Radiant energy – Photocells; circuits and apparatus – Photocell controls its own optical systems
Reexamination Certificate
1999-07-26
2002-02-26
Bowers, Charles (Department: 2813)
Radiant energy
Photocells; circuits and apparatus
Photocell controls its own optical systems
C250S201900, C250S203700, C348S308000
Reexamination Certificate
active
06350977
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to an improved pattern distortion detecting method and apparatus that are applied to pattern formation in a semiconductor manufacturing process, as well as to a recording medium on which a program to be used for performing such pattern distortion detection by using a computer is recorded.
2. Background Art
At present, the design rules of semiconductor devices have reached the 0.15-&mgr;m level and this value is smaller than light source wavelengths (0.248 &mgr;m in the case of using an excimer laser) of steppers for transfer of such patterns. Since the resolution performance deteriorates significantly in this circumstance, it is attempted to improve the resolution performance by using a special transfer technique such as a modified illumination technique.
Where such a special transfer technique is used, the pattern fidelity deteriorates though the resolution performance is improved. Also in other processes such as an etching process, variations in pattern dimensions may occur owing to differences in the density of miniaturized patterns.
To solve the above problems, the OPC (Optical Proximity Effect Correction) is widely employed which modifies design layout patterns so as to obtain desired patterns. It is assumed here that the OPC means the method to correct pattern distortions in general that result from not only optics-related causes but also process-related ones.
The OPC is classified into three kinds of methods. The first method is a model-based OPC in which patterns are modified based on simulation results. The second method is a rule-based OPC in which rules (OPC rules) according to which to modify design layout patterns are set in advance in consideration of geometrical features of the design layout patterns such as widths of respective patterns, distances between adjacent patterns, and distances from corner portions, and the design layout patterns are modified according to those rules. The third method is a combination of the first and second methods.
The pattern distortion detection for checking whether desired finish patterns will be obtained correctly by using OPC results is important.
In all of the above processes except the rule-based OPC, it is necessary to perform a simulation by using, as an input, design layout patterns or post-OPC patterns.
FIG. 20
shows the configuration of a conventional pattern distortion detecting apparatus. In
FIG. 20
, a layout pattern holding section
1
stores layout patterns to be used for prediction of pattern distortions. An on-edge sampling point generation rule holding section
7
holds rules according to which to generate sampling points on pattern edges of layout patterns. An on-edge sampling point generating means
2
generates sampling points on pattern edges. A pattern distortion predicting means
3
predicts pattern distortion amounts by performing simulations for the respective generated sampling points. A pattern distortion correcting means
4
corrects the layout patterns in accordance with the predicted pattern distortion amounts. Processing results are output to a pattern distortion information holding section
5
and/or a corrected layout pattern holding section
6
.
When pattern distortion detection is performed by using the above conventional pattern distortion detecting apparatus, the processes from the layout pattern holding section
1
to the pattern distortion predicting means
3
are executed and, as a result, pattern distortion information is obtained in the pattern distortion information holding section
5
. When a pattern distortion correction is performed, the processes from the layout pattern holding section
1
to the pattern distortion correcting means
4
are executed and, as a result, corrected layout patterns are obtained in the corrected layout pattern holding section
6
.
Next, the operation of the conventional pattern distortion detecting apparatus, that is, a conventional pattern distortion detecting method, will be described.
FIG. 21
is a flowchart of the conventional pattern distortion detecting method.
Steps
21
-
23
shown in
FIG. 21
are executed for a pattern distortion check and steps
21
-
22
and
24
-
26
are executed for a pattern distortion correction. All the steps are executed when both are needed.
First, at step
21
shown in
FIG. 21
, sampling points are generated on the edge portion of a layout pattern stored in the layout pattern holding section
1
according to on-edge sampling point generation rules that are stored in the on-edge sampling point generation rule holding section
7
shown in FIG.
20
. The sampling point generation rules will be described later.
At step
22
, a pattern to be generated from the layout pattern is simulated for the respective sampling points and pattern distortion amounts are thereby calculated. At step
23
, pattern distortion information is output to the pattern distortion information holding section
5
shown in FIG.
20
.
At step
24
, the pattern is modified so as to cancel out the pattern distortion amounts calculated at step
22
. The most typical method is to move the pattern edge in the direction opposite to the direction of the pattern distortion by the pattern distortion amounts. (There is a method in which after the pattern edge is moved by a predetermined amount at the first attempt, corrections and checks (the checking method will be described below) are repeated.)
At step
25
, to detect whether the correction has been made correctly, it is judged whether to re-execute steps
22
-
24
. Usually, a judgment criterion is set such that the process is to proceed to step
26
if the number of iterations of steps
22
-
24
has exceeded a predetermined number or the pattern distortion amounts have fallen within a specified value.
At step
26
, a corrected pattern is output to the corrected layout pattern holding section
6
shown in FIG.
20
.
The conventional pattern distortion check and correction are performed according to the above process.
A specific example of a conventional sampling point generation method will be described below. The sampling point generation is important in the present invention.
FIG. 22
shows an example in which sampling points are set all over an input layout pattern to be simulated and simulations are performed for the respective sampling points. In
FIG. 22
, a layout pattern
1
is hatched and sampling points
3
are indicated by mark “x.” Reference symbol
1
a
denotes pattern edges. This method has a problem of a very long processing time because calculations need to be performed for an enormous number of sampling points.
In view of the above, as shown in
FIG. 23
, it is a common practice to set sampling points
3
only on the pattern edges
1
a
that greatly affect the accuracy of pattern dimensions, whereby the number of sampling points is reduced and the calculation time is shortened.
FIG. 24
shows sampling points that are set in a case where a plurality of layout patterns exist. In
FIG. 24
, reference symbol
1
denotes a layout pattern;
1
a,
its pattern edges;
3
, sampling points;
7
, a layout pattern in another layer; and
8
, another layout pattern in the same layer as the layout pattern
1
.
In this manner, it is attempted to further reduce the number of sampling points and increase the processing speed by setting the sampling points
3
on the pattern edges
1
a
selectively in accordance with presence/absence of the adjacent layout pattern
8
and the layout pattern
7
in another layer and conditions relating to corners etc. and causing simulation results of the sampling points
3
to represent values of the entire edges.
However, the above conventional method cannot detect and correct a pattern distortion in a portion distant from a pattern edge that occurs, for example, in a halftone mask or in the case of an abnormally large pattern distortion.
FIG. 25
shows a specific example ofa halftone mask (described later in detail). In
FIG. 25
, reference numeral
1
denotes layout patterns and
9
denotes a
Bowers Charles
McDermott & Will & Emery
Mitsubishi Denki & Kabushiki Kaisha
Schillinger Laura M
LandOfFree
Pattern distortion detecting method and apparatus and... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Pattern distortion detecting method and apparatus and..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Pattern distortion detecting method and apparatus and... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2972871