Defect monitoring method and defect monitoring system

Abstract

The present invention provides a defect monitoring method and defect monitoring system. The defect monitoring method includes: establishing a digital signal database; acquiring a first optical image of a test wafer layer by layer, manually defining nuisance defects; digitally analyzing the first optical image containing the nuisance defects; and storing the obtained first digital signals in the digital signal database; acquiring a second optical image of a monitoring wafer and digitally analyzing it, outputting a second digital signal of the monitoring wafer; and comparing the second digital signal with a first digital signal in the digital signal database. If the second digital signal is identical to a first digital signal in the digital signal database, the second optical image is filtered out. By establishing a digital signal database and using digital signal analysis and comparison, defect filtering is intelligent, efficient, and highly accurate, effectively improving wafer yield.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor manufacturing technology, and in particular to a defect monitoring method and a defect monitoring system. Background Art

[0002] In recent years, with the rapid development of semiconductor integrated circuits and the scaling down of critical dimensions, their manufacturing processes have become increasingly complex. Currently, advanced integrated circuit manufacturing processes typically involve hundreds of steps. A problem in any one step can cause problems throughout the entire semiconductor integrated circuit chip, and in severe cases, even lead to chip failure. Therefore, timely detection of process defects is crucial in the semiconductor integrated circuit manufacturing process. Based on these considerations, the industry generally uses defect scanning machines for image processing chips, performing chip defect detection to control defects in the manufacturing process.

[0003] Multiple unit chips (dies) are arranged on a wafer. The semiconductor industry's defect scanning machines typically use a die-to-die comparison method (a defect scanning method that uses unit chips as comparison units) to determine whether defects exist on the chip units. Specifically, when using the die-to-die comparison method, two adjacent unit chips are compared to determine whether there are differences between the two unit chips, thereby detecting the presence of defects. When performing die-to-die comparison on the same wafer by converting image signals into digital signals, manual parameter settings are required to filter out some nuisance defects. Nuisance defects can be irregularities or false defects on the chip. This method relies on manual parameter settings. When the parameter settings are too large, real defects can be easily overlooked. Furthermore, when the digital signals of real defects are close to those of nuisance defects, they are also easily overlooked, resulting in yield loss. Summary of the Invention

[0004] The object of the present invention is to provide a defect monitoring method and a defect monitoring system to improve the filtering accuracy of defects and avoid yield loss.

[0005] In order to achieve the above object, the present invention provides a defect monitoring method, comprising:

[0006] Establishing a digital signal database: collecting first optical images of the test wafer layer by layer, manually defining nuisance defects, digitally analyzing the first optical images containing the nuisance defects, and storing the obtained first digital signals in the digital signal database;

[0007] A second optical image of the monitoring wafer is collected and digitally analyzed, a second digital signal of the monitoring wafer is output, and the second digital signal is compared with a first digital signal in the digital signal database. If the second digital signal is the same as one of the first digital signals in the digital signal database, the second optical image is filtered out.

[0008] Optionally, when counting the detection areas where defects occur, the defect monitoring method further includes:

[0009] Analyzing the first digital signal to obtain a variation range of the first digital signal;

[0010] If the second digital signal is different from any of the first digital signals in the digital signal database, it is further determined whether the second digital signal is within the variation range of a certain first digital signal. If so, the second optical image is filtered out; if not, the defect distribution map of the monitored wafer is output.

[0011] Optionally, when analyzing the first digital signal, a median value of the first digital signal is selected for fitting to obtain a variation range of the first digital signal.

[0012] Optionally, if the second digital signal is within a variation range of a certain first digital signal in the digital signal database, the second digital signal is stored in the digital signal database.

[0013] Optionally, the nuisance defect is set by manual editing to form a judgment standard, and the judgment standard can be adjusted in real time according to the manufacturing standard.

[0014] Based on this, the present application also provides a defect monitoring system, including:

[0015] An optical image acquisition module, configured to acquire a first optical image of the test wafer layer by layer and a second optical image of the monitoring wafer;

[0016] a digital analysis module, configured to digitally analyze the first optical image and the second optical image having nuisance defects, and obtain corresponding first digital signals and second digital signals;

[0017] a digital signal database, configured to store the first digital signal;

[0018] The comparison module is configured to compare the second digital signal with a first digital signal in the digital signal database, and filter out the second optical image if the second digital signal is identical to a first digital signal in the digital signal database.

[0019] Optionally, if the second digital signal is different from any of the first digital signals in the digital signal database;

[0020] The digital analysis module is further configured to analyze the first digital signal to obtain a variation range of the first digital signal;

[0021] The comparison module is further configured to determine whether the second digital signal is within a certain variation range of the first digital signal; if so, the second optical image is filtered out; if not, a defect distribution map of the monitored wafer is output.

[0022] Optionally, if the second digital signal is within a variation range of a certain first digital signal in the digital signal database, the digital signal database is further used to store the second digital signal.

[0023] Optionally, the nuisance defect is set by manual editing to form a judgment standard, and the judgment standard can be adjusted in real time according to the manufacturing standard.

