Wet Process Stations Cited for Non-Particulate Contamination
By LISA A. COLEMAN
Phoenix, AZ–Cleanrooms work well to control particle contamination but how can contaminants from a wet bench be prevented from getting into the cleanroom air? That`s the question Don Tolliver, Motorola assignee at SEMATECH at the University of Arizona, has labeled a significant and emerging issue.
“This is something the industry hasn`t studied enough to understand the degree to which a wet process station will contaminate a cleanroom,” says Tolliver. As he describes, the pressure inside a cleanroom should be at a steady state that`s greater than the outside air pressure. Therefore, when a door opens, for example, the overall balance inside the cleanroom and the air being exhausted from a wet bench is disrupted. Says Tolliver, “A little bit of that air that was going to be exhausted didn`t get exhausted. It came back into the cleanroom.”
Traditionally, pressure measurements inside a cleanroom are not taken very often. Because large cleanrooms have several air flows moving in different directions, cleanroom operators aren`t aware of pressure changes near any particular tools at any one time. “But those short intermittent pressure changes are causing these tools to contaminate the cleanroom,” warns Tolliver. “Are cleanrooms clean? I say `no.` We`re beginning to get more and more data which says cleanrooms are loaded with cation, anion and organic contamination.”
A New Frontier
For the last 25 years, the cleanroom industry has, in general, been measuring the performance of cleanrooms by one basic parameter–particles. Tolliver faults the industry for not having a technology roadmap that specifies cleanness via metallic, cation, vapor phase and organic concentrations in addition to the typical particle counts. “We`re at a new frontier in cleanroom technology where we`re beginning to discuss non-particulate contamination. That`s the big difference,” says Tolliver. “If organic contamination is depositing on my wafers and my tools then I`m starting to hurt the product directly.”
Tolliver believes that mini-environments offer a contamination control solution for wet process stations used in microelectronics processing, noting that in Asia, at least five fabs have moved away from cleanrooms and are using and building only mini-environments. Although there are three to four fabs in the United States that use mini-environments companies such as Intel and Motorola have yet to agree that mini-environments are worthwhile. Says Tolliver, “It`s hard for companies to believe that what they`re doing isn`t quite appropriate when they`re making a lot of money. There`s risk in changing. My conclusion is that there`s also a risk if you don`t go to mini-environments.”
Tolliver does acknowledge that some specific challenges will have to be met. These include real-time control of airflow balance; co-mingling of acid fumes with recirculation air, compatibility with conventional HEPA and ULPA filters; and molecular contamination and particle control. Also, process automation will be required along with design customization.
“Mini-environments will offer improved operational safety, process control and manufacturing flexibility as well as reduce the cost of ownership,” says Tolliver. He also notes that there could be a 50 to 70 percent reduction in exhaust and isolation from the cleanroom`s contamination.
Tolliver recommends designing in and installing mini-environments where possible for wet process stations. Other recommendations include: not ignoring molecular contamination; watching co-mingling issues over time or age of the mini-environment; exploring new filtration media technologies; making pressure change readings a primary control parameter; looking at laminarity as a key design feature; and looking for nitrogen ambients by the year 2000. n