Using spherical indentation to measure the strength of copper-chromium-zirconium

Highlights

• The hardness of CuCrZr was measured using spherical nano-indentation.

• A strong indentation size effect was seen at low indentation strains.

• The indentation size effect was suppressed at higher strains.

Abstract

Precipitation hardened CuCrZr will be used in heat-sink components in the ITER tokomak and is a primary candidate for EU DEMO. The measurement of mechanical properties of irradiated CuCrZr using conventional, standardised techniques is difficult due to the challenges involved in working with radioactive material and the relatively large specimen size required. Spherical nano-indentation offers a technique to measure stress-strain properties from far smaller volumes than conventional tests. In this work, CuCrZr has been heat-treated at different temperatures to vary the Cr precipitate size and spacing. Spherical nano-indentation using multiple tip radii was then used to produce stress-strain curves for all samples, from which values of initial flow stress were calculated. It was found that there was a strong indentation size effect (ISE) in the stress required to initiate plasticity, however at higher indentation strains the flow stress became constant for tip radii, R, ≥8 μm. This suggests that at the initiation of plastic deformation the ISE is dominated by dislocation source activation but in later stages the interaction with microstructural material length-scales dominate the measured mechanical strength. The mechanical response of these small-scale tests is governed by multiple mechanisms, which convolute interpretation of data and must be considered when measuring the effects of irradiation on the mechanical properties.

Introduction

Due to its high thermal conductivity, high strength at elevated temperature, and commercial availability [1,2], precipitation hardened (PH) copper-chromium-zirconium (CuCrZr) alloy has for many years been used to make actively cooled plasma facing components in fusion reaction vessels [3]. To this effect, CuCrZr has also been selected as the heat-sink material for the International Thermonuclear Experimental Reactor (ITER) and is a primary candidate for EU DEMO and future power plants. The optimised chemical composition for ITER-grade CuCrZr specifies the alloying elements as being in the range 0.6–0.9 wt.% chromium and 0.07–0.15 wt.% zirconium with ≤0.15 wt.% impurities; peak strength is achieved by aging at 475 °C ± 5 °C for 3 h [4]. When used for this purpose, the alloy response under neutron irradiation must be considered. As previously driven by the fission industry, small-scale testing techniques are preferred for neutron-irradiated material due to the induced radioactivity of samples and the associated difficulty and cost in handling.

There are a range of small-scale experimental techniques that can be used to measure the mechanical properties of materials and investigate the mechanisms responsible for plastic deformation. Micro- and nano-pillar compression have extensively been used for this purpose [[5], [6], [7]], with in-situ studies providing additional information on fundamental processes such as slip step formation, and dislocation generation and movement [8,9]. Cantilever flexure is commonly used for experiments in fracture mechanics and cyclic fatigue [[10], [11], [12]]. Both these techniques require fabrication of miniaturised test pieces, typically carried out using a focussed ion beam (FIB), which can be expensive in terms of both time and facility costs, and data are subject to variation due to the current lack of a standardised approach for these techniques. One method that does not require intensive sample preparation is instrumented indentation testing (IIT), which is supported by ISO 14577 [13]. This technique is now highly automated and can perform a large number of tests in a short amount of time. Also, when sample size is limited (e.g. in thin films or ion-irradiated material), nano-indentation can be used to probe the very first few tens to hundreds of nanometres.

IIT using a spherical indenter tip has some advantages over using pointed indenters; due to the blunt geometry, spherical indentation can be initially elastic [14], therefore the whole strain range from elastic, elastic-plastic, and finally fully plastic deformation can be observed. This has made it is possible to generate indentation stress-strain curves comparable to those from traditional tensile tests (see e.g. Refs. [15,16] and, more recently [17]). Results from indentation tests are, however, influenced by the indentation size effect (ISE), whereby the smaller the interaction volume between tip and sample the harder the material appears. The ISE has been the subject of many studies over the last two decades [[18], [19], [20], [21]]. For Berkovich indentation, ISE is exhibited by a reduction in the measured hardness with an increase in penetration depth. For spherical indentation, the interaction volume and ISE are a function of both the tip radius and penetration depth, with a decrease in tip radii measuring a higher hardness. The extrinsic length-scale imposed by the test is directly related to this interaction volume and results in the commonly observed ISE; the underlying mechanisms responsible for this size effect in plasticity are yet to be fully understood and are a matter of debate [22].

Initial reasoning for the ISE was hypothesised by Fleck et al. [23] who suggested that since geometrically necessary dislocations (GNDs) are required for material deformation in a restricted volume, small indentations appear stronger because GND density scales with the plastic strain gradient, which is higher for smaller interaction volumes. This theory was later expressed by a model by Nix and Gao [24], which showed good agreement with experimental data for Berkovich indentation in Ag and Cu. However strain gradients cannot be the cause of all extrinsic size effects; for example, (perfect) uniaxial micro-pillars have no strain gradient yet still exhibit a ‘smaller is stronger’ behaviour [25]. There have since been many studies exploring further plasticity mechanisms that account for extrinsic size effects, which show that dislocation starvation and curvature also contribute to this phenomenon [[26], [27], [28], [29]]. The fundamental mechanisms controlling the ISE are distinguished separately from the well-established strengthening due to microstructural intrinsic length-scales, e.g. grain size, obstacle spacing or film thickness [[30], [31], [32], [33]].

In this work, indentation using spherical tips with a range of radii between 2 and 90 μm have been used to probe CuCrZr samples that have been heat-treated at different temperatures. The heat-treatments have the effect of changing the dominant internal length-scale of the material, in this case the average distance between Cr precipitates, which was characterised using transmission electron microscopy (TEM). The aim of the study was to observe the effect of varying contributions from extrinsic and intrinsic length-scales on the observed size effect to give insights into the dominant mechanisms of dislocation plasticity in a PH alloy.