JP2000077496A - Electronic device inspection system and method for manufacturing electronic device using the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To shorten analysis time and increase analysis accuracy by arranging and providing a vast amount of defect inspection data ready for use. SOLUTION: Inspection processes 31, 32, etc., and 3N are provided for every prescribed manufacturing processes 21, 22, etc., and 2N of a wafer 1, and a historical defect list 8 is created in an analysis unit 6 by each time inspection results 51, 52, etc., and 5N are obtained from the respective inspection processes 31, 32, etc., and 3N. The historical defect list 8 is created for each wafer 1 and comprises a coordinates position, the number of detects, size, a cluster information, an index information of an image capturing, and the like of each defects for every inspection processes 31, 32, etc., and 3N, and every information in relation to the defects of the wafer. The use of the historical defect list enables to obtain accurately within a short period of time, a selection of each manufacturing process with defects, a variation of defects in the manufacturing process, and the analysis result of a growth process and the like.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electronic device inspection system applied to a production line of electronic devices such as semiconductors and TFT liquid crystal displays, and a method of manufacturing an electronic device using the inspection system. The present invention relates to an electronic device inspection system for shortening data analysis time and improving analysis accuracy, and a method for manufacturing an electronic device using the inspection system.

[0002]

2. Description of the Related Art Conventionally, electronic devices, for example, semiconductors, are formed by repeating a plurality of processing steps such as exposure, development, and etching on a wafer. On the other hand, the wafer processed in a predetermined processing step of the plurality of processing steps is inspected by a foreign substance inspection device or a visual inspection device if necessary, and the position and size of the foreign substance or the appearance defect attached to the wafer are determined. , The number, type, etc. are collected. Hereinafter, a foreign substance which is a detection target of the foreign substance inspection device and an appearance defect which is a detection target of the appearance inspection apparatus are collectively referred to as a defect.

[0003] Monthly “Semiconductor World” 1996.8 p
As described on pages 88, 99, and 102, an enormous amount of defect data is usually sent from an inspection device to an analysis system via a network, where it is managed and analyzed.

[0004]

However, in many of such conventional analysis systems, a user extracts data stored in a database at the time of analysis and performs a calculation process. Even in an analysis system that performs calculations in advance, since individual analysis processes are performed independently, when all pieces of information relating to defects on a wafer are freely combined and analyzed, processing time for search and calculation is separately required. . Therefore, the time required for preparation for analysis such as search / calculation becomes longer, and the feedback of the analyzed data to the manufacturing process tends to be longer.

Further, when a specific defect position is selected from the data obtained by the inspection apparatus and observed to obtain an image, the past defect information of the wafer cannot be easily extracted. It is impossible to select a newly generated defect or a defect that grows through a process. In the future, if the wafer diameter increases to 300φ, the amount of data will increase and the preparation time for analysis will further increase, and delays in feedback work will have a significant effect on improving the production line yield. Is done.

SUMMARY OF THE INVENTION An object of the present invention is to solve such a problem, to reduce the time required for analysis preparation, to improve the analysis time, and to improve the analysis accuracy by organizing a large amount of defect inspection data to make it easy to use. And a method for manufacturing an electronic device using the same.

[0007]

In order to achieve the above-mentioned object, the present invention relates to a method of producing inspection data stored in an analysis unit in advance on a production line of an electronic device, for each wafer, using defect coordinates and their attributes. A defect history list including cluster information and image acquisition index information as a result of determining a certain detection process, the number of detected defects, a defect size, a defect type, and the density of defects is collected, and one defect history list relates to defects on a wafer. All history information can be referenced.

As a result, the data search and calculation work at the time of data analysis have been set up, and the analysis preparation time has been greatly reduced.

Further, the defect history list can be transmitted to an inspection apparatus connected via a network.

This makes it possible for the inspection apparatus to remove all the defect data previously detected on the wafer from the current inspection data, thereby enabling an inspection focusing only on newly detected defects. For example, the number of defects newly generated in the processing step can be grasped on the spot where the inspection is performed, and abnormalities in the processing step are detected, and quick feedback to the processing step becomes possible.

[0011] Similarly, the defect history list can be transferred to a defect observation device connected via a network.

This makes it possible to observe a defect growing through a processing step or a defect newly generated in a new processing step, thereby enabling more effective observation and acquisition of an image in which the origin of the observed defect is clear. Was.

