Unimolecular reactions of CH2BrCH2Br, CH 2BrCH2Cl, and CH2BrCD2Cl: Identification of the Cl-Br interchange reaction

Article abstract of DOI:10.1021/jp9116134

The recombination reactions of CH₂Br and CH₂Cl radicals have been used to generate vibrationally excited CH ₂BrCH₂Br and CH₂BrCH₂Cl molecules with 91 kcal mol^-1 of energy in a room-temperature bath gas. The experimental unimolecular rate constants for elimination of HBr and HCl were compared to calculated statistical rate constants to assign threshold energies of 58 kcal mol^-1 for HBr elimination from C₂H ₄Br₂ and 58 and 60 kcal mol^-1, respectively, for HBr and HCl elimination from C₂H₄BrCl. The Br-Cl interchange reaction was demonstrated and characterized by studying the CH ₂BrCD₂Cl system generated by the recombination of CH ₂Br and CD₂Cl radicals. The interchange reaction was identified from the elimination of HBr and DCl from CH₂ClCD ₂Br. The interchange reaction rate is much faster than the rates of either DBr or HCl elimination from CH₂BrCD₂Cl, and a threshold energy of ?43 kcal mol^-1 was assigned to the interchange reaction. The statistical rate constants were calculated from models of the transition states that were obtained from density functional theory using the B3PW91 method with the 6-31G(d′,p′) basis set. The model for HBr elimination was tested versus published thermal and chemical activation data for C₂H₅Br. A comparison of Br-Cl interchange with the Cl-F and Br-F interchange reactions in 1,2-haloalkanes is presented.

Full text of DOI:10.1021/jp9116134

4138
J. Phys. Chem. A 2010, 114, 4138–4147
Unimolecular Reactions of CH2BrCH2Br, CH2BrCH2Cl, and CH2BrCD2Cl: Identification of
the Cl-Br Interchange Reaction
Laura Friederich,† Juliana R. Duncan,† George L. Heard,† D. W. Setser,‡ and
Bert E. Holmes*,†
Department of Chemistry, UniVersity of North Carolina-AsheVille, One UniVersity Heights,
North Carolina 28804-8511 and Department of Chemistry, Kansas State UniVersity, Manhattan, Kansas
ReceiVed: December 7, 2009; ReVised Manuscript ReceiVed: February 12, 2010
The recombination reactions of CH2Br and CH2Cl radicals have been used to generate vibrationally excited
CH2BrCH2Br and CH2BrCH2Cl molecules with 91 kcal mol-1 of energy in a room-temperature bath gas. The
experimental unimolecular rate constants for elimination of HBr and HCl were compared to calculated statistical
rate constants to assign threshold energies of 58 kcal mol-1 for HBr elimination from C2H4Br2 and 58 and 60
kcal mol-1, respectively, for HBr and HCl elimination from C2H4BrCl. The Br-Cl interchange reaction was
demonstrated and characterized by studying the CH2BrCD2Cl system generated by the recombination of CH2Br
and CD2Cl radicals. The interchange reaction was identified from the elimination of HBr and DCl from
CH2ClCD2Br. The interchange reaction rate is much faster than the rates of either DBr or HCl elimination
from CH2BrCD2Cl, and a threshold energy of =43 kcal mol-1 was assigned to the interchange reaction. The
statistical rate constants were calculated from models of the transition states that were obtained from density
functional theory using the B3PW91 method with the 6-31G(d′,p′) basis set. The model for HBr elimination
was tested versus published thermal and chemical activation data for C2H5Br. A comparison of Br-Cl
interchange with the Cl-F and Br-F interchange reactions in 1,2-haloalkanes is presented.
I. Introduction
The experimental rate constants for molecules with 91 kcal
mol-1 of vibrational energy were compared to calculated
statistical (RRKM) unimolecular rate constants to assign
threshold energies (E0) to HCl and HBr elimination and to
Cl-Br interchange. The channel with the lowest threshold
energy for CH2BrCH2Cl is actually the Cl-Br interchange. The
RRKM rate constants are based upon density functional theory
(DFT) calculations for the vibrational frequencies and moments
of inertia of the molecule and transition state. We have continued
to use the B3PW91 method with either the 6-31G(d′,p′) or
6-311+G(2d,p) basis sets.3-10,12 The complex torsional mode
of these molecules was treated as an asymmetric, hindered,
internal rotation for the calculation of the rate constants.10 The
rate constants for DCl and DBr elimination from CH2BrCD2Cl
and CH2ClCD2Br were calculated from the models developed
for CH2BrCH2Cl.
Because DFT calculations for transition states involving HBr
elimination or other processes involving Br-atoms have not been
extensively explored,9,13,14 we decided to test the method that
our laboratory has employed for Cl-F interchange and HCl
and HF elimination reactions by comparing calculated rate
constants to published experimental data for the C2H5Br reaction.
Although some uncertainty may exist for the experimental
thermal15-18 and chemical activation11 rate constants, the C2H5Br
reaction can serve as a baseline test for the transition-state
structure for HBr elimination. Comparison also can be made
for trends in the C2H5F, C2H5Cl, and C2H5Br series. For
example, the threshold energies are known to decline by about
3 kcal mol-1 for each member in the series, that is, 58, 55, and
52 kcal mol-1. Modest agreement between experimentally
determined Arrhenius pre-exponential factors and those from
electronic-structure calculations exist for HF elimination when
the torsional mode(s) are treated as hindered internal rotations.10,12
However, the calculated transition-state models for HCl elimina-
The Cl-F interchange reaction for chlorofluoroalkane mol-
ecules with adjacent fluorine and chlorine atoms has been
characterized for several fluorochloroethanes,1,2 -propanes,3-7
and -butanes.8 The threshold energies are in the 60-70 kcal
mol-1 range. The Br-F interchange reaction also has been
characterized for CF2BrCF2CH3 for comparison to Cl-F
interchange in similar compounds.9 We now have turned our
attention to the Cl-Br interchange reaction in 1,2-chlorobro-
moethane. The chemical activation technique was used to
generate CH2BrCH2Br, CH2BrCH2Cl, and CH2BrCD2Cl mol-
ecules with 91 kcal mol-1 of vibrational energy by the
recombination reactions of CH2Br and CH2Cl (CD2Cl) radicals
at room temperature. These radicals were obtained from the
photolysis of CH2BrI and CH2ClI (CD2ClI). The Cl-Br
interchange was detected from the conversion of CH2BrCD2Cl
into CH2ClCD2Br. Just as the Cl-F interchange reaction was
in competition with HCl and HF elimination reactions for the
fluorochloroalkanes, Cl-Br interchange is in competition with
HBr (DBr) and HCl (DCl) elimination reactions. Therefore, the
CH2BrCH2Br and CH2BrCH2Cl systems were used to character-
ize the HBr and HCl elimination processes as a necessary
part of the study of Cl-Br interchange. The HCl elimination
reaction from CH2BrCH2Cl also can be compared to that of
CH2ClCH2Cl10 to gain more information about the unimolecular
reactions of 1,2-dihaloethanes.10 The chemically activated
CH2BrCH2Br and CH2BrCH2Cl molecules have been studied
previously,11 but the chemical system was complicated, and the
new data are an improvement.
* To whom correspondence should be addressed.
† University of North Carolina-Asheville.
‡ Kansas State University.
10.1021/jp9116134  2010 American Chemical Society
Published on Web 03/10/2010

Identification of the Cl-Br Interchange Reaction
J. Phys. Chem. A, Vol. 114, No. 12, 2010 4139
tion may overestimate the Arrhenius pre-exponential factors by
a factor of 2,10,12 which is near the limit of the experimental
uncertainty given the limited temperature ranges of the indi-
vidual investigations. Thus, it is of interest to extend the
comparison to the HBr elimination reaction of C2H5Br.
The experiments with C2H4Br2 utilized the photolysis19 of
CH2BrI to generate CH2Br radicals in a clean environment; C4F10
was added to achieve higher pressures. The recombination
reaction gives CH2BrCH2Br*; the asterisk denotes vibrational
excitation.
