CN106971955A - The detection method of wafer defect - Google Patents
The detection method of wafer defect Download PDFInfo
- Publication number
- CN106971955A CN106971955A CN201710353548.8A CN201710353548A CN106971955A CN 106971955 A CN106971955 A CN 106971955A CN 201710353548 A CN201710353548 A CN 201710353548A CN 106971955 A CN106971955 A CN 106971955A
- Authority
- CN
- China
- Prior art keywords
- region
- detection zone
- mensuration
- pixel
- mensuration region
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L22/00—Testing or measuring during manufacture or treatment; Reliability measurements, i.e. testing of parts without further processing to modify the parts as such; Structural arrangements therefor
- H01L22/10—Measuring as part of the manufacturing process
- H01L22/12—Measuring as part of the manufacturing process for structural parameters, e.g. thickness, line width, refractive index, temperature, warp, bond strength, defects, optical inspection, electrical measurement of structural dimensions, metallurgic measurement of diffusions
Abstract
The present invention provides a kind of detection method of wafer defect, including：The image of crystal column surface is obtained, and the image of the crystal column surface is divided into multiple detection zones；Multiple first mensuration regions are chosen in each detection zone, and the first mensuration region in each detection zone is subjected to differentiation comparison, the result compared according to differentiation obtains the threshold value that each detection zone is used to characterize defect；Each detection zone is divided into multiple second mensuration regions, and carry out the second mensuration region in each detection zone after differentiation comparison, the result that differentiation is compared is compared with the threshold value of corresponding detection zone, if result of the comparison meets preparatory condition, that is, obtain defect existing in corresponding detection zone.The present invention obtains the defect counts corresponding from different detection zones by setting different threshold values to different detection zones so that more residual polycrystalline silicon defects are captured to, and improve the degree of accuracy of detection.
Description
Technical field
The present invention relates to field of semiconductor manufacture, and in particular to a kind of detection method of wafer defect.
Background technology
In field of semiconductor manufacture, wafer to the defects count on wafer after chemical mechanical polishing, it is necessary to be scanned
Judge, the whether satisfactory purpose of technique is examined to reach.
In actual production, the grinding head of work-table of chemicomechanical grinding mill can control difference by the different zones to wafer
Pressure optimizes the flatness of wafer.During being ground to wafer, grinding head is pressed downward wafer to be occurred with grinding table
Relative motion, because grinding head is different to the pressure of wafer different zones, the grinding thickness for easily causing some regions of wafer is poor
Different big, the grinding thickness difference in other regions of wafer is small.And then detection device to wafer when being scanned detection, difference in thickness
The chromatic aberration defect number in big region will often be more than the chromatic aberration defect number in the small region of difference in thickness.Fig. 1 is coherent detection
The distribution schematic diagram for the wafer surface defects that device scan is arrived, as shown in figure 1, after being scanned to crystal column surface, by the wafer
The image that surface is presented is divided for four detection zones, is detection zone 201, detection zone 202, detection zone respectively
203 and detection zone 204.Obviously, the defect counts in detection zone 201, detection zone 202 and detection zone 203 are equal
Less than the defect counts in the detection zone 204 of crystal round fringes, the stain in Fig. 1 is the schematic diagram of defect.
At present when carrying out defects detection, it is common practice to table will be used in the serious detection zone 204 of chromatic aberration defect
The threshold value of defect is levied as the baseline threshold for searching defect in whole wafer.Although can so reduce aberration to greatest extent to lack
Fall into, but largely also filtered out the other defect on wafer, especially residual polycrystalline silicon defect, therefore, defect
The degree of accuracy of detection is low.It is additionally, since that some regions aberration is too serious, is easily caused the scanning that defect counts reach detection device
The upper limit and scan termination, so as to require high to the hardware configuration of detection device, be unfavorable for defects detection.
The content of the invention
It is an object of the invention to provide a kind of detection method of wafer defect, to solve in the prior art largely
On filtered out other defect in addition to chromatic aberration defect, and some detection zones due to defect counts it is excessive, cause detection
The problem of device scan is terminated.
To achieve these goals and other purposes, the invention provides a kind of detection method of wafer defect, including：
The image of crystal column surface is obtained, and according to the distribution situation of crystal column surface pressure, by the crystal column surface
Image is divided into multiple detection zones；
Choose multiple first mensuration regions in each detection zone, and by first in each detection zone
Mensuration region carries out differentiation comparison, and the result compared according to differentiation, which obtains each detection zone, to be used to characterize defect
Threshold value；
Each detection zone is divided into multiple second mensuration regions, and by second in each detection zone
Mensuration region is carried out after differentiation comparison, and the result that differentiation is compared is compared with the threshold value of corresponding detection zone, if
Result of the comparison meets preparatory condition, that is, obtains defect existing in corresponding detection zone.
Optionally, multiple reference zones are chosen in each detection zone, each reference zone is by multiple the
One mensuration region is constituted, and the center of the multiple reference zone is symmetrical on the center of the image of crystal column surface.
Optionally, the differentiation comparison result of the first mensuration region in each detection zone meets threshold value and defect
The functional relation of quantity, the threshold value of each detection zone is obtained by inquiring about the functional relation of threshold value and defects count.
Optionally, the reference zone is eight.
