Optics: measuring and testing – By alignment in lateral direction – With light detector
Reexamination Certificate
2001-01-18
2004-02-17
Lee, Diane I. (Department: 2876)
Optics: measuring and testing
By alignment in lateral direction
With light detector
Reexamination Certificate
active
06693713
ABSTRACT:
TECHNICAL FIELD
The present invention relates to a mark detection method, an exposure method, a device manufacturing method, a mark detection apparatus, and an exposure apparatus. More particularly, the present invention relates to a mark detection method for detecting a mark for position measurement that is formed on an object such as a semiconductor substrate or a liquid crystal display device, an exposure method for transferring a predetermined pattern on a substrate aligned by the use of the mark detection method, a device manufacturing method using the exposure method, a mark detection apparatus for detecting a mark for position measurement which is formed on an object such as a semiconductor substrate or a liquid crystal display device, an exposure apparatus for transferring a predetermined pattern onto a substrate aligned by the mark detection apparatus, and a device manufactured by the exposure apparatus.
BACKGROUND ART
In manufacturing a semiconductor device and a liquid crystal display device, a variety of planar techniques are utilized. In the planar techniques, a finely patterned image formed on a photomask and a reticle (hereinafter, referred to as reticle) by the use of an exposure apparatus is projected and exposed on a substrate such as a semiconductor wafer or glass plate on which a photosensitive agent such as photoresist is coated (hereinafter, referred to as wafer).
The reticle pattern is projected and exposed by the use of, for example, an exposure apparatus of a step-and-repeat system in such a manner that a position of the reticle and a position of the wafer are adjusted (aligned) with high accuracy and the reticle pattern is superposed on a pattern already formed on the wafer.
Particularly in recent years, high densification has been required for semiconductor circuits. Accordingly, also in the alignment of the exposure apparatus, as the pattern of the semiconductor circuit or the like becomes finer, a demand for an alignment performed with higher accuracy has increased, and various processes for alignment have been made.
In general, the alignment of the reticle is performed using exposure light.
Among alignment systems for the reticle, there is a Visual Reticle Alignment (VRA) system or the like, in which an alignment mark drawn on a reticle is irradiated with exposure light, and image data of the alignment mark picked up by a CCD camera or the like is subjected to image processing, and a mark position is measured.
The following are types of alignment sensors for wafers.
(1) Laser Step Alignment (LSA)
This sensor is a sensor for irradiating alignment marks arranged as a line of dots on a wafer with a laser beam in order to detect a mark position by the use of light diffracted or scattered by the mark.
(2) Field Image Alignment (FIA)
This sensor is a sensor for irradiating alignment marks arranged as a line of dots with light having a large wavelength bandwidth using a halogen lamp or the like as a light source, and performing image processing of the image data of an alignment mark imaged by a CCD camera or the like in order to measure a mark position.
(3) Laser Interferometric Alignment (LIA)
This sensor is a sensor for irradiating alignment marks arranged in a diffraction grating pattern on a wafer from two directions using laser beams having slightly different frequencies and causing the two generated diffraction lights to interfere with each other in order to measure a position of the alignment mark from the phase obtained through the interference.
In the alignment by these optical systems, first, an alignment mark on the reticle is detected and processed to measure a position coordinate thereof. Next, an alignment mark on the wafer is detected and processed to measure a position coordinate thereof, thus position of a shot to be superposed is determined. Based on these results, the wafer is moved by a wafer stage to perform an alignment so that a pattern image of the reticle can be superposed on the shot position, and the pattern image of the reticle is projected and exposed on the wafer.
In some of the above-described alignment systems, processing is performed after a one-dimensional image or a two-dimensional image is obtained as an alignment signal.
For the case of a two-dimensional image, by adding mark portions in the measured direction, it can also be treated as a one-dimensional image.
These signals are originally signals that are continuously distributed with respect to the position, but for convenience of signal transmission of an image pickup device, the signals will be extracted as signals sampled at a predetermined interval. For example, when an image processing sensor, such as a CCD camera or a line sensor, is used as an image pickup device, since the pixel size is limited, the signals will be sampled at an interval determined by the pixel size. Ideally, it is desirable that signals output from the image pickup device be sampled by a sampling apparatus at an interval corresponding to the pixel size of the image pickup device.
Edge detection, a correlation method or the like is used for these sampled signals in order to measure the mark position.
Incidentally, in general, the accuracy required for the alignment sensor is extremely high in comparison to the minimum resolving unit of the image pickup device. For this reason, the position must be finally determined with an accuracy equal to or less than the sampled interval.
Heretofore, in edge detection and the correlation method, processing has been performed for the sampled signals, and when the final position result is calculated, the interval between the sampled points is fitted by an appropriate function such as a linear or a quadratic function, and by solving the function, a resolving power less than the sampled interval has been obtained. Typically, the finer the sampling interval, the more the accuracy is improved.
On the other hand, when the magnification of the optical system is increased in order to reduce the sampled interval on an object, the visual field is narrowed due to a limitation in the number of pixels of the CCD camera.
Considering the constitution of the apparatus, the visual field of the sensor must be ensured to some extent by conditions such as size of the alignment mark or the accuracy of the pre-alignment performed before the alignment measurement.
In addition, in order to prevent the conversion of a high-frequency component of a signal into a low-frequency component by sampling (aliasing), a necessary condition is that the minimum resolving unit of the image in the image pickup device is 0.5 times or less of the minimum periodic component.
The minimum periodic component of the signal is given by, for example, in the case of an image processing sensor using an optical microscope, the lower limit of the minimum periodic component of the image as P
min
as follows:
P
min
=&lgr;(2
×NA
)
&lgr;: wavelength of light
NA: NA of optical system
However, this value will also vary depending on the illumination conditions. By using this value, the sampling interval P
s
is given as:
P
s
<0.5
×P
min
and, the above-described conditions can be satisfied.
However, when the sampling interval P, is increased, the error in sampling when performing the edge measurement or correlation measurement becomes significantly worse before the sampling interval P
s
reaches 0.5×P
min
, that is, from about 0.2×P
min
.
FIG. 14
is a diagram for explaining the process during the execution of the edge detection.
In a typical edge detection algorithm, first, the point of maximum inclination, slope point is found. Typically, since the sampling interval is a fixed value, a point is obtained where a difference &Dgr;V between adjacent sampled points in the V direction in the drawing is at a maximum. In the example of
FIG. 14
, the point denoted by the reference symbol P
0
is the point of maximum inclination.
From this point, the closest relative maximum and minimum are found by hill-climbing and hill-descending. In the example shown in
FIG. 14
, with reference sym
Lee Diane I.
Nikon Corporation
Oliff & Berridg,e PLC
LandOfFree
Mark detection method, exposure method, device manufacturing... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Mark detection method, exposure method, device manufacturing..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Mark detection method, exposure method, device manufacturing... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3317521