Sealing components need standards!
Surprisingly, no real standard exists for sealing elements in this industry due to the lack of understanding and awareness of the nature of seals and their potential dramatic effect on process integrity and manufacturing productivity. To achieve the latest trends towards the 22nm world and beyond, and avoid costly failures, this must change.
The facts: 1) Seals were invented in the late 1800s and first patented in the US in 1937 for use in air brakes in street cars. Most applications are intended for pressures of 500 PSI and above. 2) There is NO SEMI specification or adherence call out for seals in the manufacturing process. 3) New materials such as the perfluoroelastomer family of materials (known as FFKMs in the sealing industry) are borne to be chemically resistant first, and to be sealing components, second; they indeed mostly adhere to basic chemical requirements utilized in both wet and dry chemistries in the manufacturing of wafers. However, these materials do not have the strength or resiliency to match the performance of FKM seals invented for a completely different industry. Yet, these seals are ???measured' and described in the same manner as the seals used for industrial manufacturing and similar, lower technology products. The seals are covered by the same ASTM's (American Society for Testing and Materials) specifications and were originally created for a world of rods and pistons extensively utilized in aerospace and industrial applications- not lid seals, door gates, slit valves and vacuum chambers used in semiconductor processing.
It is both fascinating and disturbing that three out of the five basic ASTMs utilized in every seal component manufacturer's data sheets have little to do with the actual application used in semiconductor processing and do not describe any seal performance that could be associated with those applications. Even worse, these same seals do not discern any differences in the environments where they are used: plasma (etch, CVD, etc.), wet or lithography. Moreover, it is not called out whether the seals are utilized in the facilities or the main process chamber.
Of greater concern ??? the dynamic seals ??? those most critical seals that can impact tools in particle loss or maintaining leak rates, are the most sensitive to the physical property variances in FFKMs. The highest rates of failure are associated WITH these types of sealing components.
Today on 300mm tools, with process nodes attempting to reach 22nm and beyond, new facets of particle sizes as well as what type of particles become relevant in successful processing. Subcomponents will become limiting factors in the drive towards those new technologies.
To address the challenges outlined above, a sealing task force already sponsored in SEMI's Facilities group, named the F51-0200 Task Force, and chaired by ASNA, is tasked with moving this situation into the spot light and draw attention to the magnitude of the issue. This forum also must have industry participation.
This industry effort will bring about the following: 1) an agreement on the type of measurements for a few of the current ASTMs (as there are several types of each), and 2) development of unique ASTMs, possibly for various levels of processes, such as etch or plasma, non-plasma or wet applications, as examples.
Consistency of measurements for sealing elements that are representative of the semiconductor industry's requirements, will have to be considered if the 22nm and beyond world becomes our reality.
|Dalia Vernikovsky is chief executive and founder of Applied Seals North America, Inc., 39899 Balentine Drive, Suite #368, Newark, CA 94560; ph.: 510-623-8307; email email@example.com|
Solid State Technology | Volume 54 | Issue 10 | November 2011