Method of measuring the probability of failure caused only by defects, method of measuring defect limited yield, and system using the same

• US 6,985,830 B2
• Filed: 11/06/2002
• Issued: 01/10/2006
• Est. Priority Date: 12/20/2001
• Status: Expired due to Fees

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Lee, Jae-cheol, Lee, Dae-sung, Son, Gook-tae, Kim, Jung-hee
• Primary Examiner(s): Hoff, Marc S.
• Assistant Examiner(s): Gutierrez, Anthony
• Application Number:
  US10/288,424
• Publication Number:
  US 20030120459A1
• Time in Patent Office: 1,161 Days
• Field of Search: 702/181, 438/16, 438/17, 438/14
• US Class Current: 702/181
• CPC Class Codes: H01L 22/20 Sequence of activities consisting of a plurality of measurements, corrections, marking or sorting steps

Abstract

A method of calculating a probability of failures caused only by defects, a method of calculating a defect limited yield using the classification of pattern parameters extracted only from the defects, and a system for calculating the probability of failure and the defect limited yield are provided. In one exemplary embodiment for calculating a probability of failures caused only by defects, defects are detected in inspected blocks that have defects and in blocks located around the inspected blocks to measure the number of inspected blocks that have failures caused by reasons other than the defects in the blocks located around the inspected blocks having defects (n1), the number of inspected blocks having no failures in the blocks located around the inspected blocks having the defects (n2), the number of inspected blocks having failures caused by defects in the inspected blocks having defects (n3), and the number of inspected blocks having no failures in the inspected blocks having defects (n4). The data (n1) through (n4) is then substituted in the following formula: $\mathrm{KR}=1-\frac{1-{\mathrm{KR}}_1}{1-{\mathrm{KR}}_0},$
where ${\mathrm{KR}}_0=\frac{\mathrm{n1}}{\mathrm{n1}+\mathrm{n2}}$
and ${\mathrm{KR}}_1=\frac{\mathrm{n3}}{\mathrm{n3}+\mathrm{n4}}.$

Claims

1. A method of calculating a probability of failures (KR) caused only by defects in blocks in a wafer chip comprising:

inspecting said blocks to detect defects in said blocks and in blocks located around said inspected blocks;

measuring the number of said inspected blocks having failures caused by a reason other than said defects in said inspected blocks located around said inspected blocks having said defects (n1);

measuring the number of said inspected blocks having no failures in said inspected blocks located around said inspected blocks having said defects (n2);

measuring the number of inspected blocks having failures caused by said defects in said inspected blocks having said defects (n3);

measuring the number of inspected blocks having no failures in said inspected blocks having said defects (n4); and

substituting data (n1) through (n4) in formula 1a $\begin{array}{cc}\mathrm{KR}=1-\frac{1-{\mathrm{KR}}_1}{1-{\mathrm{KR}}_0};<br><br><br><br><br><br>\mathrm{wherein}<br><br><br><br>{\mathrm{KR}}_0=\frac{\mathrm{n1}}{\mathrm{n1}+\mathrm{n2}}<br><br><br><br>\mathrm{and}<br><br><br><br>{\mathrm{KR}}_1=\frac{\mathrm{n3}}{\mathrm{n3}+\mathrm{n4}}.& \left(1<br><br>a\right)\end{array}$

2. The method of claim 1, wherein said inspected blocks located around said inspected blocks are positioned in upper left, upper right, lower left, and lower right directions from said inspected blocks.

3. The method of claim 1, wherein said inspected blocks located around said inspected blocks are eight inspected blocks surrounding inspected block.

4. The method of claim 1, wherein when KR₀ is one, failures occur in all of said inspected blocks located around said inspected blocks having said defects.

