Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

SAMPLING METHODOLOGY APPLICABLE FOR INLINE METROLOGY

IP.com Disclosure Number: IPCOM000131166D
Publication Date: 2005-Nov-09
Document File: 3 page(s) / 399K

Publishing Venue

The IP.com Prior Art Database

Related People

Arie Glazer: INVENTOR [+5]

Abstract

Inline metrology capabilities such as CD and Overlay measurements, when integrated with Inline AOI systems, provide added value to FPD customers and can strongly differentiate an AOI system.In practice, to realize this added value, both the TACT of the system and the process control quality level achieved must be at least maintained, and preferably, improved. In this article we examine the trade off between sampling rate and the measurement accuracy required to achieve a particular quality level. Specifically, we show how the higher sampling rates achievable in an integrated inline metrology-AOI system allow a lower accuracy sample measurement, without sacrificing quality, thus providing a cost effective solution.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 29% of the total text.

 

Sampling Methodology Applicable for Inline Metrology

 

by

 

Arie Glazer

,

Mannie Dorfan

,

Noam Cohen

, Ben-Zion Poupko

Display Division, Orbotech Ltd,

    

P.O. Box 215

,

Yavne

81102

,

Israel

Inline metrology capabilities such as CD and Overlay measurements, when integrated with Inline AOI systems, provide added value to FPD customers and can strongly differentiate an AOI system.

In practice, to realize this added value, both the TACT of the system and the process control quality level achieved must be at least maintained, and preferably, improved.

In this article we examine the trade off between sampling rate and the measurement accuracy required to achieve a particular quality level. Specifically, we show how the higher sampling rates achievable in an integrated inline metrology-AOI system allow a lower accuracy sample measurement, without sacrificing quality, thus providing a cost effective solution.

 


Offline (i.e. dedicated standalone) Critical Dimension (CD) and Overlay Monitoring  - Methodology and Sample Size

 

Based on surveys conducted with FPD array manufacturers, for a five-mask process, two to four substrates are measured for CD/Overlay out of every 5 lots (100 glasses).

The average number of points measured per glass (up to 6th generation) is 40. The sample size is therefore, at minimum, ~80 measured points per 100 substrates. This sample size is assumed sufficient, from a statistical point of view, to determine that the material is in spec with a 99% degree of confidence.

 

Features that have a cross-section (width) of 2mm to 30mm are typical for FPD line CD’s, spacings, and overlay feature sizes. The control limits for acceptance (6s process variation) are in the range of ±10%.

The typical offline measurement system accuracy is ~ ±1% and the average measuring time is ~10 seconds per point.

 

Determining Theoretical Sample Size

 

Let us now compare the actual sample size used by customers to the theoretical sample size needed to obtain the quality level required. On the theoretical side, we will assume normal distribution. In this case, the sample size (n) for a 99% degree of confidence, which corresponds to a=0.01 (1.00-0.99; a actually expresses the relative amount of measurements that are allowed to be out of spec.) is given by the following equation (1,2)  (1)

(1)

  

 

 

Where:

Za/2 - is the Standard Normal (Z) Distribution =2.58 (also known as the ‘Critical Value’)*

s- is the Standard Deviation: s= [(R p2+ A m2)0.5]/6

Where: Rp - is the random process variation (or “

Process

Range

”) and Am is the measurement accuracy (calculated as twice the “3s of measurement repeatability”; so if the repeatability of an Offline CD/Overlay is ±30 nanometers, then Am=60 nanometers.

An example calculation of s:

For a CD of 2mm, and for a

Process

Range

of ±10% (thus Rp=2x[2x0.1]=~0.4mm), and for a CD measurement accuracy (Am) of 0.06mm, we obtain

s= [(0.42 + 0.062) 0.5]/6 = 0.067mm

E - is theMargin of Error i.e. the maximum diff...