Altermagnets are compensated magnets belonging to spin symmetry groups that allow alternating spin polarizations both in the coordinate space of the crystal and in the momentum space of the electronic structure. In these materials the anisotropic local crystal environment of the different sublattices lowers the symmetry of the system so that the opposite-spin sublattices are connected only by rotations. This results in an unconventional spin-polarized band structure in the momentum space. This low symmetry of the crystal structure is expected to be reflected in the anisotropy of the anomalous Hall effect. In this work, we study the anisotropy of the anomalous Hall effect in epitaxial thin films of ${\mathrm{Mn}}_5{\mathrm{Si}}_3$, an altermagnetic candidate material. We first demonstrate a change in the relative Néel vector orientation when rotating the external field orientation through systematic changes in both the anomalous Hall effect and the anisotropic longitudinal magnetoresistance. We then show that the anomalous Hall effect in this material is anisotropic with the Néel vector orientation relative to the crystal structure and that this anisotropy requires high crystal quality and unlikely stems from the magnetocrystalline anisotropy. Our results thus provide further systematic support to the case for considering epitaxial thin films of ${\mathrm{Mn}}_5{\mathrm{Si}}_3$ as an altermagnetic candidate material.

Some magnetically ordered phases of the ${\mathrm{Mn}}_5{\mathrm{Si}}_3$ crystal are proving to be prototypes for the study of the new fundamental spin physics related to the spontaneous breaking of the time-reversal symmetry despite a zero net magnetization. Here, we report on a route to grow epitaxial ${\mathrm{Mn}}_5{\mathrm{Si}}_3$ thin films on Si(111). To this end, we use Mn and Si codeposition in a molecular beam epitaxy system and carefully tune the deposition rates, the growth temperature, and the annealing temperature. We assessed the silicide phase-formation and morphology using reflection high-energy electron diffraction, x-ray diffraction, high-resolution transmission electron microscopy (HRTEM) and atomic force microscopy. Layers containing only ${\mathrm{Mn}}_5{\mathrm{Si}}_3$ could be stabilized under very restrictive conditions, by tuning the Mn/Si flux ratio to match the compound stoichiometry and adjusting the substrate temperature during growth to 443 K. HRTEM imaging revealed the existence of an interfacial amorphous layer of few nanometers thickness. Annealing the heterostructure up to 573 K led to the formation of MnSi at the vicinity of the ${\mathrm{Mn}}_5{\mathrm{Si}}_3$/Si(111) interface, which significantly reduced the nucleation barrier of ${\mathrm{Mn}}_5{\mathrm{Si}}_3$. High-quality crystalline ${\mathrm{Mn}}_5{\mathrm{Si}}_3$ thin films were then formed with the following epitaxial relationships: ${\mathrm{Mn}}_5{\mathrm{Si}}_3$(0001)$\left[01\overline{1}0\right]//\mathrm{MnSi}\left(111\right)\left[\overline{2}11\right]//\mathrm{Si}\left(111\right)\left[1\overline{1}0\right]$. Our experiments showed that the formation of MnSi is enhanced at a growth temperature above 473 K or for a longer annealing step, while the crystalline quality of the ${\mathrm{Mn}}_5{\mathrm{Si}}_3$ overlayer is correspondingly degraded leading to textured thin films. The growth pathways and structural properties of the manganese silicides can be rationalized in terms of reactions maximizing the free-energy lowering rate. Moreover, we found that the magnetic and the magnetotransport properties can be used as an efficient tool to track both ${\mathrm{Mn}}_5{\mathrm{Si}}_3$ crystallinity and proportion in the deposited layers.

High crystalline quality Mn5Ge3 films with thicknesses ranging 4\char21{}200 nm have been grown on Ge(111) substrates by solid phase epitaxy. The basal hexagonal plane of Mn5Ge3 is in epitaxy with the Ge(111) plane. Magnetic properties of the films have been investigated as a function of the film thickness and the magnetization curves have been analyzed using a theory that includes a description of magnetic domains in uniaxial thin films. The results clearly indicate the existence of a critical thickness below which the magnetic stripe phase disappears. We have determined the value of this thickness to lie between 10 and 25 nm from the analysis of experimental magnetization curves and the theoretical fit of the in-plane remanent magnetization. Although analogies can be drawn between the behavior observed in our system and that of hcp Co, we have shown that the critical thickness is considerably smaller in Mn5Ge3; this has the potential to open new fields of applications for Mn5Ge3 thin films in magnetic recording and spintronics.

Mn₅Ge₃ compound, with its room-temperature ferromagnetism and possibility to epitaxially grow on Ge, acts as a potential spin injector into group-IV semiconductors. However, the realization of Ge/Mn₅Ge₃ multilayers is highly hampered by Mn segregation toward the Ge growing surface. Here, we show that adsorption of some monolayers of carbon on top of the Mn₅Ge₃ surface prior to Ge deposition allows to greatly reduce Mn segregation. In addition, a fraction of deposited carbon can diffuse down to the underneath Mn₅Ge₃ layers, resulting in an enhancement of the Curie temperature up to ∼360 K. The obtained results will be discussed in terms of the formation of a diffusion barrier by filling interstitial sites of Mn₅Ge₃ by carbon.

We present a comprehensive study of magnetization reversal process in thin films of . For this investigation, we have studied the magnetic anisotropy of layers as a function of the film thickness using VSM and SQUID magnetometers. The samples grown by molecular beam epitaxy exhibit a reorientational transition of the easy axis of magnetization from in-plane to out-of-plane as the film thickness increases. We provide evidence that above a critical thickness estimated as 20 nm, the magnetic structure is most probably constituted of stripes with out-of-plane magnetization pointing alternately up and down. We have analyzed our results using different phenomenological models and all the calculations converge towards values for magnetocrystalline anisotropy constant and saturation magnetization that are in excellent agreement with the reported values for bulk . This study has also led to the first estimation in of the exchange constant, the surface energy of domain walls as well as their width. These parameters are essential for determining whether this material can be used in the next generation of spintronic devices.