Infrared spectroscopic studies of the conformation in ethyl α-haloacetates in the vapor, liquid and solid phases

Article abstract of DOI:10.1016/j.saa.2010.03.022

Infrared spectra of ethyl α-fluoroacetate, ethyl α-chloroacetate, ethyl α-bromoacetate and ethyl α-iodoacetate have been measured in the solid, liquid and vapor phases in the region 4000-200 cm^-1. Vibrational frequency assignment of the observed bands to the appropriate modes of vibration was made. Calculations at DFT B3LYP/6-311+G** level, Job: conformer distribution, using Spartan program '08, release 132 was made to determine which conformers exist in which molecule. The results indicated that the first compound exists as an equilibrium mixture of cis and trans conformers and the other three compounds exist as equilibrium mixtures of cis and gauche conformers. Enthalpy differences between the conformers have been determined experimentally for each compound and for every phase. The values indicated that the trans of the first compound is more stable in the vapor phase, while the cis is the more stable in both the liquid and solid phases. In the other three compounds the gauche is more stable in the vapor and liquid phases, while the cis conformer is the more stable in the solid phase for each of the second and third compound, except for ethyl α-iodoacetate, the gauche conformer is the more stable over the three phases. Molar energy of activation Ea and the pseudo-thermodynamic parameters of activation ΔH^?, ΔS^? and ΔG^? were determined in the solid phase by applying Arrhenius equation; using bands arising from single conformers. The respective E_a values of these compounds are 5.1 ± 0.4, 6.7 ± 0.1, 7.5 ± 1.3 and 12.0 ± 0.6 kJ mol^-1. Potential energy surface calculations were made at two levels; for ethyl α-fluoroacetate and ethyl α-chloroacetate; the calculations were established at DFT B3LYP/6-311+G** level and for ethyl α-bromoacetate and ethyl α-iodoacetate at DFT B3LYP/6-311G* level. The results showed no potential energy minimum exists for the gauche conformer in ethyl α-fluoroacetate.

Full text of DOI:10.1016/j.saa.2010.03.022

Spectrochimica Acta Part A 76 (2010) 213–223
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and
Biomolecular Spectroscopy
journal homepage: www.elsevier.com/locate/saa
Infrared spectroscopic studies of the conformation in ethyl ␣-haloacetates in the
vapor, liquid and solid phases
∗
Naserallah A. Jassem, Muhsin F. El-Bermani
Department of Chemistry, College of Science, University of Baghdad, Al-Jadryah–Baghdad, Iraq
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 25 June 2009
Received in revised form 2 March 2010
Accepted 16 March 2010
Infrared spectra of ethyl ␣-fluoroacetate, ethyl ␣-chloroacetate, ethyl ␣-bromoacetate and ethyl ␣-
−1
iodoacetate have been measured in the solid, liquid and vapor phases in the region 4000–200 cm
.
Vibrational frequency assignment of the observed bands to the appropriate modes of vibration was
made. Calculations at DFT B3LYP/6-311+G** level, Job: conformer distribution, using Spartan program
‘
08, release 132 was made to determine which conformers exist in which molecule. The results indicated
Keywords:
that the first compound exists as an equilibrium mixture of cis and trans conformers and the other three
compounds exist as equilibrium mixtures of cis and gauche conformers. Enthalpy differences between
the conformers have been determined experimentally for each compound and for every phase. The val-
ues indicated that the trans of the first compound is more stable in the vapor phase, while the cis is the
more stable in both the liquid and solid phases. In the other three compounds the gauche is more stable
in the vapor and liquid phases, while the cis conformer is the more stable in the solid phase for each
of the second and third compound, except for ethyl ␣-iodoacetate, the gauche conformer is the more
stable over the three phases. Molar energy of activation Ea and the pseudo-thermodynamic parameters
Infrared absorption spectra
Frequency assignment
Conformers
Absorptivity
Enthalpy difference between conformers
Molar energy of activation
‡
‡
‡
of activation ꢀH , ꢀS and ꢀG were determined in the solid phase by applying Arrhenius equation;
using bands arising from single conformers. The respective Ea values of these compounds are 5.1 ± 0.4,
−1
6
.7 ± 0.1, 7.5 ± 1.3 and 12.0 ± 0.6 kJ mol . Potential energy surface calculations were made at two levels;
for ethyl ␣-fluoroacetate and ethyl ␣-chloroacetate; the calculations were established at DFT B3LYP/6-
11+G** level and for ethyl ␣-bromoacetate and ethyl ␣-iodoacetate at DFT B3LYP/6-311G* level. The
results showed no potential energy minimum exists for the gauche conformer in ethyl ␣-fluoroacetate.
2010 Elsevier B.V. All rights reserved.
3
©
1
. Introduction
[6] reports presence of cis and gauche conformation. On the
other hand, NMR works on rotational isomerism of methyl ␣-
fluoroacetate in the vapor phase and in various solvents suggest
presence of cis and trans conformers [7]. While a Raman and
infrared work on the same compound in the vapor and liq-
The early investigations of conformations in esters were on ethyl
-chloroacetate and ethyl ␣-bromoacetate using infrared spec-
␣
troscopy [1,2]. These works reported presence of cis and gauche
◦
◦
conformers with X–C–C O dihedral angles ˚ = 0 and 120 respec-
tively, the cis being more stable than the gauche. The results
suggest that the repulsion between the halogen and ether oxygen
is stronger than that between the halogen and the carbonyl oxy-
gen. Later work on infrared intensity of couples of C O bands at
different temperatures on ethyl ␣-fluoroacetate in the vapor phase
uid phase together with ab initio calculations reports trans and
gauche conformation [8]. An 1H NMR and
13
C NMR work on
␣-monosubstituted ethyl acetates (substituent = F, Cl, Br, and I)
together with relative energy calculations carried out at B3LYP/6-
31G level for the first three compounds and at B3LYP/3-21 G level
for ethyl ␣-iodoacetate suggest presence of cis and trans con-
formers in ethyl ␣-fluoroacetate with the trans being the more
stable and cis and gauche in each of ethyl ␣-chloroacetate, ethyl ␣-
bromoacetate and ethyl ␣-iodoacetate with the gauche being the
more stable conformer [9].
[
3] reported presence of trans and gauche conformers. While, two
groups of work on unsubstituted acetates suggest presence of cis
and trans, with the cis being the more stable conformer [4,5].