Optionally, the threshold value is gray value, and it is poor that first mensuration region by each detection zone is carried out
The process that alienation is compared includes：
Step one：Using the center of first mensuration region as first origin of coordinates, the first plane coordinate system is set up；
Step 2：Take with the center of one first mensuration region on same straight line and adjacent another first
The center of mensuration region is second origin of coordinates, sets up second plane coordinates consistent with the first plane coordinate system direction
System；
Step 3：Take with the center of one first mensuration region on same straight line and adjacent another first
The center of mensuration region is the 3rd origin of coordinates, sets up threeth plane coordinates consistent with the first plane coordinate system direction
In system, one first mensuration region, another described first mensuration region and another described first mensuration region
Heart line is on same straight line；
Step 4：The first pixel is taken in one first mensuration region, in another described first mensuration region
The second pixel is inside taken, the 3rd pixel, and first pixel, described are taken in another described first mensuration region
The coordinate of two pixels and the 3rd pixel in corresponding coordinate system is identical；
Step 5：By the gray value of first pixel respectively with the gray value of second pixel and described
The gray value of three pixels subtracts each other, to obtain two differences；
Wherein, the first mensuration region in each detection zone is more than three, for any three it is adjacent and in
First mensuration region of the heart on same straight line performs step one to step 5, to obtain two or more difference；With according to many
Individual difference obtains the threshold of correspondence detection zone.
Optionally, the process of the threshold value of correspondence detection zone is obtained according to multiple differences to be included：
Take the absolute value of the multiple difference, defects count determined by the absolute value of the multiple difference meet threshold value with
The functional relation of defects count, with the functional relation according to the threshold value and defects count, obtains the correspondence detection zone
Threshold value.
Optionally, the threshold value is gray value, and it is poor that second mensuration region by each detection zone is carried out
The process that alienation is compared includes：
Step one：Using the center of second mensuration region as first origin of coordinates, the first plane coordinate system is set up；
Step 2：Take with the center of one second mensuration region on same straight line and adjacent another second
The center of mensuration region is second origin of coordinates, sets up second plane coordinates consistent with the first plane coordinate system direction
System；
Step 3：Take with the center of one second mensuration region on same straight line and adjacent another second
The center of mensuration region is the 3rd origin of coordinates, sets up threeth plane coordinates consistent with the first plane coordinate system direction
In system, one second mensuration region, another described second mensuration region and another described second mensuration region
Heart line is on same straight line；
Step 4：The first pixel is taken in one second mensuration region, in another described second mensuration region
The second pixel is inside taken, the 3rd pixel, and first pixel, described are taken in another described second mensuration region
The coordinate of two pixels and the 3rd pixel in corresponding coordinate system is identical；
Step 5：By the gray value of first pixel respectively with the gray value of second pixel and described
The gray value of three pixels subtracts each other, to obtain two differences；
Wherein, the second mensuration region in each detection zone is more than three, for any three centers same
On straight line and adjacent the second mensuration region performs step one to step 5, to obtain two or more difference；
Afterwards, each differentiation is compared to the obtained absolute value of two differences with the threshold value of corresponding scanning area to be compared
Compared with if comparative result meets the preparatory condition, that is, judging corresponding first pixel as defect.
Optionally, the preparatory condition is：
Each differentiation compares the threshold value that two obtained respective absolute values of difference are all higher than the correspondence scanning area.
Optionally, obtain after multiple the first pixels for defect, adjacent first pixel is merged, it is same to obtain
One defect area.
Optionally, the multiple detection zone at least includes：
Positioned at the border circular areas of wafer surface inconocenter；And
Multiple annular regions concentric with the border circular areas positioned at border circular areas periphery.
Compared with prior art, the detection method for the wafer defect that the present invention is provided, by being examined for the difference of crystal column surface
Survey region and set different threshold values, so as to obtain the defect counts mutually echoed with corresponding detection zone, such way causes crystalline substance
Different defects, especially chromatic aberration defect and residual polycrystalline silicon defect can be captured on circular surfaces, improve defect
The degree of accuracy of detection, by adjusting production technology, also improves product yield.Meanwhile, detection method of the invention passes through to not
Rational threshold value is set with detection zone so that the defect counts in each detection zone are unlikely to excessive, reduce to correlation
The hardware configuration requirement of detection device, associated assay devices will not be excessive because of defect counts, reaches the scanning upper limit and causes scanning
Terminate, therefore, the detection method is easily achieved on hardware.
Brief description of the drawings
Fig. 1 is the distribution schematic diagram for the wafer surface defects that associated assay devices are scanned；
Fig. 2 is the flow chart of the detection method of wafer defect provided in an embodiment of the present invention；
Fig. 3 is the distribution schematic diagram of detection zone on the image of crystal column surface provided in an embodiment of the present invention；
Fig. 4 is the schematic diagram provided in an embodiment of the present invention that reference zone is chosen on the image of crystal column surface；
Fig. 5 is the schematic diagram that mensuration region provided in an embodiment of the present invention carries out differentiation comparison；
Fig. 6 be it is provided in an embodiment of the present invention by it is multiple be that the pixel of defect merges into the signal of a defect area
Figure；
Fig. 7 is the function relation figure of threshold value provided in an embodiment of the present invention and defects count
Description of reference numerals is as follows：
201/202/203/204- detection zones；
301/302/303/304- grinding pressures region；
2041/2042/2043/2044/2045/2046/2047/2048- reference zones；
The mensuration region of 501/502/503- the first mensuration region/second；
601/602/603/604- pixels.
Embodiment
The embodiment of the present invention is described in more detail below in conjunction with schematic diagram.According to description below and
Claims, advantages and features of the invention will become apparent from.It should be noted that, accompanying drawing uses very simplified form and equal
Using non-accurately ratio, only for the purpose of facilitating and clarifying the purpose of the embodiments of the invention.
