November 11, 2015 Articles

Emhart v. United States: Another Failed Attempt to Prove Divisibility under CERCLA

Add the Centredale Manor Restoration Project in North Providence, Rhode Island, to the list of sites where a court has denied a CERCLA divisibility defense.

James P. Ray and Megan E. Baroni – November 11, 2015

Add the Centredale Manor Restoration Project in North Providence, Rhode Island, to the list of sites where a court has denied a CERCLA divisibility defense. On September 17, 2015, Judge William Smith of the Rhode Island District Court, in a 186-page opinion, found Emhart Industries jointly and severally liable for past and future response costs at the Centredale Manor site.

The Centredale Manor Restoration Project Superfund Site

The Centredale Manor Restoration Project Superfund Site consists of a peninsula of about nine acres in the Woonasquatucket River in North Providence, Rhode Island, as well as portions of the river and ponds downstream of the peninsula. In the 1950s and 1960s, the peninsula was the home of Metro-Atlantic Chemical Company, Emhart’s predecessor, which operated a chemical manufacturing facility on a portion of the peninsula, and New England Container Company (NECC), a barrel reconditioning company. In the late 1960s and early 1970s, Metro-Atlantic and NECC relocated, and ultimately two apartment buildings were constructed on the site.

In the 1990s, the government found dioxins in fish in the river. The U.S. Environmental Protection Agency (EPA) subsequently found dioxins, polychlorinated biphenyls (PCBs), and other contaminants in river and pond sediments, floodplain soils, and soils at the peninsula. The EPA concluded that the dioxins and other contaminants came from the chemical manufacturing and barrel reclamation operations at the site.

The EPA placed the site on the National Priorities List in 2000. In the early 2000s, Emhart and NECC performed several removal actions at the site, including placing soil caps over dioxin-contaminated areas and removing dioxin-contaminated soil and sediments from residential properties along the river. In 2009, Emhart conducted a removal action to address the discharge of contaminated groundwater to the river. Emhart excavated contaminated soil in the area of one of the historic chemical manufacturing buildings and installed a Resource Conservation and Recovery Act (RCRA) cap in that area. In 2012, the EPA issued a record of decision, which contains a cost estimate of $104.6 million for the final remedy. The final remedy will include waste removal, placement of a RCRA cap on the remaining portion of the peninsula, dredging a substantial portion of the river and downstream ponds, construction of an upland confined disposal facility, and long-term maintenance and monitoring. Total costs, including past costs, will likely approach or exceed $150 million.


The Litigation 

Emhart sued NECC and NECC’s insurers under the Comprehensive Environmental Response, Conservation, and Liability Act (CERCLA) in 2006 after NECC sold its remaining assets. Emhart initially claimed that dioxin, the principal driver at the site, was from incineration operations associated with NECC’s barrel reclamation process. What followed was nearly 10 years of litigation involving over 30 experts who generated nearly a hundred initial, supplemental, and rebuttal expert reports. The court engaged the services of a technical advisor to assist the court with the “extraordinary line-up of scientific and technical experts.”

In 2011, Emhart commenced a separate action against the United States and several branches of the military, this time claiming that the dioxin came from residual tactical military herbicides, including Agent Orange, found in empty drums that several military installations sent to NECC. The court stated that this theory was the “centerpiece of Emhart’s argument” prior to trial, but at trial, “this theory seemed to fade to black.”

The actions against NECC and the government were ultimately consolidated. The United States, on behalf of the EPA, then asserted claims against Emhart and NECC, seeking past costs and to compel a cleanup at the site. The parties asserted numerous counterclaims and cross-claims, and NECC and the United States brought third-party actions against a number of NECC’s former customers. The court entered a phased case-management order under which, in the first phase, the court would determine the liability of Emhart and NECC, including divisibility, and the proper allocation between Emhart and NECC if they were both found jointly and severally liable. Subsequent phases would deal with costs, remedy challenges, and the contribution claims against the NECC customers.

In early 2015, all parties participated in a mediation, and NECC reached an “ability to pay” settlement with the EPA, just before the Phase I trial. Emhart and the government proceeded to trial on the issue of Emhart’s liability and its divisibility defense. The court presided over a 20-day trial, after which it found Emhart jointly and severally liable.

