Evaluation methodology for 300mm wafer carrier performance
09/01/2000
Wafer Handling
Tracy Niebeling, Entegris Inc., Chaska, Minnesota
overview
In 300mm wafer fabs, the transport wafer carrier will play a vital role in manufacturing success. Because of this role, it has become increasingly important to quantitatively test and measure wafer carrier performance. The wafer carrier architecture of choice for 300mm is the front opening unified pod or FOUP. The FOUP and its application bring a whole new set of critical functions and metrics that require careful quantitative evaluation. In addition, emerging new performance factors will require further evaluation.
Front opening unified pods (FOUPs) for 300mm wafers have three critical performance functions:
- The FOUP must exhibit precise wafer access for reliable, high throughput wafer transfer.
- The FOUP must provide secure wafer protection, keeping the wafer isolated from contamination and damage due to mechanical or electrical effects.
- The FOUP must enable reliable interoperability with process tools and automated material handling systems.
At Entegris we have developed a matrix of tests that can be performed on 300mm FOUPs to measure their performance against these three critical functions.
Precise wafer access
The prime function of any FOUP is to hold its wafers in the proper location and orientation, presenting them for speedy and reliable automated transfer into and back out of a process tool. The goal is for the FOUP to enable wafer transfer at the highest possible throughput with the least amount of sliding contact with the wafer. Sliding contact between the wafer and the FOUP or between the wafer and the transfer mechanism will result in particle generation or wafer damage.
 Figure 1. Wafer plane tolerance. |
The measure of a FOUP's ability to hold the wafers correctly is wafer plane. Semi standard E1.9 details the required position for the bottom surface of each wafer. The tolerance for this position is ±0.5mm (Fig. 1). Measurement of the position of the wafers in a given FOUP is difficult to perform accurately. The most repeatable method for measuring wafer position is to use a noncontact, optical method. As it is desirable to measure many wafers on many FOUPs, it is essential that this measurement be automated. We are using a commercially available computer-controlled instrument based on video technology (Fig. 2) to measure the wafer plane of all 25 pockets of our FOUPs as a quality control measure. Further, we evaluate our ability to meet the specification using statistics. Our minimum requirement is to demonstrate 6s capability or a Cpk > 1.0.
Secure wafer protection
Data on particles/wafer pass (PWP) are the most requested parameter of a FOUP. However, there are many methods used to insert and extract wafers from a FOUP and each method provides a different PWP result. Application of each method can also vary and affect results.
At Entegris, we have adopted a Tencor SP1 laser particle counter for wafer particle measurement (Fig. 3). In addition, we do PWP studies in two different ways: Our basic test determines PWP based on FOUP door open-close cycles. A second test determines PWP based on FOUP door open-close with wafer transfer out of and back into the FOUP for each cycle.
The first test gives us a measure of the cleanliness of the FOUP's door mechanism, as well as a measure of the effectiveness of the FOUP load port interface; particles can be attributed to both the FOUP and the load port. The second test includes particles that are added due to the wafer transfer mechanism and its ability to transfer wafers without sliding contact with the carrier.
 Figure 2. Computer-controlled wafer plane measurement via the August Technology CV9812. |
With the first test — door cycle only — an acceptable PWP result is anything <0.009 at a measured particle size of >0.09mm. In the second test — door cycle plus wafer transfer — an acceptable PWP result is anything <0.05 at a particle size of >0.09mm.
We also measure particles added to wafers while being stored in the FOUP. This tests the FOUP's ability to isolate the wafer from particles even in a dirty environment. Currently, there is no industry consensus on the methodology for these environmental challenge tests. However, at Entegris we have done tests where FOUPs with wafers are stored in a relatively dirty office setting for 72 hr. We make wafer particle counts before and after the storage period and calculate a particle/wafer/day (PWD) value. This test measures the ability of the FOUP to protect the wafers from particles in an unclean environment.
We have found that good door seals and efficient breather filters are needed to obtain acceptable results in this test. Acceptable
PWD results for this type of office ambient test are anything <0.15 PWD at a measured particle size of >0.09mm.
It is important for any FOUP to have a reliable path to ground from the wafer contact areas. In the case of a 300mm FOUP, there must be a path to ground from the wafer supports down to the kinematic coupling grooves and up to the robotic lifting flange. We can reasonably expect that these equipment interfaces will allow the FOUP to be grounded through the equipment. The path to ground through the FOUP should have resistivity in the static dissipative range, rather than being in the conductive range. The common specification for this static dissipative range of resistivity is 105-109 W.
