Numerical Simulation of an Entire Wafer Surface during Ozone-Based Wet Chemical Etching

In microelectromechanical system manufacturing and especially in the photovoltaic industry, wet-chemical baths are used for surface structuring, conditioning, and cleaning. Ozone-based wet-chemical cleaning processes show, in addition to the cleaning of the silicon wafers, the ancillary effect of intended material etch back. Previous studies observed inhomogeneities over the wafer surface occurring during the ozone-based wet chemical process. Since a detailed observation during the process in the basin is not possible, simulations are carried out to understand what causes these inhomogeneities. Therefore, an existing microscopic two-dimensional (2D) model for reaction modelling is extended stepwise from a length of several microns to the full length of 157 mm of an industrial solar cell. Subsequently, the 2D model is transferred to a three-dimensional (3D) model. In addition, an initial water layer, originating from the previous rinsing step, and the movement of the handling system were taken into account for modeling. This approach allows the estimation of the etching process over an entire wafer surface for the first time and enables better understanding of the reasons for the inhomogeneities. The water layer in connection with the fast movement of the handling system and the masking, originating from carrier rods, leads to stripes on the wafer surface, shown by differences in reflection. As the water layer is not removed homogeneously, the occurring chemical reaction in the masked areas is delayed and results in an uneven etching of the wafer surface. The removal over the entire wafer surface in the simulation (8.51 ± 0.37 nm) is in accordance with the experimental data (9.02 ± 1.10 nm). This presented model can serve as a base for future work to estimate the effects of parameter changes, e.g., plant design or the composition of the etching solution on the homogeneity of the wet chemical cleaning and etch back process, thus enabling a cost efficient process optimization.