UNSEEN SCIENCE: 3M’s PFAS Ethical Debacle, Part I

A lighthouse hidden by fog (Duluth, MN). Photo by C. Bohacek

In the 2018 peer-reviewed article, “Nonstick Science: Sixty years of research and (in) action on fluorinated compounds,” Richter et al define UNSEEN SCIENCE as “research that is conducted but never shared outside of institutional boundaries.”  Although the article focuses primarily on Dupont, a review of 3M documents made public on the Attorney General of Minnesota’s website, demonstrates many ethical lapses at 3M as well.  If you’d like to review the primary data cited here, go to the website and check out the document numbers I’ve provided in [brackets] below.  Within quotation marks, words in parenthesis with italics have been added by me for clarity.

In 1975, academic researchers Guy & Taves contacted 3M in an effort to pinpoint the identity and source of organic fluorine compounds in the blood of the general population of the US.  The researchers had tentatively identified the organic fluorine compound as PFOA, but noted that the match wasnt perfect and asked for 3M’s help in refining their identification [2771, 1118, 1121].  Scientists and managers at 3M agreed to collaborate with Guy & Taves and then….didnt.  In 1975, several weeks after conferring with Guy & Taves, 3M matched the researchers unknown organic fluorochemical spectra to that of PFOS, a compound produced solely by 3M at that time.  The PFOS identification work was shared internally at 3M but WAS NOT shared with the Guy & Taves, despite an ongoing relationship and continued interest in the topic [1123, 1144, 2534 ].

3M may argue that they weren’t 100% sure that the PFOS identification was a perfect fit for the compound identified by Guy & Taves and that’s why they didnt share their insights, but subsequent actions initiated within 3M belie this defense.   Soon after confirming the widespread, organic fluorine compound was PFOS, 3M initiated a study in rats characterizing the absorption, metabolism and excretion for a fluorochemical compound called FC807, the main building block of 3M’s Scotchban chemistry that was used to coat food packing materials.  In 1977, this study demonstrated that FC807 was absorbed by the mice and that the main metabolite, which bioaccumulated, was PFOS [1133].  Additionally, 3M conducted metabolic studies of a fluorochemical compound called “Ethyl FOSE alcohol,” another component of consumer products, including Scotchban [1166].  As in the study of FC807, the main Ethyl FOSE alcohol metabolite found to accumulate in the mice was PFOS.

Had 3M truly believed that the compound identified by Guy & Taves was PFOA, there would have been no reason to conduct a study on the metabolism of FC807 or Ethyl FOSE alcohol, both present in consumer products.  These studies were also inconsistent with the notion that 3M was trying to understand exposure to their workers: FC807 and Ethyl FOSE alcohol are building blocks of consumer products.  In the words of one 3M scientist, the data from these two studies, collectively “implies that any 3M product bearing the C8F17SO2NCH2- group (e.g. Scotchgard and Scotchban)…would generate C8F17SO3- (PFOS) which accumulates in the animal’s blood and tissue”.  In 1978, less than 30 months after 3M correctly identified PFOS as the compound in the blood of the general population, 3M toxicologists agreed that “FC-95 (PFOS)...and FM-3422 (Ethyl FOSE alcohol) should be regarded as toxic [1166, 1172, 1174].”

Despite direct inquiries, none of these data were shared outside of 3M for decades.

*Unseen science* was even practiced within the corporation.  In 1979, 3M sought PFAS toxicology advice from a well known toxicology expert outside of 3M.  Following that meeting, the toxicologist specifically asked that his note be added to the minutes of the meeting:  “The study of levels of FC-807 or its metabolites is of utmost importance in determining possible future problems. It should be determined if FC-807 or its metabolites are present in man, what level they are present, and the degree of persistence (half- life) of these materials. If the levels are high and widespread and the half-life is long, we could have a serious problem (underline added) [1204].”

When the final meeting minutes were published 6 weeks later, the underlined portion of that ominous warning from the expert had been struck from the record [1210].

It’s clear from 3M’s own documents that 1) PFOS had been identified as the most likely component of widespread organic fluorine contamination in the general population in the 1970s; 2) 3M knew that PFOS bioaccumulated; and 3) 3M knew PFOS was toxic.  The documents summarized here also indicate that 3M had a strong suspicion that the source of PFOS in the general population was from Scotchban food packaging.  

These examples meet the definition of Unseen Science and contributed to 3M’s ability to continue to manufacture PFAS compounds between 1975 and 2000.