We predict that unpolarized charge current injected into a ballistic thin film of prototypical topological insulator (TI) ${\mathrm{Bi}}_2{\mathrm{Se}}_3$ will generate a noncollinear spin texture $\mathbf{S}\left(\mathbf{r}\right)$ on its surface. Furthermore, the nonequilibrium spin texture will extend into an $\simeq 2$-nm-thick layer below the TI surfaces due to penetration of evanescent wave functions from the metallic surfaces into the bulk of TI. Averaging $\mathbf{S}\left(\mathbf{r}\right)$ over a few angstroms along the longitudinal direction defined by the current flow reveals a large component pointing in the transverse direction. In addition, we find an order of magnitude smaller out-of-plane component when the direction of injected current with respect to Bi and Se atoms probes the largest hexagonal warping of the Dirac-cone dispersion on the TI surface. Our analysis is based on an extension of the nonequilibrium Green's functions combined with density functional theory (NEGF+DFT) to situations involving noncollinear spins and spin-orbit coupling. We also demonstrate how DFT calculations with a properly optimized local orbital basis set can precisely match putatively more accurate calculations with a plane-wave basis set for the supercell of ${\mathrm{Bi}}_2{\mathrm{Se}}_3$.

Helicons are transverse electromagnetic waves propagating in three-dimensional (3D) electron systems subject to a static magnetic field. We present a theory of helicons propagating through a 3D Weyl semimetal. Our approach relies on the evaluation of the optical conductivity tensor from semiclassical Boltzmann transport theory, with the inclusion of certain Berry curvature corrections that have been neglected in the earlier literature (such as the one due to the orbital magnetic moment). We demonstrate that the axion term characterizing the electromagnetic response of Weyl semimetals dramatically alters the helicon dispersion with respect to that in nontopological metals. We also discuss axion-related anomalies that appear in the plasmon dispersion relation.

Currently there is a lively discussion concerning Fermi velocity renormalization in twisted bilayers and several contradicted experimental results are reported. Here we study electronic structures of the twisted bilayers by scanning tunneling microscopy (STM) and spectroscopy (STS). The interlayer coupling strengths between the adjacent bilayers are measured according to energy separations of two pronounced low-energy van Hove singularities (VHSs) in the STS spectra. We demonstrate that there is a large range of values for the interlayer interaction not only in different twisted bilayers, but also in twisted bilayers with the same rotation angle. Below the VHSs, the observed Landau quantization in the twisted bilayers is identical to that of massless Dirac fermions in graphene monolayer, which allows us to measure the Fermi velocity directly. Our result indicates that the Fermi velocity of the twisted bilayers depends remarkably on both the twisted angles and the interlayer coupling strengths. This removes the discrepancy about the Fermi velocity renormalization in the twisted bilayers and provides a consistent interpretation of all current data.

We investigate the properties of electronic states and optical transitions in hexagonal GaAs quantum dots within ${\text{Al}}_{0.3}{\text{Ga}}_{0.7}\text{As}$ nanowires, grown in axial direction [111]. Such dots are particularly interesting due to their high degree of symmetry. A streamlined postsymmetrization technique based on class operators (PTCO) is developed which enables one to benefit from the insight brought by the maximal symmetrization and reduction of fields (MSRF) approach reported by Dalessi et al. [Phys. Rev. B 81, 125106 (2010)], after having solved the Schrödinger equation. Definite advantages of the PTCO are that it does not require modification of existing code for the calculation of the electronic structure, and that it allows to numerically test for elevated symmetries. We show in the frame of a four-band $k·p$ model that despite the fact that the $D_{6h}$ symmetry of the nanostructure is broken at the microscopic level by the underlying zinc-blende crystal structure, the effect is quite small. Most of the particularities of the electronic states and their optical emission can be understood by symmetry elevation to $D_{6h}$ and the presence of approximate azimuthal and radial quantum numbers.