Fig. 2 is the flow chart of the detection method of wafer defect provided in an embodiment of the present invention, as shown in Fig. 2 the wafer lacks
Sunken detection method includes step S101, step S102 and step S103.
Wherein, step S101：The image of crystal column surface is obtained, and according to the distribution situation of crystal column surface pressure, will
The image of the crystal column surface is divided into multiple detection zones.
Specifically, during due to Poly CMP (multi-crystal silicon floating bar cmp), grinding head is by one, center circle
Shape region, the multiple annular region compositions concentric with border circular areas in periphery.When being ground to wafer, grinding head can be according to certainly
Body is constituted, and the different zones to wafer control different pressure, to be ground to wafer.It is subject to due to wafer different zones
Pressure is different, and the flatness that such lapping mode often results in wafer different zones is different.Normally, detection device is to wafer
Surface is scanned after the image for obtaining crystal column surface, the detection device further according to wafer area pressure distribution situation,
Different detection zones are marked off for the image of the crystal column surface.
For example, as shown in figure 3, the image of the crystal column surface of the present embodiment is divided into following detection zone：In being located at
A round pressure distributed areas at the heart, as detection zone 201, while being also the region of grinding pressure 301；It is close to inspection
The region in region 201 is surveyed, is the detection zone 202 of an annular, while being also the region of grinding pressure 302；It is close to detection
The region in region 202, is another annular detection zone 203, while being also the region of grinding pressure 303；Then, it is close to
The region of detection zone 203, is another annular detection zone 204, while being also the region of grinding pressure 304.
Step S102：Multiple first mensuration regions are chosen in each detection zone, and will each detection zone
The first mensuration region in domain carries out differentiation comparison, and the result compared according to differentiation obtains the threshold of each detection zone
Value.
Specifically, the threshold value is used to characterize defect, such as defect is characterized using the gray value of image, and the threshold value is
Gray value.First mensuration region is in corresponding detection zone, detection device is scanned acquisition crystal column surface to wafer
Image after, the minimum image-region delimited on the image of crystal column surface, such as the square or square including some pixels
Shape region, the size of the minimum image-region is determined according to the precision of detection device.More particularly, to wafer defect
During carrying out compartmentalization detection, defect is searched by setting a desired value for each detection zone, the desired value is
For the threshold value in step 102.
Detection device is drawn in each detection zone by reasonably choosing multiple reference zones, then by each reference zone
It is divided into the region being made up of multiple first mensuration regions, afterwards, these first mensuration regions is separately subjected to differentiation comparison, according to
Comparison result, is that each detection zone determines a rational threshold value.It is preferred that, multiple reference zones in each detection zone
Center it is symmetrical on the picture centre of crystal column surface.It is furthermore preferred that the quantity of the reference zone in each detection zone
For eight.
In the present embodiment, detection device is by center on same straight line and adjacent any three the first mensuration regions are entered
Row differentiation is compared.It should be appreciated that each first mensuration region is made up of some pixels, by being surveyed corresponding first
Determine region to choose a pixel and each choose a pixel in other two the first mensuration regions adjacent thereto
Differentiation comparison (i.e. three pixels) is carried out, the result after differentiation is compared is obtained, the result can be gray value difference
Absolute value.
Step S103：Each detection zone is divided into multiple second mensuration regions, and will each detection zone
The second mensuration region in domain carries out differentiation comparison, and the threshold value of the result after differentiation is compared and corresponding detection zone
It is compared, if result of the comparison meets preparatory condition, you can obtain defect existing in corresponding detection zone.
Specifically, second mensuration region is in correspondence detection zone, it is brilliant that detection device is scanned acquisition to wafer
After the image of circular surfaces, the minimum image-region delimited on the image of crystal column surface, i.e. the second mensuration region include some
Pixel, its can be square region can also be rectangle region.That is, the size of the minimum image-region
Determined according to the precision of detection device, after the image of crystal column surface is obtained, detection device is by each detection zone in image
The region being made up of multiple second mensuration regions is divided into, and these the second mensuration region arrays are set.
Further, detection device by center on same straight line and adjacent any three the second mensuration regions are carried out
Differentiation is compared, if the result that differentiation comparison is obtained every time is all higher than the threshold value of corresponding detection zone setting, detection is set
It is standby that the corresponding pixel of the second mensuration region is just defined as defect.Should know, if each second mensuration region by
Dry pixel composition, by corresponding second mensuration region choose a pixel and with other two adjacent thereto second
The pixel each chosen in mensuration region carries out differentiation comparison, the result of differentiation comparison is obtained, if the result is equal
During more than threshold value, then judge corresponding one pixel in centre for defect.With the differentiation comparison process one of the first mensuration region
Sample, the result that the second mensuration region differentiation is compared can be the absolute value of gray value difference.
In the present embodiment, the differentiation comparison result of the first mensuration region in each detection zone meets threshold value and defect
The functional relation of quantity, the threshold value of each detection zone is obtained by inquiring about the functional relation of the threshold value and defects count.Example
Such as, referring to Fig. 7, Fig. 7 is the function relation figure of threshold value provided in an embodiment of the present invention and defects count, and Fig. 7 is drawn by testing.Such as
Shown in Fig. 7, OX axles are threshold value, and OY axles are defects count, and curve 701 is the functional relation of threshold value and residual polycrystalline silicon defects count
Curve；Curve 702 is the function relation curve of threshold value and chromatic aberration defect quantity, wherein, A points are the flex point on curve 702, threshold value
Selected point is A points point corresponding in X-axis.
