Issue



Vibration control in nanotechnology research environments


11/01/2003







With geometries dipping into molecular and atomic scales, it's time to "design in" contamination control solutions

By J. Byron Davis and Ahmad Bayat, P.E.

In the high-tech research and manufacturing world, existing contaminants of any measure can mean the difference between success and failure. As technology progresses and length scales shrink, vibration and noise are increasingly being viewed as contaminants in the research and manufacturing environment—rising in the ranks with particles, microbes and waterborne and airborne molecular contaminants (AMCs).

Now that the relevant geometries in the research community have reached the molecular and atomic scale—with the manufacturing community not far behind—it's becoming more important to control vibration and noise. In an effort to move the nanotechnology community forward in this control, it's time to take a closer look into the kinds of sensitivities that exist as well as what can be done to solve vibration and noise contamination in nanotechnology research facilities.

Where's the nano action?

Nanotechnology encompasses applications on a nanometer (10-9 m) scale. It's a field that's poised to impact nearly every applied technical discipline, from computing to medicine to materials. Think of a process that enables us to snap together the fundamental building blocks of nature easily, inexpensively and in most of the ways permitted by the laws of physics.

Nanotechnology is not merely another stop on the miniaturization trendline; rather, the field represents a fundamental enabling technology that will undoubtedly revolutionize science and engineering.


Local and environmental sources affecting facility vibration include rail and automobile traffic, base building mechanical equipment, turbulent flow in piping and ducting, tool hookup equipment, and the movement of people and materials.
Click here to enlarge image

null

Currently, no large-scale nano-manufacturing exists. Probably the first products to be mass-manufactured will be materials rather than devices. For example, nanoscale particles have been developed as additives to paints and dyes. But to keep moving development along, many research laboratories are being established in the private sector as well as within universities and national labs systems in the U.S., Canada, Taiwan, Japan and Europe.

The research in these facilities will eventually lead to the development of mass-manufactured products. For now, however, nanoscale technology is confined to the research community.

Conceptually, vibration control in these facilities is no different than in other advanced technology facilities. Although research is conducted on length scales considerably smaller than in other high-technology facilities, the modes of sensitivity, vibration generation and vibration response are the same.

Through a careful process of design, with attention to the needs of the facility, very low vibration environments can be achieved. When even the "ambient" vibration environment—the environment without the building in place—is not quiet enough, non-conventional structures may be used to create "islands" of quiet in the facility.

How much is too much?

In the 1980s, generic vibration criteria were developed for use in advanced technology facilities, especially microelectronics manufacturing. These Vibration Criteria (VC) curves are widely used in industry and are now published as a standard by the Institute of Environmental Sciences and Technology (IEST).1

The VC curves are expressed in terms of 1/3-octave band velocity spectra. The curves are defined between 4 and 80 Hz, incorporating a constant velocity regime between 8 and 80 Hz and are relaxed slightly between 4 and 8 Hz.

The curves are named VC-A (the most relaxed, allowing a maximum velocity of 2,000 micro-inches/sec) through VC-E (the most stringent, allowing a maximum velocity of 125 micro-inches/sec). For reference, the "Residential Day" ISO curve at 8,000 micro-inches/sec is often taken to be a conservative approximation of the threshold of human perception.

Notes about the application and interpretation of the VC curves, including generic detail sizes, are given in Table 2 (page 19). A detailed discussion of the VC curves is given in Reference 2. Today, many general laboratory spaces are designed to meet VC-A or VC-B, while the current state-of-the-art in semiconductor manufacturing takes place in facilities meeting VC-D or VC-E.

It's notable that the relevant feature sizes referenced in Table 1 extend down to only 0.1 microns (µm), or 100 nanometers. It is likely that the VC-E criterion is acceptable for some uses below this feature size. Several changes and extensions to the VC Curves have been proposed over time, and other criteria exist.

In fact, we extend the constant velocity portion of the VC curve spectrum down to 1 Hz for design purposes in state-of-the-art facilities. Most notable of the alternative criteria is the NIST-A criterion used in facilities for the National Institute of Standards and Technology.

Click here to enlarge image

null

The NIST-A curve is the same as the most demanding VC curve—VC-E, for frequencies greater than 20 Hz. Below 20 Hz, the NIST-A curve is considerably more stringent. An even more demanding criterion, NIST-A1, typically cannot be achieved even at the foundations of most facilities and requires special structures.