The melting curve of osmium up to a pressure $P$ of 500 GPa is obtained from an extensive suite of ab initio quantum molecular dynamics (QMD) simulations using the Z method. The ab initio $P=0$ melting point of Os is $3370\pm 75$ K; this range encompasses all of the available data in the literature and corroborates the conclusion of J. W. Arblaster [Platinum Metals Rev. 49, 166 (2005)] that the melting temperature of pure Os is $3400\pm 50$ K and that the 3300 K typically quoted in the literature is the melting point of impure Os. The $T=0$ equation of state (EOS) of Os and the $P$ dependence of the optimized $c/a$ ratio for the hexagonal unit cell, both to pressures $\sim 900$ GPa, are obtained in the ab initio approach as validation of its use. Although excellent agreement with the available experimental data ($P\lesssim 80$ GPa) is found, it is the third-order Birch-Murnaghan EOS with $B_0^{\prime }=5$ rather than the more widely accepted $B_0^{\prime }=4$ that describes the QMD data to higher pressures, in agreement with the more recent experimental EOS by Godwal et al. The theoretical melting curve of Os obtained earlier by Joshi et al. is shown to be inconsistent with our QMD results, and the possible reason for this discrepancy is suggested. Regularities in the melting curves of Os and five other third-row transition metals (Ta, W, Re, Pt, Au) could be used to estimate the currently unknown melting curves of Hf and Ir.

Recent progress of high-pressure technology enables the synthesis of novel metal borides with diverse compositions and interesting properties. A precise characterization of these borides, however, is sometimes hindered by multiphase intergrowth and grain-size limitation in the synthesis process. Here, we theoretically explored new transition-metal borides $(M=\mathrm{Co}$, Rh, and Ir) using a global structure searching method and discovered a series of stable compounds in this family. The predicted phases display a rich variety of stoichiometries and distinct boron networks resulting from the electron-deficient environments. Significantly, we identified a new $\mathrm{Ir}{\mathrm{B}}_{1.25}$ structure as the long-sought structure of the first synthesized Ir-B compound. The simulated x-ray diffraction pattern of the proposed $\mathrm{Ir}{\mathrm{B}}_{1.25}$ structure matches well with the experiment, and the convex hull calculation establishes its thermodynamic stability. Results of the present paper should advance the understanding of transition-metal borides and stimulate experimental explorations of these new and promising materials.

We study the nature of spin excitations of individual transition metal atoms (Ti, V, Cr, Mn, Fe, Co, and Ni) deposited on a ${\mathrm{Cu}}_2\mathrm{N}/\mathrm{Cu}\left(100\right)$ surface using both spin-polarized density functional theory (DFT) and exact diagonalization of an Anderson model derived from DFT. We use DFT to compare the structural, electronic, and magnetic properties of different transition metal adatoms on the surface. We find that the average occupation of the transition metal $d$ shell, main contributor to the magnetic moment, is not quantized, in contrast with the quantized spin in the model Hamiltonians that successfully describe spin excitations in this system. In order to reconcile these two pictures, we build a zero bandwidth multi-orbital Anderson Hamiltonian for the $d$ shell of the transition metal hybridized with the $p$ orbitals of the adjacent nitrogen atoms, by means of maximally localized Wannier function representation of the DFT Hamiltonian. The exact solutions of this model have quantized total spin, without quantized charge at the $d$ shell. We propose that the quantized spin of the models actually belongs to many-body states with two different charge configurations in the $d$ shell, hybridized with the $p$ orbital of the adjacent nitrogen atoms. This scenario implies that the measured spin excitations are not fully localized at the transition metal.

