Poison Valley 

Part 1

 

http://archive.salon.com/tech/feature/2001/07/30/almaden1/index.html

Salon.com

Poison Valley
Is workers' health the price we pay for high-tech progress? First of two parts.

- - - - - - - - - - - -
By Jim Fisher

July 30, 2001 | At the south end of Silicon Valley in the foothills of the Santa Cruz Mountains, beside a creek thick with buckeye and sycamore, lie the ruins of California's first and richest mine. For over a century the red ore known as cinnabar, first roasted for its metal in 1845, was burned in furnaces at New Almaden Mine and reduced through a series of condensation chambers into approximately 100 million pounds of liquid mercury, used to extract silver in Nevada's Comstock mines and gold in the mother lode.

The mine is also the single greatest source of mercury pollution in the San Francisco Bay Area. After the mining companies sweated the quicksilver from the rock, they dumped an estimated 800,000 cubic yards of burnt cinnabar into nearby Alamitos Creek: To this day, drops of liquid mercury and cinnabar slag are readily found in samples collected anywhere between New Almaden Mine and the city of San Jose. If you follow that creek into Silicon Valley you'll pass signs showing a fish on a poker, and the warning:

FISH IN THESE WATERS ARE CONTAMINATED WITH DANGEROUS LEVELS OF POISONOUS MERCURY. DO NOT EAT FISH CAUGHT IN THESE WATERS.

The creek flows down through the serpentine foothills, through blue oak and California laurel, until it passes under McKean Road at the valley floor. A couple of hundred yards northeast on McKean takes you past a walnut orchard and across Calero Creek, also flowing down from the New Almaden hills and contaminated with mercury, until you hit the foot of the Santa Teresa Hills and the checkpoint for IBM's Almaden Research Center. IBM ARC, the first computer research lab west of the Mississippi, is the birthplace of a host of technological innovations as valuable as mother lode gold, and, according to a wave of recent lawsuits, as toxic as New Almaden Mine mercury.

If you make a U-turn at the check gate to IBM ARC and follow the run of Calero Creek as it flows along Camden Avenue, the pepper trees and flowering plums thin and shift from the creek bed to the center divide. After approximately five miles you reach the stoplight at Blossom Hill Road. To your right is the parking lot for the Santa Clara Valley Water District, and behind it are the Alamitos Groundwater Recharge Ponds, where the joined creeks of the eastern New Almaden hills (Alamitos and Calero) meet the Guadalupe River, flowing in from the west side of Almaden Quicksilver County Park. The combined water is then spread across the ponds, seeping through the earth's porous layers until it reaches underground aquifers, where it is stored until tapped by the county. Signs posted around the pools warn again of poisoned fish.

You are now officially in Silicon Valley. A few blocks ahead is the West Valley Freeway. Go east on the West Valley Freeway and after five miles you'll have driven over one of the largest plumes of poisoned groundwater in the United States, over 3 miles long and 180 feet deep, contaminated with xylene, toluene and other volatile organic compounds, including the chlorinated solvent trichloroethane (TCA). Pump-and-treat groundwater cleanup operations continue to this day. The original source of this poison? Underground Tank Farm No. 1 of IBM's Cottle Road Disk Drive Manufacturing Facility.

Built just three years after the disk drive was invented at IBM ARC in 1956, the Cottle Road plant was the first among dozens of manufacturing facilities -- including those operated by Intel, Hewlett-Packard, Applied Materials and National Semiconductor -- discovered in the early 1980s to have collectively leaked tens of thousands of gallons of organic solvents and other toxic contaminants into the groundwater of Silicon Valley. Today, the valley is home to more EPA Superfund sites (29) than any other county in the nation, with the most notorious of those sites -- from a leaking tank at a Fairchild Semiconductor fabrication plant -- poisoning a well that served the south San Jose neighborhood of Los Paseos. A subsequent study by the state's Department of Health Services found 2.5 to three times the expected rate of miscarriages and birth defects among pregnant women exposed to the contaminated drinking water, leading to a lawsuit and multimillion-dollar settlement in 1986 with over 250 claimants.

