Elsevier

Acta Materialia

Volume 59, Issue 1, January 2011, Pages 308-316
Acta Materialia

An investigation of hardness homogeneity throughout disks processed by high-pressure torsion

https://doi.org/10.1016/j.actamat.2010.09.034Get rights and content

Abstract

Experiments were conducted on high-purity aluminum to determine the degree of hardness homogeneity attained during processing by high-pressure torsion (HPT). The HPT processing was conducted at room temperature using several disks of identical initial thickness with an applied pressure of 6.0 GPa and torsional straining from 1/4 to 20 turns. Following HPT, the flow patterns were observed by optical microscopy and values of the Vickers microhardness were recorded on different planes. The results show that, for any selected processing conditions, the distributions of microhardness values and the appearance of the etched surfaces in optical microscopy are independent of the plane of sectioning within the HPT disks. There is also a gradual evolution in homogeneity with increasing numbers of revolutions so that the microhardness values reach a constant value after 20 turns.

Introduction

The processing of bulk metals by severe plastic deformation (SPD) provides the potential for achieving significant grain refinement to the submicrometer or nanometer range [1]. Several SPD processing techniques are now available but the two most important procedures are equal-channel angular pressing (ECAP) in which a rod or bar is pressed through a die constrained within a channel [2], and high-pressure torsion (HPT) where a disk is subjected to a high applied pressure and concurrent torsional straining [3]. In practice, HPT is an especially attractive procedure because it leads to greater grain refinement than ECAP.

Processing by HPT was first developed over 70 years ago by Bridgman [4], [5] but the procedure has only begun to receive significant attention within the last two decades. The principle of HPT is illustrated in Fig. 1: the sample, in the form of a thin disk, is placed within a depression on the upper face of a lower anvil, the anvil is brought upwards so that the disk is also contained within a corresponding depression on the lower face of an upper anvil, and the disk is subjected to a severe compressive pressure, P. In this final position, there is a small gap between the faces of the two anvils and therefore a small volume of material is forced outwards around the periphery of the sample. Processing by HPT is conducted by applying torsional straining through rotation of the lower anvil. This type of processing is termed quasi-constrained HPT and it contrasts with constrained HPT where the sample is held in place within a depression without any possibility of outward flow, and unconstrained HPT where the disk is placed on a flat anvil surface and it is free to flow outwards under the applied pressure [3], [6].

The principle of HPT is relatively simple. The shear strain, γ, imposed on the disk during torsional straining is given by the expression [7], [8], [9]:γ=2πNrh,where N is the number of revolutions, and r and h are the radius and height (or thickness) of the disk, respectively. It follows from Eq. (1) that the imposed strain varies across the disk from a maximum at the edge of the disk to zero in the center where r = 0. In practice, there is also a small additional strain imposed by the applied compressive pressure, P. For example, recent experiments on an Al-6061 aluminum alloy showed that the applied pressure introduces a measurable increase in hardness even in the absence of any torsional straining [10].

It is reasonable to anticipate from Eq. (1) that the microstructures and microhardness values introduced by HPT will be extremely inhomogeneous. Nevertheless, very early experiments revealed the possibility of achieving a gradual evolution towards a homogeneous structure in disks processed by HPT by increasing either the applied pressure, P, and/or the total numbers of revolutions, N [11], [12]. Subsequently, numerous reports confirmed this gradual evolution towards homogeneity [6], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23] and the evolution has now been successfully modeled using strain gradient plasticity theory [24].

Very recent experiments have shown that there are also additional complexities occurring in the flow patterns associated with disks processed by HPT. First, experiments on a Zn–22% Al eutectoid alloy revealed agglomerates of the Al-rich (α) and Zn-rich (β) phases in the early stages of HPT with these agglomerates lying approximately parallel to the directions of flow in torsional straining around the peripheries of the disks [19]. Second, unusual flow patterns were visible in the centers of disks of high-purity aluminum after processing through small numbers of turns using a cyclic condition where the direction of torsional straining was periodically reversed [17]. Third, experiments on a duplex stainless steel revealed unusual but well-defined flow characteristics in the early stages of processing by HPT including double-swirl patterns, shear vortices and the presence of very significant local turbulences [25], [26]. Fourth, similar vortices were observed in a Cu–28% Ag alloy where they were clearly revealed by markings in the decorative eutectic regions [27]. These observations of vortex microstructures suggest the occurrence of Kelvin–Helmholtz (K–H) instabilities which may arise within the HPT disks due to local shear velocity differences between adjacent positions. The occurrence of K–H instabilities is well established in many areas of physics, including fluid flow [28], plasma physics [29] and meteorology [30], but they have not been reported previously in processing by HPT.

