Florian Thibaut Gallien
Mass is one of the crucial parameters for hardware that has to be placed in Earth orbit. Due to its harsh environment, a material with highest specific properties is desired to achieve space missions. The rise and development of new technologies, such as additive manufacturing (AM), opened new opportunities in part-design complexity, periodic cellular structures (PCS) being one of them. The present thesis investigates the potential implementation of PCS in space applications, particularly for structures and micro-meteoroids and orbital debris (MMOD) impact shields. This was achieved in three steps:
Four different types of AlSi12 PCS manufactured by selective laser melting (SLM) were tested under quasi-static compression to measure the mechanical properties dependency versus topology and to characterize the failure mode. Properties ranging from 3 to 4 GPa for the compressive modulus, 5 to 12 MPa for the yield stress, 12 to 20 MPa for the plateau stress, and 2 to 8 MJ/cm3 for the absorbed energy were obtained. An unexpected failure mode was observed when compared to classical cellular metals, namely a brittle failure occurring by global shearing. A predictive failure criterion was established based on topology considerations and correlated to most of the reported results in the literature. A preliminary test campaign on tensile specimens was performed to compute numerical models that were fed into a finite element analysis. Good agreement with experimental data was shown, and the importance of microplasticity effects in this class of material was highlighted.
An alternative process was developed to produce AlSi1 PCS by investment casting. The process is based on replication of a polymer preform used to build a NaCl mold. It was observed that the quality of the final cast part depends mainly on the grain size of the salt, with an optimum identified for distributions between 125 and 180 um. Optimization of the process allowed to reduce the drying time by a factor 6. Main process parameters include a drying temperature of 80C and infiltration at 660C under 300 mbar. From this process, PCSs having an energy absorption capacity of 15 MJ/m3 with an efficiency of 80% were produced.
Hypervelocity impact tests were conducted on cast PCS and stochastic structures. The objective being to hit the structures with a 2mm-diameter aluminum sphere at velocities close to 7 km/s. Influence of the sample topology, the orientation, and the bumper material was assessed. Stochastic structures successfully stopped the projectile in all configurations. The beneficial effect of the bumper was measured reducing the crater depth from 20 mm to 14 mm. This type of structure exhibited a comparable areal density (0.8 g/cm2) to simple Whipple shield design. PCS poorly performed in mitigating the impact as the debris passed through all the structures, independently of the test configuration due to the open-channels present.
PCS are good candidates to be used in space hardware, but their design and the manufacturing process need to be carefully chosen depending on the specific application. AM PCS are suitable for structural application with a high compressive modulus and yield stress. Cast PCS would perfectly fit in shock absorbers. A more random design would be preferable for MMOD shielding applications.
EPFL2018
