Radiation imagery chemistry: process – composition – or product th – Including control feature responsive to a test or measurement
Reexamination Certificate
1999-06-11
2001-02-06
Young, Christopher G. (Department: 1756)
Radiation imagery chemistry: process, composition, or product th
Including control feature responsive to a test or measurement
C430S022000, C356S370000
Reexamination Certificate
active
06183919
ABSTRACT:
TECHNICAL FIELD
The present invention relates generally to manufacturing processes requiring lithography and, more specifically, to monitoring of lithographic processes used in microelectronics manufacturing.
BACKGROUND OF THE INVENTION
Lithography is used in microelectronics manufacturing, particularly in the manufacture of semiconductors, to transfer an integrated circuit pattern to a photoresist film on a substrate. Radiation, such as light, is spatially modulated through a pattern on a mask or reticle to form an aerial image that exposes the radiation-sensitive photoresist film according to the mask pattern. The photoresist is developed, or the pattern may be transferred to the substrate by an etching step, followed by removal of the photoresist.
Because the exposure and focus of the aerial image relative to the photoresist may vary due to variations in substrate reflectivity, film thickness, or topography, it is necessary to continually monitor the transferred patterns to verify that the dimensions of the patterns are within an acceptable range. The importance of such monitoring increases as the size of the features being produced decreases. The difficulty in monitoring such features also increases, however, as the size of the features decreases. This difficulty is exacerbated for features having a size on the order of one micron or less. This is because the preferred method of using a scanning electron microscope (SEM) for performing inspections tends to be relatively slow in operation and difficult to automate for features of a smaller size. The use of optical tools permits faster and more readily automated operations to be implemented, but such optical techniques are inadequate to resolve features of a smaller size, particularly those having dimensions of less than about one micron.
To overcome this problem, U.S. Pat. No. 5,629,772 (issued to Ausschnitt, assigned to the assignee of the present invention, and incorporated by reference) discloses an optical metrology method used in the manufacture of microelectronics. Essentially, referring now to
FIGS. 1 and 2
, this method comprises using a lithographic process to create a pattern
28
comprising an array of elements
44
on a substrate
45
, each element
44
in the array having a length “L” and width “W,” with spaces
46
between adjacent elements
44
also having a width W. Substrate
45
may have other layers on its surface, such as layers
45
′ and
45
″ as shown in
FIG. 2
, which may typically comprise silicon dioxide and silicon nitride, respectively.
Typically, width W of elements
44
and spaces
46
between adjacent elements
44
corresponds to the minimum feature dimension for the lithographic process used to create the elements
44
. In contrast, length L is larger than the minimum feature. The known method comprises measuring the larger length L of elements
44
as created by the lithographic process, calculating the change in length L of the elements
44
from the nominal length of the elements
44
, and calculating the lithographic process bias of the minimum feature from the change in length of the array elements
44
. The term “bias” is used to describe the difference between the dimensions of a feature as actually created by the lithographic process and the corresponding nominal dimensions of the pattern
28
desired to be created.
One potential problem with monitoring minimum features in this way is the interaction of sub-layer films with the estimate of the minimum feature size. As the thickness and optical characteristics of the underlying films change, a bias in the minimum feature width is often caused by interference effects between the light reflected from the various regions within and outside the minimum feature structure. Such a bias can be substantially reduced by eliminating the zero order or specular component in the image formation process, such as by using darkfield imaging methods.
A number of darkfield methods have been proposed that provide zero order rejection in the context of optical alignment of wafer structures and for detection of minimum features using optical imaging and polarization rejection. See, generally, Bobroff et al., “Alignment Errors from Resist Coating Topography,”
J. Vac. Sci. Tech.,
Vol. B6(1) (Jan/Feb 1988), and co-pending U.S. patent application Ser. No. 09/159,240 (Progler et al.). Methods for higher-resolution measurement of minimum feature size while accounting for sub-layer film bias, however, as well as edge-detection methods that may be implemented on existing brightfield microscopy equipment with minimum modifications, are still desired.
SUMMARY OF THE INVENTION
The present invention provides a method for optically measuring lithographic process bias of a minimum feature formed by a lithographic process. The method comprises creating on a substrate an array of elements having spaces between the elements. Each element in the array has a nominal length and a nominal width, and each space has a nominal width. The nominal width of each element and of each space corresponds to the minimum feature. The nominal length is larger than the minimum feature.
The method further comprises measuring the length of the elements in the array as created by the lithographic process. The length is measured by generating a darkfield optical image of the array of elements, detecting the darkfield optical image with an optical detector, generating electronic information corresponding to the optical image, and processing the electronic information to determine the created length of the elements in the array. The method still further comprises calculating the difference between the created length versus the nominal length and determining lithographic process bias of the minimum feature from that difference in length.
Generation of the darkfield image may further comprise generating a double-lobe darkfield optical image, in which case the method comprises generating a corresponding double-lobe intensity profile. The electronic information corresponding to the intensity profile may be processed and evaluated to calculate the created element length using a matched filter, application of a threshold slice, or definition of redundant measures of feature width as explained below.
The method of the present invention may be implemented on a system comprising a pattern for creating an array of elements to be formed by a lithographic process, an array of elements actually formed by the lithographic process, a darkfield imaging system, an optical detector, and a signal processor. The pattern comprises an array of elements having spaces between the elements, each element having a nominal length and a nominal width, and each space having a nominal width. The nominal width of each element and of each space corresponds to the minimum feature of the lithographic process. The nominal length is larger than the minimum feature.
The array of elements is actually created by the lithographic process on a substrate, each element with an actual length and an actual width and each space with an actual width. The darkfield imaging system is adapted to create a darkfield image of the array of elements. The optical detector is adapted to detect that darkfield image and produce corresponding electronic information from that image. The signal processor is adapted to receive the electronic information and process it to determine the lithographic process bias corresponding to the difference between the actual width and the nominal width as calculated based on a difference between the actual length and the nominal length.
The signal processor may further comprise a program storage device readable by a machine, tangibly embodying a program of instructions executable by the machine to perform the method steps for processing electronic information corresponding to the darkfield image to determine the lithographic process bias. The tangibly embodied, machine-executable method steps comprise receiving electronic information corresponding to the darkfield image from the optical detector, processing th
Ausschnitt Christopher P.
Progler Christopher J.
International Business Machines - Corporation
Ratner & Prestia
Young Christopher G.
LandOfFree
Darkfield imaging for enhancing optical detection of edges... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Darkfield imaging for enhancing optical detection of edges..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Darkfield imaging for enhancing optical detection of edges... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2611229