The detection method for the wafer defect that the present embodiment is provided, according to pressure of the chemical mechanical grinding head to wafer different zones
Power distribution situation, different detection zones are marked off to wafer, are set corresponding threshold value for each detection zone, are made each inspection
Surveying the different defects in region can be captured to, it is to avoid in conventional methods where, with crystal round fringes be used for characterize defect
Threshold value is the threshold value of whole detection zone, so that causing the residual polycrystalline silicon defect of wafer some regions can not be captured by more
The problem of.Meanwhile, by setting rational threshold value to different detection zones so that the defect counts of each detection zone are unlikely to
Excessively, the requirement to detection device hardware configuration is reduced, therefore, the detection method is easily achieved on hardware.
On the basis of above-described embodiment, in each detection zone, multiple reference zones are preferably along the circumferential direction uniform
Distribution, and ten first mensuration regions are chosen in each reference zone respectively.By multiple ginsengs in each detection zone
Examination district domain is uniformly distributed circumferentially, and contributes to the threshold value obtained to be capable of the truth of closer correspondence detection zone, and lifting lacks
Fall into the degree of accuracy of detection.
In other embodiments, a reference zone, but a reference area can also be chosen in each detection zone
Domain is made up of multiple first mensuration regions.
As shown in figure 4, in detection zone 201, detection zone 202, detection zone 203 and detection zone 204 respectively
It has chosen eight reference zones.By taking detection zone 204 as an example, respectively choose eight reference zones, respectively reference zone 2041,
Reference zone 2042, reference zone 2043, reference zone 2044, reference zone 2045, reference zone 2046, reference zone 2047
And reference zone 2048, and reference zone 2041, reference zone 2042, reference zone 2043, reference zone 2044, reference area
Constituted in domain 2045, reference zone 2046, reference zone 2047 and reference zone 2048 by ten the first mensuration regions.
And then, in step s 102, the first mensuration region in each detection zone is carried out to the process of differentiation comparison
Specifically include：
Step one：Using the center of first mensuration region as first origin of coordinates, the first plane coordinate system is set up；Such as
Right-angle plane coordinate system, it has transverse axis and the longitudinal axis vertical with transverse axis；
Step 2：Take with the center of one first mensuration region on same straight line and adjacent another first
The center of mensuration region is second origin of coordinates, sets up second plane coordinates consistent with the first plane coordinate system direction
System；
Step 3：Take with the center of one first mensuration region on same straight line and adjacent another first
The center of mensuration region is the 3rd origin of coordinates, sets up threeth plane coordinates consistent with the first plane coordinate system direction
In system, one first mensuration region, another described first mensuration region and another described first mensuration region
Heart line is on same straight line, and another described first mensuration region and another described first mensuration region are positioned at described
The both sides of one the first mensuration region；
Step 4：The first pixel is taken in one first mensuration region, in another described first mensuration region
The second pixel is inside taken, the 3rd pixel, and first pixel, described are taken in another described first mensuration region
The coordinate of two pixels and the 3rd pixel in corresponding coordinate system is identical；
Step 5：By the gray value of first pixel respectively with the gray value of second pixel and described
The gray value of three pixels subtracts each other, to obtain two differences.
Further, when the first mensuration region in each detection zone is more than three, for any three phases
First mensuration region of the Lin Qie centers on same straight line performs step one to step 5, to obtain two or more difference.
Further, first origin of coordinates, second origin of coordinates and the 3rd origin of coordinates can be with
, can also be on the edge point of corresponding first mensuration region at the center of corresponding first mensuration region.In addition, above-mentioned steps
One the differentiation of the sequencing performed is not present to step 3.
For example, as shown in figure 5, take three the first mensuration regions 501,502,503 in reference zone 2041, wherein,
Four summits that one mensuration region 501 has, respectively A, B, E and F；First mensuration region 502 separately has four summits, point
Wei not B, C, G and F；First mensuration region 503 has four summits, respectively C, D, G and H.Using A points as the first measurement region
The origin in domain 501, the second plane coordinate system for setting up using AB as X-axis, by Y-axis of AE the first mensuration region 501；Similarly, with B points
The first plane coordinates of the first mensuration region 502 is set up for the origin of the first mensuration region 502, using BC as X-axis, by Y-axis of BF
System；Similarly, using origin that C points are the first mensuration region 503, the first mensuration region 503 is set up using CD as X-axis, by Y-axis of CG
3rd plane coordinate system.Take three pixels A1, B1 and C1 of coordinate identical in three plane coordinate systems.And by B1 points
The gray value of gray value and A1 points subtracts each other, and obtains a difference；Whether so may determine that pixel B1 is defect.Similarly, if
Whether be defect, can all be operated as above the different pixels point in the first mensuration region 502 if judging pixel A1, can be with
Obtain two other difference.Each first mensuration region in mode successively, reference zone 2041 carries out aforesaid operations, can obtain
To many differences.Similarly, each first mensuration region of the reference zone 2041 into reference zone 2048 is carried out above-mentioned
Operation, can obtain many differences.
