AmericanChemicalSociety.com

Our studies on the effects of oxygenated hydrocarbons on soot formation have focused on two areas which are discussed below:  (1) identifying the mechanisms of synergistic effects that have been observed between oxygenated fuel additives and the baseline fuel, and (2) measuring sooting tendencies of oxygenated hydrocarbons that can be used to determine the oxygen-containing moieties that are most effective in reducing soot emissions.

1.0       Numerical Simulation of Ethylene/Ethanol and Ethylene/DME Flames

In earlier work we made measurements in coflow nonpremixed ethylene flames with up to 10 % ethanol or dimethyl ether (DME) added to the fuel [1].  In the current research we numerically simulated these flames using detailed chemical kinetic mechanisms.  These simulations were done in collaboration with our Yale colleagues Beth Anne Bennett and Mitchell Smooke.  The attached paper [2] provides a detailed discussion of these simulations, but the conclusions are summarized here.

(1)      The computational results agreed well with the experimental results.  These are the first comparisons of experimental and numerical results in two-dimensional flames, which are an important intermediate step between one-dimensional flames and real systems such as engines.  Furthermore, they are the first comparisons to include large hydrocarbons such as benzene that serve as soot precursors.  Accurate simulations of flames require detailed chemical kinetic mechanisms for the fuel molecules.  Our work shows that the mechanisms available in the literature [3], even without tuning, are adequate for accurate simulations.

(2)      In our experimental results we observed that – although dimethyl ether and ethanol are considered soot reducing additives – in small amounts they increased the amount of soot formed in ethylene nonpremixed flames.  The numerical results in the current work helped to identify the chemical mechanism responsible for these synergistic effects. This synergistic effect is dependent upon the special nature of ethylene flame chemistry, and is unlikely to be important for alkane-based fuels such as gasoline or diesel.

2.0       Sooting Tendencies of Oxygenated Hydrocarbons

A sooting tendency is a parameter that quantitatively describes the relative propensity of hydrocarbons to produce soot in flames.  In an earlier study we proposed an alternative definition of sooting tendency:  the maximum soot volume fraction f_v,max in a methane coflow nonpremixed flame with a fixed concentration of the test hydrocarbon added to the methane [4].  With this procedure we have measured sooting tendencies for over 130 aromatic hydrocarbons [4], which is more than the total sum of smoke point sooting tendencies that have every been reported.  In this study we extend these measurements to aliphatic regular and oxygenated hydrocarbons.  Only a few sooting tendencies for oxygen-containing fuels have been reported.

The results were converted into apparatus-independent Yield Sooting Indices (YSI's) defined by the equation

YSI                              =          C * f_v,max   +   D,                                             (1)

where C and D are apparatus-specific parameters chosen so that YSI-hexane = 0 and YSI-benzene = 100.

2.         Experimental methods

The procedures were the same as in our earlier study [4].  Atmospheric-pressure coflow nonpremixed flames were generated with a burner in which the fuel mixture flows out of an 11 mm diameter tube and reacts with air that flows from the annular region between this tube and an outer chimney.  A syringe pump injected the test compounds into the gas-phase fuel mixture. The nominal reactant flowrates were 330 cm³/min (CH₄), 275 cm³/min (N₂), 0.61 cm³/min (test compound), and 30,000 cm³/min (air).  The resulting concentration of the test compounds in the fuel mixture was 1000 ppm.  Soot volume fractions were measured with laser-induced incandescence (LII).

3.         Results

The data was collected in sets where f_v,max was measured for hexane, benzene, isooctane, and three test compounds. Values of C and D in equation (1) were then calculated and used to determine YSI's for the test compounds.  This procedure was repeated, with slightly different values of C and D arising each time, until at least three YSI determinations had been obtained for each test compound.  These results were then averaged to produce the final value.  The calibration constant for the f_v measurements factors out when f_v,max is converted to YSI, but the absolute volume fractions were in the range 0.2 to 1 ppm.  Isooctane was included in each set as an internal check on the measurements.

Table 1 reports the final values of YSI.  Regular hydrocarbons are listed first and then oxygenated compounds.  The YSI of isooctane was measured 25 times during this study; the average was 48.6, the standard deviation was 0.71, and the range was 47.4 to 50.1; we estimate that the total uncertainties in the YSI's are plus or minus 2.

Table 1. Yield-Based Sooting Tendencies of Hydrocarbons

3.0       References

1. C. S. McEnally, L. D. Pfefferle, Proceedings of the Combustion Institute 31:603-610 (2007).

2. B.A.V. Bennett, C.S. McEnally, L.D. Pfefferle, M.D. Smooke, M.B. Colket, Combustion and Flame 156:1289-1302 (2009).

3. K.H. Song, P. Nag, T.A. Litzinger, D.C. Haworth, Combustion and Flame 135:341-349 (2003).

4. C.S. McEnally, L.D. Pfefferle, Combustion and Flame 148:210-222 (2007).