"Looking Into the Fire," Feature Article, April 2007

looking into
the fire

Is biodiesel a potential smog source? Models of engine combustion aim to assure that it doesn't become one.

by Jean Thilmany, Associate Editor

biodiesel has supporters on two different fronts. There are those who point out that blending animal or vegetable oil with diesel fuel can reduce the United States' reliance on petroleum imports. After all, biofuels can be grown in the heartland.

Proponents of reducing carbon dioxide emissions of the world's combustion point to the biodiesel as a desirable option. Burning a 20 percent biodiesel blend produces about 15 percent less CO₂ than burning regular diesel fuel.

So you might say that cooking grease is a candidate to help save us from an uncomfortably warm future and to make Western countries more energy-independent.

A diesel engine at Iowa State helped develop CFD models. They're being adapted for biodiesel using data from two Iowa power plants.

But others say it won't be so easy. Studies done by Oak Ridge National Laboratories in Tennessee, for example, show that biodiesel emissions are rich in hydrocarbons, carbon monoxide, and a mixture of nitrogen monoxide and nitrogen dioxide. In short, the smog-forming content of biodiesel is higher than that of unmodified diesel fuel.

Let it be understood that there isn't exactly universal agreement with the Oak Ridge findings. Amber Thurlo Pearson, a spokesperson for the Biodiesel Board, cites different studies, some carried out by another national lab, the National Renewable Energy Laboratory in Golden, Colo., that have found no conclusive evidence that biodiesel in a 20 percent mix with regular diesel produces more smog gases than unmodified diesel. The Biodiesel Board in Jefferson City, Mo., is the trade association that represents the biodiesel industry.

Still, the Oak Ridge researchers—and other scientists and engineers concerned about biodiesel emissions—say the U.S. can't afford to wait to find out that biodiesel is a smog producer. They're already at work training their computers on the problem of biodiesel exhaust. Specifically, they're building computational fluid dynamics models to give them insight into ways to reduce emissions.

Biodiesel, a blend of animal or vegetable oil and diesel fuel, can be used in place of pure petroleum fuel in unmodified diesel engines. The bio-to-petroleum percentage can vary, and sometimes the petroleum content is omitted and the fuel is 100 percent biologically derived oil.

For widespread adoption, the biodiesel of the future will have to produce less NO_x than it does now, according to Oak Ridge researchers. That's why projects like those at Oak Ridge and at the Iowa Energy Center are looking at ways to cut the NO_x emissions from biodiesel-burning engines that can run anything from automobiles to power plants.

Studies at Oak Ridge and the Iowa Energy Center focus on low-temperature combustion as a means of cutting pollutants. Research led by the National Transportation Research Center in Knoxville, Tenn., and the University of Wisconsin's Engine Research Center in Madison, Wis., has shown that low-temperature combustion can cut emissions from conventional petroleum-fueled diesel engines. If low-temperature combustion reduces diesel NO_x, it should do the same for biodiesel, the reasoning goes.

At the NTRC, an Oak Ridge National Laboratory research partner, scientists are looking at a variety of ways to lower combustion temperature: mainly by making changes to a diesel engine's fuel-injection timing and throttling, and by recirculating and immediately reusing the engine's exhaust gases, which contain water and carbon dioxide that can cool combustion. In order to find the best emissions profile possible, researchers want to know how changing the internal operating factors of these low-temperature combustion engines affects the amount of NO_x released.

Biodiesel research has generated interest because oil from soybean crops, like those above, are used to make the fuel, which often blends vegetable oil and diesel fuel.

This is where CFD comes in. To figure out how to best configure a combustion engine burning biodiesel, scientists have to get a close look at what's happening inside. With combustion simulated, they can run digital simulations to experiment with various timing and injection schemes while running differing mixes of biodiesel and consult the simulation to see emissions results, said Valmor de Almeida, an Oak Ridge researcher.

In other words, they could run a variety of what-if scenarios through the combustion engine burning biodiesel to see which best affects emissions.

The fuel injection systems in diesel engines are tightly controlled environments where the tiniest of many variables might influence engine performance. The key to cleaner-burning engines depends on a better understanding of the interplay of many factors—timing, pressure, flow rate, and temperature among them—that influence fuel mixing and combustion dynamics.

The diesel combustion process is complex and difficult to visualize because a variety of factors affect it, including turbulence, spray characteristics, and air/fuel mixing. Understanding the split-second physics and flash chemistry in a running diesel engine is a feat of computational fluid dynamics that can give researchers a look inside the engine.

Problem is, they don't yet have simulations of biodiesel combustion. Currently, de Almeida is working backward from hard-and-fast information to build the models. The Iowa researchers are informing CFD models built for diesel engines with data returned during a biodiesel experiment at two Iowa power plants.

The NTRC carried out its low-emissions biodiesel studies using a 1.9-liter passenger-car engine donated by General Motors. The diesel engine hasn't been altered. It's testing various blends of soybean oil and diesel fuel.