More specifically, the present invention comprises a plurality of processing steps for processing a work to be an electronic device, and a plurality of inspection apparatuses for inspecting each of the works processed in the different processing steps. An analysis unit connected to the plurality of inspection devices via a network and having at least a storage unit for storing results of inspection by the inspection devices, wherein inspection data stored by the analysis units is classified based on defect coordinates It is. Inspection data of a certain process is subjected to a cluster recognition process and a process recognition process. The cluster recognition process is a process of determining the density of defects. Further, a process recognizing process is performed in which the coordinates of the detected defect and the defect coordinates detected in the past are compared to recognize in which process the defect was detected. Then, the inspection process name and the defect size accompanying the processing result and the defect data are added to the defect history list. Therefore, the data amount of the defect history list of one wafer increases as the manufacturing process proceeds.

Further, the present invention is to transmit a defect history list created in advance by the analysis unit to a data analysis terminal, an inspection device or a defect observation device to give defect history information of a wafer to be analyzed.

[0015]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing one embodiment of an electronic device inspection system and an electronic device manufacturing method using the same according to the present invention, wherein 1a is an unfinished wafer, 1b is a completed wafer, 2 ₁ , 2 ₂ ,. _{..., 2 N, 2 N +} 1 manufacturing _{process, 3 1, 3 2, ......} , 3 N the inspection process, 4 probe test _{step, 5 1, 5 2, ......} , 5 N inspection results, 6 Is an analysis unit, 7 is a defect history list creation process, 8 is a defect history list, 9 is a defect history list management table, 10 is category data, 11 is an analysis device, 12 is an inspection device, 1
Reference numeral 3 denotes an image acquisition device.

In FIG. 1, unfinished wafers 1a are sequentially formed in manufacturing steps 2 ₁ , 2 ₂ ,..., 2 _N , 2 _{N + 1} (when these manufacturing steps are not specified, they are referred to as manufacturing step 2). Film, exposure,
A process such as etching is performed, and if necessary, a process such as ion implantation is performed to obtain a finished wafer 1b as a product. Then, the respective manufacturing processes 2 ₁ , 2 ₂ ,.
2 _N (these may include a plurality of sub-steps) is provided with an inspection step for performing various defect inspections such as a foreign substance inspection and an appearance inspection as necessary, and a wafer processed in the manufacturing process is provided. 1 is sent to this inspection process to perform a defect inspection process. In this case, the manufacturing process 2 ₁ , 2 ₂ ,.
..., 2 _N examined every step _{3 1, 3 2, ......,} 3 N ( In the following, these test steps 1, 2, and the fact that N) and what is provided, These are connected to the analysis unit 6 by a network,
2, ..., test results 5 _1, 5 ₂ N, ..., 5 _N (if no identify these test results, the test result of 5) such defect coordinate data is to be sent to the analysis unit 6.
The completed wafer 1b obtained by the above processing is
Further, in the probe inspection step 4, the quality of each chip is determined, and the category data indicating the determination result is also sent to the analysis unit 6 via the network.

In the analysis unit 6, the inspection result 5 from each of the inspection processes 1, 2,.
_{1, 5 2, ......, 5} N is processed are collected in the order of steps, the defect history list 8 for each wafer is created. Also,
A defect history list management table 9 for managing the defect history list 8 for each wafer 1 is also created. These are held in the analysis unit 6, and the category data 10 obtained in the probe inspection step 4 is also held in the analysis unit 6.

The analysis unit 6 is further provided with an analysis device 11 for performing analysis processing of inspection results using the defect history list 8 generated by the analysis unit 6 via a network, An inspection device 12 for performing an inspection using the defect history list 8 and an image acquiring device 13 for acquiring a defect image using the defect history list 8 generated by the analysis unit 6 are connected.

Here, the inspection apparatus 12 executes each of the inspection steps 1, 2,..., N, and compares the defect detection results with the inspection results 5 ₁ , 5 _{2 ,.} As well as obtaining a defect history list 8 from the analysis unit 6 and using this to process a defect detection result to evaluate an inspection or the like. Further, the image acquisition device 13 includes the inspection device 1
A defect is appropriately selected using the defect detection result obtained in step 2 and the defect history list 8 in the analysis unit 6, an image of the selected defect is captured, and the image is stored. Further, the analysis device 11 uses the defect history list 8, the defect history management table 9, and the category data 10 of the analysis unit 6 to search for a defect occurrence state caused by each manufacturing process or a defect that causes a defective product. An analysis process is performed.

