Oxidation effect on the shear strength of graphene on aluminum and titanium surfaces

We report the interfacial shear strength of graphene on pure and oxidized Ti and Al metal surfaces using density functional theory calculations. Our results show significant changes to the graphene-metal bonding properties in the presence of an oxide phase. In particular, the strongly chemisorbed interface between graphene and pure Ti is drastically weakened by the formation of a metal-oxide phase, while the weakly physisorbed interface between graphene and pure Al is significantly strengthened through the metal oxide formation. These oxidation effects can be modulated to some extent by the presence of vacancy or Stone-Wales defects which increases the binding interactions of weaker graphene-metal interfaces. These dramatic changes to the interfacial properties by surface oxidation explain the results of recent carbon nanotube pull-out experiments from Al and Ti metal-matrix nanocomposites.

Figure 1

Cross-sectional views of the (a) atomic configurations and (b) electron-localized function (ELF) contours of graphene on bare metal surfaces of Al (Al-graphene) and Ti (Ti-graphene), O atoms bonded to exposed Al (Al-O-graphene) and Ti (Ti-O-graphene) metal surfaces, and O-terminated bulk $\alpha -\mathrm{A}{\mathrm{l}}_2{\mathrm{O}}_3$ ($\mathrm{A}{\mathrm{l}}_2{\mathrm{O}}_3$-graphene) and rutile $\mathrm{Ti}{\mathrm{O}}_2$ ($\mathrm{Ti}{\mathrm{O}}_2$-graphene). Atoms colored in red, green, yellow, and orange represent O, Ti, Al, and C, respectively. Dashed black lines denote in-plane periodicity of the supercells. An ELF contour value of close to 0.5 (green) corresponds to a uniform smeared-out electron cloud as in metallic bonding, while an ELF value of 1.0 (red) denotes high probability of finding electron localization as in covalent bonding.

Figure 2

Sliding of graphene on the bare Ti surface. (a) In-plane atomic configuration of Ti-graphene at the initial minimum-energy state, as viewed from the bottom. (b) Energy contours $\mathrm{\Delta }\gamma$ associated with in-plane sliding of graphene for Ti-graphene. (c,d) Evolution of $\mathrm{\Delta }\gamma$ and shear stress $\tau$ along possible minimum-energy sliding pathways a–c marked in (b).

Figure 3

Sliding of graphene on the bare Al surface. (a) In-plane atomic configuration of Al-graphene at the initial minimum-energy state, as viewed from the bottom. (b) Energy contours $\mathrm{\Delta }\gamma$ associated with in-plane sliding of graphene for Al-graphene. (c, d) Evolution of $\mathrm{\Delta }\gamma$ and shear stress $\tau$ with sliding displacements $\delta$ along possible minimum-energy pathways a–c marked in (b).

Figure 4

Sliding of graphene on oxidized Ti and Al. (a) In-plane atomic configurations of graphene on surface-oxidized (Ti-O, Al-O) and bulk oxide ($\mathrm{Ti}{\mathrm{O}}_2$, $\mathrm{A}{\mathrm{l}}_2{\mathrm{O}}_3$) metals, as viewed from the bottom. (b) Energy contours $\mathrm{\Delta }\gamma$ associated with in-plane sliding of graphene on the respective metal substrates.

Figure 5

Evolution of energy change $\mathrm{\Delta }\gamma$ vs sliding displacement $\delta$ of graphene along possible minimum-energy sliding pathways of (a) Ti-O, (b) $\mathrm{Ti}{\mathrm{O}}_2$, (c) Al-O, and (d) $\mathrm{A}{\mathrm{l}}_2{\mathrm{O}}_3$ substrates.

Figure 6

Evolution of shear stress $\tau$ vs sliding displacement $\delta$ of graphene along possible minimum-energy sliding pathways of (a) Ti-O, (b) $\mathrm{Ti}{\mathrm{O}}_2$, (c) Al-O, and (d) $\mathrm{A}{\mathrm{l}}_2{\mathrm{O}}_3$ substrates.

Figure 7

Sliding of graphene with Stone-Wales defects on bare-metal Ti and Al surfaces and the respective bulk oxide surfaces. (a) In-plane atomic configurations of graphene on the respective metal substrates, as viewed from the bottom. (b) Energy contours $\mathrm{\Delta }\gamma$ associated with in-plane sliding of graphene on the respective metal substrates.

Figure 8

Sliding of graphene with vacancy defects on bare-metal Ti and Al surfaces and the respective bulk oxide surfaces. (a) In-plane atomic configurations of graphene on the respective metal substrates, as viewed from the bottom. (b) Energy contours $\mathrm{\Delta }\gamma$ associated with in-plane sliding of graphene on the respective metal substrates.

Figure 9

Carbon nanotube (CNT) pull-out experiments from Ti and Al matrices. (a) Schematic of the pull-out process for a CNT with diameter of 3.12 nm subjected to applied pull-out force $F_x$. (b, c) Pull-out force vs nanotube length for Ti- and Al-CNT nanocomposites. Symbols in blue denote results from room-temperature experiments conducted without thermal annealing, while symbols in red denote results for thermally annealed nanocomposites. Results for thermally annealed Ti-CNT are currently unpublished, while the remaining results are published in [16, 18].