60
J. Phys. Chem. A, Vol. 103, No. 1, 1999
Hynes et al.
increased but the mode of inhibitor decomposition was found
to be similar to that observed at the 1.0% loading. Clearly, the
chemical kinetics of inhibitor destruction in rich flames are not
well understood and need to be studied in more depth.
Conclusion
Inspection of the product yields and reaction-path analysis
suggests that CF3CHFCF3 decomposes primarily by C-C bond
fission at temperatures above 1300 K, pressures between 16
and 18 atm, and residence times of 650-850 µs. Reaction-path
analysis also shows that the abstraction of the secondary H atom
by F was found to be the next most important destruction route,
whereas the 1,2-HF elimination channel was found to be slower.
Kinetic modeling showed that the fate of the CF3 radicals
produced in the major initiating reaction was recombination to
form C2F6; only traces of CF3H and CF4 were observed. The
fate of the CF3CHF radicals was loss of an F atom from the
CF3 group to form the stable CF2dCHF molecule. Formation
of C2F4 proceeds from the CF3CFCF3 radical which undergoes
C-C bond scission to form the carbene, CF3CF: This undergoes
a 1,2-F-shift isomerization reaction to form the product.
The yields of all major and minor products have been
satisfactorily described by the detailed kinetic model involving
the aforementioned radicals. For 3% reactant, product yields
of CF3CHFCF3 were lower, suggesting either a non-first-order
dependence of the reactant decomposition rate on the reactant
concentration or the rapid reactive cooling that takes place
behind the reflected shock at this high reactant concentration.
3 3
Figure 9. Reaction fluxes for CF CHFCF decomposition and product
formation. The largest arrows indicate the reactions with the largest
flux. The conditions are 0.5% initial reactant concentration and a
temperature of 1500 K.
noted above, the reformation of the reactant is also significant
and further enhances the heat capacity of the product mixture.
Sensitivity analysis also shows that the major products C2F6,
C2F4, and CF2dCHF are very sensitive to reactions 37 and 27
at temperatures above 1400 K. The product C3F6 is sensitive to
reaction 38 as expected and also to the two abovementioned
reactions.
As noted above, the production of F atoms is important in
this model. The principal source of F is from the CF3CHF radical
Acknowledgment. We thank Dr. G. B. Bacskay and Mr.
M. Smith for performing the G2-MP2 calculations used in this
work.
(44%). As noted above, reactions 39 and 23 account for 18%
and 23% of the F destruction routes, respectively.
Discussion
References and Notes
In terms of flame chemistry, this study is most relevant to
fuel-rich flames with a high CF3CHFCF3 loading. In rich,
uninhibited flames, the fuel competes with O2 for H radicals;
(1) Nyden, M. D.; Linteris, G. T.; Burgess, D. R., Jr.; Westmoreland,
P. R.; Tsang, W.; Zachariah, M. R. Flame Inhibition Chemistry and the
Search for Additional Fire Fighting Chemicals. In EValuation of AlternatiVe
In-Flight Fire Suppressants for Full-Scale Testing in Simulated Aircraft
Engine Nacelles and Dry Bays; Grosshandler, W., Gann, R., Pitts, W., Eds.;
NIST Spec. Pub. 861; National Institute of Standards and Technology:
Gaithersburg, MD, 1994; pp 467-641.
7
this behavior is self-inhibitory, as described by Linteris et al.
The higher concentration of CH3 radicals in these fuel-rich
flames gives rise to higher rates of chain termination, leading
to the formation of C2H6 and C2H4. Addition of inhibitor to
these flames provides more competition for H but at the same
time it is expected that pyrolysis of the inhibitor will occur given
the lower radical concentrations. As suggested by Linteris et
(2) Halon Replacements: Technology and Science; Miziolek, A., Tsang,
W., Eds.; ACS Symposium Series; American Chemical Society: Washing-
ton, DC, 1995; Vol. 611.
(
3) Zachariah, M. R.; Westmoreland, P. R.; Burgess, D. R., Jr.; Tsang,
W.; Melius, C. F. J. Phys. Chem. 1996, 100, 8737.
4) Burgess, D. R., Jr.; Zachariah, M. R.; Tsang, W.; Westmoreland,
(
7
al., the initiation step at lower loadings in fuel-rich flames
P. R. Prog. Energy Combust. Sci. 1996, 21, 453.
(5) Sanogo, O.; Delfau, J.-L.; Akrich, R.; Vovelle, C. Combust. Sci.
Technol. 1997 , 122, 33.
5
,6
should proceed by abstraction of the secondary H atom by
CH3 and other radicals to form CF3CFCF3. At higher loadings,
the same initiation step will take place, however, the 1,2-HF
elimination reaction forming CF3CFdCF2 and C-C bond
rupture will also occur. The formation of unsaturated fluoro-
carbon compounds observed in this study can undergo addition
reactions with CH3 and other radicals, forming larger fluorinated
species or soot precursors. In fact, most of the two- and three-
carbon products observed in the pyrolysis study have also been
observed in laminar counterflow CH4-diffusion flames inhibited
with CF3CHFCF3:33 CF3-CF3, CF2dCF2, CF3CFdCF2, and
CHFdCF2. Also observed was CF2dCH2 and heavier species
including hexafluorobutene, benzene, fluorobenzene, 1,2-dif-
luorobenzene, trifluoromethylbenzene, and toluene. This indi-
cates that the chemistry of inhibitor consumption in rich
environments is quite different to that observed in fuel-lean
environments, where partially oxidized species including CF2O
(
6) Hynes, R. G.; Mackie, J. C.; Masri, A. R. Combust. Flame 1998,
13, 554.
(7) Linteris, G. T.; Burgess, D. R., Jr.; Babushok, V.; Zachariah, M.;
Tsang, W.; Westmoreland, P. Combust. Flame 1998, 113, 164.
1
(8) DiFelice, J.; Ritter, E. R. Combust. Sci. Technol. 1996, 116-117,
5.
(9) Doolan, K. R.; Mackie, J. C. Combust. Flame 1983, 49, 221.
(10) Mackie, J. C.; Colket, M. B.; Nelson, P. F. J. Phys. Chem. 1990,
4, 4099.
9
9
9
(11) (a) Terentis, A.; Doughty, A.; Mackie, J. C. J. Phys. Chem. 1992,
6, 10334. (b) Doughty, A.; Mackie, J. C. J. Phys. Chem. 1992, 96, 10339.
(12) Chen, Y.; Rauk, A.; Tschuikow-Roux, E. J. Chem. Phys. 1991,
4, 7299.
(13) Smith, M. BSc (Hons) Thesis, School of Chemistry, University of
Sydney, 1998.
(14) Gl a¨ nzer, K.; Maier, M.; Troe, J. J. Phys. Chem. 1980, 84, 1681.
(15) Dixon, D. A. J. Phys. Chem. 1986, 90, 54.
(
(
16) O’Gara, J. E.; Dailey, W. P. J. Am. Chem. Soc. 1992, 114, 3581.
17) Atkinson, B.; McKeagan, D. J. Chem. Soc., Chem. Commun. 1966,
1
89.
18) Kee, R. J.; Miller, J. A.; Jefferson, T. H. CHEMKIN; A General
Purpose, Problem Independent, Transportable FORTRAN Chemical Kinetics
6
and CF3CFO were observed at 1.0% loading. At a higher
(
loading of 3.2% in lean flames, the temperature of the flame