Enthalpy-entropy correlations in reactions of aryl benzoates with potassium aryloxides in dimethylformamide

Article abstract of DOI:10.1002/kin.20763

Temperature dependences of the relative reactivity of potassium aryloxides XC₆H₄O^-K⁺ toward 4-nitrophenyl (1), 3-nitrophenyl (2), 4-chlorophenyl (3), and phenyl (4) benzoates in dimethylformamide (DMF) were studied using the competitive reactions technique. The rate constants k_X for the reactions of 1 with potassium 4-cyanophenoxide, 2 with potassium 3-bromophenoxide, 3 with potassium 3-bromo-, 4-bromo-, and unsubstituted phenoxides, 4 with potassium 4-methoxy- and 3-methylphenoxides were measured at 25°C. Correlation analysis of the relative rate constants k_X/k_H(3-Me) and differences in the activation parameters (δδH^≠and δ δS ^≠) of competitive reactions revealed the existence of six isokinetic series. We investigated the substituent effect of X on the activation parameters for each isokinetic series and concluded that the reactions of aryl benzoates PhCO₂C₆H₄Y with potassium aryloxides in DMF proceed via a four-step mechanism. The large ρ⁰(Y) and ρ_XY values at 25°C obtained for the reactions of 1-3 with potassium aryloxides with an electron-donating substituent refer to the rate-determining formation of the spiro-σ-complex. The Hammett plots for the reactions of 1 and 2 exhibit a downward curvature, causing the motion of the transition state for the rate-determining step according to a Hammond effect as the substituent in aryloxide changes from electron-donating to electron-withdrawing. Analysis of data in the terms of two-dimensional reaction coordinate diagrams leads to the conclusion that significant anti-Hammond effects arise in the cases of ortho-substituted and unsubstituted substrates. It was shown that the isokinetic and compensation effects observed for the reactions of aryl benzoates with potassium aryloxides in DMF can be interpreted in the terms of the electrostatic bonding between the reaction centers.

Full text of DOI:10.1002/kin.20763

Enthalpy–Entropy
Correlations in Reactions
of Aryl Benzoates with
Potassium Aryloxides in
Dimethylformamide
IRINA A. KHALFINA, VLADISLAV M. VLASOV
N. N. Vorozhtsov Novosibirsk Institute of Organic Chemistry, the Siberian Branch of Russian Academy of Sciences,
Novosibirsk, 630090, the Russian Federation
Received 11 July 2012; revised 10 October 2012; accepted 21 October 2012
DOI 10.1002/kin.20763
Published online 22 February 2013 in Wiley Online Library (wileyonlinelibrary.com).
ABSTRACT: Temperature dependences of the relative reactivity of potassium aryloxides
−
+
XC6H4O K toward 4-nitrophenyl (1), 3-nitrophenyl (2), 4-chlorophenyl (3), and phenyl (4)
benzoates in dimethylformamide (DMF) were studied using the competitive reactions tech-
nique. The rate constants kX for the reactions of 1 with potassium 4-cyanophenoxide, 2 with
potassium 3-bromophenoxide, 3 with potassium 3-bromo-, 4-bromo-, and unsubstituted phe-
◦
noxides, 4 with potassium 4-methoxy- and 3-methylphenoxides were measured at 25 C. Cor-
relation analysis of the relative rate constants kX/kH(3-Me) and differences in the activation
ꢀ
ꢀ=
parameters (ꢀꢀН and ꢀꢀS ) of competitive reactions revealed the existence of six isoki-
netic series. We investigated the substituent effect of X on the activation parameters for each
isokinetic series and concluded that the reactions of aryl benzoates PhCO2C6H4Y with potas-
0
sium aryloxides in DMF proceed via a four-step mechanism. The large ρ (Y) and ρXY values at
◦
2
5 C obtained for the reactions of 1–3 with potassium aryloxides with an electron-donating
substituent refer to the rate-determining formation of the spiro-σ-complex. The Hammett plots
for the reactions of 1 and 2 exhibit a downward curvature, causing the motion of the transition
state for the rate-determining step according to a Hammond effect as the substituent in ary-
loxide changes from electron-donating to electron-withdrawing. Analysis of data in the terms
of two-dimensional reaction coordinate diagrams leads to the conclusion that significant anti-
Hammond effects arise in the cases of ortho-substituted and unsubstituted substrates. It was
shown that the isokinetic and compensation effects observed for the reactions of aryl benzoates
with potassium aryloxides in DMF can be interpreted in the terms of the electrostatic bonding
between the reaction centers. ꢁC 2013 Wiley Periodicals, Inc. Int J Chem Kinet 45: 266–282, 2013
Correspondence to: V. M. Vlasov; e-mail: vmvlasov@nioch
.
ꢀ
nsc.ru.
C
2013 Wiley Periodicals, Inc.

