Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

METHOD FOR DETERMINING STEPPER FOCUS AND OFFSETS

IP.com Disclosure Number: IPCOM000006561D
Original Publication Date: 1992-Aug-01
Included in the Prior Art Database: 2002-Jan-15
Document File: 4 page(s) / 157K

Publishing Venue

Motorola

Related People

Sandeep Malhotra: AUTHOR [+3]

Abstract

Individual semiconductor layer features must be the right size and have tight critical dimension (CD) con- trol, or have low CD variation, for integrated circuits to function to performance and yield specs. One of the factors governing CD control is the stepper focus set- ting or the stepper focus offset (see Figure 1; ref. 1). A baseline stepper focus setting is determined for every stepper every day using the stepper self-metrology sys- tem, In-situ (2). Since this focus value is determined on a perfect substrate, namely flat silicon wafers, it may not be an appropriate value for real product wafers at some layers, where a significant amount of vertical fabrication layer depositions have occurred, or where due to reduced feature size required, a better centering of the focus set- ting is necessary to ensure good overall CD control within wafer and wafer to wafer. When these effects occur the data of Figure 1 would sbii? on the focus axis, the amount of focus offset induced by the substrate (see ref. 1 for an example). Even worse, the data of Figure 1 can be asymmetrical in defocus. A better method of obtaining the required layer-dependent focus offset value has been developed and is described here.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 47% of the total text.

Page 1 of 4

0 M

MOTOROLA INC. Technical Developments Volume 16 August 1992

METHOD FOR DeSERMlNlNG STEPPER FOCUS AND OFFSmS

by Sandeep Malhotra, Chris Smith and John Helbert

  Individual semiconductor layer features must be the right size and have tight critical dimension (CD) con- trol, or have low CD variation, for integrated circuits to function to performance and yield specs. One of the factors governing CD control is the stepper focus set- ting or the stepper focus offset (see Figure 1; ref. 1). A baseline stepper focus setting is determined for every stepper every day using the stepper self-metrology sys- tem, In-situ (2). Since this focus value is determined on a perfect substrate, namely flat silicon wafers, it may not be an appropriate value for real product wafers at some layers, where a significant amount of vertical fabrication layer depositions have occurred, or where due to reduced feature size required, a better centering of the focus set- ting is necessary to ensure good overall CD control within wafer and wafer to wafer. When these effects occur the data of Figure 1 would sbii? on the focus axis, the amount of focus offset induced by the substrate (see ref. 1 for an example). Even worse, the data of Figure 1 can be asymmetrical in defocus. A better method of obtaining the required layer-dependent focus offset value has been developed and is described here.

  Although In-Situ is an adequate method for deter- mining focus on ideal set up wafers for simple machine stability checks, it can experience trouble functioning due to the presence of wafer fabrication thin fdms(l). In Figure 2b, it is easily seen that the ability of the In-situ system can be severely hampered by the presence of a thin film, and focus offset measurement errors can be as large as 1 pm. Figure 2a contains no thin films and there the resulting data has no unwanted signal excursions and better overall signal to noise ratio, hence, the focus value is accurately obtained. To overcome this major limitation, a simple operator compatible empirical focus offset method has been developed.

  The method entails the application of a snake patterned (Figure 3) stepper reticle, a tightly pitched large pattern, which provides an easily identified via a LX naked eye inspection based on the defocus dependent wafer pattern. After development of the resist snake pattern on a wafer with all the device sublayers, best focus can

0 Motorola, 1°C. ,992

be visually determined from the focus stepped wafer pattern without the use of a microscope or the sopbisti- cated In-situ metrology and/or other general computer analysis software, which take extended amounts of time and skill to employ to get an offset number. The advan- tages of this system vs. a system like In-situ are its 1) simplicity, 2) low cost, 3) operator compatibility, 4) real time nature, and 5) accuracy to measure layer-dependent wafer focus offsets. Another key advantage of the sys- tem is it takes into account thin fdm resi...