High temperature nanoindentation of tungsten: Modelling and experimental validation

Highlights

• Hardness measured up to 600C agrees with earlier published results.

• Temperature induced softening is observed.

• FEM simulations agree well with nanoindentation.

Abstract

Knowledge of mechanical properties of the tungsten surface region is extremely important for its application as first wall materials in plasma-facing components for nuclear fusion devices (e.g. ITER). Since tungsten is intrinsically brittle at room temperature, characterization of its ductile properties is possible only above the so-called ductile-to-brittle transition temperature (DBTT), which is above 500–700 K. This is why the development and qualification of instrumented hardness measurements at elevated temperature is an important task to enable the characterization of tungsten properties after exposure to heat shocks, plasma beam and ion irradiation, which all together mimic the actual operation conditions of nuclear fusion. We have performed nanoindentation measurements on tungsten in the constant stiffness mode using Bruker stage developed for high temperature operation with oxygen protective environment. Commercially pure tungsten of ITER specification is studied in the as-produced and as-recrystallized conditions to deduce the impact of the texture and forging on the hardness. The obtained results are analysed by means of crystal plasticity finite element method (CPFEM) model to subtract the constitutive laws for the elasto-plastic deformation and derive the strengthening term attributed to the contribution coming from statistically stored dislocations and grain boundaries.

Introduction

Tungsten is material with unique properties for high temperature applications including shielding in particle accelerators, armour in space applications, heating elements in furnaces, source for high resolution electron microscopes and heat transfer body in plasma operating components (PFC) in fusion devices. Thanks to its high strength, thermal conductivity and melting point, tungsten is selected as PFC material in the world's largest fusion device under construction –ITER [1,2]. As the first wall material, tungsten will protect the functional components, required to evacuate the heat generated by nuclear fusion reaction, and therefore will be exposed to combined thermal loads and 14 MeV neutron flux [3,4]. The capacity of tungsten to absorb thermo-mechanical stresses by dissipating them into plastic deformation up to large extend will determine lifespan of the PFC operation (see recent review [5]), while ageing is caused by cyclic thermal fatigue and accumulation of the neutron irradiation damage.

Even in non-irradiated conditions, commercially pure tungsten suffers from intrinsic brittleness, as its ductile-to-brittle transition temperature (DBTT) is around 500–700 K [6], which depends on the particular microstructure and composition (i.e. on the initial purity, fabrication route and final heat treatment). The onset of ductility appears to be controlled by the dislocation mobility and microstructural features limiting their multiplication and recombination. Under neutron irradiation, DBTT will increase further prompting a risk of brittle fracture and cracking directly under operation [7]. The detrimental effect of neutron irradiation on the reduction of ductility of tungsten is noted by several authors [[8], [9], [10], [11], [12]]. This is why the current strategy for the development of durable and efficient PFCs for DEMO reactor especially underlines importance of understanding and mitigation of the risk related to the neutron irradiation embrittlement [13].

In that respect, the development of advanced tungsten-based alloys and composites with an optimized microstructure offering high sink strength and reduced DBTT is an important step forward [[14], [15], [16], [17], [18], [19], [20]]. However, the validation and down selection of new concepts requires an accelerated approach to enable increasing the technological readiness level of material/composite up to a level sufficient for the fabrication of the components. Application of heavy ion irradiation is one alternative to mimic the neutron irradiation damage, which is advantageous in terms of budget, implementation time, lack of activation, flexibility in terms of fluence and temperature. The main disadvantage of the ion irradiation is the penetration depth and non-homogenous profile of the damage. Due to intensive scattering, heavy ion irradiation generates very shallow damage deposition profile (typically limited to 1–2 μm), so that conventional means of mechanical characterization cannot be applied [[21], [22], [23]] and up to day, nanoindentation (NI) technique is the only way to extract mechanical properties from the ion irradiate material [[24], [25], [26]]. On the other hand, mechanical characterization of the neutron irradiated materials requires application of specially shielded equipment (so called Hot Cells) to protect operators and researchers from the irradiation immersing by the neutron irradiated samples. These tests are rather expensive and lengthy as they require remote manipulation, transportation of active material, air contamination control and other safety measures. Nanoindentation techniques might become a relevant tool to investigate the mechanical properties of sub-sized neutron irradiated samples which can be fabricated simply from the fragments of the standardized samples. However, the ability to perform nanoindentation at elevated temperature and inside of Hot Cells (i.e. mounting of samples using tele-manipulators) remains critical.

In the field of first wall fusion materials, the application of NI for tungsten (W) was successful to characterize the impact of heavy ion irradiation [[27], [28], [29], [30], [31], [32], [33], [34], [35], [36]] and low to high flux plasma exposure together with He ion irradiation [28,[34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47]], including grades produced in accordance with ITER specification. A number of computational studies were dedicated to modelling of NI processes in tungsten [[48], [49], [50], [51], [52], [53]] to help extraction of the physical properties of tungsten and assist interpretation of the experimental data. Conventional NI measurements are applied up room temperature, which in the case of tungsten is questionable since the material becomes ductile well above 300 K. The basic approach for the interpretation of the NI results assumes classical plastic deformation underneath the indenter tip [54,55]. Thus, characterization of the mechanical properties of tungsten by NI requires further development of the techniques to enable indentation at elevated temperature. The latter is much more complicated procedure due to the small-scale instrumentation, temperature gradients and thermal drifts caused by the heating [56]. Recently, two studies have been performed for the NI of polycrystalline tungsten at the elevated temperatures [[57], [58], [59]]. The measurements were performed with high temperature vacuum nanoindentation systems supplied by NanoTest Xtreme [59] and Anton Paar [58]. Other recent studies involving high temperature NI for steels [60], high entropy alloys [61], chromium [62], have also been published, which highlight importance and complexity of the development of instrumented hardness tests at elevated temperature.

In this paper, we extend the currently available knowledge on the high temperature NI of tungsten and perform study using Bruker system with the high temperature stage. The xSOL stage for high temperature NI operates without vacuum chamber and utilizes protective oxygen atmosphere. The absence of the vacuum chamber might be very important advantage for the application of such instrument in Hot Cells, as it simplifies the maintenance and speeds up the operation. It is also important to demonstrate the ability to perform fast and accurate data acquisition on miniaturized samples without their embedding/gluing (resin or cementite), to ensure that the handling of the samples can be arranged by manipulators (important for active samples). The main purpose of this work is to test this heating stage where the oxygen reduced environment is created locally without application of vacuum chamber. NI is performed using forged and recrystallized tungsten samples in the temperature range 300 - 773 K. The microstructure and mechanical properties of this material are well characterized as this material figures in many experimental works performed in a frame of the European fusion project (i.e. EUROfusion see e.g. [18,[63], [64], [65], [66], [67], [68], [69]]).

The obtained results are treated by finite element method (FEM) employing Peirce-Asaro-Needleman model by decomposing the plastic flow into several contributions accounting separately for the impact of lattice friction, dislocation forest hardening and dislocation grain boundary pinning (i.e. the Hall-Petch effect). Thus, we can account for the impact of the microstructural changes induced by the recrystallization and deduce the strengthening coefficients for the crystal plasticity model which will be used in future to assess the impact of ion and neutron irradiation on mechanical properties at elevate temperature. To ensure the industrial relevance of this work, we carried out the experiments on the commercial tungsten grade of 99.97 wt% purity produced according to the ITER specification.