Emhart’s Liability

The government alleged that Metro-Atlantic’s manufacture of hexachlorophene was the primary source of dioxin, in particular the highly toxic dioxin congener 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD), found on the peninsula and in downstream water bodies and floodplain sediments. Metro-Atlantic manufactured hexachlorophene on the western side of the peninsula adjacent to the river for a period of time in the 1960s. To manufacture hexachlorophene, Metro-Atlantic used trichlorophenol contaminated with 2,3,7,8-TCDD. (Metro-Atlantic obtained the trichlorophenol from Diamond Alkali in New Jersey, the alleged source of much of the dioxin at the Passaic River Superfund Site.)

Emhart argued that its evaluation of the site environmental data showed that Metro-Atlantic’s hexachlorophene operation could not have been the source of the dioxin contamination. The site samples exhibited dioxin and furan congener patterns that were heavily dominated by 2,3,7,8-TCDD compared with other dioxins and furans. Emhart argued that Metro-Atlantic purchased “crude” trichlorophenol from Diamond Alkali and that “crude” trichlorophenol would have contained a multitude of dioxin and furan congeners. Emhart argued the contamination must have come from a “purified” trichlorophenol containing only the 2,3,7,8-TCDD congener with little or no other dioxin and furan congeners.

Emhart conducted extensive discovery of most of NECC’s customers in an attempt to find a potential source of “purified” trichlorophenol. Among these customers were two military bases that sent empty drums to NECC. Ultimately, Emhart’s experts opined that those drums must have contained tactical herbicides or other chemicals made from “purified” trichlorophenol. At trial, Emhart seemingly abandoned its theory that the military was the likely source of the “purified” trichlorophenol, though it could not identify another source of “purified” trichlorophenol.

The court reviewed a voluminous factual record that included stipulated deposition and trial testimony from former Metro-Atlantic and NECC employees, neighbors, contractors, customers, and government personnel. Numerous expert witnesses also testified about the chemical and mechanical processes used by both industrial operations, the contamination found at and in the area of the former operations on the peninsula, the fate and transport of the contaminants, and chemical fingerprinting techniques used to identify the source or sources of those contaminants.

After thoroughly reviewing the evidence, the court found that Metro-Atlantic generated numerous solid and liquid waste streams in the hexachlorophene manufacturing operation and other operations. The court found the evidence was sufficient to conclude that those waste streams, contaminated with 2,3,7,8-TCDD, were released to the river, the ground, and in a waste disposal area on the peninsula. The court also found that hexachloroxanthene found at the site was a reliable marker of releases from the hexachlorophene process.

The court rejected Emhart’s theory that the dioxin and furan congener patterns at the site were inconsistent with a “crude” trichlorophenol source. The court held that (1) the composition of the “crude” trichlorophenol from Diamond Alkali was not known, and (2) the evidence was not sufficient to prove that site soil samples could not have been contaminated by the trichlorophenol from Diamond Alkali. And, despite the fact that Emhart did not raise it at trial, the court also rejected Emhart’s theories that barrels from the military bases could have been the source of the 2,3,7,8-TCDD contamination at the site.


Like most CERCLA potentially responsible parties (PRPs), Emhart attempted to limit its exposure by arguing that even if it was found liable under CERCLA, it should not be found jointly and severally liable because the harm is divisible. In support of its divisibility argument, Emhart advanced four separate theories. First, Emhart argued that it should have a 0 percent share of the remaining harm at the site because any incidental releases of 2,3,7,8-TCDD from Metro-Atlantic’s manufacturing operations would have been in the area of the hexachlorophene building and would have been remediated during the 2009 removal action. Second, Emhart argued that even if the 2,3,7,8-TCDD contamination in the vicinity of the hexachlorophene building migrated from that area, Emhart’s divisible share of the harm should still be 0 percent because any amount it spent on the 2009 removal action in relation to the total cost of the proposed remedy exceeds its volumetric contribution of 2,3,7,8-TCDD to the site. Third, Emhart argued that if liquid waste from the hexachlorophene building was discharged to the river, harm can be apportioned because the liquid waste contained a known quantity of 2,3,7,8-TCDD. Finally, Emhart argued that even if the court could not apportion by volume, other considerations—such as geography, time, and the type of contaminant—could provide a rational basis for divisibility.