REACTIONS OF ARYL BENZOATES WITH POTASSIUM ARYLOXIDES IN DMF
267
INTRODUCTION
Hammett plot has been understood as a change in the
RDS within a given mechanism upon changing the sub-
stituent from the electron-donating substituent (EDS)
to electron-withdrawing substituent (EWS) [49].
We have now expanded our study to the reactions
of 4-nitrophenyl, 3-nitrophenyl, 4-chlorophenyl, and
phenyl benzoates (1, 2, 3, and 4, respectively) with the
series of potassium aryloxides in DMF at four temper-
One of the most common and important reactions in or-
ganic chemistry is that of aryl esters with nucleophiles.
Acyl group transfer reactions have been studied by sev-
eral groups, and there is now considerable data on the
kinetics and mechanisms of those processes [1–49].
Two distinct reaction mechanisms, concerted ANDN
and stepwise addition-elimination AN + DN, have been
suggested. As a rule, the aminolysis reactions of aryl
esters proceed through the addition-elimination mech-
anism, in which the rate-determining step (RDS) is de-
pendent on the basicity of the entering amine and the
leaving group [14,21–27,29–40,42–44]. The evidence
provided is a nonlinear Brønsted-type plot that has of-
ten been observed for the reactions of esters with a good
leaving group [25,26,29,30,32–34]. However, the step-
wise mechanism for the reactions of 2,4-dinitrophenyl
acetate with secondary alicyclic amines shifts to the
concerted process when the solvent is changed from
water to ethanol/water mixtures [45]. Moreover, the
corresponding reactions of 2,4,6-trinitrophenyl acetate
in water proceed via the concerted mechanism [45].
The reactions of carboxylic esters with anionic
nucleophiles have not been completely elucidated.
Williams and coworkers have concluded that the nucle-
ophilic substitution reactions of 4-nitrophenyl acetate
with a series of aryloxides proceed through the con-
certed mechanism. The evidence consisted mainly of
the absence of a break in the Brønsted-type plot when
the pKa value of the incoming aryloxide corresponded
to that of the leaving 4-nitrophenoxide [1,5,9,11,13].
The concerted mechanism has been supported by Jenck
and colleagues [6], Rossi and coworkers [12,15], and
Castro et al. [18–20], as well as the kinetic isotope
effect studies of Hengge [4,7,8,17].
◦
atures in the range from −15 to 70 C to obtain more
conclusive information about the reaction mechanism
(Scheme 1).
EXPERIMENTAL
Materials
Aryl benzoates 1–4 (see Fig. 1) used in the present
study were prepared as described [50]. Their purity was
checked by means of their melting point and spectral
1
data such as IR and H NMR characteristics. Commer-
cial substituted phenols of chemically pure grade were
purified by standard methods. Potassium aryloxides
were prepared from the reactions of substituted phenols
with potassium methoxide in absolute methanol [51].
Commercial DMF of analytical grade was dried over 4
˚
A molecular sieves and distilled over calcium hydride
under reduced pressure.
Procedures
The direct-measurement kinetic method has been used
to determine the rate constants kX for the relative slow
reactions of 1 with potassium 4-cyanophenoxide, 2
with potassium 3-bromophenoxide, 3 with potassium
3-bromo-, 4-bromo-, and unsubstituted phenoxides, 4
with potassium 4-methoxy- and 3-methylphenoxides
(t1/2 > 4 s). In a typical procedure, the glass reac-
tor was initially charged with 5 mmol of aryloxide in
50–100 mL of DMF, placed in the circulating constant-
On the contrary, Buncel et al. have concluded that
the reactions of phenoxide with a series of substituted
phenyl acetates proceed through the stepwise mecha-
nism on the basis of the kinetic result that the σ con-
stants exhibit a better Hammett correlation than the σ
0
−
◦
constants [2,3,10,41]. A similar conclusion has been
drawn from the kinetic studies of the reactions of aryl
temperature bath and heated to 25 ± 0.2 C. The mag-
netic stirrer was turned on. The solution of 0.5 mmol
of aryl benzoate (0.2 mmol for the reactions of 3 with
3-bromo- and 4-bromophenoxides) in 1 mL of DMF
was adjusted to the same temperature and added in one
portion. From this moment, the reaction was timed.
Samples were drawn at prespecified time intervals.
Approximately, 10 samples were collected during the
course of each reaction. The frequency of sample col-
lection varied and was dictated by the reaction condi-
tion. Samples (3–5 mL) were collected in 10-mL test
tubes filled with 5 mL of the cooled mixture of diethyl
ether and water (2:3). The test tubes were then shaken
−
−
−
benzoates with OH , CN , and N3 ions [16,28,40].
The stepwise mechanism has been supported through
the enthalpy–entropy correlations for the reactions of
4-nitro- and 2,4-dinitrophenyl benzoates with substi-
tuted phenols in the presence of potassium carbon-
ate [46–48].
In the previous work, we have reported that the re-
actions of 2,4-dinitrophenyl benzoate (5) with a series
of aryloxides in dimethylformamide (DMF) proceed
through the stepwise mechanism in which the spiro-
σ-complex is an intermediate. The downward-curved
International Journal of Chemical Kinetics DOI 10.1002/kin.20763