Novel imaging spin-filter techniques, which are based on low-energy electron diffraction, are currently of high scientific interest. To improve the spin-detection efficiency a variety of new materials have been introduced in recent years. A new class of promising spin-filter materials are represented by multiferroic systems, as both magnetic and electric ordering exist in these materials. We have investigated $\text{Fe}\text{/}{\mathrm{BaTiO}}_3$(001) for spin-filter applications, which is a prominent candidate due to its moderate spontaneous polarization of the ${\mathrm{BaTiO}}_3$ substrate. Therefore we calculated diffraction patterns for spin polarized electrons incident on the Fe surface. Motivated by the fact that spin polarized low-energy electron diffraction is a surface sensitive method, we investigated the alterations of exchange and spin orbit scattering induced by switching the ${\mathrm{BaTiO}}_3$ polarization. As shown by our studies the system obviously offers the possibility of realizing a multiferroic spin filter and manipulating the spin-orbit and exchange scattering by an external electric field. The calculations have been done for a large range of kinetic energies and polar angles of the diffracted electrons considering different numbers of Fe monolayers.

Defect-induced ferromagnetism has triggered a lot of investigations and controversies. The major issue is that the induced ferromagnetic signal is so weak that it can be sufficiently accounted for by trace contamination. To resolve this issue, we studied the variation of the magnetic properties of SiC after neutron irradiation with fluence covering four orders of magnitude. A large paramagnetic component has been induced and scales up with defect concentration, which can be well accounted for by uncoupled divacancies. However, the ferromagnetic contribution is still weak and only appears in the low fluence range of neutrons or after annealing treatments. First-principles calculations hint towards a mutually exclusive role of the concentration of defects: A higher defect concentration favors a larger magnetic interaction at the expense of spin polarization. Combining both experimental and first-principles calculation results, the defect-induced ferromagnetism can probably be understood as a local effect which cannot be scaled up with the volume.

Ultrafast magnetization dynamics in metallic heterostructures consists of a combination of local demagnetization in the ferromagnetic constituent and spin-dependent transport contributions within and in between the constituents. Separation of these local and nonlocal contributions is essential to obtain microscopic understanding and for potential applications of the underlying microscopic processes. By comparing the ultrafast changes of the polarization rotation and ellipticity in the magneto-optical Kerr effect we observe a time-dependent magnetization profile $M(z,t)$ in Co/Cu(001) films by exploiting the effective depth sensitivity of the method. By analyzing the spatiotemporal correlation of these profiles we find that on time scales before hot electron thermalization ($<100$ fs) the transient magnetization of Co films is governed by spin-dependent transport effects, while after hot electron thermalization ($>200$ fs) local spin-flip processes dominate.

The origin of the suppression of the longitudinal spin Seebeck effect by applied magnetic fields is studied. We perform numerical simulations of the stochastic Landau-Lifshitz-Gilbert equation of motion for an atomistic spin model and calculate the magnon accumulation in linear temperature gradients for different strengths of applied magnetic fields and different length scales of the temperature gradient. We observe a decrease of the magnon accumulation with increasing magnetic field and we reveal that the origin of this effect is a field dependent change of the frequency distribution of the propagating magnons. With increasing field the magnonic spin currents are reduced due to a suppression of parts of the frequency spectrum. By comparison with measurements of the magnetic field dependent longitudinal spin Seebeck effect in YIG thin films with various thicknesses, we find qualitative agreement between our model and the experimental data, demonstrating the importance of this effect for experimental systems.

Elementary excitations of the $S=\frac{1}{2}$ one-dimensional antiferromagnet ${\mathrm{KCuGaF}}_6$ were investigated by inelastic neutron scattering in zero and finite magnetic fields perpendicular to the $(1,1,0)$ plane combined with specific heat measurements. ${\mathrm{KCuGaF}}_6$ exhibits no long-range magnetic ordering down to 50 mK despite the large exchange interaction $J/k_{\mathrm{B}}=103$ K. At zero magnetic field, well-defined spinon excitations were observed. The energy of the des Cloizeaux and Pearson [J. des Cloizeaux and J. J. Pearson, Phys. Rev. 128, 2131 (1962)] mode of the spinon excitations is somewhat larger than that calculated with the above exchange constant. This discrepancy is mostly ascribed to the effective $XY$ anisotropy arising from the large Dzyaloshinsky-Moriya [T. Moriya, Phys. Rev. 120, 91 (1960)] interaction with an alternating $\mathit{D}$ vector. ${\mathrm{KCuGaF}}_6$ in a magnetic field is represented by the quantum sine-Gordon model, for which low-energy elementary excitations are composed of solitons and antisolitons and their bound states called breathers. Unlike the theoretical prediction, it was found that the energy of a soliton is smaller than that of the first breather, although the energy of the first breather coincides with that observed in a previous electron spin resonance measurement.

