Optics: measuring and testing – By light interference – For dimensional measurement
Reexamination Certificate
2000-05-11
2002-09-10
Turner, Samuel A. (Department: 2877)
Optics: measuring and testing
By light interference
For dimensional measurement
C356S512000
Reexamination Certificate
active
06449048
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention is related in general to the field of interferometry and, in particular, to a novel scanning approach based on lateral translation of the sample stage or interferometric objective.
2. Description of the Related Art
Accurate profilometry is a very important component of modern technology. As the art moves towards smaller parts and more accurate manufacturing methods, there is an increased requirement for fast and reliable methods of shape measurement, quality control and process monitoring. Advances in the semiconductor and data storage industries and new technologies, such as MEMS (micro-electro mechanical systems), require measurement resolution in the order of nanometers and ever faster processing.
Among the various measurement methods, optical techniques are preferred because of their non-contact nature, their high accuracy and the three-dimensional surface information they produce. Several optical phenomena have been exploited for this purpose. Depending on the shape, size and material of the test object, these techniques preferentially use structured light, focusing properties of optics, interference of light, etc., to achieve the best possible results in an economical and practical way. Moire' techniques, ESPI (electronic speckle-pattern interferometry), laser scanning, photogrammetry, and interferometry are only few of the many techniques developed for conducting three-dimensional shape measurements. The accuracy of these methods depends on many factors and varies from the sub-nanometer to the millimeter range. For large objects, such as car bodies, mechanical parts, etc., fringe projection, photogrammetry, or ESPI systems are usually utilized. Smaller objects, such as magnetic heads, require better absolute accuracy and methods such as interferometry or confocal microscopy are normally used.
Among these methods white-light vertical scanning interferometry (VSI), also commonly referred to as white-light interferometry or coherence radar, is used for small objects with roughness that does not exceed a few micrometers. The method is based on detection of the coherence peak created by two interfering, polychromatic wavefronts. It has many advantages such as absolute depth discrimination, fast measurement cycle, and high vertical resolution. It has proven to be very effective in measurements of objects with surface dimensions in the range of 15 mm down to sub-millimeter sizes.
Since its early development by Flourney at al. (P. A. Flourney, R. W. McClure and G. Wyntjes, “White-light Interferometric Thickness Gauge,” Appl. Opt. 11, 1907-1915, 972), the procedure has been significantly improved and commercially available systems have measurement repeatability of 0.1 nm or better. One of the important advantages of VSI is the ease with which it can be combined with other measurement techniques, such as phase-shifting interferometry (PSI), which are superior in accuracy but lack the scanning depth of VSI. For example, both techniques were combined in an system described as “PSI on the fly” (see U.S. Pat. No. 5,133,601 to Cohen et al.), and in U.S. Pat. No. 5,471,303 to Ai et al., wherein information from PSI is used to refine the height resolution of the VSI technique.
As is well understood in the art of vertical-scanning and phase-shifting interferometry, the optical path difference (OPD) between a test beam and a reference beam is varied in order to make a measurement. This is typically accomplished by shifting either the test surface or the reference surface of the interferometer along the optical axis by a predetermined distance during or between times of acquisition of data frames. The shift is normally carried out in steps or by continuous motion at a known, ideally constant, speed.
In the stepping method, the sample surface (or, alternatively, the reference surface) is moved between data frames and held still during data acquisition; thus, the OPD is kept constant during acquisition of each data frame. In practice, the shift-and-hold motion of the stepping method is mechanically undesirable because a finite amount of time is required for the shifted portion of the apparatus to settle into a static condition, thereby slowing down the process of data acquisition. Therefore, this method is no longer generally preferred in the industry.
In the ramping method, the OPD is varied in a continuous, smooth fashion, typically by scanning either the sample surface or the reference surface at constant speed throughout the measurement sequence. This approach is more common for vertical scanning interferometry and phase shifting because it allows faster measurements than the stepping method.
A typical vertical scanning interferometer
10
is shown in
FIG. 1. A
test sample or object
12
is placed in an interferometric setup illuminated with a “white” light source, such as an incandescent light bulb or LED
14
. The beam from the source, which may be passed through a broadband filter (not shown), is collimated in collimator optics
16
and divided into object and reference beams OB and RB, respectively, using a beam splitter
18
. The light reflected from the surface S of the object and from a reference surface
20
is then combined and the resulting image is projected on a CCD camera
22
through appropriate imaging optics
24
for registration and further processing by a computer (not shown).
The contrast of registered fringes, resulting from the interference between the object and reference beams, is proportional to the modulus of the complex degree of the mutual coherence of the wavefronts OB and RB. Since the light source is polychromatic, it has a low degree of time coherence and, therefore, the interference can occur only in a limited space around the coherence plane P defined by a zero optical path difference (OPD).
This property is used for retrieval of the object's shape. The object is scanned along the optical axis of the instrument (referred to as the z direction in the figure) such that different heights pass through the coherence plane. In this limited coherence space the interference fringes become visible as the modulation of intensity along the scanning path, the peak of the fringe contrast corresponding to a vertical position where the OPD is zero. During the scan a number of intensity frames is acquired at defined locations and the intensity from each detector pixel (corresponding to a single point on the object) is analyzed as a function of the object's vertical position. Normally the maximum of the intensity modulation envelope, or its center, defines the relative height of the observed object point. The profile of the correlogram is defined by the equation:
I
(
x,y,z
)=
a
(
x,y
)+
m
(
x,y
)
c[z
−2
h
(
x,y
)]cos [2
&pgr;w
o
z
−&agr;(
a,y
)] (1)
where I(x,y,z) is the intensity at location (x,y) and height z on the sample surface; a(x,y) is the average intensity; m(x,y) is the modulation; c(z) is the envelope function defined by the spectral properties of the light source; h(x,y) is the height of the object; w
o
is the mean wavelength of the source; and &agr;(x,y) is the initial phase difference between the object and reference beams. See G. S. Kino and S. S. C. Chim, “Mirau correlation microscope,” Appl. Opt. 29, 3775-3783 (1990).
A typical correlogram C from a single pixel of an object obtained by the vertical scanning method is shown in FIG.
2
. The high frequency modulating signal arises from the interference of light while the envelope corresponds to the contrast of fringes, i.e., the modulus of the mutual degree of coherence of the interfering wavefronts.
Subsequent data processing is used to retrieve the accurate position of the peak of the envelope. Many algorithms have been devised in the art to detect the location of the maximum. In most cases, the intensity signal detected at the camera
22
is differentiated to remove the DC term and subsequently a center of mass, Fourier analysis, or curve fitting approach is used
Durando Antonio R.
Durando Birdwell & Janke P.L.C.
Turner Samuel A.
Veeco Instruments Inc.
LandOfFree
Lateral-scanning interferometer with tilted optical axis does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Lateral-scanning interferometer with tilted optical axis, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Lateral-scanning interferometer with tilted optical axis will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2896730