Compared to ceramics that have high strength but are brittle, metals are weaker but have higher ductility. Materials researchers designing advanced structural materials have long sought ways to combine the best features of each, and the quest continues unabated. Today’s frontier has moved to the nanoscale, where a team led by Lei Lu of the Shengyang National Laboratory for Materials Science in China and Huajian Gao of Brown University has taken a significant step forward with its report in Science of gradient nanotwinning in copper.

“By combining the huge strengthening effect of nanotwinning with designed gradients in the twinned structures, Lu and Gao et al. have created nanostructures that give copper strength and ductility that exceed those in all comparable metals,” says William Nix of Stanford University. “It should be possible to apply the principles revealed in this work to metallic alloys and to create a host of other metallic materials with high strengths and ductilities.”

When metals are stressed beyond the yield point where plastic deformation begins, dislocation generation and motion allow the material to stretch, compress, or twist without breaking, at least for a while before it ultimately fails under a still relatively low stress, hence the high ductility at the expense of strength. Work hardening, which allows the material to sustain a higher stress before failure in the plastic-deformation regime, is also associated with the generation and motion of dislocations.

In the armamentarium of metallurgical techniques to improve mechanical properties of materials, strategies to modify dislocation motion are therefore king. For example, grain boundaries and other kinds of interfaces block their motion, so shrinking the grain size is one approach. Twin boundaries, where the crystal structure on each side is the mirror image of that on the other, represent another option because they block dislocation motion but do not completely stop it.

These traditional techniques are what Gao calls “top-down” in that after synthesis, the materials as a whole are thermally or mechanically treated in more modern versions of the old-time “heat and beat” metallurgy to promote the desired structures and dislocation configurations. “In today’s nanoscale age, the research momentum has shifted toward ‘bottom-up’ fabrication,” he says, “which allows more precise control of the structure during growth.” Combining with nanostructures in modern-day metallurgy is the concept of a structural gradient, which derives from the ubiquity of gradient and other inhomogeneous structures in the natural world.

Drawing on these new ideas, materials researchers have made nanograined materials with high strength while retaining ductility by introducing a gradient in the grain size through the thickness of the material, and they have investigated nanotwinned materials in which twin boundaries in the grains are spaced at nanoscale intervals. Of these, nanotwinning has been the most effective to date in producing a material with high strength while retaining some ductility, but the complicated dislocation configurations that come with nanotwinning have clouded the understanding of the underlying cause.

The Shengyang-Brown team thought to try a combination of gradients and nanotwinning. To this end, the Shengyang contingent developed a direct-current electrodeposition method to construct a multilayered structure in which both the lateral grain size and the vertical twin thickness increased in steps from layer to layer by stepping the electrolyte temperature. For their investigation, the group first synthesized four homogeneous materials (labeled A, B, C, and D) with the average twin thickness increasing from 29 nm to 72 nm and the corresponding average grain size varying from 2.5 µm to 15.8 µm. Then they constructed four kinds of gradient nanotwinned samples from multiples (1×, 2×, 3×, 4×) of the sequence ABCDDCBA of these homogeneous layers. Overall sample thickness remained constant at 400 µm for all samples, so that multiple layers increased the structural gradient.

From stress-strain measurements on individual homogeneous samples, the researchers found the expected tradeoff between strength and ductility with the A samples having the highest strength but the lowest strain before failure and the D samples the reverse. For the gradient samples, the tradeoff was different. Here, the highest strength but lowest ductility was associated with the highest structural gradient (4 × ABCDDCBA) and lowest structural gradient (1 × ABCDDCBA) the reverse. But while the 4 × ABCDDCBA sample had higher strength than the homogeneous sample A, it also had far better ductility. Moreover, the average strength of all the gradient samples exceeded the average of the four homogeneous samples. Finally, consistent with the higher strength and ductility, the researchers found that the work-hardening rate increased with the structural gradient.

Pleasantly surprised by these results, the researchers examined the microstructure, in particular the dislocation configurations, with transmission and scanning electron microscopy in search of an explanation for their discovery. What the imaging unexpectedly showed was a new kind of dislocation configuration that the group named bundles of concentration of dislocations (BCDs) consisting of parallel, aligned dislocation bundles in the interior of the columnar grains. Each dislocation bundle spanned multiple nanotwins and extended through almost the entire length of the grain perpendicular to the twin boundaries. The BCDs were found only in the gradient nanotwinned samples.

Computer simulations of dislocation configurations confirmed both the existence of the BCDs at high strains and, along with the electron microscopy images, suggested a mechanism for their formation as the strain increased. The mechanism is based on the interaction of specific types of dislocations migrating to the grain interior after being generated at grain boundaries, twin boundaries, and interfaces by the inhomogeneously distributed strain throughout the gradient-nanotwinned material. “It had been known for a long time that inhomogeneous microstructures with plastic strain gradients can lead to collections of more dislocations that contribute to enhanced strength and ductility. Lu and Gao et al. have managed to do this in a controllable way with graded nanotwins,” Nix says. 

Is gradient nanotwinning feasible in materials other than copper? “Copper was a good choice for testing the concept because of the relative ease of introducing twin boundaries, but the principle is not material specific,” Gao says. “One way to generalize it is to design nanotwin gradients into high-entropy alloys in which there is no dominant element, as there is in steels, but more equal concentrations of five or more elements. Such an approach may open a new dimension of materials research in the next decade,” he suggests.

Read the abstract in Science.