With standard ionizers, positive and negative ions are created by running very high voltage, either AC or DC, to a sharp point (emitter), causing a high-voltage corona discharge at the end of the tip. The field generated around the tip of the emitters causes very small white particles to accumulate on the tip. This buildup commonly is referred to as a “fuzz ball” because of the appearance.
Often, this buildup is mistaken as a byproduct of eroding material from the end of the emitter. This is not the case.
The buildup really is a combination of chemicals—nitrogen and oxygen from the air and hydrogen from the moisture in the air. These three elements combine in the field of the emitter tip to form ammonium nitrate (NH4NO3). The field also attracts the ammonium nitrate to the tip.
The crystal formations on the positive and negative emitters have a different appearance in respect to each other. The positive points grow very fancy, fuzzy crystal branches off the center ball that forms on the tip. The negative points have the same ball on the tip but exhibit only short stubs coming out of the ball.
The reattraction of the positive ions to the emitter breaks off the branches. As the crystalline growth builds up on the emitter point, the shape of the electrical field becomes distorted, and the efficiency of the point is adversely affected.
The ion output will begin to decline. Although most ion-producing systems can maintain a near-balanced output, as the total output decreases, the neutralization times will increase.
Cleaning Processes
Two approaches can deal with this problem:
Cleaning the emitter tips frequently. If this is done often, it can reduce the amount of buildup on the tip.
But where do the ammonium nitrate crystals go after being removed from the tip? Most methods of cleaning emitter tips involve brushing the end of the point, which knocks the debris off the emitter. The ammonium nitrate crystals then fall onto the worksurface.
Another way to clean the emitter uses a special tip cleaner that contains a mixture of isopropyl alcohol and deionized water. This material dissolves the ammonium nitrate and simultaneously captures any other debris, such as dust particles, so they do not spread onto the worksurface. The frequency at which these cleanings must be performed is largely a function of the humidity level of the area. The higher the level, the more frequent the cleaning cycle.
Flowing clean dry air or clean dry nitrogen across the emitter tip. This prevents the buildup of the nitrate compound. Supplying the dry gas around the emitter tip eliminates the moisture in the immediate vicinity of the tip.
Eliminating the moisture gets rid of the associated hydrogen that is needed to form the ammonium nitrate. Without the hydrogen to recombine with the nitrogen and oxygen, the crystals do not form on the emitter.
Purged-System Ionization
This latter method of reducing or eliminating the formation of the fuzz balls has become known as purged-system ionization. Significant work by IBM personnel resulted in developing systems fitted with manifolds for delivering very dry air or nitrogen (with <5 to 6 ppm moisture) to the emitter points.1 With an ample supply of the dry gas, only very small amounts of moisture will entrain in the area of the emitter point.Many companies use ammonia-gas dissociation to produce large quantities of nitrogen for these processes, and a portion of this can be tapped to provide the gas for the purged ionizing systems. The volume of clean, dry gas that is required will vary according to the design of the ionizing system. Usually, approximately 2 to 6 CFH/point is sufficient to prevent the buildup of the ammonium nitrate.
The gas must be free of any moisture or other contaminants such as pump oils. The point-cleaning maintenance requirement is greatly reduced. Since the shape of the field surrounding the emitter point no longer becomes distorted by the growth of the fuzz ball, the ion output of the system remains more stable over longer periods of time.
Major reductions in particulate counts, especially in the 0.05- to 0.5-micron size particles, have been observed where purged systems have been installed. With geometries shrinking into the submicron range in the semiconductor and other industries, particle contamination has become critical.
During photolithography processes, for example, yield losses can be attributed to particles deposited on masks, pelicles, and reticles. With less particles being deposited on the masks, subsequent cleaning efforts will be more successful.
Reference
1. Murray, K. D., Ainsworth, G. F., and Gross, V. P., “Hood Ionization in Semiconductor Wafer Processing: An Evaluation,” 1988 EOS/ESD Symposium Proceedings, pp. 195-200.
About the Authors
Albert Breidegam is vice president and director of technology at Semtronics. His background includes more than 25 years in solid-state design and engineering management with RCA. Mr. Breidegam, who joined Semtronics in 1982, holds several patents in ESD control products and is a member of the ESD Association and several of the group’s standards committees. He received a B.S. degree in chemistry from Ursinus College and completed graduate work in electrical engineering at the University of Pennsylvania.
Jeff Salisbury is a member of the engineering group at Semtronics. Before joining the company, he was ESD coordinator for domestic and international head gimbal assembly manufacturing at Quantum. Mr. Salisbury, a member of the ESD Association Standards Committee Ionization Work Group 3.0, received a B.S. degree in electronic engineering from ITT Technical Institute.
Semtronics, 211 Prospect Park, Peachtree City, GA 30269, (770) 487-6700.
Copyright 1998 Nelson Publishing Inc.
April 1998