Browse Prior Art Database

Extending to Shorter Wavelengths and Improving the Accuracy of Spectroscopic Thickness Measurements of Very Thick Films

IP.com Disclosure Number: IPCOM000089263D
Original Publication Date: 1977-Oct-01
Included in the Prior Art Database: 2005-Mar-04
Document File: 3 page(s) / 28K

Publishing Venue

IBM

Related People

Pliskin, WA: AUTHOR

Abstract

On using CARIS (constant angle reflection interference spectroscopy), it is often desired to make the measurements at shorter wavelengths because at the shorter wavelengths the separation between minima is less, and thus one can get quicker measurements at the shorter wavelengths. However on very thick films, even those which are very uniform, the fringe intensity decreases so much at shorter wavelengths that they tend to get "washed out", and are not accurately defined even in a good spectrophotometer. The fringe intensity can be increased throughout the spectrum and especially at shorter wavelengths by masking the amount of reflected light at a distance from the sample being examined. This mask is in addition to the usual mask used to define the area of the film being examined.

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 52% of the total text.

Page 1 of 3

Extending to Shorter Wavelengths and Improving the Accuracy of Spectroscopic Thickness Measurements of Very Thick Films

On using CARIS (constant angle reflection interference spectroscopy), it is often desired to make the measurements at shorter wavelengths because at the shorter wavelengths the separation between minima is less, and thus one can get quicker measurements at the shorter wavelengths. However on very thick films, even those which are very uniform, the fringe intensity decreases so much at shorter wavelengths that they tend to get "washed out", and are not accurately defined even in a good spectrophotometer. The fringe intensity can be increased throughout the spectrum and especially at shorter wavelengths by masking the amount of reflected light at a distance from the sample being examined. This mask is in addition to the usual mask used to define the area of the film being examined.

One normally thinks of the light reflected from a sample being examined, as shown in Fig. 1, at the center. Actually, there is a cone of angles, as shown in Fig. 1. Thus, in effect, there is not just one angle of incidence Theta as shown in the simplified equation D = Ni Lambda(i) over 2 (n(i)/2/ - sin/2/ Theta) 1/2. where D is the film thickness, N is the order (half integral for minima) which decreases with an increase in wavelength, and n is the refractive index at each minimum corresponding to Lambda(i). In the actual case one would have to integrate over the angular region Theta(a) to Theta(b) to get the proper applicable formula. In practice, this cone of angles causes the fringes to get washed out especially at shorter wavelengths.

This problem was essentially overcome by use of the following setup on a GCA-McPherson spectrophotometer, as shown in Fig. 2. In the set-up of Fig. 2 the mask opening (vertical only, the horizontal dimension being significantly larger) was 0.40(4) cm, and it was about 8.5 cm from the sample. Thus, the HCA (half of the cone angle or half of the divergence in the plane of incidence) is 1.4 Degrees. Before putting in and testing this set-up, a set-up was tested where the mask opening was about 0.6 cm, corresponding to HCA = 2 Degrees. A very thick thermal oxide...