The court disagreed with all of Emhart’s theories. Stating that Emhart’s divisibility arguments rely on “the triumph of hope over reason,” the court just did not agree with Emhart’s rendition of the facts surrounding Metro-Atlantic’s operations. The court disagreed that Metro-Atlantic’s releases at the site were as limited as Emhart suggested. For example, the court disagreed that all of the solid waste (e.g., filter cakes) from Metro-Atlantic’s hexachlorophene operations were disposed off-site. The court also disagreed that any spills of trichlorophenol would have been limited to the vicinity of the hexachlorophene building. Because many of Emhart’s theories of divisibility relied on it successfully proving these facts, its failure to do so was fatal to its ability to prove that the harm was capable of apportionment.

The court also criticized Emhart’s expert’s opinions on the total amount of 2,3,7,8-TCDD at the site. One of Emhart’s experts attempted to approximate the amount of 2,3,7,8-TCDD that could have been released through leaks and spills on the peninsula and discharges of liquid waste to the Woonasquatucket River. But this expert analysis was plagued by the same factual assumptions that plagued Emhart’s divisibility theories at trial. The expert assumed that certain Metro-Atlantic wastes were not disposed of on-site; therefore, he did not consider them in his calculations. For this reason, the court found the estimates so uncertain as to be unusable to apportion liability.

With volumetric and geographic divisibility on the ropes, Emhart tried to argue that the harm was divisible by the type of contaminant. Specifically, Emhart argued that there were six contaminants of concern at the site, and because Metro-Atlantic was associated with only two of them—2,3,7,8-TCDD and perchloroethylene—it should only be 33 percent liable. The court found this “‘calculation’” to be “wholly arbitrary.” The court noted that there is no reason why all of the constituents should be treated equally, especially because 2,3,7,8-TCDD is the most toxic. The court further noted that Emhart was unable to point to a case that similarly apportioned the harm by type of contaminant when there was no reasonable basis to apportion by geography or volume.

Finally, Emhart argued that the harm was divisible by time. This argument was based on what Emhart argued was a limited time period over which Metro-Atlantic manufactured hexachlorophene (alleged to be less than one year). First, the court stated that the duration of the hexachlorophene operation was unclear. But, more important, regardless of the time period for the operation, it seemed likely that Metro-Atlantic was responsible for a disproportionate volume of the 2,3,7,8-TCDD.

The court noted that NECC was likely responsible for hazardous substances at the site. But, because this was a divisibility analysis, the court could not evaluate equitable considerations that are typically part of a claim for contribution. Given the evidentiary record, any determination of divisibility would have been arbitrary and therefore not appropriate.


Ultimately, every divisibility argument hinges on the facts of a particular case. Site conditions, the mix of past and current operations, the reliability and completeness of the historical record, and expert analysis all play a significant role in a PRP’s success or failure in proving divisibility. But when parties disagree so vastly as to the facts that led to the contamination, divisibility can become impossible to prove. A PRP in this type of case develops its theories of divisibility based on its own interpretation of the facts. But it has to prove at trial that its interpretation is correct. If it cannot, the PRP risks creating a house of cards that, when collapsed, it cannot recover. Without these facts, the PRP will be hard pressed to prove that there is a reasonable basis for apportionment under CERCLA.

Keywords: environmental litigation, liability, divisibility, apportionment


James P. Ray is the director of the Career Planning Center at the University of Connecticut School of Law. Megan Baroni is counsel with Robinson & Cole, LLP, in Stamford, Connecticut. They represented NECC in the Centredale Manor litigation, with their colleague John Peloso.

Copyright © 2015, American Bar Association. All rights reserved. This information or any portion thereof may not be copied or disseminated in any form or by any means or downloaded or stored in an electronic database or retrieval system without the express written consent of the American Bar Association. The views expressed in this article are those of the author(s) and do not necessarily reflect the positions or policies of the American Bar Association, the Section of Litigation, this committee, or the employer(s) of the author(s).