[0024] Optionally, the optical image acquisition module is an optical microscope or a high-resolution camera.

[0025] In the defect monitoring method and defect monitoring system provided by the present invention, a digital signal database is established to store the digital signals corresponding to various optical images with nuisance defects, and defects are filtered out by using digital signal analysis and comparison to avoid misjudgment caused by nuisance defects, and a defect distribution map with real defects is output. This method is intelligent, efficient, and highly accurate, effectively improving the yield of wafers. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A diagram showing the steps of a defect monitoring method provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0027] To make the objects, advantages, and features of the present invention more clearly apparent, the present invention is further described below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the drawings are all in a very simplified form and are not drawn to scale. They are only used to conveniently and clearly assist in illustrating the purposes of the embodiments of the present invention. In addition, the structures shown in the drawings are often part of the actual structure. In particular, different drawings may need to illustrate different focuses and sometimes use different scales.

[0028] The singular terms "one", "an" and "the" used in the present invention may include plural objects, unless the content clearly indicates otherwise. The term "or" used in the present invention is generally used to include the meaning of "and / or", unless the content clearly indicates otherwise. The term "several" used in the present invention is generally used to include the meaning of "at least one", unless the content clearly indicates otherwise. The term "at least two" used in the present invention is generally used to include the meaning of "two or more", unless the content clearly indicates otherwise. In addition, the terms "first", "second" and "third" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first", "second" and "third" may explicitly or implicitly include one or at least two of the features.

[0029] like Figure 1 As shown, Figure 1 This is a step diagram of a defect monitoring method provided by an embodiment of the present invention. This embodiment provides a defect monitoring method, including:

[0030] Establishing a digital signal database: collecting first optical images of the test wafer layer by layer, manually defining nuisance defects, digitally analyzing the first optical images containing the nuisance defects, and storing the obtained first digital signals in the digital signal database;

[0031] A second optical image of the monitoring wafer is collected and digitally analyzed, a second digital signal of the monitoring wafer is output, and the second digital signal is compared with a first digital signal in the digital signal database. If the second digital signal is the same as one of the first digital signals in the digital signal database, the second optical image is filtered out.

[0032] By establishing a digital signal database and using digital signal analysis and comparison to filter defects, it is intelligent, efficient, and highly accurate, effectively improving the yield of wafers.

[0033] Specifically, step S1 is first performed to establish a digital signal database: first optical images of the test wafer are acquired layer by layer, nuisance defects are manually defined, the first optical images containing the nuisance defects are digitally analyzed, and the resulting first digital signals are stored in the digital signal database. In this embodiment, the first optical images containing the nuisance defects are digitally analyzed, the first digital signals are set to A, B, C, ..., and the first digital signals are packaged and stored in the digital signal database. Each first digital signal is stored as M0, M1, M2, ..., so that the first digital signal can be stored as M0 = {A0, B0, C0, ...}.

[0034] Then, step S2 is executed to capture a second optical image of the monitoring wafer and perform digital analysis, outputting a second digital signal of the monitoring wafer. The second digital signal is then compared with a first digital signal in the digital signal database. If the second digital signal is identical to a first digital signal in the digital signal database, the second optical image is filtered out. For example, the second optical image of the monitoring wafer is input, and its digital signal MN = {AN, BN, CN...}} is output. MN is then compared with M0, M1, M2, etc. in the digital signal database. If they are identical, the second optical image is filtered out.

[0035] If the second digital signal is different from any of the first digital signals in the digital signal database, the defect monitoring method further includes:

[0036] Analyzing the first digital signal to obtain a variation range of the first digital signal;

[0037] It is further determined whether the second digital signal is within a certain variation range of the first digital signal. If so, the second optical image is filtered out; if not, a defect distribution map of the monitored wafer is output.

[0038] In this embodiment, when analyzing the first digital signal, the median value of the first digital signal is selected for fitting to obtain a variation range of the first digital signal, which is as follows:

[0039] Analyze the first digital signal and obtain its median values QA, QB, QC, ...;

[0040] Fitting the appropriate variation ranges PA, PB, PC, ... of various first digital signals, defining RA = QA ± PA, RB = QB ± PB, ...;

[0041] The variation range of the first digital signal MR={RA, RB, RC...} is obtained.

[0042] At this time, it is further determined whether the second digital signal MN is within a certain variation range MR of the first digital signal. If so, the second optical image is filtered out; if not, the defect distribution map of the monitored wafer is output.

[0043] Furthermore, if the second digital signal is within a variation range of a certain first digital signal in the digital signal database, the second digital signal is stored in the digital signal database to enrich the digital signal database.

[0044] In this embodiment, the nuisance defects are edited and set by engineers to form a judgment standard, and the judgment standard can be adjusted in real time by the engineers according to the manufacturing standard.