Historically, the VC-E and NIST-A criteria have been looked upon as being close to a lower bound on facility vibration. This is particularly true for manufacturing facilities where the need for large-scale cleanroom support and high product volume places intense demands on building mechanical systems. This is also largely true for modern urban and semi-urban campuses, with their corresponding high density of facilities and traffic.

For uses that require stability beyond that presented by the ambient vibration environment, it's possible to create "islands" of quieter floor space in the facility. Using non-conventional floor structures, vibration isolation may be built into the floor. The most common arrangement is to raise part of the ground-level slab on pneumatic springs. Experiments and equipment are attached to this floated slab, while personnel and support equipment are located on a secondary raised floor not attached to the floating slab.

What kinds of criteria should apply?

In approaching the design of these nanotech facilities as well as those future manufacturing facilities, the first issues to be considered will be the tools and processes the facility will house.

One of the most interesting aspects of nanotech facility design is the highly interdisciplinary nature of the work. For example, it is not uncommon to see biologists, electrical engineers, mechanical engineers and materials scientists come together to form research teams. Given the mix of existing tools from disparate fields, as well as the development of new instrumentation and tools, a variety of vibration sensitivities exist. A sample list of tools commonly found in larger nanotech research facilities is given in Table 1 (page 18).

For example, some laboratories engaged in nanoscale work employ "traditional" imaging tools pushed to new limits, such as scanning electron microscopes (SEMs), transmission electron microscopes (TEMs) and optical lithography tools. These tools exhibit the same modes of sensitivity as they always have, including increasingly important sensitivity to acoustical noise.

Also important are the "wet" or "bulk" tools, such as chemical vapor deposition (CVD), epitaxy and etching. In some applications, processes that previously required explicit imaging techniques—patterning by lithography—might someday be accomplished using wet techniques—patterning by self-assembly. As in semiconductor manufacturing facilities, these wet tools are typically quite insensitive to vibration.

A few nano-specific tools have also been developed, such as atomic force microscopes (AFMs) and alternative lithographic processes. Also, some of these tools are dual use; for example, AFMs may be used to either passively image a surface or to actively move atoms around on the same surface. Due to the mechanism of the probe used in AFMs, these tools are less sensitive than expected, given the length scales on which they operate.

Perhaps the most sensitive work in nanotech research facilities is that of new instrumentation. Because tools under development operate on ever-finer length scales, utilizing novel mechanisms, this comes as no surprise. In addition, the fact that new tools under development don't necessarily incorporate extensively engineered isolation systems serves to further increase the sensitivity of these experimental devices.

Finally, contaminants such as vibration and noise easily confound experimental results, hindering instrument development; in turn, providing high quality environments to these uses is reasonable. In some cases, this means employing complicated non-conventional structures to achieve extremely quiet environments.

Requirements for the facility can be formulated once current tool needs and processes are assessed. For nanotechnology facilities, this might include a variety of spaces with a variety of criteria. As illustrated in Figure 2 (at left), these requirements drive three broad aspects of facility design, by which vibration control is built into the facility.

Incorporating vibration control

For some uses, it might be possible to isolate a sensitive tool or piece of equipment from the environment. This is often the approach taken in small laboratories, where only a single experiment or piece of equipment has some vibration sensitivity; for example, optics labs, where small components can be reasonably isolated on relatively inexpensive air-sprung tables.

In larger facilities, however, isolation of receivers becomes impractical. Furthermore, passive isolation—such as the vibration reduction afforded by air-sprung tables—cannot be applied to some types of tools and equipment.

Active tools that generate their own dynamic forces, such as modern photolithographic scanners, require very stiff bases rigidly mounted to the structure for their own stability. In addition, many vendor-supplied tools already incorporate pneumatic vibration isolators. Adding an isolation base typically results in unfavorable interaction between the two in-series isolators.

It should also be noted that applying vibration isolation to a tool or piece of equipment does not eliminate vibrations; rather, it reduces the vibration levels transmitted to the tool by some factor. Particularly in nanotechnology facilities, where the future direction of research and experimentation is unknown, the inflexibility and limited utility of the receiver-isolation approach is unpalatable. In short, addressing receivers of vibration is an incomplete solution to the problem.

