Mechanical alloying (MA) has recently been used for material processing methods: the preparation of amorphous phases, non-equilibrium solid solution phases, high pressure phases, etc. [1–4]. The metastable solid solution powders of V–Cu [5], Hg–Cu [6], Al–Fe [7] systems, etc., which can not appear in an equilibrium state, were prepared by the MA treatment. The amorphous powders of Ta–Cu [5], Ni–Zr [8] systems, etc. were prepared also by the MA treatment. The tungsten (W)-copper (Cu) system is a typical immiscible system in the solid and even in the liquid phase. Several studies of the MA treatment of the copper-based W–Cu system were reported for the pre-treatment of sintering, but they did not suggest the formation of solid solutions [9–11]. Gaffet et al. [12] reported that the W–Cu system metastable phase could be prepared by the MA treatment, but the crystal state was not investigated in detail [12]. The preparation of the non-equilibrium bulk body from the MA-treated W– Cu system powder has not been reported. Shock compression has different features from static compression: pulsed short duration, shear stress, heterogeneous state, etc. and can be used as a consolidation method of non-equilibrium materials without recrystallization or decomposition [13, 14]. It is important to consolidate non-equilibrium powders for the evaluation of physical properties and for industrial applications. In this study, the MA treatment and shock compression recovery experiments were performed on a W–Cu mixture powder, to prepare non-equilibrium W–Cu system bulk bodies including solid solutions, which can not appear in an equilibrium state. Starting powders were provided by Rare Metallic Co., Ltd. These tungsten and copper powders consisted of irregular particles of 2–3 and ,45 im diameter with a total impurity value of less than 0.1 and 0.01 wt %, respectively. The starting powder mixture was prepared by mixing tungsten and copper powders in a 60:40 molar ratio. The MA experiments were performed by using the planetary micro ball mill (P-7 of Fritsch Co., Ltd.) in an argon atmosphere [4]. A mill capsule with an inner-diameter of 41 mm and a depth of 38 mm and balls with a diameter of 15 mm, which were both made of silicon nitride, were used. The powder specimen with a weight of about 20 g and seven silicon nitride balls were contained into the capsule. The rotation speed of the ball mill was 2840 r.p.m. The resultant acceleration was estimated to be about 12 g. Small amounts of powder were recovered for several different durations of the MA treatment for X-ray analyses. Shock-compression recovery experiments were conducted using a propellant gun [15]. The MA treatment powder samples were enclosed in an iron capsule (SS-41) with an inside diameter of about 12 mm and with an inside height of about 3.5 mm. The porosity of the powder was about 49%. Shock loading was carried out by impacting the sample capsule with a tungsten flat flyer plate whose thickness was 1 mm. Shock pressures achieved in the capsule and in the sample were estimated by the impedence-matching method from the measured flyer-plate velocity, the Hugoniot parameters of tungsten, steel and copper [16]. The recovered specimens were investigated by chemical analysis and powder X-ray diffraction (XRD). Powder XRD analyses were carried out using monochromatized Cu-KAE radiation with a Rigaku goniometer. Calibration of the goniometer was performed by measuring diffraction peaks of pure silicon powder. The lattice parameter was refined by the least squares method up to an accuracy of greater than 0.01%. The composition mapping analysis was carried out using an electron probe micro analyser (EPMA) apparatus, CXA-733 of JEOL Ltd. Instrumental chemical analysis of nitrogen (N), oxygen (O) and carbon (C) contents were carried out by the inert-gas fusion thermal conductivity method and the combustion in oxygen non-dispersive infrared absorption method using the TC-436 and WR-112 of LECO Corp. Chemical analysis of silicon (Si) was done by the inductively coupled plasma (ICP) method using the SPS-1200 of Seiko Electric Co., Ltd. For each element, the measurements were performed 2–3 times on each specimen. The experimental errors for N, O, C and Si contents were estimated to be less than 0.07, 0.02, 0.01 and 0.0001 wt %, respectively, by the experiments using the Si3N4 and TiN as standard samples. Fig. 1 shows the powder XRD patterns of the starting powder, the MA treated powder (360 min) and the shock-consolidated bulk body with an impact velocity of 0.938 km sy1. The driving pressure in the capsule and the first pressure in the powder were estimated to be 24.5 and 13 GPa,
Read full abstract