LBNL Masthead A-Z IndexBerkeley Lab mastheadU.S. Department of Energy logoPhone BookJobsSearch
Enol Intermediates Unexpectedly Found in Flames Print

For those studying flame chemistry and the properties of combustion intermediates by means of molecular beam mass spectrometry, the addition of tunable vacuum ultraviolet (VUV) from a synchrotron to photoionize the beam for mass spectrometry makes for a powerful technique capable of differentiating between isomers with the same molecular weight and composition. With the help of a unique experimental apparatus, an international team of American, Chinese, and German researchers has exploited this selectivity to identify chemical compounds known as enols as apparently ubiquitous intermediates in flames burning a variety of fuels. This surprising observation will require combustion modelers to revise their models to account for the presence of these compounds.

A Combustion Surprise

Taming fire is hailed as one of the key steps that set humans on the road to civilization. More recently, chemists have been intensively studying combustion (more generally, hydrocarbon oxidation) for more than 150 years. Over the decades, they have uncovered an intricate network involving thousands of reaction pathways through hundreds of short-lived transient molecules (reaction intermediates) as fuel burns and transforms into the final products, which often include lung-damaging particulates like soot, smog-causing nitrogen oxides, and other pollutants. Nowadays, detailed mathematical models get equal billing with advanced experimental techniques in the quest to master combustion at the molecular level needed for improving efficiency and reducing pollution.

The accuracy of these models depends on knowing all the chemical species involved and having quantitative values for the reaction constants that describe how the reactions proceed. Taatjes et al. have added a new ingredient to the swirling mix of intermediates with their discovery that a class of compounds known as enols (ethenol or vinyl alcohol is the simplest of these), only recently observed as a combustion intermediate, is in fact widely present in flames burning a variety of fuels, including commercial blends, such as gasoline. Combustion modelers will have to take this new finding to heart, as will those studying other forms of hydrocarbon oxidation important in such widely varied settings as fuel cells, planetary atmospheres, and interstellar space.

After more than 150 years of study, combustion seems to be well understood in terms of average energy output, high-concentration intermediates, and major products. However, for improving combustion efficiency and controlling pollution, it is necessary to understand flame chemistry at the parts-per-million level while simultaneously facing the turbulent fluid dynamics of a "real" flame. As a result, many important rate constants have never been measured directly, nor have all the species included in mathematical flame models been directly observed. To this end, a joint collaboration between Sandia National Laboratories, Cornell University, the University of Massachusetts (Amherst), and Berkeley Lab has developed a low-pressure-flame photoionization mass spectrometer that allows experimenters to isolate the chemistry.

A schematic of the flat-flame burner and molecular beam sampling assembly at ALS Chemical Dynamics Beamline 9.0.2.

In the apparatus, premixed reagent gases enter the flame chamber through the porous flat face of a burner that translates horizontally relative to a fixed quartz sampling cone and nickel skimmer, which allows the temperature profiles and concentration profiles to be mapped to very high precision. A well-collimated molecular beam from the skimmer enters a differentially pumped (10-6 Torr) chamber, where it is photoionized by a crossed tunable VUV beam. Photoions are mass-analyzed using a time-of-flight (TOF) mass spectrometer (MS).

The luminous zone of the flame shown here (just to left of the glowing red sampling cone) has the typical blue-violet or blue-green color associated with chemiluminescence from electronically excited CH and C2.

Among enols, the simplest is ethenol (vinyl alcohol). It is thermodynamically unstable relative to its isomer acetaldehyde and has only recently been observed as an intermediate in an ethene flame [T.A. Cool et al., J. Chem. Phys. 119, 8356 (2003)]. The team launched a systematic search for enols among 24 different flames of 14 prototypical single fuels found in modern fuel blends; they also studied commercial gasoline. Experiments were conducted with similar flame chambers operating at a branch of ALS Chemical Dynamics Beamline 9.0.2 and at the National Synchrotron Radiation Laboratory (Hefei, China).

Photoionization efficiency curves taken for m/z = 44 ions sampled from four representative flames showed that ethenol is present in all. The 0.9-eV difference between the photoionization thresholds for ethenol and its isomer acetaldehyde made them easily distinguishable. In addition to ethenol, larger enols also occur in flames of both simple fuels and commercial gasoline. Photoionization measurements of m/z = 58 ions show the presence of propenols, while butenols occur in measurements for m/z = 72 ions sampled from a gasoline flame.


Enols are widespread intermediates in flames. Left: Photoionization efficiency curves for m/z = 44 ions sampled from four flames burning representative fuel compounds show that ethenol is present in all. Right: Larger enols also occur in flames. Photoionization measurements of m/z = 58 ions show the presence of propenols, while butenols occur in measurements of m/z = 72 ions.

In the case of ethenol, not only are the concentrations far too high to be explained as the isomerization of acetaldehyde, but the data suggest that ethenol kinetics in flames are distinct. Markedly differing distributions of ethenol and acetaldehyde with distance from the flame burner suggest either separate formation mechanisms or differential removal of ethenol as it diffuses toward the burner. And the increasing fraction of ethenol relative to acetaldehyde with distance from the burner for two flames suggests that the chemical fates of the two are not at all the same.


Chemistries of ethenol and acetaldehyde in the flame are not the same. Left: Differing distributions of the two isomers with distance from the flame burner. Right: Increasing fraction of ethenol relative to acetaldehyde with distance from the burner.

While the practical impact of these findings on combustion remains speculative for the moment, understanding the fundamental chemistry of enols, important not only in combustion but also in other forms of hydrocarbon oxidation important in such widely varied settings as fuel cells, planetary atmospheres, and interstellar space, clearly requires much more theoretical and experimental study.



Research conducted by C.A. Taatjes (Sandia National Laboratories and JILA); N. Hansen, A. McIlroy, J.A. Miller, J.P. Senosiain, and S.J. Klippenstein (Sandia National Laboratories); F. Qi (Sandia National Laboratorires and National Synchrotron Radiation Laboratory, China); L. Sheng and Y. Zhang (National Synchrotron Radiation Laboratory, China); T.A. Cool and J. Wang (Cornell University); P.R. Westmoreland and M.E. Law (University of Massachusetts, Amherst); T. Kasper and K. Kohse-Höinghaus (Universität Bielefeld, Germany).

Research funding: U.S. Department of Energy, Office of Basic Energy Sciences (BES) and National Nuclear Security Administration; the U.S. Army Research Office; the Chinese Academy of Sciences; the National Natural Science Foundation of China; and the Deutsche Forschungsgemeinschaft. Operation of the ALS is supported by BES.

Publication about this research: C.A. Taatjes, N. Hansen, A. McIlroy, J.A. Miller, J.P. Senosiain, S.J. Klippenstein, F. Qi, L. Sheng, Y. Zhang, T.A. Cool, J. Wang, P.R. Westmoreland, M.E. Law, T. Kasper, and K. Kohse-Höinghaus, "Enols are common intermediates in hydrocarbon oxidation," Science 308, 1887 (2005).