5. The method of claim 1, wherein KR₀ represents a probability of failures caused by reasons other than said defects.

6. The method of claim 1, wherein KR₁ represents a probability of failures caused by said defects.

7. A method of calculating a probability of failures caused only by defects in each chip of a wafer comprising the steps of:

determining the total number of wafer chips (m1);

detecting defects in each said wafer chip;

determining the total number of chips having said defects (m2);

determining the total number of chips which fail due to a reason other than said defects (m3);

classifying said wafer chips having defects using probable parameters in consideration of area, size, and number of said defects;

determining the total number of wafer chips (m4) and the total number of wafer chips which fail (m5) when said wafer chips are classified according to each said probable parameter; and

calculating a probability (KR) that said wafer chips fail only due to said defects by substituting the data m1, m2, m3, m4, and m5 in the formula;

$\mathrm{KR}=1-\frac{1-{\mathrm{KR}}_1}{1-{\mathrm{KR}}_0},<br><br>\mathrm{wherein}<br><br><br><br>{\mathrm{KR}}_0=\frac{\mathrm{m3}}{\mathrm{m1}-\mathrm{m2}}<br><br><br><br>\mathrm{and}<br><br><br><br>{\mathrm{KR}}_1=\frac{\mathrm{m5}}{\mathrm{m4}}.$

8. The method of claim 7, wherein said inspection of said defects is performed by an array method which inspects said defects in cell regions.

9. The method of claim 8, wherein said probable parameters of said array method include a maximum area of defects in a chip, a maximum value of the sum of areas of said defects in a segment, a maximum value of the sum of sizes of said defects in said segment, a maximum value of the total number of said defects in said segment, a sum of areas of said defects in said chip, a sum of sizes of said defects in said chip, a total number of said defects in said chip.

10. The method of claim 7, wherein said inspection of said defects is performed by a random method which inspects defects in peripheral regions.

11. The method of claim 10, wherein said probable parameters include at least one of maximum area of defects in a chip, a maximum value of the sum of areas of said defects in a segment, a maximum value of the sum of sizes of said defects in said segment, a maximum value of the total number of said defects in said segment, a sum of areas of said defects in said chip, a sum of sizes of said defects in said chip, a total number of said defects in said chip, a maximum value of areas of said defects in peripheral regions of an X axis, a maximum value of the size of said defects in said peripheral regions of the X axis, a maximum value of the sum of the areas of said defects in peripheral regions around left and right regions of the X axis, a maximum value of the sum of sizes of said defects in said peripheral regions around said left and right regions of the X axis, a maximum value of the number of said defects in said peripheral regions around said left and right regions of the X axis, a maximum value of areas of said defects in peripheral regions of a Y axis, a maximum value of the size of said defects in said peripheral regions of the Y axis, a maximum value of the sum of areas of said defects in peripheral regions of upper and lower regions of the Y axis, a maximum value of the sum of sizes of said defects in said peripheral regions of upper and lower regions of the Y axis, and a maximum value of the number of said defects in said peripheral regions of said upper and lower regions of the Y axis.

12. The method of claim 11, wherein a process performer obtains data on chips which fail and chips which do not fail in each of said probable parameters and wherein a decision tree is obtained from said data which classifies said chips which fail and said chips which do not fail using said probable parameters.

13. The method of claim 7, wherein said probable parameters include a maximum value of areas of said defects in peripheral regions around the X axis, a maximum value of sizes of said defects in said peripheral regions around the X axis, a maximum value of the sum of said areas of said defects in peripheral regions around left and right regions of the X axis, a maximum value of the sum of sizes of said defects in said peripheral regions around said left and right regions of the X axis, a maximum value of the number of said defects in said peripheral regions around said left and right regions of the X axis, a maximum value of areas of said defects in peripheral regions around the Y axis, a maximum value of sizes of said defects in said peripheral regions around the Y axis, a maximum value of the sum of areas of said defects in peripheral regions of upper and lower regions of the Y axis, a maximum value of the sum of sizes of said defects in said peripheral regions of said upper and lower regions of the Y axis, and a maximum value of the number of said defects in said peripheral regions of said upper and lower regions of the Y axis.

