Fig. 5: Tunability under asymmetric loading conditions.

a Change of knockdown factor, Δκ_d, at Λ_m = 0.22 (B^r ⋅ B^a > 0) and Λ_m = − 0.22 (B^r ⋅ B^a < 0), versus angle of the shell magnetization, φ^r, and of the applied field, φ^a, which are consistently aligned in the initial configuration (φ^a = φ^r). In the experiments, 16 shell specimens containing a defect at the pole with identical geometry are fabricated (with average defect amplitude \(\overline{\delta }=1.64\pm 0.04\) and defect width \({\overline{\varphi }}_{\circ }=3.2\)). The shells are magnetized at an angle φ^r with respect to the axis of symmetry (2 or 3 specimens at each angle, as indicated by the number of data points). Each specimen (magnetized at φ^r) is tested with the magnetic field oriented at φ^a=φ^r. b, Change of knockdown factor, Δκ_d, at Λ_m = 0.22 (B^r ⋅ B^a > 0) and Λ_m = − 0.22 (B^r ⋅ B^a < 0), as a function of φ^a, with fixed φ^r = 0. The two shell specimens used for the data in b are the same as those from a that are magnetized at φ^r = 0.