Data processing: measuring – calibrating – or testing – Measurement system – Performance or efficiency evaluation
Reexamination Certificate
2003-02-21
2004-06-29
Niebling, John F. (Department: 2812)
Data processing: measuring, calibrating, or testing
Measurement system
Performance or efficiency evaluation
Reexamination Certificate
active
06757640
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method and apparatus for determining and assessing chamber inconsistency in a tool. In particular, the present invention relates to determining and assessing chamber inconsistency by processing the measurement data obtained by measuring the units of a product.
2. Description of the Related Art
Semiconductor fabrication normally encompasses hundreds of operations to process a blank wafer for the generation of logic integrated circuitry (IC), dynamic random access memory (DRAM), static random access memory, flash memory, and other chips. Each operation responds for a single process, such as lithography, etching, deposition, and oxidation. As part of mass-production, each operation usually has several tools that can execute a corresponding operation. Each tool is usually designed to have multiple chambers to process the same operation in parallel. Theoretically, a single operation needs to produce the same process results. Thus, each chamber of a single tool should produce equal and identical results.
In semiconductor fabrication, yield, whether wafer yield or chip yield, signifies the cost and profit of a fabrication process. In order to maximize yield, the most unproductive operation or tool is first located, and then proper action, such as adjustment or repair, must be performed. Incidents, however, in which production by one or more chambers of a tool differs significantly form that of the other chambers, defined as chamber inconsistency, is very hard to identify.
For the purpose of convenient transportation and management, wafers in semiconductor fabrication are usually gathered in lots. Presently, 8-inch fabrication puts 25 wafers in a cassette, referred to as a lot. The cassette bears a lot number. Furthermore, each wafer has a wafer number marked on its surface. For example, the wafer number 05 in the lot “65038” means that the corresponding wafer is the 5
th
wafer in the lot numbered 65038. Lots may not be full, that is, having fewer than 25 wafers. During operations, some wafers in a lot may be abandoned due to misoperation. Therefore, the number of wafers in a lot can be anywhere from 1 to 25.
Determination of this chamber inconsistency faces the following impediments:
1. Chamber inconsistency is not identifiable from commonly accessed data, such as average yield or average bit loss, of lots.
2. Wafer numbers have no fixed correlation to the chambers of a tool. Each wafer is randomly chosen to be processed in a chamber. Chamber inconsistency thus cannot be identified by checking wafers with the same wafer number.
SUMMARY OF THE INVENTION
An object of the present invention is to determine and assess whether a tool with multiple chambers has chamber inconsistency.
The method for determining chamber inconsistency of a single tool according to the present invention is applied to a process flow for manufacturing a product. The tool has a plurality of process chambers. A plurality of product groups is provided. Each the product groups has a plurality of units and has been processed in the process chambers of the tool. The product groups are measured to obtain measurement data groups. Each measurement data group comprises a plurality of measured data corresponding to the units in a corresponding product group. The measured data of the corresponding product group are sorted to obtain a corresponding monotonic sequence. The monotonic sequences corresponding to the product groups are element-by-element averaged to obtain a mean monotonic sequence. If the mean monotonic sequence fits in with a predetermined criterion, the tool is indicated to have chamber inconsistency.
The product groups and the units can be lots and wafers, respectively. The measured data can be resistance, capacitance, chip yield or failed bit number of a lot.
The predetermined criterion, for example, can be that the mean monotonic sequence has a step variation in a medial position, or that the mean monotonic sequence is out of a predetermined distribution.
This invention further provides a method for determining chamber inconsistency in several tools. First, the mean monotonic sequences corresponding to the tools are ascertained according to the present invention. The chamber inconsistency of the tools can be determined by checking these mean monotonic sequences.
Except via step variation and predetermined distribution, a tool can be determined to have chamber inconsistency if its corresponding mean monotonic sequence is unique among these mean monotonic sequences.
The present invention further provides an apparatus for determining chamber inconsistency in a tool. This apparatus is applied to a process flow for manufacturing a product. The tool has a plurality of process chambers. The apparatus comprises a plurality of product groups, measurement data provider, a sorter, a means for averaging and a system of analysis. Each the product groups has a plurality of units of the product and has been processed in the process chambers of the tool. The measurement data provider provides measurement data groups, each comprising a plurality of measured data obtained by measuring the units in a corresponding product group. The sorter sorts the measured data of the corresponding product group to obtain a corresponding monotonic sequence. The monotonic sequences corresponding to the product groups are averaged, element by element, to obtain a mean monotonic sequence. The system of analysis indicates that the tool has chamber inconsistency if the mean monotonic sequence fits in with a predetermined criterion.
The advantage of the present invention is that the chamber inconsistency in a tool is clearly and precisely determined so proper action can be quickly taken to improve the yield, thereby increasing profits.
REFERENCES:
patent: 6073501 (2000-06-01), Rohner
patent: 6195621 (2001-02-01), Bottomfield
Klaus Enk
Weng Hung-Jen
Birch & Stewart Kolasch & Birch, LLP
Niebling John F.
Promos Technologies Inc.
Stevenson Andre′ C.
LandOfFree
Method and apparatus for determining and assessing chamber... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for determining and assessing chamber..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for determining and assessing chamber... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3294793