Targeting AMC

SPECIAL REPORT

Process engineers are now being charged with the task of controlling contamination control's most daunting menace.

by Chris Anderson

Click here to enlarge image

First addressed as a manufacturing concern by the Japanese more than a decade ago, airborne molecular contamination (AMC) has captured the attention of the semiconductor, microelectronics and optics industries here in the United States over the past half dozen years.

And while industry experts estimate that we are in the very early phases of learning about AMC, the affects on equipment, finished products and how to tackle specific forms of contamination, it is clear that in many environments the time to begin addressing AMC as a serious yield-limiting threat is now.

“We know that there are certain products and factors that are at risk of AMC,” says Chris Muller, technical services manager for Purafil Inc., a filter manufacturer based in Doraville, GA. “Something like the move to a copper process in semiconductor manufacturing is going to move AMC more to the forefront. We are getting to the point where AMC control is not just talked about, but is going to be required.”

Yet unlike the straightforward process of particle control, AMC takes many forms. From acids and bases, to dopants and condensibles, the damage from the most common types of AMC range from T-topping of integrated circuits and hazing of optics to chemical reactivity and oxidation of metals. Combine these dangers with the increasingly delicate and precise nature of the newer 300-mm wafer fabs and deep ultraviolet (UV) photolithography and it's easy to understand why so much attention has now turned to controlling trace elements and gasses in a variety of clean environments.

Mark Huza, carbon segment manager of Riverdale, NJ-based Camfil Farr Inc., has witnessed this shift in priorities. “There is a recognition today that AMC is a significant issue that needs to be addressed,” Huza says. “Five or six years ago, when I would give a presentation on AMC, I would look out and see blank looks on faces. It has now gotten to the point where AMC control is seen as a process enabler in that it adds value to the process.” Part of the evolution of effective AMC control has been shifting responsibilities. Huza notes that more and more companies are charging their process engineers with tackling the problem as opposed to those responsible for contamination control. “It is such a complex issue and there are so many different kinds of contamination that it is absolutely essential we get the support of the process people if we want to effectively get a handle on AMC,” he says.

Establishing a baseline
The relatively easy days of particle control, when the simple questions of “how small and how many” set a clear path for action, are long gone. When it comes to controlling AMC, the playing field of potential contaminants is much wider, and targeting which of them to control or remove from the environments is decidedly more complicated.

Click here to enlarge image

“There are more than 300 separate processes involved to make a chip and in many cases the materials used in one step can have a harmful effect on another down the line,” says Dick Matthews, technical director of Filtration Technology Inc. (Greensboro, N.C.). “Because of this, the semiconductor industry can be its own worst enemy when it comes to AMC.” For this reason, the first step in establishing a solid AMC program is an audit of the manufacturing environment to establish a baseline detailing what compounds and gases are present and at what levels.

“Knowing the baseline of contaminants in your facility is the starting point for any comprehensive AMC plan,” says Mark Camenzind, senior technical advisor, Air Liquide-Balazs Analytical Services (Fremont, CA). “It allows not only the targeting of specific compounds that may affect particular critical processes, but can also help return the facility to normal after a disaster or spill.” Further, Camenzind notes, insurance companies are now requiring manufacturers to collect this data as a condition of granting insurance, as it allows them to accurately determine expenses for clean up and abatement claims.

“It only makes sense to survey the environment, both to understand how it affects your process and to target areas for improvement,” says Muller. “If you are going to spend money on AMC control you need to understand what you need to remove, and that is why we do a reactivity survey as the first step. It allows for a greater focus.”

A comprehensive survey should also not be a simple snapshot in time, rather one that samples the environment over a week or even at various times over a month. That's because the weather outside the plant can cause molecular contamination levels to shift, literally with the direction of the wind. Proximity to carbon-fired industry or high sulfur dioxide emissions nearby can increase contamination levels by riding downwind and into a facility if not picked off in the makeup air. “We recommend that companies establishing their baseline monitor continuously for a couple of days,” says Morgan Polen, vice president of application technology with Lighthouse Worldwide Solutions, a provider of monitoring equipment and services based in Milpitas, CA. “We had a customer in Singapore who would literally see levels of certain contaminants spike whenever the wind was blowing from the East.”

