Thermodynamic properties of benzyl halides: Enthalpies of formation, strain enthalpies, and carbon-halogen bond dissociation enthalpies

Article abstract of DOI:10.1039/b301771k

The molar enthalpies of formation of benzyl halides PhCH₂-X, PhCH(CH₃)-X, and PhC(CH₃)₂-X, where X = F, Cl, Br, and I are in the gaseous state, have been derived by combination of our own thermochemical measurem

Full text of DOI:10.1039/b301771k

View Online / Journal Homepage / Table of Contents for this issue
Thermodynamic properties of benzyl halides: enthalpies of
formation, strain enthalpies, and carbon–halogen bond
dissociation enthalpiesy
Sergey P. Verevkin,* Eugen L. Krasnykhz and James S. Wrightx
Institute of Physical Chemistry, University of Rostock, Hermannstr. 14, 18051, Rostock,
Germany. E-mail: sergey.verevkin@chemie.uni-rostock.de
Received 14th February 2003, Accepted 16th April 2003
First published as an Advance Article on the web 8th May 2003
2 3 3 2
The molar enthalpies of formation of benzyl halides PhCH -X, PhCH(CH )-X, and PhC(CH ) -X, where
X ¼ F, Cl, Br, and I are in the gaseous state, have been derived by combination of our own thermochemical
measurements together with the data available from the literature. These new results have been shown to be
internally consistent, and they were used to derive strain enthalpies of benzyl halides as well as for estimation of
the carbon–halogen bond dissociation enthalpy in benzyl halides. Strain effects are discussed in terms of
deviations of gaseous enthalpies of formation from group additivity rules. Bond dissociation enthalpies (BDE’s)
of the benzyl halides studied are generally indistinguishable from their alkyl analogs having the same
substituent X.
Introduction
Quantification of the bond dissociation enthalpies (BDEs) and
energetics of free radicals form a vital part of our understand-
ing of the influence of thermodynamic properties on chemical
reactivity. Differences in these energies, while small in percen-
tage terms, obviously have a profound influence on the path-
way of chemical reactions. There is a considerable interest in
the precise enthalpies of formation of radicals of benzyl deriva-
tives since there are probably no other radicals upon which so
1
many bond energy values depend. Thus, a systematic study of
thermochemistry of benzyl derivatives (see Fig. 1) seems to be
important for practical and theoretical reasons. The carbon–
halogen BDE is defined in the usual way from reaction (1),
ꢀ
ꢀ
PhCH2-Hal ! PhCH2 þ Hal
ð1Þ
according to eqn. (I):
ꢁ
ꢀ
ꢁ
ꢀ
BDEðPhCH2-HalÞ ¼ Df H mðg; PhCH2 Þ þ Df H mðg; Hal Þ
ꢁ
ꢂ Df H m ðg; PhCH2-HalÞ
Precise values of enthalpies of formation of halogen-radicals
ðIÞ
ꢁ
ꢀ
2
(g, Hal ) are assigned by CODATA, as well as the
D
f
H
enthalpy of formation of the benzyl radical,
m
ꢁ
Fig. 1 Structures of benzyl derivatives studied.
D
f
H
m
(g,
ꢀ
ꢂ1
PhCH ) ¼ (207 ꢃ 4) kJ mol , which have been recom-
mended recently. Hence, bond dissociation enthalpies of ben-
2
3
ꢁ
4–8
enthalpies of formation D H (l) of benzyl bromide
differ
by 15 kJ mol and for benzyl iodide these values are spread
f
m
zyl halides, BDE(PhCH
provided that reliable enthalpies of formation, D H m(g,
2
-Hal), could be derived from eqn. (I),
ꢁ
ꢂ1
7–9
ꢂ1
g
ꢁ
f
over 17 kJ mol . The vaporization enthalpies, D H , listed
l
48
m
PhCH -Hal), of benzyl halides are available.
Unfortunately, thermochemistry of benzyl halides is in
disarray. For example, according to Table 1, experimental
2
in a recent database are available only for 1a–d and most
ꢂ1.
of them are uncertain by ꢃ2–4 kJ mol
Taking this into
ꢁ
account, available gaseous enthalpies of formation, D
f
H
m
ꢁ
(
g), derived from the sum of contributions D H
(l) and
m
f
g
l
ꢁ
y Electronic Supplementary Information (ESI) available: Experimen-
D H
thermochemistry of benzyl halides requires additional experi-
, seem to be very uncertain. Thus, it is obvious that
m
tal results (temperature, number of determinations, mole fractions at
equilibrium in the liquid phase, x , and equilibrium ratios K ) of the
i
x
17
mental work. Therefore, we systematically measured a set
equilibrium study (Tables S1–S4). See http://www.rsc.org/suppdata/
cp/b3/b301771k/
z On leave from the State Technical University, Samara, Russia.
x Permanent address: Department of Chemistry, Carleton University,
Ottawa, Canada K1S 5B6.
g
l
ꢁ
of molar enthalpies of vaporization D H (298.15 K) for ben-
m
1
zyl halides (see Fig. 1) by using the transpiration method and
7
1
8,25,46
GC-correlation method. These new results (see column
1
3, Table 1) were checked for internal consistency and have
7
DOI: 10.1039/b301771k
Phys. Chem. Chem. Phys., 2003, 5, 2605–2611
This journal is # The Owner Societies 2003
2605

View Online
ꢂ1
Table 1 Experimental results for benzyl halides at 298.15 K in kJ mol (values selected for the calculation of the gaseous enthalpies of formation
are given in italic)
ꢁ
g
l
ꢁ
ꢁ
g
S
D H
f m
(l)
D H
m
D
f
H
m
(g)
H
1
2
3
4
5
1
0
3
11
Benzyl fluoride (1a)
ꢂ172.6 ꢃ 0.7
ꢂ32.6 ꢃ 2.6
44.5 ꢃ 0.4
17
4
6.22 ꢃ 0.26
ꢂ126.4 ꢃ 0.7
17.5 ꢃ 2.6
14.5
1
11
Benzyl chloride (1b)
Benzyl bromide (1c)
50.1 ꢃ 0.5
k
5
1.33
1
3
5
1.1 ꢃ 2.0
i
17
f j
l
(
ꢂ21.7)
50.12 ꢃ 0.27
(28.4 ꢃ 1.5)
(6.3)
a
k
22.5
3
53.14
b
11
0.8 ꢃ 3.9
50.5 ꢃ 0.5
47.3 ꢃ 4.2
5
48
3
0 ꢃ 14
6
2
1
2.0 ꢃ 4.4
6.2 ꢃ 2.2
h
c
17
(
26.3 ꢃ 4.8)
53.27 ꢃ 0.67
79.6 ꢃ 4.9
6.4
a
11
9
benzyl iodide (1d)
67.7
50.6 ꢃ 1.4
127.3 ꢃ 1.3
d
7
5
0.3 ꢃ 1.4
3.0 ꢃ 1.8
h
c
17
17
(
69.0 ꢃ 2.6)
57.38 ꢃ 0.37
52.79 ꢃ 0.24
126.4 ꢃ 2.6
ꢂ5.4 ꢃ 1.8
0.9
5.2
14
(1-Chloroethyl) benzene (2a)
ꢂ58.2 ꢃ 1.8
ꢂ60.6 ꢃ 7.2
ꢂ18.9 ꢃ 7.2
e
e
17
17
17
(
1-Bromoethyl) benzene (2b)
1-Iodoethyl) benzene (2c)
56.38 ꢃ 0.28
59.90 ꢃ 0.35
54.70 ꢃ 0.50
37.5 ꢃ 7.2
(96.4 ꢃ 1.5)
ꢂ35.9 ꢃ 4.8
1.5
2.3
i
f
f
(
(36.5)
e
4
e
a,a-Dimethylbenzyl chloride (3a)
ꢂ90.6 ꢃ 4.7
ꢂ76.0 ꢃ 3.0
ꢂ47.2 ꢃ 5.5
12.1
7
17
a,a-Dimethylbenzyl bromide (3b)
a,a-Dimethylbenzyl iodide (3c)
58.00 ꢃ 0.50
63.30 ꢃ 0.50
10.8 ꢃ 5.6
11.2
4.4
i
17
(3.1)
(66.4 ꢃ 1.5)
a
8
b
Value obtained using data from Gellner and Skinner and results from this work (see text). Value obtained using data Benson and Buss and
4
c
enthalpy of vaporization from this work. Value has been calculated as the average from the available results; the uncertainty is twice the sd of the
d
mean. Value obtained using data from Walsh et al. and enthalpy of vaporization from this work. Derived in this work (see text) using the
9
e
15,16
values of experimental enthalpies of reactions (V–VIII)
f
g
Calculated in this work using eqn. (11). Strain enthalpy of benzyl derivatives
.
