1995•Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIERequires access

Improved photomask metrology through exposure emulation

James E. Potzick

Open publisher page 1 citations

Abstract

The ultimate purpose of the photomask in IC manufacture is to define the image to be printed on a silicon wafer. Of the many factors which affect this aerial image in the wafer stepper, some are properties of the stepper projection system and some are properties of the photomask. The purpose of photomask metrology is to help optimize photomask properties to produce the desired aerial image in the stepper. Those factors attributable to the photomask include chrome feature size, phase shifter accuracy (if used), chrome edge roughness and runout, defects, etc. Most of these are usually imaged differently in the exposure tool and the metrology tool because these tools' optical systems are different. Two ways to predict the stepper aerial image from photomask measurements are simulation (modeling) and emulation. In the first, the values of the relevant photomask measurements and properties (if they are known) can be inserted into the models for both tools, and then the stepper aerial image can be predicted from the photomask measurements. Or, the stepper's aerial image of the photomask can be emulated in the metrology tool and magnified for the purpose of measurement. Present mask metrology tools, when used in transmission, duplicate the optical morphology of the projection tool. The projected stepper aerial image can be emulated by adjusting the few major generic optical parameters of the metrology tool to match those of the wafer stepper to be used. Then the measured image in the metrology tool emulates the image in the stepper, and photomask properties affect both images in the same ways. Diverse measurements of photomask characteristics can be made accurately and fast by emulating photomask performance instead of measuring the chrome geometry. All of the significant optical performance parameters of the photomask can be measured with a single tool, at the small expense of shifting to a more appropriate philosophy of mask measurement.

About this research paper

What this paper is about

The ultimate purpose of the photomask in IC manufacture is to define the image to be printed on a silicon wafer. Of the many factors which affect this aerial image in the wafer stepper, some are properties of the stepper projection system and some are properties of the photomask. The purpose of photomask metrology is to help optimize photomask properties to produce the desired aerial image in the stepper. Those factors attributable to the photomask include chrome feature size, phase shifter accuracy (if used), chrome edge roughness and runout, defects, etc. Most of these are usually imaged differently in the exposure tool and the metrology tool because these tools' optical systems are different. Two ways to predict the stepper aerial image from photomask measurements are simulation (modeling) and emulation. In the first, the values of the relevant photomask measurements and properties (if they are known) can be inserted into the models for both tools, and then the stepper aerial image can be predicted from the photomask measurements. Or, the stepper's aerial image of the photomask can be emulated in the metrology tool and magnified for the purpose of measurement. Present mask metrology tools, when used in transmission, duplicate the optical morphology of the projection tool. The projected stepper aerial image can be emulated by adjusting the few major generic optical parameters of the metrology tool to match those of the wafer stepper to be used. Then the measured image in the metrology tool emulates the image in the stepper, and photomask properties affect both images in the same ways. Diverse measurements of photomask characteristics can be made accurately and fast by emulating photomask performance instead of measuring the chrome geometry. All of the significant optical performance parameters of the photomask can be measured with a single tool, at the small expense of shifting to a more appropriate philosophy of mask measurement.

Why it matters

OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

The ultimate purpose of the photomask in IC manufacture is to define the image to be printed on a silicon wafer. Of the many factors which affect this aerial image in the wafer stepper, some are properties of the stepper projection system and some are properties of the photomask. The purpose of photomask metrology is to help optimize photomask properties to produce the desired aerial image in the stepper. Those factors attributable to the photomask include chrome feature size, phase shifter accuracy (if used), chrome edge roughness and runout, defects, etc. Most of these are usually imaged differently in the exposure tool and the metrology tool because these tools' optical systems are different. Two ways to predict the stepper aerial image from photomask measurements are simulation (modeling) and emulation. In the first, the values of the relevant photomask measurements and properties (if they are known) can be inserted into the models for both tools, and then the stepper aerial image can be predicted from the photomask measurements. Or, the stepper's aerial image of the photomask can be emulated in the metrology tool and magnified for the purpose of measurement. Present mask metrology tools, when used in transmission, duplicate the optical morphology of the projection tool. The projected stepper aerial image can be emulated by adjusting the few major generic optical parameters of the metrology tool to match those of the wafer stepper to be used. Then the measured image in the metrology tool emulates the image in the stepper, and photomask properties affect both images in the same ways. Diverse measurements of photomask characteristics can be made accurately and fast by emulating photomask performance instead of measuring the chrome geometry. All of the significant optical performance parameters of the photomask can be measured with a single tool, at the small expense of shifting to a more appropriate philosophy of mask measurement.

Key concepts: Photomask, Stepper, Metrology, Aerial image, Optical proximity correction, Wafer, Photolithography, Computer science

Related papers

Back to paper searchBrowse research topicsOriginal source
Improved photomask metrology through exposure emulation — Research Paper | ScholarLens