FIG. 2 is a diagram showing a specific example of the defect history list 8 shown in FIG. The defect history list 8 is shown in FIG.
2 shows the history of defects occurring on one wafer 1 in the production line shown in FIG. Therefore, one defect history list 8 is generated for one wafer 1.

In FIG. 2, in the defect history list 8, first, identification information of the wafer 1, a defect history list creation processing parameter 204, and an inspection process name 205 are described at the beginning. Here, the wafer identification information is assumed to be composed of a type name 201, a lot number 202, and a wafer number 203 of the wafer 1. The defect history list creation processing parameter 204 is a parameter of a cluster recognition process and a process identification process described later. The inspection process name 205 is described in the order in which the manufacturing process name immediately before the wafer 1 is inspected is inspected as the inspection process name. Here, it is assumed that the wafer 1 has been inspected in the inspection steps 1, 2, and 3 subsequent to the manufacturing steps 1, 2, and 3 in FIG. , 3 are described.

Next, an item 206 of "defect coordinates" is provided on the left side of the defect history list 8, and all defect coordinate positions on the wafer 1 detected in the three inspection processes 1, 2, and 3 are provided therein. (For convenience, the order of defects is indicated by a number in parentheses). However, in consideration of the coordinate detection accuracy of the inspection apparatus, for the defect coordinates detected earlier, for example, the defect coordinates detected in a small radius area referred to as a comparison radius in the later process are the same coordinates. I consider it.
That is, in FIG. 1, a defect history list creation step 7 is a step in which the inspection step detects defects one by one and sequentially sends the inspection results 5 to the analysis unit 6.
Processes the inspection result 5 each time it receives the inspection result 5 and adds it to the defect history list 8. In FIG. 2, the “defect coordinate position” in this case is a coordinate position in the x, y coordinate system with the center position of the wafer 1 as the origin (0, 0). The left side of the item 206 is the x coordinate position (distance unit = μm), and the right side is the y coordinate position (distance unit = μm).
Is shown. Therefore, the coordinate position (x, y) of the defect detected first is (23115.70, 45894.32).

Next to the item 206 of "defect coordinates", there is provided an item 207 of "number of defects by process". This indicates how many of the defects have been detected in any of the inspection processes 1, 2, and 3 in which the “inspection process name” 205 is described. The inspection processes 1, 2, and 3 are shown in order from the left. "0" indicates that no defect was detected, "1" indicates that one defect was detected, and "2" indicates that two defects were detected.

For example, the first coordinate position (23115.7)
0, 45894.32), one defect is detected in the inspection steps 1 and 3, and is not detected in the inspection step 2. At the fourth coordinate position (10778.87, 627.435), one defect is detected in the inspection steps 1 and 2,
In the inspection process, two defects have been detected. In the sixth coordinate position (−18781.02, 20487.29),
One defect is detected only in the inspection process 1, and at the eleventh coordinate position (−917.65, −2617.23), one defect is detected only in the inspection process 2 and the seventeenth coordinate position ( 6938.41, 941.94), there are some defects that are detected only in one certain inspection step, as one defect is detected only in the inspection step 3. Further, as in the first defect described above, there is a defect in which the defect detected in the first inspection step 1 is not detected in the next inspection step but is detected in the next inspection step 3. . This can be seen from the item “number of defects by process” 207.

It should be noted that the above two defects are detected when
This means that two defects are detected at the same coordinate position. In this case, two extremely close defects are regarded as defects at one coordinate position. Other defects detected within a small radius area called a radius are regarded as defects having the same position coordinates as the already detected defects. The comparison radius is set in advance as one processing parameter corresponding to the detection sensitivity, and is set to, for example, about 250 μm.

Subsequent to the item “number of defects by process” 207, an item 208 of “defect appearance process” is provided. This is an 8-bit numerical value indicating which inspection step the defect to be detected is initially detected in each inspection step (indicated by a decimal number for easy understanding). ), “0” indicates the first detection in the first inspection step, “1” indicates the detection in the next inspection step 2, and “2” indicates the first detection in the next inspection step 3, respectively. The 8-bit maximum value ("255" in decimal) indicates that no defect is detected. Also in this case, inspection steps 1, 2, and 3 are shown in order from the left.

