Semiconductor Defect Signal Capturing and Statistical System and Method

• US 20120323506A1
• Filed: 10/13/2011
• Published: 12/20/2012
• Est. Priority Date: 11/23/2010
• Status: Abandoned Application

• Current Assignee: Andrew Payshin King
• Original Assignee: Andrew Payshin King
• Inventors: King, Andrew Payshin
• Application Number:
  US13/214,609
• Publication Number:
  US 20120323506A1
• Time in Patent Office: Days
• US Class Current: 702/58
• CPC Class Codes: G01R 31/2656 using non-ionising electromagnetic radiation, e.g. optical radiation
• CPC Class Codes: G01R 31/2831 Testing of materials or semi-finished products, e.g. semiconductor wafers or substrates G01R31/318511 takes precedence; testing during manufacture H01L22/00
• CPC Class Codes: G01R 31/2894 Aspects of quality control [QC] G01R31/31718 takes precedence; program control for QC G05B19/41875
• CPC Class Codes: G06T 2207/30148 Semiconductor; IC; Wafer
• CPC Class Codes: G06T 7/001 using an image reference approach
• CPC Class Codes: H01L 22/12 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 electrical measurement of diffusions H01L22/14
• CPC Class Codes: H01L 22/20 Sequence of activities consisting of a plurality of measurements, corrections, marking or sorting steps

Abstract

This invention, embodied in software, is a defect signature detection and analysis system to group and classify defects received from semiconductor inspection tools into categories which identify the defect source. This system includes on-line monitoring devices, signal analytical and statistical devices and information database. The signal analytical device includes an analysis unit, capturing unit and an information queue. The capturing unit includes a number of sub-units, including sequential, overlap and iterative detection, grouping, and undefined signal detection. The system uniquely uses image processing techniques on non-image data to group individual defects into larger clusters, thereby increasing the probability of correct classification of the defect modes on semiconductor wafers, while decreasing incorrect or missing classifications. The system improves accuracy and integrity of these captured defect signals, enabling the statistical methods of the present invention to substantially replace the manual inspection of the prior art, and to decrease semiconductor manufacturing cost.

Claims

1. A semiconductor defect signal capturing and statistical system, including an on-line monitoring device, characterized in that, said system further includes a defect signal analytical device, a defect information bank and a defect signal statistical device, the defect signal analytical device is connected with the on-line monitoring device, the defect information bank and the defect signal statistical device respectively, the defect information bank contains several pre-determined defect signal modes, the defect signal analytical device includes a defect signal analysis unit, a defect signal capturing unit and a defect information queue, the defect signal capturing unit includes a sequential detection sub-unit, an overlap detection sub-unit, an iterative detection sub-unit, a grouping detection sub-unit and an undefined signal detection sub-unit.

2. A semiconductor defect signal capturing and statistical method carried out with the system according to claim 1, characterized in that, the method comprises the following:

Scanning semiconductor wafers passing through a production line and grouped by grids, and generating a KLA format wafer defect scanning result file containing corresponding detect signal information with the on-line monitoring device;

Analyzing the KLA format wafer defect scanning result file and capturing detect signals according to the defect information bank with the defect signal analytical device;

Performing a defect signal statistical process according to the above-processed result with the defect signal statistical device.

3. The semiconductor defect signal capturing and statistical method according to claim 2, characterized in that, the information contained in the KLA format wafer defect scanning result file includes grid identifications, slot identifications, processing step identifications, processing device identifications and corresponding defect signal binary information of semiconductor wafers.

4. The semiconductor defect signal capturing and statistical method according to claim 3, characterized in that, the defect signal capturing unit includes a sequential detection sub-unit, that analyzing the KLA format wafer defect scanning result file, and capturing detect signals according to the defect information bank with the defect signal analytical device comprises the following steps:

(step 2-1) Reading the KLA format wafer defect scanning result file and analyzing the content of the file with the defect signal analysis unit;

(step 2-1) Identifying and capturing corresponding defect signal modes from the defect signal binary information obtained through analysis according to the pre-determined defect signal modes in the defect information bank with the sequential detection sub-unit of the defect signal capturing unit;

(step 2-3) Storing the grid identifications, the slot identifications, the processing step identifications, the corresponding defect signal mode types and the defect signal locations the semiconductor wafers captured correspond to into the defect information queue.

