Defect and Diffusion Forum Vol. 430

Abstract: As semiconductor processing technologies continue to advance, semiconductor wafers are becoming more densely packed and intricate, resulting in a higher incidence of surface imperfections. Therefore, it is crucial to detect these defects early and accurately classify them to pinpoint the root causes of the defects in the manufacturing process, ultimately leading to improved yield. Therefore, defect detection is critical in the industrial production of monocrystalline silicon. This study employs deep learning techniques to propose a framework for detecting defects on silicon wafers, focusing on optimizing the hyperparameters of support vector machines (SVM). Three methods were utilized to fine-tune the SVM parameters: Bayesian optimization, grid search, and random search techniques. This study demonstrates how selecting optimal values for SVM parameters can lead to better classification. Additionally, real manufacturing data were utilized to evaluate the performance of the proposed SVM classifier, with a comparison to state-of-the-art techniques in the field. By using deep features from ResNet 101 and a support vector machine, this work achieves 74.5% accuracy in identifying wafer defects without employing any optimization technique. However, the performance of the model was further improved by utilizing the random search optimization technique, which yielded the best result among the three optimization techniques tested, with an accuracy of 88.1%.

Abstract: This paper studies the two-dimensional unsteady incompressible Ag-water and CuO-water nanofluid flow in a semi-porous expanding-contracting channel in the presence of thermal radiation effect. The continuity equation, Navier-Stokes equation, and energy equation governing the model are transformed into a set of non-dimensional ordinary differential equations using appropriate transformations. These dimensionless governing equations are solved using power series with the aid of the Hermite-Padé approximation method. The influences of physical parameters such as Reynolds number, expansion ratio, solid volume fraction, Prandtl number, Magnetic parameter, and shape factor are depicted in velocity and temperature profiles. Moreover, the average Nusselt number and skin friction coefficient are also investigated with the effect of Reynolds number, solid volume fraction, and expansion ratio. It is observed that the heat transfer rate decreases significantly as the shape factor increases.

Abstract: The numerical work presented in this paper focuses on the influence of the magnetic field and the nanoparticles metallic nature on the hydrodynamic and thermal behavior of a nanofluid flowing in an extended curved duct. It deals with a semi-toroidal curved duct with an expanded circular section. The narrowed part of this duct from which the nanofluid enters with a cold temperature, is considered to be thermally insulated. However, the extended part is kept at a constant hot temperature. The nanoparticles used in the present study respectively are Alumina (Al₂O₃), copper oxide (CuO) and iron trioxide (Fe₃O₄). In this study, the effects of inertia, buoyancy and Lorentz forces as well as the metallic nature of the nanoparticles suspended in the pure water have been highlighted on the thermal, hydrodynamic and economic levels. The study is based on the resolution of the classical monophasic equations governing the non-isothermal flow of nanofluids by the use of finite element method, namely: the mass, momentum and energy equations. The obtained results have shown that the buoyancy and inertia forces strongly favor the global heat exchange rate. Moreover, the magnetic force acts negatively on these thermal exchanges. Furthermore, the CuO nanoparticles have demonstrated a better heat transfer rate, approximately 7% higher than that of pure water. Nevertheless, according to the economic needs, we suggest we suggest using alumina nanoparticles, as their transfer rate is comparable to that of CuO nanoparticles. It should be noted, that this study provides important insights for many industrial applications where the curved ducts are strongly presented.

Abstract: The subject of unsteady convective flow with non-linear thermal radiation has become an important issue of research, due to its implications in advanced energy conversion systems operating at high temperature, solar energy technology and chemical process at high operation temperature. Due to the importance of this issue, a time dependent incompressible viscous fluid flow, heat and mass transfer over a curved stretching surface has been numerically analysed by taking into account the heat flux due to concentration gradient and mass flux due to temperature gradient. Together with this the Rosseland approximation is being employed for the nonlinear thermal radiation impact in presence of thermal slip. With the aid of non-dimensional variables and the corresponding physical boundary conditions, the leading nonlinear momentum, energy, and species equations are converted into a set of coupled ordinary differential equations. These equations are then resolved using the MATLAB bvp4c solver. The stability of the numerical technique has been verified and compared with available literatures. The resultant parameters of engineering interest and the boundary layer flow field parameters and have been presented using tables and graphically plots. The study concludes that for lesser curvature parameter (0.5≤K≤0.7) the surface drag force, heat and mass transfer rates can improve by about 9.59%, 2.87% and 1.67% each respectively. The presence of the temperature ratio parameter and the non-linear thermal radiation are found to greatly influence the temperature profile and the heat transfer rate of the system. Results show that the heat transfer rate improves by about 24.39% and 16.66% for varying non-linear thermal radiation (1≤Rd≤1.5) and temperature ratio parameter (1.2≤θ_w≤1.4) respectively. Results obtained also show that improving the thermal slip parameter (0.4≤L≤0.6) can reduce heat transfer rate by about 13.62% and reduce the surface temperature profile.