Process conditions change monitoring systems that use...

Details Process conditions change monitoring systems that use... Process conditions change monitoring systems that use...

2003-01-22

2004-09-14

Wells, Nikita (Department: 2881)

Radiant energy

Inspection of solids or liquids by charged particles

Methods

C250S310000, C250S311000

Reexamination Certificate

active

06791082

ABSTRACT:

BACKGROUND OF THE INVENTION
The present invention relates to the systems and methods in which, during lithography, whether pattern exposure to the resist film on a wafer has been provided under the appropriate exposure conditions by use of electron beam images of the resist patterns. The invention relates particularly to the technology for controlling such an exposure process and maintaining the appropriate exposure conditions.
The flow of conventional lithography is described below.
The formation of a resist pattern is accomplished by coating a semiconductor wafer or a similar substrate with a resist (photosensitive material) to the required thickness, then exposing a mask pattern to light using an exposure unit, and conducting a developing process. The resist pattern that has thus been formed is dimensionally checked using a scanning-type electronic microscope provided with a length measuring function (this microscope is called “length-measuring SEM or CD-SEM”). An example of processing with conventional length-measuring SEM is described below. First after an electron beam image of the area which includes the section requiring stringent dimensional accuracy has been acquired in process
1
, dimensions are measured in process
2
, then whether the dimensions satisfy reference values is judged in process
3
, and if the reference values are not satisfied, the exposure level of the exposure unit is corrected in process
4
(the amount of correction of the exposure level is represented as &Dgr;E). For example, in the case of a positive type of resist, if the resist size is too large, the exposure level is increased, and if the resist width is too small, the exposure level is reduced. It is not rare that the amount of correction of the exposure level is determined in accordance with the experience and working knowledge of the operator.
FIG. 17
represents the relationship between a resist pattern and an after-etching film pattern (data source: “Handbook of Electronic Beam Testing”, p. 255, a research document cited at the 98th Study Session of the 132nd Committee on the Application of Charged Beams to Industries, held under the auspices of the Japan Society for the Promotion of Science). Given the same etching conditions, there is a relationship of invariableness between the shape of the resist pattern and that of the film pattern. To obtain a film pattern of the required shape, therefore, the resist pattern also needs to have the required shape. For example, during the comment of new processes, “conditions establishing operations” for identifying the focus and exposure level at which the required resist pattern shape can be obtained are performed by, after creating a wafer on which a pattern has been printed by changing the focus and the exposure level with each shot (unit of exposure) [an example of a wafer is shown in
FIG. 18
; such a wafer is usually called the focus exposure matrix (FEM)], measuring the dimensions of the resist pattern for each shot, then cutting the wafer, and examining its cross-sectional shape. A system for supporting the conditions establishing operations is set forth in Japanese Application Patent Laid-Open Publication No.Hei11-288879. These operations are performed to determine the exposure level (E0) and focus value (F0) at which greater margins can be obtained, and the product wafer undergoes exposure based on the corresponding conditions. However, changes in the photosensitivity of the resist, changes in the thickness of the reflection preventive film under the resist, drifts in the various sensors of the exposure unit, and various other changes in process conditions, may prevent the required resist pattern shape from being obtained under the E0 and F0 conditions that have been determined during the conditions establishing operations. Dimensional measurement (process
2
) described above takes place to detect these changes in process conditions, and the prior art described above is intended to compensate for changes in resist shape, caused by changes in process conditions, through correcting the exposure level.
SUMMARY OF THE INVENTION
Under the prior art, the line width and other dimension values are examined using length-measuring SEM to detect changes in process conditions and undertake corrective measures, and if the dimension values do not satisfy reference values, the exposure level is corrected. The prior art, however, poses the following three problems:
The first problem is that changes in process conditions, not associated with any changes in the dimension values, more specifically, changes in the focus value during exposure cannot be detected. The resist has an approximately trapezoidal cross-sectional shape. Since inclined portions are greater than flat portions in terms of secondary electron signal intensity, the signal waveform peaks at the portion corresponding to the edge of the trapezoid as shown in FIG.
19
(
a
). An example of dimensional measurement with length-measuring SEM is described below. As shown in FIG.
19
(
b
), a straight line is drawn along both the outer portion and base portion of the peak, then the crossing point of the two lines is derived, and after the same has also been performed on the other side, the distance between the two crossing points is taken as the line width.
FIG. 20
is a graph on which the line width was plotted for each exposure level (from “e0” to “e8”) with the focus value plotted along the horizontal axis in order to represent how the line width would change when the exposure level and the focus value changed. The magnitude of the exposure level increases in the order from “e0” to “e8”, and there is the relationship that the line width decreases with increases in the exposure level (this relationship applies to a positive resist, and the opposite relationship is established for a negative resist). Changes in the exposure level can therefore be detected by examining the line width. However, as is obvious from the graph, changes in the line width are not too significant with respect to those of the focus value, and near the appropriate exposure level of “e4”, in particular, even if the focus value changes, the line width suffers almost no changes. Changes in the focus value, therefore, cannot be detected by examining the line width. On the other hand, even if the line width does not change, when the focus value changes, the cross-sectional shape of the resist will change as shown in FIG.
20
(
b
). Since, as described earlier in this document, changes in the cross-sectional shape also affects the shape of the film pattern existing after etching, the use of the prior art which does not enable changes in the focus value to be detected is likely to create large quantities of defects in the shape of the film pattern existing after etching.
The second problem is that deviations in focus value cannot, of course, be accommodated by merely correcting the exposure level only. For example, for situation A shown in FIG.
20
(
a
), since the line width is greater than its normal value, the exposure level will be increased according to line width measurement results. However, since the deviation in focus value must be corrected, situation B shown in FIG.
20
(
b
) will only result and the cross-sectional shape of the resist will not return to normal. Consequently, defects in the shape of the film pattern existing after etching are likely to be created in great quantities in this case as well.
The third problem is that such quantitative information on process conditions changes that is required for the maintenance of a normal exposure process cannot be obtained with the above-described prior art. The tolerances for the exposure level and focus value are being narrowed very significantly with the decreases in pattern rule in recent years. For example, for a semiconductor pattern whose design rule is 180 nm, the rate of change of pattern size is required to be controlled below 10%, and to implement this, it is necessary to acquire information that quantitatively represents changes in process conditio

Affiliated with

Also associated with

Rating

Process conditions change monitoring systems that use... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Process conditions change monitoring systems that use..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Process conditions change monitoring systems that use... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3218646