August 2001
Volume 43 Number 8

Filtration: Activated Carbon Remedy for Private Well MTBE Contamination
by Mark Hawkins & Gina Hess   Pages: 

Photographs, figures and/or graphics that may illustrate this article are visible in the printed or .pdf file version of the article only. For the link to the .pdf file, go to the end of the article. For a printed copy, please make a request at Be sure to include the article title, author(s) name(s), the issue, your name and fax number or full address in the email.

Summary: Due to a leak at a gas station, two Pennsylvania communities had MTBE released into their drinking water in nearby wells. The station hired a private firm to assist in implementing a solution, and it chose activated carbon. Here’s why.

The most widely used oxygenated gasoline additive, methyl tertiary butyl ether (MTBE), has made itself known as a way to combat air pollution as a fuel oxygenate that helps gas burn cleaner and, more recently, causing a major water-pollution problem. This problem is exacerbated because MTBE is highly soluble, moves quickly, is slow to decompose and a small dose can contaminate a large amount of water with a foul taste and smell—not to mention suspected health effects. This article highlights use of activated carbon by an environmental consulting firm hired to remediate the MTBE contamination of private wells in two Pennsylvania communities.

The U.S. Environmental Protection Agency (USEPA) hasn’t established drinking water standards for MTBE, which is under review for such a federal maximum contaminant level (MCL). In December 1997, the USEPA issued a "Drinking Water Advisory" for MTBE of 20 to 40 parts per billion (ppb), or micrograms per liter (µg/L), primarily for taste and odor considerations. The taste threshold is generally accepted to be 5 ppb.

In May 2000, California—which has led the nation in regulation of MTBE—set a primary MCL for MTBE at 13 ppb. California is scheduled to discontinue use of MTBE as a fuel oxygenate by Dec. 31, 2002, according to an executive order from Gov. Gray Davis based on a University of California at Davis study that found MTBE poses an environmental threat to groundwater and drinking water. In June 2001, a state waiver from the federal minimum oxygen requirement in gasoline was denied.

In Pennsylvania, where the case study in this article is based, the Department of Environmental Protection has set 20 ppb as the maximum allowable limit in water, with higher levels considered unsafe for drinking.

Extent of contamination

While the federal government and states debate what regulation should be enacted, large and small cities have had to deal with the immediate task of cleaning up water supplies. In April 2000, a few residents of Bedminster and Hilltown, both Pennsylvania communities 35 miles north of Philadelphia, were informed that MTBE had seeped into their wells. No other contaminants were observed above their practical quantitative limits. In October 1999, the company that owned a gas station nearby had discovered the site was contaminated with MTBE at a level nearly 1,000 times greater than the 20 ppb maximum level recommended by the Pennsylvania Department of Environmental Protection. As a result, the gas station hired a firm to investigate the problem and propose a solution. The firm tested residential wells within a roughly 4,000-foot radius of the station and found 14 of 80 wells contaminated with MTBE levels greater than 20 ppb; the most severe case contained 1,400 ppb. After testing monthly for four months, it was apparent that MTBE concentration varied greatly even within the same well.

After considering other options including extension of two public water mains, the firm chose to install activated carbon point-of-entry (POE) filters for each of the 14 contaminated wells in question, as well as in nine additional homes. Two vessels charged with a coconut shell-based activated carbon designed specifically for removing MTBE—or an MTBE-specific carbon—were installed in each home in a series configuration. Each POE adsorber had a 180 pound (lb.) granular activated carbon (GAC) capacity with a maximum 15 gallons per minute (gpm) flow rate. Every three months, samples were collected directly following the first adsorber and tested. After a year, none of the carbon beds has shown MTBE breakthrough.

Behind the success

The success of the POE filters lies in the unique characteristics of the activated carbon. MTBE-specific carbon is a coconut-shell carbon with both a high capacity for difficult-to-adsorb organics and high MTBE retentivity. Both traits are important because high capacity enables the carbon to remove a large overall amount of MTBE, while high retentivity ensures that the MTBE doesn’t de-sorb when the influent concentration dips or when other contaminants contact the carbon. (This proved to be a crucial feature because of influent MTBE concentration fluctuations.)

Both the large capacity and high retentivity of MTBE-specific carbon can be traced back to the fact that coconut shell activated carbon has a greater number of high-energy pores (micropores) than coal, wood or peat carbons. In order for any contaminant to be adsorbed, the forces exerted by the carbon pore walls must overcome the forces of solvation. This is why MTBE isn’t easily adsorbed by traditional activated carbons—most carbons cannot exert the necessary force to pull MTBE out of solution. To make MTBE-specific carbon, coconut shell from a consistent source is activated under specialized conditions. This yields carbon with a high volume of micropores smaller than 6 angstroms, and an optimized apparent density.

In general, the results of lab filter testing—using NSF Standard 53’s protocol—showed coconut shell carbon to have a higher capacity for MTBE than coal carbon (see Chart 1). Chart 2 displays the difference in MTBE retentivity between MTBE-specific coconut shell carbon and a high-quality coal-based carbon. A column of each carbon was equilibrated with a 500-ppb MTBE solution. After equilibrium was reached, de-ionized water passed through the column, and effluent samples were collected and analyzed by gas chromatography and mass spectrometry (GC-MS). Not all coconut shell carbons perform equally in removing MTBE. Chart 3 shows the effect that raw material source and activation conditions can have on MTBE capacity, even between different coconut shell carbons.

The firm decided to use MTBE-specific carbon for POE adsorbers, not only for its MTBE capacity, but also because of low dust levels and metals leaching associated with coconut carbons.


High levels of MTBE contamination in private wells have created great concern among residents of Hilltown and Bedminster, Pa. The owners of the gas station suspected of leaking the MTBE hired the firm to determine the extent of contamination and recommend a solution. As many as 14 wells were tainted with levels above Pennsylvania’s 20 ppb limit with one site reaching 1,200 ppb. The firm opted to install vessels outfitted with MTBE-specific coconut shell activated carbon. After one year, the systems are still delivering water with MTBE levels below the practical quantitative limit to the homes.

About the authors

Mark Hawkins is a project scientist for Harding ESE Inc., a subsidiary of MACTEC Inc., of Plymouth Meeting, Pa. Hawkins holds a bachelor’s degree in chemical geology from Juniata College. He can be reached at (610) 941-9700, (610) 941-9707 (fax) or email:

Gina Hess is the customer support chemist for Barnebey Sutcliffe, a division of Columbus, Ohio-based Waterlink. Its CNS-MTBE carbon is used in the Protect FX-7 filter vessels discussed in the above article. Hess holds a master’s degree in chemistry from Ohio State University. She can be reached at

Click here to view additional information: hawkins.pdf

To view in PDF format, you will need the FREE Adobe Acrobat Reader.