CH2ClCD2Br* f HBr + CHCldCD2 (D1)
f DCl + CH2dCDBr (D2)
f CH2BrCD2Cl*
(4e)
(4f)
(4g)
(4h)
+ SF6 f CH2ClCD2Br (S)
The recombination reactions also give C2D4Cl2* and
C2H4Br2*, and their decomposition products must be considered
in the analysis of vinyl chloride and vinyl bromide. The
decomposition of CH2BrCD2Cl* gives DBr (+ CH2dCDCl) and
HCl (+ CD2dCHBr), whereas CH2ClCD2Br* gives HBr (+
CD2dCHCl) and DCl (+ CH2dCDBr) and each channel can
be identified from the vinyl chloride-d1 or -d2 and vinyl bromide-
d1 or -d2 by gas chromatography with mass spectrometric
detection. In anticipation of the results to be presented, the
Cl-Br interchange reaction is ≈8 times faster than the sum of
the elimination reactions of CH2BrCD2Cl.
CH2Br + CH2Br f CH2BrCH2Br*
CH2BrCH2Br* f HBr + CH2dCHBr (D)
+ C4F10 f CH2BrCH2Br (S)
(1a)
(1b)
(1c)
The threshold energy for Br atom dissociation is 70 kcal
mol-1, and dissociation is not competitive with HBr elimination
from CH2BrCH2Br molecules with 91 kcal mol-1 of energy.
Reaction 1b is 20 kcal mol-1 endothermic, and CH2dCHBr will
be stabilized by collisions for the pressures of our experiments.
The experiments for CH2BrCH2Cl (and CH2BrCD2Cl) utilized
the photolysis of CH2BrI and CH2ClI (CD2ClI) in SF6 bath
gas. The radical recombination reactions are 1, 2, and 3.
II. Experimental Methods
The CH2BrI and CH2ClI were purchased from Acros and
CD2ClI was purchased from CDN Isotopes with a stated isotopic
purity of 99.9% D. The C4F10 and SF6 were both obtained from
PCR (now SynQuest). The reagents and bath gases were
measured in calibrated volumes on a grease-free vacuum line
and transferred cryogenically to Pyrex vessels ranging in size
from 3.88 to 24.88 cm3 for photolysis. The ratios of bath gas to
reagent were g50. Some experiments were done in larger
vessels to achieve low pressure for the experiments with
CH2BrCD2Cl. A small amount of mercury(I) iodide was added
to each vessel to aid in the generation of radicals22 and to remove
HCl and HBr. Pressures were measured with a MKS electronic
manometer. Each vessel was irradiated for about 5 min with
the output of a 200 W high-pressure mercury lamp. The Pyrex
vessel acted as a filter to limit the effective photolysis
wavelength to 290-330 nm. In this wavelength range the upper
electronic state dissociates only by rupture of the CH2Br-I and
CH2Cl-I bonds.19
After photolysis, the samples for the CH2BrCH2Br or the
CH2BrCH2Cl systems were analyzed for products with a
Shimadzu 14A gas chromatograph fitted with a 105 m RTX-
VGC column of 0.53 mm diameter. For measurement of
decomposition to stabilization ratios (D/S), a flame-ionization
detector (GC-FID) was used. The temperature program with
the GC-FID for analysis of both the C2H4Br2 and CH2BrCH2Cl
systems was 20 min at 30 °C followed by a temperature increase
of 6 °C/min until the final temperature of 180 °C was attained.
The characteristic retention times (in minutes) were 11.1 for
CH2dCHCl, 14.9 for CH2dCHBr, 39.5 for C2H4ClBr, and 44
for C2H4Br2. The problem of interference in the measurements
of vinyl chloride or vinyl bromide from reactions 1b or 3b in
the study of C2H4ClBr was avoided by using large starting ratios
of CH2BrI/CH2ClI when observing vinyl chloride and large
ratios of CH2ClI/CH2BrI when observing vinyl bromide. A
similar approach has been described10 for the CH2FCH2Cl
system, and more details are provided in Section III-A.
Calibration factors for the GC-FID were experimentally deter-
mined from prepared mixtures of CH2dCHCl, CH2dCHBr,
C2H4ClBr, C2H4Cl2, and C2H4Br2. The factors were 1.10 ( 0.12
(CH2dCHBr/C2H4Br2), 1.09 ( 0.11 (C2H3Cl/C2H4ClBr), and
1.11 ( 0.12 (C2H3Br/C2H4ClBr), which were applied to the
experimentally measured ratios to obtain the actual ratios of
concentrations. The SF6 is not sensed by the GC-FID, the
CH2Br + CH2Cl f CH2BrCH2Cl*
CH2BrCH2Cl* f HBr + CH2dCHCl (D1)
f HCl + CH2dCHBr (D2)
(2a)
(2b)
(2c)
(2d)
+ SF6 f CH2BrCH2Cl (S) + SF6
CH2Cl + CH2Cl f CH2ClCH2Cl*
CH2ClCH2Cl* f HCl + CH2dCHCl (D)
+ SF6 f CH2ClCH2Cl (S) + SF6
(3a)
(3b)
(3c)
The unimolecular rate constants were measured relative to
collisional deactivation. The slopes from plots of the ratio of
decomposition to stabilization products (D/S) versus pressure-1
for a range of pressure such that D/S e 1.0 give rate constants,
after conversion to units of s-1, that are equivalent to k<E>. The
assumption implied by eqs 1c, 2d, and 3c for efficient collisional
deactivation will be addressed in the section discussing studies
of the C2H5Br reaction. The rate constant for C2H4Cl2* previ-
ously has been measured20,21 in SF6 bath gas as 21 ( 3 torr,
which corresponds to 3.1 × 108 s-1
.
The Cl-Br interchange reaction was identified from deute-
rium labeling of CH2BrCD2Cl and the possible reactions are
enumerated in 4.
CH2BrCD2Cl* f DBr + CH2dCDCl (D1)
f HCl + CHBrdCD2 (D2)
f CH2ClCD2Br*
(4a)
(4b)
(4c)
(4d)
+ SF6 f CH2BrCD2Cl (S)

4140 J. Phys. Chem. A, Vol. 114, No. 12, 2010
Friederich et al.
Figure 1. Plot of C2H3Br/C2H4Br2 vs pressure-1 for the C2H4Br2
system. The slope and intercept of the line are 112 ( 8 torr and -0.05
( 0.04, respectively. The correlation coefficient is 0.94.
Figure 2. Plots of D/S vs pressure-1 for the C2H4ClBr system. 9:
CH2dCHCl/C2H4ClBr; the slope and intercept of the line are 75.7 (
5.8 torr and -0.03 ( 0.04, respectively, the correlation coefficient is
0.95. b: CH2dCHBr/C2H4BrCl; the slope and intercept of the line are
32.6 ( 0.7 torr and -0.06 ( 0.01, respectively, the correlation
coefficient is 0.99.
detector is not very sensitive to C4F10, and CH2BrI and CH2ClI
were not monitored.
For analysis of vinyl chloride and vinyl bromide in the
CH2BrCD2Cl system, the FID detector was replaced by a mass-
spectrometer detector (GC-MS) containing a 0.25 mm by 105 m
RTX-VMS column. The temperature program for the GC-MS
analysis of vinyl chloride-d1, -d2, and vinyl bromide-d1, -d2
began with 15 min at 35 °C followed by a temperature increase
of 12 °C/min to a final temperature of 180 °C. The retention
times were 7.0 and 9.4 min for vinyl chloride and vinyl bromide,
respectively. Additional details about the mass spectrometric
analysis are given in Section III-B.
obtain the rate constants for just reactions 2a and 2b, two series
of experiments were done with 8:1 ratios of CH2BrI vs CH2ICl
to measure vinyl chloride from C2H4BrCl* and 20:1 ratios of
CH2ClI vs CH2BrI to measure vinyl bromide from C2H4BrCl*.