Therefore, for example, the first and fourth defects are detected in the detection step 3, but are described as "0", and thus are the defects detected in the first inspection step 1. You can see that there is. In addition, the defect detected in the inspection step 3 at the 16th coordinate position (−27068.01, −7597.72) is described as “1”, so that the defect is detected first in the inspection step 2. I understand. In this manner, by referring to the defect appearance process, it is possible to determine in which process of the manufacturing line the defect detected in a certain process has occurred.

Subsequent to the item 208 of the "defect appearance step", an item 209 of the "defect size" is provided.
This indicates the size of the detected defect, and the size is classified into classes, 1, 2, 3,.
It is represented by the number of ..., and the larger the number, the larger the defect.

Following the item 209 of “defect size”,
An item 210 of “defect type” is provided. The item 210 is a result of classification according to a predetermined category, and is indicated by defect classification numbers 1 to 9 in this embodiment.

Subsequent to the item 210 of “defect type”, an item 211 of “cluster” is provided. This item 21
Numeral 1 indicates the result of determining the density of defect coordinates.

The density of defect coordinates is determined, for example, by the following method. First, the wafer 1 is divided into two-dimensional minute rectangular areas, and a rectangular area having a defect is replaced with 1 and a rectangular area without a defect is replaced with 0.
Next, after expanding the image to determine the connectivity (for example, determining that the rectangular areas of the adjacent image 1 have connectivity), a group of connected rectangular areas (connected area) is obtained. If the number of defects exceeds the threshold value, it is determined that the density is high. Such a defect group determined to have high density is called a cluster defect.

In the "Cluster" item 211, a different identification number is assigned to each cluster defect, and the value is described for each defect coordinate. Here, the fifth, sixth, and seventh defects in the inspection step 1 constitute cluster defects, and these are assigned identification numbers of “1”. If there is another cluster defect, the identification number “2” is assigned to the defect constituting the cluster defect. In the inspection step 2, it has been determined that the twelfth and thirteenth defects constitute a cluster defect.

Subsequent to the item 211 of “cluster”, an item 212 of “image index” is provided. This item 212 indicates, for each inspection step, the index (numerical value) of the defect from which the image was obtained, and this index also indicates the number of the obtained image. Here, image acquisition refers to copying an image of a defect on the wafer 1 into a photograph or the like by the image acquisition device 13 in FIG. 1 and storing the image, and the index is also a storage number in this case.

Here, from the inspection result of the inspection step 1, 4
The image is obtained in the order of the fifth defect, the fifth defect, and the tenth defect. From the inspection result of the inspection process 2, the image is obtained in the order of the fourth defect, the twelfth defect, and the fourteenth defect. The image is obtained in the order of the fourth defect, the eighth defect, and the seventeenth defect from the inspection result of the inspection process 3. For the fourth defect, an image is obtained for each of the inspection steps 1, 2, and 3. Defects for which image acquisition is performed in this manner include, for example, those determined to be large in size in the item 209 of “defect size”. Among the defects detected in the first inspection process 1, , Noticeable defects (size, location, etc.)
Regarding the above, as in the case of the fourth defect, an image can be obtained in each inspection step.

FIG. 3 is a flowchart showing a specific example of the defect history list creation step 7 in the analysis unit 6.

In the figure, first, the analysis unit 6
When the inspection result 5 is received from the inspection process 3 (step 10)
0), this is read and the defect history list creating step 7 is entered (step 101). In this step, the wafer to be inspected 1
Based on the identification number, the defect history list 8 already created for the inspection wafer 1 is retrieved and acquired (step 102), and is read (step 103).

Next, the analysis unit 6 determines whether or not the coordinate position of the defect of the read inspection result 5 is registered in the read defect history list 8,
A process recognition process for recognizing whether the defect is newly obtained in the inspection process 3 or a defect detected in any of the past inspection processes 3 is performed (step 104). Perform processing (step 10
5). The defect history list 8 is updated with the new inspection result 3 based on the processing result and the defect size information based on the inspection result 5 (step 106).

