Thermal Decomposition of 2-Fluoroethanol: Single Pulse Shock Tube and ab Initio Studies

Article abstract of DOI:10.1021/jp027323x

The thermal decomposition of 2-fluoroethanol (FEOH) was studied at 1000-1200 K behind reflected shock waves in a single pulse shock tube. The total pressures behind the reflected shocks varied between 13 and 23 atm. The products were CH₃CHO, C₂H₃F, CH₄, CO, C₂H₄, and C₂H₆. The unimolecular eliminations of HF and H₂O were the major channels through which FEOH decomposed under these conditions. The rate constants for HF and H₂O elimination were presented. The CH₃CHO produced by HF elimination through the vinyl alcohol intermediate was chemically active and decomposed leading to CH₄ and CH₃CH₃ products. Ethylene formation could be explained by considering C-O bond dissociation equally well. Direct real-time spectroscopic observation of HOF would be needed to choose between the two pathways. Ab initio (Hartree-Fock [HF] and second-order Moller-Plesset perturbation theory) and density functional theory [DFT] computations overestimated the barrier by 18 kcal/mole for HF elimination and 22 kcal/mole for H₂O elimination, and including electron correlation improved the agreement. DFT predictions for activation energies for HF and H₂O elimination reactions were within 1 kcal/mole of the experimental values.

Full text of DOI:10.1021/jp027323x

9782
J. Phys. Chem. A 2003, 107, 9782-9793
Thermal Decomposition of 2-Fluoroethanol: Single Pulse Shock Tube and ab Initio Studies
B. Rajakumar,†,# K. P. J. Reddy, and E.Arunan*
‡
,†
Department of Inorganic and Physical Chemistry and Department of Aerospace Engineering,
Indian Institute of Science, Bangalore 560012, India
ReceiVed: October 28, 2002; In Final Form: September 8, 2003
The thermal decomposition of 2-fluoroethanol (FEOH) has been studied in the temperature range of 1000-
1200 K behind reflected shock waves in a single pulse shock tube. The total pressures behind the reflected
shocks varied between 13 and 23 atm. The products observed in the covered temperature range are CH
3
CHO,
C
2
H
3
F, CH , CO, C , and C . The unimolecular eliminations of HF and H O are found to be the major
4
2
H
4
2
H
6
2
channels through which 2-fluoroethanol decomposes under these conditions. The rate constant for HF
1
3.17(0.33
-1
elimination is found to be 10
2
exp[-(59.5 ( 1.7)/(RT)] s , and the rate constant for H O elimination
is found to be 101
The CH
decomposes leading to CH
the CH CHO pyrolysis mechanism. It is most likely formed directly from the reactant through HOF elimination
4.30(0.34
exp[-(69.7 ( 1.7)/(RT)] s , where the activation energies are given in kcal mol .
-1
-1
3
CHO produced by HF elimination through the vinyl alcohol intermediate is chemically active and
and CH CH products. The production of ethylene could not be explained from
4
3
3
3
or by radical processes beginning with C-O bond dissociation. Ab initio (Hartree-Fock [HF] and second-
order Møller-Plesset perturbation theory [MP2]) and density functional theory [DFT] calculations have been
2
carried out to find the transition state and activation barrier for HF and H O elimination reactions. The HF
-1
-1
calculations overestimate the barrier by 18 kcal mol for HF elimination and 22 kcal mol for H
2
O elimination,
and including electron correlation improves the agreement. In particular, DFT predictions for activation energies
-
1
2
for HF and H O elimination reactions are within 1 kcal mol of the experimental values.
I. Introduction
elimination, and indeed the report on CH2Cl-CH2OH has
1
4
identified only the HCl elimination channel. It would be
interesting, as well, to study the effect of halogen substitution
on the barrier to H2O elimination. With these considerations,
we have initiated studies on thermal decomposition of halo-
ethanols in a single pulse shock tube, and this manuscript reports
our results on 2-fluoroethanol. A future publication will address
in detail the thermal decompositions of 2-chloroethanol and
ethanol.
Recently, 2-fluoroethanol (FEOH) has been suggested as a
replacement for chlorofluorocarbons (CFC). It is important to
1
know all of the physical and chemical properties of molecules
that are being considered for such applications before they are
put into commercial use. Often incineration is the method of
choice for disposing of used chemicals, and experimental data
on the mechanism of pyrolysis would be valuable. However,
there are no experimental or theoretical reports yet on the
thermal decomposition of FEOH. Moreover, we have been
interested in HX elimination reactions from haloethanes and
recently reported our results on HCl elimination from 1,2-
There have been numerous reports, both theoretical and
2
4-31
experimental, on FEOH recently.
All of these reports have
considered the structure, vibrational frequencies, isomerization,
and intramolecular vibrational energy redistribution (IVR) of
the ground state and rotational isomers. In this manuscript, we
discuss the high-barrier HF and H2O elimination reactions from
FEOH. Single pulse shock tube and ab initio and DFT methods
have been used to study the thermal decomposition of FEOH
between 1000 and 1200 K. Theoretical studies have been carried
out to characterize the transition state (TS) for both reactions.
Transition-state theory (TST) calculations using theoretical
results have been performed for comparison with experimental
results.
2
dichloroethane. The substitution effect on the barrier to HX
elimination from haloethanes has been studied in detail both
2
-14
15,16
experimentally
and theoretically.
However, the effect
of OH substitution on HX elimination has rarely been addressed,
except for one report on 2-chloroethanol.14
Thermal decomposition of ethanol has attracted much interest
as well.1
7-21
However, as is evident from a recent paper by Lin
and co-workers,22 there is still a need for reliable high-
temperature kinetic data for ethanol. Both C-C dissociation
and H2O elimination channels are important for ethanol, and
most of the thermal studies1
7-21
showed the dominance of C-C
II. Materials and Methods
23
dissociation. Chemical activation studies and unpublished
shock tube results from Lin and co-workers22 have identified
H2O elimination as well. Halogen substitution on ethanol opens
up the possibility for another low-barrier channel, namely, HX
II.A. Experimental Details. The shock tube facility used in
2
this study has been described in detail in an earlier publication ,
and only a brief summary is given here. It is an aluminum tube
of 51 mm internal diameter and 25 mm wall thickness. The
lengths of the driven section and driver section can be varied
by attaching/detaching small segments of the tube. In these
experiments, the length of the driver section was 1850 mm and
that of the driven section was 2581 mm. The dwell time was
*
To whom correspondence should be addressed. E-mail: arunan@
ipc.iisc.ernet.in.
†
Department of Inorganic and Physical Chemistry.
Current Address: Aeronamy Laboratory, NOAA, Boulder, CO 80305.
Department of Aerospace Engineering.
#
‡
1
0.1021/jp027323x CCC: $25.00 © 2003 American Chemical Society
Published on Web 10/24/2003

Thermal Decomposition of 2-Fluoroethanol
J. Phys. Chem. A, Vol. 107, No. 46, 2003 9783
Figure 1. Gas chromatogram of a postshock mixture of 2-fluoroethanol in argon heated to 1154 K obtained on a 2-m Porapak Q column using
FID: (A) methane; (B) ethylene; (C) ethane; (D) vinyl fluoride; (E) acetaldehyde; (F) 2-fluoroethanol.
typically 1200 µs. The experimental procedure is very similar
was obtained by summing up the concentrations of all of the
species, except for ethane. Ethane concentration was multiplied
by 2 because two FEOH are required to form ethane according
to the kinetic mechanism that fits all experimental observation.
The sensitivity of the detector toward each compound was
determined with standard samples: 2-fluoroethanol (Lancaster
chemicals Inc.), acetaldehyde (Merck), ethyl chloride and vinyl
fluoride (Fluka), ethylene (Hydrogas), and methane, ethane, and
propylene (Bhoruka Gases). All of the other gases used in GC
analysis such as argon, helium, oxygen, and hydrogen were from
Bhoruka Gases, India (UHP grade 99.999%).
to that suggested by Tsang for single pulse shock tube
1
3
experiments. A 1% mixture of 2-fluoroethanol in argon was
prepared manometrically in a 10 L Pyrex glass bulb. This
mixture was used in all of the runs. The shock tube was
-
6
evacuated to 10 Torr before the experiment with a 6 in.
diffusion pump. The ball valve was kept closed, and the FEOH
mixture was loaded into the sample chamber (between the ball
valve and the end flange) to 20-40 Torr using a mercury
manometer. The sample was further diluted with argon until a
desired pressure is reached. The final concentration of the
sample was in the range of 500-1000 ppm. The segment of
the driven section from the diaphragm to the ball valve was
filled only with argon to a slightly larger pressure (4-6 Torr
higher) than the sample chamber to avoid the back diffusion of
the test gas. In all of the experiments, this pressure (P1) was
varied between 400 and 650 Torr. Just before rupturing the
diaphragm, the ball valve was opened, and soon after the
experiment, the ball valve was closed. The reaction dwell time
was determined from the experimental pressure trace, and the
reflected temperature was estimated using an external standard,
that is, HCl elimination from ethyl chloride under similar
II.C. Computational Details. All calculations have been
32
performed with the Gaussian 94 program suite. Geometry
optimizations have been carried out at Hartree-Fock (HF),
second-order Møller-Plesset perturbation theory (MP2(FULL)),
and density functional theory (DFT) levels with the standard
6-31G*, 6-31G**, 6-31+G**, and 6-311++G** basis sets that
are internally available in the program suite. Equilibrium and
transition-state structures were fully optimized using Hartree-
Fock, second-order Møller-Plesset perturbation theory with all
the electrons correlated, and the density functional theory with
the B3LYP correlation functional. All transition states have been
characterized with frequency analysis. Vibrational frequencies
and zero-point energies were not scaled. These results were used
in performing transition-state theory (TST) calculations to
predict the activation energy and the preexponential factors for
the HF and H2O elimination reactions for comparison with
experimental measurements. For additional comparison, calcula-
tions on H2O elimination from ethanol were carried out as well,
and the results are included. The enthalpy of formation of FEOH
is not available to the best of our knowledge, and it was
determined at the MP2/6-311++G** level. Our main interest
in determining this quantity was to get a rough estimate of the
internal energy in the products.
2
conditions.
II.B. Materials and Analysis. Spectroscopic grade 2-fluoro-
ethanol (>99.5%, Lancaster chemicals Inc.) was used as
received. For degassing and further purification, the freeze-
pump-thaw method was used several times before making the
sample. The purity of the reactant was routinely confirmed by
gas chromatography. The postshock gas mixture was analyzed,
plus
quantitatively, by using an HP6890 gas chromatograph (GC)
with flame ionization detector and, qualitatively, by IR spec-
trometer (Bruker Equinox 55 FTIR). The sample section of the
shock tube was connected to the GC, and 0.5 mL of the mixture
was injected through an online sampling valve. Porapak (6 ft
length) Q column was found to be suitable for both the reactant
and products. The major products formed in the process of
thermal decomposition are CH2dCHF and CH3CHO. The other
observed products were CH4, C2H4, and C2H6. Some trace
amounts of propylene were also formed at temperatures above
III. Experimental Results and Discussion
Thirty-four experiments were carried out between T ) 940
and 1200 K, and postshock mixtures were quantitatively
analyzed with GC. Details of the experimental conditions and
the product distribution are given in the Table 1. It includes
the temperature, T5, the pressure, P5, the dwell time, and
concentrations of all the observed species. The P5 was above
1
150 K but were too small to be quantified. The GC analysis
was carried out at a constant oven temperature (413 K). A typical
chromatogram of the reaction mixture at the temperature, 1154
K, is shown in Figure 1. The initial concentration of reactant