Recent numerical calculation of the intrinsic thermal Hall conductivity of nodal $d$-wave superconductors in the mixed state revealed a rapid increase of this quantity above an onset temperature. Interestingly, this defines a measurable energy scale in an otherwise gapless state. Using the mathematics of magnetic coherent states, in this paper such energy scale is related to a dynamical process associated with the Andreev scattering of an electron wave packet moving along the constant energy contours in the momentum space. This energy scale is then used to obtain an improved scaling collapse of numerically calculated thermal Hall conductivity in a tight-binding model as a function of temperature, magnetic field, and the $d$-wave pairing amplitude at various band fillings. The results indicate that the mentioned onset temperature is associated with the ability of the quasiparticle wave packet to complete its semiclassical orbit before it is appreciably scattered by the superconducting condensate.

The upper critical field $H_{c2}$ in superconducting ${\mathrm{BaFe}}_{2-x}{\mathrm{Ni}}_x{\mathrm{As}}_2$ single crystals has been determined by magnetotransport measurements down to 0.6 K over the whole superconducting dome with $0.065\le x\le 0.22$ for both the interplane $(H\parallel c,H_{c2}^c)$ and in-plane $(H\parallel ab,H_{c2}^{ab})$ field directions in static magnetic fields up to 16 T and pulsed magnetic fields up to 60 T. The temperature dependence of $H_{c2}^{ab}$ follows the Werthamer-Helfand-Hohenberg model incorporating orbital and spin paramagnetic effects, while $H_{c2}^c\left(T\right)$ can only be described by the effective two-band model with unbalanced diffusivity. The anisotropy of the upper critical fields, $\gamma \left(T\right)=H_{c2}^{ab}/H_{c2}^c$, monotonically increases with increasing temperature for all dopings, and its zero-temperature limit $\gamma \left(0\right)$ has an asymmetric doping dependence with a significant enhancement in the overdoped regime, where the optimally doped compound has the most isotropic superconductivity. Our results suggest that the anisotropy in the superconductivity of iron pnictides is determined by the topology of the Fermi surfaces together with the doping-induced impurity scattering.

The properties of the interface between solid and melt are key to solidification and melting, as the interfacial free energy introduces a kinetic barrier to phase transitions. This makes solidification happen below the melting temperature, in out-of-equilibrium conditions at which the interfacial free energy is ill defined. Here we draw a connection between the atomistic description of a diffuse solid-liquid interface and its thermodynamic characterization. This framework resolves the ambiguities in defining the solid-liquid interfacial free energy above and below the melting temperature. In addition, we introduce a simulation protocol that allows solid-liquid interfaces to be reversibly created and destroyed at conditions relevant for experiments. We directly evaluate the value of the interfacial free energy away from the melting point for a simple but realistic atomic potential, and find a more complex temperature dependence than the constant positive slope that has been generally assumed based on phenomenological considerations and that has been used to interpret experiments. This methodology could be easily extended to the study of other phase transitions, from condensation to precipitation. Our analysis can help reconcile the textbook picture of classical nucleation theory with the growing body of atomistic studies and mesoscale models of solidification.

The absence of energy dissipation leads to an intriguing out-of-equilibrium dynamics for ultracold polar gases in optical lattices, characterized by the formation of dynamically bound on-site and inter-site clusters of two or more particles, and by an effective blockade repulsion. These effects combined with the controlled preparation of initial states available in cold-gas experiments can be employed to create interesting out-of-equilibrium states. These include quasiequilibrated effectively repulsive 1D gases for attractive dipolar interactions and dynamically bound crystals. Furthermore, nonequilibrium polar lattice gases can offer a promising scenario for the study of quasi-many-body localization in the absence of quenched disorder. This fascinating out-of-equilibrium dynamics for ultracold polar gases in optical lattices may be accessible in on-going experiments.

