Pressure measurement in ion implanters
02/01/2000
Eric Bopp, MKS Instruments Inc., Vacuum Products Group, Boulder, Colorado
A quick way to increase uptime and lower the cost of consumables on an ion implanter is to replace the hot cathode ion gauge in the ion source with a cold cathode gauge. The latter typically lasts three times longer, representing a significant cost advantage over time.
Ion implant tools can be tough on pressure gauges. Usually the pressure is measured with a conventional hot-filament Bayard-Alpert (B-A) ionization gauge, and a typical implanter has four such gauges. Gauges in the ion source chamber of an implanter are exposed to high concentrations of dopant gases.
 |
Figure 1. The effect of gauge outgassing — typical vacuum pumpdown curves for a hot cathode gauge and a cold cathode gauge. The cold cathode gauge, having no filament to cause outgassing within the gauge, indicates ultimate pressure at ~15 mins. The hot cathode gauge takes ~1 hr to indicate the actual state of the system.
The usual filament in a B-A gauge is thorium-oxide-coated iridium, chosen for its burnout resistance. This may not be a good choice for the ion source region, where it can be exposed to halogen vapors from BF3 sources, which degrade the electron-emission ability of the thorium oxide. The alternative is a tungsten filament, which quickly fails if it is exposed to air or oxygen. With either filament material, the hot filament gauge in an ion source chamber typically needs to be replaced every three months. Gauge failure is unpredictable and causes unscheduled maintenance downtime, which can be as long as 24 hrs depending on the gauge location. Replacing the source end gauge takes the most time, because of the care required to avoid exposure of personnel to hazardous gas residues. In the worst case, one or more wafers could be lost, depending on when the gauge fails and whether or not a gauge failure triggers an interlock-controlled shutdown.
 |
High pressures are also incurred in the "business" end of the implanter — the end station where the wafers are implanted. Pressure bursts during wafer exchange (mainly air) and during implant (mainly water and hydrogen) can exceed the upper pressure limit of a hot cathode gauge and reduce filament life. As mentioned, filament life is unpredictable and depends on the amount of abuse incurred by the filament, such as exposure dose (pressure x time) to high pressure or halogen-containing gases.
Figure 2. Electronic configuration of a modern cold cathode gauge with dual electrical connections to increase stability.
Because of the inherent problems associated with burnout in the hot cathode ion gauge (HCIG), some manufacturers of ion implant equipment have been looking for a longer lifetime gauge. In some instances, cold cathode ion gauges (CCIGs) may be the answer.
Cold cathode gauges
The cold cathode Penning (or Phillips) ionization gauge does not have a hot filament to burn out. It is sometimes called a glow-discharge gauge. It is simple and has no glass envelope to break, no fragile electrode structure, and negligible outgassing (Fig. 1). Kendall concluded that "cold-cathode gauges are free of ... thermally produced outgassing effects" and "...the modern designs are very competitive with hot-cathode gauges in many applications. Their advantages increase as the pressure falls" [1].
Many engineers perceive hot cathode gauges to be more accurate than cold cathode gauges. In reality, the two types of gauges can be equally accurate. The original Penning gauge has evolved into the modern inverted magnetron (IM) cold cathode gauge that is linear over the pressure range from ~10-9 Torr to ~5 mTorr (see Fig. 2). The repeatability (~±5%) and sensor-to-sensor matching (±20-25%) of IM gauges is equivalent to conventional glass-envelope hot cathode gauges [2, 3].
Although a cold cathode gauge is reluctant to start at low pressures (starting depends on a random ionization event triggered by a cosmic ray), there is no starting delay at 1 mTorr or higher. The lower the pressure, the longer the starting delay will be (up to several minutes at 10-6 Torr). The MKS HPS
IgniTorr, for example, shines ultraviolet light into the IM gauge to trigger ionization to address this problem. It is a simple retrofit that uses a small UV lamp that can be replaced without breaking a vacuum seal.