9784 J. Phys. Chem. A, Vol. 107, No. 46, 2003
Rajakumar et al.
TABLE 1: Summary of the Experimental Results of 2-Fluoroethanol
dwell
T
5
P
5
time
no (K) (atm) (µs) [CH
4
]
t
/[FEOH]
0
[C
H
2 4
]
t
/[FEOH]
0
[C
2 6
H
]
t
/[FEOH]
0
[C
2 3
H F]
t
/[FEOH]
0
[CH
3
CHO]
t
/[FEOH]
0
t 0
[FEOH] /[FEOH]
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
936 13.2 1150
961 14.2 1210
963 13.9 1160
969 14.2 1140
983 15.0 1230
986 15.0 1200
986 15.1 1220
987 15.2 1160
1009 16.6 1380
1013 15.8 1180
1014 16.2 1240
1018 16.6 1240
1028 16.7 1240
1033 16.9 1230
1055 17.8 1230
1055 17.7 1240
1056 18.0 1230
1062 18.1 1260
1064 17.7 1230
1067 17.8 1230
1078 18.4 1220
1098 19.3 1200
1099 19.3 1140
1115 20.0 1200
1119 20.3 1180
1123 20.2 1200
1126 20.3 1210
1136 20.4 1310
1146 20.9 1210
1154 21.4 1200
1166 22.3 1280
1176 22.3 1160
1186 23.0 1270
1196 23.4 1260
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0.013 400
0.002 050
0.002 160
0.001 700
0.006 160
0.003 250
0.002 330
0.002 210
0.002 970
0.007 200
0.004 100
0.003 500
0.006 270
0.006 270
0.009 230
0.007 610
0.008 560
0.009 370
0.011 660
0.041 510
0.014 370
0.028 050
0.034 210
0.033 950
0.035 120
0.068 820
0.059 060
0.073 550
0.075 160
0.077 080
0.098 830
0.132 180
0.162 100
0.152 720
0.986 600
0.997 950
0.997 840
0.998 060
0.990 340
0.996 490
0.997 500
0.997 670
0.996 870
0.988 860
0.995 560
0.996 170
0.993 120
0.993 140
0.989 360
0.991 100
0.989 900
0.988 490
0.985 720
0.938 980
0.982 170
0.963 530
0.955 260
0.954 980
0.952 730
0.904 160
0.917 530
0.894 910
0.891 890
0.889 150
0.853 420
0.794 190
0.738 550
0.724 060
0.000 240
0.003 500
0.000 260
0.000 168
0.000 127
0.000 155
0.003 940
0.000 141
0.000 212
0.000 286
0.000 345
0.000 444
0.000 043
0.000 471
0.000 582
0.000 698
0.006 140
0.001 010
0.001 960
0.002 430
0.002 620
0.002 880
0.005 800
0.004 690
0.007 190
0.007 32
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
0.000 201
0.000 113
0.000 326
0.000 246
0.000 645
0.000 644
0.000 707
0.001 030
0.001 190
0.009 500
0.001 680
0.004 140
0.005 180
0.005 230
0.005 860
0.010 940
0.009 170
0.014 050
0.013 100
0.013 570
0.018 820
0.027 090
0.031 020
0.034 010
0.000 311
0.000 211
0.000 366
0.000 517
0.000 737
0.003 880
0.000 765
0.001 950
0.002 410
0.002 840
0.002 830
0.008 560
0.008 170
0.008 610
0.010 320
0.010 400
0.014 860
0.023 660
0.034 740
0.044 110
0.000 360
0.000 501
0.000 381
0.000 569
0.001 730
0.001 370
0.001 690
0.002 210
0.002 360
0.003 460
0.005 850
0.009 630
0.013 350
0.007 43
0.010 6
0.017 02
0.023 95
0.031 75
13 atm in all runs, and our analysis assumed that the FEOH
unimolecular reactions are in the first-order limit. In addition
to the products listed in Table 1, CO was identified by using
the infrared spectrum of the postshock gas mixture but was not
quantified. Acetaldehyde was observed at all T in our study,
whereas methane was observed only above 970 K. Ethylene
and ethane are found to be products above 1050 and 1100 K,
respectively. Vinyl fluoride is observed above 1015 K. The
concentrations of ethane, ethylene, and methane are significant
only at higher temperatures, that is, above 1150 K. Less than
28% of the reactants are consumed within the temperature range
and dwell times of our experiments. Figure 2 shows the
Arrhenius plot of the rate constant for overall decomposition
of FEOH calculated as follows:
ktotal ) -ln{[FEOH] /[FEOH] }/t
(1)
t
0
The result is ktotal ) 101
3.55(0.32
exp[-(60.6 ( 1.6)/(RT)] s-1,
-
1
where Ea is given in kcal mol . [FEOH]t and [FEOH]0 are the
experimentally measured concentration of FEOH at the end of
the reaction time t and initial concentration, respectively. This
plot is linear within the temperature range of analysis (1000-
Figure 2. Arrhenius plot for the overall decomposition of 2-fluoro-
ethanol.
of HF and H2O:
1
2
200 K) and suggests that the overall decomposition of
-fluoroethanol is unimolecular. In the case of nitromethane,
F-CH CH -OH f HF + CH dCH-OH
(R1)
(R2)
2
2
2
33
Zhang and Bauer noted curvature in the Arrhenius plot, which
was attributed to the consumption of nitromethane by secondary
reaction with radicals. From the Arrhenius plot given in Figure
F-CH CH -OH f H O + CH dCH-F
2
2
2
2
2
, it appears that the FEOH consumption under our experimental
(The reaction numbers here and throughout the text correspond
to the list in Table 2, which gives the complete mechanism used
for simulating the profiles of all products). The vinyl alcohol
formed in reaction R1 presumably isomerizes to acetaldehyde,
conditions is unaffected by secondary reactions.
III.A. Major Channels. The major channels through
which 2-fluoroethanol decomposes are unimolecular elimination

