Read the EPA memo. 

CHARLESTON, W.Va. -- Federal government scientists originally sought a much tougher standard for the toxic chemical C8 than was included in a nationwide health advisory issued last month, according to an internal U.S. Environmental Protection Agency memo.

More than three years ago, EPA staffers proposed that the agency rewrite a deal with DuPont Co. to mandate a C8 limit of 0.2 parts per billion, according to the memo.

"EPA's intent is to seek consent through negotiations with DuPont," the memo said. "However, the order could be issued unilaterally if agreement cannot be reached."

EPA staffers wrote the undated memo, called a "briefing paper," sometime in March 2006. At the time, the EPA was preparing to negotiate with DuPont. Agency officials wanted to tighten a four-year-old deal that required the company to provide alternative water for Wood County residents whose drinking water contained higher levels of C8.

C8 is a common name for perfluorooctanoate acid, or PFOA. The chemical is a processing agent used for years at DuPont's Washington Works plant south of Parkersburg to make Teflon, other nonstick products, oil-resistant paper packaging and stain-resistant textiles.

A limit of 0.2 parts per billion of C8 would have been twice as stringent as a C8 drinking water health advisory issued by the EPA in late January.

And the proposed limit would have been much stronger than the number the EPA eventually agreed to - 0.5 parts per billion - when it revised the Wood County consent order with DuPont in November 2006.

The EPA would not answer detailed questions about the internal memo, or provide a full explanation of how it went from the more stringent number to the limits it later issued.

Mark Pollins, the EPA's water enforcement division director, would not explain how his agency went from the 0.2 part-per-billion number to the 0.5 figure included in the updated deal with DuPont.

"I can't disclose anything," Pollins said during a Jan. 23 interview, "but in these discussions, there is always a wide range of assumptions that go into it."

Pollins said the EPA memo was "part of the iterative process and discussions that go on within the agency.

"I wouldn't call it a final recommendation or anything like that," Pollins said. "What you saw in that memo was a small piece of that conversation."

Pollins said the EPA memo was exempt from the Freedom of Information act, and agency officials "probably made a mistake in releasing it."

The EPA turned the memo over to lawyers for DuPont plant neighbors in response to a Freedom of Information Act request. Those lawyers provided a copy to the Gazette.

After the Gazette started asking questions about the memo, EPA attorneys demanded that the residents' lawyers return "all copies" of the document.

The EPA had touted its revision of the C8 consent order as a tough measure "to protect drinking water near the DuPont Washington Works."

 Read the EPA memo. 

CHARLESTON, W.Va. -- Federal government scientists originally sought a much tougher standard for the toxic chemical C8 than was included in a nationwide health advisory issued last month, according to an internal U.S. Environmental Protection Agency memo.

More than three years ago, EPA staffers proposed that the agency rewrite a deal with DuPont Co. to mandate a C8 limit of 0.2 parts per billion, according to the memo.

"EPA's intent is to seek consent through negotiations with DuPont," the memo said. "However, the order could be issued unilaterally if agreement cannot be reached."

EPA staffers wrote the undated memo, called a "briefing paper," sometime in March 2006. At the time, the EPA was preparing to negotiate with DuPont. Agency officials wanted to tighten a four-year-old deal that required the company to provide alternative water for Wood County residents whose drinking water contained higher levels of C8.

C8 is a common name for perfluorooctanoate acid, or PFOA. The chemical is a processing agent used for years at DuPont's Washington Works plant south of Parkersburg to make Teflon, other nonstick products, oil-resistant paper packaging and stain-resistant textiles.

A limit of 0.2 parts per billion of C8 would have been twice as stringent as a C8 drinking water health advisory issued by the EPA in late January.

And the proposed limit would have been much stronger than the number the EPA eventually agreed to - 0.5 parts per billion - when it revised the Wood County consent order with DuPont in November 2006.

The EPA would not answer detailed questions about the internal memo, or provide a full explanation of how it went from the more stringent number to the limits it later issued.