5. The semiconductor defect signal capturing and statistical method according to claim 4, characterized in that, the defect signal capturing unit further includes an iterative detection sub-unit, and after the step (2-3), this method further comprises the following steps:

(step 2-4) According to the grid identifications, the slot identifications, the processing step identifications and the processing device identifications the semiconductor wafers captured correspond to, searching for semiconductor wafers with the same grid identifications, the same slot identifications, the same processing step identifications and the same processing device identifications in the semiconductor wafers uncaptured with the iterative detection sub-unit;

(step 2-5) Detecting the semiconductor wafers searched out according to the defect signal modes of the semiconductor wafers captured with the iterative detection sub-unit.

6. The semiconductor defect signal capturing and statistical method according to claim 5, characterized in that, after the step (2-5), this method further comprises the following steps:

(step 2-6) Identifying the grid identifications, the slot identifications, the processing step identifications and the processing device identifications the semiconductor wafers uncaptured correspond to with the iterative detection sub-unit;

(step 2-7) According to the grid identifications, the slot identifications, the processing step identifications and the processing device identifications identified, searching for semiconductor wafers with the same grid identifications, the same slot identifications, the same processing step identifications and the same processing device identifications in the semiconductor wafers captured with the iterative detection sub-unit;

(step 2-8) Detecting the semiconductor wafers uncaptured according to the defect signal modes of the semiconductor wafers searched out with the iterative detection sub-unit.

7. The semiconductor defect signal capturing and statistical method according to claim 6, characterized in that, the defect signal capturing unit further includes a grouping detection sub-unit, and after the step (2-8), this method further comprises the following steps:

(step 2-9) Creating a grouping condition information according to each of the defect signal modes with the grouping detection sub-unit, wherein the grouping condition information includes grouping characteristic parameters of the defect signal mode and the wafer information of the semiconductor wafer the defect signal mode corresponds to;

(step 2-10) According to the defect signal modes the semiconductor wafers captured correspond to, searching for semiconductor wafers with the same defect signal modes and the same wafer information with the grouping detection sub-unit;

(step 2-11) Comparing the wafer information of the semiconductor wafers searched out with the wafer information of the semiconductor wafers captured, if the comparison result meets the grouping characteristic parameters, the semiconductor wafers captured and the semiconductor wafers searched out are grouped into one group.

8. The semiconductor defect signal capturing and statistical method according to claim 7, characterized in that, the defect signal capturing unit further includes an overlap detection sub-unit, and after (step 2-11), this method further comprises the following steps:

(step 2-12) Identifying the grid identifications, the slot identifications, the processing step identifications and the processing device identifications the semiconductor wafers uncaptured correspond to with the overlap detection sub-unit;

(step 2-13) Searching for KLA format wafer defect scanning result file with the same grid identifications, the same slot identifications, the same processing step identifications and the same processing device identifications with the overlap detection sub-unit;

(step 2-14) Overlapping the current semiconductor wafer file and the KLA format wafer defect scanning result file searched out to generate a new semiconductor wafer file;

(step 2-15) Returning to the step (1).

9. The semiconductor defect signal capturing and statistical method according to claim 7, characterized in that, the defect signal capturing unit further includes an overlap detection sub-unit, and after (step 2-11), this method further comprises the following steps:

(step 2-12′

) According to the grid identifications, the slot identifications, the processing step identifications and the processing device identifications pre-determined, identifying grid identifications, slot identifications, processing step identifications and processing device identifications semiconductor wafers correspond to and same as those predetermined in the semiconductor wafers uncaptured with the overlap detection sub-unit;

(step 2-13′

) Adding up the total number of semiconductor wafers identified by the overlap detection sub-units;

(step 2-14′

) When the total number of semiconductor wafers added up is up to a pre-determined value, and after a pre-determined period of time, overlapping the semiconductor wafer file identified to generate a new semiconductor wafer file;

(step 2-15′

) Returning to the step (1).

10. The semiconductor defect signal capturing and statistical method according to claim 8 or claim 9, characterized in that, the defect signal capturing unit further includes an undefined signal detection sub-unit, the defect information bank includes a defect information database and an undefined defect information database, and after (step 2-15) or (step 2-15′

), this method further comprises the following steps;

(step 2-16) After a pre-determined period of time extracting the wafer information of the semiconductor wafers uncaptured during this period of time with the undefined signal detection sub-unit;

(step 2-17) Forming an undefined defect signal group using the wafer information having one or more same identifications among the extracted wafer information with the undefined signal detection sub-unit;

(step 2-18) Storing the undefined defect signal group into the undefined defect information database with the undefined signal detection sub-unit.

11. The semiconductor defect signal capturing and statistical method according to claim 10, characterized in that, the period of time is the scanning or detecting time of semiconductor wafers.