Thermal Decomposition of 2-Fluoroethanol
J. Phys. Chem. A, Vol. 107, No. 46, 2003 9785
TABLE 2: The Reaction Scheme Proposed for the Thermal Decomposition of 2-Fluoroethanola
reaction
number
reaction
rate expression
reference
) 1.507 × 10 exp[-59 456/(RT)] s-1
1
3
this work
this work
39
38
38
53
53
54
55
34
53
34
56
54
53
57
58
53
53
53
53
53
54
54
53
53
53
53
53
54
55
53
53
54
53
54
54
53
54
59
60
61
55
54
62
this work
62
54
40
40
63
54
64
this work
65
40
40
40
40
40
40
66
66
67
68
63
R1
R2
R3
R4
R5
R6
R7
R8
F-CH
F-CH
2
-CH
-CH
2
-OH f CH
-OH f CH
3
CHO + HF
2
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
k
1
2
3
4
5
6
7
8
9
1
4
-1
2
2
dCH-F + H
2
O
) 2.0067 × 10 exp[-69 690/(RT)] s
14 -1
CH
CH
CH
2
3
3
dCH-F f CHtCH + HF
) 2.398 × 10 exp[-81 400/(RT)] s
13 -1
CHO f CH
CHO f CH
3
+ CHO
+ CO
) 1.21 × 10 exp[-57 200/(RT)] s
1
2
-1
4
) 1.21 × 10 exp[-57 200/(RT)] s
2
1
-2.14
-1
CHO f CO + H
CHO + H f CO + H
) 5.11 × 10 (T)
exp[-20 425/(RT)] s
1
4
3
-1 -1
2
) 1.2 × 10 cm mol
s
CH
CH
CH
H + CH
H + CH
3
3
3
3
+ CHO f CH CHO
+ CH
+ CH
) 1.8129 × 10 cm mol s-1
1
3
3
-1
-
6
5.64
3
-1 -1
R9
3
CHOf CH
CHOf CH
4
+ CH
+ CH
3
2
CO
CHO
) 1.98 × 10 (T) exp[-2464/(RT)] cm mol
s
-
1
4.0
3
-1 -1
R10
R11
R12
R13
R14
R15
R16
R17
3
4
10 ) 4.5 × 10 (T) exp[-8280/(RT)] cm mol
s
1
3.5
3
-1 -1
3
CHO f H
CHO f H
2
+ CH
2
CHO
CO
11 ) 2.4 × 10 (T) exp[-5167/(RT)] cm mol
s
13
3
-1 -1
3
2
+ CH
CO
3
12 ) 8.7 × 10 exp[-4200/(RT)] cm mol
s
13
-1
CH
CH
CH
CH
CH
CH
2
3
3
3
3
3
CHO f H + CH
CO f CH + CO
2
13 ) 1.58 × 10 exp[-34 974/(RT)] s
42
-8.62
-1
3
14 ) 8.73 × 10 (T)
exp[-22 455/(RT)] s
a
14
-0.64
3
-1 -1
+ CH
+ CH
+ CH
+ C
2
3
3
3
f C
f C
f C
2
2
2
H
H
H
6
4
5
4
15 ) 6.8 × 10 (T)
cm mol
s
a
15
3
-1 -1
+ H
+ H
2
16 ) 9.9 × 10 exp[-32 987/(RT)] cm mol
s
s
a
13
3
-1 -1
17 ) 2.4 × 10 exp[-12 877/(RT)] cm mol
R18
R19
R20
R21
R22
R23
R24
R25
R26
R27
R28
R29
R30
R31
R32
R33
R34
R35
R36
R37
2
H
6
f CH
+ C
2
H
5
18 ) 0.548(T)4 exp[-8285/(RT)] cm mol
.0
3
-1 -1
s
H + C
H
6
f C
2
H
5
+ H
2
19 ) 5.54 × 10 (T) exp[-5167/(RT)] cm mol s-1
2
3.5
3
-1
8
1.19
-1
2
C H
2
C H
2
C H
2
C H
2
C H
2
C H
2
C H
2
C H
2
C H
2
C H
2
C H
2
C H
2
C H
2
C H
2
C H
2
C H
2
C H
2
C H
5
5
5
5
5
5
5
5
5
5
4
4
4
4
4
4
3
3
f C
+ H f C
+ H f C
+ C
+ C
+ H f CH
+ CH
+ CH
+ CHO f C
+ CH
f C
f C
H
2 4
+ H
20 ) 4.8 × 10 (T) exp[-37 205/(RT)] s
12
3
-1 -1
2
H
H
4
+ H
2
21 ) 1.81 × 10 cm mol
22 ) 4.86 × 10 cm mol
23 ) 1.39 × 10 cm mol
s
12
3
-1 -1
2
6
s
12
3
-1 -1
2
2
H
H
5
f C
f C
2
4
H
H
4
+ C
10
2
H
6
s
1
3
3
-1 -1
5
24 ) 1.078 × 10 cm mol
s
13
3
-1 -1
3
+ CH
3
25 ) 1.68 × 10 cm mol
s
12
-0.5
cm mol s-1
3
-1
3
f C
f C
2
H
H
4
+ CH
4
26 ) 9.8 × 10 (T)
14
-0.5
3
-1 -1
3
4
8
27 ) 2.45 × 10 (T)
cm mol
s
14
3
-1 -1
2
H
6
+ CO
+ CH
28 ) 1.2 × 10 cm mol
s
H
6
3
CO
29 ) 2.25(T)3 exp[-9141/(RT)] cm mol
.65
3
-1 -1
s
3
CHO f C
2
12
0.44
-1
2
H
H
2
+ H
2
30 ) 7.94 × 10 (T) exp[-88 769/(RT)] s
17
-1
2
3
+ H
31 ) 2.58 × 10 exp[-96 578/(RT)] s 3
+ CH
3
f C
f C
3
H
H
6
+ H
+ CH
32 ) 2.0 × 10 exp[-10 000/(RT)] cm mol s-1
13
-1
+ CH
3
2
3
4
33 ) 6.62(T)3 exp[-9499/(RT)] cm mol
.7
3
-1 -1
s
6
2.53
3
-1 -1
+ H f C
2
H
3
+ H
+ C
2
34 ) 1.32 × 10 (T) exp[-12 241/(RT)] cm mol
s
+ C
f C
2
H
5
f C
+ H
2
H
6
2
H
3
35 ) 6.32 × 10 (T) exp[-18 011/(RT)] cm mol s-1
2
3.13
3
-1
4
1
-7.49
-1
2
H
2
36 ) 4.1 × 10 (T)
exp[-45 506/(RT)] s
13
3
-1 -1
+ H f C
f CH + H
+ H f CH
CO f CH + CO
2
H
2
+ H
2
37 ) 9.6 × 10 cm mol
s
a
14
-1
R38
R39
R40
R41
R42
R43
R44
R45
CH
CH
CH
CH
CH
CH
CH
CH
4
4
2
2
2
3
3
2
3
38 ) 7.8 × 10 exp[-103 826/(RT)] s
a
39 ) 2.244 × 10 (T) exp[-8763/(RT)] cm mol s-1
4
3
3
-1
3
+ H
2
14
-1
2
40 ) 3.0 × 10 exp[-70 943/(RT)] s
a
13
3
-1 -1
CO + H f CH
3
+ CO
41 ) 1.8 × 10 cm mol
s
13
3
-1 -1
+ CH
+ CH
+ CHO f CH
+ CH
2
f C
f C
2
H
H
2
+ H
+ H
2
42 ) 2.99 × 10 cm mol
s
43 ) 4.2221 × 10 cm mol s-1
1
3
3
-1
2
2
4
14
3
-1 -1
4
+ CO
44 ) 1.21 × 10 cm mol
s
12
3
-1 -1
2
f C
2
H
4
45 ) 1.0 × 10 cm mol
s
b
17
-1
R46
F-CH
2
-CH
2
-OH f HOF + C
+ CO
+ CO
OH f CH + FCHCH
OH f CH + FCHCH
OH f CH dCHF + OH
+ OH f CH OH
OH
OH f FCH CH
f C + F
2
H
4
46 ) 2.01 × 10 exp[-85 929/(RT)] s
R47
R48
R49
R50
R51
R52
R53
CH
2
2 2 4
+ CH CO f C H
47 ) 3.981 × 10 exp[-8365.2/(RT)] cm mol s-1
1
3
3
-1
13
3
-1 -1
2
C H
3
+ CHO f C
+ FC
+ FC
FCHCH
CH
2
H
4
48 ) 9.05 × 10 cm mol
s
11
3
-1 -1
CH
3
2
H
4
4
2
OH
OH
49 ) 3.98 × 10 exp[-9697.7/(RT)] cm mol
s
s
11
3
-1 -1
2
C H
5
2
H
4
4
2
50 ) 3.98 × 10 exp[-9697.7/(RT)] cm mol
11
-1
2
2
51 ) 6.19 × 10 exp[-23 647/(RT)] s
1
3
3
-1 -1
3
3
52 ) 6.023 × 10 cm mol
s
1
3
3
-1 -1
2
C H
5
+ OH f C
2
H
5
53 ) 7.709 × 10 cm mol
s
b
17
-1
R54
FCH
FCH
OH + FCH
OH + FCH
2
CH
CH
2
2
2
+ OH
54 ) 2.01 × 10 exp[-86 095/(RT)]s
13
-1
R55
R56
R57
R58
R59
R60
R61
R62
R63
R64
R65
R66
2
2
2
H
4
55 ) 3.90 × 10 exp[-21 660/(RT)]s
11
3
-1 -1
2
CH
2
OH f H
2
O + FCH
O + FCHCH
CHOH
OH f HF + FCHCH OH
CHOH
OH f HO + FCHCH OH
2
CHOH
56 ) 3.98 × 10 exp[-9697.7/(RT)] cm mol
s
OH f H
2
57 ) 3.98 × 10 exp[-13 697.7/(RT)] cm mol s-1
11
3
-1
2
CH
2
2
OH
11
3
-1 -1
F + FCH
F + FCH
O + FCH
O + FCH
F + OH f HF + O
HF + O f OH + F
2
CH
CH
2
OH f HF + FCH
2
58 ) 3.98 × 10 exp[-9697.7/(RT)] cm mol
s
59 ) 3.98 × 10 exp[-13 697.7/(RT)] cm mol s-1
11
3
-1
2
2
2
11
3
-1 -1
2
CH
CH
2
OH f HO + FCH
2
60 ) 3.98 × 10 exp[-9697.7/(RT)] cm mol
s
11
3
-1 -1
2
2
2
61 ) 3.98 × 10 exp[-13 697.7/(RT)] cm mol
s
1
2
3
-1 -1
62 ) 5.902 × 10 exp[-5683.3/(RT)] cm mol
s
1
0
3
-1 -1
63 ) 1.210 × 10 cm mol
s
O + O + Ar f O
F + F + Ar f F
FCH CHOH f F + CH
2
+ Ar
64 ) 2.74 × 10 exp[-0.971/(RT)] cm mol s-1
13
6
-2
14
6
-2 -1
2
+ Ar
65 ) 1.00 × 10 cm mol
s
11
-1
2
3
CHO
66 ) 6.19 × 10 exp[-23 647/(RT)] s
a
For reactions R15, R17, R25, R38, R39, and R41 revised rate constants are available. Reactions R16, R21, and R22 are negligible (refs 69-
7
4). See note added at the end of section III in text. b Reactions R54 and R55 can replace reaction R46 to explain the ethylene formation equally
well. Reactions R56-R66 are added to account for the radical reactions that are likely as the result of reactions R54 and R55. With our experimental
results, the two options, that is, direct HOF elimination [reaction R46] or successive bond dissociation [reactions R54 and R55], cannot be distinguished.