[0038] Therefore, in FIG. 2, but shows a defect history list 8 that registered the inspection result to the inspection process 3 in Figure 1, the test results 5 ₄ of the following inspection process 4 (inspection step 3 ₄₎ When sent, the analysis unit 6 sets the items “defect number per process”, “defect appearance process”, “defect size”, “defect type”, “cluster”,
For "image index", the section of this inspection step 4 provided, defects located compare within a radius of already defect coordinates that are registered in the inspection results 5 ₄ as the same defect as the defect this has already been registered register describes a predetermined data in the column of the inspection step 4 of the above items, also defects already coordinate positions different from the defects registered in the inspection results 5 _4, new in this inspection step 4 As a defect that occurred in
The position coordinates of this defect are added to the item 206 of “defect coordinates”, and predetermined data is described and registered in the column of the inspection step 4 of each item.

In this manner, as the inspection steps 5, 6,... Proceed, the inspection results 5 of these inspection steps are sequentially registered in the defect history list.

Next, the evaluation results of the processing time and the file capacity when the defect history list 8 is created will be described.

FIG. 4 is a diagram showing the results of an experiment on the relationship between the total number of defective particles on one wafer and the processing time for creating the defect history list 8. For evaluation, Sun SPARCstation10 (memory 3
2MByte).

In the figure, 200 defects / 2 in the inspection process
In the case of zero-step inspection, the total number of defects is 4000, and the processing time for creating the defect history list 8 is estimated to be about 25 seconds. When the total number of defects exceeds 10,000 in the visual inspection, the defect history list 8 is created in about 70 seconds.

FIG. 5 is a diagram showing the results of experiments on the relationship between the total number of defects on one wafer and the file (data) capacity of the defect history list 8.

In the figure, the function of the file capacity with respect to the total number of defects in the defect history list 8 created from the inspection data for i times is a function y of the history i, where x is the total number of defects.
= Ax. For example, the file capacity for the total number of defects in the defect history list 8 created from inspection data for 20 times is expressed as a function y = 0.2474x of the history 20. Therefore, the file capacity when 20 processes are inspected in 200 defect / inspection processes is estimated to be about 1 MB.

FIG. 6 is a view showing an experimental result of a file compression effect when the defect history list 8 is compressed by a file using the data compression technique.

In the figure, the file can be compressed about 1/10 by the compress command of UNIX. Now, assuming that the function of the file capacity before compression with respect to the total number of defects in the defect history list 8 is y = 0.164x, by compressing the file, about 1/10 of y = 0.0
153x. According to this, assuming that 50 wafers are produced per day, 20 wafers are produced per wafer.
1MByte defect history list 8 from inspection data of inspection process
If a hard disk with a capacity of 2 GBytes is used, only 2000/50 = 40 days of a wafer defect history list 8 can be stored. 400 days of
That is, storage for one year is possible, and there is no practical problem.

FIG. 7 is a diagram showing a specific example of an analysis operation using the defect history list of the analyzer 11 in FIG.
Parts corresponding to those in FIG. 1 are denoted by the same reference numerals.

In the figure, when the inspection result is analyzed by the analyzer 11, the analysis unit 6 sends a defect history list 8 and a defect history list management table 9 for managing the defect history list 8, and category data as probe inspection data. 1
0 is taken. The defect history list management table 9 includes
The identifier of each wafer, for example, the type and lot, the name of the process, the number of inspected processes, and the summary of each inspection result (inspection process name, foreign matter / appearance identifier, image presence / absence flag, inspection date / time, cluster presence / absence flag, etc.) are described. I have. Category data 10
Shows the results of the electrical inspection for each chip of the completed wafer 1b (FIG. 1), which makes it possible to determine whether the chips in the wafer 1b are good or bad.

When these data are fetched, first, an analysis wafer selection screen 11a is displayed. On the analysis wafer selection screen 11a, a wafer to be analyzed (that is, an analysis wafer) can be selected based on the information in the defect history list management table 9. In this case, a plurality can be selected.

When a desired wafer is selected as an analysis wafer on the analysis wafer selection screen 11a, a defect map list display screen 11b is displayed, and the defect distribution status of each selected analysis wafer in each inspection process can be viewed. . The defect distribution status for each inspection step is displayed using the coordinate position (x, y) registered in the item 206 (FIG. 2) of “defect coordinates” in the defect history list 8.

