Optics: measuring and testing – Dimension – Thickness
Reexamination Certificate
2000-07-14
2003-06-03
Epps, Georgia (Department: 2873)
Optics: measuring and testing
Dimension
Thickness
C356S630000
Reexamination Certificate
active
06573999
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to the measurement of a film thickness and in particular to a method of the measurement of thin film thickness using light.
BACKGROUND
The use of light to measure the thickness of a film, e.g., a film that overlies a substrate during the production of an integrated circuit or flat panel display, is well known. For example, Fourier Transform Infrared (FTIR) spectrometers operating in the infrared (IR) wavelength range is a well-known metrology tool useful in a variety of scientific or technological fields, for example, biology, geology, forensics, nutrition science, medicine, and semiconductor processing.
A Fourier Transform Infrared (FTIR) spectrometer typically uses a Michelson interferometer, which includes an IR source, a beam splitter, and two plane mirrors (one fixed and one moving) and after mathematical processing, produces a spectrum of the light coming from the sample. A FTIR spectrometer detects the absorption of the IR light that is either passed through or reflected by the sample. In the IR range, absorption of light is associated with chemical bonds in molecular structures and, thus, valuable compositional information can be obtained. Because the spectrum variations can also be caused by interference effects of light reflecting from different interfaces, film-thickness information can also be extracted. Examples of FTIR spectrometers are the Century Series FT-IR Spectrometers made by Bio-Rad located in Cambridge, Mass., the Epitaxial Layer Thickness Monitor MappIR by PIKE Technologies, located in Madison Wis., the MB Series of FTIR Spectrometers by Bomem, located in Quebec, Canada, the Genesis Series FTIR by Mattson, located in Madison, Wis., the M-Research Series and SPR Prospect IR spectrometer by Midac Corporation, located in Irvine Calif.
There are multiple conventional methods to extract layer thickness from a mid-IR spectrum (2.5 &mgr;m to 25 &mgr;m). One method is based on interference phenomena, which results from the constructive and destructive interference that occurs at different wavelengths between the reflection at the top surface of a layer and the successive reflection at bottom surface of the layer. By curve-fitting this spectrum over a wide range, the thickness of the layer may be determined. This method requires that two prerequisites be satisfied. First, the layer to be measured must be partially transparent. Second, the optical constant dispersion of the layer and substrate needs to be pre-determined or expressed as a known function of the wavelength. Due to the complexity of optical responses from a variety of materials in the mid-IR spectral region, it may take an enormous analysis effort and consume a large amount of time in order to determine the optical constants of the material.
Another conventional method is based on empirical calibration of the IR transmittance. In this method, the transmittance at a wavelength or transmittance spectrum over a band of wavelengths is measured. The material absorbs the infrared light at those wavelengths. The thickness of the material can be determined by interpolating the measured transmittance according to a standard set of transmittance values of a series of known thickness samples. The general principle behind the second method is to have a set of standard samples (called calibration set) of known thickness values determined using an independent means of thickness measurement from which can be generated a calibration table which correlates measured thickness values from the calibration set to IR transmittance values. As long as an unambiguous correlation can be established between thickness and IR transmittance, thickness of any thin film can be determined using this calibration method.
In the calibration method, the thickness of a film is determined by monitoring the attenuation of the transmitted light intensity by the film being tested at a predetermined wavelength. The attenuation is caused by the absorption of the light by the material of the film being tested and is related to the thickness of the film through Lambert's law, which is I=l
0
e
−60 z
, where z is the distance light travels inside the material, &agr; is the absorption coefficient of the material, I
0
is the light intensity just inside the material, and I is the remaining light after traveling a distance Z inside the material. To calibrate the test, a reference scan is used prior to the measurement of the film. The reference scan determines the light intensity through the ambient atmosphere and a substrate without the film. This method of determining the thickness of a film, however, requires that the substrate is transparent in the interested wavelength region, or the film is free-standing.
SUMMARY
A method of measuring the thickness of a top layer in a stack, which includes the top layer and underlying material, includes measuring the reflectance spectrum produced by the stack and determining the thickness of the top layer based on the top layer's attenuation of absorption band of the underlying material in the reflectance spectrum or its derivative. This absorption band of the underlying material is separate from the top layer's absorption band and the top layer has finite absorption in this wavelength region. A correlation between the thickness of the top layer and the strength of the absorption band of the underlying material in the reflectance spectrum or its derivative is used. The correlation is determined by identifying an absorption band of the underlying material in the reflectance spectrum or its derivative, which may be done by collecting a reflectance spectrum on a sample without the top layer or by consulting a library of absorption data. The reflectance spectra of calibration samples, each having a top layer with a different thickness, are taken. If needed, the derivative of each reflectance spectrum is then calculated. The strength of the absorption band is measured by measuring a peak height or a valley depth, or by computing a peak or a valley area, and then correlated to the thickness of the top layer for each calibration sample. Based on the correlation, any thickness of a top layer having the same composition of material as the calibration samples, may be determined based on the attenuation or the strength of the absorption band characteristic of the underlying material in the reflectance spectrum or its derivative. The underlying material may be a single layer or a composite of multiple layers.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1
shows a cross-sectional view of a stack comprising a substrate with overlying layers, the top of which is to be measured using a method in accordance with an embodiment of the present invention.
FIG. 2
is a flow chart illustrating a method of deriving the relationship between the strength of an absorption band of the underlying material in the reflectance spectrum of the stack and the thickness of the film being measured.
FIG. 3
shows a cross-sectional view of a sample with the underlying material layer disposed on a substrate.
FIG. 4
is a graph showing the absorption band R
0
(&lgr;) of the underlying material in the reflectance spectrum of the underlying material.
FIG. 5
shows one example of a calibration sample with an overlying top film, the underlying material layer and substrate.
FIG. 6
is a graph showing the absorption band R
0
(&lgr;) of the underlying material in the reflectance spectrum of the stack being attenuated by several different thicknesses of top films.
FIG. 7
is a graph showing the correlation between the peak area of the absorption band R
0
(&lgr;) in the reflectance spectrum of the stack and the thickness of the top film.
FIG. 8
is a graph showing the correlation between the peak area of the absorption band R
0
(&lgr;) in the derivative of the reflectance spectrum of the stack and the thickness of the top film.
REFERENCES:
patent: 4243882 (1981-01-01), Yasujima et al.
patent: 4421983 (1983-12-01), Fogle et al.
patent: 4707611 (1987-11-01), Southwell
patent: 4984894 (
Dinh Jack
Epps Georgia
Halbert Michael J.
Nanometrics Incorporated
Silicon Valley Patent & Group LLP
LandOfFree
Film thickness measurements using light absorption does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Film thickness measurements using light absorption, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Film thickness measurements using light absorption will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3131293