Ashta Chemicals, LLC, in Ashtabula, OH,  one of four remaining mercury-based chlor-alkali plants in the United States, is the source of mercury measured in air surrounding the plant, yet the company claims to be releasing no mercury.  Tests conducted in June 2009 by Oceana scientists directly contradict Ashta Chemicals' reported "zero mercury emissions," which appear to be the result of creative accounting by the company.  While Ashta claims to have installed some of the most stringent mercury pollution controls in the industry, Oceana's results demonstrate that shifting to mercury-free technology is the only way to eliminate mercury releases. 

Oceana's air monitoring results are described in a white paper, Ashta Chemicals' Zero Mercury Emission Claim, Refuted by Air Monitoring, which is released today.  Oceana suspects the discrepancy in mercury emissions identified by its measurements is due to the company's inappropriate use of mathematical estimation methods. Key findings from Oceana's air testing include:

• The Ashta plant is emitting toxic mercury into the air, demonstrating that its "mass balance" accounting reports are incorrect, and that the plant's pollution control system fails to eliminate emissions.
• During two days of testing air mercury levels, Oceana detected levels up to 35 times higher than background levels immediately downwind of the Ashta chlor-alkali plant, while concentrations upwind of the plant were similar to background levels. This demonstrates that the plant releases mercury and its emissions are far greater than zero, contrary to what the company has reported.
• The maximum mercury level detected downwind of the Ashta plant (105 ng/m3) was similar to the maximum mercury level detected around another chlor-alkali plant (103 ng/m3) with less stringent mercury controls, which reported 500 pounds of mercury emissions. This demonstrates that the technology purported to achieve zero emissions does not reduce the plant's emissions below those of plants with less stringent controls.

"We had high hopes that Ashta's strict pollution controls would work but these results demonstrate that the only reliable way to eliminate mercury pollution from these chlor-alkali plants is to end the use of mercury cell technology," said report co-author Dr. Kimberly Warner. "It is clear that their creative accounting practices did not make the mercury disappear," she added.

To achieve zero mercury emissions at the Ashta and other U.S. chlor-alkali plants, Oceana recommends phasing out mercury use. Fortunately, alternative mercury-free technology is available and being used by 95 percent of the chlorine industry.  A bill that would require the remaining four plants to upgrade to modern technology, the Mercury Pollution Reduction Act (H.R. 2190, S. 1428), is currently being considered by Congress.
In the interim, Oceana recommends the following:

• EPA should require that all mercury-based chlor-alkali plants monitor, quantify and report all mercury air emissions.
• EPA and the states should more closely scrutinize mass balance accounting methods, which can produce unreliable measurements of mercury use and releases.

The "mass balance" method of accounting is based on the principle that mass can neither be created nor destroyed, and therefore the company should be able to account for all of the mercury that goes into the process. This amount of mercury should be equal to the amount leaving the process or staying in the plant, since no mercury is consumed during the process of manufacturing chlorine.  Historically, mercury-based chlorine factories lost tons more mercury than they could account for annually. But recently, Ashta Chemicals has used this method to report unexplained mercury gains instead of losses. In fact, through this method, Ashta has reported gaining more than a ton of mercury above and beyond what it brought into the plant. These large unaccounted for gains were the apparent basis for the company's report of zero mercury emissions, and astonishingly, neither the federal EPA nor the Ohio EPA has addressed this error.