CN117197021A - Semiconductor defect identification method based on AI image identification - Google Patents
Semiconductor defect identification method based on AI image identification Download PDFInfo
- Publication number
- CN117197021A CN117197021A CN202210583619.4A CN202210583619A CN117197021A CN 117197021 A CN117197021 A CN 117197021A CN 202210583619 A CN202210583619 A CN 202210583619A CN 117197021 A CN117197021 A CN 117197021A
- Authority
- CN
- China
- Prior art keywords
- defect
- image
- recognition
- defects
- accuracy
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000007547 defect Effects 0.000 title claims abstract description 202
- 238000000034 method Methods 0.000 title claims abstract description 50
- 239000004065 semiconductor Substances 0.000 title claims abstract description 31
- 230000008569 process Effects 0.000 claims abstract description 17
- 239000002245 particle Substances 0.000 claims description 13
- 238000013135 deep learning Methods 0.000 claims description 4
- 238000011282 treatment Methods 0.000 claims description 4
- 238000004519 manufacturing process Methods 0.000 abstract description 7
- 238000012552 review Methods 0.000 abstract description 4
- 238000004131 Bayer process Methods 0.000 abstract 1
- 230000000875 corresponding effect Effects 0.000 description 6
- 229940095676 wafer product Drugs 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 101100476210 Caenorhabditis elegans rnt-1 gene Proteins 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000004626 scanning electron microscopy Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000007103 stamina Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Testing Or Measuring Of Semiconductors Or The Like (AREA)
Abstract
The invention provides a semiconductor defect identification method based on AI image identification, which calculates the accuracy of primary defects according to the original coordinate positions of the defects, and judges whether the accuracy of the primary defects reaches a primary threshold value or not; if the primary threshold value is not reached, searching the defect offset and adjusting the defect offset; calculating secondary defect accuracy according to the final defect coordinates, and judging whether the secondary defect accuracy reaches a secondary threshold value or not; if the secondary threshold is reached, the types of defects are further identified, classified and ranked. The automatic Bayer process control system has the advantages that automation is realized under the condition that equipment hardware is not transformed or upgraded, dependence on personnel is supported, misjudgment and misoperation in the operation process are greatly reduced, the production efficiency is improved, the success rate and stability of defect review are improved, and the labor cost is directly reduced.
Description
Technical Field
The invention relates to a semiconductor defect identification method, in particular to a semiconductor defect identification method based on AI image identification.
Background
In the semiconductor industry, defect recognition equipment (SEM Review scanning electron microscopy) is currently operated in a manual operation mode, the steps are fixed and repeated, but links of defect recognition and classification have high requirements on quality of people, and the links depend on experience and service level of operators. The flow of the manual operation mode is as follows: manually selecting parameters by an on-line operator; after the equipment runs to find the related data, intervention is started manually, and whether the equipment accurately searches the defects is observed; if the equipment is inaccurate in finding the defects, the defects are manually found by people, and the positions are marked; then the equipment searches the related data; until all defects are manually confirmed to be found accurately; and classifying the defects manually by experience, and performing various detection manually. The whole process has very high requirements on personnel, and the personnel is required to have abundant defect identification, classification experience and sharp judgment, and has very high requirements on the stamina, bearing capacity and responsibility of the personnel. In the past work, the problems of missing finding and error finding of the defect position, error judging of the defect type, error EDX operation and the like often occur. Meanwhile, the whole industry is also faced with the problems of difficulty in newly-recruiting staff and high staff leaving rate, and further the problems of long staff training period, high difficulty, high training cost, and the like are brought to enterprises.
Disclosure of Invention
The invention provides a semiconductor defect identification method based on AI image identification; under the condition of not modifying or upgrading equipment hardware, the defect review process is completely automated, the dependence on personnel is supported, misjudgment and misoperation in the operation process are greatly reduced, the production efficiency is improved, the success rate and stability of the defect review are improved, and the labor cost is directly reduced; overcomes the defects in the prior art.
The invention provides a semiconductor defect identification method based on AI image identification, which comprises the following steps:
step S1, obtaining defect information of a product, wherein the defect information comprises a defect original coordinate position;
s2, calculating the primary defect accuracy according to the original coordinate position of the defect, and judging whether the primary defect accuracy reaches a primary threshold value or not; if the primary threshold is reached, the step S5 is entered; if the primary threshold is not reached, the step S3 is entered;
s3, searching a defect offset and adjusting; the method comprises the following steps:
s3-1, acquiring a surrounding image of a position where the defect is not identified;
s3-2, judging whether a surrounding image has defects or not through AI image recognition; if the defect is found in the surrounding image, recording the actual coordinates of the position of the found defect, and then entering step S3-3;
step S3-3, calculating vector difference values of the actual coordinate position and the original coordinate position of each defect, and calculating their average.