In this case, the display data and the display condition can be selected. In this case, the display data is set to “Defect”, and the item 20 of the “defect coordinates” in the defect history list 8 is set.
The defect detected for each inspection process is displayed by using the coordinate position of No. 6 and the data of the item 207 of “defect number by process”, and the display condition is “L”.
By setting “M” and “S”, defects of all sizes are displayed. When "Adder" is selected as the display data, the item 2 of the "defect coordinates" in the defect history list 8 is displayed.
Item 208 of “defect appearance step” among the coordinate positions 06
Is selected using the data described above, and only the newly detected defect is displayed for each inspection step. When “Cluster” is selected, among the coordinate positions of the item 206 of the “defect coordinates” in the defect history list 8, The selection is made using the data of the item 210 of “Cluster”, and only the cluster defects are displayed for each inspection step. When “OR” is selected, each of the coordinate positions of the item 206 of “Defect coordinates” of the defect history list 8 is displayed. By performing the process of sequentially adding the data of the item 208 of the “defect appearance step”, all the defects detected in the previous inspection steps including the inspection step are displayed for each inspection step (for example, In the inspection step i, all the defects detected in the inspection steps 1 to i are displayed). The display condition “L”,
By selecting one of "M" and "S",
By using the coordinate position of the item 206 of “defect coordinates” of the defect history list 8 and the data of the item 209 of “defect size”, a distribution display for each defect size becomes possible.

When a predetermined operation is performed by the analysis apparatus 11 in the display state of the defect map list display screen 11b, the number of detected defects for each selected analysis wafer is divided into the vertical stack bar graphs in the defect appearance process. (Stack chart) can be displayed (stack chart display screen 11c). This stack chart is obtained by adding the data of the item 208 of “defect appearance step” in the defect history list 8 for each defect appearance step. As the display unit, it is possible to display all data (each wafer), lot average, and total average.

When a predetermined operation is performed by the analysis device 11 in the display state of the defect map list display screen 11b, each of the selected analysis wafers is subjected to each manufacturing process using the category data 10 fetched from the analysis unit 6. For each time, the number of chips (the number of lost chips) that can become defective due to a defect newly detected in the process is displayed (the fatality rate / number of lost chips display screen 11d).

In the display state of the defect map list display screen 11b, a desired analysis wafer is designated by a cursor on the screen 11b and a selection button is designated, and a wafer detailed analysis display relating to the designated analysis wafer is displayed. The screen 11e is displayed.

On the wafer detailed analysis display screen 11e,
The defect distribution status of each of the designated analysis wafers in each inspection process is displayed, and the ranks A and C indicating the quality of each chip obtained based on the category data 10.
Probe inspection data indicating B, C,..., A defect map indicating the distribution of defects detected in the specified inspection process, a graph indicating the number of defects newly detected in the inspection process, a new A graph showing the probability (fatal rate) that a detected defect causes a chip failure is displayed (note that
In the graph of the fatality rate, a chip having a higher failure rate is more likely to have a defective chip in the inspection process. In the graph shown in the drawing, the manufacturing process 3 ₃ inspected by the inspection process ₃ (FIG. 1)
Indicates that a defective chip is likely to occur).

As described above, the analysis apparatus 11 uses the defect history list 8 and the category data 10 to immediately analyze and evaluate the defect occurrence status of a predetermined process on a desired wafer and the process that caused a defect. Can do it.

FIG. 8 is an explanatory view of the inspection device 12 in FIG. 1, and portions corresponding to FIG. 1 are denoted by the same reference numerals.

The inspection apparatus 12 executes a wafer inspection process, sends the inspection result obtained to the analysis unit 6, and processes the inspection result to select a defect or the like. . FIG. 8 shows a case where the inspection apparatus 12 executes an n-th (where n = 1, 2,..., N) inspection step n.
The inspection result 5 _n obtained for the wafer processed in the manufacturing process 2 _n is supplied to the analysis unit 6 to prepare the defect history list 8 as described above, and the n-1st inspection is performed. steps 1 to n-1 of the test results 5 ₁ ~5 _n-1 defect history list 8 of the wafer based on the wafer identifier was performed test, for example, by using varieties, lot, wafer number, the analysis unit 6 The inspection result 5 _n is used to process the inspection result 5 _n, and a desired defect is selected from the inspection result 5 _n to obtain information on the desired defect.

