Experts: Spin coat design put workers at risk

02/01/2004

By MARK A. DeSORBO

SANTA CLARA, CALIF.—While the spin coater has been described as makeshift ingenuity, two plaintiff witnesses in the cleanroom cancer case against IBM Corp. and one semiconductor equipment design engineer say a phonograph-like contraption did everything to preserve the integrity of the disk drives, but nothing to protect the workers from exposure to harmful chemicals.

Spin coaters are perhaps as old as the semiconductor and microelectronics industries, and experts who spoke with CleanRooms claim they are not only the technological brainchild of IBM Corp., but an un-patented platform that was adopted worldwide and is still used today—safety hazards and all.

"The people who designed the machine...had little to no knowledge of ventilation," says Robert Morris, a cleanroom ventilation expert. "[IBM] created the spinner technology. It uses nasty chemicals, and the exhaust was not designed correctly. What stuns me is that IBM's design for the spinners was copied by the industry."

Morris, president and chief executive of Flow Safe Inc., a Denville, N.J.-based manufacturer of airflow and control systems, is a key witness for the plaintiffs in the ongoing case against IBM. He, along with Scott Reynolds, an engineer at Binghamton, N.Y.-based Computer Aided Engineering Solutions, created computational fluid dynamic models to simulate airflow and working conditions in the cleanrooms, where two plaintiffs, Alida Hernandez and James Moore, worked.

Hernandez, 73, and Moore, 62, worked at IBM's disk drive and printed circuit board manufacturing facility in San Jose starting in the late 1960s and early 1970s. Both allege that exposure to chemicals at IBM later caused their cancers.

Morris and Reynolds say the virtual reality models they created would have unveiled not only severe cleanroom airflow problems, but also design flaws in spin coaters that, despite being packaged inside fume hoods, still put workers at an unusually high risk of acute chemical exposure.

"But it doesn't take a rocket scientist to figure out that you have clogged exhaust systems and you're exposing people," Reynolds says.

The spinners, he says, were clogged, and in all cases Reynolds and Morris were able to figure out chemical concentrations in the cleanrooms, which ranged from threshold limits to on-the-spot-poisoning. And whether there were 30 air changes per hour or 100 air changes per hour, it didn't make any difference.

"The faster the air was turned over, the more it was mixed and spread around and the concentration stayed the same," Reynolds adds "The cleanrooms were working just fine in terms of particulate control, but there was an added burden of chemical control, and they did not do a good job at controlling that."

Reynolds explains that the spin coater, or spinner as it is often called, operated at low revolutions so it could apply such chemicals as isophorone, acetone, formaldihide, xylene and ethyl amyl ketone to the disk or wafer.

"Then, there's a spin-off cycle, at high revolutions per minute to spread the coating uniformly to very thin layers," he says. "Then, they slow it way down to magnetize and dry it."

The machine, Reynolds says, has a spindle that holds the disk in place as it is spun at revolutions as high as 5,000 RPMs. It is at the perimeter of the disk where problems start to arise, Reynolds says.

"There is a trap around the perimeter that catches the fluid that flies off the disc," he explains. "This clogs the drain lines, and workers would have to push a wooden dowel to push the gunk out of the lines. Then there is an outer tray that serves as an exhaust plenum. That outer tray is designed to capture the fumes, but some of that gunk flies off, blocks that up. It was a very bad design, and there did not appear to be any effort to isolate the product and the chemicals being applied to the product from the people."

The most alarming aspect of the design, Morris adds, was the unavoidable law of physics.

"Fumes are trapped within the spinning substrate, and air is spinning with it and creates a vortex. It was like a tornado," he explains. "At that kind of RPM, the fumes get caught in the eye of the storm, but as soon as that machine is shut down, the fumes are released into the breathing zone of the worker."

And the complaints about the fumes in cleanrooms alone should have sparked IBM officials to take actions, Morris says, citing specific Occupational Safety and Health Administration (OSHA) regulations. Morris says he regularly corresponds with OSHA, and in a letter dated April 4, 2001, Richard E. Fairfax, director of compliance programs for OSHA, told Morris that employers are responsible for ensuring that fume hoods are functioning properly and implementing feasible control measures to reduce employee exposures.

"If an employer discovers, through routine monitoring and/or employee feedback, that fume hoods are not effectively reducing employee exposures, it is the employer's responsibility to adjust controls or replace hoods as necessary," Fairfax wrote, adding that OSHA does not "promulgate specific fume hood testing."

Fairfax continued, "If an employee believes that he or she is routinely overexposed to hazardous substances while working in or around fume hoods, he or she may want to file a formal complaint to the local OSHA area office."

Morris and Reynolds claim IBM cleanroom workers, like Hernandez and Moore, were denied the rights outlined in the OSHA regulations. "We know that the chemicals did get in the room because all the workers I talked with mentioned the odors," Reynolds says. "IBM's objective was to keep the product safe. Employee health was not part of the equation at all."

A semiconductor equipment design engineer who spoke with CleanRooms, on the condition of anonymity, says it has been a well-known fact for many years that filtration and exhaust systems of spin coaters are ineffective. The devices, he says, are built with one goal in mind: To protect the product.

"It's all specked, but there is no pollution abatement," he says. "There is a complete lack of participation on the management's part in the health of their workers. It is a lack of attention to detail, and due diligence. It's negligence."

When it comes to spin coaters, says the design engineer, the laminar flow theory does not work, and the belief that it does is just another example of the "legend and mythology that goes into manufacturing these machines."

He also says that while there are a lot of spin coaters in the industry that are clean because the process demands it, there are still many machines in operation worldwide that continue to put cleanroom workers as risk.

"We have all these laws on the books to stop stuff like this from happening, but how do these get enforced?," the design engineer asks. "Employers, by law, are supposed to instruct their employees about the chemicals and inform them that they may also call OSHA to file a complaint. So, then you have the whistleblower phenomenon, and they suffer egregiously because of their actions, even though they were encouraged by OSHA. It's a Catch-22, and this is a wake-up call, or at least, it ought to be."