M.Sc Thesis


M.Sc StudentBarak Yaakobovitz
SubjectDevelopment of Detection Methods for Line Edge Roughness
DepartmentDepartment of Chemical Engineering
Supervisor Professor Tsur Yoed
Full Thesis textFull thesis text - English Version


Abstract

The semiconductor fabrication process is a combination of many processes: photolithography, ion implantation, metal deposition and etching. Line Edge Roughness (LER) is one of the problems that reported to affect the final fab products and most of the time it occurs in the photolithography process. The industry requirements call for an in-line nondestructive and very fast measuring tool. In this work a possible in-line integrated metrology solution for alarming when LER has grown out of a specified range was developed. The solution is based on optical scatterometry combined with new algorithms. Once the integrated metrology tool starts alarming, the operators can initiate a more accurate and time consuming procedure at offline stand alone tools to identify and quantify the problem. In order to develop the detection method, Wafers with structures of lines that intentionally contain roughness at different sizes have been prepared by e-beam process. The structures were characterized both by direct imaging tools (SEM and AFM with high aspect ratio tip) and by a scatterometric method using NovaScan 3090. It is shown that for specific structures and by using the best model, scatterometry can alarm when the nominal roughness is between 7.5-10 nm. Further development of the method may lead to even smaller detectable range. Hence it is recommended to use this tool for process alarming by preparing few simple test structures that will be combined in scribe lines (in unused areas within the mask) or in test chips.

The photolithography process step is important since the pattern of the structure is transferred in this step. The available tools today for detection of Line Edge Roughness are CD- SEM, and AFM. These tools are very reliable but are not fast. In this work we propose a new solution to the problem: detection based on an optical scatterometry combined with new algorithms. This method could be used as an in-line nondestructive method and it also has the advantage of a very fast measuring and response.