Next-generation aluminum vacuum systems
05/01/1998
Next-generation aluminum vacuum systems
Glenn Tisdale, Judith A. Offerle, UHV Aluminum Company
Richard Bothell, Justin Bothell, Atlas Technologies Inc.
The semiconductor fabrication equipment industry is aiming to produce equipment that can process devices with 180-nm features on 300-mm wafers. As feature sizes shrink, molecular impurities incorporated into the devices during processing will increasingly limit device performance. Both the process materials and the vacuum environment contribute to these molecular impurities. The fabrication equipment community will need to improve the purity of the process materials and lower the base pressure of its processing tools to the ultrahigh vacuum (UHV) operating region.
Current fabrication equipment uses aluminum as a standard vacuum system material. The semiconductor industry is typically unaware that aluminum, however, is also UHV capable. During the past 15 years, aluminum UHV system development techniques have been perfected by the high-energy physics community. These techniques are being adapted for semiconductor processing applications and will allow aluminum processing tools to become UHV compatible without extensive design changes.
The semiconductor industry is constantly targeting new levels of production capability. The new process technologies that are at the heart of this advance will be based on fabrication tools that can handle 300-mm wafers with feature sizes <150 nm. However, devices with features of this scale are not in high-volume production and the processing techniques used for fabrication are still being extensively researched. As processing research proceeds, contaminating gases will induce defects in the deposited layers that will affect the functioning of the devices being constructed. These defect phenomena could be ignored in previous device generations, but will become increasingly detrimental as device geometries shrink [1].
Next-generation processing systems will need to provide contamination-free manufacturing at both the particle and molecular level. Molecular-level chemical contamination will become important in all processing technologies, particularly physical vapor deposition, CVD, and etch. While a given process may take place in an environment with gas partial pressures in the millitorr range, the vacuum system base pressure determines the level of molecular contamination inherent in that process.
 |
Automated welding hardware joins a bimetal flange to a prototype process module chamber. Bimetal flanges and automated aluminum welding are essential for producing alumninum ultrahigh vacuum system.
At high vacuum levels (10-6-10-9 torr), a monolayer of metal or gas contaminants forms in seconds. At UHV levels (<10-9 torr), however, a monolayer will typically form over the course of several hours. Upcoming device generations require deposited films that are not contaminated with impurities. In a UHV environment, the background impurity gas levels are low enough that, during a normal processing cycle, a wafer will be exposed to minimal levels of molecular contamination.
 |
The process tool community now faces the challenge of developing UHV-capable process tools, while continuing to improve their cost-effectiveness. Aluminum has long been a standard vacuum chamber material due to its ready machinability and low manufacturing costs. As UHV becomes a necessity, semiconductor tool manufacturers must decide to continue to embrace aluminum or to return to stainless steel, which has served as the de facto standard for UHV research hardware. Developments in aluminum instrumentation at synchrotron laboratories have demonstrated that stainless steel is no longer the only UHV material (Table 1). The technologies for aluminum UHV systems are now available for transfer into mass production environments. Stainless steel also has a number of drawbacks for semiconductor tool fabrication. First, it is more difficult to outgas than aluminum. Second, it is a potential source of iron and chromium contamination. Finally, because the metal is very expensive to machine, stainless steel systems are typically not geometrically compact and will require excess cleanroom floor space.
Aluminum, by contrast, is a less-contaminating vacuum material and lends itself to the production of relatively compact instruments. Aluminum systems may be fabricated at low cost and their overall cost of ownership is well understood. Consequently, aluminum has become the dominant vacuum system material for both low- and high-vacuum cluster tools and processing reactors. The semiconductor fabrication industry, however, is not aware that aluminum is now very capable of UHV pressures as well. In fact, aluminum is a natural platform for the development of semiconductor processing tools that are both cost-effective and UHV compatible.
Aluminum UHV systems
Until the early 1980s, the technology required to process aluminum into a UHV system had not been established. However, developments over the past 15 years in the accelerator physics community have allowed aluminum to become a UHV-compatible material [2]. These methods, processes, and components, which have been used to build large UHV systems for particle accelerators and synchrotrons, are gradually adapting to semiconductor processing applications.
Three primary technical challenges have been met to make aluminum UHV capable: the development of aluminum surface treatment schemes and outgassing reduction; the development of UHV-compatible welding techniques; and, finally, the development of robust, metal-sealed aluminum flanges. These technologies improve the outgassing and permeation performance of aluminum vacuum hardware [3]. While the new technologies provide superior vacuum performance, they do not require substantial modification to existing vacuum system production.
 |
Figure 1. Aluminum surface morphologies: a) carbon-based contaminants collected on a raw machined surface; and b) dense and nonporous thin native oxide formed from aluminum.
Aluminum surface treatment and outgassing reduction. As in processing untreated or machined stainless steel, we must first strip the off-the-shelf, mill-grade surface of aluminum in preparation for UHV application. We cannot use extruded and rolled plate stock directly, as it is covered with a relatively thick (typically 150-200 ?) porous oxide. In addition, during aluminum extrusion, alloy impurities migrate to the surface, where they combine to produce oxide impurities. Finally, when aluminum is machined, carbon-based contaminants tend to collect on the raw machined surface (Fig. 1a). To make aluminum UHV compatible, we must properly cleanse these surfaces of contaminants before we can produce a thin and nonporous surface oxide.
 |
Figure 2. Differentially pumped sealing geometry.
Hajime Ishimaru, of KEK Japan, was the first to concentrate on building aluminum vacuum systems for particle accelerators. For extruded surfaces, Ishimaru extruded aluminum into an argon-oxygen atmosphere. For machined components, he used ethanol as a machining fluid [4]. Both of these techniques allow the aluminum to form a thin native oxide that is dense and nonporous (Fig. 1b). The results for these techniques have been highly successful, achieving outgassing rates of 10-13 torr liter/sec cm2 after appropriate baking [5].
These techniques have been refined during the past 15 years. Work at a number of particle accelerator facilities has shown that surface cleaning with simple alkaline solutions is as effective as argon-oxygen extrusion and ethanol machining [6]. This alkaline solution cleaning removes impurity oxides as well as the top layer of aluminum oxide, forming a thin, dense, and nonporous native oxide on the newly exposed aluminum surface.
Aluminum vacuum flanging. Aluminum high vacuum systems have typically used elastomeric O-ring seals. Elastomers outgas and are permeable to atmospheric gases. Permeation occurs when atmospheric gases dissolve into the elastomer and then diffuse through the material, where they escape into the vacuum chamber. Unlike outgassing, permeation cannot be eliminated by baking the system. Gas permeation is a continuous system gas load that will not improve over time. Consequently, eliminating O-rings can significantly reduce pumping requirements and manufacturing cost.
There are two methods for eliminating O-ring gas permeation, both of which will be required in future aluminum UHV systems. The first is an old solution involving differentially pumped seals (Fig. 2). A separately pumped secondary channel isolates the main vacuum chamber from the atmosphere. Since permeation is proportional to the pressure differential, even a rough vacuum in the pump channel can reduce the permeation rate by several orders of magnitude. This scheme requires an auxiliary channel for each vacuum flange, as well as a pump to evacuate these channels. Differential pump schemes do provide the required vacuum performance for ports that require quick access, such as sputter target ports.
The other solution is to replace the elastomeric seals with metal-sealed flanges. Early metal-sealed flanges, composed entirely of aluminum, had a ConFlat-style knife edge that could seal an aluminum gasket. A coating of titanium nitride at the knife edges provided strength. However, these flanges face three limiting factors. First, the knife edge can be easily damaged due to the softness of aluminum. Second, overheating the flange during weld-up or heat cycling will anneal the aluminum, making the knife edge more subject to failure. The third is a more subtle system issue. Since most vacuum components are made of stainless steel, an aluminum flange will serve as a mount for a stainless steel component. These types of joints do not tolerate temperature cycling well because the differential thermal expansion between aluminum and stainless steel will loosen the knife-edge seal to the gasket and form leaks. The standard solution has been to provide a temperature differential during system bakeout between the two materials to match the thermal expansions.
 |
Figure 3. Bimetal flanges.
Atlas Technologies has constructed a bimetal flange to solve the problem [7]. This flange consists of a stainless steel surface and a ConFlat knife edge bonded to an aluminum nipple used for weld-up (Fig. 3). While it provides a robust sealing surface, it may be readily welded to an aluminum vacuum system. Because the stainless and aluminum surfaces meet at the bimetal bond rather than at the gasket, the sealing surfaces are not subjected to differential thermal expansion and consequent leaking. These flanges will serve as ready replacements for elastomer-sealed flanges that are not routinely opened.
Aluminum welding. The aluminum surface of a vacuum system can be treated to minimize outgassing and the system can be sealed with metal-sealed flanges to minimize permeation. The flanges, however, must still be bonded to the remainder of the system with welds that are nonporous and mechanically strong to maintain ultrahigh vacuum.
 |
Figure 4. Automated welding equipment.
Producing nonporous welds in aluminum is difficult for a number of reasons [8]. First, hydrogen becomes highly soluble to molten aluminum at high temperatures. When the aluminum subsequently cools, the hydrogen bubbles through the melt and forms pores. Second, aluminum oxide has a high melting temperature. Any aluminum oxide present at the weld surface will collect within the weld and form a defect layer. Third, the heat of the weld tends to dissipate within the material bulk due to aluminum`s high thermal conductivity, enlarging the heat-affected zone of the weld. Finally, since aluminum shrinks when it solidifies, aluminum welds are more subject to cracking than their steel counterparts.
While the above effects do not imply that aluminum welds cannot be produced, their production will require adequate material preparation and tight control over the processing parameters to insure repeatability. To meet the challenges of aluminum welding, the vacuum community has developed automated tungsten inert gas (TIG) welding processes, which, when coupled with careful material preparation, yield reliable UHV welds. Figure 4 presents an example of automated welding equipment.
Aluminum as a vacuum chamber construction material
When learning how to use aluminum for UHV system design, a vacuum system engineer experienced in stainless steel UHV system design will first be skeptical, then interested, and then excited by the powerful capabilities that aluminum offers. Our discussion of important Al properties uses stainless steel as a convenient benchmark.
 |
Alloys. Aluminum comes in a wide range of alloys with varying characteristics [9, 10, 11]. Wrought aluminum alloys are suitable for vacuum chamber fabrication and are designated by a four-digit number, followed by a temper specification (Table 2). The first digit specifies the primary alloy composite. Wrought alloys have been used for vacuum chamber fabrication, except for the 7000 series, which has zinc, an element with a high vapor pressure at low temperature. The 2000 series alloys are highly weldable. The 6000 series, particularly 6061 and 6063, have been used for UHV systems [9].
Machinability. One of the driving forces for fabricating vacuum systems from aluminum is that aluminum is inherently more machinable than materials such as stainless steel, which have been traditionally used for vacuum component fabrication. For example, the machining cost for 300 series stainless steel is 5.5? that of aluminum [12]. Because of this ready machinability, cluster tool components can be machined from a single plate of aluminum. These tools are exceptionally rigid, have a minimal vacuum surface area, and occupy minimal floor space.
Mechanical properties. Typical elastic modulii for aluminum alloy 6061T6 and stainless steel alloy 304 [9] are 7470 kgf/mm2 and 19700 kgf/mm2, respectively. If these values are used in mechanical formulae for standard geometries, the ratios of critical thickness for the two materials are [10]:
 |
Here, tAS(flat plate), tAS(long cylinder), and tAS(short cylinder) are the minimum thickness ratios to avoid buckling in flat plates, long cylinders, and short cylinders, respectively. Note that the ratios are close to unity. An aluminum vacuum system will not require parts that have appreciably greater thickness than similar ones manufactured from stainless steel.
Thermal conductivity. Aluminum`s thermal conductivity, depending on the alloy, ranges between 170 W/mK and 230 W/mK. Stainless steels, by contrast, have thermal conductivities that are between 14 W/mK and 16 W/mK. High thermal conductivity is an advantage when designing systems that require temperature cycling. This is the case for vacuum systems that must be baked to reach UHV levels. An aluminum chamber may be baked and then cooled much more rapidly than a stainless chamber. Furthermore, aluminum`s high conductivity allows a complete bakeout without recondensation of gases on local cool spots, a common problem in stainless steel systems.
Weight. Aluminum is roughly 1/3 the weight of stainless steel (2.8 g/cm3 [Al] vs. 8.0 g/cm3 [stainless steel alloys]). The cost burden associated with excess weight begins when the raw materials are handled and progresses throughout the manufacturing process. It affects all production steps, including shipping, installation, and even the architectural engineering and construction of the environment surrounding a process tool.
Magnetic properties. Aluminum is not magnetic, whereas stainless steel, being essentially an alloy of iron, exhibits residual magnetism. The absence of magnetic properties in aluminum is advantageous for applications involving charged particle beams, because the vacuum system will not modify the fields from the beam control magnets.
Radioactivity. Aluminum, in comparison to stainless steel, has a much more rapid decay of induced radioactivity. If both types of materials are bombarded with the same flux of charged particles, the residual radioactivity will typically be one to two orders of magnitude less for an aluminum sample than for an identical stainless steel sample [13]. The nuclear half-life of elements that make up stainless steel suggests that a-particle contamination is always present in stainless steel and a possible source of circuit damage.
Corrosion. The corrosion of both aluminum and stainless steel alloys in reactive gases is complicated. Experimental work performed on various alloys in different reactive gaseous environments shows that both aluminum and stainless steel are subject to attack by reactive gases; halogen-containing species are typically the most damaging; and the corrosion of any given compound is usually no worse than that of its halogen component alone [9, 10].
Aluminum is not a worse corroder than stainless steel. It simply has different reaction dynamics that do not serve as a source of iron and nickel contamination, one of the most significant yield-limiting factors for silicon IC production.
Outgassing properties. One of the most important properties of a vacuum material is the outgassing rate, as this determines the ultimate pressure that may be obtained in the vacuum chamber. Repeatable outgassing rates of <10-13 torr liter/sec cm2 are now possible in aluminum UHV systems [13], comparable to the best outgassing rates obtainable with stainless steel [14]. This improvement in outgassing performance has been one of the principal breakthroughs that has allowed aluminum to become a competent material for the construction of UHV systems.
Conclusion
The next generation of aluminum vacuum systems will be capable of UHV performance. Surface treatment, automated welding processes, and metal-sealed flange technologies have made this possible. The impact of UHV environments for contamination-free manufacturing processing has yet to be determined in a quantitative fashion. Current studies, however, indicate that they will be essential components of semiconductor materials processing and control and key aspects of 300-mm processing systems.n
Acknowledgment
We wish to pay our respects to Hajime Ishimaru, who passed away this year, and who pioneered the use of aluminum for UHV systems. We also thank James Garner of SMC Corp. for many helpful discussions and for his pivotal role in focusing our efforts within the semiconductor industry; the Vacuum Group of the Advanced Photon Source (APS) at Argonne National Laboratory, in particular, John Noonan, John Crandall, Joseph Gagliano, George Goeppner, Richard Rosenberg, and Dean Walters for being responsible for much of the foundation work in the area of aluminum UHV systems (Crandall has been especially instrumental in bringing these ideas to practice); and David Moncton, Russell Huebner, John Galayda, and Rudolph Damm of the APS for their continued support. ConFlat is a registered trademark of Varian Inc.
References
1. I. Hashim , I. J. Raaijmakers, "Vacuum Requirements for Next Wafer Size Physical Vapor Deposition System," Journal of Vacuum Science and Technology A, Vol. 15, No. 3, p. 1305, 1997.
2. R.C. Niemann et al., "APS Storage Ring Vacuum System," in Vacuum Design of Synchrotron Light Sources, American Vacuum Society Series 12, Conference Proceedings No. 236, p. 84, 1990.
3. J. O`Hanlon, "Ultrahigh Vacuum in the Semiconductor Industry," Journal of Vacuum Science and Technology A, Vol. 12, No. 4, p. 921, 1994.
4. H. Ishimaru, "Ultimate Pressure of the Order of 10-13 torr in an Aluminum Alloy Vacuum Chamber," Journal of Vacuum Science and Technology A, Vol. 7, No. 3, p. 2439, 1989.
5. J.R. Chen et al., "Thermal Outgassing from Aluminum Alloy Vacuum Chambers," Journal of Vacuum Science and Technology A, Vol. 3, No. 6, p. 2188, 1985.
6. R. Rosenberg et al., "X-ray Photoelectron Spectroscopy Analysis of Aluminum and Copper Cleaning Procedures for the Advanced Photon Source," Journal of Vacuum Science and Technology A, Vol. 12, No. 4, p. 1755, 1994.
7. Atlas Technologies, Port Townsend, WA, www.atlasbimetal.com.
8. G. Goeppner, TIG Welding of Aluminum Alloys for the APS Storage Ring - A UHV Application, Argonne National Laboratory Light Source Note 254, 1990.
9. G.S. Brady, H.R. Clauser, Materials Handbook, Thirteenth Edition, McGraw Hill Inc., p. 39, 1998.
10. J. Garner, "Aluminum-based Vacuum Systems," in Handbook of Vacuum Technology, pp. 509-552, eds. D. Hoffman, B. Singh, J. Thomas, Academic Press, Harcourt Brace and Co., 1997.
11. G. Goeppner, "APS Storage Ring Vacuum Chamber Fabrication," in Vacuum Design of Synchrotron Light Sources, American Vacuum Society Series 12, p. 124, ed. Y.G. Amer, American Institute of Physics, 1990.
12. Profile Milling Requirements for Hard Metals, 1965-1970. Report of the Ad Hoc Machine Tool Advisory Committee to the Department of the US Air Force, May 1965.
13. H. Ishimaru, "All-aluminum-alloy Ultrahigh Vacuum System for a Large-scale Electron-positron Collider," Journal of Vacuum Science and Technology A2, No. 2, p. 1170, 1984.
14. R. Nuvolone, "Technology of Low-pressure Systems - Establishment of Optimum Conditions to Obtain Low Degassing Rates on 316 L Stainless Steel by Heat Treatments," Journal of Vacuum Science and Technology, Vol. 14, No. 5, p. 210, 1977.
GLENN TISDALE is director of business development and VP of technology at UHV Aluminum Co., an ARCH Development Corp. company founded to promote the development of aluminum UHV hardware. UHV Aluminum Co., 1360 North Lake Shore Drive, Suite 1918, Chicago IL 60610; ph 630/323-8489, fax 630/323-8604, e-mail [email protected], www.uhvaluminum.com.
JUDITH A. OFFERLE is president and CEO of UHV Aluminum Co. Her background is in operations management and strategic planning for domestic and international businesses in development.
RICHARD BOTHELL is president of Atlas Technologies and a coinventor of the Atlas Flange. He has also been coprincipal investigator on DOE SBIR projects, developing explosion-formed, superconducting RF cavities, and NSF SBIR projects in semiconductor temperature measurement, using remote diffuse reflectance spectroscopy techniques. Atlas Technologies Inc., 301 10th Street No.8, Port Townsend, WA 98368; ph 360/385-3123, fax 360/379-5220, e-mail [email protected], www.atlasbimetal.com.
JUSTIN BOTHELL is a graduate of UC Berkeley and founder and VP of Atlas Technologies. He is a coinventor of the Atlas Flange. Bothell has been coprincipal investigator on DOE SBIR projects, developing explosion-formed superconducting RF cavities.