As the digitization of the manufacturing process is accelerating, various data-driven approaches using machine learning are being developed in chemical mechanical polishing (CMP). For a more accurate prediction in contact-based CMP, it is necessary to consider the real-time changing pad surface roughness during polishing. Changes in pad surface roughness result in non-uniformity of the real contact pressure and friction applied to the wafer, which are the main causes of material removal rate variation. In this paper, we predicted the material removal rate based on pressure and surface roughness using a deep neural network (DNN). Reduced peak height (Rpk) and real contact area (RCA) were chosen as the key parameters indicative of the surface roughness of the pad, and 220 data were collected along with the process pressure. The collected data were normalized and separated in a 3 : 1 : 1 ratio to improve the predictive performance of the DNN model. The hyperparameters of the DNN model were optimized through random search techniques and 5 cross-validations. The optimized DNN model predicted the material removal rate with high accuracy in ex-situ CMP. This study is expected to be utilized in data-driven machine learning decision making for cyber-physical CMP systems in the future.

Finite element analysis model was fabricated to confirm stress concentration phenomenon occurring in the wafer edge region in the CMP process, and it was confirmed if it corresponds to the measurement result of the actual pressure sensor. First, contact stress distribution at the edge of the wafer was calculated by the finite element analysis model in which material properties and boundary conditions were set up. As a result, an engineering contact stress distribution profile was obtained. Next, the pressure generated in the edge region of the wafer was measured using a pressure sensor that detects resistance change of the polymer. To compare with the result of the finite element analysis, the non-dimensional sensor signal unit was converted into the pressure unit, and correlation between the analysis and measurement results was obtained. As a result, the finite element analysis result, the actual pressure measurement, and the trend of the results were more than 90%. The results show that the finite element analysis model produced and modified in this study is consistent with the actual behavior trend of the components.

The chemical mechanical planarization (CMP) process combines the chemical effect of slurry with the mechanical effect of abrasive (slurry)-wafer-pads The slurry delivery system has a notable effect on polishing results, because the slurry distribution is changed by the supply method. Thus, the investigation of slurry pumps and nozzles with regard to the slurry delivery system becomes important. This paper investigated the effect of a centrifugal slurry pump on a spray nozzle system in terms of uniform slurry supply under a rotating copper (Cu) wafer, based on experimental results and computational fluid dynamics (CFD). In conventional tools, the slurry is unevenly and discontinuously supplied to the pad, due to a pulsed flow caused by the peristaltic pump and distributed in a narrow area by the tube nozzle. Adopting the proposed slurry delivery system provides a higher uniformity and lowered shear stress than usual methods. Therefore, the newly developed slurry delivery system can improve the CMP performance.