Reinforcing effect of graphene oxide on cement mortar under high strain rate loadings

Abstract

Reinforcing cement-based materials by nanomaterials has attracted extensive attention over the past decades. It has been reported that the mechanical properties of cement-based materials can be greatly improved by incorporating a small amount of nanomaterials. However, experimental and theoretical works have all focused on the enhancement of cement-based materials subjected to static loading. To the author’s knowledge, no studies have reported on the mechanical properties of nanomaterial- reinforced cement-based composites under high strain rate loadings. This study investigates the reinforcing effect of graphene oxide (GO) on cement mortar under high strain loadings. GO is an emerging nano-scale candidate for enhancing cement-based materials, which is valued for its two-dimensional geometry, superior mechanical properties and large surface area. Tensile splitting and compression tests were conducted on GO-reinforced cement mortar under both static and high strain rate loadings. The high strain rate testing was performed using a Split Hopkinson Pressure Bar (SHPB) apparatus. Static experiments showed that incorporation of GO (of 0.02 wt% of cement) improved the tensile splitting and compression strength of plain cement mortar by 8.9% and 10.8%, respectively, and there was little difference between the 0.02 wt% and 0.04 wt% GO addition. The SHPB tests showed that the reinforcing effect of GO under high strain rate is directly related to the number of cracks. In tensile splitting tests, because the major crack is always localized in the centre of the specimen regardless of the strain rate, the reinforcing effects of GO on cement mortar were the same for different strain rates. In compression tests, however, because the cracks were diffused and there were more cracks under high strain rates, the reinforcing effect of GO was also more significant under higher strain rates. To understand the reinforcing mechanisms of the nanomaterials, the key is to determine the interaction properties between the nano-scale reinforcements and the cement matrix. However, understanding of the governing forces in the reinforcement’s pull-out behaviour at such a small scale is still very limited. The present study is the first to identify the governing force during the pull-out of atomically thin two dimensional (2D) nanosheets (e.g. graphene and its derivative GO) and to develop corresponding theoretical models. In molecular dynamic (MD) simulations, friction was found to make negligible contribution to the pull-out force because of the lack of asperities on atomically thin materials (ATM). The pull-out force was revealed to be governed by a “crack plane adhesion”. Unlike frictional pull-out, crack plane adhesion produces a pull-out force independent of the embedded length of ATM. The magnitude of pull-out force and its affecting factors were investigated by MD simulations. On the basis of crack plane adhesion, the relations between the pull-out force and the pull-out displacement for ATMs was formulated. Furthermore, a new theoretical model was developed to predict the crack bridging stress (σB) of 2D ATMs. The magnitude of maximum σB was estimated to be 3MPa for 0.2 wt% of GO in cement matrix, which was used in the finite element simulations as discussed herein. The fundamental theory and new model proposed here serve as theoretical support for understanding and development of ATM-reinforced composite materials. Finite element simulations for static and SHPB compression tests were conducted using a micromechanical model, the microplane model. First, the microplane model was examined by comparative study with the commonly used concrete damaged plasticity model. Then finite element simulation models of static and SHPB compression tests were built in ABAQUS. For simulation of the static test, by including the calculated crack bridging stress into microplane model, the experimental compressive strength of GO-reinforced cement mortar could be accurately simulated, which validated the calculation of the crack bridging stress by the proposed model in Chapter 4 and the effectiveness of the microplane model to simulate the static behaviour of GO-reinforced cement-based composites. For simulation of the SHPB test, the microplane model with a logarithmic relation between the dynamic increase factor (DIF) and the strain rate to account for the strain rate effect worked for the simulations of plain cement mortar samples. However, for simulation of the GO-reinforced mortar samples, the simulation results give a significant underestimation compared with the experimental results under higher strain rates.