There are many accepted ways to measure the resistivity of a path to ground directly. One way to measure the effectiveness of a FOUP's path to ground is to do a charge and decay test. In this test, the FOUP is charged to ±5000V and then grounded at either the kinematic coupling or the top robotic flange. A decay to 0V in <0.1 sec is a measure of effective path to ground.
Reliable interoperability
Unlike its 200mm predecessor, a FOUP is no longer a simple plastic part. It is really a small machine. We have worked from the start with I300I on FOUP door marathon testing. Testing has been performed both at I300I and in our own facilities. Most of this testing has been simply open and close cycles on actual load ports. Some testing at Entegris has been done using a door cycling mechanism built for the purpose that speeds up cycling and reduces the time necessary to do a full marathon test. In all of our FOUP marathon testing, we have used an acceptance benchmark of 161,000 cycles, indicating a 100,000-cycle life with an 80% confidence level.
 Figure 3. Tencor SP1 laser particle counter for wafer particle measurement. |
Reliability also requires demonstration of a FOUP's interoperability on many load ports. This testing is similar to a marathon test, except the same FOUP is tested to at least 20,000 cycles on multiple load ports.
It is important that each load port be set up to Semi standards, including E57 for the kinematic coupling pins and E62 for the FIMS door opening features. With the load port set up properly, if the FOUP fails during cycling, the cause can be attributed to the FOUP rather than to a misaligned load port.
A successful interoperability test is evident when a single model of FOUP can be cycled at least 20,000 cycles on collection of different makes of load port.
A FOUP also has many critical equipment interface features. The ability of the FOUP to hold wafers precisely depends on the dimensional stability of these features. We routinely measure the position of the top robotic lifting flange relative to the datum structure defined by the kinematic coupling. This feature must be compliant with Semi standard E47.1 to allow reliability and interoperability of the FOUP in automated materials handling system (AMHS) applications.
FOUP test matrix
The cumulative result of our work with evaluating 300mm FOUPs is a matrix of tests (see table) that measure performance against the three critical functions we listed above.
To address additional emerging critical factors and needed performance characterization requirements, we are looking at future FOUP test plans in four main areas:
Airborne molecular contamination (AMC). Because the FOUP is an enclosed environment, airborne molecular contamination of the wafers is a potential issue. We have been using advanced capabilities to evaluate both raw materials and FOUP assemblies for out-gassed contaminants for many years. However, the wafer level effects of a certain amount of a certain molecular contaminant are difficult to determine. Currently, the industry realizes the critical nature of AMC, but has yet to form a consensus on the methodology for testing.
Effect of time and use. We have plans to test the effect of time and use on wafer plane performance and on other critical equipment interfaces. We do not currently know the effects on the FOUP of multiple years of use, including thousands of wash cycles, thousands of hot wafers inserted, and hundreds of thousands of load port cycles and AMHS transfer cycles. We plan to conduct tests that simulate normal and extraordinary use of FOUPs and to study the effects on wafer plane and interface dimensions.
 |
Challenge environment. To gain the full advantage of isolated wafer transport using FOUPs, load ports, and minienvironments, users will begin to allow the outside environment to degrade, while keeping the wafer environment inside the FOUPs and process tools at Class 1. To address this, we intend to design tests to challenge the FOUP with a controlled Class 1000 or Class 10,000 environment. Simulated AMHS transport of the FOUP in this challenge environment, followed by a test for particle adders, will provide more realistic results than testing that has been performed to date.
Wafer level effect. We also intend, as much as possible, to perform future tests so that the wafer level effect of FOUP performance is measured. We currently rely on laser particle counting to estimate various PWP cases. But there is a need to develop even more measures of wafer level effect so as to come closer to predicting the performance of a 300mm FOUP in production use.
Conclusion
The FOUP will be a critical component in 300mm manufacturing success. Because of this, thorough FOUP performance evaluation and characterization is increasingly crucial, including evaluation of precise wafer access for reliable high throughput wafer transfer; secure wafer protection and isolation from contamination and damage due to mechanical or electrical effects; and reliable interoperability with process tools and automated material handling systems. Successful evaluation of these is already drawing on today's known methodologies, but additional work will be required to address yet unknown issues, including AMC, the effects of time and use, and wafer level effects.
Acknowledgments
This article was written with contributions from the entire Entegris 300mm team and, in particular, the efforts of the product test laboratory.
Tracy Niebeling received his degree in mechanical engineering from the University of Minnesota and his MBA from the University of St. Thomas. He serves as a volunteer in the Semi standards development process, and has served as co-chair of the wafer carriers subcommittee. Niebeling is involved with the design, technical field support, and marketing of wafer management products at Entegris Inc., 3500 Lyman Blvd., Chaska, MN 55318; ph 952/448-3131, fax 952/556-1880.