Elsevier

Acta Materialia

Volume 59, Issue 10, June 2011, Pages 3903-3914
Acta Materialia

Three-dimensional shear-strain patterns induced by high-pressure torsion and their impact on hardness evolution

https://doi.org/10.1016/j.actamat.2011.03.015Get rights and content

Abstract

The shear strain imposed on austenite/ferrite duplex stainless steel discs at different stages of high-pressure torsion (HPT) processing was imaged in plan-view and cross-section using optical microscopy and scanning electron microscopy. The effect of the shear strain was correlated to the hardness evolution of the discs. The shear-strain patterns are complex and are different on the top and bottom surfaces of the discs. A double-swirl pattern emerged on the top surface in the early stages of HPT. These two centres of the swirl moved towards the centre of the disc as the numbers of HPT revolutions was increased and ultimately the double-swirl evolved into a single-swirl. Less regular shear-strain patterns were observed on the bottom surfaces of the discs. Multiple ring-like patterns with mirror symmetry over the central axes of the discs were visible from cross-sectional observations. Nanoindentation testing on the two surfaces and a cross-section of HPT discs showed that the hardness is insensitive to specific shear-strain patterns, but is closely related to the widths of the austenite and ferrite phase domains. Late in the deformation process, the hardness in the interior of an HPT disc may be higher than at either of the disc surfaces because of the development of finer microstructural phase distributions.

Highlights

► Three-dimensional shear strain patterns were evaluated during high-pressure torsion. ► Double-swirl patterns are visible on the top surfaces of discs in early stages of HPT. ► Double-swirls ultimately evolve into a single swirl with increasing revolutions. ► Microstructural evolution in HPT may deviate initially from rigid-body situation

Introduction

High-pressure torsion (HPT), which was first proposed by Bridgman in the 1940s [1], has attracted significant attention in recent years because it is the most effective severe plastic deformation technique [2], [3], [4], [5], [6] in grain refinement for producing bulk ultrafine-grained (<1 μm) and nanocrystalline (nc, <100 nm) metallic materials having superior mechanical properties, including a combination of high strength and good ductility [7], [8], [9]. During HPT processing, a sample in the form of a disc is placed between two anvils, and a torsional strain is imposed on the sample by applying a very high pressure of several GPa to the upper anvil and simultaneously rotating the lower anvil [3].

Under the ideal rigid-body condition, the HPT-imposed shear strain γ can be calculated using the equation γ = 2πNr/h, where N is the number of revolutions, r is the distance from the centre of the disc and h is the thickness of the disc [10]. Based on this equation, the following conclusions may be drawn: (i) the direction of HPT shear strain is expected to be perpendicular to the radial directions throughout the disc; (ii) the shear strain, and therefore the hardness, should be the same at all points having the same radial value; and (iii) the shear strain at the centre of the disc is zero. However, deviations from these conclusions have been widely reported [11], [12], [13], [14], [15], [16]. For example, it was reported that uniform refined grains, and therefore uniform hardness, can be achieved throughout an HPT disc by processing through a high number of HPT revolutions [11], [16]. On the other hand, an undulating microhardness evolution has been observed during HPT processing [12], [13], [16].

Extensive theoretical and experimental efforts have been made to explain the deviations of HPT shear strain from the ideal situation. Estrin et al. [17] developed a detailed analytical model to explain the formation of a uniform microstructure in HPT discs after processing through large numbers of revolutions. Vorhauer and Pippan [18] conducted very careful HPT experiments and obtained a near-ideal torsional deformation with different numbers of revolutions. They proposed that possible reasons for the disappearance of the near-undeformed central region of HPT samples [11], [16] included the misalignment of the axes of the anvils or other deviations from idealized HPT processing. Zhilyaev et al. [11] proposed that HPT deformation develops in a repetitive manner throughout the outer ring of the disc and thereafter the shearing spreads inwards towards the centre, which leads to an undulating evolution of microhardness.

The above-mentioned reports fail to directly reveal the shear strain imposed by HPT because of the difficulty of visualizing the shear strain in single-phase materials or multiphase alloys with only a small amount of secondary phases [19], [20], [21]. By choosing a duplex stainless steel in which two phases coexist with approximately the same volume fractions, the HPT-induced shear-strain pattern was successfully imaged using microscopy [22], [23], revealing several surprising features of the shear strain, including shear straining parallel to the radial directions in the very early stages of HPT processing, local shear vortices and a double-swirl shear-strain pattern at the centre of the HPT discs. These remarkable shear strain features indicate that the microstructural evolution during HPT is more complex than the ideal rigid-body situation. In a previous report [22], only the two-dimensional shear-strain patterns at the top surfaces of HPT discs were investigated. The three-dimensional nature of the shear-strain patterns at both surfaces and in the disc interior was not previously investigated and the relationship between shear-strain patterns, local hardness and the evolution of hardness during HPT processing is not known. This report addresses these questions in a comprehensive investigation of the shear-strain patterns and hardness changes at the top and bottom surfaces and over the cross-sections of duplex stainless steel discs subjected to HPT processing for different numbers of revolutions. The results provide new insight into the microstructural development and the changes in hardness of materials processed by HPT.