For example, as shown in FIG.
In the inspection results 5 _n obtained in the above, referring to the position coordinates of the defect already registered in the defect history list 8 (item 206 of “defect coordinates” in FIG. 1), the inspection steps 1 to n−1 By removing the defects detected in the above step, the defects newly detected in the inspection step n are selected, and the information 14 is obtained. By displaying this on the screen, the distribution of such defects on the wafer can be known, and the tendency of defects occurring in the manufacturing process 2 _n (FIG. 8) (in FIG. 9, the tendency of defects occurring at the peripheral portion of the wafer). ) Can be known, and measures can be taken against this.

FIG. 10 is an explanatory diagram relating to the image acquisition device 13 in FIG. 1, and portions corresponding to FIG. 1 are denoted by the same reference numerals.

In the figure, an image acquisition device 13 captures an image of a desired defect to be detected by a camera and saves it on a film, or captures an image by a video camera and saves it on a recording medium. (Saving an image in this manner is referred to as image acquisition), and a desired defect is selected from the defects detected by the detection device 12 to acquire an image.

FIG. 11A shows a specific example of the defect distribution on the wafer 1 newly detected in the inspection step n in FIG. 10. Here, the new defects are defects A, B, C is three. Such defect selection is performed by referring to the defect history list 8 of the wafer 1 created using the inspection result 5 _n detected in the inspection step n. In the image acquisition device 13, an image as shown in FIG. 11A is displayed, and the user can select a defect to acquire an image from among the defects A, B, and C.

FIG. 11B shows the distribution of the defects D, E, and F which have been detected and acquired in any of the inspection steps 1 to n-1 among the defects detected in the inspection step n. Things. This is selected by the image acquisition device 13 using the registered data of the item “defect coordinates” 206 and the item “image index” 212 (FIG. 2) of the defect history list 8 of the wafer 1. Also in this case, an image as shown in FIG. 11B is displayed, and the user can select a defect from among these defects D, E, and F from which an image is to be obtained.

For the defect obtained by the image obtaining operation 13, an index unique to the defect is set in the analysis unit 6, and the “defect history list 8” is updated in the defect history list 8 based on the inspection result 5 _n of the inspection process n. This index is registered in the column for this defect in the item 211 of the "image index" (FIG. 2), that is, a column uniquely determined by the coordinates of the defect of interest and the inspection process name, and a film on which an image of this defect is recorded. Is stored with this index.

Therefore, the analyzer 11 can observe the image of the defect using the film stored in this way, and, for example, as shown in FIG. The process of change and the process of growth can be observed. The film on which the image of the same defect is recorded can be easily found from the registered data of the item 206 of “defect coordinates” and the item 212 of “image index” (FIG. 2) in the defect history list 8 of the wafer.

In FIG. 10, the image acquisition device 13
With reference to the defect history list 8 created in advance, the defect for which an image is to be obtained is selected, but the image obtaining apparatus 13 also performs the defect selection as shown in FIGS. 11A and 11B by itself. You may do so. In this case, in FIG. 10, the image acquisition device 13 takes in the inspection result 5 _n from the inspection device 12 and the defect history list 8 from the analysis unit 6 to the inspection result 5 _n−1 , respectively. By using the registration data of No. 8, a desired defect as shown in FIGS. 11A and 11B may be selected from the defects of the inspection result 5n.

As described above, in this embodiment, since the defect history list includes all past information on defects on one wafer, it is possible to obtain useful defect image data with a clear defect origin. It is possible.

Although the above embodiments have been centered on semiconductor manufacturing, the present invention is not limited to this, and products (for example, magnetic disks and circuit boards, etc.) that perform work appearance inspection and final product inspection It is needless to say that the present invention can be applied to a production line such as the above.

[0069]

As described above, according to the present invention,
Since an enormous amount of defect inspection data is organized and stored in an easy-to-use form as a defect history list, analysis preparation work time can be reduced, analysis time can be reduced, and analysis accuracy can be improved.

[Brief description of the drawings]

FIG. 1 is a system configuration diagram showing one embodiment of an electronic device inspection system and an electronic device manufacturing method using the same according to the present invention.

FIG. 2 is a configuration diagram showing a specific example of a defect history list in FIG.

FIG. 3 is a flowchart showing a specific example of a defect history list creation process of the analysis unit in FIG. 1;

4 is a diagram showing an experimental result of a relationship between a total number of defects in the defect history list shown in FIG. 2 and a creation processing time.

FIG. 5 is a diagram showing an experimental result of a relationship between a data capacity and a total number of defects in the defect history list shown in FIG. 2;

6 is a diagram showing an experimental result of a relationship between the total number of defects and the data capacity when the data of the defect history list shown in FIG. 2 is compressed.