13
A part from the above studies a
C NMR work on ␣-
monosubstituted methyl acetates (substituent = F, Cl, Br, and I)
These results raise interesting questions regarding the num-
ber of conformers 2n that would exist, where n is the number of
asymmetric centers in the molecule, which conformers exist in
which molecule, the conformation preferences and the interactions
responsible for the presence of cis-trans or cis-gauche or trans-
gauche equilibriums. To explore these further, we have undertaken
∗ Corresponding author at: 238 Charlebois Cr, Saskatoon, SK, and S7K 5J5
.Saskatchewan, Canada. Tel.: +1 306 9781574; fax: +1 306 956 3323.
E-mail address: m f el bermani@Yahoo.com (M.F. El-Bermani).
1
386-1425/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.saa.2010.03.022

2
14
N.A. Jassem, M.F. El-Bermani / Spectrochimica Acta Part A 76 (2010) 213–223
Scheme I. Some of the possible conformers of ethyl ␣-haloacetate (X = F, Cl, Br, and I).
4000–200 cm 1 using CsI windows. Controlled variable tempera-
ture cell Specac type fitted with thermocouple wire was used in
two temperature ranges 248–123 K and 298–538 K for the solid and
−
infrared studies of ethyl ␣-fluoroacetate, ethyl ␣-chloroacetate,
ethyl ␣-bromoacetate and ethyl ␣-iodoacetate in the solid, liq-
uid and vapor phases. It was felt that by studying such a series
as this can one hope to make a reasonable account for the factors
that determine stability of the conformers for the members of the
series and to measure their enthalpy differences and molar energy
of activation of the interconverting conformers. To the best of our
knowledge, no previous infrared work has been published on these
compounds in the same program as ours.
◦
liquid phases, respectively. It was controlled to ± 1 C in every case.
Liquid nitrogen was used for cooling. In the vapor phase, a 10 cm
path length glass gas cell was used with CsI windows [Pye–Unicam]
fitted with the thermocouple wire and wound by an asbestos heat-
ing tape gated to temperature controller unit type tem-1.
The spectral bands that are identified to arise from two different
conformers were measured five times at each temperature. Aver-
age values of the absorbance were taken in the calculation. The IR
instrument was programmed to high resolution so that to provide
a spectral slit width of 1/5 ꢀꢁ1/2, with ꢀꢁ1/2 being the band width
at half height of the absorption [12]. It was calibrated by using a
pair of bands of 1,2-dichloroethane vapor phase spectrum arising
from two different conformers measured at different temperatures.
2
. Experimental
Ethyl ␣-fluoroacetate and ethyl ␣-chloroacetate are commer-
cial liquids furnished by Fluka Buchs S.G. of stated purity 98%. They
were repurified by drying over magnesium sulfate for 2 days. After
filtrations, fractional distillation was carried out at normal condi-
tions. Middle portion was collected for each compound at their
◦
−1
It gave ꢀH = 4.26 ± 0.28 kJ mol which is in good agreement with
◦
−1
◦
◦
published values ꢀH = 4.28 and 4.6 kJ mol [13,14].
b.p. 117 C and 143.5 C, respectively. Ethyl ␣-bromoacetate was
prepared by conventional method [10] through esterification of
bromoacetic acid by mixing with absolute ethyl alcohol. The dry
Molar energy of activation was estimated by measuring the rate
constants k through tracing the growth of a band identified to arise
from a single conformer in the crystalline phase. The same method
of cooling and annealing was used in these measurements. While
cooling the crystalline sample, the relevant band was recorded to
observe its intensity changes. When no significant reduction in
band intensity is observed, the sample was allowed to heat. By set-
ting the spectrophotomer at fixed wavenumber and starting time
drive program to record time in minutes; the band maximum was
recorded at certain temperature. The same procedure was repeated
for the subsequent measurements of intensity of this peak at higher
product was purified by fractional distillation at reduced pressure
◦
[
6 mm Hg]. Middle portion was collected at its b.p. 58 C. Ethyl ␣-
iodoacetate was prepared by displacement reaction between ethyl
-bromoacetate and potassium iodide in acetone base at room tem-
␣
perature [11]. After drying, fractional distillation was carried out at
reduced pressure [12 mm Hg]. Middle portion was collected at its
◦
b.p. 73 C.
The infrared spectra of these compounds were recorded
on Perkin-Elmer599B Infrared Spectrophotometer in the region
Table 1
Calculated parameters at level DFTB3LYP/6-311+G** of conformer distribution for the cis, gauche and trans conformations of ethyl ␣-halacetates.
Parameters
Ethyl ␣-fluoroacetate
Ethyl ␣-chloroacetate
Ethyl ␣-bromoacetate
Ethyl ␣-iodoacetate
Cis
Trans
Cis
Gauche
Cis
Gauche
Cis
Gauche
Energy (Hartrees)
Relative energy (kJ/mol)
Dipole moment ꢂ (D)
D(X1–O1 = C1) (Å)
D(X1–O2) (Å)
−407.1384
0.64
−407.1386
0.000
1.19
3.549
2.612
180.0
58.9
113.9
120.6
114.0
116.7
125.7
1.385
1.093
1.521
1.208
1.331
1.455
−767.5643
0.000
3.81
−767.5646
−0.74
2.35
−2881.8880
0.000
3.66
−2881.8895
−4.05
2.42
−7228.0936
9.35
−7228.0972
0.000
2.41
3.98
3.43
2.731
3.541
0.00
3.021
3.943
0.00
3.642
3.161
112.8
128.3
111.4
123.5
111.5
116.5
124.9
1.811
1.085
1.518
1.206
1.338
1.453
3.135
4.096
0.00
3.687
3.342
104.1
166.1
110.6
123.2
111.4
116.4
124.7
1.971
1.085
1.513
1.206
1.340
1.453
3.249
4.311
0.00
3.820
3.600
97.6
˚
˚
(X1–C2–C1 O5)
(H1–C2–C1 O5)
120.3
110.7
125.4
109.4
116.6
125.4
1.379
1.093
1.522
1.200
1.346
1.454
120.7
112.9
127.0
108.1
117.2
125.3
1.790
1.089
1.523
1.200
1.347
1.454
59.6
119.8
112.6
126.9
109.0
115.7
124.2
2.176
1.088
1.521
1.201
1.348
1.456
18.8
ꢃ(X1–C6–C1)
ꢃ(C6–C1 O1)
ꢃ(C6–C1–O2)
ꢃ(C1–O2–C3)
ꢃ(O2–C1 O1
D (X1–C6)
D (H3–C2)
D (C6–C1)
D (C1 O5)
D (C1–O2)
113.5
126.9
108.1
117.3
124.1
1.947
1.088
1.521
1.200
1.348
1.455
111.1
124.1
113.3
116.5
124.6
2.199
1.084
1.508
1.206
1.343
1.453
D(O2–C3)
˚, dihedral angle in degrees.
