by Robert P. Donovan
 |
No one would consider manufacturing many of today's precision products without a cleanroom. Difficult as it may be to believe, there was a time when cleanrooms were not used in fabricating semiconductor chips. Jack Kilby did not build the first integrated circuit in a cleanroom.
The first design of a filtered, well-controlled airflow to achieve improved air quality in a workspace was led by Willis Whitfield at Sandia National Laboratories in the early 1960s.
This development was in direct response to the need to produce components with greatly improved long-term reliability for use in the nation's weapons program. Sandia built a prototype cleanroom in 1961 that featured what was called laminar airflow between the inlet and outlet ducts of the protected space. The idea was that the makeup air entering through filters establishes a flow of clean air that sweeps any contamination created within the workspace into the exit ducts and away from product surfaces.
While Whitfield is now universally credited with developing the first cleanroom, no patent application was ever filed. I've heard many sighs around Sandia concerning this oversight.
The next step in developing the cleanroom concept grew through the actions of a Federal committee initially organized by Sandia workers. With the approval of the General Services Administration (GSA), this group prepared the first Federal Standard 209 (Fed-Std-209), dated December 1963.
It was during the early 1960s that awareness was building concerning the sensitivity of semiconductor structures to contaminants. This was forcefully illustrated by the experience of Bell Labs' Telestar I satellite.
After a period of successful operation in space, this satellite went dead for no apparent reason. Shutting off the electrical bias on the transistors and then repowering restored operation.
The explanation was that electrically charged surface impurities-ions-had drifted along the transistor surfaces under the influence of their operating voltages, creating accumulations of surface charges adjacent to the electrodes. This charge buildup, in turn, induced a surface inversion layer in the transistor collector region, much like the action of a gate voltage in controlling the source-drain current of today's MOS transistors.
In the Telestar transistor, however, the inversion layer acted as a shunting path, shorting or severely degrading the collector-base junction. Removing the transistor voltages eliminated the surface electric fields and allowed the ions to redistribute themselves more uniformly so that the inversion layer disappeared as did the junction-shorting path it created.
This experience foretold the importance of controlling surface alkali ions in MOS technology, as detailed in seminal publications by the research team of Grove, Deal, Snow and Sah at Fairchild Semiconductor Inc. in the mid-1960s. These publications and others made it clear that successful semiconductor manufacturing required new cleanliness standards and practices and that the existing specifications of what constituted an acceptable environment for semiconductor manufacturing needed significant upgrading.
The nascent cleanroom technology under development within the federal establishment thus proved most timely.
In subsequent years, GSA assigned the task of updating Fed-Std-209 to the Institute of Environmental Science and Technology (IEST), an organization representative of a broad spectrum of industrial activities concerned with clean workspaces. Five updated versions followed, the latest being Fed-Std-209E, September 1992. Since then, the IEST has led an international group in preparing international contamination standards, including those specifying cleanroom classification and verification procedures. With the issuance of the international standards, ISO 146644-1 and -2 in the late 1990s, the GSA, earlier this year, officially recognized them as the rightful successors to Fed-Std-209E, signaling that the 209 saga has ended.
Robert P. Donovan is a process engineer assigned to the Sandia National Laboratories as a contract employee by L & M Technologies Inc., Albuquerque, NM.