Answers are grounded in this paper and cite their sources.
Modelling the anhysteretic magneto-elastic behaviour of a giant magnetostrictive material such as Terfenol-D is not straightforward. The discrete energy-averaged model (DEAM), developed and validated for Galfenol, is cubic and routes the large λ₁₁₁ constant into shear, so it predicts a near-zero longitudinal piezomagnetic coefficient for Terfenol-D unless it is re-parameterized ad hoc. An analytical energy-averaged model was proposed in 2022 for finite-element use, but its closed-form solution, built on the Dawson function, becomes singular under compressive stress and needs an empirically fitted validity limit. We show that this singularity is not physical: the partition integral is always finite, and a direct one-dimensional quadrature evaluates it, together with the magnetization, the magnetostriction and the full piezomagnetic tensor, including the shear coefficient d₁₅, exactly and in closed form for any field and stress. The same three-parameter law reproduces the longitudinal Galfenol response of the DEAM, while the shear channel of cubic materials keeps the DEAM necessary: the two laws are complementary by channel. The law is validated against measured Terfenol-D curves, where adding an anisotropy term to the energy balance improves the agreement. Embedding it at every integration point of an axisymmetric multiphysics finite-element model then quantifies the structure effect, the gap between the material curve and the device response, governed by two dimensionless numbers, χ_0\ N_d for the magnetic effect and k/E/h for the mechanical one, and illustrated on an annular Terfenol-D/PZT current sensor. A frequency study closes the loop: eddy-current screening multiplies the dynamic coefficient by a complex factor, the associated Joule losses heat the material, and the Bloch-law temperature dependence of the constitutive parameters derates the coefficients; thin laminated layers push this limit to the megahertz range.
No takes yet. Share an insight, caveat, or question.
Hakeim Talleb (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: