No Arabic abstract
Disordered heterostructures stand as a general description for compounds that are part of homologous series such as bismuth chalcogenides. In device engineering, van der Waals epitaxy of these compounds is very promising for applications in spintronic and quantum computing. Structural analysis methods are essential to control and improve their synthesis in the form of thin films. Recently, X-rays tools have been proposed for structural modeling of disordered heterostructures [arXiv:2107.12280]. Here, we further evaluate the use of these tools to study the compound Mn$_x$Bi$_2$Te$_{3+x}$ in the grazing incidence region of the reflectivity curves, as well as the effect of thickness fluctuation in the wide angle region.
Synthesis of new materials demands structural analysis tools suited to the particularities of each system. Van der Waals (vdW) materials are fundamental in emerging technologies of spintronics and quantum information processing, in particular topological insulators and, more recently, materials that allow the phenomenological exploration of the combination of non-trivial electronic band topology and magnetism. Weak vdW forces between atomic layers give rise to composition fluctuations and structural disorder that are difficult to control even in a typical binary topological insulators such as Bi2Te3. The addition of a third element as in MnBi2Te4 makes the epitaxy of these materials even more chaotic. In this work, statistical model structures of thin films on single crystal substrates are described. It allows the simulation of X-ray diffraction in disordered heterostructures, a necessary step towards controlling the epitaxial growth of these materials. On top of this, the diffraction simulation method described here can be readily applied as a general tool in the field of design new materials based on stacking of vdW bonded layers of distint elements.
Alloys of Bi$_2$Te$_3$ and Sb$_2$Te$_3$ ((Bi$_{1-x}$Sb$_x$)$_2$Te$_3$) have played an essential role in the exploration of topological surface states, allowing us to study phenomena that would otherwise be obscured by bulk contributions to conductivity. Thin films of these alloys have been particularly important for tuning the energy of the Fermi level, a key step in observing spin-polarized surface currents and the quantum anomalous Hall effect. Previous studies reported the chemical tuning of the Fermi level to the Dirac point by controlling the Sb:Bi composition ratio, but the optimum ratio varies widely across various studies with no consensus. In this work, we use scanning tunneling microscopy and Landau level spectroscopy, in combination with X-ray photoemission spectroscopy to isolate the effects of growth factors such as temperature and composition, and to provide a microscopic picture of the role that disorder and composition play in determining the carrier density of epitaxially grown (Bi,Sb)$_2$Te$_3$ thin films. Using Landau level spectroscopy, we determine that the ideal Sb concentration to place the Fermi energy to within a few meV of the Dirac point is $xsim 0.7$. However, we find that the post- growth annealing temperature can have a drastic impact on microscopic structure as well as carrier density. In particular, we find that when films are post-growth annealed at high temperature, better crystallinity and surface roughness are achieved; but this also produces a larger Te defect density, adding n-type carriers. This work provides key information necessary for optimizing thin film quality in this fundamentally and technologically important class of materials.
We study the chemical ordering in Bi$_2$Te$_{3-x}$Se$_x$ grown by molecular beam epitaxy on Si substrates. We produce films in the full composition range from x = 0 to 3, and determine their material properties using energy dispersive X-ray spectroscopy, X-ray diffraction and Raman spectroscopy. By fitting the parameters of a kinetic growth model to these results, we obtain a consistent description of growth at a microscopic level. Our main finding is that despite the incorporation of Se in the central layer being much more probable than that of Te, the formation of a fully ordered Te-Bi-Se-Bi-Te layer is prevented by kinetic of the growth process. Indeed, the Se concentration in the central layer of Bi$_2$Te$_2$Se$_1$ reaches a maximum of only $approx$ 75% even under ideal growth conditions. A second finding of our work is that the intensity ratio of the 0 0 12 and 0 0 6 X-ray reflections serves as an experimentally accessible quantitative measure of the degree of ordering in these films.
We have investigated the optical properties of thin films of topological insulators Bi$_{2}$Te$_{3}$, Bi$_{2}$Se$_{3}$ and their alloys Bi$_2$(Te$_{1-x}$Se$_x$)$_3$ on BaF$_{2}$ substrates by a combination of infrared ellipsometry and reflectivity in the energy range from 0.06 to 6.5 eV. For the onset of interband absorption in Bi$_2$Se$_3$, after the correction for the Burstein-Moss effect, we find the value of direct bandgap of $215pm10$ meV at 10 K. Our data supports the picture that Bi$_2$Se$_3$ has a direct band gap located at the $Gamma$ point in the Brillouin zone and that the valence band reaches up to the Dirac point and has the shape of a downward oriented paraboloid, i.e. without a camel-back structure. In Bi$_2$Te$_3$, the onset of strong direct interband absorption at 10 K is at a similar energy of about 200 meV, with a weaker additional feature at about 170 meV. Our data support the recent GW band structure calculations suggesting that the direct interband transition does not occur at the $Gamma$ point but near the Z-F line of the Brillouin zone. In the Bi$_2$(Te$_{1-x}$Se$_x$)$_3$ alloy, the energy of the onset of direct interband transitions exhibits a maximum near $x=0.3$ (i.e. the composition of Bi$_2$Te$_2$Se), suggesting that the crossover of the direct interband transitions between the two points in the Brillouin zone occurs close to this composition.
We used low-energy, momentum-resolved inelastic electron scattering to study surface collective modes of the three-dimensional topological insulators Bi$_2$Se$_3$ and Bi$_{0.5}$Sb$_{1.5}$Te$_{3-x}$Se$_{x}$. Our goal was to identify the spin plasmon predicted by Raghu and co-workers [S. Raghu, et al., Phys. Rev. Lett. 104, 116401 (2010)]. Instead, we found that the primary collective mode is a surface plasmon arising from the bulk, free carrers in these materials. This excitation dominates the spectral weight in the bosonic function of the surface, $chi (textbf{q},omega)$, at THz energy scales, and is the most likely origin of a quasiparticle dispersion kink observed in previous photoemission experiments. Our study suggests that the spin plasmon may mix with this other surface mode, calling for a more nuanced understanding of optical experiments in which the spin plasmon is reported to play a role.