9786 J. Phys. Chem. A, Vol. 107, No. 46, 2003
Rajakumar et al.
which has been observed. Water elimination leads to the
formation of vinyl fluoride. By monitoring the concentrations
of C2H3F and CH3CHO at various temperatures, it is possible
to extract the rate constants for the H2O and HF elimination
reactions. However, the CH3CHO is consumed by secondary
reactions, as is evident from the formation of CH4, even at lower
temperatures. Hence, first the concentration of C2H3F was
followed to estimate the rate constant for H2O elimination, and
Figure 3a shows the Arrhenius plot. The rate constant for the
dominant HF elimination channel was then estimated by
monitoring the FEOH and CH3CHO. It involved kinetic
simulation with fewer steps and data points, and Figure 3b gives
the Arrhenius plot for this reaction. A detailed kinetic simulation
of all of the major and minor products and comparisons with
their experimentally observed profiles further refined and
confirmed these rate constants.
III.B. Minor Channels and Secondary Reactions. Methane
has been observed along with acetaldehyde, and it is most likely
formed by acetaldehyde decomposition. The FEOH cannot lead
to methane either directly or through any other simple radical
processes. Direct studies on the thermal decomposition of
34
acetaldehyde in reflected shock waves in the temperature range
of 1400-1800 K reveal that the primary step in the decomposi-
tion process is C-C bond dissociation; the activation energy
-
1
for the process is 81.9 kcal mol .
CH CHO f CH + CHO
3
3
15
-1
k ) 7.08 × 10 exp[-81 872/(RT)] s (R4)
The rate constant for this reaction at 970 K, at which temperature
-
1
CH4 has been noted, is 0.002 s . Hence, its contribution to the
formation of methyl radicals and CH4 is negligible. Also, the
fact that C2H6 has not been observed until T > 1100 K suggests
that CH3 radicals are not present in significant concentrations
below this temperature. Direct CH4 formation from CH3CHO
-
1
has been considered, and it is a high-barrier (>80 kcal mol )
3
5
process as well.
Figure 3. Arrhenius plot for the unimolecular elimination of (a)
-
1
2
hydrogen fluoride and (b) H O from 2-fluoroethanol (k is in s ).
CH CHO f CH + CO
(R5)
3
4
Vinyl fluoride decomposition behind the reflected shock
Because the preexponential factor for this molecular elimination
process should be significantly smaller than that of reaction R4,
this reaction is unlikely. However, CH3CHO that is produced
by the isomerization of CH2dCHOH is “chemically active”,
and the effective barriers for reactions R4 and R5 may be less.
3
9
waves has also been reported, and the activation energy is
nearly the same as that for acetaldehyde decomposition.
Unimolecular elimination of HF is the major channel through
which it decomposes to give acetylene.
36
-1
The enthalpy of formation for CH3CHO (-39.7 kcal mol )
CH dCHF f C H + HF
-
1
2
2
2
is 10 kcal mol less than that of CH2dCHOH (-29.8 kcal
1
4
-
1
37
mol ). However, the barrier for isomerization is estimated
k ) 2.398 × 10 exp[-81 400/(RT)] (R3)
3
-1
to be 55 kcal mol . The enthalpy of formation for FEOH is
not available to the best of our knowledge. From our MP2/6-
Even at the higher temperature range of this present study, the
-1
3
11++G** calculations, it is determined to be -109.4 kcal
rate constant for this reaction is 0.3 s . Hence, the contribution
of reaction R3 toward the formation of acetylene is expected
to be negligible. Acetylene was not observed in our experiments
at all, and this confirms that, within our experimental conditions,
these high-barrier reactions are not contributing. Evidently, vinyl
fluoride, which is directly formed through reaction R2, is not
“chemically active”.
-
1
mol . At this level, the FEOH f CH3CHO + HF reaction is
6
-
1
kcal mol exoergic. Considering the fact that the activation
energy for HF elimination from FEOH is determined to be 60
-
1
kcal mol in this work, it is clear that most of the available
energy from the HF elimination reaction will remain as internal
energy in the polyatomic product. Lifshtiz and Ben-Hamou
38
reported the thermal decomposition of ethylene oxide, the first
step of which was found to be isomerization to give chemically
active acetaldehyde. Coincidentally, they have observed the
same side products, namely, CH4, C2H4, and C2H6. Considering
the fact that CH3CHO was chemically active, they used lower
barriers for both reactions R4 and R5 to account for the
formation of these products in the lower-temperature range.
To account for the formation of these products, a detailed
kinetic simulation was carried out. In addition to reactions R1
and R2, the acetaldehyde pyrolysis steps were included.38
A
reaction scheme containing 27 species and 53 reactions for the
decomposition of 2-fluoroethanol was used for the simulation.
Table 2 contains all of the proposed reactions and their rate
expressions, along with the references. The scheme is composed

Thermal Decomposition of 2-Fluoroethanol
J. Phys. Chem. A, Vol. 107, No. 46, 2003 9787
Figure 4. Comparison between the experimental and calculated concentrations of (a) methane, (b) ethylene, (c) ethane, (d) vinyl fluoride,
e) acetaldehyde, and (f) 2-fluoroethane. Filled circles on the plot are the experimental concentrations, and the open circles are simulated concen-
(
trations.
of unimolecular dissociation of the reactant molecule, uni-
molecular dissociation of the radical intermediates, abstraction,
and recombination reactions. The rate constants were varied
within the limits of the reported uncertainties. The rate expres-
sions for reactions R1, R2, and R46 (vide infra) were evaluated
from our experimental results. The simulation was carried out
for the dwell time, and not surprisingly, all of the radicals were
not consumed within this time. The kinetic simulation was
extended for a few more milliseconds, but the rate constants
were evaluated at room temperature (298 K). We note again
that Zhang and Bauer have used a similar approach to model
nitromethane decomposition.33 Consumption of FEOH by reac-
tions with radicals, reactions R49 and R50, was included, but
it was found that these reactions did not play a significant role.
Simulation was carried for 20 ms, and it was noted that the
radicals were nearly completely consumed. The concentration
profiles for all of the products were compared to the experi-
mental results. Figure 4a-f shows the comparisons between
the concentrations of products based on analyses of post-
shock mixtures and those from kinetic simulation with the
reaction scheme given in Table 2. The solid symbols in the
figure are experimental points, and the open symbols show the
simulated values. The agreement between the experimental
concentrations and the simulated values is reasonable. All of
the steps that lead to methane, ethane, and ethylene are discussed
below.
1. Methane. Methane is formed directly from excited acet-
aldehyde by either direct molecular elimination or by
C-C dissociation followed by the reaction between CH3 and
HCO:
CH CHO f CH + HCO
3
3
13
-1
k ) 1.21 × 10 exp[-57 200/(RT)] s (R4)
4
CH CHO f CH + CO
3
4
1
2
-1
k ) 1.21 × 10 exp[-57 200/(RT)] s (R5)
5
CH + HCO f CH + CO
3
4
k ) 1.21 × 10 cm mol s-1 (R44)
1
4
3
-1
44
The reaction R44 competes with stabilization to CH3CHO
[reaction R8] and also CH3 radical recombination [reaction R15].
Some other reactions were considered in the reaction scheme
for the formation of CH4:

9788 J. Phys. Chem. A, Vol. 107, No. 46, 2003
Rajakumar et al.
CH + C H f CH + C H
Though the reaction steps R22, R23, R28, and R35 form ethane,
their contribution toward the formation of ethane is negligible
because all of these reactions depend on the ethyl radical
concentration, which is low. Reactions R18 and R19 are
included in the scheme in which ethane is consumed.
3
2
6
4
2
5
k ) 0.548(T) exp[-8285/(RT)] cm mol s-1
4.0
3
-1
1
8
(R18)
C H + CH f C H + CH
4
2
5
3
2
4
3
. Ethylene. Among all of the other products, ethylene
12
-0.5
cm mol s-1 (R26)
3
-1
k ) 9.8 × 10 (T)
2
6
concentration is the highest. About 4% of ethylene is found at
the highest temperature. Initially it was assumed that ethylene
could be the product from reactions of methyl and other radical
products. Hence the following reactions were included in the
scheme.
The contribution of these two reactions depends on the
concentration of the ethane, as well as methyl radicals.
Contributions from both of these reactions are small, but they
were included for completeness because both ethane and ethyl
are formed in subsequent steps.
CH + CH f C H + H
2
3
3
2
4
k ) 9.9 × 10 exp[-32 987/(RT)] cm mol s-1
1
5
3
-1
The reaction of CH3 with FEOH was considered because the
reactant is present in large concentrations. The rate constant
for this reaction is not available in the literature, and the rate
constant for hydrogen abstraction reaction from ethanol was
therefore used [reaction R49]. It was found that the FEOH
consumption was insignificant. For the CH3 + CH3CH2OH
reaction, it appears that only H abstraction mechanism has been
1
6
(R16)
CH + CH f C H + H
3
3
2
5
k ) 2.4 × 10 exp[-12 877/(RT)] cm mol s-1
13
3
-1
1
7
(R17)
40
considered. Both in ethanol and in FEOH, the C-O bond (∼90
C H f C H + H
-
1
2
5
2
4
kcal mol ) is weaker than the C-H bonds (∼90-100 kcal
-
1
41
8
1.19
-1
mol ). It appears that abstraction of OH could be important
in these reactions, but we are not aware of any reports
considering this possibility. In our experiments, methanol was
not observed, and it is likely that CH3 + FEOH reaction does
not contribute significantly.
k ) 4.8 × 10 (T) exp[-37 205/(RT)] s (R20)
2
0
C H + H f C H + H
2
2
5
2
4
k ) 1.81 × 10 cm mol s-1 (R21)
12
3
-1
2
1
Reaction of CH3 with CH3CHO and subsequent reactions are
included for completeness, and their contribution to CH4
formation is very small. The consumption of the methane formed
in the reaction scheme is taken into account in the following
reactions.
C H + C H f C H + C H
6
2
5
2
5
2
4
2
k ) 1.39 × 10 cm mol s-1 (R23)
1
2
3
-1
2
3
C H + CH f C H + CH
4
2
5
3
2
4
1
2
-0.5
cm mol s-1 (R26)
3
-1
k ) 9.8 × 10 (T)
2
6
CH f CH + H
4
3
14
-1
All of these reactions involve methyl and ethyl radicals. Out of
all of the above-mentioned reactions, reaction R16 is expected
to contribute more because the methyl radical concentration is
high when compared to ethyl and other radicals. However, it
was found that the ethylene concentration predicted was much
smaller than the experimental observation. The main reason is
that the rate constant for reaction R15 (CH3 recombination
leading to ethane) is 2-3 orders of magnitude larger than that
of reaction R16 or R17, which produce ethylene or ethyl
radicals, respectively.
k ) 7.8 × 10 exp[-103 826/(RT)] s (R38)
3
8
CH + H f CH + H
4
3
2
4
k ) 2.244 × 10 (T) exp[-8763/(RT)] cm mol s-1
3
3
-1
3
9
(R39)
Reaction R38 has very high activation energy, and it is very
unlikely in the present experimental conditions.
2. Ethane. The only major channel for the formation of ethane
is the recombination of methyl radicals. The following reactions
are included in the reaction scheme that can contribute to the
production of ethane.
The CH2 recombination to form ethylene is well-known. In
our experiment, CH2 can be formed through the following
reaction scheme:
CH + CH CHO f CH + CH CHO
CH + CH f C H
6
3
3
4
2
3
3
2
4
.0
1
4
-0.64
cm mol s-1 (R15)
3
-1
k10 ) 0.45T exp[-8280/(RT)] (R10)
CH CHO f CH CO + H
k ) 6.8 × 10 (T)
1
5
12
3
-1 -1
C H + H f C H
k ) 4.86 × 10 cm mol
s
2
2
2
5
2
6
22
(R22)
13
k ) 1.58 × 10 exp[-34 974/(RT)] (R13)
1
3
C H + C H f C H + C H
6
2
5
2
5
2
4
2
CH CO f CH + CO
k ) 1.39 × 10 cm mol s-1 (R23)
12
3
-1
2
2
2
3
14
k ) 3.0 × 10 exp[-70 943/(RT)] (R40)
4
0
C H + CHO f C H + CO
2
5
2
6
12
k ) 1.2 × 10 cm mol s-1 (R28)
14
3
-1
CH + CH f C H
4
k45 ) 1.0 × 10
(R45)
2
2
2
2
8
However, even with all of these reactions, not even 10% of the
total ethylene found experimentally could be accounted for. It
appeared that there must be a direct reaction from FEOH
C H + C H f C H + C H
3
2
4
2
5
2
2
6
3.13
2
k ) 6.32 × 10 (T) exp[-18 011/(RT)] cm mol s-1
3
-1
3
5
4
2
(R35)
producing ethylene. Klaassen et al. have observed HOI

Thermal Decomposition of 2-Fluoroethanol
J. Phys. Chem. A, Vol. 107, No. 46, 2003 9789
are 101
3.1(0.3
exp[(-59.1 ( 1.7)/(RT)] s
-1
and 1014.6(0.4
-
1
exp[(-71.3 ( 2.0)/(RT)] s . Obviously, more experimental
results are needed to verify the complete mechanism of
pyrolysis.
4. Propylene. The only channel through which propylene can
be formed using the present scheme is reaction R32].
C H + CH f C H + H
2
4
3
3
6
k ) 2.0 × 10 exp[-10 000/(RT)] cm mol s-1
1
3
3
-1
3
2
(R32)
We could not quantify propylene in the experimental analysis
because of its insignificant amounts. The present reaction
scheme was also predicting very small amounts of propylene.
Note Added in Revision. Since completion of this work, it
has been noted that rate constants for several steps, especially
reactions R17, R25, R38, R39, and R41 have been refined.
6
9
Sutherland, Su, and Michael have reported direct measure-
ments for reactions R38 and R39. For reaction R38, their rate
constant at 1100 K is an order of magnitude higher than what
we have used, but the contribution of R38 is negligible in our
experiments. Davidson and co-workers have compiled and
revised kinetic parameters for methane and ethane decomposi-
Figure 5. Arrhenius plot for the unimolecular elimination of HOF
from 2-fluoroethanol (k is in s-1).
formation from C2H5OI, which is obtained by reacting ethyl
3
iodide with atomic oxygen ( P). Leone and co-workers have
58
reported recently43 that HOI elimination competes with the
tion reactions, especially reactions R15 and R17. For reaction
R17, their value has been used, and it is in close agreement
3
elimination of HCl in the reaction of O( P) with CH2ICH2Cl.
7
0
with Lim and Michael. They have also shown that reaction
R16 is not important at all. The rate constant that we have used
for reaction R16 in comparison to reactions R15 and R17 is
much smaller and so this would not seriously affect our
conclusion. However, for reaction R15, the rate constant of
Davidson and co-workers is 2.2 times larger than that given in
Table 2, and we suspect that using these rate constants may
improve the agreement between simulation and experiment,
This led us to consider the possibility of the unimolecular
elimination of HOF from the reactant, that is, 2-fluoroethanol.
F-CH -CH -OH f HOF + C H
4
(R46)
2
2
2
When this reaction was added in the scheme, the formation of
ethylene in the thermal decomposition of 2-fluoroethanol could
be explained quantitatively. The rate constant for this reaction
was determined by simulation with fewer steps following FEOH,
CH3CHO, C2H3F, and C2H4. The Arrhenius plot for the
unimolecular elimination of HOF from 2-fluoroethanol, which
is obtained from the simulated fits, is shown in Figure 5. The
7
1
especially for C2H6. Michael and co-workers have reported
direct measurements for reaction R41, which is 5 times smaller
7
2-74
than what we have used. Hidaka et al.
have also reported
some revised rate constants for reactions R17, R39, and R41.
None of these rate parameters are directly determined in this
work, and they affect the minor products only. Hence, major
conclusions from this work, that is, rate constants for reactions
R1 and R2 are unlikely to be affected.
1
7.30(0.66
rate constant for this process is determined to be 10
-
1
exp[(-85.9 ( 3.4)/(RT)] s . The preexponential factor is
significantly larger than HF or H2O elimination. It is conceivable
that both C-F and C-O bonds weaken, which could lead to
several low-frequency stretching and bending modes at the TS.
However, the activation energy of nearly 86 kcal mol- suggests
that C-O bond dissociation reaction of FEOH cannot be ignored
as a possible source of ethylene. Hence, the following steps
were added instead of unimolecular elimination of HOF from
FEOH to explain the formation of ethylene.
IV. Computational Results and Discussions
1
Our main objective in the computational work was to identify
the transition states for HF and H2O elimination and estimate
the preexponential factor and activation energy for the processes
using conventional transition-state theory. For this reason, both
ground-state and the transition-state structures were fully
optimized at HF, MP2(FULL), and DFT(B3LYP) levels with
the standard 6-31G*, 6-31G**, 6-31+G**, and 6-311++G**
basis sets. Our attempts to get the transition state for HOF
elimination were not successful. The FEOH has five rotational
isomers due to single-bond rotation about both C-C and C-O
bonds, and there have been numerous discussions on their
FCH CH OH f FCH CH + OH
(R54)
(R55)
2
2
2
2
FCH CH f CH CH + F
2
2
2
2
The rate constant obtained for the C-O bond scission by fitting
1
7.32(0.6
the concentration of C2H4 is k54 ) 10
exp[(-86.1 ( 2.3)/
-
1
24-26
(
RT)]s . Reaction R55 is fast under our conditions, and it will
structures, vibrational frequencies, and relative energies.
2
4
be difficult to distinguish between these two possibilities as
source for C2H4. Not surprisingly, the fitted rate constants for
reactions R46 and R54 are nearly the same. Reactions R56-
R66 were also included in the mechanism to account for the
radical consumptions. With these steps added, instead of the
HOF elimination [reaction R46], the concentration profiles of
all of the products could be simulated equally well. After
addition of these steps, the rate constants for HF elimina-
tion and H2O elimination are refined marginally, and they
The Gg′ structure is reported to be the most stable structure
compared to other forms of 2-fluoroethanol, namely, Gg, Gt,
Tg, and Tt (G and T refer to gauche and trans geometry for
rotation about the C-C bond and g and t refer to gauche and
trans about the C-O bond. For gauche geometry about C-O
bond, g′ has the OH proton on the F side enabling hydrogen
bonding and g has the OH proton on the other side). The other
four conformers have more or less the same energy (within 1
-
1
kcal mol ) according to the best theoretical results available,

9790 J. Phys. Chem. A, Vol. 107, No. 46, 2003
Rajakumar et al.
HF level changes these distances in one direction (2.055 and
1
.250 Å), but inclusion of electron correlation brings them back
closer to the HF/6-31G** predictions. The C-F distance is
.983 Å at B3LYP/(6-311++G**) and 1.895 Å at MP2(FULL)/
1
(
(
6-311++G**). The C-H distance is 1.310 Å at MP2(FULL)/
6-311++G**) and 1.293 Å at B3LYP/(6-311++G**). The
H-F bond distances are very different at different level of
calculations. With 6-311++G** basis set, HF, MP2, and
B3LYP calculations lead to H-F distances of 1.390, 1.313, and
1
.362 Å. At all levels, these distances are significantly larger
Figure 6. Optimized structures of the transition state for (a) HF
elimination and (b) H O elimination from 2-fluoroethanol at B3LYP/
-311++G**.
than that of the free HF molecule (0.918 Å). However,
unimolecular HX elimination reactions do not lead to significant
energy release in the HX vibrational mode. All of these four
2
6
45
distances show very similar trend for HF elimination from ethyl
fluoride and trifluoroethane as well. It appears that the
transition states for HF elimination from ethyl fluoride and other
substituted fluoroethanes are very similar in nature.
all of them 3-4 kcal mol-1 higher than the Gg′ isomer depending
on the level of theory used. Because our main emphasis is on
the high-barrier chemical reactions, we consider the global
minimum Gg′ to represent the reactant. Initially, extensive
calculations were done for Tt and Gt conformers as well, and
our results are in good agreement with earlier published work.
To verify the â-substitution effect of F on the barrier for H2O
elimination, calculations were carried out for ethyl alcohol as
well. These results are compared with our experimental results.
The fully optimized structures of Gg′, Gt, and Tt conformers
are available on request from the authors, and they are available
44
B. Transition State for H2O Elimination. The TS for H2O
elimination has longer C-C and C-H bond distances and
shorter H-X and C-X distances (X ) O/F) compared to that
of HF elimination at all levels, though the difference is smaller
at the higher level calculations. These changes point to a subtle
difference in the four-center transition states for HF and H2O
elimination reactions. Table 3 gives a compact summary of these
distances. In particular, the C-H bond is significantly longer
at the TS for H2O elimination than that for the HF elimination.
As a result, the H-O bond is more fully formed in H2O
elimination than the H-F bond in HF elimination. It is clearly
2
4-26
elsewhere
as well. Our results are in complete agreement
with earlier reports at similar levels of theory. The complete
TS structures for HF elimination and H2O elimination are given
in Tables IS and IIS (Tables XS are to be found in Supporting
Information). To the best of our knowledge, there is no prior
report on these structures for comparison. The TS structures
46
shown in Figure 6. Butkovskaya and Setser have pointed out
such a trend between HF elimination and H2O elimination from
ethyl fluoride and ethanol, respectively. Our results show that
15
for ethyl fluoride (HF elimination) and ethanol (H2O elimina-
tion)2 have been reported, and they are used for comparison
with our results. Figure 6 shows structures of the transition states
of both HF and H2O eliminations at B3LYP/6-311++G** level.
The TS for HX elimination from ethyl chloride and ethyl
fluoride have an effective plane of symmetry that contains all
of the four atoms (H, C, C, and X) directly involved in the
2
-fluoroethanol is a good example to identify this trend within
2,23
the same molecule. In the past, the TS for HX elimination has
been characterized as “loose” or “tight” depending primarily
on the C-X distance. Looking at the C-O and C-F bond
distances, it is clear that the TS for H2O elimination is tighter
than that of HF elimination. It affects the preexponential factor
significantly and will be discussed in detail in the next section.
C. TST Calculations. Frequency calculations were carried
out at all levels of calculations for both the ground and transition
states to perform TST calculations. The ground-state frequencies
have been reported earlier by others and are also available from
the authors on request. Tables IIIS and IVS list the frequencies
at various levels for the transition states of HF and H2O
elimination. The transition states are characterized by one
imaginary frequency corresponding to the reaction coordinate.
The reaction coordinate corresponds to the motion of H away
from C toward F/O in both cases. The motion along the reaction
coordinate is very similar to that described by Holmes and co-
2
reaction, that is, the dihedral angle HCCX is zero. For FEOH,
the HF elimination TS is not planar with ∠ FCCH dihedral angle
around 6°-7° at all levels of theory. Similarly, for the TS for
H2O elimination, the ∠ OCCH dihedral angle is around 4°-5°
at HF and MP2 levels, but at DFT (B3LYP) level, it is 1°-3°.
It appears that the â-substitution of F/OH distorts the planar
TS slightly. Other structural details for both the TSs are
described separately next.
IV.A. Transition State for HF Elimination from FEOH.
Four bond distances are important at the TS for HF elimination,
that is, the newly partially formed CdC and H-X and the
breaking C-X and C-H bonds. These have been considered
extensively by several workers, and the recent paper by Holmes
42
workers for the HF elimination reaction of CF3CH3. The
47
motion of the reaction coordinate when visualized with Molden
44
and co-workers gives a good summary. At the TS, the C-C
bond distance at HF level is 1.393 Å with 6-31G** basis set,
and adding the diffuse functions (6-311++G**) decreases it
to 1.388 Å. Adding electron correlation results in an increase
in C-C bond length, and it is 1.400 Å at both MP2(FULL)/6-
clearly shows that the transition state corresponds to the reaction
of interest.
The thermal rate constants for both HF and H2O elimination
from 2-fluoroethanol were calculated using the transition -state
48
theory expression.
311++G** and B3LYP/6-311++G** levels. This C-C bond
distance corresponds to a bond order of ∼1.5, which is very
close to that of benzene. It is very similar to the TS found for
HF elimination from ethyl fluoride15 and trifluoroethane as
well. The C-F and C-H (the hydrogen that is involved in the
HF elimination) bond distances follow a similar trend. At the
lowest level of calculations reported here, that is, HF/6-31G**,
they are 1.927 and 1.303 Å, respectively. Improving the basis
set to triple-ú with polarization functions (6-311++G**) at the
kBT Q*
E0
RT]
k(T) ) l h QR exp -
(2)
[
44
where l is the reaction path degeneracy, kB is the Boltzmann
constant, Q* and QR are partition functions for the transition
state and reactant, respectively, and E0 is the zero-point barrier
for the reaction. The thermodynamic formulation of TST is then

Thermal Decomposition of 2-Fluoroethanol
J. Phys. Chem. A, Vol. 107, No. 46, 2003 9791
TABLE 3: Summary of C-C, C-X, C-H, and H-X Distances and Percent Changes Calculated for the Transition States for
HF/H
O Elimination Reactions from 2-Fluoroethanola
2
bond (HX)
HF/6-31G**
HF/6-311++G**
MP2/6-31G**
MP2/6-311++G**
B3LYP/6-31G**
B3LYP/6-311++G**
C-C (HF)
1.393 (-7.8)
1.442 (-4.6)
1.944 (+41.9)
1.702 (+21.7)
1.305 (+20.0)
1.588 (+46.1)
1.271 (+41.7)
1.100 (+16.9)
1.388 (-8.2)
1.457 (-3.4)
2.055 (+50.0)
1.629 (+16.4)
1.250 (+15.0)
1.645 (+51.3)
1.390 (+55.0)
1.078 (+14.6)
1.397 (-7.4)
1.412 (-6.4)
1.854 (+30.5)
1.803 (+26.9)
1.349 (+23.5)
1.494 (+36.8)
1.245 (+35.2)
1.196 (+24.9)
1.400 (-7.5)
1.412 (-6.7)
1.895 (+33.4)
1.787 (+25.8)
1.310 (+19.6)
1.497 (+36.7)
1.313 (+42.6)
1.195 (+24.9)
1.406 (-7.4)
1.421 (-6.5)
1.900 (+36.7)
1.845 (+29.9)
1.346 (+22.4)
1.467 (+33.4)
1.267 (+37.7)
1.227 (+27.5)
1.399 (-7.8)
1.411 (-7.0)
1.983 (+42.7)
1.900 (+29.9)
1.293 (+18.0)
1.412 (+28.8)
1.362 (+48.0)
1.268 (+31.8)
2
C-C (H O)
C-F (HF)
C-O (H O)
2
C-H (HF)
2
C-H (H O)
HF (HF)
2
HO (H O)
a
For C-C, C-X, and C-H bonds, the percent changes (in parentheses) give the change in these distances compared to the reactant. For the
O.
H-X bond, percent change gives the change compared to free HF and H
2
TABLE 4: Comparison of the Rate Parameters for the Unimolecular Elimination of HF and H
2
2
O from 2-Fluoroethanol and
H
O Elimination from Ethyl Alcohola
FCH CH
2
2
OH f HF + CH
3
CHO
2 2
FCH CH OH f H
2
O + CH
2
CHF
C
2 5
H
2
OH f H O + C
2 4
H
theory/basis set
log A
E
a
(kcal mol-1)
log A
E
a
(kcal mol-1
)
log A
E
a
(kcal mol-1
)
HF/6-311++G**
14.726 ( 0.006
14.635 ( 0.006
14.532 ( 0.005
13.178 ( 0.14
77.72 ( 0.03
65.64 ( 0.02
58.91 ( 0.02
59.45 ( 1.7
14.318 ( 0.005
14.721 ( 0.006
14.714 ( 0.006
14.302 ( 0.13
92.43 ( 0.02
75.52 ( 0.03
70.07 ( 0.03
69.69 ( 1.7
15.700 ( 0.006
14.448 ( 0.006
14.885 ( 0.006
13.84
85.95 ( 0.01
68.87 ( 0.01
64.61 ( 0.01
67.9
MP2(FULL)/6-311++G**
B3LYP/6-311++G**
b
Experiment
a
0 2
The E O elimination, the
values are 74.69, 62.61, and 56.13 kcal mol-1 at HF, MP2, and DFT levels for HF elimination, respectively. For H
q
-
1
-1
-1
corresponding values are 89.69, 72.17, and 66.80 kcal mol . The entropy of activation, ∆S , at 1100 K are 2.9, 2.5, and 2.01 cal K mol for
2
HF elimination at HF, MP2, and DFT levels, respectively. For H .
O elimination, the corresponding values are 1.0, 2.9, and 2.8 cal K mol-1
-
1
b
FEOH data is from present work and ethanol data is from ref 22. Butkovskaya et al.23 quote A ) 10 and E
) 69.2 kcal mol at 1100 K.
14.4
-1
a
used to estimate the Ea and A based on theory.48 The rate
constant for a unimolecular reaction is given by
for hydrogen fluoride elimination and 13 and 6 kcal mol-1 for
H2O elimination from 2-fluoroethanol, respectively.
The activation energy for HF elimination from ethyl fluoride49
is 59.5 ( 1 kcal mol . This estimate is also from single pulse
q
-1
k T
Ea
B
∆S
R
k ) le
exp
exp -
(3)
[
]
[
]
shock tube studies at the same temperature range (996-1137
K). It is nearly the same as that for FEOH, that is, OH
substitution does not seem to have increased the activation
energy for HF elimination. However, theoretical calculations
h
RT
q
Here, ∆S is the entropy of activation calculated using the
partition functions of reactant and the transition state. The A
and Ea were calculated between 1000 and 1200 K (at 50 K
intervals) using the HF, MP2(FULL), and DFT results using
the largest basis set, 6-311++G**, and average values for this
temperature range were estimated. These parameters could be
determined by empirically fitting the rate constants, k(T), to the
Arrhenius expression. The fit was excellent for all cases
suggesting that the Arrhenius expression is valid within this
temperature range, and it yielded results in agreement with eqs
-
1
at all three levels predict an increase of 1-4 kcal mol on OH
substitution. The activation energy for HF elimination from
CH3CH2F at HF, MP2, and DFT levels with 6-311++G** basis
-
1
75
sets are 74.1, 62.3, and 57.5 kcal mol , respectively. These
values can be compared to those for HF elimination from FEOH
-
1
at similar levels, 77.7, 65.6, and 58.9 kcal mol . Similarly,
between ethanol and FEOH, all levels of theory predict an
-
1
increase of 6-7 kcal mol in barrier for H2O elimination on F
substitution. Direct experimental results on the activation energy
for H2O elimination from ethanol are still unavailable, and the
results given in Table 4 are predictions from electronic structure
and TST calculations.
2
and 3. The same method is followed for H2O elimination from
ethanol. All of these results are compiled in Table 4. The
experimentally obtained values in the present work are also
included for the comparison.
At all levels, theory predicts the correct order for activation
energies for the two elimination reactions considered here. The
Ea for H2O elimination is significantly higher than that for HF
elimination. The agreement between experimental activation
energies for both HF and H2O elimination and the DFT
predictions appears too good to be true. It may be noted that
Holmes and co-workers have concluded that DFT level theory
gives good results for HF elimination reactions.4 In this
particular case, it should be emphasized that FEOH has several
The preexponential factors (A) estimated by all levels of
theory are significantly larger than the experimental observation
in our temperature range, especially for HF elimination reaction.
At the HF level of theory, the TS for HF elimination was
significantly looser than that of H2O elimination, and this is
evident from the preexponential factors listed in Table 4. For
HF elimination from CF3CH3, Holmes and co-workers pointed
out that the experimental and theoretical preexponential factors
are in good agreement at 800 K. Our results show that the
experimental value for HF elimination from FEOH is signifi-
cantly smaller than theoretical predictions at 1100 K. On the
other hand, it appears that the calculated preexponential factors
are in good agreement for H2O elimination at 1000-1200 K.
However, one needs to be very cautious in comparing the
experimental and theoretical preexponential factors. First, FEOH
has two internal rotors that have been treated as harmonic
oscillators in our model. Chuang and Truhlar have clearly shown
that the partition functions for the hindered rotors can vary by
4
4
4
-
1
other conformers that are higher in energy by 3 kcal mol . If
their contributions are included, the effective activation energy
calculated will be slightly smaller. For HCl elimination from
ethyl chloride and 1,2-dichloroethane, it was found that the DFT
2
predictions were underestimating the activation energy. Our
results on ethanol appear similar with the DFT calculations
underestimating the activation energy. In any case, the HF and
MP2(FULL) level theories differ from experimental values more
-1
significantly, overestimating the barrier by 18 and 6 kcal mol

9792 J. Phys. Chem. A, Vol. 107, No. 46, 2003
Rajakumar et al.
orders of magnitude depending on the model.50 At 1100 K,
though, harmonic oscillator partition functions are closer to the
partition functions calculated more rigorously. For example, for
CH2FCH2Cl at 1100 K, the partition functions are 5.9, 41.5,
and 8.8 for the harmonic oscillator, free rotor, and hindered
rotor models, respectively.
Institute of Science. We also acknowledge Dr. R. B Sunoj of
The Ohio State University, Columbus, OH, for fruitful discus-
sions in running Gaussian 94. Mr. D. Anandraj is acknowledged
for his help in the experiments. Prof. D. W. Setser is acknowl-
edged for comments on an earlier version. An anonymous
referee is acknowledged for critical comments that were helpful.
McGrath and Rowland have pointed out the importance of
51
tunneling in the HCl elimination reaction from ethyl chloride.
However, for comparison with experimental data between 700
Supporting Information Available: Four tables containing
the optimized transition-state structures for HF and H2O
elimination and the normal-mode frequencies for both transition-
state structures for HF, MP2(FULL), and DFT (B3LYP)
calculations with 6-31G*, 6-31G**, 6-31+G**, and 6-311++G**
basis sets. This material is available free of charge via the
Internet at http://pubs.acs.org. The completely optimized struc-
tures and the normal mode vibrational frequencies of both
ground and transition state for H2O elimination of ethyl alcohol
can be obtained from the authors on request.
5
2
and 800 K, they considered the rate constant directly and did
not try to fit the preexponential factor and activation energy
independently. The experimental activation energy was 56.3 (
-
1
0
.3 kcal mol , and they were able to fit the rate constants
accurately with a calculated classical barrier of 62.7 ( 0.4 kcal
-
1
mol . We note that the rate constant calculated at the MP2
level is in good agreement with the experimental value for HF
elimination from FEOH between 1000 and 1100 K because both
the preexponential factor and activation energy are larger
compared to experiment. For HF and HCl elimination from
CH2FCH2Cl also, similar agreement between experimental and
theoretical rate constants has been noted, though both A and Ea
References and Notes
(1) Kelly, T.; Manning, M.; Bonard, A.; Wenger, J.; Treacy, J.;
Sidebottom, H. In Transport and Chemical Transformation in the Tropo-
sphere, Proceedings of EUROTRAC Symposium 2000, 6th; Garmisch-
Partenkirchen, Germany, 27-31 March, 2000; Midgley, P. M., Reuther,
M., Williams, M., Eds.; Springer: Berlin, 2001; pp 410-413.
(2) Rajakumar, B.; Reddy, K. P. J.; Arunan, E. J. Phys. Chem. A 2002,
106, 8366.
7
5
differed significantly between theory and experiment. Cer-
tainly, there is a need for more experimental and theoretical
work on these reactions covering a wider temperature range for
proper understanding of the kinetic parameters.
(
3) Maccoll, A. Chem. ReV. 1969, 69, 33.
As mentioned earlier, our attempts to identify the transition
state for HOF elimination were not successful. Recently, Zhu
(4) Cadman, P.; Day, M.; Trotman-Dickenson, A. F. J. Chem. Soc. A
1
1
971, 1356.
(5) Evans, P. J.; Ichimura, T.; Tschuikow-Roux, E. Int. J. Chem. Kinet.
978, 10, 855.
76
and Bozzelli considered Cl2 elimination reaction from CH2Cl-
CH2Cl and determined the barrier to be very large, 91.5 kcal
-
1
75
(6) Weissman, M.; Benson, S. W. Int. J. Chem. Kinet. 1984, 16, 941.
mol at B3LYP/6-31++G** level. Our results on ClF
(
7) Jones, Y.; Holmes, B. E.; Duke, D. W.; Tipton, D. L. J. Phys. Chem.
-1
elimination from CH2F-CH2Cl gave a barrier of 145 kcal mol
1
990, 94, 4957.
at HF/6-311++G** level. These results imply that the HOF
elimination from FEOH is unlikely and C-O bond dissociation
may be the route to C2H4 formation. However, direct observation
or the lack of it) of HOF can resolve this question unambigu-
ously.
(8) Rakestraw, D. J.; Holmes, B. E. J. Phys. Chem. 1991, 95, 3968.
(
9) McDoniel, J. B.; Holmes, B. E. J. Phys. Chem. 1996, 100, 3044.
(
10) Srivatsa, A.; Arunan, E.; Manke, G., II; Setser, D. W.; Sumathi,
R. J. Phys. Chem. A. 1998, 102, 6412.
(11) Sudbo, Aa. S.; Schulz, P. A.; Shen, Y. R.; Lee, Y. T. J. Chem.
Phys. 1978, 69, 2312.
(
(
(
(
12) Quick C. R.; Wittig, C. J. Chem. Phys. 1980, 72, 1694.
13) Tsang, W. J. Chem. Phys. 1964, 41, 2487.
14) Skingle, D. C.; Stimson, V. R. Aust. J. Chem. 1976, 29, 609.
V. Conclusions
The thermal decomposition of 2-fluoroethanol has been
studied in the temperature range of 1000-1200 K. The
unimolecular elimination reactions of HF and H2O have been
reported both experimentally and theoretically. Both ab initio
and DFT methods have been employed to characterize all of
the transition states and ground states. The TST calculations
have been performed to obtain the rate parameters for all of
the unimolecular elimination processes. From the experimental
results, the rate constants for HF, H2O, and HOF eliminations
(15) Toto, J. L.; Pritchard, G. O.; Kirtman, B. J. Phys. Chem. 1994, 98,
359.
16) Chuchani, G.; Martin, I.; Rotinov, A.; Hernandez, A.; Reikonnen,
N. J. Phys. Chem. 1984, 88, 1563.
(17) Borisov, A. A.; Zamanskii, V. M.; Konnov, A. A.; Lisyanskii. V.
8
(
V.; Rusakov, S. A.; Skachkov, G. I. SoV. J. Chem. Phys. 1992, 9, 2527.
(18) Natarajan, K.; Bhaskaran, A. Proc. 13th Int. Shock Tube Symp.
1
981, 834.
19) Li, J.; Kazakov, A.; Dryer, F. L. Int. J. Chem. Kinet. 2001, 33,
859.
(
(20) Dunphy, M. P.; Patterson, P. M.; Simmie, J. M. J. Chem. Soc.,
Faraday Trans. 1991, 87, 2549.
13.17(0.33
are estimated to be 10
exp[-(59.5 ( 1.7)/(RT)],
exp[-(69.7 ( 1.7)/(RT)], and 1017.30(0.66 exp[-(85.9
3.4)/(RT)] respectively. The overall decomposition rate
(21) Marinov, N. M. Int. J. Chem. Kinet. 1999, 31, 183.
14.30(0.13
1
(
0
(22) Park, J.; Zhu, R. S.; Lin, M. C. J. Chem. Phys. 2002, 117, 3224.
(23) Butkovskaya, N. I.; Zhao, Y.; Setser, D. W. J. Phys. Chem. 1994,
13.55(0.32
98, 10779.
24) Gounev, T. K.; Bell, S.; Zhou, L.; Durig, J. R. J. Mol. Struct. 1998,
47, 21.
(25) Buemi, G. J. Chem. Soc., Faraday Trans. 1994, 90, 1211.
26) Dixon, D. A.; Smart, B. E. J. Phys. Chem. 1991, 95, 1609.
constant is given by 10
exp[-(60.6 ( 1.6)/(RT)]. The
(
activation energies calculated for HF elimination at HF,
MP2(FULL), and DFT methods with 6-311++G** basis set
differ from the experimental values by +18.2, +6.1, and -0.6
4
1
(
-
1
(27) McWhorter, D. A.; Hudspeth, E.; Pate, B. H. J. Chem. Phys. 1999,
10, 2000.
kcal mol , respectively. The activation energies for H2O
elimination at the same levels of theory differ from experimental
(
28) Green, D.; Hammond, S.; Keske, J.; Pate, B. H. J. Chem. Phys.
1999, 110, 1979.
29) Rasanen, M.; Murto, J.; Bondeybey, V. E. J. Phys. Chem. 1985,
9, 3967.
-
1
values by 22.7, 5.9, and 0.4 kcal mol . Our attempts to get the
transition-state structure for HOF elimination at all levels of
theories were not successful. Ethylene formation could be
explained by considering C-O bond dissociation equally well.
Direct real-time spectroscopic observation of HOF would be
needed to choose between the two pathways.
(
8
(30) Shirk, J. S.; Marquardt, C. L. J. Chem. Phys. 1990, 92, 7234.
(31) Brummel, C. L.; Mork, S. W.; Philips, L. A. J. Chem. Phys. 1991,
95, 7041.
(32) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T.; Petersson, G.
A.; Montgomery, J. A.; Raghavachari, K.; Al-Laham, M. A.; Zakrzewski,
V. G.; Ortiz, J. V.; Foresman, J. B.; Cioslowski, J.; Stefanov, B. B.;
Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala, P. Y.; Chen, W.;
Acknowledgment. We acknowledge the financial support
from IISc-ISRO Space Technology Cell and the Director, Indian

Thermal Decomposition of 2-Fluoroethanol
J. Phys. Chem. A, Vol. 107, No. 46, 2003 9793
Wong, M. W.; Andres, J. L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart, J. P.; Head-
Gordon, M.; Gonzalez, C.; Pople, J. A. Gaussian 94, revision C.2; Gaussian,
Inc.: Pittsburgh, PA, 1995.
(54) Tsang, W.; Hampson, R. P. J. Phys. Chem. Ref. Data 1986, 15,
1087.
(55) Baulch, D. L.; Cobos, C. J.; Cox, R. A.; Esser, C.; Frank, P.; Just,
Th.; Kerr, J. A.; Pilling, M. J.; Troe, J.; Walker, R. W.; Warnatz, J. J.
Phys. Chem. Ref. Data 1992, 21, 411.
(56) Colket, M. B., III; Naegeli, D. W.; Glassmann, I. Int. J. Chem.
Kinet. 1975, 7, 223.
(57) Kern, R. D.; Singh, H. J.; Wu, C. H. Int. J. Chem. Kinet. 1988, 20,
731.
(
(
33) Zhang, Y.-X.; Bauer, S. H. J. Phys. Chem. B 1997, 101, 8717.
34) Kern, R. D.; Singh, H. J.; Xie, K. AIP Conf. Proc. (Curr. Top.
Shock WaVes) 1990, 487.
(
(
35) Martell, J. M.; Yu, H.; Goddard, J. D. Mol. Phys. 1997, 92, 497.
36) Cioslowski, J.; Schimeczek, M.; Liu, G.; Stoyanov, V. J. Chem.
Phys. 2000, 113, 9377.
37) Suenobu, K.; Nagaoka, M.; Yamabe, T. J. Mol. Struct. (THEOCHEM)
999, 461-462, 581.
(
58) Davidson, D. F.; DiRosa, M. D.; Chang, E. J.; Hanson, R. K.;
Bowman, C. T. Int. J. Chem. Kinet. 1995, 27, 1179.
59) Warnatz, J. In Combustion Chemistry; Gardiner, W. C., Jr., Ed.;
Springer-Verlag: New York, 1984; p 197.
60) Frank, P.; Bhaskaran, K. A.; Just, Th. J. Phys. Chem. 1986, 90,
226.
(
1
(
(
(
(
(
38) Lifshitz, A.; Ben-Hamou, H. J. Phys. Chem. 1983, 87, 1782.
39) Cadman, P.; Engelbrecht, W. J. Chem. Commun. 1970, 453.
40) Gray, P.; Herod, A. A. Trans. Faraday Soc. 1968, 64, 1568.
41) McMillen, D. F.; Golden, D. M. Annu. ReV. Phys. Chem. 1982,
(
2
(
61) Frank, P.; Just, Th. Proc. Int. Symp. Shock Tubes and WaVes 1984,
3
3, 493.
1
4, 706.
(
42) Klaassen, J. J.; Linder, J.; Leone, S. R. J. Chem. Phys. 1996, 104,
(
(
(
62) Mackie, J. C.; Doolan, K. R. Int. J. Chem. Kinet. 1984, 16, 525.
63) Diau, E. W.-G.; Lee, Y.-P. J. Chem. Phys. 1992 96, 377.
64) Fagerstrom, K.; Lund, A.; Mahmoud, G.; Jodkowski, J. T.;
7
403.
(
43) Marcy, P. T.; Reid, J. P.; Qian, C. X. W.; Leone, S. R. J. Chem.
Phys. 2001, 114, 2251.
44) Martell, J. M.; Beaton, P. T.; Holmes, B. E. J. Phys. Chem. A 2002,
06, 8471.
Ratajczak, E. Chem. Phys. Lett. 1993, 208, 321.
(
(
(
(
(
(
65) Barat, R. B.; Bozzelli, J. W. J. Phys. Chem. 1992, 96, 2494.
66) Park, C. J. Phys. Chem. 1977, 81, 499.
67) Kondratev, V. N. Kinet. Catal. 1961, 8, 831.
68) Lloyd, A. C. Int. J. Chem. Kinet. 1971, 3, 39.
1
9
(45) Arunan, E.; Wategaonkar, S. J.; Setser, D. W. J. Phys. Chem. 1991,
5, 1539.
(
(
46) Butkovskaya, N. I.; Setser, D. W. J. Chem. Phys. 1996, 105, 8064.
47) Schaftenaar, G.; Noordik, J. H. Molden: a pre- and postprocessing
69) Sutherland, J. W.; Su, M.-C.; Michael, J. V. Int. J. Chem. Kinet.
2
001, 33, 669.
program for molecular and electronic structures. J. Comput.-Aided Mol.
Des. 2000, 14, 123.
(
(
70) Lim, K. P.; Michael, J. V. 25th Symp. (Int.) Combust. 1994, 713.
71) Hranisavljevic, J.; Kumaran, S. S.; Michael, J. V. 27th Symp. (Int.)
(
48) Wright, M. R. Fundamental Chemical Kinetics: An exploratory
introduction to the concepts; Horwood Publishing: Chichester, U.K., 1999.
49) Okada, K.; Tschuikow-Roux, E.; Evans, P. J. J. Phys. Chem. 1980,
4, 467.
Combust. 1998, 159.
72) Hidaka, Y.; Sato, K.; Henmi, Y.; Tanaka, H.; Inami, K. Combust.
(
(
Flame 1999, 118, 340.
8
(
(
(
50) Chuang, Y.-Y.; Truhlar, D. G. J. Chem. Phys. 2000, 112, 1221.
51) McGrath, M. P.; Rowland, F. S. J. Phys. Chem. A 2002, 106, 8191.
52) Heydtmann, H.; Dill, B.; Jonas, R. Int. J. Chem. Kinet. 1975, 7,
(73) Hidaka, Y.; Sato, K.; Hoshikawa, H.; Nishimori, T.; Takahashi,
R.; Tanaka, H.; Inami,; Ito, N. Combust. Flame 2000, 120, 245.
(74) Hidaka, Y.; Kimura, K.; Kawano, H. Combust. Flame 1994, 99,
18.
9
73.
(
53) Lifshitz, A.; Tamburu, C.; Suslensky, A. J. Phys. Chem. 1990, 94,
(75) Rajakumar, B.; Arunan, E. Phys. Chem. Chem. Phys. 2003, 5, 3897.
(76) Zhu, L.; Bozzelli, J. W. Chem. Phys. Lett. 2002, 357, 65.
2
966.