14. A method of calculating a defect limited yield (DLY) comprising:

inspecting said blocks to detect defects in said blocks and in blocks located around said inspected blocks;

measuring the number of said inspected blocks having failures caused by a reason other than said defects in said inspected blocks located around said inspected blocks having said defects (n1);

measuring the number of said inspected blocks having no failures in said inspected blocks located around said inspected blocks having said defects (n2);

measuring the number of inspected blocks having failures caused by said defects in said inspected blocks having said defects (n3);

measuring the number of inspected blocks having no failures in said inspected blocks having said defects (n4);

substituting data (n1) through (n4) in formula 1a $\begin{array}{cc}\mathrm{KR}=1-\frac{1-{\mathrm{KR}}_1}{1-{\mathrm{KR}}_0};<br><br><br><br><br><br>\mathrm{wherein}<br><br><br><br>{\mathrm{KR}}_0=\frac{\mathrm{n1}}{\mathrm{n1}+\mathrm{n2}}<br><br><br><br>\mathrm{and}<br><br><br><br>{\mathrm{KR}}_1=\frac{\mathrm{n3}}{\mathrm{n3}+\mathrm{n4}};<br><br>\mathrm{and}& \left(1<br><br>a\right)\end{array}$

substituting KR into formula;

$\mathrm{DLY}=\frac{k-\underset{i}{\overset{k}{\sum <br><br>}}<br><br><br><br>\mathrm{KRi}}{k}\times <br><br>100<br><br>\left(\%\right),$

wherein i is 1 to k;

wherein k is the total number of wafer chips; and

wherein KR is the probability that wafer chips fail.

15. A method of calculating a defect limited yield (DLY) comprising:

determining the total number of wafer chips (m1);

detecting defects in each said wafer chip;

determining the total number of chips having said defects (m2);

determining the total number of chips which fail due to a reason other than said defects (m3);

classifying said wafer chips having defects using probable parameters in consideration of area, size, and number of said defects;

determining the total number of wafer chips (m4) and the total number of wafer chips which fail (m5) when said wafer chips are classified according to each said probable parameter;

calculating a probability (KR) that said wafer chips fail only due to said defects by substituting the data m1, m2, m3, m4, and m5 in the formula;

$\mathrm{KR}=1-\frac{1-{\mathrm{KR}}_1}{1-{\mathrm{KR}}_0},<br><br>\mathrm{wherein}<br><br>,{\mathrm{KR}}_0=\frac{\mathrm{m3}}{\mathrm{m1}-\mathrm{m2}}<br><br><br><br>\mathrm{and}<br><br><br><br>{\mathrm{KR}}_1=\frac{\mathrm{m5}}{\mathrm{m4}};<br><br>\mathrm{and}$

substituting KR into formula;

$\mathrm{DLY}=\frac{k-\underset{i}{\overset{k}{\sum <br><br>}}<br><br><br><br>\mathrm{KRi}}{k}\times <br><br>100<br><br>\left(\%\right)$

wherein i is 1 to k;

wherein k is the total number of wafer chips; and

wherein KR is the probability that wafer chips fail.

16. A system of calculating a probability of failures (KR) caused only by defects in blocks in a wafer chip comprising:

a defect detector for inspecting said blocks to detect defects in said blocks and in blocks located around said inspected blocks; and

a controller for measuring;

the number of said inspected blocks having failures caused by a reason other than said defects in said inspected blocks located around said inspected blocks having said defects (n1);

the number of said inspected blocks having no failures in said inspected blocks located around said inspected blocks having said defects (n2);

the number of inspected blocks having failures caused by said defects in said inspected blocks having said defects (n3); and

the number of inspected blocks having no failures in said inspected blocks having said defects (n4); and

substituting data (n1) through (n4) in formula 1a $\begin{array}{cc}\mathrm{KR}=1-\frac{1-{\mathrm{KR}}_1}{1-{\mathrm{KR}}_0};<br><br><br><br>\mathrm{wherein}<br><br><br><br>{\mathrm{KR}}_0=\frac{\mathrm{n1}}{\mathrm{n1}+\mathrm{n2}}<br><br><br><br>\mathrm{and}<br><br><br><br>{\mathrm{KR}}_1=\frac{\mathrm{n3}}{\mathrm{n3}+\mathrm{n4}}.& \left(\text{1a}\right)\end{array}$