The toxic details of Silicon Valley's mercury-laden streams and contaminated aquifers are relatively well known. But another, even more troubling potential vector of deadly pollution has required more time to come to light -- the "clean rooms" in which high-tech workers come into direct contact with a vast array of chemicals as they manufacture semiconductor-laden circuit boards and computer hard drives. According to a lawsuit filed in 1998 in Santa Clara County Superior Court on behalf of four cancer-stricken IBM employees and the families of five deceased workers -- the number of plaintiffs has since quintupled to 45 -- Big Blue and its chemical suppliers, including Union Carbide, Shell Oil and Eastman Kodak, fraudulently concealed from their employees the risks of adverse health effects, including fetal toxicity and cancer, arising from chronic, low-level exposures to chemicals used in the manufacture of disk drives and related circuitry. Solvents named in the complaint include many of the toxic compounds leaked into the groundwater two decades before.

In January, IBM and two chemical suppliers (Union Carbide and Ashland Chemical) settled a separate case in a similar wave of lawsuits involving about 200 current, former and deceased IBM employees, most of whom worked at a huge chip-making plant in East Fishkill, N.Y. But the amount of the settlement was not made public and IBM admitted no guilt.

And yet, IBM's own corporate mortality statistics, charges the Santa Clara lawsuit, record a death rate from brain cancer among its employees about 2.5 times that of the general public. Did the chemicals involved in high-tech manufacturing cause the cancers? No one, not even experts who have long been critical of the potential safety hazards associated with clean-room workplaces, can say for certain. But numerous scientific studies have established that certain chemicals used in manufacturing semiconductors are statistically associated with increased rates of reproductive problems and various types of cancers. And the heart of the Santa Clara suit is the assertion that IBM repeatedly assured its workers that those workplaces were safe.

To the handful of experts occupied with the dismayingly difficult challenge of assessing the health threats of semiconductor manufacturing, IBM's alleged confidence could not possibly have been merited. There simply hasn't been enough testing and research into the health hazards posed by low-level exposure to combinations of toxic chemicals. If anything, the experience of the semiconductor industry should be sobering -- the complexity of the chemical cocktails at use in modern high-tech industrial manufacturing is mind-boggling, and it is always getting more so. There is little chance, warn these experts, of ever catching up with the public health challenges inherent in new advances in technology, especially when the rate of change continues to accelerate. We may know that mercury is deadly, we're pretty sure that drinking water contaminated with trichloroethane isn't a good idea and we may finally be waking up to the dangers of making clean-room workers breathe the same recirculated air, laden with complex chemicals, all day long. But what do we know about the explosion of research in biotech, and microelectronic machines, or the next wave of advances in semiconductor manufacturing?

Is the price of technological advancement, and its consequent economic growth, to be paid in workers' health? The legacy evident in Silicon Valley, since at least the 1850s, might hint at such a conclusion, although it also raises an obvious question: What alternatives do we have, if we are intent on technological progress? The lawsuits against IBM -- the consummate symbol of high-tech prowess -- might also give pause to the Silicon Valley's more ardent advocates of high-tech progress. But instead of attempting to help public health officials and their own workers keep up with the challenges of accelerating technological change, for years the semiconductor industry has been more interested in investing its dollars in pretending that problems don't exist.

Cottle Road, which today forms the western boundary of IBM's disk drive manufacturing facility in San Jose, is named after one of the valley's pioneer ranching families and forms part of what was once a vast Spanish land grant rancho. Orchards planted by the Cottles and dozens of other 19th century growers turned the valley into a world-famous provider of prunes and apricots, inspiring its first commercial nickname: the Valley of Heart's Delight. 

In the 1950s the prune and apricot orchards began to disappear to make space for the more than 2,500 electronics manufacturing firms that, by the early 1980s, had come to dominate the valley and would eventually lend it a new name, after the most common semiconductor substrate: silicon. The IBM campus, occupying approximately 1 square mile below Coyote Creek to the north and above the West Valley Freeway to the south, was built in 1959 on the commercial promise of the disk drive and solid-state electronics. At its peak it employed between 10,000 and 15,000 workers.

