Browse Prior Art Database

Second and Third Layer Quartz Via Etch End Point Detect Program

IP.com Disclosure Number: IPCOM000086245D
Original Publication Date: 1976-Aug-01
Included in the Prior Art Database: 2005-Mar-03
Document File: 3 page(s) / 42K

Publishing Venue

IBM

Related People

Smith, EG: AUTHOR

Abstract

Described is a reflective measurement data processing technique which reliably and repeatably signals the end of second and third-level quartz etch processes, i.e., processes for etching holes in second and third-level insulative layers in multilevel metallurgy integrated circuits, thereby enhancing product uniformity and process yield.

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

Second and Third Layer Quartz Via Etch End Point Detect Program

Described is a reflective measurement data processing technique which reliably and repeatably signals the end of second and third-level quartz etch processes, i.e., processes for etching holes in second and third-level insulative layers in multilevel metallurgy integrated circuits, thereby enhancing product uniformity and process yield.

The present method is summarized as follows. A typical integrated circuit chip, envisioned here as centered on a circular-shaped notched wafer (but not necessarily restricted to the centered position), is employed to collect sampled reflectance data using standard reflectance detect apparatus (see Fig. 1). The data is then processed by the 2nd (and 3rd) quartz Via Etch Algorithm (described below).

The algorithm outputs an end-of-etch process (EEP) signal which actuates process quenching processes after a suitable choice of overetch time has elapsed. (It is important to note here that the previously mentioned overetch time is a parameter of choice which is specified by the algorithm user.)

With reference to Fig. 2, let the algorithm parameters be denoted b, T, and N where:. b = Smoothing filter constant (b is greater than zero and less than unity); T = Algorithm threshold constant (T is greater than zero); and, N = Number of data samples required for smoothing filter to reach steady-state operation (nominally set at N = 150).

Let x(n) denote the n-th reflectance data sample, a digitized voltage sample at a time equal to n times the sampling period (nominally assumed to be one second), where n = 0,1,2,... In what follows:. X(n) = Predicted value of x(n+1), g = 1 - b, and m(n) and M(n) denote computed statistics. Write: (1) X(n) = X(n-1) + g[x(n) - X(n-1)], X(0) = x(1), (2) m(n) = m(n-1) + g[E(n) - m(n-1)], m(0) = 0, and
(3) M(n) = M(n-1) + (1/n)[E(n) - M(n-1)], M(0) = 0 where; (4) E(n) = absolute value of x(n) - X(n-1) and where n = 1,2,...

In terms of the foregoing relations, Fig. 2 is a flow chart of the 2nd (and 3rd) Quartz Via Etch Recursive Filter Algorithm. It should be noted here that the algorithm has been exp...