People on CEAD Lab

Mon, 01 Jan 0001 00:00:00 +0000

Prashant leads the CEAD Lab. His work asks how much physics and computational fidelity a model needs to make a reliable engineering prediction. That question has led from mathematical analysis of fracture, through particle-resolved and multiphysics modeling of heterogeneous materials, to multiscale design and scientific-AI methods whose errors can be quantified and acted upon.

The mechanics came first: a master’s at the Indian Institute of Science with Chandrashekhar S. Jog, on monolithic compressible fluid–structure interaction, where much of the nonlinear continuum mechanics and finite element method that the rest of this rests on was learned.

Mon, 01 Jan 0001 00:00:00 +0000

Alison worked on how a crack crosses the boundary between a particle and the matrix holding it, comparing phase-field, diffuse-interface and cohesive-zone descriptions of interface fracture, which disagree about exactly the case that matters. She also looked at whether a double-cantilever-beam test could supply fracture energies consistent enough to tell them apart.

Mon, 01 Jan 0001 00:00:00 +0000

An optimized design is a field of numbers until something is made from it. Caleb works on the step between: taking designs that come out of the material-and-structure optimization and getting them printed, which means confronting what a printer can actually resolve, what the cured material actually is, and how far a fabricated specimen departs from the geometry that was optimized.

That gap is the useful measurement. It is what will say which parts of a design are worth optimizing to three decimal places and which are being erased in fabrication anyway.

Mon, 01 Jan 0001 00:00:00 +0000

Between a stiff particle and the soft matrix around it there is an interface, and under load that interface is a likely place for things to go wrong first. It can carry load, or it can slip, or it can debond and unload the particle entirely, and which of those happens changes the composite’s stiffness and its strength together.

Henry models that interface with cohesive-zone descriptions, in the setting of magnetic soft materials, where the particles are also what the material responds with. Get the interface wrong and the actuation prediction can be wrong for reasons that have nothing to do with the magnetics.

Mon, 01 Jan 0001 00:00:00 +0000

Ian works on designing hard-magnetic soft materials where the geometry and the material are chosen together: where to put material, how much magnetic particle it should carry, and which way its remanent magnetization should point. Optimizing any one of those against fixed values of the others finds a worse design than optimizing them jointly, and doing it honestly means settling first on which constitutive description to trust.

That is the order the 2026 manuscript works in: twenty-one pairings of effective shear-modulus relation and strain-energy function, compared first on the actuation problems themselves; then experimental stress-strain data to settle which shear-modulus relation is worth carrying forward; only then the joint optimization. He is its first author, and came to it from undergraduate research in the group.