Evidence of substrate roughness surface induced magnetic anisotropy in Ni80Fe20 flexible thin films

Abstract

Experimental and computational evidence of a surface roughness induced magnetic anisotropy in NiFe thin films coated onto substrates of various surface roughnesses is reported. Magnetic coercive fields of 15 nm NiFe thin films coated on substrates with approximately 7 nm average roughness were remarkably 233% larger than identical thin films coated onto smooth substrates with < 1 nm average roughness. The NiFe films coated onto rough substrates developed hard and easy axes, normally non-existent in NiFe Permalloy. A linear correlation of the incline angles of the hard axis hysteresis loops to the average roughness values of the individual substrates was observed, with 99% correlation level. Using a modified micromagnetics theory that incorporates the effects of surface roughness, it is shown the observed magnetic anisotropy arises due to the spatial anisotropy of the surface roughness, resulting in an effective in-plane uniaxial magnetic anisotropy with energy density up to 15 kJ/m 3 .

Study of roughness effect in Fe and Co thin films prepared by plasma magnetron sputtering

“…We see that Co, having the highest KU is almost unaffected by the roughness of substrates, whilst the other two materials, Fe and Ni80Fe20, scale well with KU and appear to be greatly influenced by the roughness values. We believe this is due to a surface roughness induced magnetic anisotropy as predicted by Lepadatu [14] and proven experimentally in other studies [15,17,20].…”

“…As shown in previous works on NiFe thin films [14,15,17], the effect of topographical surface roughness is to induce an effective anisotropy term due to modification of the magnetostatic field. Locally, the induced anisotropy strongly depends on the roughness profile, however when averaged over the entire sample a uniaxial anisotropy contribution results, with easy axis aligned to the average roughness texture orientation.…”

“…This was found to be in direct contrast with the grain size effect for which the HC is expected to be lower in thinner films due to a smaller grain size [16]. This remarkable and contradicting result was attributed to a change in effective magnetic anisotropy, induced by the substrate's surface roughness, as predicted theoretically [14] and later proven experimentally [17].…”

“…Magnetic thin films were fabricated using three different substrates: a) Silicon (Si) wafer (100); b) Kapton® thermal film 50 µm thick; c) 110 µm polyvinylidene fluoride (PVDF) film. These materials were chosen because of their diverse average surface roughness values (Ra) that allow studying the roughness effect onto the magnetic properties of the final films, but also to carry out a comparative study with the previous work on Ni80Fe20 thin films [15,17]. The roughness values as measured after plasma etchings are 0.6 nm, 2.2 nm and 6.8 nm for Si, Kapton and PVDF, respectively as reported previously [17].…”

“…These materials were chosen because of their diverse average surface roughness values (Ra) that allow studying the roughness effect onto the magnetic properties of the final films, but also to carry out a comparative study with the previous work on Ni80Fe20 thin films [15,17]. The roughness values as measured after plasma etchings are 0.6 nm, 2.2 nm and 6.8 nm for Si, Kapton and PVDF, respectively as reported previously [17]. All samples have the following structure: Substrate/Cr (5nm)/Ferromagnetic material (t nm), where t = 15 nm and 100 nm represents the film thickness.…”