In one embodiment, because the absolute value of these differences meets the functional relation of threshold value and defects count, therefore,
The function relation figure that threshold value and defects count can be inquired about obtains the threshold value of reference zone 2041,
In step s 103, the second mensuration region in each detection zone is carried out to the process bag of differentiation comparison
Include：
Step one：Using the center of second mensuration region as first origin of coordinates, the first plane coordinate system is set up；
Step 2：Take with the center of one second mensuration region on same straight line and adjacent another second
The center of mensuration region is second origin of coordinates, sets up second plane coordinates consistent with the first plane coordinate system direction
System；
Step 3：Take with the center of one second mensuration region on same straight line and adjacent another second
The center of mensuration region is the 3rd origin of coordinates, sets up threeth plane coordinates consistent with the first plane coordinate system direction
In system, one second mensuration region, another described second mensuration region and another described second mensuration region
Heart line is on same straight line, and another described second mensuration region and another described second mensuration region are positioned at described
The both sides of one the second mensuration region；
Step 4：The first pixel is taken in one second mensuration region, in another described second mensuration region
The second pixel is inside taken, the 3rd pixel, and first pixel, described are taken in another described second mensuration region
The coordinate of two pixels and the 3rd pixel in corresponding coordinate system is identical；
Step 5：By the gray value of first pixel respectively with the gray value of second pixel and described
The gray value of three pixels subtracts each other, to obtain two differences.
When the second mensuration region in each detection zone is more than three, for any three centers same
On bar straight line and adjacent the second mensuration region performs step one to step 5, to obtain two or more difference.Similarly, herein
The step of one priority being not carried out to step 3 distinguish.
Differentiation herein compares the implementation process compared referring specifically to the differentiation of the first mensuration region, no longer has herein
State.
After more than two differences are obtained, the process for obtaining corresponding defect is：Differentiation comparison each time is obtained
Two differences take respective absolute value, if the absolute value of two differences is all higher than the threshold value of corresponding detection zone, that is, judge
The first pixel for carrying out differentiation comparison is defect.First pixel be in differentiation comparison process each time in
Between pixel.
For example, as shown in Figure 5, it is assumed that the threshold value of one of detection zone is 20.Intermediary image vegetarian refreshments B1 gray value is
80, the pixel A1 positioned at the left side adjacent with intermediary image vegetarian refreshments gray value are 50, and adjacent with intermediary image vegetarian refreshments is located at the right side
The pixel C1 on side gray value is 120.The gray value of A1 pixels and the absolute value of the difference of the gray value of B1 pixels are
The absolute value of the difference of the gray value of 30, A1 pixels and the gray value of C1 pixels is 40, the absolute value 30 of two differences,
40 are all higher than the threshold value 20 of the detection zone, it may be determined that intermediary image vegetarian refreshments B1 is defect.
Further, as shown in fig. 6, pixel 601, pixel 602, pixel 603 and pixel 604 pass through difference
Change to compare and is judged as defect pixel point, and defect pixel point 601, defect pixel point 602, defect pixel point 603 and lack
Fall into pixel 604 adjacent in mode as shown in Figure 6, then by defect pixel point 601, defect pixel point 602, defect pixel point
603 and defect pixel point 604 merge into same defect.So the area of defect area can be made sufficiently large, be easy to human eye
Observation.
Present pre-ferred embodiments as described above, but be not limited to the scope disclosed in above-described embodiment, such as according to crystalline substance
When circular surfaces pressure distribution situation divides detection zone, except centre is the form that center of circle areas outside is annular region
It is outer or middle for delta-shaped region or square region.In addition, the quantity of detection zone is not limited to four.In addition, first
The shape of mensuration region and the second mensuration region is not limited to rectangle or square.
In summary, the detection method for the wafer defect that the present invention is provided, is examined by the difference of the image for crystal column surface
Survey region and set different threshold values, so as to obtain the defect counts mutually echoed with corresponding detection zone, such way causes crystalline substance
Different defects, especially chromatic aberration defect and residual polycrystalline silicon defect can be captured on circular surfaces, improve defect
The degree of accuracy of detection, by adjusting production technology, also improves product yield.Meanwhile, detection method of the invention passes through to not
Rational threshold value is set with detection zone so that the defect counts in each detection zone are unlikely to excessive, reduce to correlation
The hardware configuration requirement of detection device, makes detection device not cause scanning eventually because defect counts excessively reach the scanning upper limit
Only, therefore, the detection method is easily achieved on hardware.
The preferred embodiments of the present invention are above are only, any restriction effect is not played to the present invention.Belonging to any
Those skilled in the art, in the range of technical scheme is not departed from, to the invention discloses technical scheme and
Technology contents make the variation such as any type of equivalent substitution or modification, belong to the content without departing from technical scheme, still
Belong within protection scope of the present invention.
Claims (10)
1. a kind of detection method of wafer defect, it is characterised in that including：
The image of crystal column surface is obtained, and according to the distribution situation of crystal column surface pressure, by the image of the crystal column surface
It is divided into multiple detection zones；
Multiple first mensuration regions are chosen in each detection zone, and first in each detection zone is determined
Region carries out differentiation comparison, and the result compared according to differentiation obtains the threshold that each detection zone is used to characterize defect
Value；
Each detection zone is divided into multiple second mensuration regions, and second in each detection zone is determined
Region is carried out after differentiation comparison, and the result that differentiation is compared is compared with the threshold value of corresponding detection zone, if comparing
Result meet preparatory condition, that is, obtain in corresponding detection zone existing defect.
2. the detection method of wafer defect as claimed in claim 1, it is characterised in that chosen in each detection zone
Multiple reference zones, each reference zone is made up of multiple first mensuration regions, and the center of the multiple reference zone
Center on the image of crystal column surface is symmetrical.
3. the detection method of wafer defect as claimed in claim 2, it is characterised in that first in each detection zone
The differentiation comparison result of mensuration region meets the functional relation of threshold value and defects count, by inquiring about threshold value and defects count
Functional relation obtains the threshold value of each detection zone.
4. the detection method of wafer defect as claimed in claim 2, it is characterised in that the reference zone is eight.