ꢁ
h
7
i
H
S
¼ {D
f
H
m
(g) ꢂ S Benson’s increments}. This result was disregarded by calculation of the average value. Assessed in this work from the
ꢁ
g
l
ꢁ
j
from colmns 4 and 3. Calculated in this work using eqn. (11) and considered to be more reliable
difference of selected values D
H
f m
(g) ꢂ D H
m
k
l
as the existing experimental value. Calculated in this work using Antoine coefficients from ref. 12 and the procedure developed in ref. 17.
Expected value of strain (see text).
2
2
ꢂ1
ꢂ1
been used in this work for calculation of the gaseous enthalpies
ꢁ
Papina et al., ꢂ(154.9 ꢃ 3.0) kJ mol , and an older one from
23
of formation D H (g) of the benzyl derivatives (see column 4,
Parks et al., ꢂ(161.0 ꢃ 1.3) kJ mol . Both were measured by
combustion calorimetry in the laboratories of impeccable
reputation, but their disagreement is larger than the experi-
mental uncertainties! In order to resolve this contradiction,
we decided to obtain the enthalpy of formation of benzyl alco-
hol in an independent way. Indeed, a study of the chemical
equilibrium in the reactive system
f
m
Table 1). Due to the internal consistency and the high preci-
ꢁ
ꢂ1
g
sion (ca. ꢃ0.5 kJ mol ) of our new results for D H
l
298.15 K), the new values of the gaseous enthalpies of forma-
m
(
tion (column 4, Table 1) should lend themselves to better inter-
pretation.
In order to get insight into reasons for scatter in the data-
base of enthalpies of formation of the benzyl halides in the
liquid phase (column 2, Table 1), a careful analysis of the pri-
4
–10
mary experimental data
has been performed. This analysis
ðIVÞ
revealed some possible ways for correction of the available
results. The first is referred to enthalpies of formation of ben-
zyl bromide 1b and benzyl iodide 1c, which were obtained from
ꢁ
ꢁ
gives the value of the reaction enthalpy, D (l) which,
according to the second law of thermodynamics, could serve
to obtain the enthalpy of formation, D H
r
H
m
the calorimetric measurements of enthalpies, D
following reactions:
r
H
m
, of the
ꢁ
(l) of benzyl
m
f
C6H5-CH2-Br þ H2O ! HBr þ C6H5-CH2-OH
alcohol 1e, provided that enthalpies of formation of the other
reaction participants (tert-amyl benzyl ether 1f and 2-methyl-
butene-2) are known. The enthalpy of formation of 2-methyl-
butene-2 is well established. Thus, combustion experiments
on tert-amyl benzyl ether 1f should provide the knowledge of
the enthalpy of formation of benzyl alcohol 1e independently,
ꢁ
ꢂ1
DrH m ¼ ꢂ7:9 kJ mol
;
ðIIÞ
2
4
C6H5-CH2-I þ H2O ! HI þ C6H5-CH2-OH
ꢁ
ꢂ1
DrH m ¼ 12:6 kJ mol
There reactions were carefully studied by Gellner and Skin-
:
ðIIIÞ
and will help to resolve the contradiction between the two
8
ner in the liquid phase at 298.15 K. Unfortunately, the origi-
22,23
values available from the literature.
lished a reliable enthalpy of formation for benzyl alcohol,
Then, having estab-
ꢁ
nal values of D
Skinner and cited in the literature
f
H
m
(l) for 1b and 1c derived by Gellner and
8
30,48
8
the results from Gellner and Skinner can be corrected and
are not particularly
valid, because they were calculated using an uncertain value
used for estimation of reliable values of enthalpies of
formation of the bromide 1b and iodide 1c.
The second helpful way to understand trends in thermo-
chemistry of benzyl halides 1a–d is to consider their parent
compounds, such as a-methyl-substituted benzyl halides
th
for benzyl alcohol, C H -CH -OH, measured in the 19 cen-
6
5
2
1
tury. A literature search for benzyl alcohol has revealed that,
9
2
0,21
besides some other archival values of ambiguous quantity,
there are only two values to merit attention: a recent one from
2
606
Phys. Chem. Chem. Phys., 2003, 5, 2605–2611

View Online
ꢁ
ꢂ1
;
2
halides) 3a–c. Thus, in this work we have derived enthalpies
a–c, and a,a-dimethyl-substituted benzyl halides (cumyl
DrH m ðlÞ ¼ ð42:7 ꢃ 0:8Þ kJ mol
ꢁ
ꢂ1
ꢂ1
ꢁ
DrS m ðlÞ ¼ ð155:2 ꢃ 2:4Þ J K mol
In spite of the fact that equilibrium studies have been per-
formed at elevated temperatures (303 to 371 K), any correc-
tions of the reaction enthalpies are negligible when taking
of formation, D
f
H
m
(l), of the benzyl derivatives 2 and 3
see Table 1) using experimental results from equilibrium stu-
(
dies.
1
5,16
Hence, our additional efforts provide a broader data
ꢁ
set on the gaseous enthalpies of formation D
f
H
m
(g) of benzyl
26
derivatives, which can be used to derive carbon–halogen bond
dissociation enthalpies.