S3-4, taking the original coordinate position as a defect offset according to the average number, and obtaining a final coordinate of the defect position;
s4, calculating secondary defect accuracy according to the final defect coordinates, and judging whether the secondary defect accuracy reaches a secondary threshold value or not; if the secondary threshold is reached, the step S5 is entered;
step S-5, further identifying the type of the defect in step S2 or step S4, classifying, and sorting.
Further, the present invention provides a semiconductor defect recognition method based on AI image recognition, which may further have the following features: step S2 comprises the steps of:
s2-1, selecting a set number of defects, and judging whether the product has defects at the selected original coordinate position or not through AI image recognition according to the selected defect information;
s2-2, counting the number of defects judged in the step, and calculating to obtain the primary defect accuracy;
s2-3, judging whether the accuracy of the primary defect reaches a primary threshold value; if the primary threshold is reached, the step S5 is entered; if the primary threshold is not reached, the process proceeds to step S3.
Further, the present invention provides a semiconductor defect recognition method based on AI image recognition, which may further have the following features: in step S2-1, according to the original coordinate position of each defect center point, an image centered on the coordinate position is acquired, and a trained model is adopted to judge whether defects exist in the image.
Further, the present invention provides a semiconductor defect recognition method based on AI image recognition, which may further have the following features: step S4 comprises the steps of:
s4-1, selecting a set number of defects, and judging whether the product has defects at a final coordinate position or not through AI image recognition according to the selected defect information;
s4-2, counting the number of defects judged in the step, and calculating to obtain the secondary defect accuracy;
s4-3, judging whether the accuracy of the secondary defect reaches a secondary threshold value; if the secondary threshold is reached, the process proceeds to step S5.
Further, the present invention provides a semiconductor defect recognition method based on AI image recognition, which may further have the following features: in step S4-1, according to the final coordinate position of each defect center point, an image centered on the coordinate position is acquired, and a trained model is adopted to judge whether defects exist in the image.
Further, the present invention provides a semiconductor defect recognition method based on AI image recognition, which may further have the following features: image recognition adopts image recognition based on a traditional computer vision classical open source library OpenCV and/or image recognition based on AI deep learning.
Further, the present invention provides a semiconductor defect recognition method based on AI image recognition, which may further have the following features: the types of defects can be classified as: one or more of raised particles, scratches and depressions.
Further, the present invention provides a semiconductor defect recognition method based on AI image recognition, which may further have the following features: types of defects include raised particles; the defect information further includes a size of the defect; and S-6, selecting the largest selected number of the raised particles for scattering treatment, and collecting and analyzing the components of the scattered particles.
Further, the present invention provides a semiconductor defect recognition method based on AI image recognition, which may further have the following features: in step S3-1, the original coordinate position of the defect not identified in step S2 is enlarged, the shooting range of the image is widened, or 8 surrounding pictures centered on the picture are acquired centered on the image.
The invention provides a semiconductor defect identification method based on AI image identification, which fully realizes automation of the operation process of the existing equipment, and the whole process does not need to be manually participated or intervened under the condition that no abnormal condition exists in the whole process, thus truly realizing automation and unmanned production links. The concrete steps are as follows: the defects are searched, the shooting accuracy is judged and counted, and different operation treatments are carried out on corresponding counting results according to a threshold value set in advance; the defect type is automatically identified and classified, and the classified result can be stored in the background, so that the follow-up manual secondary confirmation and historical record tracing are facilitated; the function of automatically screening out a preset defect of a certain specific type is realized, and the next EDX action is automatically carried out; the EDX result and the corresponding photo are automatically stored; the function of automatic alarm after the failure of the GA link machine is realized, error information is prompted to corresponding personnel through an RCM interface, and further measures are directly taken through the RCM; the function of automatically triggering corresponding actions or alarming according to a threshold value set in advance in the Run1 link is realized in the EDX link; the device has the function of accessing an automatic control system, provides automatic production, improves the production efficiency and reduces production errors; the method has the advantages of no reconstruction of equipment software, low cost and quick time, and has absolute advantages compared with the feasibility of reconstructing equipment hardware and equipment software by equipment manufacturers, realization speed, cost and later maintenance.
Drawings
Fig. 1 is a flowchart of a semiconductor defect recognition method based on AI image recognition in an embodiment.
Detailed Description
In order to more clearly illustrate the implementation of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and examples. The specific examples described herein are intended to be illustrative of the present disclosure and are not intended to be limiting of the present embodiments.
Examples
The method comprises the following steps: detecting a defect of a lot of wafer products lot, and recording the position of the defect and the size of the defect when the defect is found. Specifically, the center of the circle of the wafer is used as the coordinate center, the original coordinate positions of all defect center points are recorded, and the radius size of the defect is recorded. And then correspondingly storing the lot of wafer products lot and defect information.
This step is the prior art for automatically detecting defects in wafer products, and will not be described in detail.
In this embodiment, a semiconductor defect identification method based on AI image identification includes the following steps:
the preparation steps are as follows: the equipment operator places a lot of wafer products lot on the detection input port of the equipment, manually selects a lot identification lotID, a recipe number recope and a wafer identification wafer ID, and can also adopt modes of automatically identifying the lot identification lotID and the like. And then clicking a start button on an interface of the automated control scheduling system RCM of the flexible manufacturing unit to start a real-time monitoring device picture.
Step S1, obtaining defect information of lot of wafer products lot.
And reading defect information of the wafer product lot of the lot, namely the original coordinate position of the center point of each defect and the radius size of the defect according to any one or more information of the lot identification lotID, the recipe number record and the wafer identification wafer ID of the wafer product lot of the lot. Of course, if the obtained batch information cannot read the corresponding defect information, an error is reported, an alarm is sent, and a manual intervention process is sought.
S2, calculating the primary defect accuracy according to the original coordinate position of the defect, and judging whether the primary defect accuracy reaches a primary threshold value or not; if the primary threshold is reached, the step S5 is entered; if the primary threshold is not reached, the process proceeds to step S3.
S2-1, selecting a set number of defects, and judging whether the defects exist in the original coordinate positions of the wafer product lot or not through AI image identification according to the selected original coordinate positions of the defects.
Acquiring an image with the coordinate position as a center according to the original coordinate position of each defect center point, and judging whether defects exist in the image by adopting a trained model; if the defect is identified in the picture, the defect identification of the position is considered to be accurate, and if the defect is not identified in the picture, the defect identification of the position is considered to be inaccurate.
In this embodiment, the set number is 20.
And S2-2, counting the correct number/set number judged by the steps, and obtaining the primary defect accuracy.
S2-3, judging whether the accuracy of the primary defect reaches a primary threshold value; if the primary threshold is reached, the step S5 is entered; if the primary threshold is not reached, the process proceeds to step S3.
In this embodiment, the primary threshold is equal to or greater than 70%; the primary defect accuracy is more than or equal to 70%, namely, when 14 defects are correct, the step S5 is carried out; and when the primary defect accuracy is less than 70%, entering a step S3.
A primary defect accuracy below 70% of the primary threshold value indicates that there is a certain error in semiconductor defect identification, and adjustment is required.
And S3, searching a defect offset and adjusting.
And step S3-1, acquiring a surrounding image of the position where the defect is not identified.
And (3) expanding the shooting range of the image or taking the image as the center to acquire 8 surrounding pictures taking the picture as the center by taking the original coordinate position in which the defect is not identified in the step S2-1.
S3-2, judging whether the surrounding image of the wafer product lot has defects or not through AI image identification; if no defect exists in the surrounding images, reporting errors, sending out an alarm, and searching for manual intervention; if a defect is found in the surrounding image, the actual coordinates of the location of the found defect are recorded and then step S3-3 is entered.
Step S3-3, calculating vector difference values of the actual coordinate position and the original coordinate position of each defect, and calculating their average.
And S3-4, taking the original coordinate position as a defect offset according to the average number, and obtaining the final coordinate of the defect position.
S4, calculating secondary defect accuracy according to the final defect coordinates, and judging whether the secondary defect accuracy reaches a secondary threshold value or not; if the secondary threshold is reached, the step S5 is entered; if the secondary threshold is not reached, an error is reported, an alarm is sent out, and a manual intervention process is sought.
And S4-1, selecting a set number of defects, and judging whether the wafer product lot has defects at the final coordinate position according to the selected defect information through AI image identification.
Acquiring an image with the coordinate position as a center according to the final coordinate position of each defect center point, and judging whether defects exist in the image by adopting a trained model; if the defect is identified in the picture, the defect identification of the position is considered to be accurate, and if the defect is not identified in the picture, the defect identification of the position is considered to be inaccurate.
In this embodiment, the set number is 20.
And S4-2, counting the correct number/set number judged by the steps, and obtaining the secondary defect accuracy.
S4-3, judging whether the accuracy of the secondary defect reaches a secondary threshold value; if the secondary threshold is reached, the step S5 is entered; if the secondary threshold is not reached, an error is reported, an alarm is sent out, and a manual intervention process is sought.
In this embodiment, the secondary threshold is greater than or equal to 70%; namely, the secondary defect accuracy is more than or equal to 70%, and the step S5 is carried out; and stopping operation when the secondary defect accuracy is less than 70%.
A secondary defect accuracy below 70% of the secondary threshold represents a large error in semiconductor defect identification, a failure to adjust to the correct position by the defect offset, a need for re-identification, or a human error.
Step S-5, further identifying the type of the defect in step S2-1 or step S4-1, classifying, and sorting.
In step S2-1, according to the original coordinate position of each defect center point, an image centered on the coordinate position is acquired, and the type of the defect in the image is judged by adopting a trained model.
Or in step S4-1, according to the final coordinate position of each defect center point, acquiring an image with the coordinate position as the center, and judging the type of the defect in the image by adopting a trained model.
The types of defects can be classified as: raised particles, scratches, depressions, etc. And the defects are ordered according to the radius size, and are arranged in sequence according to the defect size. Can be used for subsequent manual work or program to judge whether the batch of products are qualified or not.
S-6, selecting the largest selected number of the convex particles for scattering treatment, and collecting and analyzing the components of the scattered particles.
In this example, the energy dispersive X-ray spectrometer EDX was used to break up and analyze the composition of the largest 3 raised particles, thereby obtaining the main components forming the particle defects.
Finally, the semiconductor defect identification method based on AI image identification stores the acquired information of the batch of products in the process so as to be convenient for subsequent operation or inspection and other conditions to be reused.
In the embodiment, the image recognition part respectively adopts the image recognition technology based on the traditional computer vision classical open source library OpenCV and the image recognition technology based on AI deep learning, and is respectively suitable for different scenes, the OpenCV is suitable for scenes with single picture types and backgrounds, simple template matching, positioning and the like are performed, rapid deployment is facilitated, and no special requirement is imposed on hardware; in contrast, the AI deep learning technology has strong processing capability on images, high processing speed, capability of coping with complex and changeable scenes, high accuracy, high corresponding deployment difficulty, capability of achieving ideal accuracy only by repeated training of a model, certain requirement on hardware (GPU), and the number of hardware deployment being basically proportional to the scale and complexity of application.
It should be noted that, under normal conditions, no error is usually reported, and an alarm is triggered. This is the case in extreme cases, or when the automation device has a BUG or when a pre-step has an error, belonging to the protection program.
While the present invention has been described in detail through the foregoing description of the preferred embodiment, it should be understood that the foregoing description is not to be considered as limiting the invention.
Claims (9)
1. A semiconductor defect identification method based on AI image identification is characterized in that: the method comprises the following steps:
step S1, obtaining defect information of a product, wherein the defect information comprises a defect original coordinate position;
s2, calculating the primary defect accuracy according to the original coordinate position of the defect, and judging whether the primary defect accuracy reaches a primary threshold value or not; if the primary threshold is reached, the step S5 is entered; if the primary threshold is not reached, the step S3 is entered;
s3, searching a defect offset and adjusting; the method comprises the following steps:
s3-1, acquiring a surrounding image of a position where the defect is not identified;
s3-2, judging whether a surrounding image has defects or not through AI image recognition; if the defect is found in the surrounding image, recording the actual coordinates of the position of the found defect, and then entering step S3-3;
step S3-3, calculating vector difference values of the actual coordinate position and the original coordinate position of each defect, and calculating their average.
S3-4, taking the original coordinate position as a defect offset according to the average number, and obtaining a final coordinate of the defect position;
s4, calculating secondary defect accuracy according to the final defect coordinates, and judging whether the secondary defect accuracy reaches a secondary threshold value or not; if the secondary threshold is reached, the step S5 is entered;
step S-5, further identifying the type of the defect in step S2 or step S4, classifying, and sorting.
2. The AI-image-recognition-based semiconductor defect recognition method of claim 1, wherein: step S2 comprises the steps of:
s2-1, selecting a set number of defects, and judging whether the product has defects at the selected original coordinate position or not through AI image recognition according to the selected defect information;
s2-2, counting the number of defects judged in the step, and calculating to obtain the primary defect accuracy;
s2-3, judging whether the accuracy of the primary defect reaches a primary threshold value; if the primary threshold is reached, the step S5 is entered; if the primary threshold is not reached, the process proceeds to step S3.
