Critics say chemicals used in making chips cause serious illnesses
David Lazarus / SF Chronicle 3dec00
The popular image of semiconductor companies -- the backbone of Silicon Valley -- is one of spotless work environments where technicians wear special "bunny suits" and operate in so-called clean rooms.
But critics charge that the chip-making business is anything but clean and that many companies, through their use of toxic materials, continue to expose workers to a variety of potential health problems, including cancer, miscarriages and birth defects.
Such problems are only now coming to light due to the decades it frequently takes for problems to appear after initial exposure to toxins.
A flurry of recent lawsuits -- some settled out of court, others still awaiting trial dates -- underlines the threat that toxins pose to the semiconductor industry.
"There have always been carcinogens used in the making of chips," said Joe LaDou, director of the International Center for Occupational Medicine at the University of California at San Francisco. "They're still in use today."
More than 220 billion chips are manufactured annually, requiring up to 1, 000 separate chemicals and metals. Many of these materials are known or suspected carcinogens.
To be sure, the U.S. semiconductor industry, employing more than 280,000 people worldwide, has made significant improvements over the years in workplace safety and automation. Machines now handle many tasks that once required a human touch.
Chip manufacturers say that their factories met industrial standards of the time and that they have consistently improved working standards as more was learned about the possible health effects of various production materials.
The lawsuits charge that these companies knowingly exposed workers to a wide variety of hazardous toxins. They allege that the semiconductor industry has yet to acknowledge what critics say is a lethal legacy of employee mistreatment.
Jim Harris, 59, worked as a technician at Santa Clara's National Semiconductor Corp. in the mid-1970s. He now has leukemia -- something no one else in his family has ever contracted.
While Harris can't prove a direct link between his illness and the chemicals he once worked with, he is one of nearly 300 people -- former workers, family members, next of kin -- suing National Semiconductor, IBM and their chemical suppliers in three separate lawsuits for illnesses allegedly caused by exposure to harmful materials.
"Every day I would grab a machine, take it apart, spray it with chemicals and put it back together," Harris said. "I wasn't the best mechanic in the world, but I was adequate."
These chemicals likely included trichloroethylene, xylene, benzene and toluene -- each known to cause cancer or other major sicknesses.
For protection, Harris wore nothing more than a cloth smock over his street clothes and a helmet with a plastic visor.
Several years ago, he found that his concentration was starting to waver; he had troubling focusing on a single task. "I also began to cough a lot," Harris said. "I was lethargic. I got antagonistic for no good reason."
After leukemia was diagnosed, his mood darkened. The coughing had grown so severe that he found getting through each day an almost intolerable ordeal. He read a book about how to commit suicide. He planned to put an end to his suffering last fall.
Harris' doctor was subsequently able to prescribe medication that relieved his symptoms, and his illness has since gone into remission. "I've beaten the odds," he said.
Mike Gutberlet, 45, has yet to turn that corner. He worked as a technician at IBM's South Bay complex from 1981 to 1994 and then again from 1998 until March.
"Here I was working for Big Blue," he said. "I thought I was set for life."
Gutberlet never complained when asked to clean the "thick, gooey, brown, nasty-smelling stuff" that would gum up the chip-making gear. He used powerful solvents to scrape away the residue, and, following company rules, wore gloves and a cap but no face protection.
The solvents he used included isophorone, a suspected carcinogen, as well as xylene and iron oxide, which have been linked to a variety of neurological ailments.
"When you're cleaning, this stuff gets everywhere," Gutberlet said. "It's like playing in oil. I don't care what you're wearing -- it's going to get all over you."
He was diagnosed with testicular cancer last year. Surgery and radiation treatments followed. While Gutberlet had hoped the cancer was gone, recent tests have not been promising.
"We were an IBM family," he said, the anger in his voice clear. "They were supposed to take care of us, but they couldn't have cared less if I have cancer or not."
Gutberlet is also suing his former employer.
IBM declined to comment on the litigation. LuAnn Jenkins, a spokeswoman for National Semiconductor, said the lawsuit "is absolutely without merit."
In legal papers filed in June, National Semiconductor formally denied playing any role in workers' sickness. IBM said the same in papers filed earlier this year.
According to the Semiconductor Industry Association, a San Jose trade group,
the chip-making business has ranked in the top 5 percent for workplace health and safety among U.S. companies since 1972.
"Using chemicals that have the potential to cause harm doesn't mean it's not a safe industry. It is," said George Scalise, the association's president. "These chemicals are tightly controlled.
"Our environmental health and safety program is second to none," he said.
The association reported this month that global chip sales reached a record $18.4 billion in September, a 45 percent increase from $12.7 billion a year earlier.
While Scalise noted that bunny suits and clean rooms are designed to protect chips and not workers from contamination, he stressed that other safeguards are in place to prevent workers from exposure to potentially lethal chemicals.
"There is no possibility of exposure," he said.
UCSF's LaDou, who has studied chip-making and other tech-industry manufacturing processes for about three decades, laughed when told what Scalise had said.
"It constantly amazes me that people can call up the SIA and be told there's no problem," he said. "And they've been getting away with it for years. "
In fact, LaDou insisted, chipmakers are well aware of the dangers inherent to their production process.
But he said nearly all past efforts to hold the industry accountable for worker ailments hit a brick wall when it came to establishing a direct link between specific manufacturing materials and employees' illnesses.
