SPIN

In this work, we studied doping of pure hafnium disulfide (HfS²) by nonmagnetic atoms C and N. This doping was carried out by the substitution of a host atom S by doping atom C or N. The purpose of this work is to search new half-metallic material for the field of spintronics. This theoretical study is based on density functional theory (DFT) and GGA-PBE approximation. Our calculation revealed that the HfS² binary material doping with carbon and nitrogen induces strong magnetism and predicts new doped materials HfS²-C and HfS²-N. Their magnetic moments are respectively 2${\mu }_B$ and 1${\mu }_B$. The host atoms Hf and the doping atoms induce a strong magnetic moment, while the host atoms S introduce a weak magnetic moment. This magnetism results fundamentally from the coupling between the Hf-5d states and the X-2p states (X = C and N). These two new materials will be promising for the spintronic domain.

With the substantial increase in the amount of data, the mismatch between the processing speed of the hardware and the software results in the ‘Memory Wall’ problem. Processing-in-memory (PIM), in which the compute and memory units are integrated, can avoid frequent data transmission. Binary neural network (BNN) uses binary weights and activations instead of full-precision weights and activations in the convolutional neural network, which reduces computational complexity with minor influence on accuracy. In this paper, we used a one-step operation to write a pair of SOT-MRAM cells and verified the two basic operations in BNN: XNOR and bitcount. Then, we employed an external control circuit with FPGA and accomplished ‘I’ single-character recognition based on vector-matrix multiplication in the SOT-MRAM array.

This work is focused on determining the electrical resistance, which is induced by single domain wall (DW) in magnetic nanowire (NW) with a negligible defect. The provided model covers a wide range of NW diameters. The obtained result demonstrates a few-order rapid reduction of the DW resistance accompanied by its possible deviations versus the diameter growth ranging from 1.2nm to 15.2nm. The origin of these deviations, which are also identified as oscillations, is referred to the nonuniform electron scattering on the DW due to the intermixing electron scattering conditions: ballistic for one spin channel and quasi-ballistic for other one with opposite spin direction. It may happen when the DW width by value is approximately in between two lengths: a mean free path (MFP) with the spin down and spin up. The indirect evidence of this finding also comes from the fact that homogeneous NWs show the most valuable DW resistance oscillations by magnitude rather than segmented magnetic NWs. In addition to the approach, where DW width is constant, the other reasonable model is used when the DW can be constrained for some conditions. The same results are valid for magnetic junctions with DW. Finally, resistance simulation in the diffusive range, when a diameter of the NW (or point-like junction) is larger than any of spin-resolved MFP of electrons, successfully follows experimental data for the single and double DW resistances available in literature.

Due to the exponential increase of transistor leakage currents, especially in the technologies lower than 90nm, the static power consumption has dramatically increased. So, it is regarded as one of the main problems of the CMOS circuits. The spintronic devices, such as magnetic tunnel junction (MTJ), have notable advantages including low-static power consumption, nonvolatility, high endurance, compatibility with the CMOS transistors and the possibility of fabrication in high scales. Hence, the hybrid MTJ/CMOS circuits are considerable options to mitigate the problem of high-static power consumption in CMOS circuits. In this paper, a nonvolatile and low-power hybrid MTJ/CMOS full adder is proposed for implementation of in-memory computing. The drawback of existing full adders is their low-speed and high-power consuming for MTJ switching, which terribly affect the performance of multi-bit adders using these full adders. In this work, a new full adder is designed by eliminating the MTJ and transistor related to the input carry bit and modifying the tree structure of the MTJ network associated with the other inputs. In the proposed full adder, the effect of carry propagation has drastically decreased and its power consumption is also improved.

We have employed the Oguchi approximation (OA) to investigate the effects of a random crystal field which is active (inactive) with probability $p$ $(1-p)$ for given lattice sites on the thermodynamic properties of the mixed spin-$(\frac{5}{2},\frac{1}{2})$ anisotropic Heisenberg model in the presence of an external magnetic field. By exploiting the thermal variations of the order parameters, interesting phase diagrams are produced on various planes. The phase diagrams show first- and second-order transition lines and also tricritical points. In addition to the reentrant behavior which is localized in a certain range of $p$ at very low temperature, the magnetic exponent $\beta$ calculated near the critical point exhibits interesting behaviors depending on the crystal field interaction. The hysteresis behavior of the model is also investigated and it reveals that the width of hysteresis loop decreases or increases, respectively, for negative or positive crystal field region as the probability $p$ increases.

