Ton-scale metal–carbon nanotube composite: The mechanism of strengthening while retaining tensile ductility

One-dimensional carbon nanotubes (CNT), which are mechanically strong and flexible, enhance strength of the host metal matrix. However, the reduction of ductility is often a serious drawback. Here, we report significantly enhanced plastic flow strength, while preventing tensile ductility reduction, by uniformly dispersing CNTs inAlmatrix. Nanoscale plasticity and rupturing processes near CNTs were observed by in-situ mechanical tests inside Transmission Electron Microscope (TEM). CNTs act like forest dislocations and have comparable density (∼1014/m2), and such 1D nano-dispersion hardening is studied in detail by in situ TEM and molecular dynamics simulations. Rupture-front blunting and branching are seen with in situ TEM, which corroborates the result from macro-scale tension tests that our Al + CNT nanocomposite is quite damageand fault-tolerant. We propose a modified shear-lag model called ‘‘Taylor-dispersion’’ hardening model to highlight the dual roles of CNTs as load-bearing fillers and ‘‘forest dislocations’’ equivalent that harden the metal matrix, for the plastic strength of metal + CNT nanocomposite. © 2016 Elsevier Ltd. All rights reserved. In this paper we show how the Nano can impact the Macro at 103 kg level, fifteen years after the highly speculative concept of carbon nanotube (CNT) cables for space elevatorwas proposed [1].Metal+ CNT composites (MCC), specifically Aluminum+ CNT composite, can now bemass produced (Fig. 1) at less than twice the cost of bulk Al (including raw material cost of CNTs and processing costs), ∗ Corresponding author at: Department of Mechanical Science and Bioengineering, Osaka University, Osaka 560-8531, Japan. ∗∗ Corresponding authors. E-mail addresses: [email protected] (S. Ogata), [email protected] (Y.H. Lee), [email protected] (J. Li). with 150% the mechanical strength of Al but with no loss in tensile ductility (Fig. 2(a)). Analyzing its mechanism is the subject of this Letter, based on our in situ mechanical deformation experiments inside Transmission Electron Microscope (TEM), molecular dynamics (MD) simulations, and standard dislocation mechanics modeling. The shear lag models [2] commonly used for fiber-(polymer)matrix composites are unlikely towork verywell here, as the CNTs are dispersed at approximately the same length scale as the dislocations, and therefore strengthen the composite not only by carrying the load themselves (direct effect, described by the shear lag model), but also by strengthenhttp://dx.doi.org/10.1016/j.eml.2016.04.002 2352-4316/© 2016 Elsevier Ltd. All rights reserved. 2 K.P. So et al. / Extreme Mechanics Letters ( ) – Fig. 1. A proof-of-concept of the mass produced Al + CNT composite. As-cast Al + CNT composite billet with 0.4 vol% CNT, inset: Al + CNT wheels. Fig. 2. Mechanical properties of Al + CNT composite after powder metallurgy and extrusion. (a) The stress vs. strain curves for different CNT contents (ASTM E8 tensile specimen). (b) Young’s modulus (c) tensile strength and fracture strain at different CNT contents. ing the metal matrix through 1D nano-dispersion hardening (indirect effect, based on size-dependentmetal plastic-