5. the detection method of wafer defect as claimed in claim 1, it is characterised in that the threshold value is gray value, described to incite somebody to action
The process that the first mensuration region in each detection zone carries out differentiation comparison includes：
Step one：Using the center of first mensuration region as first origin of coordinates, the first plane coordinate system is set up；
Step 2：Take with the center of one first mensuration region on same straight line and adjacent another first is determined
The center in region is second origin of coordinates, sets up second plane coordinate system consistent with the first plane coordinate system direction；
Step 3：Take with the center of one first mensuration region on same straight line and adjacent another first is determined
The center in region is the 3rd origin of coordinates, sets up threeth plane coordinate system consistent with the first plane coordinate system direction, institute
State the line of centres of first mensuration region, another described first mensuration region and another first mensuration region
On same straight line；
Step 4：The first pixel is taken in one first mensuration region, is taken in another described first mensuration region
Second pixel, takes the 3rd pixel, and first pixel, second picture in another described first mensuration region
The coordinate of vegetarian refreshments and the 3rd pixel in corresponding coordinate system is identical；
Step 5：By the gray value and the 3rd picture of the gray value of first pixel respectively with second pixel
The gray value of vegetarian refreshments subtracts each other, to obtain two differences；
Wherein, the first mensuration region in each detection zone is more than three, adjacent and center exists for any three
The first mensuration region on same straight line performs step one to step 5, to obtain two or more difference；With according to multiple differences
Value obtains the threshold of correspondence detection zone.
6. the detection method of wafer defect as claimed in claim 5, it is characterised in that obtain correspondence detection according to multiple differences
The process of the threshold value in region includes：
The absolute value of the multiple difference is taken, defects count determined by the absolute value of the multiple difference meets threshold value and defect
The functional relation of quantity, with the functional relation according to the threshold value and defects count, obtains the threshold value of the correspondence detection zone.
7. the detection method of wafer defect as claimed in claim 1, it is characterised in that the threshold value is gray value, described to incite somebody to action
The process that the second mensuration region in each detection zone carries out differentiation comparison includes：
Step one：Using the center of second mensuration region as first origin of coordinates, the first plane coordinate system is set up；
Step 2：Take with the center of one second mensuration region on same straight line and adjacent another second is determined
The center in region is second origin of coordinates, sets up second plane coordinate system consistent with the first plane coordinate system direction；
Step 3：Take with the center of one second mensuration region on same straight line and adjacent another second is determined
The center in region is the 3rd origin of coordinates, sets up threeth plane coordinate system consistent with the first plane coordinate system direction, institute
State the line of centres of second mensuration region, another described second mensuration region and another second mensuration region
On same straight line；
Step 4：The first pixel is taken in one second mensuration region, is taken in another described second mensuration region
Second pixel, takes the 3rd pixel, and first pixel, second picture in another described second mensuration region
The coordinate of vegetarian refreshments and the 3rd pixel in corresponding coordinate system is identical；
Step 5：By the gray value and the 3rd picture of the gray value of first pixel respectively with second pixel
The gray value of vegetarian refreshments subtracts each other, to obtain two differences；
Wherein, the second mensuration region in each detection zone is more than three, for any three centers in same
On straight line and adjacent the second mensuration region performs step one to step 5, to obtain two or more difference；
Afterwards, the absolute value for each differentiation being compared into two obtained differences is compared with the threshold value of corresponding scanning area,
If comparative result meets the preparatory condition, that is, judge corresponding first pixel as defect.
8. the detection method of wafer defect as claimed in claim 7, it is characterised in that the preparatory condition is：
Each differentiation compares the threshold value that two obtained respective absolute values of difference are all higher than the correspondence scanning area.
9. the detection method of wafer defect as claimed in claim 8, it is characterised in that obtain multiple the first pixels for defect
After point, adjacent first pixel is merged, to obtain same defect area.
10. the detection method of wafer defect as claimed in claim 1, it is characterised in that the multiple detection zone is at least wrapped
Include：
Positioned at the border circular areas of wafer surface inconocenter；And
