Toward a quantitative understanding of mechanical behavior of nanocrystalline metals

Abstract

Focusing on nanocrystalline (nc) pure face-centered cubic metals, where systematic experimental data are available, this paper presents a brief overview of the recent progress made in improving mechanical properties of nc materials, and in quantitatively and mechanistically understanding the underlying mechanisms. The mechanical properties reviewed include strength, ductility, strain rate and temperature dependence, fatigue and tribological properties. The highlighted examples include recent experimental studies in obtaining both high strength and considerable ductility, the compromise between enhanced fatigue limit and reduced crack growth resistance, the stress-assisted dynamic grain growth during deformation, and the relation between rate sensitivity and possible deformation mechanisms. The recent advances in obtaining quantitative and mechanics-based models, developed in line with the related transmission electron microscopy and relevant molecular dynamics observations, are discussed with particular attention to mechanistic models of partial/perfect-dislocation or deformation-twin-mediated deformation processes interacting with grain boundaries, constitutive modeling and simulations of grain size distribution and dynamic grain growth, and physically motivated crystal plasticity modeling of pure Cu with nanoscale growth twins. Sustained research efforts have established a group of nanocrystalline and nanostructured metals that exhibit a combination of high strength and considerable ductility in tension. Accompanying the gradually deepening understanding of the deformation mechanisms and their relative importance, quantitative and mechanisms-based constitutive models that can realistically capture experimentally measured and grain-size-dependent stress–strain behavior, strain-rate sensitivity and even ductility limit are becoming available. Some outstanding issues and future opportunities are listed and discussed.

Introduction

In the mid-1980s, Gleiter [1] made the visionary argument that metals and alloys, if made nanocrystalline, would have a number of appealing mechanical characteristics of potential significance for structural applications. This followed, quite plausibly, from what was known about the extraordinary strength of alloys such as highly cold drawn wires characterized by structural length scales of nanometer size (e.g. [2]). Compared with conventional coarser grained materials, the benefits that may be derived from nanostructuring include ultrahigh yield and fracture strengths, superior wear resistance, and possibly superplastic formability at low temperatures and/or high strain rates. The deformation mechanisms are also predicted to be radically different, as plasticity at the nanoscale may be mediated mostly by grain-boundary deformation processes. These provocative thoughts stimulated widespread interest in the mechanical properties and novel deformation mechanisms of nanostructured materials over the past two decades. Many research articles have been published in this area. Unfortunately, most of the experimental findings documented in this literature up to the late 1990s were not representative of intrinsic material response, due to the problems and difficulties associated with preparing full-density and flaw-free nanocrystalline samples [3]. Improvements in materials processing, discussed briefly herein, have led to enhancements in properties, but yet still further refinements are needed.

Strengthening with grain size refinement in metals and alloys with an average grain size of 100 nm or larger has been well characterized by the Hall–Petch (H–P) relationship, where dislocation pile-up against grain boundaries (GBs) along with other transgranular dislocation mechanisms are the dominant strength-controlling processes. When the average, and entire range of, grain sizes is reduced to less than 100 nm, the dislocation operation becomes increasingly more difficult and grain boundary-mediated processes become increasingly more important [3], [4], [5], [6].

With these observations in mind, we continue to use the terminology proposed in the earlier literature [3]: nanocrystalline (nc) materials are defined as those with their average and entire range of grain size typically finer than 100 nm; ultrafine crystalline (ufc) materials are defined as those with grain sizes on the order of 100 nm–1 μm; and microcrystalline (mc) materials are defined as those with average grain sizes greater than 1 μm [3], [4], [5], [7], [8], [9]. When one or more dimensions on average is smaller than 100 nm, the material is often termed a nanostructured (ns) material [7], [8], [10]. Another category may be termed nc/ufc metals, whose grain sizes are characterized by averages near 100 nm, but with grain size distributions spanning the range from nc to 500 nm. This class is included to highlight the fact that recent methods utilizing severe plastic deformation methods have produced high-density bulk Ti metal with grain sizes (d) in the range of 50 nm ⩽ d ⩽ 150 nm.

A number of reviews have been written since Gleiter [1] first summarized the pioneering ideas, e.g. by Gleiter [11], [12], Weertman [13], Kumar et al. [3], Koch [4], Cheng et al. [14], Wolf et al. [15], Meyers et al. [9], Ma [10], etc. Specific references to these are made throughout the text.

This overview highlights some of the most recent experimental advances in property improvements and mechanisms-based quantitative analyses, rather than attempting to provide a detailed account for all developments in this field. Recent experimental studies, discussed herein, on the one hand point out promising routes to optimize mechanical properties, yet on the other hand reveal challenges to the understanding of intrinsic nc behavior that require further careful quantitative examination. Examples of such effects or phenomena include grain size distribution vs. overall mechanical response and properties, the unusual size dependence of nanoscale growth twins in terms of ductility and the (apparent) stress-induced grain growth observed during deformation. All of these have significant effects on the macroscopic mechanical response and (therefore) implications for potential use of nc or even ufc metals and alloys. Parallel to these new developments in experimental investigations, several recent mechanisms based and physically motivated models have provided quantitative insights into the deformation mechanisms as well as possible routes to mechanical property improvements.

This paper is organized as follows. Section 2 briefly highlights several of the most important and commonly used methods for processing bulk nc/ns materials. Severe plastic deformation is included in the discussion owing to the aforementioned ability to process nc/ufc metals. Although not intended as a comprehensive review of processing, this discussion provides needed perspective for our subsequent presentation of nc metal properties. Section 3 summarizes experimental observations on strength, ductility, strain rate and temperature dependence of strength, fatigue and tribological properties of nc materials. Here, along with the discussion of recently reported phenomenology, we note several key findings that challenge our understanding of nc metal behavior. Nanocrystalline grain size stability during deformation is one example of such critical behavior. Improvements in fatigue performance, including crack initiation vs. fatigue crack growth, are additional examples. Section 4 reviews recent developments in dislocation based and physically motivated continuum models. The modeling is discussed with an aim to quantitatively explore the critical phenomenology highlighted in Section 3. This focus explains the choice of our title. A summary and concluding remarks follow in Section 5.