An X-based spatial oxide growth visualization software for furnace characterization

Abstract

used for developing the temperature contours inside the furnace. The resulting oxide thickness profiles generated can be viewed either as a color coded graph or portrayed in 3 dimensions. Several conveniences like a user friendly data entry module and PostScript conversion for hard copies are provided. The software is expected to be useful to process engineers in understanding their furnace better and thus reduce the time taken to stabilize the process line. Demand for high quality control and stringent manufacturing standards continue to escalate in the VLSI industry. The term "manufacturing performance" and C4 manufacturing capability" are becoming by words of quality control in the VLSI process industry. Equipment characterization has emerged as a competent way of increasing the throughput in the VLSI factory with minimal time overhead in trial runs and maturity delays. The detailed analysis required for such techniques has paved way for computer software tools. Computer interpretations through simulation and modeling can fulfill a critical need in managing the voluminous data and provide timely feedback for manufacturing quality control. A perfect example to illustrate this point is the cylindrical furnace referred to as the "diffusion" furnace. The furnace is one the most important and complex process equipment in the process line and efforts towards characterizing it will be well worth the effort. This thesis will outline a software to characterize a furnace based on its single most important utility-oxide growth on the silicon wafer. The software developed is completely generic and has no affiliation to any make of furnace. The data required for simulation can easily be obtained from the actual piece of equipment. Simulation capabilities include almost all possible environments for oxidation such as wet, dry, with/without HCI, various levels of doping etc. Interpolation techniques with acceptable accuracy are used for developing the temperature contours inside the furnace. The resulting oxide thickness profiles generated can be viewed either as a color coded graph or portrayed in 3 dimensions. Several conveniences like a user friendly data entry module and PostScript conversion for hard copies are provided. The software is expected to be useful to process engineers in understanding their furnace better and thus reduce the time taken to stabilize the process line.