DuPont knew for decades that the Teflon chemical C8 was contaminating drinking water and harming health, and hid it

A chemical built to last, in the wrong places

In 1951, DuPont started using a compound it called C8, more formally perfluorooctanoic acid or PFOA, at its Washington Works plant on the Ohio River near Parkersburg, West Virginia. C8 was a processing aid in the manufacture of Teflon and related nonstick and stain-resistant products. The chemical itself was made by 3M, which sold it to DuPont for decades.

The property that made C8 useful in industry is the same property that made it dangerous in the environment. It is exceptionally stable: the carbon-fluorine bonds at its core do not readily break down in nature or in the human body. That persistence is why PFOA and its chemical relatives are now known as “forever chemicals.” Once released, C8 does not go away on any human timescale. It accumulates: in soil, in rivers, in tap water, and in blood.

For years, DuPont released C8 into the air and water around Washington Works and disposed of C8-laden waste on and near its property. The chemical migrated into the groundwater and into the public and private water supplies drawn from the Ohio River and local wells, on both the West Virginia and Ohio sides. Tens of thousands of people drank it, for years, without knowing it was there.

What the company knew, and when

The heart of this case is not that contamination happened; industrial pollution is common. It is that DuPont's own internal research had been flagging problems for decades, and the company kept that knowledge to itself while continuing to release the chemical and reassure the public.

The internal record, later pried loose in litigation and detailed in reporting such as The Intercept's Teflon Toxin investigation, runs a long way back. Company and supplier studies from the 1960s and 1970s pointed to toxic effects in laboratory animals. DuPont monitored its own workforce and found C8 building up in employees' blood. In 1981, after 3M reported a study finding birth defects in the offspring of exposed rats, DuPont tested women in its Teflon operations, detected C8 in their blood, and moved some of them out of the area, without laying the broader hazard before regulators or the surrounding communities.

Perhaps the most telling single episode came in 1984, when DuPont quietly collected tap water from communities near the plant and found PFOA in local drinking water, in places above the company's own internal guideline. Internal deliberations weighed cost and liability. The public was not told, and the releases continued. This is the pattern the documents establish: not ignorance, but knowledge that was measured, recorded, and withheld.

The farmer, the lawyer, and the paper trail

The case might have stayed buried without an unlikely pairing: a West Virginia cattle farmer and a corporate-defense lawyer. In the late 1990s, Wilbur Tennant, whose cattle were dying near a DuPont waste site, brought his story to Robert Bilott, a Cincinnati attorney whose firm had historically represented chemical companies. Bilott took the case, and then kept pulling the thread.

Through discovery, Bilott obtained an enormous trove of DuPont's internal documents, the records that showed how much the company had known about C8, for how long. His years-long pursuit became the subject of a 2016 New York Times Magazine investigation, “The Lawyer Who Became DuPont's Worst Nightmare,” and later a book and a feature film. The significance of Bilott's work is not rhetorical: it is that the concealed record became a public record, admissible and citable, rather than a suspicion.

In 2001, the Tennant case settled. But Bilott had by then grasped the scale of the contamination, and in the early 2000s he filed a class action, Leach v. DuPont, on behalf of the tens of thousands of people whose water carried C8. That case would produce the settlement that, in turn, produced the single most important piece of science in the whole story.

The C8 Science Panel and the 'probable link' standard

The 2004 Leach settlement did something unusual. Both sides agreed to fund an independent C8 Science Panel of epidemiologists, and to be bound by its conclusions on whether PFOA was linked to disease. To give the panel data, a related effort, the C8 Health Project, gathered questionnaires and blood samples from roughly 69,000 people in the affected water districts. It remains one of the largest studies of its kind ever conducted on a single contaminant in a single population.

Between 2011 and 2012, the panel issued its findings. It concluded there was a “probable link” between PFOA exposure and six conditions: kidney cancer, testicular cancer, ulcerative colitis, thyroid disease, pregnancy-induced hypertension, and diagnosed high cholesterol. For a number of other conditions it examined, it found no probable link.

The exact meaning of that phrase matters, and it is easy to overstate. The panel defined a “probable link” as a determination that, given the available science, it was more likely than not that exposure was connected to the disease at the level of the exposed population. It is an epidemiological association, not a verdict that C8 caused any specific person's cancer. That distinction is precisely why the settlement structure used the finding as a gateway: a probable link opened the door to individual personal-injury suits, in which causation for a particular plaintiff still had to be argued.

Penalties, settlements, and a national limit

Accountability arrived in stages. In December 2005, the EPA settled its enforcement case against DuPont for $16.5 million, for failing to report health and environmental information about PFOA that federal law required it to disclose. The agency called it the largest civil administrative penalty it had ever obtained under a US environmental statute. In the context of DuPont's revenues it was a modest sum, but as an official finding that the company had unlawfully withheld information, it was significant.

The litigation kept compounding. In 2017, DuPont and its spinoff Chemours agreed to pay $671 million to resolve about 3,550 personal-injury claims from the Ohio valley. In 2023, DuPont, Chemours, and Corteva reached a roughly $1.18 billion settlement with public water systems, and 3M agreed to a deal valued at up to $10.3 billion, paid over 13 years, to help utilities across the country remove PFAS from drinking water.

The company measured C8 in its workers' blood and in the town's water, and did not say so. That is not a theory; it is in the record the courts and the EPA acted on.

The regulatory endpoint came in April 2024, when the EPA finalized the first enforceable national drinking-water limits for PFAS, setting the maximum for PFOA and PFOS at 4 parts per trillion, with a health-based goal of zero, reflecting the agency's conclusion that there is no level of exposure to these particular chemicals free of risk. The rule extends far beyond Parkersburg, an acknowledgment that the problem DuPont's files first documented is now a national one.