Search for: All records

This work introduces a computational method for designing ceramic scaffolds fabricated via direct ink writing (DIW) for maximum bone growth, whereby the deposited rods are curvilinear. A mechanobiological model of bone adaptation is used to compute bone growth into the scaffold, taking into account the shape of the defect, the applied loading, and the density distribution of bone in which the scaffold is implanted. The method ensures smooth, continuously varying rod contours are produced which are ideal for the DIW process. The method uses level sets of radial basis functions to fully define the scaffold geometry with a small number of design variables, minimizing the optimization’s computational cost. Effective elastic and diffusive properties of the scaffold as a function of the scaffold design and the bone density are obtained from previously constructed surrogates. These property surrogates are in turn used to perform bone adaptation simulations of the scaffold-bone system. Design sensitivities of the bone ingrowth within the scaffold are efficiently obtained using a finite difference scheme implemented in parallel. A demonstration of the methodology on a scaffold implanted in a pig mandible is presented. The scaffold is optimized to maximize bone ingrowth with geometric constraints to conform to the manufacturing process. 

Larval settlement in wave-dominated, nearshore environments is the most critical life stage for a vast array of marine invertebrates, yet it is poorly understood and virtually impossible to observe in situ . Using a custom-built flume tank that mimics the oscillatory fluid flow over a shallow coral reef, we isolated the effect of millimeter-scale benthic topography and showed that it increases the settlement of slow-swimming coral larvae by an order of magnitude relative to flat substrates. Particle tracking velocimetry of flow fields revealed that millimeter-scale ridges introduced regions of flow recirculation that redirected larvae toward the substrate surface and decreased the local fluid speed, effectively increasing the window of time for larvae to settle. Regions of recirculation were quantified using the Q -criterion method of vortex identification and correlated with the settlement locations of larvae for the first time. In agreement with experiments, computational fluid dynamics modeling and agent-based larval simulations also showed significantly higher settlement onto ridged substrates. Additionally, in contrast to previous reports on the effect of micro-scale substrate topography, we found that these topographies did not produce key hydrodynamic features linked to increased settlement. These findings highlight how physics-based substrate design can create new opportunities to increase larval recruitment for ecosystem restoration.