Multiple annular regions concentric with the border circular areas positioned at border circular areas periphery.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710353548.8A CN106971955B (en) | 2017-05-18 | 2017-05-18 | Wafer defect detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710353548.8A CN106971955B (en) | 2017-05-18 | 2017-05-18 | Wafer defect detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106971955A true CN106971955A (en) | 2017-07-21 |
| CN106971955B CN106971955B (en) | 2019-12-24 |
Family
ID=59326026
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710353548.8A Active CN106971955B (en) | 2017-05-18 | 2017-05-18 | Wafer defect detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106971955B (en) |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107784660A (en) * | 2017-11-02 | 2018-03-09 | 武汉新芯集成电路制造有限公司 | Image processing method, image processing system and defect detecting device |
| CN107907821A (en) * | 2017-10-28 | 2018-04-13 | 天津大学 | The advanced method of discrimination of attack detecting sensor threshold value for safety chip |
| CN107910276A (en) * | 2017-11-24 | 2018-04-13 | 上海华力微电子有限公司 | A kind of wafer defect detection method |
| CN108362780A (en) * | 2017-12-07 | 2018-08-03 | 中航复合材料有限责任公司 | A kind of method of evaluating combined material structure Zone R defect ultrasound detection probability |
| CN108776966A (en) * | 2018-06-12 | 2018-11-09 | 成都银河磁体股份有限公司 | A kind of method and system of magnet appearance defect detection |
| CN109585325A (en) * | 2018-12-05 | 2019-04-05 | 上海华力集成电路制造有限公司 | The method for detecting crystal edge side washing boundary |
| CN109978839A (en) * | 2019-03-08 | 2019-07-05 | 浙江大学 | The detection method of the low texture defect of wafer |
| CN109994398A (en) * | 2019-04-18 | 2019-07-09 | 上海华力微电子有限公司 | A kind of wafer defect scanning control methods |
| CN110190007A (en) * | 2019-05-07 | 2019-08-30 | 上海华虹宏力半导体制造有限公司 | It is a kind of for investigating the localization method of backside of wafer exception |
| CN110189300A (en) * | 2019-04-22 | 2019-08-30 | 中国科学院微电子研究所 | Detection method, detection device, storage medium and the processor of pass structure processing quality |
| CN110487801A (en) * | 2019-07-30 | 2019-11-22 | 厦门三安光电有限公司 | Identify method, apparatus, equipment and the storage medium of graphite plate high-temperature baking result |
| CN111239152A (en) * | 2020-01-02 | 2020-06-05 | 长江存储科技有限责任公司 | Wafer detection method, device and equipment |
| CN111855663A (en) * | 2019-04-30 | 2020-10-30 | 芯恩（青岛）集成电路有限公司 | Equipment and method for detecting wafer defects |
| WO2020224612A1 (en) * | 2019-05-07 | 2020-11-12 | 徐州鑫晶半导体科技有限公司 | Method and system for automated detection and control of defects on wafer |
| CN112345544A (en) * | 2020-09-24 | 2021-02-09 | 海克斯康制造智能技术(青岛)有限公司 | Multifunctional connecting rod automatic detection line |
| CN113838003A (en) * | 2021-08-30 | 2021-12-24 | 歌尔科技有限公司 | Speckle detection method, device, medium, and computer program product for image |
| CN113901286A (en) * | 2021-12-09 | 2022-01-07 | 广州粤芯半导体技术有限公司 | Processing system and processing method for detection data and detection system for substrate |
| WO2023109463A1 (en) * | 2021-12-14 | 2023-06-22 | 盛美半导体设备(上海)股份有限公司 | Electroplating cavity skip plating warning method and system |
| CN116912242A (en) * | 2023-09-12 | 2023-10-20 | 青岛天仁微纳科技有限责任公司 | Method for positioning defects of nanoimprint wafer |
| CN117046735A (en) * | 2023-10-12 | 2023-11-14 | 江苏芯德半导体科技有限公司 | Method and system for merging and distributing patterns of multiple chips in wafer |
| TWI825721B (en) * | 2021-06-04 | 2023-12-11 | 日商Ｓｕｍｃｏ股份有限公司 | Wafer appearance inspection device and wafer appearance inspection method |
| CN113838003B (en) * | 2021-08-30 | 2024-04-30 | 歌尔科技有限公司 | Image speckle detection method, apparatus, medium and computer program product |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090304261A1 (en) * | 2008-06-10 | 2009-12-10 | Fujitsu Microelectronics Limited | Defect inspection apparatus, defect inspection method, and manufacture method for semiconductor device |
| CN101996908A (en) * | 2009-08-14 | 2011-03-30 | 中芯国际集成电路制造(上海)有限公司 | Method and device for detecting wafer |
| CN101996855A (en) * | 2009-08-21 | 2011-03-30 | 中芯国际集成电路制造(上海)有限公司 | Wafer defect analysis method |
-
2017
- 2017-05-18 CN CN201710353548.8A patent/CN106971955B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090304261A1 (en) * | 2008-06-10 | 2009-12-10 | Fujitsu Microelectronics Limited | Defect inspection apparatus, defect inspection method, and manufacture method for semiconductor device |
| CN101996908A (en) * | 2009-08-14 | 2011-03-30 | 中芯国际集成电路制造(上海)有限公司 | Method and device for detecting wafer |
| CN101996855A (en) * | 2009-08-21 | 2011-03-30 | 中芯国际集成电路制造(上海)有限公司 | Wafer defect analysis method |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107907821B (en) * | 2017-10-28 | 2020-01-14 | 天津大学 | Attack detection sensor threshold advanced discrimination method for security chip |
| CN107907821A (en) * | 2017-10-28 | 2018-04-13 | 天津大学 | The advanced method of discrimination of attack detecting sensor threshold value for safety chip |
| CN107784660A (en) * | 2017-11-02 | 2018-03-09 | 武汉新芯集成电路制造有限公司 | Image processing method, image processing system and defect detecting device |
| CN107784660B (en) * | 2017-11-02 | 2020-12-04 | 武汉新芯集成电路制造有限公司 | Image processing method, image processing system and defect detection device |
| CN107910276A (en) * | 2017-11-24 | 2018-04-13 | 上海华力微电子有限公司 | A kind of wafer defect detection method |
| CN108362780A (en) * | 2017-12-07 | 2018-08-03 | 中航复合材料有限责任公司 | A kind of method of evaluating combined material structure Zone R defect ultrasound detection probability |
