The present invention relates to a method and an apparatus of reviewing defects for supporting a condition decision and an apparatus performance check of a review apparatus which observes defects obtained by an external inspection apparatus which detects foreign matters, pattern defects, and the like on the surface of a semiconductor wafer, a photo mask, a magnetic disc, a liquid crystal substrate, and the like.
In a semiconductor manufacturing process, a foreign matter or a pattern defect on the wafer surface causes a defective product. Accordingly, it is necessary to constantly quantify foreign matters, pattern defects, and the like (hereinafter, referred to simply as defects) and monitor whether the production apparatus and the production environment have no problem. Furthermore, by reviewing a defect shape, it is necessary to check whether the defect gives a fatal affect to the product.
Conventionally, such a review work is performed visual inspection by a person. For this, there has been a problem that depending on the person who performs review, the defect position and the defect type of the object to be reviewed may be varied and the defect to be reviewed cannot be constant. Recently, in order to solve these problems, the image processing technique is used to automatically judge the defect size, the defect type, and the like, such as the automatic defect review (ADR) technique and the automatic defect classification (ADC) technique are used.
For example, JP-A-10-135288 discloses a system for effectively observing, i.e., reviewing a pattern formed on an inspected part such as a wafer by using a scanning electron microscope (SEM) type review apparatus while reducing a load on an operator of the apparatus.
These years, the semiconductor device processing dimensions are reduced and defect sizes are also reduced. Accordingly, it is necessary to change the inspection condition of the inspection apparatus for extracting defects and output a plurality of defects extracted under respective conditions all at once. Moreover, as the inspection apparatus sensitivity is increased, the output noise of the inspection apparatus is increased and there is a case that the number of defects detected in one inspection exceeds several tens of thousands. For removing the noise, there is known a method for classifying the defects being inspected by the real-time defect classification (RDC) function and removing the noise. For example, JP-A-2001-156141 discloses a technique for easily performing defect analysis by deciding the defect detection condition in the inspection apparatus and the condition when using the real-time defect classification (RDC) function for removing the noise. That is, as much as information outputted from the inspection apparatus are ordered according to the defect detection condition and the real-time defect classification (RDC) function use condition. Moreover, the technique orders the defect identification number and coordinate information outputted from the review apparatus as well as automatic defect review (ADR) information and automatic defect classification (ADC) outputted from the review apparatus.
As has been described above, the work to detect external defects and attached foreign matters (hereinafter, referred to simply as defects) is very important so as to improve the yield. On the other hand, as the semiconductor device size is reduced, the inspection apparatus should have ability/performance to detect more minute defects and an inspection apparatus capable of inspecting defects with a high sensitivity is now used. As the sensitivity is increased, the number of detected defects becomes enormous and a plenty of time is required.
Moreover, the semiconductor manufacturing step uses, i.e., manages and operates several inspection apparatuses having such a high sensitivity. Since inspection of the same step is performed by using a plurality of inspection apparatuses, even if the apparatus types are identical, the defect detection sensitivity may vary depending on the apparatuses. For this, the number of defects and the defects size differs and the apparatus management requires a great care by an operator. Currently, the data processing is performed by matching data one by one, which requires a complicated manual operation.
Concerning the aforementioned technique, the inventor suggests a technique for supporting a worker by data processing through information delivery between an external view inspection apparatus and a review apparatus and displaying on the screen, a defect map expressing defect distribution and an ADR image of defects.
As has been described above, defects detected by an inspection apparatus includes detection of noise and in order to remove the noise, feedback should be performed to the inspection condition setting, which increases the information amount and makes it difficult to accurately decide the inspection condition. For this, an enormous time is required for setting the inspection condition.
Moreover, the aforementioned data has a problem that the number of detected foreign matters and defects is increasing and the characteristic amount is increasing, which requires a plenty of time for data processing and data ordering. In particular, an enormous amount of image data is outputted from the review apparatus as the processing speed of the apparatus is increased and it is becoming more and more difficult to process the automatically outputted images.
The aforementioned JP-A-2006-173589 by the inventor displays a defect map expressing a defect distribution state and an ADR image of defects so as to support the operator. However, as the information amount further increases in the future, it is necessary to provide a method and apparatus capable of sufficiently supporting the operator.
It is therefore an object of the present invention to provide a defect review method and apparatus having the function to rapidly search a hint to find a cause by improving the operability and user-friendliness.
An aspect of the present invention provides a defect review method using a review apparatus for reviewing defects of an object to be inspected according to information obtained by an external view inspection apparatus which inspects an external view of the object, the method comprising: a step of displaying on a screen, a defect map explicitly indicating existence of defects in an inspection region of the object; a step of displaying on the screen, a defect image list of defect image display planes prepared in correspondence to a plenty of defects in the defect map together with the defect map; a step of receiving an operation input signal for specifying an arbitrary defect in the defect map; a step of receiving an operation signal for specifying a display plane corresponding to an arbitrary defect in the defect image list; an image distinguishing display step performed upon reception of a signal specifying an arbitrary defect in the defect map, for displaying a display plane corresponding to the specified defect in the defect image list in such a way that it is distinguished from the other defect display screen; and a map defect distinguishing display step performed upon reception of a signal specifying an arbitrary defect in the defect image list, for displaying the specified defect on the defect map in such a manner that it is distinguished from the other defects. In a preferred embodiment of the present invention, for the defects specified by the defect map side, an unseen display plane corresponding to the defects specified out of the display range of the defect image list screen is shifted into the display range of the screen. Moreover, according to another preferred embodiment of the present invention, in still another embodiment of the present invention, for the defects specified by the defect image list screen side, the corresponding defects are blinked when displayed on the defect map. According to a yet another embodiment of the present invention, the display plane of the defect image list displays an image of corresponding defects obtained by the review apparatus.
According to the present invention, it is possible to effectively support a work of a review worker by improving the display of the enormous images and the defect map. According to a preferred embodiment of the present invention, it is possible to check an enormous amount of defect images and check whether a desired defect has been detected and simplify the procedure for optimizing the check condition. Moreover, it is possible to significantly reduce the time and labor required for detecting a defect of importance (DOI) and optimizing the check condition.
Other objects, features and advantages of the invention will become apparent from the following description of the embodiments of the invention taken in conjunction with the accompanying drawings.
Explanation will be given on the entire configuration of the present invention with reference to
Here, description will now be directed to
Firstly, the maximum gray level difference (1) is an absolute value of brightness of a defective portion when a differential image is obtained by performing image processing on a image of a portion which has been judged to be a defect and an image of its reference portion. The reference image average gray level (2) is an average value of brightness of the pixel portion judged to be a defective portion on the reference image. The defect image average gray level (3) is an average value of brightness of the pixel portion judged to be a defective portion on the defect image. The polarity (4) indicates whether the defective portion is brighter or darker than the reference image: “+” indicates a bright defect while “−” indicates a dark defect. The inspection mode (5) is an image comparison method used when the defect was detected: die comparison, cell comparison, or a mixture of them. The defect size (6), the number of defective pixels (7), the defect size width (8), the defect size height (9) show the size of the detected defect; the unit of the defect size, the width, and the height is a micron and the unit of the number of defective pixels is a pixel. The defect size ratio (10) indicates the width/height ratio of the defect size using parameters such as 1 if the width is identical to the height and 2 if the width is twice as much as the height. Lastly, the defective portion pixel differentiation value (11), (12) represents a differential value of the pixel portion judged to be defective on the defect image or the reference image and indicates the degree of the change of the shading. The value of the defect image portion is called a defective portion pixel differential value in defect image (11) and that of the reference image portion is called a defective portion pixel differential value in the reference image (12).
Here, returning to
Since the inspection apparatus 13 outputs enormous data as the defect information 23, only defect information 24, 25 extracted by the data processing apparatus 15 by using a plurality of filter functions are sent via a communication line 16 to the optical review apparatus 21 and the SEM-type review apparatus 22. The format of the defect information 24, 25 is generally identical to that of the defect information 23.
According to the extracted defect information 24, 25, an image of the defect detection unit is acquired by the optical review apparatus 21 and the SEM-type review apparatus 22. By using the image, the automatic defect classification (ADC) mounted on each review apparatus is used to classify the defects. The information is sent as the automatic defect review (ADR)/automatic defect class (ADC) information 26, 27 via the communication line 16 to the data processing apparatus 15.
The screen 50 shown in
Next, referring to
On the other hand, by clicking an arbitrary defect image in the image list 720, it is possible to display the corresponding defect by the blinking dot in the map 710. In the selection of a dot and an image, if the selection is performed while pressing the shift key, it is possible to select a plurality of defects.
When an arbitrary defect is selected by a defect identifier input column 701, the position of the defect is indicated by the blinking dot on the map 710 and a photograph (image) of the defect is explicitly indicated in the image list 720. Moreover, when an image in the image list 720 is selected by click, an arbitrary class (numeric) is inputted in column 702 of the classification input (Class# Input), and the return key (not depicted) is pressed, it is possible to input the class number corresponding to the defect. When the check mark in the map-on window 703 is removed, the map 710 disappears from the screen and as shown in
Referring back to the basic screen of
Secondary, when an arbitrary defect image shown in
According to the present embodiment, it is possible to provide a defect review apparatus capable of acquiring data outputted from an external view inspection apparatus into the data processing apparatus 15 and easily reviewing a large amount of image data by improving the image data display and operation units.
It should be further understood by those skilled in the art that although the foregoing description has been made on embodiments of the invention, the invention is not limited thereto and various changes and modifications may be made without departing from the spirit of the invention and the scope of the appended claims.
Number | Date | Country | Kind |
---|---|---|---|
2006-316903 | Nov 2006 | JP | national |