Piezoelectric biomaterials that imitate human bone’s intrinsic electromechanical capabilities represent a viable approach for tissue engineering scaffolds. Comprehensive structural and functional investigations show that the addition of strontium (Sr) and cerium (Ce) dopants to barium titanate (BaTiO₃) lattice greatly improves its biocompatibility for orthopedic applications. X-ray diffraction (XRD) and Raman spectroscopy confirmed the retention of the tetragonal phase ( P4mm ) in doped samples, with reduced tetragonality (c/a ratio) and stable vibrational modes at ∼250–720 cm⁻¹. Field-emission scanning electron microscopy (FESEM) revealed changes in the average grain size with Sr and Ce doping, while energy-dispersive spectroscopy (EDS) verified homogeneous Sr/Ce distribution. The doped samples exhibit commendable piezoelectric coefficients (d 33 ) ranging from 8–25 pC/N and 18–23 pC/N for Sr- and Ce-doped samples, respectively, which is higher than the d 33 value of human bone. The density, open porosity, and water absorption properties of the pellet scaffolds were examined, and the 3 mol% Sr- and Ce- doped BT pellets exhibited values in the optimal range. The antibacterial studies against Escherichia coli and Staphylococcus aureus suggested that all the pellet scaffolds have antibacterial properties against both gram-negative and gram-positive bacteria, with the Ce-doped sample exhibiting the best performance. In-vitro MTT assay and AO-EB staining showed >90 % MG-63 viability at 7 days for doped samples (vs. 85 % for BaTiO₃), with 3 mol% doping showing the best biocompatibility.
This study examines the free‑convection flow of a micropolar fluid in a vertical channel subjected to an inclined magnetic field while accounting for thermal radiation, internal heat generation, and chemical reactions. The analysis is performed under turbulent flow conditions in order to better represent complex transport mechanisms encountered in practical thermal systems. The nonlinear coupled governing equations are solved using the Differential Transform Method, which provides an efficient semi‑analytical framework for predicting velocity, microrotation, temperature, and concentration distributions. To further enhance the predictive capability of the study, a feed‑forward artificial neural network is developed and trained using the Levenberg–Marquardt algorithm, forming a hybrid modeling strategy that combines physics‑based analysis with data‑driven learning. The parametric investigation reveals that increasing the buoyancy parameter enhances the angular velocity of the fluid, whereas stronger magnetic effects and larger microrotation parameters suppress rotational motion due to increased resistive forces. The temperature field intensifies with stronger heat generation and higher thermal Peclet numbers, indicating enhanced thermal transport, while the concentration level increases with the solutal Peclet number but decreases as the Schmidt number and chemical reaction parameter grow. Increasing micropolar parameter from 1 to 3 decreases skin friction coefficient by 10.38%, whereas increasing microrotation parameter from 1 to 3 increases skin friction coefficient by 3.82%. Increasing solutal Peclet number from 0.1 to 0.5 decreases Sherwood number by 17.62%, whereas increasing schmidt number from 0.1 to 1 increases Sherwood number by 12.20%. The developed neural network demonstrates excellent predictive accuracy, with mean square error values between 1.0E‑8 and 5.2E‑8, performance levels ranging from 1.5E‑8 to 2.4E‑8, gradients between 9.4E‑8 and 9.9E‑8, and Mu values between 1.0E‑9 and 1.0E‑8. These results confirm that the proposed semi‑analytical and machine‑learning framework can effectively capture the complex nonlinear behavior of turbulent magneto‑micropolar flows with coupled heat and mass transfer.
Graphene oxide (GO) offers a promising hole-transport layer (HTL) for lead-free perovskite solar cells because of its favorable electronic properties and interfacial compatibility. However, its role as a HTL in cesium hexaiodostannate(IV) (Cs₂SnI₆)-based solar cells remains insufficiently explored. Here, the FTO/ZnO/Cs₂SnI₆/GO/Au architecture was systematically optimized using SCAPS-1D. The optimized configuration consists of an 800nmCs2SnI6 absorber, 60nm GO HTL, and 30nm ZnO electron-transport layer (ETL). Under ideal low-loss conditions, the optimized device achieved an ideal simulated Voc of 0.850V, Jsc of 27.772mA·cm−2, FF of 71.890%, and PCE of 16.972%. Furthermore, when incorporating realistic experimental constraints, including finite parasitic resistances and moderate interface defect densities, the cell exhibited a practical PCE of 9.878% (Voc=0.740V, Jsc=27.489mA·cm−2, and FF=48.548%). Further analysis of GO acceptor doping, ZnO donor doping, interface defects, and series resistance demonstrated their pronounced influence on photovoltaic performance. In particular, an increase in the defect density at the GO/Cs₂SnI₆ interface substantially increased the recombination losses and degraded the device efficiency. These results demonstrate the potential of GO as an HTL for Cs₂SnI₆ photovoltaics and identify interfacial recombination and resistive losses as critical factors for further device optimization.
