Ferromagnetism of Nanometer Thick Sputtered Fe₃GeTe₂ Films in the Absence of Two-Dimensional Crystalline Order: Implications for Spintronics Applications

Abstract

The discovery of ferromagnetism in two-dimensional (2D) monolayers has stimulated growing research interest in both spintronics and material science. However, these 2D ferromagnetic layers are mainly prepared through an incompatible approach for large-scale fabrication and integration, and moreover, a fundamental question of whether the observed ferromagnetism actually correlates with the 2D crystalline order has not been explored. Here, we choose a typical 2D ferromagnetic material, Fe3GeTe2, to address these two issues by investigating its ferromagnetism in an amorphous state. We have fabricated nanometer thick amorphous Fe3GeTe2 films approaching the monolayer thickness limit of crystallized Fe3GeTe2 (0.8 nm) through magnetron sputtering. Compared to crystallized Fe3GeTe2, we found that the basic ferromagnetic attributes, such as the Curie temperature which directly reflects magnetic exchange interactions and local anisotropic energy, do not change significantly in the amorphous states. This is attributed to the short-range atomic order, as confirmed by valence state analysis, being almost the same for both phases. The persistence of ferromagnetism in the ultrathin amorphous counterpart has also been confirmed through magnetoresistance measurements, where two unconventional switching dips arising from electrical transport within domain walls are clearly observed in the amorphous Fe3GeTe2 single layer. These results indicate that the long-range ferromagnetic order of crystallized Fe3GeTe2 may not correlate to the 2D crystalline order, and the corresponding ferromagnetic attributes can be utilized in an amorphous state which suits large-scale fabrication in a semiconductor technology-compatible manner for spintronics applications.

Proximity-Induced Interfacial Room-Temperature Ferromagnetism in Semiconducting Fe3GeTe2

“…On the other hand, from the point of view of practical applications, the demonstrated interfacial MPE provides a reliable approach to increase the functional temperature of those newly developed low-dimensional materials to room-temperature in the absence of magnetic ion doping or extra electrodes. Since both the amorphous and crystallize FGT show similar ferromagnetic properties, the MPE reported here may also persist in a crystallize FGT/Pt interface.…”

“…To reveal the possible ferromagnetic contribution from FGT itself, R H of the 30 nm FGT single layer was also measured. As mentioned above, the FGT single layer does not show ferromagnetic behaviors at room temperature, and the measured R H curves only show normal linear Hall signal above 200 K. As reported in previous works, even for the sputtered amorphous FGT lacking long-range crystalline order, when the temperature drops below 180 K, R H signal increases dramatically and becomes hysteretic with decreasing temperature . The ferromagnetic transition temperature is the same as the well-crystallized counterparts with the T c of about 180 K. , To clearly demonstrate the temperature dependence of ferromagnetism, the anomalous Hall resistance ( R H s ) originating from ferromagnetism was extracted by subtracting a linear normal Hall contribution.…”

“…The additional Hall contribution has been widely used to describe ferromagnetism both experimentally and theoretically . As shown in Figure a, the room-temperature R H curves show a linear H dependence for the sputtered single FGT layers, which is the same as the crystallized FGT because of no ferromagnetism. ,, Remarkably, for the FGT (10)/Pt (3) bilayer, clear hysteresis behaviors appear in the R H curve at room temperature, indicating that the ferromagnetism has been induced in the FGT/Pt bilayer with T c larger than 300 K. Since both the FGT and Pt single layers do not show ferromagnetism at room temperature, the ferromagnetism must arise from the FGT/Pt interfaces due to proximity effects. For the Sub/Pt (3)/FGT (10)/capping layer where the Pt layer was deposited first, R H shows the same behaviors as FGT (10)/Pt (3) (Figure S3), indicating that the observed room-temperature ferromagnetism does not depend on the deposition sequence of the FGT and Pt layers.…”

“…The purity of the Fe 3 GeTe 2 target is 99.9% and that of the Pt target is 99.95%. The actual atomic percentages of Fe, Ge, and Te in sputtered FGT films are 42.1%, 24.6%, and 33.3%, respectively, in which the Fe concentration is less while that of Ge is higher than their concentrations in FGT targets. The deposited multilayers were patterned into a Hall bar structure with the length of 50 μm–200 μm and width of 2 μm–10 μm by using standard photolithography and ion milling processes.…”