Although the unit cost of a cold cathode IM gauge sensor can exceed that of a glass-envelope BA gauge tube, the cost-of-ownership (COO) can be lower because of the extended lifetime in an ion implanter — typically three times that of a B-A gauge tube. The sensor could be considered a consumable; however a contaminated cold cathode gauge can often simply be cleaned and re-installed, further extending lifetime. In this case, the cost of cleaning must be weighed against the cost of a new sensor. IM gauges also provide greater repeatability. A hot cathode gauge tube with equal repeatability would cost more but would still have a limited useful life. The table above compares COO based on an IM cold cathode gauge lifetime three times that of a hot-cathode gauge. Rebuild costs include parts only and do not include labor.
Cathode choice, measurement accuracy
The Implant Division of Varian has designed its implanters to allow customers to choose the type of gauge they prefer. HCIGs and CCIGs are equivalent in form, fit, and function and are completely interchangeable. According to Morgan Evans of Varian Semiconductor Equipment Associates (VSEA), "Our recommended configuration is three hot cathode gauges and one cold cathode gauge (in our implanters) because we believe it is the combination that will give the customer his preference for a HCIG in the end station while maintaining system integrity with a CCIG in the source. Because of the contamination problem, we strongly recommend a cold cathode gauge for the ion source. But many customers still have the perception that HCIGs are more accurate."
 |
Dual tungsten filaments are used in the HCIGs that Varian ships. This provides a back-up filament and avoids using radioactive thoria-coated filaments. Glass tube breakage is avoided by using miniature nude ion gauges within a stainless steel sleeve. An auto shut-off system also protects the filament in case of a high-pressure excursion. Evans stated that the customer base is split on how to handle a burned-out gauge. Some customers rebuild, but then there is the perceived risk of an improper clean or damage to the gauge elements in addition to the downtime caused from verification that the gauge is operating correctly. Some replace it with CCIGs. In the micro-ion gauge HCIG, rebuilding is not an option; the sensor must be replaced.
Evans also commented on the requirement for accuracy in pressure measurement: "In the end station, the pressure reading is not a programmed process parameter in our machines. Rather, it is a go or no-go indicator as to whether the pressure is low enough so as to not affect the dose. The maximum tolerable pressure depends on the implant recipe and can vary from 1.0 x 10-6 Torr to 9.9 x10- 6 Torr, with ~ 5 x 10-6 Torr being typical." It is important that the end station gauge be accurate and repeatable in this pressure range. Users should also be aware that the absolute sensitivity of an ionization gauge can vary up to 25% for the same type of gauge. Also, the gauge sensitivity for hydrogen is one-third to one-half of that for nitrogen. This means that a given pressure reading in hydrogen represents a "true" pressure that is two or three times higher [4]. Sensitivity and variability specifications should be obtained from the gauge tube manufacturer. (Cold cathode gauges from MKS Instruments meet these specifications and are now starting to be seen as a viable alternative to the hot cathode in the ion implant source chamber.)
Acknowledgments
I thank Morgan Evans of Varian Semiconductor Equipment Associates for his contributions to this article. HPS and IgniTorr are trademarks of MKS Instruments Inc.
References
- B.R.F. Kendall, "Ionization Gauge Errors at Low Pressures," J. Vac. Sci. Technology., A 17 (4), pp. 2041-2049, July/Aug. 1999.
- D.L. Hyatt, N.T. Peacock, "Long Term Measurement of an Inverted Magnetron Cold Cathode Gauge," Proceedings of the 37th Technical Conference, Society of Vacuum Coaters, pp. 409-412,1994.
- R.N. Peacock, N.T. Peacock, D.S. Hauschulz, "Comparison of Hot and Cold Cathode Ionization Gauges," J. Vac. Sci. Technology., A 9 (3), pp. 1977- 1985, May/June 1991.
- J.F. O'Hanlon, "A User's Guide to Vacuum Technology," (Wiley, New York, 1982).
Eric Bopp received his BS in mechanical engineering from the University of New Mexico. He is the current secretary of the ISOTC112 subcommittee for flanges and fittings. Bopp is a senior applications engineer at MKS Instruments, Vacuum Products Group, 5330 Sterling Dr., Boulder, CO; ph 800/345-1967, fax 303/442-6880, e-mail [email protected].