First-principles and deep learning frameworks to predict the electronic, magnetic and electrical properties of V-doped SiC nanotube

Nowadays, with the miniaturization of nanoindustry, we are increasingly interested in doping effects for nanotube systems. In this study, based on density functional theory (DFT) and local spin density approximation (LDA) methods within Hubbard U corrections, the electronic, magnetic and electrical properties of single-wall silicon carbide nanotubes doped with vanadium were theoretically studied. Although the undoped SiC system is nonmagnetic, the V-doped SiC nanotube induces magnetism and the total magnetic moment of this magnetic material is equal to ~1 µB. Density of states calculations indicated that the magnetization of SiC:V single-wall nanotubes mainly comes from the 2p orbitals of carbon atoms and 3d orbitals of the V dopant. Total energy calculations of the ferromagnetic and antiferromagnetic phases of V-doped SiCNT systems revealed that the ferromagnetic phase was more stable.In this study, based on density functional theory (DFT) and local spin density approximation (LDA) methods within Hubbard U corrections, the electronic and magnetic properties of single wall silicon carbide nanotubes doped with vanadium were theoretically studied. These properties were simulated for cases in which single or double silicon atoms of the SiC nanotube were replaced with V atoms. Using deep learning (DL) algorithms is beneficial for predicting quantum-confined electronic structures; however, first-principles simulation methods are more accurate. The ML-based regression model shows the accuracy and prediction model for quantum-confined nanotubes. Among the various neural network algorithms, trilayered and medium neural network algorithms provide more accuracy and a lower error rate for these molecular nanotubes. A comparison between the ML-based approach and the DFT-based procedure reveals the similarity and accuracy of the proposed algorithm. We have used Quantum ATK version 20.0 for the DFT simulation whereas MATLAB is used to predict the best fitted model.