FIG. 7 is a diagram showing operation functions of the analysis device in FIG. 1;

FIG. 8 is a diagram showing functions of the inspection device in FIG. 1;

9 is a diagram showing one processing operation of the inspection device shown in FIG.

FIG. 10 is a diagram illustrating operation functions of the image acquisition device in FIG. 1;

11 is a diagram showing one processing operation of the image acquisition device shown in FIG.

FIG. 12 is a diagram illustrating an example of use of an acquired image obtained by the image acquiring apparatus in FIG. 1;

[Explanation of symbols]

1, 1a, 1b wafer 2 ₁ ~2 _{N + 1} manufacturing process 3 ₁ to 3 _N inspection device 4 probe test step 5 ₁ to 5 _N test results 6 analysis unit 7 defect history list creation process step 8 the defect history list 9 defect history List management table 10 Category data 11 Analysis device 12 Inspection device 13 Image acquisition device

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Junko Konishi 292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Prefecture Inside the Hitachi, Ltd. Production Technology Research Institute (72) Kazunori Nemoto 292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Inside Hitachi, Ltd. Production Technology Laboratory (72) Inventor Makoto Ono 292, Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Prefecture Inside Hitachi-Production Technology Laboratory (72) Inventor Yoko Ikeda, Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa 292 Hitachi Manufacturing Co., Ltd. Production Technology Research Laboratory (72) Inventor Yasuhiro Yoshitake 292 Yoshidacho, Totsuka-ku, Yokohama-shi, Kanagawa Prefecture Hitachi Manufacturing Co., Ltd. Production Technology Research Laboratory F-term (reference) 4M106 AA01 BA01 CA38 CA41 CA70 DA14 DB02 DB04 DH60 DJ38

Claims (13)

[Claims] 1. An electronic device inspection system comprising an inspection step of inspecting an electronic device with an inspection apparatus for each predetermined operation step of a work to be an electronic device, wherein defect information of the work obtained in the inspection step is obtained. An electronic device inspection system, further comprising: an analysis unit for creating a defect history list including defect coordinates and attribute data for each workpiece. 2. The electronic device inspection system according to claim 1, further comprising an analyzer for analyzing data of the defect history list. 3. The electronic device inspection system according to claim 1, wherein the inspection apparatus acquires the defect history list and sorts out defect information in the work. 4. The electronic device according to claim 1, further comprising an image acquisition device for acquiring an image of a desired defect among the defects in the workpiece detected in the inspection step. Inspection system. 5. The electronic device inspection system according to claim 4, wherein index information indicating a defect acquired by the image acquisition operation is registered in the defect history list. 6. The image acquisition device according to claim 5, wherein the image acquisition device acquires an image of a defect near a coordinate detected in the different inspection step by referring to index information of the defect coordinate in the defect history list. An electronic device inspection system, comprising: 7. The method according to claim 1, wherein the attribute is determined based on an inspection process and a density of detected defects, a defect size, a defect type, and defect coordinates in the inspection process. An electronic device inspection system, comprising cluster information as a result and an acquired image index. 8. The analysis apparatus according to claim 2, wherein the analysis apparatus analyzes and displays a fatality rate, which is a probability that a defect newly detected in the inspection step for each inspection step causes a chip failure. Electronic device inspection system. 9. The method according to claim 8, wherein the attribute included in the defect history list transmitted to the analysis device includes an inspection process and a number of detected defects, a defect size, a defect type, and a defect coordinate in the inspection process. An electronic device inspection system, comprising: cluster information as a result of determining the density and an acquired image index. 10. A method of manufacturing an electronic device, comprising: an inspection step of inspecting a defect of a work to be an electronic device at each predetermined manufacturing step, wherein a defect generated on the work at each inspection step is provided. Producing a defect history list in which the pieces of history information are collected, and inspecting the work with reference to the defect history list in the inspection step. 11. The method according to claim 10, wherein the inspection with reference to the defect history list is acquisition of a defect image. 12. The method for manufacturing an electronic device according to claim 11, wherein the acquisition of the defect image is an acquisition of an image of a defect near coordinates detected in different inspection steps. 13. The method for manufacturing an electronic device according to claim 10, wherein the inspection referring to the defect history table is a defect analysis.