Virtually every computer currently manufactured owes something to the research carried out by IBM ARC scientists and the products then manufactured at the Cottle Road plant. ARC researchers came up with things like thin-film inductive heads, rotary actuators and sector servos -- technologies found in most every modern hard drive, be it Quantum, Western Digital or any other brand owing its skeleton to IBM patents. Without a hard drive, no computer, not the IBM Thinkpad 600E on which this story is being typed, nor any of the rack of high-powered Web servers on which this story is being served, would be anything more than so much heavy metal and miscellaneous plastics.

Today, the Cottle Road plant is still the principal factory transforming the research and development of IBM ARC into salable product. This is where patented chemical formulations used in optical lithography -- a process in which chip circuitry patterns are transferred onto silicon wafers -- and disk-drive coating are mixed, packaged and shipped. It is where proprietary microcircuitry and subassemblies for new generations of disk drives are manufactured in the famous clean rooms -- the factory floors of high-tech production whose highly protected environments require that workers take air showers before entering the "fab," and wear head-to-toe "bunny suits" to protect the wafers from microscopic debris.

"The tiniest speck of dust on a chip could ruin thousands of transistors," reads an exhibit at the Intel Museum in Santa Clara. Nowhere in the museum is it mentioned what health professionals and activists have attempted to point out since the late 1970s: that this "clean" environment has very little to do with safeguarding worker hygiene. The bunny suits may do an excellent job of preventing particles on employee clothes from damaging silicon wafers, but they are deplorably inadequate to protect workers against skin contact with the acids, solvents and other chemicals they use as a daily part of their job. Even worse, most clean-room ventilation systems are designed to recirculate the majority of the air used in the workplace, so as to prevent new infusions of airborne dust -- in effect, workers are breathing the same chemically suffused air over and over again throughout the workday.

"Had I known that I was working with anything that could cause cancer, I would have had second thoughts about going to work there," says Alida Hernandez, a former IBM employee and plaintiff in the Santa Clara lawsuit, who began her 14-year career at IBM washing residue from the surface of disk drives. She never knew what chemicals were in the wash, but a likely suspect is trichloroethane (TCA), a so-called safe substitute for the known carcinogen trichloroethylene (TCE), which itself was once touted as a safe substitute for the carcinogen perchloroethylene (PERC). In relatively low doses TCA can damage the liver, nervous system and circulatory system, and has been associated with brain cancer in gerbils exposed through inhalation. It is one of the contaminants in the solvent plume spreading beneath the Cottle Road plant, and shows up in Cottle Road's Toxic Release Inventory data as late as 1991 -- the year Hernandez left IBM.

Most of Hernandez's 14-year career, however, was spent in the disk-coating operations, where she was exposed on a daily basis to another mix of solvents and resins that also included known or suspected carcinogens, in addition to liver and nervous-system toxicants. "We were given classes as to what to do in case of an explosion, what kind of a fire extinguisher to use if it was electrical or if it was chemical -- those were the instructions they gave us. They didn't say anything about the chemicals being bad for your [biological] system, or possibly cancer causing, or anything like that."

Before starting each shift, it was Hernandez's responsibility to inspect the back of her "operation" -- as the coating workstations were called -- to ensure the machine was running properly. If the mixers were running too fast, for example, air bubbles could end up in the coating formulation and ruin a batch of disk drives, not to mention an employee's performance record. Workers were also responsible for cleaning the coating equipment with solvents several times throughout the workday.

"In coating you could only run 50 disks at a time without having to stop your operation and clean [the machine]," Hernandez says. Machines were cleaned chiefly with acetone, a moderately toxic solvent that is rapidly absorbed by the skin and is narcotic in high concentrations. Symptoms of acute exposure include convulsions, kidney and liver damage, and coma. Lower exposure symptoms include "slight intoxication, central nervous system depression, lassitude, drowsiness, loss of appetite, insomnia, somnolence, loss of strength, shallow respiration, weakness of the limbs, lightheadedness and general malaise."

