Feedforward control for reduced run-to-run variation in microelectronics manufacturing

Increased manufacturing yields can be obtained by reducing process variatio n, One potential method to achieve lower process variance is through interp rocess feedforward control. During feedforward control, a process recipe is adjusted to compensate for measured input deviations. The potential benefi ts of feedforward control include reduced run-to-run variance, rework, and scrap. Feedforward control has been used often in manufacturing, Howe ver, there are two problematic issues associated with feedforward recipe adjustm ent: 1) there is noise in the measurement tool and adjusting for inaccurate measurements could increase the variance and 2) it is difficult to alter o ne parameter in a manufacturing process without worsening other key paramet ers. In this paper, we will address both issues using a systems approach, M easurement noise poses a significant threat to the success of feedforward c ontrol. If the measurement noise is sufficiently large, the variance under feedforward control could exceed:the variance with no control. To address t his concern, we have integrated statistics theory into the feedforward cont roller design. This detunes the recipe adjustment based on the confidence i n the accuracy of the sensor, These algorithms have the effect of filtering the noise from the measurement tool. In order to address the problem of al tering one parameter without adversely affecting others, one can use a feed forward controller that selects a recipe from within a predefined set of al lowable qualified recipes, We call this feedforward recipe selection contro l (FRSC), We have developed a design methodology for this type of controlle r, Preliminary versions of our design algorithms have been implemented into a graphical user interface (GUI)-based computer-aided design (CAD) environ ment, This interactive software package guides the engineer through the des ign of feedforward controllers using process data as inputs.