| CN108776966A (en) * | 2018-06-12 | 2018-11-09 | 成都银河磁体股份有限公司 | A kind of method and system of magnet appearance defect detection |
| CN109585325A (en) * | 2018-12-05 | 2019-04-05 | 上海华力集成电路制造有限公司 | The method for detecting crystal edge side washing boundary |
| CN109585325B (en) * | 2018-12-05 | 2021-02-02 | 上海华力集成电路制造有限公司 | Method for detecting crystal edge washing boundary |
| CN109978839A (en) * | 2019-03-08 | 2019-07-05 | 浙江大学 | The detection method of the low texture defect of wafer |
| CN109978839B (en) * | 2019-03-08 | 2020-12-08 | 浙江大学 | Method for detecting wafer low-texture defects |
| CN109994398A (en) * | 2019-04-18 | 2019-07-09 | 上海华力微电子有限公司 | A kind of wafer defect scanning control methods |
| CN110189300A (en) * | 2019-04-22 | 2019-08-30 | 中国科学院微电子研究所 | Detection method, detection device, storage medium and the processor of pass structure processing quality |
| CN111855663A (en) * | 2019-04-30 | 2020-10-30 | 芯恩（青岛）集成电路有限公司 | Equipment and method for detecting wafer defects |
| WO2020224612A1 (en) * | 2019-05-07 | 2020-11-12 | 徐州鑫晶半导体科技有限公司 | Method and system for automated detection and control of defects on wafer |
| CN110190007A (en) * | 2019-05-07 | 2019-08-30 | 上海华虹宏力半导体制造有限公司 | It is a kind of for investigating the localization method of backside of wafer exception |
| CN110190007B (en) * | 2019-05-07 | 2021-04-02 | 上海华虹宏力半导体制造有限公司 | Positioning method for investigating wafer back surface abnormity |
| CN110487801A (en) * | 2019-07-30 | 2019-11-22 | 厦门三安光电有限公司 | Identify method, apparatus, equipment and the storage medium of graphite plate high-temperature baking result |
| CN111239152B (en) * | 2020-01-02 | 2023-11-17 | 长江存储科技有限责任公司 | Wafer detection method, device and equipment |
| CN111239152A (en) * | 2020-01-02 | 2020-06-05 | 长江存储科技有限责任公司 | Wafer detection method, device and equipment |
| CN112345544A (en) * | 2020-09-24 | 2021-02-09 | 海克斯康制造智能技术(青岛)有限公司 | Multifunctional connecting rod automatic detection line |
| TWI825721B (en) * | 2021-06-04 | 2023-12-11 | 日商Ｓｕｍｃｏ股份有限公司 | Wafer appearance inspection device and wafer appearance inspection method |
| CN113838003A (en) * | 2021-08-30 | 2021-12-24 | 歌尔科技有限公司 | Speckle detection method, device, medium, and computer program product for image |
| CN113838003B (en) * | 2021-08-30 | 2024-04-30 | 歌尔科技有限公司 | Image speckle detection method, apparatus, medium and computer program product |
| CN113901286A (en) * | 2021-12-09 | 2022-01-07 | 广州粤芯半导体技术有限公司 | Processing system and processing method for detection data and detection system for substrate |
| WO2023109463A1 (en) * | 2021-12-14 | 2023-06-22 | 盛美半导体设备(上海)股份有限公司 | Electroplating cavity skip plating warning method and system |
| CN116912242A (en) * | 2023-09-12 | 2023-10-20 | 青岛天仁微纳科技有限责任公司 | Method for positioning defects of nanoimprint wafer |
| CN116912242B (en) * | 2023-09-12 | 2023-12-15 | 青岛天仁微纳科技有限责任公司 | Method for positioning defects of nanoimprint wafer |
| CN117046735A (en) * | 2023-10-12 | 2023-11-14 | 江苏芯德半导体科技有限公司 | Method and system for merging and distributing patterns of multiple chips in wafer |
| CN117046735B (en) * | 2023-10-12 | 2023-12-12 | 江苏芯德半导体科技有限公司 | Method and system for merging and distributing patterns of multiple chips in wafer |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106971955B (en) | 2019-12-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106971955A (en) | The detection method of wafer defect | |
| CN102738029B (en) | Method for detecting specific defect and system used for detecting specific defect | |
| KR101551753B1 (en) | Cad a system a method and a computer program product for cadbased registration | |
| KR101860393B1 (en) | A system, a method and a computer program product for cadbased registration | |
| CN103885217B (en) | A kind of method detecting liquid crystal panel cylindrical spacer defect and device | |
| CN103499590B (en) | Ring-shaped work pieces end face defect detection and screening technique and system | |
| CN103646893A (en) | A wafer defect detecting method | |
| US10964014B2 (en) | Defect detecting method and defect detecting system | |
| US6512843B1 (en) | Pattern comparison method and appearance inspection machine for performance comparison based on double detection without delay | |
| CN103311148B (en) | A kind of method detecting observation defect | |
| US10516822B2 (en) | Method and device for merging images of calibration devices | |
| KR20000034922A (en) | Removal of noise from a signal obtained with an imaging system | |
| CN109029327A (en) | The production method of the arc profile tolerance test method and 2.5D cover board of 2.5D cover board sample | |
| CN117058151A (en) | Hub detection method and system based on image analysis | |
| CN105014317A (en) | Manufacturing method of ring gauge | |
| CN211601854U (en) | Detection device | |
| KR20180116406A (en) | An inspection information generation device, an inspection information generation method, and a defect inspection device | |
| JP6809626B1 (en) | DIC defect shape measurement method and polishing method for silicon wafers | |
| JP6206380B2 (en) | Evaluation method of silicon single crystal wafer | |
| CN111863646A (en) | Method for detecting defects of semiconductor device | |
| JP2017034025A (en) | Processing damage evaluation method of semiconductor wafer | |
| JP4695106B2 (en) | Method for determining height of chuck top and program recording medium recording this method | |
| CN104198509B (en) | Photomask graphic defect detection system and method | |
| US20240047278A1 (en) | Method for detecting back surface of wafer | |
| KR200292152Y1 (en) | Dimension Inspection Unit of Injected E.V.A Midsole |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |