Rapid and accurate end point detection in a noisy environment

• US 6,172,756 B1
• Filed: 12/11/1998
• Issued: 01/09/2001
• Est. Priority Date: 12/11/1998
• Status: Expired due to Term

• Current Assignee: Filmetrics, Inc. (KLA Corporation)
• Original Assignee: Filmetrics, Inc. (KLA Corporation)
• Inventors: Chalmers, Scott A., Geels, Randall S.
• Primary Examiner(s): LAUCHMAN, LAYLA G
• Assistant Examiner(s): Smith, Zandra V.
• Application Number:
  US09/210,140
• Time in Patent Office: 760 Days
• Field of Search: 356/381, 356/382, 451/5, 451/6
• US Class Current: 356/630
• CPC Class Codes: B24B 37/013 Devices or means for detecting lapping completion
• CPC Class Codes: B24B 49/02 according to the instantaneous size and required size of the workpiece acted upon, the measuring or gauging being continuous or intermittent B24B49/12 takes precedence
• CPC Class Codes: B24B 49/12 involving optical means
• CPC Class Codes: G01B 11/0683 measurement during deposition or removal of the layer

Abstract

Methods and apparatus for film measurement and endpoint detection in a noisy environment, such as CMP processing of semiconductor wafers, are disclosed, characterized by the use of spectral analysis of intensity data derived from light reflected off the sample to estimate film thickness or detect an endpoint condition.

Claims

1. An end point detection apparatus comprising:

a white light source configured to generate a light signal;

a first fiber optic cable connected to said light source to receive said light signal and further configured to direct said light signal onto a sample, said light signal reflecting from said sample to create a reflected light signal having a plurality of wavelength components, each having an intensity;

a second fiber optic cable positioned to receive said reflected light signal;

a spectrometer configured to receive said reflected light signal from said second fiber optic cable, and determine therefrom a plurality of electrical signals, each representative of the intensity of a wavelength component of the reflected light;

a computer programmed to receive from said spectrometer said plurality of electrical signals, and determine therefrom the presence of an endpoint condition for at least one film included in said sample, the film comprising a material having a refractive index, by;

obtaining data representative of the intensity of at least some of said wavelength components;

arranging said data so that it is a function of a variable equal to the refractive index of the material divided by wavelength;

normalizing said data using a reference spectrum;

determining the Power Spectral Density of the normalized data as a function of potential layer thickness;

integrating the Power Spectral Density over potential layer thickness to obtain an endpoint parameter;

detecting a transition in the endpoint parameter; and

detecting the presence of the endpoint condition from the transition.

2. The apparatus of claim 1 further comprising a CMP station for processing the film.

3. An end point detection apparatus comprising:

a light source configured to generate a light signal;

a first optical director configured to receive said light signal and direct said light signal onto a sample during processing to obtain a reflected light signal having a plurality of wavelength components, each having an intensity;

a second optical director positioned to receive said reflected light signal;

an optical device configured to receive said reflected light signal from said second director, and derive therefrom a plurality of electrical signals, each representative of the intensity of a wavelength component of the reflected light;

an electronic device configured to receive said plurality of electrical signals and convert them to numeric representations of the intensity of each detected wavelength component;

a computer programmed to receive said plurality of numeric representations, and detect therefrom an endpoint condition of at least one film included in said sample by;

obtaining data representative of the intensity of at least some of said wavelength components;

performing analysis on said data to obtain thickness spectral data as a function of potential layer thickness;

integrating the thickness spectral data over a range of potential layer thicknesses to obtain an endpoint parameter;

detecting a transition in the endpoint parameter; and

detecting the endpoint condition from the transition of the endpoint parameter.

4. The apparatus of claim 3 further comprising a CMP station for processing the film, wherein the computer is programmed to signal the CMP station upon detecting the presence of the endpoint condition.