Mark Pollins, the EPA's water enforcement division director, would not explain how his agency went from the 0.2 part-per-billion number to the 0.5 figure included in the updated deal with DuPont.

"I can't disclose anything," Pollins said during a Jan. 23 interview, "but in these discussions, there is always a wide range of assumptions that go into it."

Pollins said the EPA memo was "part of the iterative process and discussions that go on within the agency.

"I wouldn't call it a final recommendation or anything like that," Pollins said. "What you saw in that memo was a small piece of that conversation."

Pollins said the EPA memo was exempt from the Freedom of Information act, and agency officials "probably made a mistake in releasing it."

The EPA turned the memo over to lawyers for DuPont plant neighbors in response to a Freedom of Information Act request. Those lawyers provided a copy to the Gazette.

After the Gazette started asking questions about the memo, EPA attorneys demanded that the residents' lawyers return "all copies" of the document.

The EPA had touted its revision of the C8 consent order as a tough measure "to protect drinking water near the DuPont Washington Works."

The 0.5-part-per-billion figure replaced a 150-ppb limit set up by the EPA and the state Department of Environmental Protection in 2002. DuPont had agreed to provide bottled water or install treatment systems for anyone with water above that trigger amount.

Then-EPA regional administrator Don Welsh said at the time the change was motivated at least in part by preliminary results of area health screenings. Those tests found mid-Ohio Valley residents had C8 levels in their blood far above the national average.

Around the world, researchers are finding that people have C8 and other perfluorochemicals, or PFCs, in their blood at low levels. People can be exposed by drinking contaminated products, eating tainted food, or through food packaging and stain-proof agents on furniture or carpet. Evidence is mounting about the chemical's dangerous effects, but regulators have yet to set a binding federal limit for emissions or human exposure.

In a Nov. 17, 2006, memo that accompanied the revised consent order, EPA senior toxicologist Christopher P. Weis noted that animal studies demonstrated C8's "potential for developmental toxicology, including mortality."

"We feel that it is particularly important to address ongoing exposure to the residential population, especially pregnant women, [newborns] and the elderly by identification and elimination of multiple pathways associated with C8 in water distribution systems," Weis wrote. "This is a precautionary recommendation to reduce exposure to the population living in the vicinity of the Washington Works facility."

However, the internal EPA "briefing paper" says Weis had actually concluded that the agency should seek a C8 limit of 0.2 parts per billion.

"There are no written briefing materials on the science supporting the 0.20 ppb proposed level at this time," the briefing paper adds.

In December, EPA regional drinking water chief Karen Johnson said the briefing paper was merely "a draft document."

"There were a lot of adjustments to the calculations, and [we] added data when we got more information," Johnson said in a phone interview.

Later, in an e-mail message edited by EPA lawyers, Johnson attributed the change in C8 numbers as "the result of a calculation error, which was corrected by the time the order was issued in November 2006."

"In March 2006, we were still getting our number set before we approached DuPont because it had to be a defensible number," Johnson wrote. "You'd have to confirm with Chris Weis, our toxicologist, but at some point he realized that the water consumption value in the calculation was not correct for the child's weight factor we were using ... when using the correct consumption value, it changed our calculation from around 0.2 to around 0.50 as the position we would take with DuPont."

EPA officials have not responded to a request for an interview in which Weis would explain the calculation error.

Last week, EPA spokesman Dave Ryan in Washington, D.C., sent the newspaper a "final statement" on the issue.

"EPA set a site-specific action limit of 0.50 parts per billion for PFOA in drinking water in the 2006 enforcement order with DuPont," Ryan said. "This limit replaces the higher limit of 150 ppb, which had been established in the 2002 enforcement order with DuPont.

"EPA set the 0.50 limit based on toxicological studies and on results of environmental sampling and monitoring conducted near the facility," Ryan said.

"In early January, EPA's Office of Water issued a provisional health advisory of 0.4 ppb for PFOA. The 2006 order allows for possible modifications based on new science. EPA will review the 2006 order and determine next steps as appropriate."

Reach Ken Ward Jr. at kw...@wvgazette.com or 304-348-1702.