ꢂ1
into account the individual error bars of ca. 1–2 kJ mol typi-
cal for the equilibrium measurements. In the above, it was
assumed that the change in the enthalpy of reaction is negli-
gible on passing from the average temperature of the experi-
mental range to T ¼ 298.15 K.
Experimental
Materials
Combustion calorimetry
Benzyl alcohol and 2-methylbutene-2 (2MB2) were of commer-
cial origin. Tert-amyl benzyl ether (1f) was synthesised via
alkylation of benzyl alcohol with 2MB2 [reaction (IV)] in the
The pure sample of tert-amyl benzyl ether 1f was obtained by
repeated distillation under N using a spinning-band column at
2
reduced pressure. No impurities could be detected in tert-amyl
benzyl ether by GC. For combustion experiments the purified
liquid samples were dried over molecular sieves and distilled
once more before combustion. This procedure provided col-
ourless material, and the absence of water was shown by Karl
Fischer titration. For measurements of the energies of combus-
tion of the benzyl ethers a rotating-bomb calorimeter described
presence of a catalytic amount of cation exchange resin in
form (Amberlist 15, Aldrich) at room temperature and
+
H
was purified by repeated distillation under reduced pressure.
GC analysis of the sample for thermochemical measurements
gave a purity > 99.99%.
2
8
elsewhere was used. The combustion products were exam-
ined for carbon monoxide (Dr a¨ ger tube) and unburned car-
bon, but none was detected. For converting the energy of the
actual bomb process to that of an isothermal process, and
Chemical equilibrium study
The chemical equilibrium of [reaction (IV)] was studied in the
liquid phase in the temperature range (303 to 371 K) using
glass vials with screw caps. Vials were filled with the initial
liquid mixture of alkanols and 2-methylbutene-2. Cation-
exchange resin Amberlist 15 (Aldrich) in H form was added
as a heterogeneous catalyst. Each vial was thermostatted at
2
9
reducing to standard states the conventional procedure was
applied. Results of a typical combustion experiment for 1f
and the mean values of the standard specific energies of com-
+
ꢁ
bustion D u , together with their standard deviations, are are
c
T
i
ꢃ 0.1 K and periodically shaken. After definite time inter-
deposited as electronical supplementary information.y We
ꢁ
vals the vial was cooled rapidly in ice and opened. A sample
for GC analysis was taken from the liquid phase using a syr-
inge. The thermostatting of the vial then proceeded at the
same temperature. Samples were taken successively until no
further change of composition was observed, indicating that
the chemical equilibrium was established. The equilibrium
ratio Kx was obtained from the current concentrations of
the reaction participants, which were analysed with a Hew-
lett-Packard gas chromatograph 5890 Series II equipped with
a flame ionisation detector and a capillary column HP-5. The
response factors of all reagents were determined using cali-
bration mixtures of the corresponding components prepared
gravimetrically.
obtained the resulting value D H
(l) ¼ ꢂ(263.8 ꢃ 3.0) kJ
f
m
ꢂ1
mol (the uncertainty assigned is twice the overall standard
deviation and includes the uncertainties from calibration, from
the combustion energies of the auxiliary materials, and the
uncertainties of the enthalpies of formation of the combustion
products H O and CO ).
2
2
Results and discussion
Enthalpy of formation of benzyl alcohol
We investigated the chemical equilibrium of reaction (IV) of
synthesis of tert-amyl benzyl ether from benzyl alcohol and
The experimental results (temperature, number of determi-
nations, mole fractions at equilibrium in the liquid phase, xi ,
and equilibrium ratios K ) of the equilibrium study are depos-
x
2
-methylbutene-2 and derived the enthalpy of this reaction in
ꢁ
the liquid phase. The enthalpy of formation D
tert-amyl benzyl ether at 298.15 K was measured by means
of combustion calorimetry in this work. Using D
ꢂ(68.1 ꢃ 1.3) kJ mol for 2MB2 available in literature we
calculated the enthalpy of formation of benzyl alcohol 1e in
the liquid phase:
f
H
m
(l) of
ited as electronical supplementary information.y It was well
ꢁ
2
6
f
H
m
(l) ¼
documented in our previous work that alkyl ether synthesis
mixtures behave non-ideally, and that the equilibrium ratio
ꢂ1
24
x
K strongly depends on the composition of the equilibrium
2
6
mixture. However, it has been observed for systems similar
to reaction (IV) that K –values are almost independent of
the mole fraction of alkanol if xAlkOH > 0.5. For such systems
it has been proven that the equilibrium ratio K (thermo-
ꢄ K
dynamic equilibrium constant) for mixtures with x
ꢁ
ꢁ
ꢁ
ꢁ
x
Df H m ðlÞ
¼ DrH m ðlÞ ꢂ Df H m ðlÞ
þ Df H m ðlÞ
ð1eÞ
ð2MB2Þ
ꢂ1
ð1fÞ
¼
This value is in very close agreement to the recent result
ꢂð153:5 ꢃ 3:4Þ kJ mol
:
x
a
>
AlkOH
0
.5, by comparison of the results of equilibrium studies in
22
ꢂ1
from Papina et al. ꢂ(154.9 ꢃ 3.0) kJ mol measured by
23
2
6
the gaseous and liquid phases simultaneously. Taking this
experience into account, we investigated reaction (IV) in an
excess of alcohols (the summarised mole amount of benzyl
alkohol and butanol was xAlkOH > 0.5.) Experimental values
combustion calorimetry and inconsistent with the older one
from Parks et al. Thus, our value helps to resolve the contra-
diction between two available results in the literature. Taking
into account the close agreement of our and Papina et al.’s
22
of K
the linear equation ln K
x
were approximated as a function of temperature by
¼ a + bꢅ(T/K) , using a least
ꢁ
result on D
precise and direct result from Papina et al.