5. An endpoint detection apparatus comprising:

a first means for directing a light signal onto a sample during processing to obtain a reflected light signal having a plurality of wavelength components, each having an intensity;

second means for receiving said reflected light signal and deriving therefrom data representative of the intensity of at least some of the wavelength components of the reflected light;

third means for performing analysis of said data to obtain thickness spectral data;

fourth means for deriving an endpoint parameter from said thickness spectral data; and

fifth means for detecting a transition of the endpoint parameter and detecting therefrom an endpoint condition of a film included within the sample.

6. A computer program medium embodying a program of instructions executable by a computer to perform a method for detecting an endpoint condition for at least one film included within a sample from light reflected from the sample having a plurality of wavelength components, each having an intensity, and the film comprising a material having a refractive index, the method comprising the following steps:

obtaining data representative of the intensity of at least some of said wavelength components;

arranging said data so that it is a function of a variable equal to the refractive index of the material divided by wavelength;

normalizing said data using a reference spectrum;

determining the Power Spectral Density of the normalized data as a function of potential layer thickness;

integrating the Power Spectral Density over a range of potential layer thicknesses to obtain an endpoint parameter;

detecting a transition in the endpoint parameter; and

detecting the endpoint condition from the transition.

7. A computer program medium embodying a program of instructions executable by a computer to perform a method for detecting an endpoint condition for at least one film included within a sample from light reflected from the sample having a plurality of wavelength components, each having an intensity, the method comprising the following steps:

obtaining data representative of the intensity of at least some of said wavelength components;

performing analysis on said data to obtain thickness spectral data as a function of potential layer thickness;

integrating the thickness spectral data over a range of potential layer thicknesses to obtain an endpoint parameter;

detecting a transition in the endpoint parameter; and

detecting the endpoint condition from the transition.

8. A computer program medium embodying a program of instructions executable by a computer to perform a method for detecting an endpoint condition for at least one film included within a sample from light reflected from the sample having a plurality of wavelength components, each having an intensity, the method comprising the following steps:

obtaining data representative of the intensity of at least some of the wavelength components of the reflected light;

performing analysis of said data to obtain thickness spectral data;

deriving an endpoint parameter from said thickness spectral data;

detecting a transition of the endpoint parameter; and

detecting the endpoint condition from said transition of said endpoint parameter.

9. A method for detecting an endpoint condition for at least one film included within a sample from light reflected from the sample having a plurality of wavelength components, each having an intensity, and the film comprising a material having a refractive index, the method comprising the following steps:

obtaining data representative of the intensity of at least some of said wavelength components;

arranging said data so that it is a function of a variable equal to the refractive index of the material divided by wavelength;

normalizing said data using a reference sample;

determining the Power Spectral Density of the normalized data as a function of potential layer thickness;

integrating the Power Spectral Density over potential layer thickness to obtain an endpoint parameter;

detecting a transition in the endpoint parameter; and

detecting the endpoint condition for the film from the transition.

10. A method for detecting an endpoint condition for at least one film included within a sample from light reflected from the sample having a plurality of wavelength components, each having an intensity, the method comprising the following steps:

obtaining data representative of the intensity of at least some of said wavelength components;

performing analysis on said data to obtain thickness spectral data as a function of potential layer thickness;

integrating the spectral data over a range of potential layer thicknesses to obtain an endpoint parameter;

detecting a transition in the endpoint parameter; and

detecting the endpoint condition from the transition.

11. The method of claim 10 further comprising processing the film using a CMP processing station, and signaling the CMP station upon detecting the endpoint condition.

12. The method of claim 10 in which the performing step includes deriving the Fourier Transform of the intensity data.

13. The method of claim 10 in which the performing step includes deriving the Power Spectral Density of the intensity data.

14. The method of claim 10 further comprising the step of processing the film from the CMP station during the foregoing steps.

15. A method for detecting an endpoint condition for at least one film included within a sample from light reflected from the sample having a plurality of wavelength components, each having an intensity, the method comprising the following steps:

obtaining data representative of the intensity of at least some of the wavelength components of the reflected light;

performing analysis of said data to obtain thickness spectral data;

deriving an endpoint parameter from said thickness spectral data;

detecting a transition of the endpoint parameter; and

detecting the endpoint condition from said transition of said endpoint parameter.