The National Toxicology Program safety data sheet on acetone recommends that workers wear "a full face chemical cartridge respirator equipped with the appropriate organic vapor cartridges" when handling this chemical. Hernandez was never provided with a respirator, or any other means of scrubbing organic contaminants from the air.

Hernandez, who was frequently in charge of running several machines at once, estimates that she passed from 350 to 375 disks through each machine per shift.

"Sometimes the [machine] lines would plug up and it was up to the operator to unplug those lines. You'd get coating all over yourself -- I mean, it went right through your clothing. It went down to your skin. After you finished cleaning you just went and changed the outside smocks -- the bunny suits -- but your own clothing was all stained. It went right through the bunny suits."

After the film had been applied, the disks were placed in drying machines that spewed mists filled with acetone and coating. That coating, states the complaint, contained the organic solvent xylene. An aromatic hydrocarbon -- like benzene -- xylene has long been implicated in toxicological literature for its adverse effects on the peripheral nervous system. Additionally, commercial formulations of xylene -- at least in the early 1980s -- contained concentrations of up to a few percent of its carcinogenic cousin benzene, according to a 1986 journal article, "Carcinogens and Cancer Risks in the Microelectronics Industry." It too is one of the chemicals found in the Cottle Road groundwater plume.

Epoxy resins were another ingredient in disk coating, made from the compounds epichlorohydrin and bisphenol-A. The former chemical is mutagenic and genotoxic, and the latter is a known endocrine disruptor. Mutagenic and genotoxic "events" -- in which genetic material is changed or damaged -- are part of the first stage of cancer development, and may be indicative of cancer-causing chemicals. Epichlorohydrin is, in fact, a carcinogen. Endocrine disruptors are associated with reproductive and developmental harm.

Even today, clean-room workers continue to breathe recirculated air throughout their shifts. Machines are still cleaned, and metal surfaces degreased, with solvents, the most common being acetone and isopropyl alcohol, though more than a few companies -- particularly the smaller, less recognizable firms -- still use the carcinogen trichlorethylene or its cousin trichloroethane, according to annual Toxic Release Inventory data. To this day, the single most important chemical formulation in the manufacture of computer chips -- the photoresist -- is almost always a mixture of xylenes, carrier solvents, formaldehyde-based resins and genotoxic photoactive compounds. Other potential exposures in modern clean rooms include hydrofluoric acid, antimony, boron, phosphorous, gallium and arsenic.

Hernandez was diagnosed with breast cancer in 1993, two years after leaving IBM. Hernandez has no family history of the disease. At the time of her departure, two of her immediate colleagues had fallen ill, says Hernandez. One female engineer was on a leave of absence as a result of breast cancer, and the employee who had trained Hernandez on disk-coating operations came down with skin cancer. Another colleague suffered a miscarriage.

Hernandez never connected the illnesses with the job until she was diagnosed with the disease herself. "It's something you tell yourself always happens to somebody else, and never to you. When it happened to me, I started to think something was wrong."

"My mother's death should not have happened," says Carmen Navarro, daughter of former IBM worker Alicia Apodaca, who rinsed and inspected silicon wafers in the clean rooms of Cottle Road from 1980 through 1989, and died of breast cancer at age 51. As with Hernandez, and the great majority of women newly diagnosed with breast cancer, there was no history of the disease in Apodaca's family. "She was vibrant, healthy. She didn't smoke, she didn't drink, she took good care of her health. She was loved by her six children, and by her grandchildren, whom she adored."

"She had friendships with fellow employees at IBM -- a few of them have also passed away with cancer," Navarro says. One acquaintance died of lung cancer, another of brain cancer, says Navarro. "And it's continuing," she says. In mid-April, Navarro says she learned of another IBM worker of more than 20 years who was diagnosed with breast cancer. (IBM declined to comment on Navarro's and Hernandez's statements, citing pending litigation.)

"I believe that [IBM] knew that the chemicals were dangerous to the employees," says Navarro. I do believe that. This should not have happened."