ꢃ, tri-atomic angle in degrees.
d: bond distance (−) and D: intra-molecular distances ( . ..) in Angstrom (Å) unit.
.

N.A. Jassem, M.F. El-Bermani / Spectrochimica Acta Part A 76 (2010) 213–223
215
temperatures and times. Then absorbances at the corresponding
temperature and time were determined from which rate constants
have been calculated.
3
. Results and discussion
The structural configuration of ethyl ␣-haloacetate molecule
predicts at least three possible conformers [cis, gauche and trans]
as depicted in Scheme I. Each conformer of the molecules in this
series of compounds is expected to display 36 fundamental modes
of vibration each of which should be infrared and Raman active.
ꢀ
ꢀ
They are distributed as 22A ˇı +14 A for each of the cis and trans
conformers (point group Cs). While for the gauche conformer (point
group C ) all the 36 modes are of symmetry class A. Hence, 72 dis-
1
tinct fundamental bands are theoretically possible for each couple
of conformers. In practice, this number of bands is never found
since the corresponding vibrational modes in the two conform-
ers usually have virtually identical frequencies which result in
an infrared band common to both species. Hence it is usually
found that the solid phase spectrum of these molecules is sim-
plified by loss of some bands compared with the liquid phase
spectrum.
−1
Fig. 1. Potential energy profile showing plots of relative energy (kJ mol ) against
the constraint dihedral angle (˚) for ethyl ␣-haloacetate: (a) ethyl ␣-fluoroacetate,
(b) ethyl ␣-chloroacetate, (c) ethyl ␣-bromoacetate, and (d) ethyl ␣- iodoacetate.
The results of Table 1 infer that the dihedral angle ˚ (X–C–C O)
of the predominant conformer decreases as the electronegativity of
◦
◦
◦
the halogen atom is decreased, the values are: 180 , 112.8 , 104.1
3.1. The theoretical calculations
◦
and 97.6 for the fluoro, chloro, bromo- and iodo-compounds,
respectively. The energy difference between conformers of these
compounds increases as the atomic mass of the halogen atom is
increased, the order of which is as follows; F < Cl < Br < I. It can be
seen from Table 1 also that, geometry of the molecules changes with
conformational interconversion and change of substituents. The
bond distances of each couple of conformers increases as the elec-
tronegativity of the halogen is decreased. Further, we have found
that the frequency of C O stretching bands of the two stable con-
formers is linearly related to the dihedral angle ˚(X–C–C O). The
plot of the experimental wavenumbers of C O stretching bands of
the conformers of these compounds in the vapor phase against the
calculated dihedral angles ˚(X–C–C O) gave a good straight line as
shown in Fig. 2. This plot can foretell which bands belong to which
conformers.
The geometry of ethyl ␣-haloacetate molecule indicates that
the repulsion between the halogen atom and ether oxygen in the
trans position is stronger than that between the halogen and the
carbonyl oxygen in the cis position, thus suggesting preference of
cis or gauche conformation over the trans [see Table 1]. The the-
oretical calculations using Spartan ‘08, release 132 a commercial
program at DFT B3LYP/6-311+G** level, Job: conformer distribu-
tion; reveal that there are two stable conformers in each compound
of this series of molecules. For ethyl ␣-fluoroacetate they are cis
and trans, with the trans being the more stable conformer, while
for each of ethyl ␣-chloroacetate, ethyl ␣-bromoacetate and ethyl
␣
-iodoacetate they are cis and gauche. The results of calcula-
tions are set out in Table 1. They infer that the energy difference
−
1
value for ethyl ␣-iodoacetate ꢀE = −9.45 kJ mol is in good agree-
◦
−1
Stability preference of the gauche conformer over the cis in the
last three compounds may be ascribed to three factors. First, the
ment with the experimental value ꢀH = −9.1 ± 1.7 kJ mol
of
the vapor phase [see Table 5]. The corresponding value for ethyl
−
−1
1
molecular orbital interactions of the type ␲*C O → ␴*
and sec-
[15,16].
O
␣
-bromoacetate ꢀE = −3.94 kJ mol is lower than the experimen-
X–C
◦
ond the hyperconjugative effect of the type ␴X–C → ␲*
tal value ꢀH = −6.7 ± 2.5 kJ mol of the vapor phase. The values
C
The experimental evidence for the latter effect can be seen from
the decreasing absorptivity ratio Dc/Dg of the carbonyl bands aris-
ing from cis and gauche conformers of the last three compounds;
the approximate values are: 0.77, 0.68 and 0.54 for the chloro-,
bromo- and iodo- compounds, respectively. These values indicate
increasing stability of the gauche conformer with the increase of
the C–X bond polarizability. The correlation, may be interpreted in
terms of interaction between unperturbed ꢄC–X and ␲*C O orbitals
for the fluoro- and chloro-compounds are still lower. They are
−1
ꢀ
E = −0.5 and −0.79 kJ mol compared with the experimental val-
◦
−1
ues of the vapor phase ꢀH = −3.8 ± 1.7 and −4.2 ± 1.7 kJ mol of
fluoro and chloro compounds, respectively. However, the values of
Table 1 are much better than those of the basis sets RHF/6-31 G*,
RHF/6-311G*, DFT B3LYP/6-31G*and DFT B3LYP/6-311++G**. The
result for the fluoro-compound has a support in [9] in that it sug-
gests presence of cis and trans conformers in ethyl ␣-fluoroacetate
[
see Table 1 of ref. [9]].
The potential energy surface calculations were carried out at
two levels; one for ethyl ␣-fluoroacetate and ethyl ␣-chloroacetate
the calculations were performed at level DFT B3LYP/6-311+G** and
the other for ethyl ␣-bromoacetate and ethyl ␣-iodoacetate at DFT
B3LYP/6-311G* level. Fig. 1 shows the potential energy surfaces of
−
1
ethyl ␣-haloacetates where plots of the relative energy (kJ mol
)
against the constraint dihedral angle ˚ (X–C–C O) were made.
Both Table 1 and Fig. 1 have established existence of the first com-
◦
◦
pound as equilibrium mixture of cis (˚ = 0 ) and trans (˚ = 180 )
conformers with slight predominance of the trans with no poten-
tial energy minimum for the gauche conformer [see Fig. 1a]. This
result has a support in Ref. [7], in that methyl ␣-fluoroacetate has
no potential energy minimum.
Fig. 2. Plot of the wavenumbers (cm−1) of the conformers against the calculated
dihedral angle ˚ (X–C–C O) of ethyl ␣-haloacetates. F: ethyl ␣-fluoroacetate, Cl:
ethyl ␣-chloroacetate, Br: ethyl ␣-bromoroacetate and I: ethyl ␣-iodoacetate.

2
16
N.A. Jassem, M.F. El-Bermani / Spectrochimica Acta Part A 76 (2010) 213–223
within the molecule in a second-order perturbation equation as a
result of considerable energy gap between them [16,17]. The values
3.3. The carbonyl stretching bands
3
3
of C–X bond polarizability are: C–F [0.67 Å ], C–Cl bonds [2.53 Å ],
The infrared spectra of the vapor phase Fig. 3c, for ethyl ␣-
fluoroacetate show two strong doublets in the region of C
3
3
C–Br [3.61 Å ] and C–I bond [5.4 Å ] [16]. Third, the intra-molecular
O
distance X· · ·O in the gauche conformer is larger than X· · ·O C in
stretching mode of vibration. The first appears at 1803 and
2
−
1
−1
and the second at 1772 and 1765 cm . In the mid-
the cis form [see Table 1].
1793 cm
−
1
dle of these doublets there is a peak at 1782 cm . Both doublets
are fundamental bands with the first being resolved into R and
P branches. The former has been reduced in intensity into a very
weak shoulder in the liquid phase, while the second fundamen-
tal band at 1772 cm and its shoulder at 1765 cm persisted in
the liquid phase [Table 2] with persistence of the middle peak in
3.2. The assignment
Frequency assignment of the infrared bands to the approx-
−
1
−1
imate modes of vibration of the trans and cis conformers of
ethyl ␣-fluoroacetate and cis and gauche conformers of each
of ethyl ␣-chloroacetate, ethyl ␣-bromoacetate and ethyl ␣-
iodoacetate are given in Table 2. They are tentative particularly
with reference to the modes of vibration involving CH3 and CH2
−
1
this phase and appears as a hump. The middle peak at 1782 cm
−
1
may be a combination band (923 + 865 cm ) deriving its intensity
from Fermi resonance with the fundamental. A report on methyl ␣-
fluoroacetate confirmed that this peak is a Fermi resonance band by
recording the first overtone of the carbonyl bands which appeared
as two well resolved bands with no first overtone of this band [7].
However, we could not confirm that this peak is a Fermi resonance
band because the first overtones of the carbonyl bands are too weak
to be recognized. In the solid phase, the first doublet 1803 and
groups. We have adopted in the assignment the order ꢁ(CH ) > ꢁ
3
(
CH ) > d(CH ) > w(CH ) > w(CH ) > tw(CH ) > tw(CH ) > r(CH ) >
2 2 3 2 3 2 3
r(CH ). The same order was followed with respect to the modes
2
of vibration involving XCH2 groups. The assignment has been
built mainly on the spectral changes that have been observed
between spectra of the different phases, thermal changes, and
solvent effects and with taking advantage of having been made
for four similar molecules and on comparisons with literature as
will be mentioned later on. One can see from Table 2 that the
assignments are internally consistent but since they must be to
some extent arbitrary they will not be discussed in detail except
for few specific points. Fig. 3 shows solid, liquid and vapor phase
spectra of ethyl ␣-fluoroacetate as a representative figure for all
spectra of the compounds of the series of molecules of the present
work.
−
−
1
1
1
793 cm has disappeared almost completely, while the band at
772 cm appeared at 1752 cm as a very strong and thin band
1
−1
(Fig. 3a).
A worthwhile point to note is that, we found a linear relationship
between the frequency difference of C O stretching bands arising
from the cis and gauche conformers of the chloro, bromo and iodo
compounds and the corresponding calculated dihedral angle ˚(X-
C–C O). It is likely from the frequency differences measured for
these compounds that a hypothetical frequency difference value
between cis and gauche conformers for ethyl ␣-fluoroacetate can
be estimated by extrapolation on the assumption that the corre-
The bands which have appeared in the vapor phase spectrum
and disappeared in the solid phase are: for ethyl ␣-fluoroacetate:
◦
−1
)
−
1
sponding dihedral angle ˚ = 120 . Thus, the respective ꢀꢁ(cm
1
1
295, 1216, 1111, 1097, and 1038 cm ; for ethyl ␣-chloroacetate:
−
1
−
1
values for these three compounds 22, 18, 17 cm and the corre-
315, 1300, 1295, 1169, 1117 and 1048 cm ; for ethyl ␣-
◦ ◦ ◦
sponding ˚ = 112.8 , 104.1 and 79.6 allowed extrapolation into
−
1
bromoacetate: 1212, 930, and 717 cm with a band decreases in
−
1
−1
a value of 24 cm . Thus, by comparing this value with the exper-
imental frequency difference ꢀꢁ = 31 cm
803 and 1772 cm , we found that the ꢀꢁ = 31 cm is between cis
intensity at 1282 cm and for ethyl ␣-iodoacetate: 1337, 1209, and
040, 1028 cm . Thus, there is no doubt that these bands should
−
1
−1
between the band at
1
−
1
−1
1
be assigned to the trans for the first compound and to the gauche
conformer for the last three compounds.
In addition, there is some bands increase in intensity
and others decrease upon the shift from phase to phase in
the trend: solid–liquid–vapor. For ethyl ␣-fluoroacetate, the
ratio D1472/D1461 takes the values: 0.46, 1.23 and 2.00; ethyl
and trans conformers. Thus, suggesting that the doublet (1803 and
−
1
1
793 cm ) arises from the cis conformer and the second doublet
−
1
at 1772 and 1765 cm arises from the trans.
The solvent effect studies on the intensity variation of ethyl ␣-
haloacetate in various solvents, can distinguish between the more
polar and less polar conformers. The plots of absorbance D against
the dielectric constant function (ε + 1)/(2ε + 1) of the solvents give
rise to two lines; one for the more polar conformer and the other
for the less polar one as shown in Fig. 4. The plots of the couple
␣-chloroacetate, D1438/D1425: 0.19, 0.58 and 1.00; ethyl ␣-
bromoacetate, D1416/D1400: 1.92, 1.22 and 1.00, and another
ratio, D556/D572: 1.22, 1.50 and 2.50, and for ethyl ␣-iodoacetate
D1392/D1386: 0.76, 1.00 and 1.33.The bands which increase in the
vapor phase were assigned to the trans in the first compound and to
the gauche for the other three compounds and those who decreased
assigned to the cis with respect to the first three compounds and
to the gauche with respect to the last compound. There are many
other bands show increase or decrease in the absorptivity ratios
which were of great help to the assignment.
−
1
of carbonyl stretching bands at 1803 and 1772 cm of ethyl ␣-
fluoroacetate in various solvents; indicate that the first band arises
from a more polar conformer and the latter arises from a less polar
one. Hence, there is no doubt according to the theoretical calcula-
tions that the first doublet can be assigned to the cis conformer and
the second to the trans which lend support to the above analysis.
Fig. 3. Infrared spectra of ethyl ␣-fluoroacetate in the: (a) solid, (b) liquid and (c) vapor phase.

Table 2
Observed infrared vibrational frequencies and approximate assignments to the appropriate fundamental modes of vibration of the conformers, cis (c), trans (t) and gauche (g) of ethyl ␣-haloacetates (cm−1). The notations are:
the first parenthesis () is for ethyl ␣-fluoroacetate and the second () is for ethyl ␣-chloroacetate, ethyl ␣-bromoacetate, and ethyl ␣-iodoacetate. In case there is no parenthesis it means for both conformers.

Table 2 (Continued )

2
20
N.A. Jassem, M.F. El-Bermani / Spectrochimica Acta Part A 76 (2010) 213–223
Fig. 4. Plots of the absorbance of ꢁ(C O) bands of the more polar and less polar
conformers of ethyl ␣-haloacetate in various solvents against the solvent dielectric
−
1
constant function (ε − 1)/(2ε + 1): (A) ethyl ␣-fluoroacetate bands (a) at 1803 cm
−
1
−1
and (b) at 1772 cm and (B) ethyl ␣-chloroacetate bands (a) at 1786 cm and (b)
−
1
at 1764 cm . The solvent key number is as given in Table 3.
The spectral data for ethyl ␣-fluoroacetate, ethyl ␣-chloroacetate
and ethyl ␣-bromoacetate in various solvents are given in Table 3.
Ethyl ␣-iodoacetate was excluded from this table because of the
bad data points.
The assignations of the fundamental bands to the appropriate
modes of vibration and to the appropriate conformers is based
on comparing spectra of these compounds with each other, phase
to phase and with spectra of ethylacetate, ethyl ␣-chloroacetate,
ethyl ␣-trichloroacetate, and with the spectrum of ethyl ␣-
trifluoroacetate [18] and ethyl ␣-cyanoacetate [19], and others in
the literature [20–27], in addition to the spectra of propanal [28], 3-
bromopropene [29], and 1,3-dichloropropane [30]. In the first step,
all the halogenated parts were figured out, namely all the XCH2
modes; the CH2 stretching, deformation, wagging, twisting, rock-
ing, C–X stretching, bending modes of XCH –C O, and XCH –C–O
2
2
and so on. In the second step, CH3 and CH2 of the ethyl group and
all relevant modes of vibration were figured out. In the third step,
the C–O, CH –O stretching and COC bending modes were detected.
2
In fact, all the bands arising from the characteristic groupings CH3,
CH2 and C O, XCH , C–O, O–CH , C–COO, CH –CH and C–X were
2
2
2
3
practically figured out through this method.
◦
3.4. Enthalpy difference ꢀH
The enthalpy differences between the conformers have been
determined in the vapor, liquid and solid phases using two bands
for each molecule identified to arise from two different conformers
by using the Van’t Hoff equation below
◦
Dm
ꢀH
RT
ln
= log C −
(1)
Dl
where Dm and Dl are absorptivities (Absorbances) of the more
stable and less stable conformers respectively, C is a constant com-
prising molar absorption coefficient ratios of the bands related
to the corresponding conformers, partition function ratios and
◦
(
ꢀS /R). Fig. 5, shows two typical temperature-dependent absorp-
tion bands b C–C O and b C–C–O assigned to cis and trans
conformers respectively for ethyl ␣-fluoroacetate at 620 and
−
1
6
06 cm . Linear regression lines were obtained from the plots
−
1
◦
of ln Dm/D against T . The calculated ꢀH values of these com-
l
pounds are given in Table 5.

N.A. Jassem, M.F. El-Bermani / Spectrochimica Acta Part A 76 (2010) 213–223
221
Table 3
Influence of solvents on the intensity of ꢁ(C O) stretching bands of ethyl ␣-fluoroacetate at 1803 and 1772 cm 1, ethyl ␣-chloroacetate at 1786 and 1764 cm−1 and ethyl
−
−1
␣
-bromoacetate at 1770 and 1752 cm in various solvents. The subscripts t, c and g denote for trans, cis and gauche, respectively.
Solvent
(ε − 1)/(2ε + 1)
Ethyl ␣-fluoroacetate
Ethyl ␣-chloroacetate
Ethyl ␣-bromoacetate
Dc
Dt
Dc
Dg
Dc
Dg
1
2
3
4
5
6
7
8
. Cyclohexane
. CCl4
. Benzene
0.202
0.226
0.23
0.166
0.147
0.168
0.166
0.141
0.18
0.106
0.096
0.09
0.106
0.08
0.082
0.075
0.079
0.087
0.097
0.106
0.103
0.104
0.102
0.108
0.123
0.100
0.102
0.092
0.083
0.093
0.079
0.077
0.084
0.075
0.069
0.087
0.069
–
0.093
0.095
0.093
–
0.123
0.125
0.131
0.119
0.127
0.121
0.120
. Toluene
0.24
. Chloroform
. Chlorobenzene
. Ethylene dichloride
. Acetonitrile
0.359
0.377
0.43
0.165
0.187
0.48
The enthalpy difference for ethyl ␣-fluoroacetate was mea-
sured in the vapor phase using the b(C–C O) and b C–C–O
3.5. Estimations of the molar energy of activation
−1
◦
bands at 620 and 606 cm . The ꢀH value for the cis–trans
The barrier to rotation, which is the molar energy of activation,
can be calculated through application of the Arrhenius equation:
−
−
1
equilibrium of −3.8 ± 1.7 kJ mol
is in good agreement with
1
reported value of −3.6 ± 0.5 kJ mol using ꢁ(C O) stretching bands
ꢀ
ꢁ
Ea
[
3,31]. The thermal spectral results for all the compounds of
k = A exp
−
(2)
RT
this work in the solid, liquid and vapor phases are set out in
Table 4.
with
One can see from Tables 4 and 5, that a reversal in stability
is observed in the first compound on passing from the vapor to
the liquid and solid phases, where the cis is the more stable in
the liquid and solid phases. Ethyl ␣-chloroacetate and ethyl ␣-
bromoacetate show reversal in stability only in the solid phase,
indicating preference of stability to the cis in these two compounds
respectively. However, the general trend of stability of the pre-
dominant conformers the trans for the first compound and gauche
for the other three compounds in the vapor phase show a regular
increase as the mass of the halogen atom is increased F [trans] < Cl
ꢀ
ꢁ
1
t
D∞ − D0
D∞ − Dt
k =
ln
(3)
where k, A, D , Dt, and D∞ are rate constant, pre-exponential fac-
0
tor, absorbances at zero time, at time t and at the end of the run
respectively. Fig. 6 shows plots of ln k (k in s ) against T (K )
for all the compounds of this series. The least squares method was
used in the calculations of Ea and the standard errors. The mea-
surements were made in the solid state only at sufficiently low
−1
−1
−1
◦
temperature (−165 C) for the conformer to manifest itself. Thus,
[
gauche] < Br [gauche] < I [gauche]. In the liquid phase, this trend
by tracing intensity change of either band with temperature and
time it is expected to obtain molar energy of activation value of
the relevant conformer. Consequently, the pseudo-thermodynamic
in the last three compounds is reversed and takes the order
Cl > Br > I.
In the solid phase spectrum [non-glassy samples], ethyl ␣-
iodoacetate an exception to the first three compounds revealed that
the gauche conformer is the more stable conformer as in the liquid
‡
parameters of activation, the enthalpy change of activation ꢀH ,
entropy change of activation ꢀS , and Gibbs free energy change
of activation ꢀG can be calculated from Ea using the following
‡
‡
◦
and vapor phases. However, the ꢀH values [Table 5] in the solid
phase are very low indicating that the conformers of each couple
for each compound coexist almost equally with slight bias of the
cis conformer in the first three compounds.
Fig. 6. Arrhenius plots of ln k against T 1, with k in s−1 and T in K, for ethyl
−
␣-haloacetates in the solid phase. Least squares method was used; the result-
ing parameters are: (A) for ethyl ␣-fluoroacetate the slope m = −0.61 ± 0.13 K,
2
intercept b = −2.94 ± 0.28, correlation coefficient r = −0.9791,
ꢆ
= 0.0514, No.
of data points = 10; (B) for ethyl ␣-chloroacetate; m = −0.80 ± 0.03 K, intercept
2
b = −2.22 ± 0.02, r = −0.98673, ꢆ = −0.0284, No. of data points = 13; (C) for ethyl
2
−4
␣
-bromoacetate; m = 0.90 ± 0.22 K, b = −1.30 ± 0.90, r = 0.970, ꢆ = −9.89 × 10 , No.
−
1
Fig. 5. Infrared absorption bands of ethyl ␣-fluoroacetate 606 and 620 cm
of data points = 4; (D) for ethyl ␣-iodoacetate; m = 1.44 ± 0.11 K, b = 0.50 ± 0.35,
2
−4
recorded at different temperatures in the vapor phase [see Table 2].
r = −0.99737, ꢆ = −1.09 × 10 , No. of data points = 4.

2
22
N.A. Jassem, M.F. El-Bermani / Spectrochimica Acta Part A 76 (2010) 213–223
Table 4
Temperature dependence of optical densities of couples of some bands arising from different conformers for ethyl ␣-haloacetates in the solid, liquid, and vapor phase.
Compound
Solid
Liquid
Vapor
Dt 590 cm 1
−
Dc 573 cm
−1
Dt/Dc
T (K)
Dt 590 cm
−1
Dc 620 cm
−1
Dt/Dc
T (K)
Dt 606 cm
−1
Dc 620 cm
−1
Dt/Dc
T (K)
108.5
123.0
133.5
142.0
156.5
170.5
182.5
193.5
0.146
0.269
0.234
0.241
0.252
0.251
0.258
0.258
0.162
0.292
0.254
0.256
0.270
0.266
0.272
0.267
0.901
0.921
0.921
0.941
0.933
0.944
0.949
0.966
239
244
254
274
284
294
309
314
0.117
0.115
0.119
0.169
0.197
0.210
0.215
0.302
0.161
0.134
0.131
0.148
0.162
0.157
0.166
0.203
0.727
0.858
0.908
1.142
1.216
1.338
1.295
1.488
313
323
333
353
363
373
388
398
0.42
0.12
3.50
3.11
2.84
2.95
2.71
2.74
2.57
2.34
0.314
0.295
0.271
0.290
0.293
0.398
0.356
0.101
0.104
0.092
0.107
0.107
0.155
0.152
FCH2COOR
Dg 590 cm 1
−
Dc 620 cm
−1
Dg/Dc
T (K)
Dg 590 cm
−1
Dc 620 cm
−1
Dg/Dc
T (K)
Dg 590 cm
−1
Dc 620 cm
−1
Dg/Dc
T (K)
1
1
1
1
08.5
19.0
28.5
38.5
0.273
0.264
0.246
0.231
0.242
0.256
0.292
0.163
0.152
0.139
0.137
0.135
0.133
0.157
1.675
1.737
1.770
1.686
1.793
1.925
1.860
244
254
259
264
274
284
294
0.157
0.195
0.230
0.227
0.221
0.223
0.224
0.227
0.278
0.068
0.084
0.116
0.116
0.128
0.130
0.134
0.138
0.184
2.309
2.321
1.983
1.957
1.727
1.715
1.672
1.645
1.511
329
349
359
369
379
389
399
0.346
0.346
0.299
0.217
0.198
0.185
0.167
0.270
0.260
0.249
0.191
0.180
0.175
0.163
1.281
1.331
1.201
1.136
1.100
1.057
1.025
ClCH2COOR
148.5
1
1
58.5
83.5
304
14
3
Dg 556 cm 1
−
Dc 620 cm
−1
Dg/Dc
T (K)
Dg 552 cm
−1
Dc 620 cm
−1
Dg/Dc
T (K)
Dg 548 cm
−1
Dc 620 cm
−1
Dg/Dc
T (K)
103.5
113.0
123.5
133.5
163.5
173.5
183.5
208.5
0.386
0.390
0.388
0.399
0.405
0.403
0.388
0.397
0.346
0.338
0.328
0.332
0.326
0.324
0.306
0.308
1.116
1.154
1.183
1.202
1.242
1.244
1.268
1.289
243.5
251.5
273.5
283.5
293.5
306.5
0.187
0.118
0.109
0.113
0.117
0.113
0.204
0.137
0.147
0.160
0.170
0.168
0.917
0.861
0.741
0.706
0.688
0.673
309
319
339
359
369
389
0.147
0.135
0.092
0.148
0.147
0.140
0.210
0.224
0.198
0.320
0.317
0.354
0.700
0.603
0.465
0.463
0.464
0.395
BrCH2COOR
Dg 548 cm 1
−
Dc 505 cm
−1
Dg/Dc
T (K)
Dg 548 cm
−1
Dc 620 cm
−1
Dg/Dc
T (K)
Dg 690 cm
−1
Dc 568 cm
−1
Dg/Dc
T (K)
1
1
1
1
1
1
1
1
17.5
28.5
38.5
48.5
58.5
68.5
78.5
88.5
0.198
0.200
0.190
0.186
0.185
0.183
0.180
0.151
0.265
0.271
0.267
0.266
0.273
0.271
0.277
0.235
0.747
0.738
0.712
0.699
0.678
0.675
0.650
0.643
263.5
273.5
283.5
293.5
303.0
313.0
0.150
0.147
0.137
0.148
0.144
0.142
0.224
0.222
0.216
0.234
0.234
0.234
0.670
0.662
0.634
0.632
0.615
0.607
319
328
339
349
359
369
379
389
0.360
0.404
0.271
0.244
0.228
0.225
0.229
0.215
0.139
0.173
0.121
0.120
0.123
0.127
0.154
0.156
2.590
2.335
2.240
2.033
1.854
1.772
1.487
1.378
ICH2COOR
‡
‡
equations [32]:
Ea, ꢀH and ꢀS values are increasing as the halogen atomic mass
‡
is increased, while ꢀG values decrease with the third and fourth
compounds being exchanging the order of the decreasing trend
‡
0
ꢀ
H = Ea − RT
(4)
(5)
ꢂ ꢃ
ꢄꢅ
−
1
AhN0
[see Table 6]. The values of Ea = 5.1, 6.7, 7.5 and 12.00 kJ mol for
the fluoro, chloro, bromo and iodo compounds respectively are
somewhat low indicating that the hindering forces to rotation are
not so strong. The calculated potential energy barriers to rotation
are: Eb = 9.2, 2.1, 4.7 and 6.7 kJ mol . They are still lower than the
experimental values except for ethyl ␣-fluoroacetate. The possible
sources of hindrances are steric, the partial double bond charac-
ter where the percentages of which are around 10%, 11%, 14% and
ꢀ
S‡ = R ln
eRT0)
(
‡
‡
0
‡
ꢀ
G = ꢀH − T ꢀS
(6)
−
1
0
where h, N , e, and T are Planck’s constant, Avogadro’s number, e
0
is 2.718 and the temperature 298.15 K, respectively.
The bands used for estimating the molar energy of activation
−1
are: 590, 590, 552 (shoulder) and 522 cm for ethyl ␣-fluoro, ethyl
-chloro, ethyl ␣-bromo and ethyl ␣-iodoacetate, respectively. The
1
6% for these molecules, respectively, repulsive forces between the
␣
halogen and oxygen atoms of the molecule and the interactions
between ␴*C–X and ␲*C O orbital’s and the hyperconjugative effects
␴C–X and ␲*C O.
first band arises from the trans conformer and the others arise from
the gauche. The resulting Ea values and the corresponding pseudo-
thermodynamic parameters of activation are given in Table 6. The
Table 5
Enthalpy differences of the cis–trans, cis–gauche, equilibriums of ethyl ␣-haloacetates in the solid, liquid and vapor phases.
◦
−1
)
Compound
ꢀH (kJ mol
Solid phase
Liquid phase
Vapor phase
Ethyl ␣-fluoroacetate
Ethyl ␣-chloroacetate
Ethyl ␣-bromoacetate
Ethyl ␣-iodoacetate
0.1 ± 0.1
0.3 ± 0.2
0.3 ± 2.0
−0.4 ± 0.1
5.0 ± 0.8
−3.9 ± 1.3
−3.2 ± 0.5
−1.7 ± 0.0
−3.3 ± 1.7; −3.6 ± 0.5, Ref. [3,30]
−4.2 ± 1.7
−6.7 ± 2.5
−9.1 ± 1.7

N.A. Jassem, M.F. El-Bermani / Spectrochimica Acta Part A 76 (2010) 213–223
223
Table 6
‡
‡
‡
Rate constant k, pre-exponential factor A, energy difference of activation Ea and pseudo-thermodynamic parameters of activation (ꢀH , ꢀS , and ꢀG ) of the trans conformer
of ethyl ␣-fluoroacetate and of the gauche conformer of ethyl ␣-chloroacetate, ethyl ␣-bromoacetate and ethyl ␣-iodoacetate in the solid phase.
ln k* (k is in s−1
)
ln A (A is s
−1
)
Ea (kJ mol
−1
)
ꢀH (kJ mol
‡*
−1
)
ꢀS (J K mol
‡*
−1
−1
)
ꢀG (kJ mol
‡*
−1
)
Compound
−
−
3
3
Ethyl ␣-fluoroacetate
Ethyl ␣-chloroacetate
Ethyl ␣-bromoacetate
Ethyl ␣-iodoacetate
(6.78 ± 2.92) × 10
(7.37 ± 0.04) × 10
0.013 ± 0.019
−2.94 ± 0.28
−2.22 ± 0.02
−1.30 ± 0.90
0.50 ± 0.35
5.09 ± 0.37
6.67 ± 0.09
7.48 ± 1.31
12.00 ± 0.62
2.61 ± 0.37
4.19 ± 0.09
5.00 ± 1.31
9.52 ± 0.62
−277.7 ± 2.3
−271.7 ± 0.2
−264.0 ± 7.5
−249.1 ± 2.9
85.4 ± 1.1
85.2 ± 0.1
83.7 ± 3.6
83.8 ± 1.5
0.013 ± 7.82 × 10−
3
*
The values are calculated at T0 = 298.15 K.
Acknowledgements
We are indebted to the head of the Department of Chemistry and
Staff members of the College of Science of the University of Baghdad
for providing the needed facilities. Thanks also due to the head of
the Chemistry Department of the University of Saskatchewan and
to Professor John A. Weil for facilities in the computer services.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.saa.2010.03.022.
References
[
[
[
[
[
[
[
[
1] T.L. Brown, J. Amer. Chem. Soc. 80 (1958) 3513–3515.
2] T.L. Brown, Spectrochim. Acta 18 (1962) 1615–1623.
3] M.V. Garcia, A. Hermanzy, J. Morcillo, An. Quim. 77 (1981) 72.
4] M. Oki, H. Nakanishi, Bull. Chem. Soc. Jpn. 43 (1970) 2558.
5] G.I.L. Jones, N.L. Owen, J. Mol. Struct. 18 (1973) 1.
6] R. Rittner, E.L. Silva, E.L. Canto, Spectrosc. Int. J. 8 (1–6) (1990) 173–182.
7] R.J. Abraham, C.F. Tormena, R. Rittner, J. Chem. Soc., Perkin Trans. 2 (2001) 815.
8] B.J. Van der Veken, S. Truyen, W.A. Herrebout, G. Watkins, J. Mol. Struct. 293
(
1993) 55.
[
9] L. Tasic, R.J. Abraham, R. Rittner, Magn. Reson. Chem. 40 (7) (2002) 449–454.
[
[
10] A.I. Vogel, Practical Organic Chemistry), Lowe and Brydone, London, 1970.
11] A.M. Dowell Jr., H.S. McCullougti, P.K. Calaway, J. Amer. Chem. Soc. 70 (1948)
226.
‡
Fig. 7. (A) Plots of the Arrhenius molar enthalpy change of activation ꢀH and
‡
[12] D.A. Ramsay, J. Amer. Chem. Soc. 74 (1952) 72.
the entropy parameter TꢀS against the polarizability of the halogen atoms
[
[
[
13] S. Mizushima, Structure of Molecules and internal Rotation, Academic Press Inc
Publishers, New York, 1954.
14] W.J. Orville Thomas, “Internal Rotation in Molecules”, London, New York, Syd-
ney, Toronto, 1972.
0
at standard temperature
T
= 298.15 K. Linear regression parameters are; the
2
slope m = 1.42 ± 0.47, intercept b = 0.99 ± 1.16, ꢆ = 0.552, r = 0.94961 and No. of
‡
2
data points = 4. For TꢀS parameters, m = 1.81 ± 0.41, b = −0.85 ± 1.01, ꢆ = 0.0248,
r = 0.97553, No. of data points = 4. (B) Plot of Arrhenius Gibbs’ molar energy of
15] C.F. Tormena, F. Yoshinaga, T.R. Doi, R. Rittner, Spectrochim. Acta Part A 63
‡
‡
activation ꢀG against the polarizability of the halogen atoms. ꢀG parameters,
m = −0.39 ± 0.23, b = 0.86 ± 0.57, ꢆ = 0.00747, r = −0.86041, No. of data points = 4.
(
2006) 511–517.
[16] P.R. Olivato, R. Rittner, Rev. Heteroatom Chem. 15 (1996) 115–158.
2
[
[
[
17] B.H. Kong, B.C. Lee, I. Lee, K. Yang, Bull. Kor. Chem. Soc. 6 (No. 5) (1985) 227–279.
18] G.A. Groder, D. Jackson, Spectrochim. Acta 27 (1971) 1973.
19] D. Sinna, J.E. Katon, Can. J. Chem. 52 (1974) 3057.
From Eq. (6), one may expect that the Gibbs isothermal energy
change ꢀG for the kinetic process is the most reliable among
the three pseudo-thermodynamic quantities due to the enthalpy-
entropy cancellation. The ꢀH and ꢀS parameters have displayed
good linear relationships with the polarizability of the halogen
atoms but in the same fashion [Fig. 7A]. On the other hand, the
plot of ꢀG against the polarizability parameter exhibits a linear
regression line of negative slope but with appreciable scattering of
the data points [see Fig. 6B].
The values of ꢀS are extremely low because of the pre-
exponential factor values A being very low as given in Table 6. This
is a result of the low rate constant values which are in the order
of 10 per second and the differences between their consecutive
values are very small. However, the low ꢀS values indicate very
‡
[20] B. Nolin, R.N. Jones, J. Amer. Chem. Soc. 75 (1953) 5626.
[
[
21] J.K. Witmushurst, J. Mol. Spectrom. 1 (1957) 201.
22] A. Gambi, S. Colgrgianni, A. Passerini, R. Visioni, Spectrochim. Acta 38A (1982)
‡
‡
871.
[23] J.J.H.S. Green, D.J. Harrison, Spectrochim. Acta 32A (1976) 1265.
[
[
[
24] R. Zwarich, J. Smolarek, L. Goodman, J. Mol. Spectrom. 38 (1971) 336.
25] D.T. Mcbee, D.L. Christman, J. Amer. Chem. Soc. 77 (1955) 755.
26] A.B.F. Duncan, F.A. Masten, Chemical Application of Spectroscopy, Inter Science
Publishers Inc, New York, 1958.
‡
[
27] R.E. Robinson, R.C. Yalor, Spectrochim. Acta 13A (1962) 1094.
‡
[28] G.A. Guirgis, B.R. Drew, T.K. Gonev, J.R. Durig, Spectrochim. Acta 54A (1998)
23–143.
1
[29] D.T. Durig, Z. Yu, J. Mol. Struct. 550–551 (2000) 481–493.
[30] J.R. Durig, C. Zheng, M.A. Qtaitat, S. Deng, G.A. Guirgis, J. Mol. Struct. 657 (1–3)
(2003) 357–373.
−4
[
[
31] C.V. Hernan, J. Morcillo, An. Quim. Ser. A 77 (1) (1981) 70.
32] R.L. Tyson, J.A. Weil, J. Phys. Chem. 94 (1990) 351.
‡
small steric effect.