Once a facility has established a baseline, it both informs managers which contaminants to target according to their specific process and provides a target for minimum contamination performance.

According to Huza, specific contamination work is much more difficult without this information. “Often, by the time we get involved, our client is having a significant problem and we have to scramble to come up with a solution based only on the conditions as they exist today.”

Monitoring the problem
Once an AMC control program is in place, companies should develop a program of environmental monitoring to validate and keep tabs on their efforts.

However, monitoring for AMC is not nearly as straightforward as setting up particle counters. At present, so-called real-time monitoring is expensive and is likely to stay that way as companies press for monitoring not just in parts per billion but parts per trillion.

Click here to enlarge image

Further, no single monitoring technology can give a complete picture of exactly which contaminants are present and at what levels. From surface acoustic wave, to reactivity and air monitors, each method can answer only a part of the molecular contamination question. “The monitoring equipment available today will give you a crude indication of contamination problems,” says Huza. “For instance, if a monitor will provide information on the level of bases in an environment, it doesn't specify which bases are present or if it is a problem.”

But Huza doesn't throw the baby out with the bath water when it comes to monitoring. “Even though there is not a single solution, an integrated monitoring approach can be a valuable tool whether you are drawing fab air through ultra pure water or using 'witness wafers.' A combination of these will help people manage their efforts.” One such company helping to provide this information is Particle Measuring Systems (PMS), a provider of surface acoustic wave monitoring technology based in Boulder, CO.

According to PMS, its monitors help determine not what molecules are in the environment, but when a range of contaminants are actually accumulating on a surface. “All the talk you hear is about AMC. What we are trying to do is look at the SMC, or surface molecular contamination, because it is on the surface that the damage is done,” says Dan Rodier, applications engineer with PMS.

Rodier says the equipment can be set to trigger an alarm if surface contamination exceeds a particular level. “This allows the user to investigate the reasons for the increase. Often it's easily connected to another event in the facility.”

Matt Middlebrooks, a senior research engineer at AQF Technologies LLC, a filtration media supplier based in Charlotte, NC, points out that while the monitoring can help sniff out process problems, it can have a practical application as well: to help determine when it is time to replace AMC filters that have stopped working at optimal levels.

“When is the right time to change out filters is a big question in the industry,” he says. “Unfortunately, sometimes filters aren't changed until it is too late and a problem crops up. Perhaps regular monitoring could be the spot check to make sure filters are operating as they should.” Monitoring for the key compounds and gases to control may not tell cleanroom managers that a filter is near the end of it's life cycle, but can provide information quickly should a filter stop being effective.

Filter developments
Many of the first filters sold to help control AMC used media that had already been developed for other applications; and, in many cases, pulled a range of molecular contaminants out of the clean environment.

Today, filter manufacturers say they are shoulder-deep into focused efforts to develop filters and filter media specifically targeting the needs of different industries. “I have seen the emergence of more targeted materials and media that were developed to filter specific chemicals,” says Middlebrooks.

While the market for AMC filters is still slow and the sales cycles are as long as one year, things are picking up steam. “We think that with the new technologies that are coming online, AMC is something everyone will have to look at one way or another,” he says.

Customers are asking filter manufacturers to provide filters that target specific compounds associated with specific manufacturing processes. “Unfortunately, many of the requests we have to design filters for these mini-environments too often revolve around the aesthetics of the environment and not function,” says Huza. “So we are often constrained by size and shape issues when designing for these environments.”

But this separation of manufacturing events using mini-environments may have real advantages in the fight against AMC. Smaller, chemically targeted filters will be less expensive, and, because they catch only specific compounds-not an entire array-they will last longer. Also, process engineers can target the manufacturing functions most sensitive to contamination instead of focusing on the entire process.

Still, for filter manufacturers and others combating AMC, technology advancements in manufacturing mean the science of AMC presents an ever-changing, moving target.


Chris Anderson is a special correspondent to CleanRooms Magazine.

POST A COMMENT

Easily post a comment below using your Linkedin, Twitter, Google or Facebook account. Comments won't automatically be posted to your social media accounts unless you select to share.