Nineteen experiments to measure the rate constant for loss of
HBr were done over the 70-600 torr pressure range with SF6
as the bath gas, and the results are shown in Figure 2. The least-
squares fit to the points give a slope of 75.7 ( 5.8 torr. Twenty-
six experiments were done over the pressure range of 25-280
torr of SF6 to measure the rate constant for reaction 2c and the
results are shown in Figure 2. The slope of the line, which
corresponds to the elimination of HCl from CH2BrCH2Cl, is
32.6 ( 0.7. The sum of the two rate constants is 108 torr, and
the total rate constants for C2H4Br2* and C2H4BrCl* are equal.
The ratio of rate constants for C2H4BrCl* is 2.3 in favor of
HBr elimination. Changing the ratios of the iodide precursors
as explained above was very successful in isolating the reaction
of interest. The rate constants for C2H4Br2 and C2H4BrCl are
summarized in Table 1. The collision diameters and Lennard-
Jones well-depth parameters that were used to calculate the
collision rate constant, for converting the slopes of the D/S plots
to s-1, are given in the footnotes of the table. The measured
rate constants also are adjusted to unit deactivation in Table 1.
Estimates of the two rate constants for CH2ClCH2Br* were also
provided in the 1967 work. They were measured by an indirect
method, and the values seem to be too small by a factor of 2,
even though the ratio of rate constants is the same as that
deduced from the data in Figure 2. The rate constant for HCl
elimination from CH2BrCH2Cl* is 1.5 times larger than that
from C2H4Cl2*, which implies that the threshold energy for HCl
elimination is lower for C2H4BrCl.
III. Experimental Results
A. Rate Constants for CH2BrCH2Br and CH2BrCH2Cl.
The analysis for the C2H3Br and C2H4Br2 products from the
photolysis of CH2BrI was straightforward, and 29 measurements
were made over the pressure range of 100-1000 torr with C4F10
as the bath gas. The resulting C2H3Br/C2H4Br2 (or D/S) ratios
are plotted versus pressure-1 in Figure 1. As is evident from
inspection of the plot, experiments in triplicate or quadruplicate
usually were done at a given pressure. The least-squares linear
fit to the plot gives a slope of 112 ( 8 torr. The linear plot has
a small negative intercept. If the line is forced to pass through
the origin, the slope would decline to 104 ( 8 torr. This
experimental result can be compared to the measurement made
in 1967 in a more complex chemical system (the reaction of
CH2 with CH2Br2) with CF4 as the bath gas. The slope of the
D/S plot reported for C2H4Br2* was 220 ( 30 torr. After
allowance for the difference in collisional efficiencies of CF4
versus C4F10, a factor of 1.4-1.5 for the high-pressure rate
constant, the 1967 result would correspond to 150 ( 20 torr
for a collider similar to C4F10. The two measurements are nearly
within their combined experimental uncertainties. However, the
current measurement is certainly preferred. The rate constant
for C2H4Br2* is approximately 6 times larger than that for
C2H4Cl2*.
B. Evidence for Cl-Br Interchange in CH2BrCD2Cl. The
rate constants for CH2BrCD2Cl* and CH2ClCD2Br* will be
reduced23-25 by approximately a factor of 2 for DCl or DBr
elimination (a combined primary and secondary kinetic-isotope
effect) and 1.5 for HCl or HBr elimination (a secondary kinetic-
isotope effect) relative to CH2BrCH2Cl*. Thus, the overall
pressure range for a similar degree of decomposition of
The photolysis of CH2BrI/CH2ClI mixtures gives C2H4Br2*
and C2H4Cl2*, as well as C2H4ClBr*. The rate constants for
formation of vinyl chloride from C2H4Cl2* and vinyl bromide
from C2H4Br2* are large enough that excluding vinyl bromide
from reaction 1b and vinyl chloride from 3b is necessary. To

Identification of the Cl-Br Interchange Reaction
J. Phys. Chem. A, Vol. 114, No. 12, 2010 4141
TABLE 1: Rate Constants and Threshold Energies for C2H4Br2 and C2H4BrCl
rate constants, (s-1
)
molecule
slopes of D/S plots (torr)a
experimentala,d
calculated
E0 (kcal mol-1
)
CH2BrCH2Br(-HBr)
112 ( 8b,d
1.7 ( 0.3 × 109
2.0 × 109
58
59
58
59
60
61
59.1
60.2
58.2
61.1
(1.5 × 109)e
1.5 × 109
CH2BrCH2Cl(-HBr)
(-HCl)
75.7 ( 5.8c, d
32.6 ( 0.7c, d
1.0 ( 0.2 × 109
(0.74 × 109)e
1.1 × 109
0.84 × 109
0.55 × 109
0.40 × 109
(0.49 × 109)f
(0.34 × 109)f
(0.74 × 109)f
(0.23 × 109)f
0.45 ( 0.07 × 109
(0.32 × 109)e
CH2BrCD2Cl (-DBr)
(-HCl)
CH2ClCD2Br (-HBr)
(-DCl)
a The slopes were converted to rate constants by changing the pressure to concentrations (molecules/cm3) and multiplying by the collision
rate constant, kM. b In C4F10 bath gas; the σ and ε/k are 6.3 Å and 247 K; see ref 26. c In SF6 bath gas; the σ and ε/k are 5.2 Å and 212 K; see
ref 26. d The σ and ε/k for C2H4Br2 are 5.5 Å and 465 K, and for C2H4BrCl they are 5.2 Å and 484 K; see ref 26. e The measured rate
constants were reduced by factors of 0.88 and 0.71 for C4F10 and SF6, respectively, to obtain unit deactivation rate constants given in
parentheses. f Calculated rate constants deduced from the model developed for CH2ClCH2Br using E0(HCl) ) 60 kcal mol-1 and E0(HBr) ) 58
kcal mol-1; see text for explanation. The calculations are for the hindered-rotor model with the three overall rotations taken as adiabatic.
CH2BrCD2Cl* will be shifted to lower pressures by a factor of
≈2 relative to the data in Figure 2. All experiments were done
with SF6 bath gas, and large vessels were used to achieve low
pressures. Two series of experiments were done. For the first
series, the pressure was sufficiently low that decomposition of
C2H2D2ClBr* molecules would have been nearly complete, for
example, S/D e 0.05. If the rate of Br/Cl interchange is
competitive with the rates of elimination, products from
reactions 4e and 4f will be present. The degree of interchange
can be measured by the ratios of HCl/DCl (CHBrdCD2/
CDBrdCH2) and HBr/DBr (CHCldCD2)/CDCldCH2). Before
describing the results of the experiments, the GC-MS analysis
of the isotopes of vinyl bromide and vinyl chloride needs to be
discussed. It should be noted that C2D4Cl2* and C2H4Br2* also
are products in these experiments and the composition of the
products from decomposition of C2D4Cl2*, C2H4Br2*, and
C2H2D2BrCl* will change with pressure.
