-based Heusler compounds are promising materials for spintronics applications due to their high Curie temperature, large spin polarization, large magnetization density, and exotic transport properties. In the present paper, we report the anomalous Hall effect (AHE) in a polycrystalline Heusler compound using combined experimental and theoretical studies. The Rietveld analysis of high-resolution synchrotron x-ray diffraction data reveals a large degree of antisite disorder between Fe and Al atoms. The analysis of anomalous transport data provides the experimental anomalous Hall conductivity (AHC) about 227 S/cm at 2 K with an intrinsic contribution of 155 S/cm, which has nearly constant variation with temperature. The detailed scaling analysis of anomalous Hall resistivity suggests that the AHE in is governed by the Berry phase driven intrinsic mechanism. Our theoretical calculations reveal that the disorder present in the compound enhances the Berry curvature induced intrinsic AHC.
We present a deep-learning framework, CrysXPP, to allow rapid and accurate prediction of electronic, magnetic, and elastic properties of a wide range of materials. CrysXPP lowers the need for large property tagged datasets by intelligently designing an autoencoder, CrysAE. The important structural and chemical properties captured by CrysAE from a large amount of available crystal graphs data helped in achieving low prediction errors. Moreover, we design a feature selector that helps to interpret the model’s prediction. Most notably, when given a small amount of experimental data, CrysXPP is consistently able to outperform conventional DFT. A detailed ablation study establishes the importance of different design steps. We release the large pre-trained model CrysAE. We believe by fine-tuning the model with a small amount of property-tagged data, researchers can achieve superior performance on various applications with a restricted data source.
16 June 2023 in The Journal of Physical Chemistry C
Accurate band gap prediction in semiconductors is crucial for materials science and semiconductor technology advancements. This paper extends the Perdew–Burke–Ernzerhof (PBE) functional for a wide range of semiconductors, tackling the exchange and correlation enhancement factor complexities within density functional theory (DFT). Our customized functionals offer a clearer and more realistic alternative to DFT+U methods, which demand large negative U values for elements like sulfur (S), selenium (Se), and phosphorus (P). Moreover, these functionals are more cost-effective than GW or Heyd–Scuseria–Ernzerhof (HSE) hybrid functional methods, therefore, significantly facilitating the way for unified workflows in analyzing electronic structure, dielectric constants, effective masses, and further transport and elastic properties, allowing for seamless calculations across various properties. We point out that such development could be helpful in the creation of comprehensive databases of band gap and dielectric properties of the materials without expensive calculations. Furthermore, for the semiconductors studied, we show that these customized functionals and the strongly constrained and appropriately normed semilocal density functional (SCAN) perform similarly in terms of the band gap.
23 June 2023 in The Journal of Physical Chemistry C