Since providing piecewise isolation for sensitive equipment is not feasible for most advanced technology facilities, considerable effort must be made during the design of the facility to minimize both the forces that cause vibrations, as well as the response of the structure to those forces. In this regard, nanotechnology facilities are no different than facilities for biotech research or semiconductor manufacturing.


Facility requirements are driven by the needs of tools and processes, as well as by anticipated needs. Requirements are met via site selection and structural/mechanical system design.
Click here to enlarge image

null

A schematic cross-section of a generic advanced technology facility is illustrated in Figure 1 (page 16), with some types of vibration sources shown. The sources may be broadly categorized as environmental sources, such as nearby rail lines, roadways and adjacent facilities; and local sources, such as rotating mechanical equipment, turbulent flow in piping and ducting, and the movement of people and materials.

Because these facilities are usually located in urban or semi-urban areas, environmental sources are often significant contributors to the building vibration environment. Especially for cleanrooms, mechanical system vibration tends to dominate the vibration environment, as tremendous amounts of energy are consumed in maintaining the clean environment. It is often said that compared with non-cleanroom research and manufacturing environments, cleanrooms consume some 100 times as much energy per unit area (in the form of vibration-producing rotating mechanical equipment) while being some 100 times more sensitive.

For conventional structures, the ambient vibration environment, or the vibrations not attributable to the facility, represents the quietest possible state. No amount of concrete or steel can reduce vibration levels below ambient in these structures; in fact, the dynamic properties of the structure itself tend to amplify ambient vibrations in some frequency ranges.

Therefore, site selection is crucial, as there is no way to isolate a building against the environment without resorting to complicated, non-conventional structures, which themselves reduce but cannot eliminate ambient vibrations.

Click here to enlarge image

It's typically assumed that the only remedy for environmental sources is location. For example, if traffic on an existing roadway generates too much vibration, then the site is unacceptable. In fact, there have been cases where potential sites were rejected due to the threat of future impact from planned rail lines. For environmental sources, the forces generated by the sources themselves are not the only consideration; the response of the soils in the area as well as the propagation of vibrations in those soils is fundamentally important.

More control may be exercised over vibration sources local to the facility. During the design phase, major mechanical systems are carefully specified for items such as dynamic balance and maximum acceptable flow rates. In addition, detailed isolation schemes are devised for these systems.

Rotating machinery is typically mounted on heavy inertia bases supported by soft springs, while major piping and ducting is supported on resilient mounts. In some cases, restrictions are placed on the routing of large pipes and ducts to avoid attaching these to sensitive floors. Careful attention to specifications, isolation design, materials selection and installation insures that the dynamic load on the structure is kept to a minimum.

Reducing the vibration-causing forces acting on the structure is only part of the solution. Given the extraordinary sensitivity of the tools and processes in advanced technology facilities, the structure must be designed to adequately resist these forces.

A typical floor in the metrology or lithography area of a semiconductor fab, for example, consists of a "waffle" or "grillage" floor some 2 to 3 feet deep, with massive columns every 12 to 16 feet.

This floor structure is designed to be stiff enough to adequately resist the forces applied by local vibration sources. Additional design issues surrounding foundations, structural isolation breaks, and more are also important.

In research facilities, vibration and noise control are contaminants best dealt with at the earliest inception of the project. Knowing the needs of the facility, and weighing such factors as future flexibility and future needs, basic criteria may be formulated for the facility.

Where necessary, non-conventional structures may be used to achieve environments even better than that afforded by the ambient condition. From site selection to structural design to mechanical system isolation, the design process can deliver a vibration and noise solution for very high-quality research environments.

J. BYRON DAVIS is an associate and AHMAD BAYAT, P.E., is president of Vibro-Acoustic Consultants based in San Francisco, CA. Contact info: [email protected]

References

[1] Institute of Environmental Sciences (IEST), "Considerations in Clean Room Design," IES-RP-CC012.1 (1993)
[2] Amick, H. (1997). "On Generic Vibration Criteria for Advanced Technology Facilities: with a Tutorial on Vibration Data Representation," Journal of the Institute of Environmental Science, v. XL, No. 5.