Abstract: A method and computer program product is described for optimizing the design of a circuit layout that assigns binary properties to the design elements according to a hierarchy of rules. For example, the design of an alternating phase shifted mask (altPSM) is optimized first according to rules that assign phase shapes that maximize image quality for critical circuit elements, and then further optimized to minimize mask manufacturability problems without significantly increasing the complexity of the design process flow. Further optimization of the design according to additional rules can be performed in a sequentially decreasing priority order. As the priority of rules decrease, some violation of lower priority rules may be acceptable, as long as higher priority rules are not violated.

Abstract: A method and computer program product is described for optimizing the design of a circuit layout that assigns binary properties to the design elements according to a hierarchy of rules. For example, the design of an alternating phase shifted mask (altPSM) is optimized first according to rules that assign phase shapes that maximize image quality for critical circuit elements, and then further optimized to minimize mask manufacturability problems without significantly increasing the complexity of the design process flow. Further optimization of the design according to additional rules can be performed in a sequentially decreasing priority order. As the priority of rules decrease, some violation of lower priority rules may be acceptable, as long as higher priority rules are not violated.

Abstract: A method is described for designing an alternating phase shifted mask (altPSM) by optimally selecting the width of phase shapes. The selection of optimal phase shape widths is achieved by providing a lithography metric that describes the relationship between phase shape width and the target image dimension such that the metric, such as process window or across chip linewidth variation (ACLV), is optimized. In a preferred embodiment, ACLV is computed by Monte Carlo simulation by providing a set of error distributions for lithographic parameters such as focus, dose, lens aberrations, and the like. Alternatively, a lookup table of optimal phase widths associated with target image dimensions may be provided. The resulting altPSM is characterized by phase shapes having widths that vary according to the widths of the target image dimensions.