"Workers are a kind of controlled experiment," says Dr. Sandra Steingraber, author of "Living Downstream: An Ecologist Looks at Cancer and the Environment," an authoritative study of the growing body of evidence linking cancer to the environment. "We know they work in certain workplaces for a certain number of hours with certain kinds of exposures. It's considered unethical to go out and do human experiments on a group of folks who aren't workers -- but this happens de facto in a lot of workplaces. Workers are the canaries in the mines."

In the East Fishkill lawsuit, former IBM workers Michael Ruffing and Faye Calton are the parents of Zachary Ruffing, 15, who was born blind and with facial deformities so severe he cannot breathe through his mouth or nose. They originally sued for $40 million in damages. Other Fishkill cases name cancers of the gastrointestinal and lymphatic systems; of the skin, bone and brain; and, most commonly, of the breast and testes. The cases filed by Cottle Road employees reflect a similar suite of cancers, the majority of which -- like the cancers listed above -- have all shown increased rates over the past 20 years and show longer-term increases that can be traced back at least 40 years, megatrends that correspond with the proliferation of synthetic chemicals following World War II.

In fact, workers' compensation statistics show that exposure to toxic chemicals -- coded as "systemic poisoning" in California -- is twice as likely to be a cause of occupational illness in electronics workers as it is for workers in other manufacturing industries. National figures from the Bureau of Labor Statistics show that the percentage of work-loss injuries and illnesses involving "exposures to caustic, noxious and allergenic substances" in recent years (1992-1998) was consistently between three and four times higher for workers in the semiconductor industry than in manufacturing industries as a whole, a group that includes manufacturers of petrochemicals, paper, petroleum, coal, steel, aluminum, plastics and rubber.

The BLS statistics do much to erode the perception that the high-tech industry is somehow "cleaner" than its predecessors. But what of the companies themselves? How much did they know about what they might be subjecting their workers to, and how hard were they trying to find out?

The simple fact is that it isn't in the high-tech industry's interest to know too much about the long-term health consequences of exposing its workers to toxic chemicals: The more it knows, the greater its legal liability. Of the few industry-funded studies of clean-room-related worker health problems, the two most significant examined workers' reproductive problems. One study was funded by the Semiconductor Industry Association, or SIA, the other by IBM. Both studies were conducted after activists raised concerns about the toxicity of a group of chemicals called ethylene glycol ethers, or EGE, used in photoresist.

The IBM-funded study, whose preliminary findings were released in 1992, found that pregnant employees at IBM's Fishkill lab who were exposed to EGE were roughly 1.5 times more likely to suffer a spontaneous abortion than unexposed workers. The authors emphasized that no conclusive causative chemical could be identified, but IBM acknowledged that it could be "inferred" that the cause of the increased miscarriages was exposure to EGE. Eventually, IBM and most of the industry stopped using EGE. (The SIA study came up with the same conclusions.)

What's noteworthy is that the gloomy results of this study didn't lead the industry to carry out more research into the long-term health consequences of exposure to other chemicals.

See no evil is a wise corporate strategy. But the Santa Clara lawsuit declares that IBM should have known that something was very wrong in its clean rooms, based on trends visible in its Corporate Mortality File, a database with work history on over 10,000 deceased IBM employees. Public access to the mortality file is currently restricted by a gag order, but the facts cited in the Santa Clara complaint are corroborated by statistics in a 1996 article in the scientific journal Epidemiology, "Brain Tumors Among Electronics Industry Workers." The file is a substantially complete (99 percent) database of all U.S. IBM workers of five or more years who died between 1975 and 1989; the records were constructed from death certificates obtained by IBM "for administrative purposes"; and the cause of death in 149 of the total 10,331 cases was primary brain cancer. (The article never specifically identifies the subject company, but a footnote identifies IBM as the funder of the research, and the mortality statistics are identical to those included in the complaint.)

That's quite a lot of brain cancer, about 2.5 times that of the general population, without factoring in biases for gender and age. More significantly, what this study found was an upward slope in brain cancer deaths among male electronics workers as duration of employment lengthened.

Because of the gag order, the other charges in the complaint -- that these records prove IBM knew that workers involved in manufacturing electronic devices were at a significant risk not only of brain cancer but of non-Hodgkin's lymphoma, gastric cancers and leukemia -- cannot be independently confirmed. (IBM will not comment on pending litigation.) But if one traces the citations in the Epidemiology article back through the scientific literature a pattern emerges that raises troubling, unanswered questions about elevated risks of cancer among workers in the manufacture and repair of electronics, and particularly among workers with long-term work histories -- specifically, 10 or more years -- and with probable exposure to solders and organic solvents.

In 1985, the same year the elevated brain cancer mortality rates began showing up in the scientific literature, Gary Adams, a chemist working in the material analysis department in Cottle Road's Building 13, where IBM disk drive coatings were developed, wrote a memo to IBM corporate headquarters. The memo alerted IBM officials to a cluster of cancers in his building. Eight out of his 14 immediate colleagues had fallen ill with some form of cancer.

Brain cancer had killed Adams' colleagues John Wong and Al Smith; lymphatic and hematopoietic cancers killed his colleagues Gordon Mol and Dwayne Johnson; and gastric cancers killed his colleagues Robert Cappell and Ken Hart, states the complaint. When Adams and another colleague, Fred Tarman, developed bone tumors, they decided it had to be more than a statistical fluke.

"All of a sudden we began to worry," Adams told "Dateline NBC" in 1998. "And then when another one [was diagnosed] and another one, it really began to hit home." Adams said the response of a staff doctor to his request that the company monitor its workers' health, particularly in Building 13, was to say such a program would be a waste of time, because "workers did not get cancer from their jobs."

The official stance of the semiconductor industry has long been similar. At the end of each year, when the Bureau of Labor Statistics releases the results of its survey on occupational health and safety, the Semiconductor Industry Association, which calls itself "the leading voice for the semiconductor industry," and whose member companies constitute more than 90 percent of U.S.-based semiconductor production, issues a press release announcing that the industry ranks among the safest manufacturing industries in the nation. (The current SIA chairman, incidentally, is John Kelly III, a senior vice president at IBM.)

Molly Maar, a spokeswoman for the SIA, says too little is known about the chemicals involved to point fingers at any particular industry. "What we're finding," she says, referring to cancer risks among clean-room workers, "is that there's not much scientific data out there ... Studies aren't inexpensive, and when you have many companies coming together, these things don't happen overnight."

IBM's short, official statement following the Fishkill settlement admits of no doubt:

"No scientific data supports the allegations of [the plaintiffs]. No evidence conclusively links the cause of [the plaintiffs' son's] birth defects to the chemicals in question or, for that matter, any specific chemical at all."

One of the toxicological literature's most detailed surveys of health risks in clean rooms, it turns out, was written by a former IBM industrial physician, Dr. Myron Harrison, in a 1992 article titled "Semiconductor Manufacturing Hazards." If there is a smoking gun for IBM, showing just how much it knew or should have known about potential health risks in clean rooms during the late 1980s, it is to be found is this exhaustive analysis of the potential hazards and exposure pathways of chemicals at every stage of chip making.

"If you look at the very early studies of chemical carcinogenesis," says Dr. Steingraber, author of "Living Downstream," "a lot of them were done by researchers who were industrial toxicologists, who might have originally worked for an oil company or something like that. They're right on the front lines ... When they have the courage and integrity to publish their findings, that's some of the best science that we have showing the relationship between chemical carcinogens and cancer."

Harrison's catalog of health risks is staggering. He lists potential exposures of workers to arsenic in the manufacture of gallium-arsenide wafers; to acid aerosols in the "wet etch" stage of chip lithography; and to toxic gases of arsine and boron in the operation of dopant implantation tools. He attests to cases of hydrofluoric acid burns during the cleaning of furnace tubes; of exposure to corrosive solvents in wet-stripping processes; and of untested photoactive compounds being sprayed by photoresist spinners. He warns of "catastrophic accidents" in the replacing of gas cylinders and the draining and refilling of wet chemical baths; of malfunctioning ventilation systems; and of widespread respiratory complaints among workers, including sinusitis, laryngitis and asthma. He documents mercury exposure from arc lamps; "relatively frequent" chemical fires at storage sinks; and solvent overflows in tool exhaust systems.

