One of the energy stories I've followed with great interest all year concerns efforts to increase the proportion of ethanol blended into ordinary US gasoline. This began last year when Growth Energy, an ethanol trade association, asked the Environmental Protection Agency for a waiver to increase the allowed percentage of ethanol in gasoline from 10% to 15%. In October EPA issued a partial approval of the request, but only for vehicles built in model year 2007 or later. However, a new report from Underwriters Laboratories (UL) indirectly casts doubt, not only EPA's ruling, but on whether the agency was assessing all the relevant issues.

I ran across the UL report on the compatibility of mid-level ethanol/gasoline blends in gasoline dispensing equipment--the pumps, hoses and tanks in gas stations--in a posting on API's EnergyTomorrow blog. It cited the UL study, which had been commissioned by the Department of Energy, as evidence that E15, the 15% blend of ethanol and gasoline that the EPA just approved for use in newer cars, could result in serious failures of gas pumps. Yet when I read the report, I immediately encountered its innocuous-sounding conclusion stating, "The overall results of the program were not conclusive insofar as no clear trends in the overall performance of all equipment could be established." It went on to say that the equipment "generally performed well." If I had stopped reading there, I'd have concluded that API was blowing the whole story out of proportion.

When I read the data included in the report, however, a different story emerged. Of the new and used gasoline dispensers and associated equipment tested, very few exhibited no problems on the 17% ethanol test fuel used. In fact, in UL's long-term exposure test, many hoses, nozzles and swivels leaked. 100% of the meter, manifold and valve assemblies tested leaked or failed to shut off. Perhaps most worryingly, two-thirds of the breakaway couplings tested leaked, failed their pressure tests, or required more than the recommended pull to separate. (A breakaway is designed to pop the hose off the dispenser when a customer forgets to remove the nozzle from his car's gas intake and attempts to drive off. This happens a surprising number of times a year, and before the deployment of breakaways such incidents imposed significant repair costs on dealers, even when the resulting spills didn't cause fires.)

The common denominator in these failures was what the report refers to as "nonmetals", gaskets, seals and parts made from various polymers. From that I would draw two conclusions: First, it ought to be possible to design new dispensers and retrofit existing dispensers with new gaskets, seals and plastic parts designed to withstand higher concentrations of ethanol, just as the fuel systems in flexible fuel vehicles are designed to tolerate blends of up to 85% ethanol. However, considering that the US has between 90,000 and 160,000 gas stations, depending how you count them, the number of dispensers that would have to be modified is at least in the high tens of thousands, if not well into the hundreds of thousands. To my knowledge the ethanol industry has not offered to defray the cost of these conversions for a retail fuel industry that operates with extremely lean margins. Nor is it obvious that dealers would qualify for federal assistance, as they do when they add E85 capability.

My second conclusion--really more of a suspicion--has nothing to do with gas pumps or gas stations, and everything to do with cars. After reading the UL report I went back and reread portions of the EPA's official waiver response, which ran to 58 pages in the Federal Register. From what I can tell, EPA wasn't really looking at whether cars would suffer damage from operating on a higher percentage of ethanol than the fuel for which they were designed. The waiver was granted on the basis of those cars not emitting more pollutants than on the fuel for which they were designed. Quoting from the EPA document:

"For MY 2007 and newer light-duty motor vehicles, the DOE Catalyst Study and other information before EPA adequately demonstrates that the impact of E15 on overall emissions, including both immediate and durability related emissions, will not cause or contribute to violations of the emissions standards for these motor vehicles. Likewise, the data and information adequately show that E15 will not lead to violations of the evaporative emissions standards, so long as the fuel does not exceed a Reid Vapor Pressure (RVP) of 9.0 psi in the summertime control season. The information on materials compatibility and drivability also supports this conclusion."

That's good as far as it goes, but from my perspective this finding reflects a necessary but hardly sufficient standard for putting a new fuel into the marketplace, particularly when the failures of the dispensers in the UL study point to the possibility of similar failures of "nonmetals" in the fuel systems of cars or other devices not designed to run on more than 10% ethanol. Even if the leaks found in the testing of product dispensers didn't result in safety hazards, they would at a minimum increase the evaporative emissions from infrastructure, aside from the automotive impact on which EPA apparently focused. I also find it interesting that a bill was introduced in Congress this summer, as EPA was considering the waiver request, that would appear to make it more difficult for consumers to recover the cost of damages resulting from compatibility problems in approved vehicles or misfueling of non-approved vehicles.

As I've noted in my previous postings on this topic, I'm sympathetic to the box into which altered circumstances have placed both the ethanol industry and the federal government with regard to ethanol blending. US gasoline sales, which stagnated after the financial crisis and are only growing by a historically modest 0.7% this year (through November) according to API's latest statistics, are not expanding fast enough to accommodate the output of all the ethanol plants that have been built or are under now construction. When the Renewable Fuels Standard was enacted as part of the Energy Independence and Security Act of 2007, the bill's architects presumably expected that E85 sales would take up any slack. The fact that that hasn't happened does not justify creating a new outlet for additional ethanol in automobiles not designed to accommodate it, any more than it would justify running a new fuel through infrastructure that has been shown not to be up to the challenge. If EPA doesn't revisit the more comprehensive aspects of this question as part of its deferred decision on allowing E15 for cars made before 2007, then perhaps it's time for another government agency with a broader charter to take over this issue.