System and method for a semiconductor lithographic process control using statistical information in defect identification
A system and method is described for evaluating a wafer fabrication process for forming patterns on a wafer based upon data. Multiple inspection regions are defined on the wafer for analysis. For each inspection region, images of patterns within the inspection region are captured, edges are detected, and lines are registered to lines of a reference pattern automatically generated from the design data. Line widths are determined from the edges. Measured line widths are analyzed to provide statistics and feedback information regarding the fabrication process. In particular embodiments defects are identified as where measured line widths lie outside boundaries determined from the statistics. In particular embodiments, lines of different drawn width and/or orientation are grouped and analyzed separately. Measured line widths may also be grouped for analysis according to geometry such as shape or proximity to other shapes in the inspection region to provide feedback for optical proximity correction rules.
1. A method of obtaining feedback information, which relates to fabrication of patterns on a wafer based upon fabrication data, the fabrication data derived from design data, comprising:
capturing for at least one inspection region at least one image of patterns within the inspection region using an imaging device;
automatically generating from data selected from the fabrication data and the design data, a reference pattern for the inspection regions, the reference pattern comprising a machine-readable representation of at least one of (a) one or more line segments and (b) one or more curves;
detecting edges within each of the images;
determining, from the detected edges, one or more measurements for each of the selected patterns; and
processing the measurements to form statistical information and obtain the feedback information;
wherein the measurements are grouped into a plurality of groups for statistical analysis by using at least one of
(i) line widths of patterns in the data,
(ii) orientations of patterns in the data,
(iii) length of patterns in the data,
(iv) space between patterns and adjacent patterns in the data;
until each distribution of the grouped measurements becomes a normal distribution, and wherein the feedback information is feedback information for an Optical Proximity Correction (OPC) rule.
2. The method of claim 1 , further comprising determining defects by comparing the measurements to limits derived from statistical information derived from the measurements and identifying outliers.
3. The method of claim 2 , wherein the statistical information comprises a mean and a standard deviation of measurements, and wherein outliers are identified as measurements that are at least a predetermined number of standard deviations from the mean.
4. A method of obtaining feedback information, which relates to fabrication of patterns on a wafer based upon fabrication data, the fabrication data derived from design data, comprising:
capturing, for at least one inspection region, at least one image of patterns within the inspection region using an imaging device;
automatically generating from data selected from the fabrication data and the design data, a reference pattern for the inspection region, the reference pattern comprising a machine-readable representation of at least one of (a) one or more line segments and (b) one or more curves;
detecting edges within each of the images;
determining, from the detected edges, one or more measurements for each of the selected patterns; and
processing the measurements to form statistical information and obtain the feedback information;
wherein the selected patterns are determined by using coordinate values of a cell coordinate system within hierarchical cell data in the data,
wherein the feedback information is subdivided into
(i) feedback information for an OPC rule,
(ii) information that the defect is caused by a pattern defect of a mask pattern or a particle on the mask, and
(iii) information that the defect is a random defect on the wafer.
5. A method of extracting a set of patterns that are sensitive to process condition on a wafer based upon fabrication data, comprising:
capturing, for an inspection region, at least one image of patterns within the inspection region using an imaging device;
automatically generating, from data having a same geometrical feature type, a reference pattern for the inspection region, the reference pattern comprising a machine-readable representation of at least one of (a) one or more line segments and (b) one or more curves;
detecting edges within each of the images;
determining, from the edges, one or more measurements for each of the selected patterns;
processing the measurements to form statistical information;
processing the measurements of patterns against the statistical information in order to identify the set of patterns;
wherein the geometrical feature type includes at least one of
(i) a cell name of design data and
(ii) a cell name of mask data.
6. The method of claim 5 , wherein the statistics include at least one of
(i) averages,
(ii) standard deviations,
(iii) maximum values, and
(iv) minimum values.
7. The method of claim 5 , further comprising controlling at least one process condition using the set of extracted patterns.
8. A system for obtaining feedback information relating to patterns on a wafer fabricated based upon fabrication data derived from design data, the system comprising:
an imaging device for providing a digital image of at least one inspection region of the wafer; and
a processing system further comprising at least one processor, and at least one memory, the processing system coupled to receive the digital image from the imaging device into the memory, the memory containing at least a portion of a reference pattern for the inspection region;
wherein the memory of the processing system contains machine readable instructions for:
locating at least one inspection region on the wafer for analysis;
selecting patterns within each inspection region;
capturing into the memory at least one image of patterns within the inspection region from the imaging device;
determining, from detected edges in the images, one or more measurements for each of the selected patterns; and
processing the measurements against the statistical information to identify defect locations; and
obtaining feedback information;
wherein the measurements are grouped into a plurality of groups for statistical analysis by using at least one of
(i) line widths of patterns in the data,
(ii) orientations of patterns in the data,
(iii) length of patterns in the data,
(iv) space between patterns and adjacent patterns in the data;
until each distribution of the grouped measurements becomes a normal distribution, and wherein the feedback information is feedback information for an Optical Proximity Correction (OPC) rule.
9. The system of claim 8 wherein the statistical information is derived for a same location relative to a die origin across multiple die on the wafer.
10. The system of claim 8 wherein the statistical information is derived for a group of locations in the same inspection region on the wafer.
11. The system of claim 8 , wherein the machine readable instructions for determining the measurements comprise instructions for determining at least one measurement selected from the group consisting of
(i) a line width of a measured line,
(ii) a line-end shrinkage of a measured line, and
(iii) a space between a measured line and adjacent shapes in the inspection region.
12. The system of claim 8 , wherein defect locations are identified by comparing the measurements to limits derived from statistical information derived from the measurements and identifying outliers.
13. The system of claim 12 , wherein the statistical information comprises a mean and a standard deviation of measurements, and wherein outliers are identified as measurements that are at least a predetermined number of standard deviations from the mean.
14. The system of claim 8 , wherein the machine readable instructions further comprise instructions for grouping the measurements into a plurality of groups, by at least one geometric feature selected from the group consisting of:
(i) line widths of patterns in the design data,
(ii) orientations of patterns in the design data,
(iii) lengths of patterns in the design data, and
(iv) space distance between patterns in the design data and adjacent patterns;
and the machine readable instructions comprise instructions for determining statistical information for each group of the grouped measurements.
15. An apparatus for obtaining feedback information relating to patterns on a wafer fabricated based upon fabrication data derived from design data, the system comprising:
an imaging device for providing a digital image of at least one inspection region of the wafer;
a device for outputting the feedback information, and
an image processing device further comprising at least one processor, and at least one memory, the image processing device coupled to receive the digital image from the imaging device into the memory, the memory containing at least a portion of a reference pattern for the inspection region, the reference pattern being derived automatically from data selected from the group consisting of design data and fabrication data; wherein the memory of the processing system contains machine readable instructions for:
locating at least one inspection region on the wafer for analysis;
selecting patterns within each inspection region;
capturing from the imaging device into the memory at least one image of patterns within the inspection region;
determining from detected edges in the images one or more measurements for each of the selected patterns; and
processing the measurements against the statistical information to identify defect locations; and
obtaining feedback information,
wherein defect locations are identified by comparing the measurements to limits derived from statistical information derived from the measurements and identifying outliers; and
the statistical information comprises a mean and a standard deviation of measurements, and wherein outliers are identified as measurements that are at least a predetermined number of standard deviations from the mean.
16. The apparatus of claim 15 wherein the statistical information is derived from similar locations in a plurality of die from a plurality of shots on the wafer, and from multiple inspection regions on the wafer.
17. The apparatus of claim 15 wherein the statistical information is derived from multiple locations within an inspection region.