Harrison begins his article with an extensively diagrammed treatment of what remains the most worrisome -- and least acknowledged -- pathway of exposure in clean rooms: the vaporized mix of organic chemicals recirculated by the ventilation systems. A rule of thumb proposed by Harrison is that 90 percent of the air in a clean room is recirculated per hour, to minimize the introduction of contaminants that might degrade semiconductors or other advanced technological fabrications. He also shows how fumes can enter into circulation through "service cores," where vapors escape during equipment maintenance and where chemical spills are most likely to occur.

On top of that, recent evidence suggests that 15 percent of new fume hoods -- the local exhaust system for clean-room workstations -- fail to operate properly, potentially blowing toxic vapors back at the worker and into the clean-room environment.

"The ventilation conditions in clean rooms are very turbulent, and they cause a lot of problems," says Tom Smith of Exposure Control Technologies, a business that tests and evaluates laboratory ventilation systems. "Fume hoods [designed for the microelectronics industry], when we've tested them in clean rooms, generally only have a capture effectiveness about six inches above the work surface. If you get above that, or if you have a very volatile process, they just spill. And the clean-room airflow is so turbulent that it competes with these hoods, and the vapors escape from these hoods and infiltrate the return air system and are recirculated with the air handler."

And there are, without question, plenty of chemical vapors that can escape into the air system during the manufacture of a single computer chip, beginning with the pulling of a silicon crystal to the apotheosized "metallization" of the wafer -- the industry's term of art for deposition of electrical connections of aluminum on silicon. Figures based on a speech by a Texas Instruments fellow at the International Symposium on Semiconductor Manufacturing in September 1993 estimate that Intel's state-of-the-art chip fabrication plant in Rio Rancho, N.M., consumes, in a single year of manufacturing, 832 million cubic feet of bulk gases, 5.72 million cubic feet of hazardous gases and 5.2 million pounds of chemicals.

These figures, though prodigal, are deceptively simple, for they do not indicate the unprecedented spectrum of chemicals used in semiconductor manufacturing. In his 1992 article, Harrison prefaces a section titled "Selected Toxic Hazards" with the disclaimer: "An attempt to review the toxicology of all the thousands of chemicals in use at a typical fabrication plant is doomed to be superficial and of little value."

And the acceleration of the use of new techniques and new chemicals in new combinations in high-tech manufacturing makes safety evaluation harder all the time. "Professionals associated with this industry," wrote Harrison, "have invariably commented on the rapid pace of change in tools and materials, and on the fact that adequate toxicologic assessment of chemicals almost never precedes their introduction into manufacturing settings."

Harrison's frustration is echoed by Joseph LaDou, director of occupational and environmental health at the University of San Francisco. LaDou calls chip making "one of the most chemical-intensive industries ever conceived."

"The air-filtering systems do not alter chemicals except to dilute and recirculate them; and smocks and head gear do not protect workers from toxic exposures," LaDou wrote in 1984. He reiterates the point in an interview 17 years later. "Not only are you recycling the vaporized chemicals, but you're presumably allowing them to react with one another and introducing reactants into the air and recycling those as well."

"Most of our [health] regulations are predicated on workers being exposed to one chemical, maybe two or three -- but what do you do when they're exposed to a hundred?" LaDou asks. "What we have here is a brand-new work setting with an almost scientifically impossible question to answer -- how do you determine if a recirculated mix of chemicals is safe? -- and there is no magic formula."

"The problem with the spectrum of chemicals used in semiconductor manufacturing is that it could conceivably cause any cancer anywhere in the body," says LaDou. "When you find a cancer in a semiconductor worker, it's almost impossible to find a smoking gun."