The vinyl chloride sample delivered to the mass spectrometer
will be a mixture of C2D3Cl, CHCldCD2 and CDCldCH2. Due
to the 35Cl (75.8%) and 37Cl (24.2%) isotopes, this mixture has
masses of 65 and 67 (for C2D3Cl), 64 and 66 (for CD2dCHCl),
Figure 3. Plots of the kHCl/kDCl (b: CHBrdCD2/CH2dCDBr) and kHBr
/
and 63 and 65 (for CDCldCH2). The contribution to mass 65
from C2D3Cl was removed from the mass spectrum of the
mixture by measuring the ratio of 65 and 67 from C2D3Cl alone,
which was generated from the decomposition of C2D4Cl2 from
the photolysis of CD2ClI. With this ratio and with the signal
from mass 67 from the mixture, the contribution to mass 65
from C2D3Cl could be subtracted to obtain just the signal for
just CDCldCH2.The ratio of CD2dCHCl/CDCldCH2 could be
assigned as the ratio of the sum of the mass peaks (64 + 66):
(63 + 65) or as just the ratio of mass peaks 64:63. The situation
for vinyl bromide is similar, since 79Br (50.7%) and 81Br (49.3%)
isotopes exist. The masses present in a photolyzed sample are
110 and 108 (for CD2dCHBr), 109 and 107 (for CDBrdCH2),
and 106 and 108 (for C2H3Br). The contribution to mass peak
108 from C2H3Br could be subtracted after measurement of
the106/108 ratio from a pure sample of C2H3Br. Therefore, the
ratio of CD2dCHBr/CDBrdCH2 could be assigned as the ratio
of the sum of masses (109 +107)/(110 + 108) or as the ratio
of 110:109.
kDBr (9: CHCldCD2/CH2dCDCl) product ratios from CH2BrCD2Cl
and CH2ClCD2Br vs pressure.
are observed with ratios of CHCldCD2/CH2dCDCl ) 1.14 (
0.03 and CHBrdCD2/CH2dCDBr ) 1.65 ( 0.07 for the
experiments below 5 torr. The next series of experiments
included photolysis at 13, 20, 32, and 62 torr to explore the
effects of partial collisional stabilization of CH2BrCD2Cl* and
CH2ClCD2Br* on the ratios of products. The D/S ratio should
be about 1 at 60 torr. Within the experimental uncertainty, the
vinyl chloride ratio is 1.14 ( 0.04 and did not change over this
pressure range. The vinyl bromide ratio for the four trials at 62
torr is 1.74 ( 0.04, which is slightly higher than the average at
low pressure. However, the vinyl bromide ratio is 1.67 ( 0.06
for the 13-62 torr pressure range, the same as the average at
low pressure. If collisoional deactivation influences the product
branching in the system, it must affect both ratios. Since loss
of DCl has the slowest rate, CH2dCDBr has the smallest yield
and is subject to the highest uncertainty. We conclude that up
to 60 torr, the ratio of HCl/DCl and HBr/DBr is nearly constant,
that is, the interchange rate constant is considerably larger than
the elimination rate constants. After the elimination constants
are assigned from calculations of the kinetic-isotope effects for
The data are presented in Figure 3 for experiments over two
pressure regimes; 0.25-3.5 torr and 13-62 torr. Below 3.5 torr
the decomposition of CH2BrCD2Cl*(CH2ClCD2Br*) will be
higher than 90%. The interchange reaction certainly has occurred
because appreciable quantities of CH2dCDBr and CHCldCD2

4142 J. Phys. Chem. A, Vol. 114, No. 12, 2010
Friederich et al.
CH2BrCD2Cl and CH2ClCD2Br, an estimate is provided for the
interchange rate constant in Section IV-C.
C. Thermochemistry. The average vibrational energy of the
C2H4Br2 molecules formed by reaction 1a can be evaluated from
eq 5, assuming that the activation energy for CH2Br recombina-
tion is nearly zero.
should be within 15-20% of the density of states using the
explicit energy levels of the asymmetric-rotor.10 In addition to
the C2H4Br2 and C2H4BrCl systems, calculations also were done
for C2H5Br as a reference for HBr elimination. In this case the
hindered internal rotation is a symmetric CH3 rotor and
evaluation of the density of states is straightforward. The
calculated frequencies, moments of inertia, and structures for
the transition states are shown in the Supporting Information.
All electronic structure calculations were done with the Gaussian
03 suite of codes.34
E(C H Br ) ) D (CH Br-CH Br) + 3RT
〈
〉
2
4
2
0
2
2
+ 2 E (CH Br) (5)
〈
〉
V
2
The calculated threshold energies, in kcal mol-1, from the
6-31G(d′,p′) basis set were 53.7, C2H5Br; 57.5, CH2BrCH2Br;
and 58.1, 57.7, 41.8 and CH2BrCH2Cl for the HCl, HBr, and
interchange channels, respectively. The predicted increase in
E0 for HBr elimination from the dihalides relative to C2H5Br
should be noted, as well as the very low E0 for the interchange
reaction.
The density of states, N*(E), for the molecules and the sums
of states for the transition state, ΣP†(E - E0), were used in the
RRKM equation below to calculate the rate constants.
The last term is the thermal vibrational of the CH2Br radicals
at room temperature. The D298(CH2-CH2Br) value can be
obtained from the enthalpies of formation of C2H4Br2 and CH2Br
at 298 K, which are -9.027 and 40.928 kcal mol-1, respectively.
These numbers give D298(CH2Br-CH2Br) ) 90.8 kcal mol-1
,
which becomes 89.3 kcal mol-1 at 0 K, and 〈E(C2H4Br2)〉 is
91.8 kcal mol-1. The same procedure for CH2BrCH2Cl gives
an average energy of 90.7 kcal mol-1, based on enthalpies of
formation at 298 K of 28.0 and -21.0 for CH2Cl29a and
C2H4ClBr,29b respectively. A recent computational study30
reported the enthalpy of formation of CH2Br at 298 K as 39.8
kcal mol-1. This value would lower the average energies of
C2H4Br2 and C2H4BrCl by 2 and 1 kcal mol-1, respectively,
and make them equal. Therefore, we will take both average
energies to be 91 kcal mol-1. Given the reliability of the
thermochemistry, the uncertainty in 〈E〉 should be ( 2 kcal
kE ) s†/h(I†/I)1/2ΣP†(E-E0)/N*(E)
(6)
The reaction path degeneracy is s†, and I†/I is the ratio of
overall moments of inertia. For C2H4Br2, s† ) 4; and for
C2H4BrCl s† ) 2 for HCl and for HBr elimination. The moments
mol-1
.
of inertia and the reduced moment for internal-rotation, Ired
,
change with conformer and with internal rotation in general.
Therefore, we used the average values of I, Ired, and vibrational
frequencies from calculations of the molecular structure of
C2H4Br2 for each 20° of rotation. The average Ired and (I†/I)1/2
values were 28.7 amu Å2 and 1.38 for C2H4Br2. For C2H4BrCl,
the frequencies and moments of inertia were calculated for the
two conformers and the two barriers; the resulting average values
were Ired ) 19.2 amu Å and (I†/I)1/2 ) 1.32 (HCl elimination),
1.36 (HBr elimination), and 0.91 (Br/Cl interchange). The
transition-state structures for HBr elimination, HCl elimination,
and Br/Cl interchange were obtained from the Gaussian code
in the usual way.5-10 The rate constants were calculated using
the Multiwell code35 after the N(E*) were obtained. As points
of reference, rate constants also were calculated from N(E*)
with the torsional mode treated as a vibration and as a free-
rotor.
After the E0(HCl) and E0(HBr) values were assigned for
C2H4BrCl from matching k〈E〉 and kexp, the calculated zero-point
energies were used to assign threshold energies to the reactions
of CH2BrCD2Cl and CH2ClCD2Br. In fact, E0(HCl) and E0(HBr)
were nearly the same as for C2H4BrCl, and E0(DCl) and E0(DBr)
were higher by 1.1 kcal mol-1. The threshold energy for Br/Cl
interchange, which was assigned from fitting the data in Section
III-B, was the same for CD2ClCH2Br and CH2ClCD2Br. The
frequencies and moments of inertia needed to obtain the sums
and densities of states in eq 6 for CH2BrCD2Cl and CD2BrCH2Cl
were obtained from the Gaussian code following the same
procedure as described for CH2BrCH2Cl.
