Inspection method and apparatus for detecting defects on...

Optics: measuring and testing – Inspection of flaws or impurities – Surface condition

Reexamination Certificate

Rate now

  [ 0.00 ] – not rated yet Voters 0   Comments 0

Details

C356S394000, C382S144000

Reexamination Certificate

active

06297879

ABSTRACT:

FIELD OF THE INVENTION
The present invention generally relates to the manufacture of semiconductor devices. In particular, the present invention relates to inspection techniques used for detecting defects on photomasks used in photolithography.
BACKGROUND OF THE INVENTION
The capacity of integrated circuits has increased primarily as the result of reductions in the size of features on a semiconductor chip. The lateral dimensions of features are generally defined by photolithographic techniques in which a detailed pattern is transferred to a reactive material by shining light through a photomask or reticle. During the photolithography process, energy is applied to photo resist deposited on a wafer, where the energy application is controlled through the use of a patterned photomask. The exposure to the wafer is made by a step and repeat procedure. In this procedure, the wafer is moved and the steppers are used to move and repeat the pattern of the photomask over the wafer. Since accuracy of the pattern is essential as it is repeated several times, the pattern of the photomask is enlarged when it is created and reduced while it is being exposed on the wafer. Although some defects are effectively eliminated by the use of a {fraction (1/10)} or a ⅕ reduction stepper, many defects remain as minimum feature size on the photomask continues to decrease. Moreover, the decrease in minimum feature size has increased the printability of certain types of defects because a slight variation of the exposure dose can cause repeating defects on the wafer.
Printable defects on photomasks and reticles have historically been a source of defects that have reduced die yields. Printable defects in the photomasks are repeated many times over the surface of a semiconductor wafer since the photomask is stepped and repeated over the wafer. For fatal defects, this can result in substantial yield losses. Accordingly, it is important to detect and correct as many defects as possible in the photomasks.
Since fatal defects in a mask or reticle are highly undesirable, it would be useful if such defects could be repaired, thereby rendering the mask free of fatal defects. One of the mask repair methods for accomplishing this purpose is laser repair of the photomask. After the defect has been corrected by the laser repair, the photomask must be inspected again to verify the repair. However, defects which continue to exist after laser repair can also remain undetected and continue to print on the wafer. Therefore, inspection of photomasks is an important step in the photolithography process.
Originally, photomasks and wafers were inspected manually with a microscope. Manual optical inspection enabled identification of a wide range of defect types on a variety of process steps but was extremely slow and strongly dependent on the operator. Manual inspection evolved into automatic inspection employing high resolution CCD imaging system performing image capturing on two similar pattern zones and image comparison.
Other approaches to inspecting photomasks include U.S. Pat. No. 4,641,353 to Kobayashi on Feb. 3, 1987, which teaches projecting an optical image of a photomask pattern on to an image sensor. The image sensor converts the image and compares it to design data. The inspection is done just prior to the process of exposing the mask pattern on the wafer.
Conventional inspection methods, however, have difficulties in detecting all potential defects. The current mask defect inspection systems, such as KLA 351, Orbot RT-8000, and Lasertec normally consist of a high magnification and high resolution imaging system, and the photomask is scanned pixel by pixel. However, even when passing incoming inspection using these types of inspection techniques, certain types of defects were continuing to print on the wafer. Locating these types of defects then involves the time consuming process of finding the defect on the wafer to determine where the defect on the mask is located.
Other particle inspection systems, such as QCO and Horiba, rely on the detection of the scattered light from the defect. The particle inspection systems are not sensitive to certain defects, such as cleaning stains or local resist misprocessing. These defects do not scatter enough light to be detected by these inspection systems. Thus, the defects are undetected by these inspection systems.
After the automatic inspection of the photomasks, the operator has to evaluate all the defects found by the automatic inspection and classify whether the defect is false or real. The current inspection systems are capable of finding some defects, such as local small CD variation, shown in
FIG. 1
or a stripe butting error (not shown). However, when an operator views the image in
FIG. 2
, the mask is inaccurately flagged defect-free since the defect is not readily apparent to humans. As a result, the defect is undetected and will print on the wafer during the photolithography process.
Certain defects remain undetected by conventional inspection methods, or are inaccurately flagged as defect-free. Thus, what is needed is an improved method of mask inspection process for detecting printable defects which are difficult to accurately detect using current inspection techniques.
SUMMARY OF THE INVENTION
One aspect of the present invention is to provide a method of photomask inspection which uses available technology in a novel fashion to detect defects on a photomask. Although this invention may detect other types of defects, the invention is particularly useful for inspecting for printable defects including local small CD variation, stripe batting error caused by the e-beam writing tool, cleaning stains, local resist misprocessing, clear extension, chrome extension, repairs, and other defects.
In one embodiment, a photomask is inspected using a modified microscope, image comparison software, and a CCD camera. The CCD camera captures an image of the photomask out of focus at a low magnification on the microscope. A reference image is generated, such as from another die or from a database containing original design data. The image of the photomask is then compared to the reference image by the image comparison software.
In another embodiment, the mask is scanned at multiple focuses to implement the inspection effectively. This image is then compared with a reference image, such as an image from another die or a database. Any discrepancies between the images indicate a defect in the photomask. In yet another embodiment, a photomask is inspected using an optical microscope at low magnification in a dark field mode, and compared to a reference image.


REFERENCES:
patent: 3909602 (1975-09-01), Micka
patent: 4628531 (1986-12-01), Okamoto et al.
patent: 4641353 (1987-02-01), Kobayashi
patent: 4659220 (1987-04-01), Bronte et al.
patent: 5038048 (1991-08-01), Maeda et al.
patent: 5790247 (1998-08-01), Henley et al.
Schwart, J., “Process and Machine Mastering Employing WF-710 Wafer Inspection System”, Metrology, Inspection, and Process Control for Microlithography X, Santa Clara, CA: SPIE Proceedings Series, vol. 2725, 242-54, (Mar. 11-13, 1996).

LandOfFree

Say what you really think

Search LandOfFree.com for the USA inventors and patents. Rate them and share your experience with other people.

Rating

Inspection method and apparatus for detecting defects on... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Inspection method and apparatus for detecting defects on..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Inspection method and apparatus for detecting defects on... will most certainly appreciate the feedback.

Rate now

     

Profile ID: LFUS-PAI-O-2567480

  Search
All data on this website is collected from public sources. Our data reflects the most accurate information available at the time of publication.