Material characterisation of additively manufactured elastomers at different strain rates and build orientations

Abstract

Material jetting, particularly PolyJet technology, is an additive manufacturing (AM) process which has introduced novel flexible elastomers used in bio-inspired soft robots, compliant structures and dampers. Finite Element Analysis (FEA) is a key tool for the development of such applications, which requires comprehensive material characterisation utilising advanced material models. However, in contrast to conventional rubbers, PolyJet elastomers have been less explored leading to a few material models with various limitations in fidelity. Therefore, one aim of this study was to characterise the mechanical response of the latest PolyJet elastomers, Agilus30 (A30) and Tango+ (T+), under large strain tension-compression and time-dependent high-frequency/relaxation loadings. Another aim was to calibrate a visco-hyperelastic material model to accurately predict these responses. Tensile, compressive, cyclic, dynamic mechanical analysis (DMA) and stress relaxation tests were carried out on pristine A30 and T+ samples. Quasi-static tension-compression tests were used to calibrate a 3-term Ogden hyperelastic model. Stress relaxation and DMA results were combined to determine the constants of a 5-term Prony series across a large window of relaxation time (10 μs–100 s). A numerical time-stepping scheme was employed to predict the visco-hyperelastic response of the 3D-printed elastomers at large strains and different strain rates. In addition, the anisotropy in the elastomers, which stemmed from build orientation, was explored. Highly nonlinear stress-strain relationships were observed in both elastomers, with a strong dependency on strain rate. Relaxation tests revealed that A30 and T+ elastomers relax to 50 % and 70 % of their peak stress values respectively in less than 20 s. The effect of orientation on the loading response was most pronounced with prints along the Z-direction, particularly at large strains and lower strain rates. Moreover, the visco-hyperelastic material model accurately predicted the large strain and time-dependent behaviour of both elastomers. Our findings will allow for the development of more accurate computational models of 3D-printed elastomers, which can be utilised for computer-aided design in novel applications requiring flexible or rate-sensitive AM materials.