A numerical calculation of the temperature dependence of the chemical potential and for various values of the filling factor $\nu$ of the Landau level is carried out. The degree of filling of each Landau level is found as a function of temperature. Analytical calculations were carried out in the model of an ideal Fermi gas both without taking into account the broadening of the Landau levels $\mathrm{\Gamma }=0$ and taking into account their broadening $\mathrm{\Gamma }>0$. An analytical formula is found that describes the dependences of $\mu (T,\nu )$ at $\mathrm{\Gamma }=0$ and low temperatures. A low-temperature formula for the dependence of $\mu (T∕\mathrm{\Gamma },{\epsilon }_F)$ based on the Sommerfeld expansion is also given. The influence of the intra- and interlevel thermal excitations on changes in the chemical potential with temperature is discussed.

Following our previous work on the magnetocaloric compounds FeMnAs_xP${}_{1-x}$, we obtain thermomagnetic properties for the barocaloric antiperovskite Mn₃GaN, which has been investigated in recent years in view of its thermal response under pressure cycling. Its structure displays canting of spins and a volume dilation in the antiferromagnetic (AFM) transition, related to magnetoelastic coupling. A random-phase-approximation (RPA) Green function theoretical treatment is applied to a two-exchange parameter Heisenberg model, and the canting of the spins orientation in the (111) planes of the structure is considered in the model. The observed first-order AFM phase transitions are reproduced by the introduction of a biquadratic spin term in the Hamiltonian, coupled to the volume dilation in the transition. The model is able to predict a greater stability for the canted spin structure as compared with conventional AFM, if the magnitudes of Mn magnetic moments in these structures are taken from the literature. The calculations of quantities of interest for barocaloric applications reproduce quite well the available data on temperature changes and latent heat under pressure cycling, at adiabatic and isothermal conditions, respectively.

We report layered TiCl₂ monolayer to be a robust antiferromagnetic (AFM) semiconductor by investigating its geometry, stability, electronic and magnetic properties using first-principles calculation and Monte Carlo (MC) simulation. The TiCl₂ monolayer exfoliated from layered TiCl₂ bulk is verified to be an AFM semiconductor with a relatively wide band gap by spin-polarized density of states and energy band structure calculations. Phonon dispersion and elastic modulus calculations show that the TiCl₂ monolayer is dynamically and mechanically stable, respectively. The Néel temperature is evaluated to be 368K by MC simulation based on Ising model. Additionally, the band gap, the magnetic moment of a single Ti atom and the Néel temperature can be significantly tuned by applied biaxial strain, but the AFM semiconducting behavior is robust against the applied strains, indicating that the TiCl₂ monolayer is a potential candidate for high-speed AFM spintronic applications.

With transistors reaching nanometer dimensions, dissipated energy has become of great importance in recent years. A practical approach to reducing energy consumption is to use logic-in-memory (LIM) structures based on magnetic tunnel junction (MTJ) devices combined with approximate computing. In this paper, we propose energy-efficient MTJ/FinFET-based approximate 5:2 compressors for error-resilient in-memory computing, providing an accuracy close to the exact design (1.54% error rate) while reducing the energy consumption by more than 50%. The innovative Boolean equations and the structure of the proposed approximate circuits based on spin-Hall effect assisted MTJs lead to a significantly more effective compromise between energy and accuracy than the previous exact and approximate counterparts. The simulation results provided using HSPICE with 7nm FinFETs and SHE-assisted MTJ models demonstrate the superior hardware parameters of the proposed designs. Furthermore, the MATLAB simulations show an average peak signal-to-noise ratio (PSNR) of more than 43 and an average structural similarity index metric (SSIM) of more than 0.99 across image multiplication, sharpening, and smoothing operations.

In this paper, we have developed models based on Deep-Learning Neural Network (DNN) for accurately predicting lattice constants of half-Heusler alloys. A commercial software WIEN2k employing the Density Functional Theory (DFT) is first used to generate data of the lattice constants of 377 half-Heusler alloys for the training, testing and validation of the models. These models use elemental symbols or/and ionic radii as input parameters. The model that uses only symbols of the constituent element and the model that uses symbols in combination with the radii of the ions predict lattice constants of half-Heusler alloys with an average error of less than 1% to the data obtained from the WIEN2k calculations. The average error stays below 2% in the prediction by the model that uses radii of the ions alone. These results show a great promise for these models to be extended for the prediction of structural and elastic properties of not only new half-Heusler alloys but also other new materials such as full-Heusler, spinels, perovskites, etc. with greater convenience, saving hours of computation time.