Any interpretation of the complex flow processes occurring in HPT is currently hampered by the fact that all observations to date, without exception, have been taken in the planes of the HPT disks and there are no parallel observations delineating any differences that may occur within the thicknesses of the disks. This is an important omission because very recent experiments suggest there are differences between observations taken on the top and bottom surfaces of HPT disks. The problem has arisen primarily because the HPT disks are very thin so that typically h  0.8 mm. It is important to note that most observations reported to date were recorded on arbitrarily selected planes and no attention was given to whether these planes are close to the upper or lower surfaces of the disk within the HPT facility. This contrasts with observations documented for samples processed by ECAP where microhardness measurements have been systematically recorded both perpendicular [31] and parallel [32] to the pressing direction.

The present investigation was initiated to provide the first information on the variation of microhardness throughout disks processed by HPT. Care was taken to distinguish between the upper and lower surface of every disk and the exact location of each plane of measurement was carefully documented. Emphasis was placed on recording values of the microhardness because earlier studies demonstrated there is a clear correlation between the individual microhardness measurements and the internal microstructure [6], [11], [12], [14], [15], [16], [20], [22], thereby suggesting that hardness measurements provide a simple and expedient procedure for reaching conclusions on the microstructural characteristics and the degree of internal homogeneity within samples processed by HPT.

Section snippets

Experimental material and procedures

The investigation was conducted using high-purity (99.99 wt.%) aluminum. This material was selected for three reasons. First, there are already extensive experimental data describing the evolution of hardness in high-purity aluminum processed by HPT [15], [17], [18], [20], [22], [33], [34], [35], [36], [37], [38] and similar sets of data are available also for high-purity aluminum processed by ECAP [39], [40], [41], [42], [43], [44], [45]. Second, slippage is often a critical problem when

Flow patterns produced by HPT

Inspection by optical microscopy of the various disks processed in this investigation revealed a consistent change and evolution between the disks processed through different numbers of revolutions. Conversely, there was no apparent difference between the flow patterns visible on the upper or lower surfaces for any of the disks processed under identical conditions. Fig. 2 shows an example of the flow patterns visible on the lower etched surfaces of disks processed through 1/4 to 20 turns. These

The significance of surface location within the HPT disk

There are two important results from these experiments.

First, the measurements show that, for any selected experimental condition of HPT processing, the distribution of microhardness values throughout disks of high-purity aluminum is independent of the plane of sectioning. Specifically, the data recorded in Fig. 3, Fig. 4, Fig. 5, Fig. 6 show identical results for measurements taken on three separate HPT disks at upper, center and lower positions where this refers to planes ∼200 μm from the top

Summary and conclusions

  • 1.

    Disks of high-purity aluminum were processed by HPT at room temperature using an applied pressure of 6.0 GPa and torsional straining from 1/4 to 20 turns. The flow patterns were observed by optical microscopy and detailed measurements were taken of the Vickers microhardness.

  • 2.

    By conducting careful experiments using disks with the same initial thickness, it is shown that, for any selected experimental condition of HPT processing, the distributions of microhardness values and the appearance of the

Acknowledgement

This work was supported by the National Science Foundation of the United States under Grant No. DMR-0855009.

References (53)

  • R.Z. Valiev et al.

    Prog Mater Sci

    (2000)
  • R.Z. Valiev et al.

    Prog Mater Sci

    (2006)
  • A.P. Zhilyaev et al.

    Prog Mater Sci

    (2008)
  • R.Z. Valiev et al.

    Acta Mater

    (1996)
  • F. Wetscher et al.

    Mater Sci Eng

    (2004)
  • A.P. Zhilyaev et al.

    Scripta Mater

    (2001)
  • A.P. Zhilyaev et al.

    Acta Mater

    (2003)
  • A. Vorhauer et al.

    Scripta Mater

    (2004)
  • A.P. Zhilyaev et al.

    Mater Sci Eng

    (2005)
  • C. Xu et al.

    Acta Mater

    (2007)
  • C. Xu et al.

    Acta Mater

    (2008)
  • M. Kawasaki et al.

    Mater Sci Eng

    (2008)
  • C. Xu et al.

    Mater Sci Eng

    (2009)
  • M. Kawasaki et al.

    Acta Mater

    (2010)
  • X.H. An et al.

    Scripta Mater

    (2010)
  • Y. Estrin et al.

    J Mech Phys Solids

    (2008)
  • Y.Z. Tian et al.

    Scripta Mater

    (2010)
  • C. Xu et al.

    Mater Sci Eng

    (2005)
  • M. Prell et al.

    Mater Sci Eng

    (2008)
  • Y. Harai et al.

    Scripta Mater

    (2008)
  • D. Orlov et al.

    Scripta Mater

    (2009)
  • D. Orlov et al.

    Mater Sci Eng

    (2009)
  • Y. Ito et al.

    Mater Sci Eng

    (2009)
  • Y. Iwahashi et al.

    Acta Mater

    (1997)
  • K. Nakashima et al.

    Acta Mater

    (1998)
  • Y. Iwahashi et al.

    Acta Mater

    (1998)
  • Cited by (172)

    • High-entropy alloys for solid hydrogen storage: a review

      2024, International Journal of Hydrogen Energy
    View all citing articles on Scopus
    View full text