17. A system of calculating a probability of failures caused only by defects in each chip of a wafer comprising:

a controller for determining the total number of wafer chips (m1);

detecting defects in each said wafer chip;

determining the total number of chips having said defects (m2);

determining the total number of chips which fail due to a reason other than said defects (m3);

classifying said wafer chips having defects using probable parameters in consideration of area, size, and number of said defects;

determining the total number of wafer chips (m4) and the total number of wafer chips which fail (m5) when said wafer chips are classified according to each said probable parameter; and

calculating a probability (KR) that said wafer chips fail only due to said defects by substituting the data m1, m2, m3, m4, and m5 in the formula;

$\mathrm{KR}=1-\frac{1-{\mathrm{KR}}_1}{1-{\mathrm{KR}}_0},<br><br>\mathrm{wherein}$${\mathrm{KR}}_0=\frac{\mathrm{m3}}{\mathrm{m1}-\mathrm{m2}}<br><br><br><br>\mathrm{and}<br><br><br><br>{\mathrm{KR}}_1=\frac{\mathrm{m5}}{\mathrm{m4}}.$

18. A system of calculating a defect limited yield (DLY) comprising:

a defect detector for inspecting said blocks to detect defects in said blocks and in blocks located around said inspected blocks; and

a controller for measuring;

the number of said inspected blocks having failures caused by a reason other than said defects in said inspected blocks located around said inspected blocks having said defects (n1);

the number of said inspected blocks having no failures in said inspected blocks located around said inspected blocks having said defects (n2);

the number of inspected blocks having failures caused by said defects in said inspected blocks having said defects (n3); and

the number of inspected blocks having no failures in said inspected blocks having said defects (n4);

substituting data (n1) through (n4) in formula 1a $\begin{array}{cc}\mathrm{KR}=1-\frac{1-{\mathrm{KR}}_1}{1-{\mathrm{KR}}_0};<br><br>& \left(\text{1a}\right)\end{array}$

substituting KR into formula;

$\mathrm{DLY}=\frac{k-\underset{i}{\overset{k}{\sum <br><br>}}<br><br><br><br>\mathrm{KRi}}{k}\times <br><br>100<br><br>\left(\%\right),$

wherein i is 1 to k;

wherein k is the total number of wafer chips; and

wherein KR is the probability that wafer chips fail.

19. A system of calculating a defect limited yield (DLY) comprising:

a controller for determining the total number of wafer chips (m1);

detecting defects in each said wafer chip;

determining the total number of chips having said defects (m2);

determining the total number of chips which fail due to a reason other than said defects (m3);

classifying said wafer chips having defects using probable parameters in consideration of area, size, and number of said defects;

determining the total number of wafer chips (m4) and the total number of wafer chips which fail (m5) when said wafer chips are classified according to each said probable parameter;

calculating a probability (KR) that said wafer chips fail only due to said defects by substituting the data m1, m2, m3, m4, and m5 in the formula;

$\mathrm{KR}=1-\frac{1-{\mathrm{KR}}_1}{1-{\mathrm{KR}}_0},<br><br>\mathrm{wherein}$${\mathrm{KR}}_0=\frac{\mathrm{m3}}{\mathrm{m1}-\mathrm{m2}}<br><br><br><br>\mathrm{and}<br><br><br><br>{\mathrm{KR}}_1=\frac{\mathrm{m5}}{\mathrm{m4}};<br><br><br><br>\mathrm{and}$

substituting KR into formula;

$\mathrm{DLY}=\frac{k-\underset{i}{\overset{k}{\sum <br><br>}}<br><br><br><br>\mathrm{KRi}}{k}\times <br><br>100<br><br>\left(\%\right)$

wherein i is 1 to k;

wherein k is the total number of wafer chips; and

wherein KR is the probability that wafer chips fail.