3. The AI-image-recognition-based semiconductor defect recognition method of claim 1, wherein: in step S2-1, according to the original coordinate position of each defect center point, an image centered on the coordinate position is acquired, and a trained model is adopted to judge whether defects exist in the image.
4. The AI-image-recognition-based semiconductor defect recognition method of claim 1, wherein: step S4 comprises the steps of:
s4-1, selecting a set number of defects, and judging whether the product has defects at a final coordinate position or not through AI image recognition according to the selected defect information;
s4-2, counting the number of defects judged in the step, and calculating to obtain the secondary defect accuracy;
s4-3, judging whether the accuracy of the secondary defect reaches a secondary threshold value; if the secondary threshold is reached, the process proceeds to step S5.
5. The AI-image-recognition-based semiconductor defect recognition method of claim 1, wherein: in step S4-1, according to the final coordinate position of each defect center point, an image centered on the coordinate position is acquired, and a trained model is adopted to judge whether defects exist in the image.
6. The AI-image-recognition-based semiconductor defect recognition method of claim 3 or 5, wherein: image recognition adopts image recognition based on a traditional computer vision classical open source library OpenCV and/or image recognition based on AI deep learning.
7. The AI-image-recognition-based semiconductor defect recognition method of claim 1, wherein: the types of defects can be classified into: one or more of raised particles, scratches and depressions.
8. The AI-image-recognition-based semiconductor defect recognition method of claim 1, wherein: types of defects include raised particles; the defect information further includes a size of the defect;
and S-6, selecting the largest selected number of the raised particles for scattering treatment, and collecting and analyzing the components of the scattered particles.
9. The AI-image-recognition-based semiconductor defect recognition method of claim 1, wherein: in step S3-1, the original coordinate position of the defect not identified in step S2 is enlarged, the shooting range of the image is widened, or 8 surrounding pictures centered on the picture are acquired centered on the image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210583619.4A CN117197021A (en) | 2022-05-25 | 2022-05-25 | Semiconductor defect identification method based on AI image identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210583619.4A CN117197021A (en) | 2022-05-25 | 2022-05-25 | Semiconductor defect identification method based on AI image identification |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117197021A true CN117197021A (en) | 2023-12-08 |
Family
ID=88987389
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210583619.4A Pending CN117197021A (en) | 2022-05-25 | 2022-05-25 | Semiconductor defect identification method based on AI image identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117197021A (en) |
-
2022
- 2022-05-25 CN CN202210583619.4A patent/CN117197021A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110709688B (en) | Method for predicting defects in an assembly unit | |
| CN105817430B (en) | Product inspection method based on machine vision | |
| CN102473660B (en) | Automatic fault detection and classification in a plasma processing system and methods thereof | |
| JP5168215B2 (en) | Appearance inspection device | |
| CN112016409A (en) | Deep learning-based process step specification visual identification determination method and system | |
| US6701204B1 (en) | System and method for finding defective tools in a semiconductor fabrication facility | |
| US20220284699A1 (en) | System and method of object detection using ai deep learning models | |
| WO2021120179A1 (en) | Product manufacturing message processing method and device, and computer storage medium | |
| CN110738630A (en) | Training method and detection system of recursive deep learning system | |
| CN117372377B (en) | Broken line detection method and device for monocrystalline silicon ridge line and electronic equipment | |
| CN117689646A (en) | High-precision defect detection method, system and medium for positive and negative sample fusion | |
| CN117669964A (en) | Scheduling method and system for performing early maintenance on machine | |
| CN117314829A (en) | Industrial part quality inspection method and system based on computer vision | |
| CN113297913A (en) | Method for identifying dressing specification of distribution network field operating personnel | |
| CN117197021A (en) | Semiconductor defect identification method based on AI image identification | |
| CN116840240A (en) | Visual detection system of power supply distributor | |
| TWI785579B (en) | Automatic model reconstruction method and automatic model reconstruction system for component recognition model | |
| CN112927170B (en) | Automatic defect removing method in semiconductor manufacturing process | |
| CN110087066B (en) | One-key automatic inspection method applied to online inspection | |
| CN113012125A (en) | Automatic pincard checking method and system for semiconductor CP (content provider) equipment based on image recognition | |
| CN117817211B (en) | Welding automation control method and system based on machine vision | |
| EP3992738A1 (en) | Fluorescent inspection visual data processing system and processing method thereof based on mes | |
| CN117690817B (en) | Identification method and system for linear defects | |
| CN118096747B (en) | Automatic PCBA (printed circuit board assembly) board detection method and system based on deep learning | |
| CN117654907B (en) | Automatic eliminating method and system for strip detector |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |