Thermal reactions of isodihydrobenzofuran: Experimental results and computer modeling

Article abstract of DOI:10.1021/jp003093f

The thermal reactions of isodihydrobenzofuran (phthalan) were studied behind reflected shock waves in a single pulse shock tube at 1050-1300 K and overall densities of ～ 3 × 10^-5 mol/cc. The total decomposition rate expressed as a first-order rate constant, was 10^12.10 exp(-53.7 x 10³/RT)/sec, where R is in given units of cal/(K mol). The isomerization product o-tolualdehyde and products resulting from unimolecular cleavage of the furan ring were obtained under shock heating. CO and toluene were the products of the highest concentration. Isobenzofuran was obtained by 1,6-H₂ elimination from the furan ring with a rate constant of k = 10^13.78 exp(-70 × 10³/RT)/sec. Benzene, styrene, ethylbenzene, ethylene, methane, and acetylene were found in the postshock mixtures. Trace amounts of allene and propyne were also observed. Phthalan decomposed via two unimolecular reactions involving cleavage of the furan ring, in addition to H-atom ejection from the furan ring. One reaction yielded stable products and one decomposition yielded unstable intermediates which were responsible for the propagation of free radical reactions in the system. A reaction scheme containing 26 species and 49 elementary reactions was constructed, and computer modeling was conducted in 25 K intervals at 1050-1300 K. The experimental results agreed well with the computed yields. Differences and similarities in the thermal behavior of the isodihydrobenzofuran, dihydrobenzofuran, 2,3-dihydrofuran, and 2,5-dihydrofuran, and the effect of fused benzene ring were elucidated.

Full text of DOI:10.1021/jp003093f

3
148
J. Phys. Chem. A 2001, 105, 3148-3157
Thermal Reactions of Isodihydrobenzofuran: Experimental Results and Computer
Modeling
Assa Lifshitz,* Aya Suslensky, and Carmen Tamburu
Department of Physical Chemistry, The Hebrew UniVersity of Jerusalem, Jerusalem 91904, Israel
ReceiVed: August 29, 2000; In Final Form: NoVember 15, 2000
The thermal reactions of isodihydrobenzofuran (phthalan) were studied behind reflected shock waves in a
-
5
single pulse shock tube over the temperature range 1050-1300 K and overall densities of ∼3 × 10 mol/
3
12.10
3
cm . The total decomposition rate, expressed as a first-order rate constant, is 10
exp(-53.7 × 10 /RT)
-1
16.50
s , where R is in given units of cal/(K mol). One isomerization product, o-tolualdehyde [kisomerization) ) 10
3
-1
exp (-80.5 × 10 /RT) s ], and products resulting from unimolecular cleavage of the furan ring were obtained
under shock heating. Carbon monoxide and toluene are the products of the highest concentration. Isobenzofuran
1
3.78
3
is obtained by a 1,6-H
2
elimination from the furan ring with a rate constant k ) 10
exp(-70.0 × 10 /RT)
-1
s . In addition, benzene, ethylbenzene, styrene, ethylene, methane, and acetylene were found in the postshock
mixtures. Trace quantities of allene and propyne were also found. In addition to H-atom ejection from the
furan ring, it is believed that phthalan decomposes via two unimolecular reactions that involve cleavage of
1
4.11
the furan ring. One reaction yields stable products, phthalan f C
H
6 5
-CH
3
+ CO [k ) 10
exp(-67.0 ×
3
-1
1
0 /RT) s ], and one decomposition yields unstable intermediates which are responsible for the propagation
•
•
16.6
3
-1
of free radical reactions in the system, phthalan f HCO + C
6
H
5
-CH
2
(k ) 10 exp(-78.0 × 10 /RT) s ).
A reaction scheme containing 26 species and 49 elementary reactions was constructed, and computer modeling
was performed in 25 K intervals over the range 1050-1300 K. The agreement between the experimental and
the calculated yields is reasonable. Differences and similarities in the thermal behavior of isodihydrobenzofuran,
dihydrobenzofuran, 2,3-dihydrofuran, and 2,5-dihydrofuran and the effect of fused benzene ring are discussed.
Introduction
We have recently published a detailed investigation on the
thermal reactions of 2,3-dihydrobenzofuran.
We present here a detailed investigation on the thermal
reactions of isodihydrobenzofuran (phthalan) and compare the
isomerization reactions and the decomposition channels of the
dihydrofurans to the dihydrobenzofurans.
1
7
The thermal reactions of the furan ring, with and without
substituents, have been the subject of extensive experimental
and theoretical (quantum mechanical) investigations. Tetrahy-
drofuran is a kinetically stable 5-membered ring compound that
undergoes ring cleavage with a high activation energy (∼83
kcal/mol) to yield mainly ethylene and ketene and formaldehyde
Experimental Section
1
and propylene. With the introduction of partial unsaturation to
the ring the two isomers 2,5-dihydrofuran and 2,3-dihydrofuran
are obtained. These molecules are considerably less stable than
tetrahydrofuran and decompose at much lower temperatures.
Apparatus. The thermal reactions of isodihydrobenzofuran
were studied behind reflected shock waves in a pressurized
driver 2 in. i.d single pulse shock tube. The tube, made of
electropolished stainless steel tubing, was heated and maintained
at 170 °C with an accuracy of (1 °C. The driven section was
4 m long and was divided in the middle by a 2 in. i.d. ball
valve. The driver had a variable length up to a maximum of
2.7 m and could be varied in small steps in order to tune for
the best cooling conditions. A 36-L dump tank was connected
to the driven section near the diaphragm holder in order to
prevent reflection of transmitted shocks and in order to reduce
the final pressure in the tube before collecting a sample for
analysis. The driven section was separated from the driver by
Mylar polyester film of various thickness depending upon the
desired shock strength.
After the tube was pumped down to approximately 10-5 Torr,
the reaction mixture was introduced into the section between
the ball valve and the end plate of the driven section, and pure
argon into the section between the diaphragm and the valve,
including the dump tank. After a shock was fired, gas samples
were transferred directly from the tube through an outlet in the
2,3
2,3-Dihydrofuran isomerizes and decomposes simultaneously
whereas the main reaction in 2,5-dihydrofuran is a H2 elimina-
4
,5
tion to form furan and no isomerization is observed. With an
additional double bond the furan molecule is formed. This is
again a very stable molecule that decomposes at higher
temperatures to yield mainly methylacetylene and carbon
6
-9
10
monoxide.
furan,
The thermal reactions of furonitrile, 2-methyl-
9
,11
12
and 2,5-dimethylfuran have also been investigated,
and the effects of the substituents on the reactions of the ring
were examined. Ruifeng Liu et al.13 and Bouchoux et al.
performed detailed ab initio quantum mechanical calculations
on the various decomposition channels of furan.
14
Recently, the decompositions of indene1 and indole16 were
investigated using the single pulse shock tube technique, to
examine the effect of fused benzene ring on the thermal reactions
of the 5-membered ring. In these two compounds benzene
replaces a double bond in cyclopentadiene and pyrrole. A
considerable stabilization of the 5-membered ring was evident.
5
1
0.1021/jp003093f CCC: $20.00 © 2001 American Chemical Society
Published on Web 02/22/2001

Thermal Reactions of Isodihydrobenzofuran
J. Phys. Chem. A, Vol. 105, No. 13, 2001 3149
driven section (near the end plate) into Hewlett-Packard model
5890A gas chromatographs using flame ionization (FID) detec-
tors. All the connecting tubes were heated to 170 °C.
Reflected shock temperatures were determined from the extent
of decomposition of 1,1,1-trifluoroethane, which was added in
small quantities to the reaction mixtures to serve as an internal
standard. Its decomposition to CH2dCF2 + HF is a first-order
unimolecular reaction that under the temperature and pressure
1
8
conditions of this investigation has a rate constant of kfirst )
1
4.85
3
-1
10
exp(-74.05 × 10 /RT) s . Reflected shock temperatures
were calculated from the relation
1
At
T ) - (E/R) ln - ln(1 - ø)
[
{
}
]
/
Figure 1. Gas chromatogram of a shock-heated sample of 0.3%
phthalan and 0.08% 1,1,1-trifluoroethane in argon, to 1215 K. The
numbers on the chromatogram are multiplication factors.
where t is the reaction dwell time, A and E are the preexponential
factor and the activation energy of the standard reaction, and ø
is the extent of decomposition defined as
1
,1,1-trifluoroethane and carbon monoxide was determined
periodically in order to verify a complete conversion of these
two compounds to methane in the methanyzer.
ø ) [CH dCF ] /([CH dCF ] + [CH CF ] )
2
2 t
2
2 t
3
3 t
Identification of the reaction products was based on their
retention times in the gas chromatograph. It was also assisted
by GC/MS measurements carried out on a Hewlett-Packard
model 5970 Mass Selective Detector. The sensitivities of the
reaction products were determined from standard mixtures. The
areas under the GC peaks were integrated with Spectra Physics
model SP4200 computing integrators and were transferred to a
PC for data reduction and graphical presentation.
Evaluation of the Product Concentrations. When a large
number of products are formed as a result of shock heating, the
evaluation of the absolute concentrations of the reaction products
from the area under their GC peaks is more complicated. To
avoid errors resulting from irreproducibility of the detectors,
concentrations of reaction products C5(pr)i were calculated
relative to the concentration of the reactant. Evaluation of the
concentrations of the individual products from their GC peak
areas were done using the following relations:
Density ratios behind the reflected shock waves were
calculated from the measured incident shock velocities using
the three conservation equations and the ideal gas equation of
state. The incident shock velocities were measured with two,
miniature, high-frequency-high-temperature pressure transduc-
ers (PCB model M113M199) placed 300 mm apart, near the
end plate of the driven section. The signals generated by the
shock waves passing over the transducers were fed through a
piezo amplifier to a Nicolet model 3091 digital oscilloscope.
Time intervals between the two signals shown on the oscil-
loscope were obtained digitally with an accuracy of (2 µs,
corresponding to approximately (15 K. A third transducer
placed in the center of the end plate of the driven section
provided measurements of the reaction dwell times (ap-
proximately 2 ms) with an accuracy of (5%. Cooling rates were
5
approximately 5 × 10 K/s.
Materials and Analysis. Reaction mixtures containing 0.3%
phthalan and 0.08% 1,1,1-trifluoroethane diluted in argon were
prepared and stored at 1 atm in 12-L glass bulbs. The bulbs
and the gas handling manifold were maintained at 170 °C and
C (pr) ) A(pr ) /S(pr ){C (reactant) /A(reactant) }
5
i
i t
i
5
0
0
C (reactant) ) {p (% reactant)(p /p ) /100RT }
-5
5
0
1
5
1
1
were pumped down to ∼10 Torr. The reactants were obtained
from Aldrich Chemical Co., Inc., and were of high purity. The
argon used was Matheson ultrahigh purity grade, listed as
A(reactant) ) A(reactant) + (1/N)Σ[n(pr )A(pr )/S(pr )]
0
t
i
i
i
9
9
9.9995%, and the helium was Matheson pure grade, listed as
9.999%. All the materials were used without further purifica-
C5(reactant)0 is the concentration of a reactant behind the
reflected shock prior to decomposition, A(reactant)0 is its
calculated GC peak area prior to decomposition, and N is the
number of its carbon atoms, 8 in isodihydrobenzofuran. A(pri)t
is the peak area of a product i in the shocked sample, S(pri) is
its sensitivity relative to reactant, and n(pri) is the number of
its carbon atoms. p5/p1 is the compression behind the reflected
shock wave, and T1 is the temperature of the reaction mixture
prior to shock heating, i.e. the temperature of the shock tube,
tion.
The gas chromatographic analyses of the products in the
postshock mixtures (except for CO) were performed on a
Porapak N column using a flame ionization detector. The
column’s initial temperature of 35 °C was gradually elevated
to 190 °C in analyses which lasted over 2 h. A typical
chromatogram of 0.3% phthalan in argon shock heated to 1215
K (not including CO) is shown in Figure 1. The Porapak N
column could not separate benzaldehyde from isobenzofuran.
They were determined quantitatively with the aid of a GC/MS
using different m/z peaks of their spectrum.
1
70 °C in the present study.
Results
To determine the distribution of reaction products, some 35
Carbon monoxide was analyzed on a 2 m molecular sieve
A column at 35 °C. It was reduced at 400 °C to methane prior
5
to its detection using a Chrompak methanyzer with a carrier
gas composed of 50% hydrogen and 50% argon. These analyses
gave the ratio [CO]/[CH3CF3]. From these ratios and the known
concentration of 1,1,1-trifluoroethane obtained in the Porapak
N analyses, the concentration of CO could be determined for
each run. The ratio [CO]/[CH3CF3] in a standard mixture of
tests were run with mixtures containing 0.3% isodihydroben-
zofuran and 0.08% 1,1,1-trifluoroethane in argon. The temper-
ature range covered was ∼1070-1300 K at overall densities
-
5
3
of (2.5-3) × 10 mol/cm . Details of the experimental
conditions and the distribution of reaction products are given
in Table 1 and are shown as an overview in Figure 2. Figure 2a

3150 J. Phys. Chem. A, Vol. 105, No. 13, 2001
Lifshitz et al.
TABLE 1: Experimental Conditions and Product Distribution (in mol %) in the Decomposition of Isodihydrobenzofuran
5
no.
T
5
10 C
5
phthalan CO
C
6
H
5
CH
3
o-tolualdehyde styrene
C
H
6 6
C
6
H
5
C
H
2 5
benzaldehyde isobenzofuran
C
2
H
4
C
2
H
2
CH
4
1
2
3
4
5
6
7
8
9
1066 2.89
1087 2.38
1098 2.86
1104 2.04
1104 2.66
1118 2.72
1122 2.70
1141 2.23
1141 2.63
1143 2.69
1143 2.63
1146 2.52
1153 2.30
1155 2.49
1159 2.55
1168 2.30
1171 2.15
1179 2.47
1181 2.12
1183 2.38
1200 2.39
1202 2.75
1211 2.43
1215 2.14
1226 2.48
1246 2.34
1246 2.58
1249 2.19
1252 2.29
1253 2.17
1269 2.42
1270 2.00
1293 2.03
1298 2.39
95.33
92.45
93.93
88.73
86.06
85.43
82.91
76.2
2.61
3.14
2.77
5.30
7.83
5.93
8.44
11.96
9.77
1.59
3.58
2.51
3.57
3.84
6.43
5.12
7.45
7.43
9.49
5.88
8.31
8.37
8.21
0.126
0.146
0.220
0.443
0.795
1.12
1.34
1.33
1.19
1.42
0.954
1.31
2.17
2.25
2.88
3.60
2.81
3.02
4.67
4.09
8.07
8.70
8.47
5.86
9.34
10.37
12.25
8.37
5.98
5.03
12.66
5.24
5.14
11.07
0.149 0.098
0.221 0.119
0.224 0.143
0.241 0.651
0.316 0.455
0.483 0.200
0.410 0.592
0.623 1.16
0.527 0.307
0.718 1.01
0.457 0.581
0.720 0.944
0.657 0.765
0.625 0.427
0.028
0.031
0.047
0.144
0.121
0.060
0.179
0.313
0.164
0.350
0.192
0.322
0.267
0.238
0.290
0.516
0.483
0.696
0.330
0.544
1.0
0.025
0.141
0.083
0.527
0.302
0.182
0.574
0.026
0.104
0.054
0.323
0.184
0.097
0.298
0.652
0.126
0.388
0.279
0.450
0.360
0.243
0.014
0.046
0.014
0.024
0.00065
0.070
0.038
0.031
0.060
0.159
0.035
0.106
0.050
0.085
0.081
0.053
0.061
0.134
0.152
0.060
0.035
0.078
0.155
80.20
0.255
0.783
0.563
0.909
0.716
0.482
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
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
74.85 10.62
81.72
0.190 0.081
0.091 0.041
0.157 0.079
0.136 0.065
0.059 0.041
0.065 0.046
0.153 0.093
0.241 0.104
0.276 0.126
0.142 0.051
0.172
9.19
74.63 12.09
72.89 13.52
76.55 10.82
73.34 11.69 10.33
58.39 17.57 15.23
61.05 19.62 11.49
55.30 20.53 14.90
0.80
1.14
0.490
1.17
1.30
1.14
1.40
1.18
2.60
1.20
0.902
1.50
2.24
2.08
1.76
1.57
1.49
2.33
2.73
1.97
3.74
2.59
2.15
0.718
0.638
0.844
0.724
1.62
0.880
0.674
1.22
1.93
1.98
1.96
1.74
1.69
2.70
3.21
2.72
5.11
4.12
3.71
0.995 1.29
1.33 1.58
0.687 0.684
0.474 1.44
73.55 12.01
68.89 12.69
5.88
7.48
0.095
0.20
48.81 15.55 20.42
53.55 17.69 15.29
45.23 19.02 19.09
26.93 30.42 24.91
49.09 17.30 15.89
27.67 27.69 23.12
27.29 26.71 23.56
2.29
1.57
2.20
2.92
2.29
3.26
3.0
2.27
3.27
3.71
3.37
3.17
4.37
3.97
1.34
0.886
1.72
2.44
1.19
2.24
2.01
1.81
3.01
4.19
2.64
5.05
4.85
3.55
0.138 0.108
0.723
0.974
1.304
0.868
1.19
0.241
0.411
0.178 0.161
0.402 0.240
0.416
0.20
0.218
0.301
0.335 0.111
0.474 0.138
0.574 0.135
0.974 0.423
0.432
2.016 0.913
2.187 0.728
1.491 0.387
1.23
30.15 32.0
20.36
1.04
19.71 33.33 27.46
10.56 39.31 28.13
17.05 29.67 28.16
12.79 37.31 21.81
6.89 37.66 30.01
8.73 33.80 29.94
1.51
1.45
0.273
1.33
2.25
0.605
0.349
1.11
1.20
monoxide and toluene are the products of the highest concentra-
tion. Both are believed to be formed in the reaction
However, the concentrations of carbon monoxide are higher than
those of toluene since there are other channels that produce CO,
•
•
particularly HCO f CO + H .
In addition to the products shown in Figure 2, trace quantities
of allene, propyne, ethane, C4H4, C4H2, and others were found
among the reaction products. Since their concentrations were
very minute, they are not reported as part of the general product
distribution.
Figure 3 shows an Arrhenius plot of the total decomposition
of isodihydrobenzofuran. The rate constants on the graph are
calculated as a first-order rate constants using the relation
ktotal ) -ln{[isodihydrobenzofuran]t/
[
isodihydrobenzofuran] }/t
0
The value obtained is ktotal ) 1012.10 exp(-53.7 × 10 /RT) s ,
where R is expressed in units of cal/(K mol). It should be
mentioned that the total decomposition of isodihydrobenzo-
furan is not a first-order process. The first-order rate constant
gives, however, a good estimate of the total decomposi-
tion rate over the temperature range covered in this investiga-
tion.
3
-1
Figures 4 and 5 show Arrhenius plots for two unimolecular
reactions that take place in the decomposition of isodihydroben-
zofuran: the isomerization of isodihydrobenzofuran to o-
tolualdehyde and the 1,6-H2 elimination from isodihydroben-
Figure 2. Product distribution: (a) without oxygen; (b) with oxygen.
shows all the reaction products that do not contain oxygen, and
Figure 2b, those that do contain oxygen. As can be seen, carbon

Thermal Reactions of Isodihydrobenzofuran
J. Phys. Chem. A, Vol. 105, No. 13, 2001 3151
TABLE 2: Arrhenius Parameters for the Decomposition of
Isodihydrobenzofuran
log
E
log
E
-
1
-1
compd
(A, s ) (kcal/mol)
compd
(A, s ) (kcal/mol)
tot. decomposn 12.10
53.7
48.4
49.6
73.2
59
benzaldehyde 12.03
isobenzofuran 13.65
ethylbenzene 12.40
60.2
69.8
64.2
73.7
CO
toluene
11.03
11.15
o-tolualdehyde 15.12
C
2
H
2
13.79
styrene
benzene
11.90
12.30
CH
4
61.8
C
2
H
4
11.25
60.6
It should be stressed again that both the preexponential factors
and the activation energies, which are obtained from the plots
for these two reactions, are lower than the true values owing to
further decomposition and the existence of parallel channels.
Values of E obtained from the slopes of the Arrhenius lines
for product formation and their corresponding preexponential
factors are summarized in Table 2 for all the products, including
those whose production is associated with free radical reactions.
For the latter, obviously, the Arrhenius parameters do not
correspond to the parameters of true first-order rate constants.
They do provide, however, a convenient way to summarize
general rates. Even for the unimolecular reactions of the reactant,
the parameters are not precisely those of the single reaction
owing to further decomposition and parallel production channels.
Figure 3. Arrhenius plot for the total decomposition of phthalan.
Discussion
Unimolecular and Free Radical Reactions. Among the
reactions that characterize the decomposition of isodihydroben-
zofuran, there are four unimolecular reactions that isodihy-
drobenzofuran undergoes. In all the four reactions the benzene
ring stays intact and changes occur only in the furan ring. Three
reactions are associated with cleavage of the furan ring, and in
one, molecular hydrogen is eliminated. There is one isomer-
ization reaction in which hydrogen migrates from position 3 to
position 1 in the ring and the C(1)-O(2) bond is broken to
yield o-tolualdehyde.
Figure 4. Arrhenius plot for production (formation minus decomposi-
tion) of the isomerization product o-tolualdehyde.
In another reaction, hydrogen molecule is eliminated from the
1
,3 positions in the furan ring in a 1,6-elimination process to
yield isobenzofuran.
There are two additional reactions in which the furan ring is
cleaved to yield two products: toluene + carbon monoxide and
•
•
benzyl + HCO .
Figure 5. Arrhenius plot for production (formation minus decomposi-
tion) of isobenzofuran.
zofuran to form isobenzofuran. Their rate constants are calculated
from the relation
[
product]t
kproduct
)
ktotal
[
reactant] - [reactant]t
0

3152 J. Phys. Chem. A, Vol. 105, No. 13, 2001
Lifshitz et al.
TABLE 3: Reaction Scheme for the Decomposition of Isodihydrobenzofuran

Thermal Reactions of Isodihydrobenzofuran
J. Phys. Chem. A, Vol. 105, No. 13, 2001 3153
TABLE 3 (Continued)
It should be added that these two reactions should not be
viewed as single-step reactions although they are pictured as
such in the figures. The reactions probably contain several
transition states and intermediates. Since, however, these are
not known, we inserted the reaction into the kinetic scheme as
a single-step reaction. The rate constants for the reactions in
the scheme are thus apparent rate constants. There is no way
that one can invent a sequence of pathways to accurately
describe the processes unless one runs quantum chemical
calculations. We plan to do so sometime soon.
Whereas the first three unimolecular processes yield stable
products, the fourth reaction yields unstable intermediates and
is responsible for the initiation of the free radical reactions in
3
the system. Ejection of an sp hydrogen atom from C(1) or C(3)
in the furan ring also initiates free radical processes. However,
without the introduction of the dissociation of isodihydroben-
•
•
zofuran to benzyl and HCO the calculated yields of almost all
of the products are orders of magnitude below the experimental
yields. Except for o-tolualdehyde, benzofuran, toluene, and
Figure 6. Calculated profiles of the important radicals in the
decomposition of isodihydrobenzofuran at 1150 K.
(partially) carbon monoxide, the production of all the other
products are associated with free radical reactions.
The detailed mechanism of the decomposition of isodihy-
drobenzofuran is reflected by the kinetic scheme shown in Table
•
The concentrations of HCO radicals, on the other hand, are by
3 orders of magnitude lower than those of benzyl although they
are formed in the same reaction (reaction 4). This results from
the extremely high decomposition rate of HCO to H and CO.
3. As has been mentioned before free radical reactions play an
•
•
important role in the decomposition. The main source for H
•
•
•
atoms is the very fast decomposition of HCO , which is formed
The concentrations of H and CH are the highest among the
3
by the unimolecular dissociation of isodihydrobenzofuran (reac-
tion 4 in Table 3) and to a much lesser extent by the ejection
of H atoms from the reactant (reaction 5). CH3 radicals are
obtained by a dissociative recombination of H atoms with
toluene and with o-tolualdehyde. The latter is also the reaction
that is responsible for the presence of benzaldehyde among the
decomposition products. Hydrogen atoms and methyl radicals
active free radicals and are involved in the production of many
products.
•
Reaction Scheme and Computer Modeling. To model the
observed product distribution we have constructed a reaction
scheme containing 26 species and 49 elementary reactions. The
scheme is shown in Table 3. The rate constants listed in the
n
table are given as k ) A exp(-E/RT) (or k ) A′T exp(-E/RT)
when the rate constant fits a wide temperature range) in units
3
-1
-1
of cm , s , kcal, and mol . The Arrhenius parameters for the
reactions in the scheme are either estimated or taken from
various literature sources, mainly from ref 19. The parameters
for the reactions that were taken from ref 19 are, in many cases,
best fits to a large number of entries. The thermodynamic
properties of the species involved were taken from various
are the main chain carriers of the free radical reactions. The
channel for production of ethylbenzene and styrene, for example,
begins with a methyl group attack on benzyl. Figure 6 shows
calculated profiles of radicals in the decomposition scheme of
isodihydrobenzofuran. Benzyl radicals being very stable and
products of the first generation are of the highest concentration.
2
0-22
literature sources
or estimated using ref 23 (Structure and
Properties program (SP)). We performed sensitivity analysis
0
with respect to variations (or rather uncertainties) in the ∆H f
of species whose thermodynamic properties were estimated or
are not known accurately enough. Incorrect values of the

3154 J. Phys. Chem. A, Vol. 105, No. 13, 2001
Lifshitz et al.
Figure 7. Calculated (lines) and experimental yields of o-tolualdehyde
and isobenzofuran. The calculations are done at 25 K intervals, marked
on the figure as crosses.
Figure 9. Calculated (lines) and experimental yields of carbon
monoxide, benzaldehyde, and methane. The calculations are done at
25 K intervals, marked on the figure as crosses.
Figure 8. Calculated (lines) and experimental yields of toluene,
benzene, and ethylene. The calculations are done at 25 K intervals,
marked on the figure as crosses.
thermodynamic functions result in erroneous values for the rate
constants of the back-reactions for a given value of the forward
rate constant. In several sensitivity tests that were performed
on uncertain values of heat of formation of various species, we
found that the results of the computer simulations were only
slightly sensitive to variations of ∼3 kcal/mol in the values of
0
the estimated ∆H f.
Figures 7-10 show experimental and calculated yields of 11
products found in the postshock mixtures. Products of trace
quantities were not dealt with in the scheme. The symbols in
the figures are the experimental points, and the lines are the
best fit to the calculated points using the scheme shown in Table
Figure 10. Calculated (lines) and experimental yields of acetylene,
styrene, and ethylbenzene. The calculations are done at 25 K intervals,
marked on the figure as crosses.
3
. The calculations were made at 25 K intervals and are shown
as + signs on the lines. The agreement for most of the reaction
products is satisfactory.
feature to many reactions is the decreased sensitivity as the
temperature increases. The number of channels that contribute
to the formation of the products increase as the temperature
increases and the effect of eliminating a single reaction
diminishes.
Table 4 shows that only a relatively small number of
elementary steps affect the product distribution in the sense that
their elimination from the scheme affects the yield of at least
Table 4 gives the sensitivity of the products to elimination
of specific reactions from the kinetic scheme, at 1100 and 1250
K, respectively. It gives the percent change in the yield of a
particular product as a result of eliminating a given reaction
from the scheme. The calculations correspond to dwell times
of 2 ms. Reactions that show an effect of less than 10% both at
1100 K and at 1250 K are not included in the table. A common

Thermal Reactions of Isodihydrobenzofuran
J. Phys. Chem. A, Vol. 105, No. 13, 2001 3155
TABLE 4: Sensitivity Spectrum at 1100/1250 K: Percent Change in the Yields for Elimination of a Reaction from the Scheme
one of the products. The majority of the elementary reactions
that compose the scheme do not affect the distribution at all.
They are left in the kinetic scheme for completeness and
applicability beyond the temperature range of the present
investigation. It should be mentioned, however, that the sensitiv-
ity analysis is performed by removing a single reaction at a
time and examining its effect. When a group of reactions is
eliminated from the scheme, there can be a strong effect on
particular products although the elimination of only one step,
as is shown in Table 4, has no affect at all. The general features
of the sensitivity spectrum are self-explanatory. Removal of
elementary steps that are the sole producers of a given product
or are part of a consecutive chain that is the only route for a
product formation reduces the concentration of that product to
zero. Examples are reaction 1 for the isomerization and for the
formation of benzaldehyde that is formed by dissociative
recombination of H atoms with the isomerization product.
Removing reaction 1 or 2 from the scheme results in the
disappearance of benzaldehyde from the collection of products.
Reactions 9 for methane and 17 for ethylbenzene are additional
example. As has been mentioned before, many of the products
are formed by free radical reactions. The most important reaction
that is responsible for the initiation of free radical reaction is
Differences and Similarities in the Thermal Behavior of
Isodihydrobenzofuran, Dihydrobenzofuran, 2,3-Dihydro-
furan, and 2,5-Dihydrofuran. Dihydrobenzofuran and isodi-
hydrobenzofuran are two structural isomers that differ from one
another in the location of the oxygen atom with respect to the
benzene ring. It is of interest to compare the thermal reactions
of these two isomers. It is also interesting to examine the role
of the fused benzene ring on the thermal behavior of the furan
ring. The thermal reactions of dihydrobenzofuran should thus
be compared to those of 2,3-dihydrofuran whereas the reactions
of isodihydrobenzofuran should be compared to those of 2,5-
dihydrofuran. In these two molecules the fused benzene ring is
removed from the molecule and is replaced by a CdC double
bond.
•
reaction 4 that produces HCO . The latter decomposes practically
•
instantaneously to CO and H (reaction 48). The removal of
reaction 4 from the scheme reduces almost completely the
concentration of all the products whose formation is associated
with free radical reactions.

3156 J. Phys. Chem. A, Vol. 105, No. 13, 2001
Lifshitz et al.
In both dihydrobenzofuran and isodihydrobenzofuran, the
benzene ring stabilizes the dihydrofuran part of the molecule
and the temperature range at which the molecules begins to
decompose (in the same reaction time of ∼2 ms) is considerably
higher, >100 K.
bonds connecting the benzene ring to position 1 in the furan
ring [oxygen atom in dihydrobenzofuran (C(9)-O(1)) and
carbon atom in phthalan (C(9)-C(1)] have a partial vinylic
character and are thus strong bonds (>100 kcal/mol), whereas
the 1-2 bonds in both molecules are much weaker and are much
more easily opened to undergo either isomerization or frag-
mentation.
An important feature by which the two isomers dihydroben-
zofuran and phthalan differ markedly from one another is the
number of channels involving unimolecular destruction of the
furan ring. This behavior is due to the fact that fragments
containing oxygen atom between two carbons, which are formed
during the process of phthalan decomposition, are very unstable
whereas the situation in dihydrobenzofuran where the oxygen
atom is at the end is considerably more stable. Thus, whereas
cleavage of the ring to form benzene and ketene, dihydroben-
zofuran f C6H6 + CH2CO, is a major decomposition channel
in dihydrobenzofuran, it does not occur at all in phthalan. No
ketene was found among the decomposition products in phthalan
decomposition. The main decomposition channel in phthalan
is phthalan f carbon monoxide + toluene. It should be
mentioned that the consecutive reactions of ketene are very
important in the decomposition mechanism of dihydrobenzo-
furan. In 2,3-dihydrofuran the main decomposition channel is
identical to that in dihydrobenzofuran where the products are
ethylene and ketene, 2,3-dihydrofuran f C2H4 + CH2CO.
Most of the products in the decomposition of the two isomers
are the result of consecutive free radical reactions that are
The 1-2 bond cleavage in both molecules accompanied by
H-atom migrations from C(3) to position 1 in the furan ring
results in an isomerization. In dihydrobenzofuran the H atom
is attached to O(1) to yield o-hydroxystyrene1
7
whereas in phthalan it is attached to C(1) to yield o-tolualdehyde.
The productions of the two different isomers in the two reactants
are similar processes, although preliminary DFT calculations
show that the details of the potential energy surfaces of the two
processes are different.24
In contrast to the identical isomerizations in dihydrobenzo-
furan and in isodihydrobenzofuran, the situation in the dihy-
drofuran isomers is entirely different. Whereas 2,3-dihydrofuran
undergoes isomerization, 2,5-dihydrofuran does not isomerize
•
initiated by the unimolecular production of HCO . Note that
the concentration of CO is higher than that of toluene,
particularly at high temperatures, although both are produced
in equal quantities by the direct decomposition of phthalan. The
reason for this behavior is that a considerable amount of the
4
,5
al all. Its main reaction is a 1,6-H2 elimination to form furan.
•
CO is produced by the very fast decomposition of HCO , leaving
after its decomposition a hydrogen atom that is the main chain
carrier in the system.
Acknowledgment. This work was supported by a grant from
the U.S.-Israel Binational Science Foundation under Grant
Agreement No. 94-00067. We wish to thank Dr. Wing Tsang
who served as the American cooperative investigator in this
reaearch.
As has been mentioned before, the equivalent process in
phthalan occurs also in addition to the isomerization, with
relatively high yield.
The isomerization process in 2,3-dihydrofuran is different
from the one in the dihydrobenzofurans. In the main isomer-
ization process, the furan ring is cleaved to yield a 3-membered
ring compound, cyclopropanecarboxaldehydele:2
References and Notes
(1) Lifshitz, A.; Bidani, M.; Bidani, S. J. Phys. Chem. 1986, 90, 3422.
(
(
(
(
(
(
2) Lifshitz, A.; Bidani, M. J. Phys. Chem. 1989, 93, 1139.
3) Lifshitz, A.; Laskin, A. J. Phys. Chem. 1994, 98, 2341.
4) Lifshitz, A.; Bidani, M.; Bidani, S. J. Phys. Chem. 1986, 90, 6011.
5) Rubin, J. A.; Filseth, S. V. J. Chem. Educ. 1969, 46, 57.
6) Lifshitz, A.; Bidani, M.; Bidani, S. J. Phys. Chem. 1986, 90, 5373.
7) Organ, P. P.; Mackie, J. C. J. Chem. Soc., Faraday Trans. 1991,
8
7, 815.
8) Bruinsma, O. S. L.; Tromp, P. J. J.; de Sauvage Nolting, H. J. J.;
Moulijn, J. A. Fuel 1988, 67, 334.
9) Grela, M. A.; Amorebieta, V. T.; Colussi, A. J. J. Phys. Chem.
(
When a benzene ring is fused to the furan ring, the equivalent
species that is obtained in the isomerization is unstable. In a
minor isomerization process in 2,3-dihydrofuran a small amount
of crotonaldehyde is also formed but the main yield of the latter
is obtained by further isomerization of cyclopropanecarboxal-
dehydele to crotonaldehyde, in a process similar to the cyclo-
propane f propylene isomerization.
(
1985, 89, 38.
(10) Laskin, A.; Lifshitz, A. Shock Waves. In Proceedings of the 20th
International Symposium on Shock Tubes and WaVes, 1995; Sturtevant, B.,
Sheferd, J. E., Hornung, H., Eds.; World Scientific: Singapore, London,
1
1
1
996; p 971.
(11) Lifshitz, A.; Tamburu, C.; Shashua, R. J. Phys. Chem. A 1997,
01, 1018.
(12) Lifshitz, A.; Tamburu, C.; Shashua, R. J. Phys. Chem. A 1998,
02, 10655.
(
(
13) Liu, R.; Zhou, X.; Zhai, L. J. Comput. Chem. 1998, 19 (2), 240.
14) Bouchoux, G.; Dagaut, J.; Fillaud, J.; Burgers, P. C.; Terlouw, J.
K. NouV. J. Chem. 1985, 9 (1), 25.
15) Laskin, A.; Lifshitz, A. Thermal decomposition of indene. Experi-
mental results and computer modeling. Proc. Combust. Inst. 1998, 27,
13.
(
3
(
(
16) Laskin, A.; Lifshitz, A. J. Phys. Chem. A 1997, 101, 7787.
17) Lifshitz, A.; Suslensky, A.; Tamburu, C. Proc. Combust. Inst. 2000,
A marked difference between the dihydrofurans and the
benzodihydrofurans is their kinetic stability. As expected, the
in press.

Thermal Reactions of Isodihydrobenzofuran
J. Phys. Chem. A, Vol. 105, No. 13, 2001 3157
(
18) Tsang, W.; Lifshitz, A. Int. J. Chem. Kinet. 1998, 30, 621-628.
(25) Robaugh, D.; Tsang, W. J. Phys. Chem. 1986, 90, 4159.
(26) Tsang, W.; Hampson, R. F. J. Phys. Chem. Ref. Data 1986, 15,
(19) Westly, F.; Herron, J. T.; Cvetanovic, R. J.; Hampson, R. F.;
Mallard, W. G. NIST-Chemical Kinetics Standard Reference Database 17,
Ver. 5.0; National Institute of Standards and Technology: Washington, DC.
1087.
(27) Tsang, W. Combust. Flame 1989, 78, 71-86.
(28) Brand, U.; Hippler, H.; Lindemann, L.; Troe, J. J. Phys. Chem.
1990, 94, 6305.
(
20) Pedley, J. B.; Taylor, R. D.; Kirby, S. P. Thermochemical Data of
Organic Compounds; Chapman and Hall: London, 1986.
21) Burcat, A.; McBride, B.; Rabinowitz, M. Ideal Gas Thermodynamic
(
(29) Muller-Markgraf, W.; Troe, J. J. Phys. Chem. 1988, 92, 4914.
(30) Fahr, A.; Laufer, A.; Klein, R.; Braun, W. J. Phys. Chem. 1991,
95, 3218-3224.
Data for Compounds Used in Combustion; TAE 657 report; Technion-
Israel Institute of Technology: Haifa, Israel, 1997.
(
22) Warnatz, J. Rate coefficients in the C/H/O system. In Combustion
Chemistry; Gardiner, W. C., Jr., Ed.; Springer-Verlag: New York, 1984.
23) Stein, S. E.; Rukkers, J. M.; Brown, R. L. NIST-Standard Reference
(31) Kiefer, J. H.; Mizerka, L. J.; Patel, M. R.; Wei, H. C. J. Phys.
Chem. 1985, 89, 2013-2019.
(
(32) 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, 411-429.
Database 25; National Institute of Standards and Technology: Washington,
DC.
(24) Dubnikova, F.; Lifshitz, A. Unpublished results.
(33) Clark, W. D.; Price, S. J. Can. J. Chem. 1970, 48, 1059.