Learning method and apparatus for predictive determination of endpoint during chemical mechanical planarization using sparse sampling

• US 6,676,482 B2
• Filed: 04/20/2001
• Issued: 01/13/2004
• Est. Priority Date: 04/20/2001
• Status: Active Grant

• Current Assignee: Novellus Systems Incorporated (LAM Research Corporation)
• Original Assignee: SpeedFam-Ipec Corporation
• Inventors: Adams, John A., Bibby, Thomas F. A. Jr.
• Primary Examiner(s): Rose, Robert A.
• Application Number:
  US09/839,631
• Publication Number:
  US 20020155789A1
• Time in Patent Office: 998 Days
• Field of Search: 451/5, 451/6, 451/8, 451/41, 451/288, 451/287
• US Class Current: 451/6
• CPC Class Codes: B24B 37/013 Devices or means for detecting lapping completion
• CPC Class Codes: B24B 49/04 involving measurement of the workpiece at the place of grinding during grinding operation
• CPC Class Codes: B24B 49/12 involving optical means

Abstract

A method and apparatus to generate an endpoint signal to control the polishing of thin films on a semiconductor wafer surface includes a through-bore in a polish pad assembly, a light source, a fiber optic cable, a light sensor, and a computer. The light source provides light within a predetermined bandwidth, the fiber optic cable propagates the light through the through-bore opening to illuminate the surface as the pad assembly orbits, and the light sensor receives reflected light from the surface through the fiber optic cable and generates reflected spectral data. The computer receives the reflected spectral data and calculates an endpoint signal by comparing the reflected spectral data with previously collected spectral reference data, calculating a trigger time based on the comparison, and predicting the endpoint time utilizing the trigger time.

Claims

1. A method for determining an endpoint during polishing of a semiconductor wafer, the method comprising:

sampling the wafer surface at time intervals to determine reflectance spectra at each time interval;

calculating a magnitude of a difference between a reflectance spectrum and a reference spectrum for each sampled time interval;

using paired data comprising the calculated magnitude and corresponding time interval to determine a best straight line curve fit;

determining a trigger time value when the magnitude difference is zero, based on the best curve fit; and

the trigger time is based on extrapolating the straight line fit to zero;

an endpoint time is determined by adding an over-polish time; and

determining a wafer polishing endpoint time based on the trigger time.

2. The method of claim 1 wherein the comparing step comprises calculating the sum of the squares of the differences between the reflected spectrum data and the reference spectrum data.

3. The method of claim 1, wherein the step of predicting the endpoint time comprises:

calculating a sum of the trigger time and a predetermined amount of time, wherein the predetermined amount of time is a constant.

4. The method of claim 1, wherein the step of predicting the endpoint time comprises:

calculating a sum of the trigger time and a predetermined amount of time, wherein the predetermined amount of time is a percentage of the trigger time.

5. The method of claim 1, wherein the step of collecting data samples is performed after a predetermined time delay, wherein the predetermined time delay is less than an expected total polish time.

6. An apparatus to generate an endpoint in the polishing of films on a semiconductor wafer for use in a chemical mechanical polishing system comprising:

a light source providing light to reflect from a film;

a light sensor receiving a spectrum of light reflected from the film, the light sensor including a processor generating, in digital form, spectral reflective data based on the reflected spectrum of light; and

a computer in communication with the light sensor receiving the generated data, the computer programmed to generate an endpoint based on the generated data, wherein the generation of the endpoint comprises;

calculating a trigger time based upon the collected data which comprises the steps of;

sampling the wafer surface at time intervals to determine reflectance spectra at each time interval;

calculating a magnitude of a difference between a reflectance spectrum and a reference spectrum for each sampled time interval;

using paired data comprising calculated magnitudes and corresponding time intervals to determine a best straight line curve fit; and

determining a trigger time value when the magnitude difference is zero, based on the best curve fit; and

determining the wafer polishing endpoint time based on the trigger time.

7. A method for detecting an endpoint during chemical mechanical polishing of a wafer surface of a wafer, the method comprising:

producing reference spectrum data corresponding to a spectrum of light reflected from a surface of a reference wafer at least at a time proximate to an estimated endpoint of the polishing;

producing reflectance spectrum data corresponding to a spectrum of light reflected from a surface of a production wafer at least at a time proximate to an expected endpoint;

comparing the reflected spectrum data with the reference spectrum data by calculating the sum of the squares of the differences between the reflected spectrum data and the reference spectrum data;

calculating a trigger time based upon a statistical analysis of the data collected; and

determining the endpoint time based on the trigger time.

8. The method of claim 7, wherein the step of calculating the trigger time comprises:

using paired data comprising calculated magnitudes and corresponding time intervals to determine a best straight line curve fit; and

determining a trigger time value when the magnitude difference is zero, based on the best curve fit.