Browse Prior Art Database

Semiconductor Process Defect Monitor

IP.com Disclosure Number: IPCOM000082772D
Original Publication Date: 1975-Feb-01
Included in the Prior Art Database: 2005-Feb-28
Document File: 3 page(s) / 35K

Publishing Venue

IBM

Related People

Ghatalia, AK: AUTHOR [+2]

Abstract

These semiconductor defect monitors enable defect distribution density to be determined for various size defects. The resulting data is useful in determining in-process semiconductor yield levels.

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 53% of the total text.

Page 1 of 3

Semiconductor Process Defect Monitor

These semiconductor defect monitors enable defect distribution density to be determined for various size defects. The resulting data is useful in determining in-process semiconductor yield levels.

If a monitoring pattern 10, as shown in Fig. 1, is provided on the surface of a semiconductor wafer using the same process steps as used by a product containing wafers, process induced defects experienced by the monitoring pattern will be the same as those experienced by the product.

There is a certain probability that defects will occur such that the continuity of the line 10 will be broken. The number of defects in the size range affecting integrated circuit chip operation varies as a monadic decreasing function with the defect size. Therefore, the susceptibility of the monitor to defects will vary with the length and width of the line 10. Assuming a Poisson distribution in defects, the probability of no defects occurring such as to break the line, i.e., the yield of the line, can be given by: Yw = exp(-W /inf/ A(x)D(x)dx); where A(x) is the susceptible area for defects of diameter x and D(x) is the defect density.

A(x), from Fig. 2 is equal to L(x-w) where x is the diameter of defect 12, w is the line width, and L is the line length. Here a circular defect has been assumed, as the probability of a noncircular defect falling in any one orientation is a random function. Substituting the above in the yield equation results in: Yw = exp (-Lw/inf/ (x-w)D(x)dx).

By constructing monitors of varying widths, it is possible to determine yields for various size defects and to construct an iterative curve fit in order to determine the defect density function, D(x), which best fits the actual defect data.

Yield data for defects which cause conductive lines to open i...