"Virtually every known solvent has been used in the industry," he said. "But that's the problem. When you have a cauldron of chemicals that's causing cancer and birth defects, it becomes even harder to find a smoking gun.
"You can't prove cause and effect unless you can narrow down the individual chemicals," LaDou said.
In other words, with so many potentially hazardous materials involved, it becomes that much harder to say that an individual worker's ailment is directly linked to a known toxin.
There have been exceptions. A 1988 study published in the Journal of Occupational Medicine showed that up to 38 percent of pregnant women working in the chip industry's clean rooms had experienced miscarriages -- well above established norms.
Amid growing negative publicity, the industry conducted its own studies and eventually pinpointed ethylene-based glycol ethers as the cause of the problem.
Leading chipmakers have since phased out use of the chemical.
But LaDou said no significant research has yet been conducted on links between the industry's production materials and workers' cancer. The catch, he said, is that such research requires extensive, costly epidemiological studies.
More importantly, it requires the industry's cooperation to supply data on workers' illnesses -- something that chipmakers so far have been reluctant to disclose.
LaDou said the semiconductor industry's claims of exceptional workplace safety are skewed by a mixing of statistics. He said the overall safety numbers are reduced by a low rate of on-the-job injuries and do not reflect illnesses among workers.
"There's talk of cancer clusters," LaDou said, "but they're brought to us by attorneys because they're the ones putting together cases. The industry doesn't let us look at their data."
Semiconductor companies strongly challenge such assertions. They say data on workers' health, including reported ailments and diagnoses by staff medical personnel, are kept secret mainly for privacy reasons.
"A lot of this is employees' confidential information," said Dave Stangis, a spokesman for Santa Clara's Intel Corp., the world's biggest chipmaker. "It's between them and their doctors."
Intel is not the subject of illness-related litigation.
To deflect criticism, the Semiconductor Industry Association said last fall that it was forming an independent panel to review chipmakers' data on potential health risks. The panel's findings are not expected until next year.
But Bruce Fowler, director of the toxicology program at the University of Maryland, noted that neither he nor UCSF's LaDou have been asked to sit on the panel. Both men are leading authorities on safety issues related to the chip- making business.
Fowler said it appeared that the industry's independent panel is merely a smoke screen to keep companies' secrets hidden. The data given to the panel probably won't be exhaustive, he said.
"There's data, and there's data," Fowler said. "They're going to be very cautious with what they disclose."
More than likely, the panel will not hear about "Maria Sosa," 65, who worked at National Semiconductor from 1969 to 1995. Sosa, who asked that her real name be withheld, recalled an occasion in the mid-1970s when she and other workers had been asked to sort through rejected chips.
"All of a sudden," she said, "we felt something wet, but we didn't think anything about it. Later on, we learned that some of the chips were cracked and were oozing out stuff."
It could have been anything, but chances are high that Sosa was handling epoxy resin, which includes epichlorohydrin, a known carcinogen.
The rashes began appearing not long after, and the coughing attacks started a few years later. Then, in 1996, Sosa said her doctor found a tumor in her colon "the size of a golf ball."
More than anything else, she's confused about the cause of her ailments. It pains Sosa to think that the company where she worked for almost 30 years may have played a role in jeopardizing her life.
"I was happy there," she said. "I really was. Now I think maybe they didn't take care of us like they should have."
Alida Hernandez, 70, feels the same way. She worked at IBM from 1977 to 1991 and was always proud of being an employee of one of the world's most famous companies.
"I never thought that someone like me would ever work for a company like IBM," Hernandez said. She described herself as the sort of employee who would ask supervisors for additional tasks after completing a job.
One of her jobs was to wash the discs used in making computer hard drives. "You would be exposed to water with the chemicals," she recalled. "We wore gloves, but they were so thin that if there was a defect on the metal, the gloves would rip and you would get it on your hands."
Hernandez also remembered the fumes, which were so strong they made her head spin when she entered the plant. "When you went to eat," she said, "you didn't taste nothing."
In 1993, Hernandez was diagnosed with breast cancer and underwent a mastectomy. Her cancer has since gone into remission, but the shame she feels about her experience continues.
Hernandez has yet to tell her children, who also work in Silicon Valley, what has happened to her. Nor has she told her granddaughter, who is graduating from San Jose State University with a degree in computer engineering and has received job offers from the likes of Intel, IBM and Cisco Systems.
"I want to protect them from worrying," Hernandez said. "I'm just trying to find the right time to tell them."
The lawsuits against IBM and National Semiconductor are working their way through the legal system. One related suit against IBM is scheduled to be heard in New York in January, while the bulk of the complaints likely will be heard in a Santa Clara court later in 2001.
That is, if they ever come to trial. UCSF's LaDou, for one, believes the companies will do their utmost to reach out-of-court settlements. A trial, he noted, would require the chipmakers to reveal internal data.
"They'll give up many millions of dollars before they go to trial," LaDou predicted.
In 1996, a Campbell chipmaker named Zilog paid $2.25 million to settle a suit from workers at an Idaho plant who claimed solvents and fumes at the facility had caused illnesses and miscarriages.
Meanwhile, LaDou said that the highly competitive semiconductor industry will be slow to change its ways and that little if any pressure for reform will come from politicians, who have grown accustomed to lavish contributions flowing from Silicon Valley.
"I think we're looking at another 10 years of manufacturing mayhem before anything changes," LaDou said.
E-mail David Lazarus at email@example.com
|If you have come to this page from an outside location click here to get back to mindfully.org|