For de Almeida, his hard-and-fast numbers come from the NTRC, which used data acquisition hardware to gather engine data-pressure and temperature information as the fuels burned, and infrared and mass spectrometers to analyze exhaust gases. He's pairing that data with information about biodiesel combustion chemistry supplied by Joanna McFarlane, a physical chemist at the lab.

After creating a CFD application, de Almeida will compare the results it returns to information about the engine to verify results.

future fuels

The scientists are starting their study with biodiesel, but have plans to look at other alternative fuels. Next up will be fuels derived from liquid gasification. Devising the proper CFD application for the job could prove vital in helping scientists pick the best alternative fuel for every engine, McFarlane said.

"If we can do the calculations for one particular system, we can use the same methodology to find, say, a fuel for a vehicle that's used by the military under extreme conditions," she said. "Or to find the best jet fuel. There really are no limits."

De Almeida said he hopes model creation will also allow researchers to predict how biodiesel's physical properties affect the way it sprays in the engine, another factor that affects combustion. The spray is a difficult physical event to model, he said.

"The fluid is injected at high pressure into the chamber where it breaks up and generates filaments, which coalesce and break up again," de Almeida said. "These flow regimes happen until the fuel is atomized within the cylinder. The whole idea is to get the entire fuel injected and vaporized so it can burn."

Meanwhile in Ames, Iowa, Song-Charng Kong, an Iowa State University assistant professor of mechanical engineering, is at work on a CFD application to simulate the workings of diesel generators that burn biodiesel at power plants. He's using information gleaned during two biodiesel experiments at Iowa plants last fall.

The Iowa Energy Center-sponsored experiment aims to reduce carbon emissions by giving utility plants the capability to burn biodiesel. Iowa, of course, is a natural place for such a study because of its agricultural makeup; farmers and politicians are hungry to sell soybeans and other vegetable crops to the makers of biodiesel. Provide a thriving industry for biodiesel, and rural states can expect to see soybean sales take off.

Kong knows of no biodiesel studies yet done on large diesel generators. "The basic operating principle is the same as smaller engines. So that's common knowledge, but magnitude is much different," he said. "The fuel injection system is a lot different."

For emissions numbers, researchers use data acquisition hardware attached to an engine in the lab.

For three weeks in late September and early October 2006, researchers powered the diesel engines at utility plants in Winterset and Story City, Iowa, with biodiesel of varied formulations. A private company, Comprehensive Emission Services, used data acquisition hardware to profile the engines' workings and emissions makeup.

At the Winterset plant, researchers first monitored emissions from diesel fuel for comparison purposes. They then ran B10 and B20 in two generators: a 1,750-kilowatt Cooper-Bessemer engine used since 1966 and a four-year-old 1,825 kW Caterpillar engine. B10 refers to a mix of 10 percent biodiesel to 90 percent petroleum-derived diesel; B20 is made up of 20 percent biodiesel to 80 percent diesel fuel.

At the Story City site, they also measured diesel fuel emissions at a 27-year-old 2,070 kW Fairbanks Morse engine to provide a baseline measure. They then ran B10, B20, and B100—that is, pure bio-derived fuel—through the engine.

The Energy Center said it believes the B100 test was the first of the kind on an older generator and the first B100 emissions test done in accordance with U.S. Environmental Protection Agency methodology.

"The three engines we chose are quite representative of those used in power plants in general," Kong said.

But pollution concerns still need to be overcome before biodiesel will provide the lamplight. As at Oak Ridge, the Iowa researchers are also studying low-temperature combustion as the most feasible means to reduce NO_x emissions.

Kong is building upon his already existing CFD diesel models when constructing the new alternative-fuel application and is informing the diesel program with the information returned by the biodiesel trials.

"We have a lot of experience in modeling regular diesel engine combustion, but information about biodiesel combustion is hard to come by even though the way biodiesel combusts is quite similar,"Kong said.

The Iowa researchers will be particularly reliant upon Kong's models going forward, as the operating parameters of large diesel generators can't be changed to see if researchers are on the right track with their ideas. The generators provide power every day and need to operate in accordance with EPA standards.

"Due to the emission regulation, the power plants can't change operating conditions," Kong said. "So we'll use the models for simulation to see how, if we did make some changes in operating conditions, NO_x would be reduced."

In addition to measuring NO_x and carbon dioxide emissions, the tests also measured power output and generator fuel consumption to provide more information about biodiesel's use in generators.

"We're very excited we could do this study on a big diesel engine because the magnitude and health effects to the public are huge," Kong said. "And if we could burn this much as biodiesel in this type of engine, it would be very helpful to the bio-economy, especially in Iowa."

And studies like the Iowa Energy Center's and the Oak Ridge National Lab's could also affect whether you'll be fueling up with vegetable oil in the near term.