The fanned out, noncollinear spin structure of the Fe(Co) sublattice in amorphous NdFeCo and PrFeCo alloys is shown to strongly affect its ultrafast laser-induced magnetization dynamics. An overshooting effect is discovered at low applied magnetic fields, where the magnetization temporarily increases above its equilibrium value. We explain this phenomenon by considering the dynamics of the noncollinear spin structure. After femtosecond laser excitation the system first reconstructs magnetic order on a time scale of 50–150 ps arriving at a state with a smaller opening angle of the fan. Subsequently, the original opening angle is restored on a time scale of nanoseconds. Increasing the field up to 0.6 T we can fully close the fan and therewith suppress the overshooting behavior.

We measured the low-temperature thermal conductivity of a layered superconductor with quasi-one-dimensional characteristics, the ternary telluride ${\mathrm{Ta}}_4{\mathrm{Pd}}_3{\mathrm{Te}}_{16}$ with a transition temperature $T_c\approx 4.3$ K. The significant residual linear term of thermal conductivity in zero magnetic field and its rapid field dependence provide evidence for nodes in the superconducting gap. By measuring resistivity under pressure, we reveal a superconducting dome in the temperature-pressure phase diagram. The existence of gap nodes and a superconducting dome suggest unconventional superconductivity in ${\mathrm{Ta}}_4{\mathrm{Pd}}_3{\mathrm{Te}}_{16}$, which may relate to a charge-density-wave instability in this low-dimensional compound.

We consider two types of magnetic Josephson junctions (JJ). They are formed by two singlet superconductors S and magnetic layers between them so that the JJ is a heterostructure of the ${\mathrm{S}}_{\text{m}}/\mathrm{n}/{\mathrm{S}}_{\text{m}}$ type, where ${\mathrm{S}}_{\text{m}}$ includes two magnetic layers with noncollinear magnetization vectors. One layer is represented by a weak ferromagnet and another one—the spin filter—is either conducting strong ferromagnet (nematic or $N$-type JJ) or magnetic tunnel barrier with spin-dependent transparency (magnetic or $M$-type JJ). Due to spin filtering only a fully polarized triplet component penetrates the normal n wire and provides the Josephson coupling between the superconductors S. Although both filters let to pass triplet Cooper pairs with total spin $\mathbf{S}$ parallel to the filter axes, the behavior of nematic and magnetic JJs is completely different. Whereas in the nematic case the charge and spin currents, $I_{\text{Q}}$ and $I_{\text{sp}}$, do not depend on mutual orientation of the filter axes, both currents vanish in magnetic JJ in the case of antiparallel filter axes, and change sign with reversal of the filter direction. The obtained expressions for $I_{\text{Q}}$ and $I_{\text{sp}}$ clearly show a duality between the superconducting phase $\phi$ and the angle $\alpha$ between the exchange fields in the weak magnetic layers.

The isothermal compressibility of a general crystal is analyzed within classical density functional theory. Our approach can be used for homogeneous and unstrained crystals containing an arbitrarily high density of local defects. We start by coarse-graining the microscopic particle density and then obtain the long-wavelength limits of the correlation functions of elasticity theory and the thermodynamic derivatives. We explicitly show that the long-wavelength limit of the microscopic density correlation function differs from the isothermal compressibility. We apply our theory to crystals consisting of soft particles which can multiply occupy lattice sites (“cluster crystals”). The multiple occupancy results in a strong local disorder over an extended range of temperatures. We determine the cluster crystals' isothermal compressibility, the fluctuations of the lattice occupation numbers and their correlation functions, and the dispersion relations. We also discuss their low-temperature phase diagram.