Modeling the Specific Surface Area of Doped Spinel Ferrite Nanomaterials Using Hybrid Intelligent Computational Method

Abstract

Spinel ferrites nanomaterials are magnetic semiconductors with excellent chemical, magnetic, electrical, and optical properties which have rendered the materials useful in many technological driven applications such as solar hydrogen production, data storage, magnetic sensing, converters, inductors, spintronics, and catalysts. The surface area of these nanomaterials contributes significantly to their targeted applications as well as the observed physical and chemical features. Experimental doping has shown a great potential in enhancing and tuning the specific surface area of spinel ferrite nanomaterials while the attributed experimental challenges call for viable theoretical model that can estimate the surface area of doped spinel ferrite nanomaterials with high degree of precision. This work develops stepwise regression (STWR) and hybrid genetic algorithm-based support vector regression (GBSVR) intelligent model for estimating specific surface area of doped spinel ferrite nanomaterials using lattice parameter and the size of nanoparticle as descriptors to the models. The developed hybrid GBSVR model performs better than STWR model with the performance improvement of 7.51% and 22.68%, respectively, using correlation coefficient and root mean square error as performance metrics when validated with experimentally measured specific surface area of doped spinel ferrite nanomaterials. The developed GBSVR model investigates the influence of nickel, yttrium, and lanthanum nanoparticles on the specific surface area of different classes of spinel ferrite nanomaterials, and the obtained results agree excellently well with the measured values. The accuracy and precision characterizing the developed model would be of immense importance in enhancing specific surface area of doped spinel ferrite nanomaterial prediction with circumvention of experimental stress coupled with reduced cost.

Properties and Simulating Research of Epoxy Resin/Micron-SiC/Nano-SiO2 Composite

“…The reduction of the breakdown field of the composite material is related to the overlap of the nano-interface. The nano-fillers have a large specific surface area and the high surface energy can restrict the molecular chains on the surface of the nano-particles to form a regular arrangement in the local area and form a crystallike structure [19], which will make the atoms in the molecular chains close to each other and the atomic potential field affect each other so that the width and height of the interatomic potential barriers are reduced and the original energy level is split. At this time, the forbidden band width of the epoxy resin matrix decreases, and under the action of the applied electric field, the high-energy electrons that accumulate energy in the crystal region can penetrate the potential barrier, which helps carriers migrate in the composite material while retaining enough energy, and finally leads to the reduction of breakdown field strength.…”

Specific Surface Area Characterization of Spinel Ferrite Nanostructure Based Compounds for Photocatalysis and Other Applications Using Extreme Learning Machine Method

“…e ferrite family is classified as garnets, hexagonal, and spinel ferrite using crystal structure as a criterion [25]. Spinel ferrites are distinguished from other ferrite family members by their distinctive features, which have led to their deployment for many technological applications [9,26,27]. Spinel ferrites have the general chemical formula MFe 2 O 4 , with M denoting divalent ions such as Fe 2+ , Co 2+ , Zn 2+ , Mg 2+ , Mn 2+ , and Ni 2+ , among others [28].…”

“…ese unique features of ELM algorithm have widened its application domain in varieties of fields and specialization [44][45][46][47]. e uniqueness of ELM algorithm is harnessed in this contribution to model specific surface area of spinel ferrite based compounds while the existing models in the literature determine specific surface area of the same compounds using stepwise regression (STWR) model and hybrid genetic algorithm based support vector regression (GBSVR) model [27]. e deficiency of STWR based model comes from its inability to account for nonlinearity between the descriptors and the measured specific surface area [27].…”

“…e developed S-ELM model can be easily implemented using Excel or standard calculator while the implementation of the existing genetic algorithm based support vector regression (GBSVR) model [27] is restricted to MATLAB environment. S-ELM model has demonstrated enhanced performance of 4.67% (using correlation coefficient, CC), 5.28% (using mean absolute error, MAE), and 58.22% (using root mean square error, RMSE) over the existing GBSVR model [27] for training set of data samples while it achieves improved performance of 11.64% (using CC metric), 61% (using MAE metric), and 61.31% (using RMSE metric) for testing set of data samples.…”

“…e developed S-ELM model can be easily implemented using Excel or standard calculator while the implementation of the existing genetic algorithm based support vector regression (GBSVR) model [27] is restricted to MATLAB environment. S-ELM model has demonstrated enhanced performance of 4.67% (using correlation coefficient, CC), 5.28% (using mean absolute error, MAE), and 58.22% (using root mean square error, RMSE) over the existing GBSVR model [27] for training set of data samples while it achieves improved performance of 11.64% (using CC metric), 61% (using MAE metric), and 61.31% (using RMSE metric) for testing set of data samples. In the same vein, the developed S-ELM model shows performance improvement of 58.29% (using CC metric), 87.66% (using MAE metric), and 84.12% (using RMSE metric) over the existing stepwise regression (STWR) model [27] on training set of samples while it demonstrates improved performance of 18.28% (using CC metric), 67.05% (using MAE metric), and 70.01% (using RMSE metric) for testing set of samples.…”