al. (2013)). The grain size e ff ect canbe described as the phenomenon, where, as the grains become smaller, accumulation of dislocations resulting from the misorientation with respect to the neighboring grains occurs more frequently, leading to an elevation in both hardness and yield strength. The lateral size e ff ect is expressed as the increase in scratch hardness when the size of the scratch declines. (see e.g., Kareer et al. (2016a)). Experiments have shown that scratch hardness is more sensitive to e ff ective grain size than indentation hardness, and that the lateral size e ff ect is reduced as the grain size is decreased. This leads to the conclusion that the lateral size e ff ect and grain size e ff ect collaborate, rather than simply superimpose, and their combined length scale contributes to the variation in scratch hardness (see e.g., Kareer et al. (2016b)). Moreover, the size e ff ect in nanoscratching has been the subject of investigation through molecular dynamics simulations. Materials such as cubic silicon carbide, polycrystalline iron, and single and polycrystalline aluminum have been examined in molecular dynamics studies (see e.g., Noreyan and Amar (2008), Gao et al. (2014), Junge and Molinari (2014), AlMotasem et al. (2017)). Crystal plasticity finite element method (CPFEM) o ff ers a strong approach for investigating the behavior of metal lic materials, such as failure characteristics of dual-phase steels and the e ff ects of grain morphology and texture in additively manufactured samples (see e.g. Acar et al. (2022), Aydiner et al. (2024) for examples). Similarly, this approach has been commonly used in the subject of nanoscratch testing. Researches conducted on polycrystalline copper that focus on material pile-up height during both nanoindentation and nanoscratching reveal the extent of plas tic deformation (see e.g., Wang et al. (2019b)). Furthermore, studies have explored the influence of crystallographic orientations and the plastic behavior around grain boundaries. CPFEM has been used in tandem with experiments on bi-crystal copper specimens to investigate the e ff ect of grain boundaries and orientations on hardness and pile-up (see e.g., Wang et al. (2022)). Studies on the e ff ect of surface roughness on gold and copper films during nanoscratching demonstrate that simulation results align with experiments, especially when normal forces are su ffi ciently high (see e.g., Nazemian and Chamani (2019)). Additionally, the choice of indenter type and orientation plays a significant role in the results, with conical indenters and pyramidal indenters like Berkovich indenters being the most commonly used. The orientation of the Berkovich indenter can vary as edge-forward, face-forward, or side-forward, and the e ff ect of this aspect on material pile-up and ploughing friction coe ffi cient has been thoroughly examined through experiments and finite element simulations (see e.g., Chamani and Ayatollahi (2016)). However, there is currently a gap in the literature concerning numerical studies utilizing the finite element method (FEM) to investigate grain size e ff ects in nanoscratching tests, representing an area that warrants further exploration and research. This paper focuses on investigating the grain size e ff ects in polycrystalline copper through the utilization of a lower-order strain gradient crystal plasticity framework, which has been incorporated as an ABAQUS user material subroutine. Strain gradient crystal plasticity frameworks have a historical application in modeling size-dependent ma terial behavior, and they can be categorized into two types: higher-order and lower-order models. Higher-order models are more complex, involving additional degrees of freedom and higher-order stresses, making their implementation more challenging (see e.g., Yalc¸inkaya et al. (2021)). In contrast, lower-order models are relatively easier to implement as they avoid these challenges, and in most cases, the accuracy loss resulting from neglecting higher-order stresses is not significant (see e.g., Han et al. (2005)). These models have previously been used to model the behaviour of additively manufactured materials (see e.g. Bulut et al. (2023)) and thin specimens (see e.g. Gu¨nay et al. (2023)). Polycrystal geometries are generated using Neper software (see e.g., Quey et al. (2011)). These geometries are then scaled to create specimens with varying grain dimensions, enabling the exploration of grain size e ff ects, and FEM simulations are carried out in ABAQUS. The structure of this paper is as follows. First, the constitutive model used in FEM simulations is described. Then, the details of the boundary value problem, such as boundary conditions, geometrical dimensions, and material prop erties, are given. Afterwards, the validation of the model through comparison with the literature is shown. Finally, the grain size e ff ect results in polycrystal copper specimens are presented and discussed.