IV. Computational Results
A. Models for Molecules, Transition States, and Calcu-
lated Rate Constants. Electronic structure calculations were
done for the molecules and transition states using DFT with
the B3PW91 method and 6-31G(d′,p′) basis set. In some cases
supporting calculations were done with the 6-311+G(2d,p) basis
set. The calculated moments of inertia and frequencies closely
matched the experimental results for C2H5Br31 and other
calculated results for C2H4Br2.32 The torsional motion of the
CH2BrCH2Br and CH2BrCH2Cl molecules is the only unusual
aspect of the structures. For calculation of the density of states,
the torsion was treated as an asymmetric, hindered, internal
rotation. The molecules have trans and gauche conformers; the
trans-conformer is 1.8 kcal mol-1 lower in energy. Wong et
al.33 evaluated the potential for the hindered internal-rotation
of C2H4Br2 and calculated the energy levels below the barriers,
which are 5.5 and 9.7 kcal mol-1, as well as those above the
barriers. They kindly provided us with a listing of the energy
levels, and we combined them with the vibrational sums of states
to obtain the total sums and density of states for C2H4Br2. These
densities also were compared to the result from treating the
torsion as a hindered symmetric rotor with a barrier equal to
the average of the two actual barriers. The energy levels for
the hindered internal rotor of CH2BrCH2Cl were not evaluated
by Wong et al. Therefore, we calculated the potential function
using the 6-31G(d′,p′) basis set. The potential energy barriers
were 4.6 and 8.6 kcal mol-1, and the energy difference between
the conformers was 1.7 kcal mol-1. We averaged the two
barriers and calculated the density of states for CH2BrCH2Cl
as for an equivalent symmetric internal rotor with a barrier of
6.6 kcal mol-1. The reliability of the barriers from the calculation
for C2H4BrCl was confirmed by doing a calculation of the
potential for C2H4Br2 and comparing the result to other
calculations.32,33 Upon the basis of four examples, including that
for C2H4Br2, the symmetric-rotor approximation for C2H4BrCl
B. Comparison of Calculated and Experimental Results
for C2H5Br. The model for the HBr elimination reaction from
C2H5Br can be tested using the published results from thermal
and chemical-activation studies. The two most recent thermal
studies reported limiting high-pressure Arrhenius parameters of
1013.62(0.22 exp(-53.4 ( 0.7 kcal mol-1/RT)15 from a temperature
range of 660-706 K and 1013.6(0.3 exp(-52.8 ( 1.0 kcal mol-1/
RT)16 from a temperature range of 950-1200 K. These results

Identification of the Cl-Br Interchange Reaction
J. Phys. Chem. A, Vol. 114, No. 12, 2010 4143
are in agreement with a shock-tube study17 over the temperature
range of 740-1000 K, which reported 1013.3 exp(-53.5 kcal
mol-1/RT) and also with early pyrolysis studies.18 All of these
Arrhenius parameters could be criticized, for example the work
of ref 15 covers a very small temperature range, and that of ref
16 is an extrapolation from very-low-pressure pyrolysis experi-
ments. Also, some early measurements may not have been in
the true high pressure limit.11 Tsang17 selected isopropyl bromide
as the reference reaction for his shock-tube work, and two
subsequent studies have confirmed his selection of Arrhenius
parameters for isopropyl bromide.36,37 Hippler et al.38 combined
their data with published results for t-butyl bromide to provide
an Arrhenius plot encompassing 9 orders of magnitude in the
rate constant. Their global fit gave Arrhenius parameters for
t-butyl bromide that were slightly higher than those from Tsang’s
work;17 this report also had included Tsang’s study of C2H5Br.
In summary, it appears that 1013.8 exp(-54 kcal mol-1/RT) is a
reasonable choice, if the upper range of values is weighted in
anticipation that C2H5Br would resemble t-butyl bromide, if data
from a wide range of temperature were available.
review of 2001, Barker, Yoder, and King40 favored a nearly
linear relationship between 〈∆Ed〉 and the vibrational energy of
the excited molecule. A modest inverse dependence of 〈∆Ed〉
on temperature is usually found, although there are examples
with a more pronounced inverse dependence.40 Based upon the
information just cited and the lack of explicit studies of
bromoethanes with 90 kcal mol-1 of vibrational energy in
collisions at room temperature, we will adopt results from
studies of collisional deactivation of fluoroethanes41 and
1,2-dichloroethane20,21 to assign relative efficiencies for deac-
tivation of C2H4Br2* and C2H4BrCl*with SF6 and C4F10. For
our purposes, we only need the factor that converts the observed
experimental rate constant to the unit deactivation rate constant,
which is then matched to k〈E〉 to obtain E0.
C. Rate Constants and Threshold Energies for C2H4Br2
and C2H4BrCl. As an additional test case for use of the
symmetric-rotor approximation to evaluate the density of states
for C2H4BrCl, the density of states for C2H4Br2 from the direct
count was compared to a calculation using a symmetric rotor
with an average barrier height (7.6 kcal mol-1). At 90 kcal mol-1
the symmetric-rotor approximation for the density was higher
than the direct-count density by a factor of 1.19. A comparison
also was made to the thermal partition-functions at 800 K; the
symmetric-rotor approximation was higher than exact-count
value by a factor of 1.7, and the symmetric-rotor approximation
is not adequate for the partition-function. At 800 K the exact-
count partition-function actually converges before levels above
the highest barrier (9.7 kcal mol-1) are needed in the summation.
Upon the basis of this comparison and even better cor-
respondence10 for the density of states between the direct-count
and symmetric-rotor approximation for C2H4F2, C2H4Cl2, and
C2H4ClF, the rate constants for C2H4BrCl should be within 15%
of a direct-count evaluation of the density of states.
The experimental rate constant for C2H4Br2 of 1.5 × 109 s-1
requires E0(HBr) ) 58 kcal mol-1 to obtain a match with the
calculated rate constant; see Table 1. This 4-5 kcal mol-1
increase in E0, relative to that for C2H5Br is consistent with the
≈7 kcal mol-1 difference between C2H5Cl and CH2ClCH2Cl.
The calculated difference in threshold energies for C2H5Br and
C2H4Br2 is 3.8 kcal mol-1 for the 6-31G(d′,p′) basis set and 4.7
kcal mol-1 for the 6-311+G(2d,p) basis set.
The threshold energies required to match the calculated and
experimental rate constants for HCl and HBr elimination from
CH2BrCH2Cl are 60 and 58 kcal mol-1, respectively. Since the
ratio of the sums of states for the HBr to HCl transition states
is only 1.06, the difference of a factor of 2 in the experimental
rate constants can be obtained only with a higher E0 for the
HCl case. These values, which are higher than the E0 values
for C2H5Cl and C2H5Br, closely match the values for
CH2BrCH2Br and CH2ClCH2Cl.10
McGrath and Rowland14 used the QCISD method with the
cc-pVDZ basis set to develop a model for HBr elimination from
C2H5Br. Their calculated E0 was 57.1 kcal mol-1; the corre-
sponding activation energy would be ≈59 kcal mol-1, which
certainly is too high. The pre-exponential factor in Arrhenius
form at 800 K from their model with the torsion treated as a
hindered internal rotation is 15 × 1013 s-1, which is two times
larger than the experimental upper limit (6.3 × 1013 s-1) selected
above. Our model for the C2H5Br reaction from the DFT
calculations gives a similar pre-exponential factor of 13 × 1013
s-1 and E0 ) 53.7 kcal mol-1. The results from the two different
calculations confirm the general thesis that transition-state
structures are not very sensitive to the computational method
even though the calculated threshold energies may be signifi-
cantly different. The possibility that the entropy of the calculated
structure of the transition states for HBr elimination may be
too large remains a question to be resolved.
The experimental rate constant reported11 for C2H5Br activated
with 91 kcal mol-1 of energy formed by recombination of CH3
with CH2Br radicals has considerable uncertainty. However, this
rate constant, 6 ( 3 × 109 s-1, still can provide a useful check
for the threshold energy. Calculations from our model for
C2H5Br with the torsion treated as a hindered internal rotor gave
kE ) 9.6 × 109 and 7.3 × 109 s-1 for E0 ) 53 and 54 kcal
mol-1, respectively. Thus, the chemical activation data are
consistent with the threshold energy obtained from thermal
activation. The extensive studies of Raff and co-workers39 with
theoretical models for the dynamics of vibrationally excited
CH2dCHBr should be mentioned. They found rapid internal
energy relaxation, relative to the rate for 1,1-HBr elimination,
although they argue that C-Br rupture is not a statistical process.
Given the internal rotational mode present in C2H5Br and
C2H4Br2, rapid relaxation of the internal energy in these
molecules can be accepted relative to the rates of HBr
elimination.
If one overall rotational degree of freedom is converted to
an active mode in the calculation of densities and sums of states
in eq 6, the rate constants would be reduced by a factor of
=1.5.10 This has the effect of reducing the assigned E0 values
by 1 kcal mol-1. For this reason the lower range of values for
E0 were selected in Table 1. Given the combined experimental
and computational uncertainties, the reliability of the assigned
E0 values is probably (3 kcal mol-1; however, the difference
between E0(HCl) and E0(HBr) in CH2BrCH2Cl should be more
reliable.
D. Rate Constants and Threshold Energies for Cl-Br
Interchange in CH2BrCD2Cl. The transition state for inter-
change has higher frequencies, smaller moments of inertia, and
a reduced s† (s† ) 1) relative to the transition states for HBr
The primary goal of the two thermal studies of C2H5Br
mentioned above was to obtain information about vibrational
energy transfer using CH2DCH2Br15 and CHD2CD2Br.16 By
fitting the falloff data, 〈∆Ed〉 was assigned as 1100 ( 100 cm-1
for self-collisions by Jung et al.15 The very low pressure
pyrolysis experiments gave 〈∆Ed〉 as 850 and 1300 cm-1 for
C2H4 and C6H6, respectively, as bath gases, which are the most
similar to SF6 and C4F10 that were studied. Similar 〈∆Ed〉 values
were reported from studies with isopropyl bromide.36 In their

4144 J. Phys. Chem. A, Vol. 114, No. 12, 2010
Friederich et al.
from CH2ClCH2Br molecules with an energy of 91 kcal mol-1
and HCl elimination. Thus, a lower (≈7 kcal mol-1) E0 is
required for the interchange rate constant to obtain a value that
is similar to the elimination rate constants. Based on the
approximate fitting of the data in Figure 3, as summarized
below, the interchange rate constant should be 8 ( 2 times larger
than the rate constant for HBr elimination from CH2ClCD2Br.
Such a value corresponds to a threshold-energy of ≈43 kcal
mol-1 for Cl/Br interchange. The DFT calculated threshold
energy was 41.8 kcal mol-1. Rate constants are first assigned
to CH2BrCD2Cl (kHCl and kDBr) and to CH2ClCD2Br (kHBr and
kDCl) based on the calculations outlined below. These calculated
isotope effects were applied to the experimental rate constants
from C2H4ClBr to obtain rate constants to interpret the ratios
in Figure 3.
The changes in zero-point energies for CH2BrCD2Cl and
CH2ClCD2Br relative to CH2BrCH2Cl (E0(HCl) ) 60 kcal mol-1
and E0(HBr) ) 58 kcal mol-1) gave values (kcal mol-1) of 60.2
for loss of HCl and 59.1 for loss of DBr from CH2BrCD2Cl
and 61.1 (DCl) and 58.2 (HBr) for CH2ClCD2Br. Combining
these values with the sums and densities of states gave rate
constants (109 s-1) of 0.34 for HCl and 0.49 for DBr for
CH2BrCD2Cl and 0.23 for DCl and 0.74 for HBr for
CH2ClCD2Br at 91 kcal mol-1 of energy. These values cor-
respond to kinetic-isotope effects of 2.1 for DCl and DBr
processes and 1.5 for HCl and HBr processes. The threshold
energy for interchange in CH2BrCD2Cl was unchanged relative
to CH2BrCH2Cl, but the interchange rate constant decreased by
a factor of 1.3 from the secondary kinetic-isotope effect.
The data in Figure 3 for Cl/Br interchange can be evaluated
by a steady-state analysis. Let [CH2BrCD2Cl] be [A] and
[CH2ClCD2Br] be [B] with total elimination rate constants kA
and kB and let kI be the interchange rate constant. In the limit
of low pressure the steady-state equations are given below with
R being the rate of recombination of CH2Br and CD2Cl.
.
In fact, the threshold energy for interchange is 15 kcal mol-1
lower than for the elimination reactions, and interchange is the
dominant reaction. The E0 for Cl/F interchange for CH2FCH2Cl
also is the lowest threshold energy, based on computational
results;10 however, 1,2-HCl and 1,2-HF elimination would be
competitive for CH2ClCH2F. Since the threshold energy for 1,2-
HX elimination is elevated when halogen atoms are on adjacent
carbon atoms, the 1,2-dihaloethane molecules provide some of
the best systems to demonstrate halogen exchange reactions.
We have previously shown10 that the threshold energy for 1,1-
HX elimination in 1,2-dihaloethanes is between 80-90 kcal
mol-1; thus, a mechanism involving 1,1-HX elimination fol-
lowed by H migration need not be considered. The entropies
of the HBr and HCl transition states are very similar, as
demonstrated by the ratios (at the same energy) of their sums
of vibrational states, which is only 1.06 in favor of the transition
state for HBr elimination. In contrast, the ratio of the sums of
states for the HBr versus the Br/Cl interchange transition states
is 2.3, and the interchange transition state has the smaller
entropy. The three transition-state structures from CH2ClCH2Br
are shown in Figure 4. One common feature is the trigonal
planar geometry for the carbon atoms in the interchange
transition state and the carbon atom from which the chlorine or
bromine atom is departing in the elimination transition states.
The positions of the bromine and chlorine atoms in the
interchange transition state are nearly the same, which seems
surprising, but such a structure is consistent with the similar
atomic radii of the two atoms, which differ by only 15%. This
similarity can be contrasted to the difference between the
positions of fluorine and chlorine atoms in the Cl/F interchange
transition state of CH2ClCH2F, which also is shown in Figure
4. The C-Cl distance is the same as for the Cl/Br case, but the
C-F distance is much shorter. According to the calculations,
the E0 for Cl/F interchange in CH2ClCH2F is 60 kcal mol-1
d[A]/dt ) 0 ) R + [B]kI - [A](kI + kA)
d[B]/dt ) 0 ) kI[A] - [B](kI + kB)
(7a)
(7b)
and that for Br/F interchange in CH2BrCH2F is 57 kcal mol-1
.
These are much larger than 43 kcal mol-1 for CH2BrCH2Cl.
Exploratory calculations also were done for CH3CHClCH2Br;
the calculated E0 for interchange was 3 kcal mol-1 lower than
for CH2ClCH2Br, and Cl/Br interchange should be expected
whenever a bromine atom and a chlorine atom are on adjacent
carbon atoms in chlorobromoalkanes.
Equation 7b gives the ratio of steady-state concentrations of A
and B, eq 8, from which the ratio of HCl/DCl and HBr/DBr
can be obtained as kAHCl[A]/kBDCl[B] and kBHBr[B]/kADBr[A].
In an effort to understand the nature of the interchange
transition states and the difference between Br/Cl and Cl/F or
Br/F cases, the charges on the atoms in the structures were
examined using the atoms-in-molecules approach.42 The charges
on the atoms of the transition state for Cl/Br interchange are
Cl(-0.40), Br(-0.28), C(+0.089), and H(+0.13). The carbon
atoms become more positive in the transition state for Cl/F
interchange; F(-0.49), Cl(-0.44), C(+0.23), and H(+0.12). The
positive H2C-CH2 backbone is surrounded by the negative
halogen atoms. We did not do calculations that would provide
information about the molecular orbitals. However, an attractive
intuitive description is Π-type bonding involving the p-orbitals
of the carbon atoms and the halogen atoms. The lines drawn in
Figure 4 between the carbon atoms and the halogen atoms
denote distances and not necessarily localized chemical bonds.
The reaction coordinate for the interchange is simply a twisting
(torsional) motion of the H2C-CH2 plane bringing each CH2
group closer to the stationary halogen atoms. An Intrinsic
Reaction Coordinate calculation confirms the conversion be-
tween CH2BrCD2Cl and CH2ClCD2Br via the interchange
transition state in Figure 4.
[A]/[B] ) (kI + kB)/kI
(8)
If kI ) 8kBHBr, the product ratios are HCl/DCl ) 1.60 and HBr/
DBr ) 1.20. The ratios are 1.68 and 1.14, if kI ) 6kBHBr. The
experimental data at low pressure in Figure 3 are in agreement
with such values for kI. The calculated rate constant for
interchange is 5.4 and 8.6 times larger than kHBr for threshold-
energies of 45 and 43 kcal mol-1, respectively. If the collision
rate becomes comparable to kB, then that term must be added
to the numerator of eq 8 and the ratio of [A]/[B] becomes larger,
and the HCl/DCl and HBr/DBr ratios increase and decrease,
respectively. The experimental ratios are nearly constant, except
possibly at the highest pressure (60 torr), and kI must be 6-10
times larger than kHBBr
.
V. Discussion
A. Cl/Br Interchange Reaction. The deuterium labeling
experiment conclusively demonstrated that the Br-Cl inter-
change reaction is faster than HBr and HCl elimination reactions

Identification of the Cl-Br Interchange Reaction
J. Phys. Chem. A, Vol. 114, No. 12, 2010 4145
Figure 4. Transition states computed from the B3PW91/6-31G(d′,p′) method for the three reactions of CH2BrCH2Cl. The diagram for Cl/F interchange
from CH2ClCH2F is included for comparison. The similar distance of the Br and Cl atoms from the plane gives a very symmetric structure, and the
planar structure for the CH2CH2 backbone of the interchange transition states should be noted. The Cl-Br interchange transition state forms by
twisting of the CH2-CH2 backbone between the immobile Br and Cl atoms whose separation in the anti configuration of the molecule (4.48 Å) and
in the transition state (4.45 Å) remains constant. The degree of planarity of the two carbons is identified by the angle between the extension of the
C-C axis and the plane defined by the three atoms, e.g. CH2, CHCl, or CHBr. As points of reference, 55 and 0° correspond to 100% tetrahedral
and trigonal planar, respectively. The lines drawn from the halogen atoms to the carbon atoms are convenient ways to denote the distances; they
do not necessarily denote chemical bonds; see text.
B. Comparison of Elimination Reactions for CH2XCH2Y
Molecules. As already noted, the ΣP†(E - E0) of the transition
states for HCl and HBr elimination from CH2BrCH2Cl are very
similar. The calculated structures for the transition states also
are very similar as illustrated by the diagrams shown in Figure
4. The long C-Cl and C-Br distances and the nearly planar
geometry around the H2CX end of the transition states are
especially noteworthy. The calculated pre-exponential factors
for both HCl and HBr elimination tend to be a factor of ≈2
larger than the experimental results as demonstrated here by
the comparison with C2H5Br and by other comparisons for
various alkyl chlorides.10,12,43 The discrepancy with the experi-
mental results is a consequence of the long C-Cl and C-Br
distances, which introduce low frequency vibrations into the
model of the transition states. Different levels of theory and
different theoretical approaches give similar transition-state
structures. Although this discrepancy is not very serious for most
aspects associated with HCl and HBr unimolecular elimination
reactions, the long C-Cl and C-Br distances in the transition
states remain an interesting question.
Except for CH2BrCH2F, the experimental data for all
CH2XCH2Y(X,Y ) Br, Cl, F) type molecules have been
analyzed recently. In every case the threshold-energy is several
kcal mol-1 higher than for the C2H5X molecule. Furthermore,
E0(HX) tends to be the similar to E0(HY), although E0(HCl) is
2 kcal mol-1 higher than E0(HBr). One possible explanation
for the elevation of E0 is the net reduced sum of the bond
energies for the HXCH end of the transition state relative to the

4146 J. Phys. Chem. A, Vol. 114, No. 12, 2010
Friederich et al.
(3) Burgin, M. O.; Heard, G. L.; Martell, J. M.; Holmes, B. E. J. Phys.
Chem. A 2001, 10, 1615.
(4) Heard, G. L.; Holmes, B. E. J. Phys. Chem. A 2001, 105, 1622.
(5) Burgin, M. O.; Simmons, J. G., Jr.; Heard, G. L.; Setser, D. W.;
Holmes, B. E. J. Phys. Chem. A 2007, 111, 2283.
(6) Zaluzhna, O.; Simmons, J. G., Jr.; Heard, G. L.; Setser, D. W.;
Holmes, B. E. J. Phys. Chem. A 2008, 112, 6090.
(7) Zaluzhna, O.; Simmons, J. G., Jr.; Setser, D. W.; Holmes, B. E. J.
Phys. Chem. A 2008, 112, 12117.
(8) Beaver, M. R.; Simmons, J. G., Jr.; Heard, G. L.; Setser, D. W.;
Holmes, B. E. J. Phys. Chem.A 2007, 111, 8445.
(9) Lisowski, C. E.; Duncan, J. R.; Heard, G. L.; Setser, D. W.; Holmes,
B. E. J. Phys. Chem. A 2008, 112, 441.
(10) Duncan, J. R.; Solaka, S. A.; Setser, D. W.; Holmes, B. E. J. Phys.
Chem. A 2010, 114, 794.
H2CH end of C2H5X, where the H subscript denotes the carbon
atom attached to the departing H atom. The calculated E0 values
for CH2BrCH2Br and CH2BrCH2Cl were very close to the
experimentally assigned values. Therefore, we also calculated
the E0 for HBr and HF elimination from CH2BrCH2F to
complete the series, and the values are 57.5 and 60.6, respec-
tively. The 3 kcal mol-1 difference is consistent with the limited
experimental data, which indicate that the HBr elimination rate
is faster than the HF elimination rate.44-46 The possibility of
Cl/Br, F/Cl, and F/Br interchange should be remembered when
considering the proposed reaction mechanisms for infrared-
multiphoton excitation of haloethanes.47-50
(11) Johnson, R. L.; Setser, D. W. J. Phys. Chem. 1967, 71, 4366.
(12) Ferguson, J. D.; Johnson, N. L.; Kekenes-Huskey, P. M.; Everett,
W. C.; Heard, G. L.; Setser, D. W.; Holmes, B. E. J. Phys. Chem. A 2005,
109, 4540.
(13) Toto, J. L.; Pritchard, G. O.; Kirtman, B. J. Phys. Chem. 1994, 98,
8357.
(14) McGrath, M. P.; Rowland, F. S. J. Phys. Chem. A 2002, 106, 8191.
(15) Jung, K.-H.; Kang, S. H.; Ro, C. U.; Tschuikow-Roux, E. J. Phys.
Chem. 1987, 91, 2354.
(16) Nguyen, T. T.; King, K. D.; Gilbert, G. G. J. Phys. Chem. 1983,
87, 494.
(17) (a) Tsang, W. J. Chem. Phys. 1964, 41, 2487. (b) Int. J. Chem.
Kinet 1970, 2, 311.
(18) References 11 and 15-17 give summaries of early pyrolysis studies
of C2H5Br.
(19) Butler, L. J.; Hintsa, E. J.; Shane, S. F.; Lee, Y. T. J. Chem. Phys.
1987, 86, 2051.
(20) Setser, D. W.; Siefert, E. E. J. Chem. Phys. 1972, 57, 3613–3623.
(21) Clark, W. G.; Setser, D. W.; Siefert, E. E. J. Phys. Chem. 1970,
74, 1670.
(22) Holmes, B. E.; Paisley, D. J.; Rakestraw, D. J.; King, E. E. Int.
J. Chem. Kinet 1986, 18, 639.
VI. Conclusions
The facile interchange of bromine and chlorine atoms located
on adjacent carbon atoms has been demonstrated by observation
of CH2ClCD2Br formation from vibrationally excited
CH2BrCD2Cl molecules that were generated by the recombina-
tion of CH2Br and CD2Cl radicals at room temperature. The
vinyl halide product ratios for loss of HCl/DCl and HBr/DBr
from CH2BrCD2Cl and CH2ClCD2Br were used to identify the
interchange reaction, which is 8 ( 2 times faster than the
elimination reactions. This information was combined with
observation of the HBr and HCl elimination reactions from
CH2BrCH2Br and CH2BrCH2Cl formed by radical combination
to assign threshold energies of 60 kcal mol-1 for HCl elimina-
tion, 58 kcal mol-1 for HBr elimination and =43 kcal mol-1
for Cl/Br interchange. These assignments are in good accord
with the threshold energies calculated from DFT with the B3PW
91/6-31G(d′,p′) method. The low threshold energy for Br/Cl
interchange suggests that this unimolecular process should be
expected whenever chlorine and bromine atoms are located on
adjacent carbon atoms in chlorobromoalkyl species. The transi-
tion state for Br/Cl interchange is very symmetric with the Br
and Cl atoms resting equidistant between the carbon atoms just
above and below the plane of the H2C-CH2 structure. Accord-
ing to the DFT calculations, the E0 for interchange of fluorine
and chlorine atoms in CH2FCH2Cl and fluorine and bromine
atoms in CH2FCH2Br are 60 and 57 kcal mol-1, respectively.
Based on comparison to data from the C2H5Br reaction, the
transition state calculated by DFT with the B3PW91/6-
31G(d′,p′) method for HBr elimination should be as reliable as
that for HCl elimination reactions. In fact, the transition states
for HCl and HBr elimination are very similar. The threshold
energies for CH2BrCH2Br and CH2ClCH2Br assigned here
follow the general trend of elevated threshold energies for 1,2-
dihaloethanes, relative to the corresponding ethyl halides.
(23) Zhu, L.; Simmons, J. G., Jr.; Burgin, M. O.; Setser, D. W.; Holmes,
B. E. J. Phys. Chem. A 2006, 110, 1506.
(24) Holmes, D. A.; Holmes, B. E. J. Phys. Chem. A 2005, 109, 10726.
(a) The calculated rate constant for 1,2-FH elimination listed in Table 2
for CF3CHFCH3 should be 2.8 ×104 s-1, and the calculated kinetic-isotope
effect for CF3CHFCD3 is 2.2 as quoted in the text.
(25) Clark, W. G.; Setser, D. W.; Dees, K. J. Am. Chem. Soc. 1971, 93,
5328.
(26) Hippler, H.; Troe, J.; Wendelken, H. J. J. Chem. Phys. 1983, 78,
6709.
(27) Seetula, J. A. Phys. Chem. Chem. Phys. 2003, 5, 849.
(28) Kerr, J. A.; Stocker, D. W. Standard Thermodynamic Properties
of Chemical Substances, In Handbook of Chemistry and Physics; Linde,
D. R., Jr., Ed; CRC Press: Boca Raton, Fl, 2007.
(29) (a) Seetula, J. A. J. Chem. Soc. Faraday Trans. 1996, 92, 3069.
(b) Lias, S. G.; Bartmess, J. E.; Liebman, J. F.; Holmes, J. L.; Levin, R. D.;
Mallard, W. G. J. Phys. Chem. Ref. Data. 1988, 17, 88.
(30) Marshall, P.; Srinivas, G. N.; Schwartz, M. J. Phys. Chem. A 2005,
109, 6379.
(31) (a) Tang, B.; Zhang, S.; Wang, Y.; Tang, Y.; Zhang, B. J. Chem.
Phys. 2005, 123, 164305. (b) Gripp, J.; Dreizler, H.; Schwarz, R. Z.
Naturforsch. 1985, 40a, 575.
(32) Wang, L. J. Phys. Chem. A 2008, 112, 4951.
(33) Wong, B. M.; Fadri, M. M.; Raman, S. J. Comput. Chem. 2007,
29, 481.
Acknowledgment. Financial support for this work was
provided by the US National Science Foundation under grants
CHE-0647582.
(34) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kuden, K. N.;
Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; Cossi, M.; Scalmani, G.; Bega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K. ; Fukuda, R.; Hasegawa,
J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Adamo, C.; Jaramillo,
J.; Gomperts, R.; Stratman, R. E.; Yazyev, O.; Austen, A. J.; Cammi, R.;
Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.;
Salvador, P.; Dannenberg, J. J.; Zakrzewksi, V. G.; Dapprich, S.; Daniels,
A. D.; Strain, M. C.; Farkas, O.; Malik, D. K.; Rabuck, A. D.; Raghavachari,
K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.;
Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.;
Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng,
C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.;
Chen.; W.; Wong, M. W.; Gonzalez, C. Pople, J. A.; Gaussian 03, ReVision
B.04; Gaussian, Inc, Pittsburg, PA, 2003.
Supporting Information Available: Tables of the molecular
and transition state structure vibrational frequencies, atomic
charges, overall moments of inertia, and the reduced moments
of inertia for the internal rotors calculated using B3PW91/6-
31G(d′,p′) for CH2BrCH2Cl, CH2BrCD2Cl, and CH2ClCD2Br.
This information is available free of charge via the Internet at
http://pubs.acs.org.
References and Notes
(35) Barker, J. R. Int. J. Chem. Kinet. 2001, 33, 232.
(36) Brown, C. T.; King, K. D.; Gilbert, R. G. Int. J. Chem. Kinet. 1987,
19, 851.
(1) Beaver, M. R.; Heard, G. L.; Holmes, B. E. Tetrahedron. Lett. 2003,
44, 7265.
(2) Dolbier, W. R., Jr.; Romelaeer, R.; Baker, J. M. Tetrahedron. Lett.
2002, 43, 8075.
(37) Gutman, D.; Braun, W.; Tsang, W. J. Chem. Phys. 1977, 67, 4291.

Identification of the Cl-Br Interchange Reaction
J. Phys. Chem. A, Vol. 114, No. 12, 2010 4147
(38) Hippler, H.; Riedel, A.; Troe, J.; Willner, J. Zeit. Physik. Chem.
1991, 171, 161.
(46) Harrison, R. G. In Laser Applications in Chemistry; Kompa, K. L.
Wanner, J., Eds.; Plenum: New York, 1984; p 141.
(39) (a) Rahaman, A.; Raff, L. M. J. Phys. Chem. A 2001, 105, 2147–
2156. (b) Doughan, D. J.; Raff, L. M.; Rockley, M. G.; Hagan, M.; Agrawal,
P. M.; Komanduri, R. J. Chem. Phys. 2006, 24, 054321.
(40) Barker, J. R.; Yoder, L. M.; King, K. D. J. Phys. Chem. A 2001,
105, 796.
(41) Richmond, G.; Setser, D. W. J. Phys. Chem. 1980, 84, 2699.
(42) Bader, R. F. W., Atoms in Molecules, A Quantum Theory; Clarendon
Press: Oxford, 1990.
(43) Rajakumar, B.; Arunan, E. Phys. Chem. Chem. Phys. 2003, 5, 3897.
(44) Richardson, T. H.; Setser, D. W. J. Phys. Chem. 1977, 81, 2301.
(45) McRae, G. A.; Evans, D. K.; Goodale, J. W. J. Chem. Phys. 1990,
93, 1689.
(47) Yokoyama, A.; Yokoyama, K.; Fujisawa, G. J. Chem. Phys. 1994,
101, 10602.
(48) (a) Pushpa, K. K.; Kumar, A.; Vasta, R. K.; Naik, P. D.; Rao, K. A.;
Mittal, J. P.; Parthasarathy, V.; Sarkar, S. K. Chem. Phys. Lett. 1995, 240,
489. (b) 1996, 249, 167.
(49) Setser, D. W.; Lee, T.-S.; Danen, W. C. J. Phys. Chem. 1985, 89,
5799.
(50) Jang-Wren, J. C.; Setser, D. W.; Ferrero, J. C. J. Phys. Chem. 1985,
89, 414.
JP9116134