Researchers Burning to Learn the Secrets That Simmer Inside Flames
A flame’s a flame, no matter how you slice it, and a Yale professor is doing just that, cutting it into manageable portions with lasers and computer technology to try to show exactly what a flame is, what it does and where it does it.
Since Prometheus snatched it from the gods, fire has been essential to humans. It cooks for us, it warms us, it makes a candlelight dinner romantic. Hidden inside machinery, it makes cars and jet planes go. It also does bad things. It can hurt, it can allow noxious pollutants to escape into the air, it can destroy huge buildings.
But while fire has been both our servant and scourge, no one has understood what goes on in the heart of the flame.
Changing Its Behavior
And scrutinizing it can be important to making it behave more the way we want it to--in making bus fumes less smelly, designing sprinkler systems to stop fires or producing highly efficient engines for supersonic planes.
“If you physically probe a flame, you can measure some of the way it works,” said Marshall B. Long, an associate professor in Yale’s department of mechanical engineering. “You can put in a wire and try to understand how it heats or cools in different parts of the flame. But this . . . is intrusive; the wire affects the turbulence of the flame you’re studying.”
A better way, as scientists found about 15 years ago, is to use a much less intrusive probe: a laser beam.
Long has been working on this problem since it formed the basis for his doctoral thesis in the late ‘70s. The job needs a team because the research cuts across the disciplines--fluid mechanics, chemistry, the physics of light-scattering and the unsolved problem of dynamic mechanisms, or turbulence.
Turbulence--shown in the way cigarette smoke rises in a straight line to the ceiling and then breaks into agitated wiggles--is the heart of the flame because it mixes things up, and burning is the combination of a fuel with oxygen. Turbulence lies at the heart of other questions, such as weather prediction and the study of the behavior of the Gulf Stream.
Because of its central role in industry, defense and people’s lives, flame research tends to get plenty of funding. Long’s team has been bankrolled by the National Science Foundation, the Department of Energy, the Department of Defense, the Air Force Office of Scientific Research, United Technologies Corp. and automobile manufacturers.
‘Equipment Intensive’
“It’s equipment intensive,” Long said. His labs have up-to-date computers on which pictures of flames can be rotated and studied, as well as flame-probing lasers, sophisticated sensors and an elegant electronic camera that can take up to 20 million frames a second. “We’ve got to keep our eye on what’s new in technology because we’re always trying to push back the limits of what we’re studying,” he said.
What Long and his colleagues have done is provide a way to look not only at a point in a flame, as one thin laser beam might be able to do, but to look at flame three dimensionally, Long said.
When a laser beam passes through a flame, it bumps into the molecules in its path, and at each bump a little bit of light bounces to the side of the beam. Most of this scattering is invisible but can be picked up by spectroscopes.
These light fragments are converted into digits and fed into a computer, which can paint a picture of the lumps and disturbances in a flame and show the character of the molecules and what they are doing. Temperature, density, concentrations of various chemicals and speed all show up.
But a thin laser beam will only provide a picture of the molecules along that beam, something like studying a dark room with a flashlight. So Long and his Yale colleagues--Brandon Yip, Joseph K. Lam and Michael Winter--aim their laser beam through lenses that broaden the beam to about one-third of an inch wide.
This broad beam can scatter light over an entire cross-section--a two-dimensional slice--of the flame. The computer makes a picture that can be colored to show different concentrations of gas.
And when it comes to fire, what is important is how abruptly those concentrations change in a flame. Long and his colleagues chart this on a slope on a graph and call it the concentration gradient. As more heat is generated, the gradient becomes steeper.
This is important, for example, in automobile engines. In the late 1970s, laser research by the Ford Motor Co. and the Combustion Research Facility in Livermore, Calif., found not enough burning going on in odd corners of engine cylinders, and the researchers redesigned the cylinders accordingly.
These researchers and their counterparts at United Technologies Research Center in East Hartford, Conn., beam lasers through engines with holes and transparent cylinders. “In Germany they have even built completely transparent quartz engines,” said Alan C. Eckbreth, manager of propulsion science at the research center.
But a two-dimensional slice of the flame is not enough help to scientists who want to look at the topography of flame and how it relates to temperature and burning efficiency. It is like studying an apple by looking at a slice that may or may not include the core.
So Long and his colleagues changed the experiment slightly. The laser beam, again broadened, is beamed into a rotating mirror that scans the beam across the flame. Because the beam has to portray all the slices of the flame as close to exactly the same moment as possible, it has to move very fast.
As the laser sweeps across the flame, the electronic camera takes its rapid-fire shots, each exposure lasting from two-millionths to 10-billionths of a second.
When this was first done, the result was an extremely pretty series of pictures--in effect, 16 slices of a flame. Pieced together by the computer like a model of a landscape made from contour maps, and colored and shaded, they made an extraordinary picture.
Cut off at the top and bottom, the three-dimensional image can be turned and examined on the screen of Long’s computer. The image looks a little like a hollow log because it is a portrait of the surface of the flame only. When you look inside the log on the computer screen, you see the back of the surface.
But Long and his colleagues were not simply looking for interesting graphics.
“This provides measurements, not just visualizations,” Long said. The image made it possible to make a precise accounting of what had until now been flickering and unmeasurable.
A flame on a candle in a perfectly still room is the combination of a fuel--candle wax, for example--and oxygen to produce carbon dioxide and water. “The chemistry is relatively well-understood, but the fluid dynamics is very complicated and not well understood,” Eckbreth said.
On the candle, burning is thought to occur in a thin, curved plane above the wick, where the concentrations of fuel and oxygen are exactly right.
But in a Bunsen burner or a jet engine, a flame is not as simple as in a candle. The plane where burning takes place bunches up into masses of crinkles. It breaks up into whorls and eddies like a rushing stream. Predicting what is going to happen along that plane becomes more and more difficult.
“As you turn up the flow of the flame, and the turbulence of the flame, you’ll still have the contour” of the plane, Long said. But something odd starts to happen. “The chemistry can’t keep up.” Something is going on along that plane that scientists cannot explain.
So some scientists have proposed a different model for when flame is created--not a two-dimensional plane but something between two and three dimensions, something that mathematicians call “fractal dimensionality,” perhaps 2.3 dimensions.
This is where the mystery remains in the flame. “In turbulence, there are more unknowns than equations,” as Eckbreth described it. “In many of these complex numerical models, certain assumptions based on intuition have never been experimentally investigated. These modern techniques--such as Long’s--enable us to do so.”