[0045] Based on this, the present invention also provides a defect monitoring system, comprising:

[0046] An optical image acquisition module, configured to acquire a first optical image of the test wafer layer by layer and a second optical image of the monitoring wafer;

[0047] a digital analysis module connected to the optical image acquisition module, configured to digitally analyze the first optical image and the second optical image having nuisance defects and obtain corresponding first digital signals and second digital signals;

[0048] a digital signal database, connected to the digital analysis module, and configured to store the first digital signal;

[0049] A comparison module is connected to the digital analysis module and the digital signal database, and is used to compare the second digital signal with the first digital signals in the digital signal database. If the second digital signal is identical to any of the first digital signals in the digital signal database, the second optical image is filtered out.

[0050] In this embodiment, the optical image acquisition module, the digital analysis module, the digital signal database, and the comparison module can all be installed or embedded in a wafer defect scanning machine. The wafer defect scanning machine is used to scan target wafers to capture defects and perform scanning and detection of various defects generated during the wafer manufacturing process. The nuisance defects are edited and set by engineers to form a judgment standard, and the judgment standard can be adjusted in real time by the engineers according to manufacturing standards.

[0051] Further, if the second digital signal is different from any of the first digital signals in the digital signal database;

[0052] The digital analysis module is further configured to analyze the first digital signal to obtain a variation range of the first digital signal;

[0053] The comparison module is further configured to determine whether the second digital signal is within a certain variation range of the first digital signal; if so, the second optical image is filtered out; if not, a defect distribution map of the monitored wafer is output.

[0054] Furthermore, if the second digital signal is within a variation range of a certain first digital signal in the digital signal database, the digital signal database is further configured to store the second digital signal.

[0055] In this embodiment, the optical image acquisition module is an optical microscope or a high-resolution camera.

[0056] In summary, the present invention provides a defect monitoring method and system. By establishing a digital signal database to store the digital signals corresponding to various optical images containing nuisance defects, and using digital signal analysis and comparison to filter defects, this method avoids misjudgments caused by nuisance defects and outputs a defect distribution map showing actual defects. This intelligent, efficient, and highly accurate system effectively improves wafer yield. Furthermore, as the digital signal database continues to improve and enrich, the accuracy of the defect monitoring method can be further enhanced.

[0057] The above description is merely a preferred embodiment of the present invention and does not limit the present invention in any way. Any person skilled in the art who, without departing from the scope of the present invention, makes any equivalent substitution, modification, or other changes to the technical solution and technical content disclosed in the present invention shall be deemed to be within the scope of the present invention and still fall within the scope of protection of the present invention.

Claims

1. A defect monitoring method, characterized in that: include: Establishing a digital signal database: collecting first optical images of the test wafer layer by layer, manually defining nuisance defects, digitally analyzing the first optical images containing the nuisance defects, and storing the obtained first digital signals in the digital signal database; collecting a second optical image of the monitoring wafer and performing digital analysis thereon, outputting a second digital signal of the monitoring wafer, and comparing the second digital signal with a first digital signal in the digital signal database; and filtering out the second optical image if the second digital signal is identical to a first digital signal in the digital signal database; The defect monitoring method further includes: Analyzing the first digital signal to obtain a variation range of the first digital signal; If the second digital signal is different from any of the first digital signals in the digital signal database, it is further determined whether the second digital signal is within the variation range of a certain first digital signal. If so, the second optical image is filtered out; if not, the defect distribution map of the monitored wafer is output.

2. The defect monitoring method according to claim 1, wherein: When analyzing the first digital signal, the median value of the first digital signal is selected for fitting to obtain a variation range of the first digital signal.

3. The defect monitoring method according to claim 1, wherein: If the second digital signal is within a variation range of a certain first digital signal in the digital signal database, the second digital signal is stored in the digital signal database.

4. The defect monitoring method according to claim 1, wherein: The nuisance defects are manually edited to form a judgment standard, and the judgment standard can be adjusted in real time according to the manufacturing standard.

5. A defect monitoring system, characterized in that: include: An optical image acquisition module, configured to acquire a first optical image of the test wafer layer by layer and a second optical image of the monitoring wafer; a digital analysis module, configured to digitally analyze the first optical image and the second optical image having nuisance defects, and obtain corresponding first digital signals and second digital signals; a digital signal database, configured to store the first digital signal; a comparison module, configured to compare the second digital signal with a first digital signal in the digital signal database, and filter out the second optical image if the second digital signal is identical to a first digital signal in the digital signal database; If the second digital signal is different from any of the first digital signals in the digital signal database; The digital analysis module is further configured to analyze the first digital signal to obtain a variation range of the first digital signal; The comparison module is further configured to determine whether the second digital signal is within a certain variation range of the first digital signal; if so, the second optical image is filtered out; if not, a defect distribution map of the monitored wafer is output.

6. The defect monitoring system according to claim 5, wherein: If the second digital signal is within a variation range of a certain first digital signal in the digital signal database, the digital signal database is further configured to store the second digital signal.

7. The defect monitoring system according to claim 5, wherein: The nuisance defects are manually edited to form a judgment standard, and the judgment standard can be adjusted in real time according to the manufacturing standard.

8. The defect monitoring system according to claim 5, wherein: The optical image acquisition module is an optical microscope or a high-resolution camera.

Images