f
H
m
(l) of benzyl alcohol, we prefer the more
ꢂ1
a
22
squares fit. The slopes of these lines when multiplied by the
ꢁ
gas constant yield the enthalpy of reaction (IV) D
r
H
m
and
. Numerical
ꢁ
Enthalpies of formation of benzyl bromide (1c) and benzyl
8
iodide (1d). Gellner and Skinner measured enthalpies of the
the intercept gives the entropy of reaction D
results are as follows:
r
S
m
hydration reactions (II) and (III) calorimetrically at 298.15
ꢁ
ln Ka ¼ ꢂ5:14ꢅð1000=TÞ þ 18:67
K. Having established in this work a more reliable D
f
H
m
(l)
Phys. Chem. Chem. Phys., 2003, 5, 2605–2611
2607

View Online
3
2
of benzyl alcohol, we were able to derive new values (see Table
ꢁ
macher equation at the reference temperature 298.15 K). The
enthalpies of vaporization estimated using this procedure were
1
) of D H (l) of benzyl bromide (1c) and benzyl iodide (1d)
f
m
8
ꢁ
g
ꢁ
ꢂ1
g
ꢁ
using calorimetric results and the chosen D
f
H
m
(l) ¼
D H
m
(298.15 K) ¼ (16.4 ꢃ 1.0) kJ mol for HCl and D H
m
l
l
ꢂ1
22
ꢂ1
ꢂ(154.9 ꢃ 3.0) kJ mol for benzyl alcohol. A summary of
(298.15 K) ¼ (17.2 ꢃ 1.0) kJ mol for HBr. Thus, the desired
ꢁ
ꢂ1
ꢁ
the results available for the benzyl halides is presented in Table
1
values D
f
H
m
(l) ¼ ꢂ(108.7 ꢃ 1.0) kJ mol for HCl D
f
H
m
ꢂ1
. As can be seen from this table, enthalpies of benzyl bromide
1c) and benzyl iodide (1d) derived in this work are in agree-
(l) ¼ ꢂ(53.6 ꢃ 1.0) kJ mol for HBr were calculated as the
ꢁ
g
l
ꢁ
(
ment with those already available from the literature, within
difference between D
f
H
m
(g) and D H
(l), of the benzyl derivatives 2a–b and 3a–b
m
. Enthalpies of for-
ꢁ
mation, D
f
H
m
the range of their experimental uncertainties. Thus, for both
ꢁ
were calculated using the enthalpies of reactions (V)–(VIII)
and enthalpies of formation of the reaction participants are
listed in Table 1. The validity of the procedure applied to
compounds the average values of D
f
H
m
(l) have been calcu-
lated and used to derive gaseous enthalpies of formation
ꢁ
D
f
H
m
(l) (column 4, Table 1). It is worth mentioning that
ꢁ
derive enthalpies of formation of 2a–b and 3a–b is confirmed
ꢁ
experimental results for D
Ashcroft et al. from calorimetrically measured hydrogenation
f
H
m
(l) of 1c and 1d obtained by
by the close agreement of our result for 2a, D
f
H
m
(l)2a
¼
7
ꢂ1
ꢂ(60.6 ꢃ 7.2) kJ mol , and that measured by combustion
1
4
ꢁ
ꢂ1
.
enthalpies of benzyl halides are systematically lower by over
0 kJ mol . An explanation of this fact is still evasive, but
calorimetry: D H (l)2a ¼ ꢂ(58.2 ꢃ 1.8) kJ mol
f
m
ꢂ1
ꢁ
1
For a,a-dimethylbenzyl chloride (3a) the value of D
f
H
m
ꢂ1
it might be due to incorrect calibration of their device. Thus,
results from this work were disregarded by calculation of the
average values given in Table 1.
(l)3a ¼ ꢂ(76.0 ꢃ 3.0) kJ mol was obtained from the calori-
metrically measured enthalpy of hydrochlorination of a-
4
7
methyl-styrene in solvent (methylene chloride). This value
is in disagreement with the result obtained in this work under
ꢂ1
Enthalpies of formation of benzyl derivatives 2 and 3. Kov-
investigated the equilibrium of hydrohalogenation of
neat conditions, ꢂ(90.6 ꢃ 4.7) kJ mol . The disagreement
1
5,16
sel
styrene and a-methyl-styrene in the liquid phase using SnCl
might be due to possible rapid polymerisation of a-methyl-
styrene in the conditions of calorimetric study. However,
4
7
4
and CuCl as catalyst. As those results are not readily available,
they are presented briefly below:
the consistency of thermochemical result for 2a, obtained from
and those from combustion calori-
metry allowed us to give preference to the value derived in
this work.
1
5,16
the equilibrium study
4
1
C6H5 CH
B
CH2 þ HCl $ C6H5 CHCl CH3;
ꢁ
ꢂ1
T-range: 353 413 K Df H m ¼ ꢂ55:7 ꢃ 7:1 kJ mol
;
;
ꢁ
ꢂ1
ꢂ1
ꢁ
ꢁ
H
m
Df S m ¼ ꢂ125:3 ꢃ 14:7 J mol
K
ðVÞ
ðVIÞ
Correlation of D
f
H
m
(g) of benzyl halides with D
f
(g) of
(g) of
ꢁ
alkyl analogs. Gaseous enthalpies of formation D
benzyl derivatives 1–3 were derived in this work as the sum
f
H
m
C6H5 CH
B
CH2 þ HBr $ C6H5 CHBr CH3;
ꢁ
ꢁ
ꢂ1
of the selected D
f
H
m
(l) and enthalpies of vaporization listed
T-range: 383 413 K Df H m ¼ ꢂ69:1 ꢃ 7:1 kJ mol
in Table 1. Taking into account the high chemical reactivity
and thermal lability of the benzyl derivatives, which could
aggravate thermochemical measurements, proof of the relia-
bility of results derived in this work would be desirable. The
ꢁ
ꢂ1
ꢂ1
K
Df S m ¼ ꢂ125:7 ꢃ 14:7 J mol
C6H5 CðCH3Þ CH2 þ HCl $ C6H5 CðCH3ÞCl CH3;
B
ꢁ
ꢂ1
;
T-range: 333 413 K Df H m ¼ ꢂ52:0 ꢃ 4:6 kJ mol
ꢁ
set of D H (g) of the benzyl derivatives derived in this work
f
m
ꢁ
ꢂ1
ꢂ1
K
Df S m ¼ ꢂ130:3 ꢃ 13:0 J mol
ðVIIÞ
could be checked for internal consistency using gaseous enthal-
pies of formation of alkyl halides. In contrast to benzyl
halides, experimental thermochemical data on alkyl halides
C6H5 CðCH3Þ CH2 þ HBr $ C6H5 CðCH3ÞBr CH3;
B
3
0
ꢁ
ꢁ
ꢂ1
;
are well established. The set of D H (g) of alkyl halides
T-range: 363 413 K Df H m ¼ ꢂ63:7 ꢃ 5:4 kJ mol
f
m
3
0
C
1
–C
4
recommended by Pedley et al. has been successfully
ꢁ
ꢂ1
ꢂ1
K
Df S m ¼ ꢂ134:1 ꢃ 13:8 J mol
ðVIIIÞ
checked for internal consistency in the review by Slayden
4
9
ꢁ
m
et al. recently. We used their set of selected values D
f
H
In spite of the fact that equilibrium studies have been per-
formed at elevated temperatures, corrections of the reaction
enthalpies to the reference temperature 298.15 K are assessed
(
g) of alkyl halides C –C (see electronic supplementary
2
4
hi naf loi dr me s a. tiony) for correlation with our results on benzyl
Now let us consider structures of the benzyl derivatives pre-
sented in Fig. 1. It is obvious, that benzyl derivatives (1) are
ꢂ1
as not larger than 1.5 to 2 kJ mol , and match the individual
error bars of the equilibrium measurements. In further calcula-
tions it was assumed that the enthalpy of reaction hardly
changes on passing from the average temperature of the
experimental range to T ¼ 298.15 K.
parent to the structure of the ethyl derivatives C
example, PhCH -Cl (1b) and C H -Cl. Benzyl derivatives (2)
2 5
H -X, for
2
2
5
are parent to the structure of the isopropyl derivatives iso-
C H -X. Benzyl derivatives (3) are parent to the structure of
Enthalpies of formation of the benzyl derivatives 2 and 3
were calculated using the enthalpies of reactions (V)–(VIII).
For this purpose some additional thermochemical data have
3
7
4 9
the tert-butyl derivatives tert-C H -X. Thus, the correlation
ꢁ
of the enthalpies of formation of 1–3 with those of alkyl deri-
vatives alkyl-X, where X ¼ halogen should be linear if the
data used for the correlation are reliable. As can be seen from
Fig. 2, the data for benzyl and alkyl derivatives fit very well
(except for benzyl chloride) in the linear correlation:
been involved: enthalpy of formation D
f
H
m
(l) ¼ (103.8 ꢃ
ꢂ1
30
ꢁ
.1) kJ mol for styrene, and D H
m
1
(l) ¼ (70.07 ꢃ 0.82)
f
ꢂ1
31
kJ mol for a-methyl-styrene. Only the gaseous enthalpy
ꢁ
ꢂ1
of formation for HCl D H (g) ¼ ꢂ(92.31 ꢃ 0.13) kJ mol
f
m
ꢁ
ꢂ1
and for HBr D
f
H
assigned by CODATA. However, in order to obtain their
m
(g) ¼ ꢂ(36.38 ꢃ 0.17) kJ mol
were
2
ꢁ
ꢂ1
Df H m ðg; 298:15 K; benzyl-XÞ=kJ mol
ꢁ
g
l
ꢁ
D
f
H
m
(l) in the liquid phase, vaporization enthalpies, D H
are required. We calculated the latter using parameters of
m
ꢁ
¼
ð0:947 ꢃ 0:014Þꢅ½Df H m ðg; 298:15 K; alkyl-XÞꢆ
þ ð134:5 ꢃ 1:1Þ
the Antoine equation for vapor pressure from a literature
compilation with the help of the equation:
ð11Þ
1
2
32
Such a relationship with correlation coefficient (r ¼ 0.999) is
g
l
ꢁ
ꢂ1
2
ꢂ2
evidence of the internal consistency of the experimental results
for the benzyl derivatives listed in Table 1; only D H (g) of
D H m ðJ mol Þ ¼ ð2:3ꢅRꢅDZꢅBꢅT ÞꢅðT þ CÞ
ꢁ
f
m
ꢂ1
ꢂ1
where R ¼ 8.31451 J K mol ; B and C are Antoine equa-
tion parameters; DZ is the difference in compression factors
of vapor and liquid (DZ ꢄ 1 was calculated using the Haagen-
benzyl chloride (1b) is definitely outside this linear correlation.
However, there are no apparent reasons for such deviation and
taking into account that its enthalpy of formation (see Table 1)
2
608
Phys. Chem. Chem. Phys., 2003, 5, 2605–2611

View Online
Fig. 3 Conformational preferences of benzyl-X compounds.
as the sum of excess geminal interactions of the phenyl and
the halogen on the central C-atom.
Let us consider the row of the benzyl derivatives 1a–d which
are parent to ethylbenzene. Ethylbenzene itself is strainless.
However, benzyl fluoride exhibits the most noticeable inter-
Fig. 2 Relation between enthalpies of formation of alkyl halides and
0
ꢂ1
benzyl halides (in kJ mol ). X,S Relation between D
f
H
of C
2
H
5
-X
ꢂ1
ðgÞ
of C
action, amounting to 14.5 kJ mol . A plausible explanation
of the observed destabilization of benzyl fluoride was sug-
0
0
and D
f
H
PhCH
2
-X; / relation between D
)-X; L relation between D
-X.
f
H
2
H
7
-X and
-X and
ðgÞ
PhCH(CH
ðgÞ
0
0
36
gested by Penner et al. By means of spectral analyses and
D
D
f
H
H
3
f
H
of C
4
H
9
ðgÞ
ðgÞ
0
ðgÞ
f
PhC(CH
3
)
2
ab initio methods they determined the conformational prefer-
ences of some benzyl-X derivatives with the aim of establishing
whether stabilising interactions enhance the conformational
preference for the perpendicular structure (see Fig. 3). Except
for X ¼ F, compounds studied [X ¼ Cl, SH, SMe, S(O)Me]
adopted mainly the conformation in which the C–H bond is
perpendicular to the plain of the benzene ring. The authors
proposed therefore the existence of a ‘‘benzylic anomeric
effect’’, and that its magnitude is S(O)Me > Cl > SH,
SMe > F. The exceptional planar conformation of benzyl
1
3
is known only from a single work, it seems to be justifiable
that benzyl chloride belongs to the line defined by eqn. (II).
ꢁ
ꢂ1
That is why D
f
H
by using eqn. (11) and the enthalpy of formation of C H -Cl,
m
(g) ¼ (28.4 ꢃ 1.5) kJ mol of 1b, assessed
2
5
seems to be more reliable than the experimentally measured
ꢁ
ꢂ1
value D H (g) ¼ (17.5 ꢃ 2.6) kJ mol presented in Table
f
m
1
. In addition,, enthalpies of formation of compounds 2c
(g)
m
3
6
ꢁ
fluoride determined by Penner et al. and the profound desta-
bilization of 14.5 kJ mol for this molecule established in our
and 3c have been estimated using eqn. (11) and D H
f
ꢂ1
3
of the appropriate alkyl iodides and are presented in
0
thermochemical study are evidence that the possible ‘‘benzylic
anomeric effect’’ is destroyed by a dipole–dipole interaction
between the phenyl and fluorine substituents. A similar expla-
nation seems to be acceptable for the less profound destabiliza-
Table 1.
S
Strain enthalpies H of benzyl derivatives. Another possible
way to establish consistency of thermochemical properties of
benzyl derivatives is by comparison of strain enthalpies, which
can be derived from their gaseous molar enthalpies of forma-
tion at 298.15 K (Table 1). Indeed, the benzyl derivatives
ꢂ1
tion (ca. 6 kJ mol ) observed in benzyl chloride and benzyl
bromide, as well as for the nearly strainless (H ¼ 0.9 kJ
S
ꢂ1
mol ) benzyl iodide.
Benzyl derivatives 2a–c, which are parent to the isopropyl-
1
a–d, 2a–c, and 3a–c listed in Table 1 present a typical example
of similarly shaped molecules, where halogen is attached to the
appropriate benzyl moiety. Hence the strain, H , of a molecule
ꢂ1
benzene (H
1
S
¼ 5.6 kJ mol ) also show similar strain (H
S
¼
ꢂ1
.5 to 5.2 kJ mol ). Hence, it is reasonable to presume that
S
methyl groups are able to disorder the optimal orientation of
dipoles in the suggested planar conformation (see Fig. 3) of
a-substituted benzyl derivatives 2a–c and no additional desta-
bilization will arise in these molecules.
is expected to provide insight into the energetic interactions of
a halogen with the phenyl ring attached to the geminal
C-atom.
We define the strain enthalpy H
ence between the experimental enthalpy of formation D
S
of a molecule as the differ-
ꢁ
ꢂ1
Benzyl derivatives 3a–c are strained by ca. 10 kJ mol (see
f
H
m
Table 1) and the observed amount of destabilization could no
doubt be attributed to the inherent strain due to the steric
repulsions of methyl groups and the benzene ring attached to
(
ments
g) and the calculated sum of the Benson type incre-
for this molecule. By using these group-additivity
3
3,34,27
ꢁ
parameters and the values of D
(Table 1), the values of strain enthalpies H ¼ D H (g) ꢂ S
increments} have been estimated (Table 1).
Almost all of the benzyl derivatives listed in Table 1 are
strained. Understanding strain in benzyl derivatives is aided
by comparison to the strain in similarly shaped alkylbenzenes,
e.g. ethylbenzene, isopropylbenzene and tert-butylbenzene.
Enthalpies of formation D H (g) and strain enthalpies HS
of ethylbenzene (29.9 kJ mol and 0.6 kJ mol ) isopropyl-
benzene; (4.0 kJ mol and 5.6 kJ mol ), and tert-butylben-
f
H
m
(g) of benzyl derivatives
ꢁ
the central quaternary carbon atom, as in the similarly shaped
¼ 9.4 kJ mol . Thus, strain enthal-
S
f
m
ꢂ1
tert-butylbenzene with H
S
pies of the benzyl derivatives discussed above in terms of
deviation from additivity rules reveal reasonable similarities
within the structural pattern 1, 2, or 3, and establish consis-
ꢁ
tency of values for D
f
H
m
(g) derived in this work.
ꢁ
Calculation of the bond dissociation enthalpies of benzyl
halides. Benzyl halides are very reactive and thermolabile,
f
m
ꢂ1
ꢂ1
ꢂ1
ꢂ1
ꢂ1
ꢂ1
which is why these compounds are often used for determina-
37,38
zene; (ꢂ24.4 kJ mol and 9.4 kJ mol ) are known from
tion of the bond dissociation enthalpy (BDE).
In this work,
the literature. The enthalpies of formation of these compounds
gaseous enthalpies of formation of benzyl halides have been
derived according to eqn. (1) and listed in Table 1. Enthalpies
3
0
were taken from Pedley et al., except for tert-butylbenzene.
35
ꢁ
of formation of halogen-radicals D H (g, Hal ) were assigned
f m
2
ꢀ
These alkylbenzenes show relevant structural patterns of strain
in the benzyl derivatives studied. Their strain enthalpies
describe the intrinsic strain of the alkylbenzenes due to steric
repulsions of the alkyl groups and the benzene ring attached
to the tertiary or quaternary carbon atom. Comparison of
strain energies in the benzyl derivatives with those of alkyl-
benzenes allows the derivation of the strain effects directly.
by CODATA. Enthalpies of formation of benzyl-radicals:
ꢀ
ꢀ
ꢀ
PhCH , PhCH(CH ) , and PhC(CH ) were recommended
3
2
3 2
3
,39
in the recent literature.
of dissociation enthalpies of the C–Hal bond in the benzyl
Thus, using these data, the values
halides 1–3 were estimated and are presented in Table 2.
The BDE (PhCH F) ¼ 412.8 kJ mol derived in this work
ꢂ1
2
The values H
S
of the benzyl derivatives could be interpreted
is in excellent agreement with the experimental value from
Phys. Chem. Chem. Phys., 2003, 5, 2605–2611
2609

View Online
Table 2 Comparison of the BDE of Alkyl Halides and Benzyl Halides
Table 3 A consistent set of theoretically calculated values, using the
LLM method (values in parentheses were reported earlier using the
ꢂ1
38
BDE/kJ mol
same (LLM) method
ꢂ1
F
Cl
Br
I
BDE/kJ mol
F
a
a
a
a
CH
PhCH
CH
PhCH(CH
3
CH
2
-X
451.5
350.2
290.8
231.4
Cl
Br
b
2
-X
CH-X
412.8 ꢃ 4.1 299.9 ꢃ 4.3 239.3 ꢃ 6.3 187.8 ꢃ 4.8
a a a a
346.4
(
3
)
2
453.1
—
—
291.6
227.2
CH
CH
3
-X
462.1 (461.9)
473.7
482.5
354.9 (354.8)
356.8
358.9
312.9 (313.4)
311.3
310.3
c
3
d
)-X
292.4 ꢃ 4.4 240.1 ꢃ 8.2 176.1 ꢃ 4.3
351.6 ꢃ 3.8 292.4 ꢃ 3.5 226.9 ꢃ 4.5
292.8 ꢃ 6.3 236.7 ꢃ 6.9 176.0 ꢃ 4.3
3
2
CH -X
(
CH
3
)
PhC(CH
3
C-X
)
(CH
(CH
3
)
)
2
CH-X
C-X
e
3
2
-X
—
3
3
483.7
359.0
304.2
PhCH
Ph(CH
Ph(CH
2
-X
416.6 (417.1)
425.1
434.0
299.6 (296.2)
297.8
294.8
256.5 (257.7)
250.8
243.6
a
b
Values were taken from ref. 45. Calculated in this work from our
results for enthalpies of formation of benzyl halides (Table 1) using
3
3
)CH-X
C-X
)
2
3
eqn. (1) and enthalpies of formation of the benzyl radical
ꢂ1
ꢁ
D
f
H
m
2
(g) ¼ (207.0 ꢃ 4.0) kJ mol and the appropriate halogen radi-
ꢂ1
ꢂ1
;
cal: F ¼ (79.38 ꢃ 0.41) kJ mol ; Cl ¼ (121.301 ꢃ 0.008) kJ mol
ꢂ1
ꢂ1
c
Br ¼ (111.87 ꢃ 0.12) kJ mol ; I ¼ (106.76 ꢃ 0.04) kJ mol
.
Calcu-
ing, when data are drawn from many different sources. Here
we present an internally consistent set of BDEs for the alkyl
and benzyl halides, all calculated (or recalculated) using the
LLM method. These values are all shown in Table 3, where
earlier values obtained using the same LLM method are
shown in parentheses. Agreement with earlier values is usually
within 1 kJ mol except for benzyl chloride, which differs by
over 3 kJ mol . Such differences can arise from different con-
formations, so it is necessary to specify the AM1-preferred
conformer used in the calculation.
lated in this work from our results for enthalpies of formation of ben-
zyl halides (Table 1) using eqn. (1) and enthalpies of formation of
39
ꢁ
ꢂ1
a-methyl-benzyl radical D
appropriate halogen radical.
f
H
m
d
(g) ¼ (165.7 ꢃ 4.0) kJ mol and the
2
38
Calculated in this work from enthal-
30
pies of formation of tert-butyl halides (see supporting information )
3
using eqn. (1) and enthalpy of formation of tert-butyl-radical
ꢂ1
ꢂ1
ꢁ
D
cal.
f
H
m
(g) ¼ (48.0 ꢃ 3.0) kJ mol and the appropriate halogen radi-
ꢂ1
2
e
Calculated in this work from our results for enthalpies of for-
mation of benzyl halides (Table 1) using eqn. (1) and enthalpies of
ꢁ(g) ¼ (135.6 ꢃ
39
formation of the a,a-dimethyl-benzyl radical
f m
D H
2
For benzyl fluoride, there are AM1 minima at the planar
and perpendicular positions of the F atom, relative to the phe-
nyl ring. The planar conformer is favoured by 3.8 kJ mol , in
ꢂ1
4.0) kJ mol and the appropriate halogen radical.
ꢂ1
agreement with the experiment. For benzyl chloride and benzyl
bromide, only the perpendicular conformer (C–X bond
perpendicular to the ring) is stable. For PhCH(CH )X, the
3
4
0
ꢂ1
Zavitsas (412.5 ꢃ 8.4) kJ mol . However, theoretical studies
of the BDE in benzyl fluoride are in disagreement. Earlier
3
B3LYP-based calculations of this quantity gave 406.7 kJ
7
F-atom lies in the ring plane, whereas the Cl and Br atoms
ꢁ
ꢂ1
ꢂ1
37
mol and a value of 402.5 kJ mol was calculated by the
have dihedral angle 60 with the ring. For PhC(CH
3
)
F atom is in the ring plane. In the chloride, the methyl carbon
2
X, the
semiempirical AM1 method. The most recent result of 417.1
kJ mol was calculated using the lowest level method
LLM) of DiLabio et al., which is a hybrid of the B3LYP
ꢁ
ꢂ1
lies in the ring plane and the C–Cl bond has dihedral angle 60 .
There is another conformer with the Cl atom planar, but this
(
is less stable by 4 kJ mol ). The bromide is identical to the
(
method (for the energy) and the AM1 method (for geometry
ꢂ1
ꢁ
3
and frequency). The BDE (PhCH
8
ꢂ1
chloride, with dihedral angle 60 and two conformers.
2
–Cl) ¼ 299.9 kJ mol
Table 3 shows that there are systematic trends for BDEs in
the alkyl halides and in the benzyl halides. In particular,
increasing methyl substitution in the alkyl halides: (i) causes
the C–F BDE to increase, although diminishing returns are
reached for the tertiary methyl group, (ii) causes the C–Cl
BDE to increase slightly then reach a plateau, and (iii) causes
derived in this work may be compared with experimental
ꢂ1
ꢂ1
values 289.1 kJ mol and 299.6 kJ mol (electrochemical
4
1
ꢂ1
ꢂ1
measurements)
and calculated values 287.0 kJ mol
ꢂ1
3
7
37
B3LYP), 278.6 kJ mol (AM1), and 296.2 kJ mol
(
(
3
LLM). For benzyl bromide, our derived C–Br BDE is
8
ꢂ1
2
39.3 kJ mol , which may be compared with the following
experimental values: 233.9 or 222.2 kJ mol (electrochemical
ꢂ1
3 3
the C–Br BDE to decrease slightly. The point for (CH ) C–Br
4
measurements), 254.0 kJ mol (PAC measurements), and
1
ꢂ1
42
seems anomalous but the other trends are smooth. Increasing
methyl substitution in the benzyl halides (i) causes the C–F
BDE to increase; (ii) causes the C–Cl BDE to decrease
(slightly); and (iii) causes the C–Br BDE to decrease (signifi-
cantly) with increasing substitution.
The agreement between our calculated BDE values and the
experimental values is essentially within experimental error for
C–F and C–Cl bonds, although there are larger errors for C–
Br bonds. There are several possible reasons for this in the the-
oretical calculation: (i) The AM1 geometries and frequencies
for C–Br do not contain as many molecules in the calibration
set for AM1, hence an inferior set of AM1 parameters; (ii) the
ꢂ1
2
54.8 kJ mol (thermolysis of the benzyl bromide in the gas
phase). Calculated values are 238.9 kJ mol (B3LYP),
4
2
ꢂ1
37
ꢂ1
37
ꢂ1
38
32.6 kJ mol (AM1), and 257.7 kJ mol (LLM). The
2
BDE (PhCH
ꢂ1
2
–I) ¼ 187.8 kJ mol derived in this work is close
ꢂ1
to another experimental value of 190.0 kJ mol from shock-
wave dissociation measurements.
Comparison of the BDE’s of alkyl halides and benzyl
halides, presented in Table 2, indicates that the benzylic bond
strength is substantially lower (30–60 kJ mol ) than those in
4
3
ꢂ1
the similarly shaped alkyl analogues. However, the general
trends, such as the weakest bond in the I-derivatives and the
strongest bond in the F-derivatives, remain common for both
chemical families. It can also be seen that values for the BDE
of PhCH -X, PhCH(CH )-X and PhC(CH ) -X, are generally
6
-311+G(2d,2p) basis set may be inadequate for atoms with
atomic number as high as Br; (iii) spin–orbit coupling may
play a role in calculation of the Br radical especially. Without
extensive testing it is difficult to be sure which of these effects is
dominant, so we consider it reasonable to expect errors to
increase as we calculate values for Br and I.
2
3
3 2
2
independent of the branching of benzylic C-atom (CH , CH,
or C) similarly with their alkyl analogues having the same
substituent X. This fact again supports the consistency of the
thermochemical data derived in this work and involved in
the estimation of the BDE of the benzyl derivatives.
Conclusions
An internally consistent set of theoretical values of the BDE.
It is difficult to be completely confident about trends in bond-
In recent years considerable activity has taken place with
3
7,38
respect to the estimation of the BDE’s
and enthalpies of
2
610
Phys. Chem. Chem. Phys., 2003, 5, 2605–2611

View Online
4
4
formation of radicals using semi-empirical as well as ab initio
methods. As a rule, the estimates should be compared with
reliable experimental values when available. An extended
study of the thermochemistry of benzyl halides has been per-
formed in this work, and a set of reliable and consistent experi-
mental data has been derived. These data could serve for
the improvement of any empirical or ab initio methodology
for estimation of the thermodynamic properties of organic
compounds.
20 P. Landrieu, F. Baylocq and J. R. Johnson, Bull. Soc. Chim. Fr.,
929, 45, 36.
M. Schmidlin, Ann. Chim. Phys., 1906, 1, 195.
1
21
2
2
T. S. Papina, S. M. Pimenova, V. A. Luk’yanova and V. P.
Kolesov, Russ. J. Phys. Chem. Transl. of Zh. Fiz. Khim., 1995,
6
9, 1951. In original 2148.
23 G. S. Parks, K. E. Manchester and L. M. Vaughan, J. Chem.
Phys., 1954, 22, 2089.
2
4
W. D. Good and N. K. Smith, J. Chem. Thermodyn., 1979,
1, 111.
1
2
5
H. M. McNair and E. J. Bonelli, Basic Gas Chromatography,
Consolidated Printers, , Oakling, California, 1967.
2
2
6
7
A. Heintz and S. P. Verevkin, Fluid Phase Equilib., 2001, 179, 85.
E. S. Domalski and E. D. Hearing, J. Phys. Chem. Ref. Data,
Acknowledgements
1
993, 22, 805–1159.
S. V. expresses his gratitude to Prof. Dr C. R u¨ chardt and Dr
H.-D. Beckhaus for supporting the combustion experiments
and for the exceedingly creative atmosphere in their labora-
tory. One of us (E.K.) gratefully acknowledges a research
scholarship from the DAAD (Deutscher Akademischer Aus-
tauschdienst).
2
8
H.-D. Beckhaus, G. Kratt, K. Lay, J. Geiselmann, C. R u¨ chardt,
B. Kitschke and H. Lindner, Chem. Ber., 1980, 113, 3441.
29 W. N. Hubbard, D. W. Scott and G. Waddington, Experimental
Thermochemistry, ed. F. D. Rossini, Interscience, New York,
1
956, pp. 75–127.
3
0
J. B. Pedley, R. D. Naylor and S. P. Kirby, Thermochemical Data
of Organic Compounds, Chapman and Hall, London, 2nd edn.,
1986.
S. P. Verevkin, Thermochim. Acta, 1999, 326, 17.
3
1
References
32 R. C. Reid, J. M. Prausnitz and T. K. Sherwood, The Properties of
Gases and Liquids, 3rd edn., McGraw-Hill, New York, 1977,
p. 249.
33 S. W. Benson, Thermochemical Kinetics, Wiley, New York, 1976,
p. 274.
1
2
D. F. McMillen and D. M. Golden, Annu. Rev. Phys. Chem.,
982, 33, 493.
CODATA Key Values for Thermodynamics, ed. J. D. Cox,
1
D. D. Wagman and V. A. Medvedev, Hemisphere, New York,
1989.
34 N. Cohen, J. Phys. Chem. Ref. Data, 1996, 25, 1411.
35 S. P. Verevkin, J. Chem. Thermodyn., 1998, 30, 1029.
36 G. H. Penner, T. Schaefer, R. Sebastia and S. Wolfe, Can.
J. Chem., 1987, 65, 1845.
37 D. A. Pratt, J. S. Wright and K. U. Ingold, J. Am. Chem. Soc.,
1999, 121, 4877.
38 G. A. DiLabio and D. A. Pratt, J. Phys. Chem. A, 2000, 104, 1938.
39 S. G. Lias, J. E. Bartmess, J. F. Liebman, J. L. Holmes, R. D.
Levin and W. G. Mallard, J. Phys. Chem. Ref. Data, 1988,
17(Suppl. 1).
3
W. Tsang, in Energetics of Organic Free Radicals, ed. J. A.
Martinho Simoes, A. Greenberg and J. F. Liebman, Blackie
Academic and Professional, London, 1996, pp. 22–58.
S. W. Benson and J. H. Buss, J. Phys. Chem., 1957, 61, 104.
P. A. Fowell and C. T. Mortimer, J. Chem. Soc., 1961, 3793.
T. Holm, J. Organomet. Chem., 1973, 56, 87.
4
5
6
7
S. J. Ashcroft, A. S. Carson and J. B. Pedley, Trans. Faraday Soc.,
1963, 59, 2713.
8
9
O. H. Gellner and H. A. Skinner, J. Chem. Soc., 1949, 1145.
R. Walsh, D. M. Golden and S. W. Benson, J. Am. Chem. Soc.,
40 A. A. Zavitsas, J. Phys. Chem., 1987, 91, 5573.
41 C. P. Andreiux, A. Le Gorande and J.-M. Saveant, J. Am. Chem.
Soc., 1992, 114, 6892.
1966, 88, 650.
1
0
F. Schaffer, S. P. Verevkin, H.-J. Rieger, H.-D. Beckhaus and
C. R u¨ chardt, Liebigs Ann. Chem., 1997, 1, 1333.
S. J. Ashcroft, J. Chem. Eng. Data, 1976, 21, 397.
D. R. Stull, Ind. Eng. Chem., 1947, 39, 517.
F. W. Kirkbride, J. Appl. Chem., 1956, 6, 11.
A. T. Hu and G. C. Sinke, J. Chem. Thermodyn., 1969, 1, 507.
E. N. Kovzel, PhD Thesis, Saratov State University, Saratov,
USSR, 1981.
T. N. Nesterova, E. N. Kovzel, S. Ya. Karaseva, A. M. Rozhnov,
Heats of reaction of the hydrohalogenation of styrene and
a-methylstyrene, Vses. Konf. Kalorim., Rasshir. Tezisy Dokl.
42 L. J. J. Laarhoven, J. P. G. Born, L. W. C. E. Arends and
P. Mulder, J. Chem. Soc., Perkin Trans. 2, 1997, 2307.
43 H. Hippler and J. Troe, J. Phys. Chem., 1990, 94, 3803.
44 X. Ma and H. H. Schobert, J. Phys. Chem. A, 2000, 104, 1064.
45 D. D. M. Wayner and D. Griller, Advances in Free Radical Chem-
istry, 1990, 1, 159–192.
46 V. Pacakova and K. Feltl, Chromatographic Retention Indices; An
Aid to the Identfication of Organic Compounds, Ellis Horwood,
London, 1992.
47 E. M. Arnett and N. J. Pienta, J. Am. Chem. Soc., 1980, 102, 3329.
48 H. Y. Afeefy, J. F. Liebman and S. E. Stein, Neutral Thermoche-
mical Data in NIST Chemistry WebBook, NIST Standard Refer-
ence Database Number 69, ed. P. J. Linstrom and W. G. Mallard,
National Institute of Standards and Technology, Gaithersburg
MD, 20899, July 2001 (http://webbook.nist.gov).
49 S. W. Slayden, J. F. Liebman and W. G. Mallard, in The chemis-
try of halides, pseudo-halides and azides, ed. S. Patai and
Z. Rappoport, Wiley, 1995, Supplement D2, pp. 361–402.
11
12
13
14
15
16
7th, Acad. Nauk USSR, Moscow, 1977, p. 132.
17
18
19
E. L. Krasnykh, T. V. Vasiltsova, S. P. Verevkin and A. Heintz,
J. Chem. Eng. Data, 2002, 47, 1372.
S. P. Verevkin, E. L. Krasnykh, T. V. Vasiltsova and A. Heintz,
J. Chem. Eng. Data, 2003, 48in press.
F. Stohmann, Z. Phys. Chem. Stoichiom. Verwandtschaftle, 1890,
6, 334.
Phys. Chem. Chem. Phys., 2003, 5, 2605–2611
2611