Studies On The Multi-axial Constitutive Model Of Closed-cell Al Foam And Its Application In Automotive Energy-absorbing Structures

Inspired by the perfect structure of natural porous materials in nature,biological closed-cell aluminum(Al)foams achieve the unity of structure and function,which are usually utilized as the buffer energy-absorbing elements in the impact energy dissipation of rail trains and vehicles applications.Closed-cell Al foams could couple the advantages of both the good toughness of dense metals and the lightweight of porous materials,whose macroscopic behavior are mainly determined by the matrix material properties and meso-structures.Al foams and the corresponding composite structures are usually subjected to multi-axial loadings in practical applications,resulting in large plastic deformation and even failure behavior.Therefore,it is of great necessity to study the multiaxial constitutive model of Al foams.However,the foam cell-wall base material properties are unable to be obtained through traditional experimental methods;the three-dimensional(3D)mesoscopic topology of Al foam is highly complex;dynamic multiaxial loading experiments are rather difficult to be performed in reality.Focused on the above-mentioned questions,microscopic computed tomography(micro-CT)scanning,nano-indentation and multi-axial loading tests are conducted at first.The meso-structure reconstruction of closed-cell Al foam and the inverse identification of foam cell-wall material properties are carried out.Then,the virtual multiaxial numerical simulations of CT-based mesoscopic model are performed;and the quasi-static yield behavior and dynamic/high-temperature failure surface of Al foams are comprehensively characterized.Further,the macroscopic constitutive model of foams involving strain-rate effect is verified numerically.Finally,the actual applications of closed-cell Al foams in novel vehicle energy-absorbing structures are explored.The main research and major findings are summarized as follows:(1)A novel general method for inverse identification and verification of the matrix material properties of porous materials is proposed.The key issue behind this approach could be that the material parameters are the intrinsic properties of one particular material;and then the cell-wall base material properties are identified based upon indentation tests,numerical simulation and optimization technique.The mesoscopic topological structure of closed-cell Al foams is achieved based on micro-CT technology.The identified cell-wall material properties are further verified through comparing uniaxial compression tests with the numerical results based upon the same foam specimen.The quasi-static virtual multiaxial experiment s are conducted on CT-based foam mesoscopic models,and sufficient yield point data are obtained in the mean stress vs.von Mises stress(σ_m,σ_e)space.The initial yield surface shape is enlarged with increasing foam relative density.However,the initial yield surface normalized by the uniaxial yield strength is almost independent of relative density,which could be well fitted by using a unified parabolic or elliptic yield expression.Compared with GAZT and Miller models,the D&F model can better characterize the initial and subsequent yield evolution behavior s of Al foams.(2)The dynamic multiaxial failure behavior of closed-cell Al foams under medium strain rates is explored.The initial failure strength exhibits evident strain-rate dependency,which is established to associate the initial failure strength with relative density and nominal strain rate.The dynamic virtual multiaxial experiments are also performed on CT-based models.It is revealed that the rate-dependence of the foam cell-wall base material is the main factor to cause the hardening of the uniaxial compressive failure strength.The initial failure surface plotted in(σ_m,σ_e)plane significantly expands with increasing loading rates or strain-rates in a geometrically self-similar manner.After normalized by the uniaxial failure strength at the corresponding strain rates,the failure surface is almost independent of strain rates.The normalized failure surface could be well characterized by using theoretical Miller and D&F models.The modified rate-dependent constitutive models are suggested for foam materials subjec t to quasi-static and dynamic loadings.(3)The mechanical behaviors of closed-cell Al foams at elevated temperatures or medium strain-rate conditions are studied based upon the dynamic compressive tests under constant strain-rates,uniaxial compression and compression-shear tests at high temperatures.The initial failure surface expands with increasing Al foam density or decreasing temperature.However,after normalized by the failure strength at the corresponding foam density or temperature,the failure surface is almost independent of foam density or temperature,which can be also characterized by using a parabolic or an elliptic yield function.Based on the experimental results of uniaxial compression at constant strain-rates,a constitutive model of Al foams with considering strain-rates is proposed and its reasonability together with D&F model involving strain-rate effects are both calibrated numerically.(4)The applications of closed-cell Al foams in the development of novel automotive energy-absorbing structures are investigated.The crashworthiness of Al foam filled hat-shaped CFRP structures is analyzed experimentally.It is indicated that three point bending and transverse compression crushing performance of Al foam filled hat-shaped CFRP structures are enhanced by the interaction effects in between Al foam filler and CFRP structure.Based upon the calibrated D&F foam constitutive model,numerical approach is utilized to study the frontal crashworthiness performance of Al foam filled bumper beam as well as the pedestrian protection capability of Al foam laminated engine hood,which divulges that the introduction of closed-cell Al foams into the key structural energy-absorbing parts of automotive can effectively improve the lightweight and impact safety performance of automotive.