Section snippets

Experimental material and procedures

The material used in this experiment was the commercial DP3W super-duplex stainless steel in the form of rectangular plates. The steel has a composition of C 0.017, Si 0.3, Mn 0.5, P 0.015, S 0.001, Ni 7.0, Cr 25, Mo 3.3, W 2.0 and N 0.28 (wt.%), and has approximately equal volume fractions of the face-centred cubic austenite (γ) phase and the body-centred cubic ferrite (α) phase. Discs having a diameter of ∼9.8 mm and a thickness of 1.7 mm were cut from the steel plates, as illustrated

Visualization of HPT-imposed shear strain in three dimensions

The as-received duplex stainless steel sample was in the shape of a rectangular plate, as illustrated schematically in Fig. 1a. The microstructure of the as-received sample is shown in Fig. 1b–d when viewed on the XY, YZ and ZX planes, respectively, as defined in Fig. 1a. The islands with bright contrast are the γ phase, whereas the matrix areas with dark contrast are the α phase. The morphology is consistent with previous reports [25], [26]. A three-dimensional reconstruction of the

Discussion

The present HPT experiments were carried out carefully and no slippage was observed during the first six HPT revolutions. Although it was not possible to check if there was slippage for higher HPT revolutions because the marks used to check for slippage then became obscured, there was no direct evidence for slippage as usually characterized by a uniform circular shining ring structure on the surface. On the contrary, the unpolished surfaces of the discs processed through medium to high numbers

Summary and conclusions

Optical microscopy and SEM were applied to characterize the three-dimensional nature of the HPT-induced shear-strain patterns in sample discs of a duplex stainless steel. The effect of the shear-strain patterns on the hardness of HPT discs was investigated using nanoindentation testing. We conclude the following:

  • 1.

    The shear-strain patterns on the top and bottom surfaces of each HPT disc are different. However, this does not directly affect the hardness distribution across the face of the disc

Acknowledgements

The authors are grateful for scientific and technical input and support from the Australian Centre for Microscopy & Microanalysis at the University of Sydney. This project is supported by the Australian Research Council (Grant No. DP0772880 (Y.C., Y.B.W., and X.Z.L.)), the National Science Foundation of the United States (Grant No. DMR-0855009, (M.K. and T.G.L.)) and the US Army Research Office and Army Research Laboratory (Y.T.Z.).

References (45)

  • R.Z. Valiev et al.

    Prog Mater Sci

    (2000)
  • A.P. Zhilyaev et al.

    Prog Mater Sci

    (2008)
  • I. Sabirov et al.

    Scripta Mater

    (2005)
  • R.Z. Valiev et al.

    Prog Mater Sci

    (2006)
  • R.Z. Valiev et al.

    J Mater Res

    (2002)
  • F. Dalla Torre et al.

    Acta Mater

    (2004)
  • R.Z. Valiev et al.

    Acta Mater

    (1996)
  • A.P. Zhilyaev et al.

    Acta Mater

    (2003)
  • A.P. Zhilyaev et al.

    Mater Sci Eng

    (2005)
  • C. Xu et al.

    Acta Mater

    (2007)
  • C. Xu et al.

    Acta Mater

    (2008)
  • Y. Todaka et al.

    Mater Sci Eng

    (2007)
  • Y. Estrin et al.

    J Mech Phys Solids

    (2008)
  • A. Vorhauer et al.

    Scripta Mater

    (2004)
  • A.P. Zhilyaev et al.

    Scripta Mater

    (2001)
  • H.G. Jiang et al.

    Mater Sci Eng

    (2000)
  • Z. Horita et al.

    Mater Sci Eng

    (2005)
  • G. Fargas et al.

    J Mater Process Technol

    (2009)
  • L.H. Zhang et al.

    Electrochim Acta

    (2009)
  • Y.Z. Tian et al.

    Scripta Mater

    (2010)
  • R. Dakhlaoui et al.

    Mater Sci Eng

    (2007)
  • M. Kawasaki et al.

    Acta Mater

    (2011)
  • Cited by (0)

    View full text