2
68
KHALFINA AND VLASOV
Scheme 1
Figure 1 The structures of aryl benzoates (1–5) considered in this study.
to stop the reaction. The organic phase was separated;
the aqueous phase was extracted with diethyl ether
and heated to the desired temperature. The solution of
0.5 mmol of aryl benzoate in 1 mL of DMF was ad-
justed to the same temperature and then added in one
portion. The reaction time was varied from 5 s to 1 h
and depended on the substrate reactivity. A sample (3–
5 mL) was collected in a 10-mL test tube filled with
5 mL of the cooled mixture of diethyl ether and water
(2:3). The test tube was then shaken to stop the re-
action. The organic phase was separated; the aqueous
phase was extracted with diethyl ether (2 × 3 mL). The
extracts were combined with the organic phase, the sol-
vent was distilled off, and the residue was analyzed by
GLC. In every case, we found only the resulting aryl
benzoates (PhCO2C6H4X), which do not practically
undergo the reverse reactions. Since for each pair of
potassium aryloxides, the competition reactions were
performed at equimolar concentrations exceeding the
concentration of the substrate by a factor of 5, so the
concentration ratio of the products is directly propor-
tional to the ratio of the second-order rate constants for
each reaction.
As the reference compound for reactions of 1−3
with potassium 4-methoxy-, 3-methyl-, 4-chloro-,
and 4-bromophenoxides, we used potassium phenox-
ides; the relative rate constants kX/kH are given in
Table I. For potassium 4-methyl-, 3-nitro-, 4-cyano-,
and 3-bromophenoxides, we determined the rel-
ative rate constants k4-MeO/k4-Me (for substrates
1−3), k4-Cl/k3-NO2 and k3-NO2/k4-CN (for substrate 1),
k4-Cl/k3-Br (for substrate 2), which were recalculated to
kX/kH (Table I). For the reactions of 4 with potassium 4-
methoxy- and 4-methylphenoxides, we determined the
relative rate constants k4-MeO/k3-Me and k4-MeO/k4-Me,
the latter was recalculated to k4-Me/k3-Me (Table I).
(
2 × 3 mL). The extracts were combined with the
organic phase, the solvent was distilled off, and the
residue was analyzed by gas–liquid chromatography
(GLC).
It has previously been shown that the reactions
of 1, 2, and 3 with potassium 4-cyano-, 3-bromo-,
and unsubstituted phenoxides, respectively, are irre-
versible. The pseudo–first-order rate constants (kobs)
were obtained from the linear plots of ln(A∞ − At)
versus t. For the reactions of 4 with 4-methoxy- and
3-methylphenoxides, kobs were obtained from the plots
of ln(A∞ − At) versus t, which were linear up to 90% of
the reactions. For the reactions of 3 with 3-bromo- and
4-bromophenoxides, kobs were obtained from the plots
of ln(A∞ − At) versus t, which were linear up to 50%
of the reactions. Four different nucleophile concentra-
tions were used to determine the second-order rate con-
stants (kN) from the slopes of the linear plots of kobs ver-
sus potassium aryloxide concentration, which showed
near-zero intercepts, indicating that potassium arylox-
ide does not exist as dimers or other aggregates [52].
The correlation coefficients of the plots were usually
higher than 0.999.
It was difficult to use this direct kinetic method for
studying fast reactions. As we showed earlier, for the
fast reactions of 5 with potassium aryloxides in DMF
the relative rate constants kX/kH were determined using
the competitive reactions technique [49]. In a typical
run, the solution of a pair of potassium aryloxides with
the pKa values larger than that of the leaving aryloxide
(
2.5 mmol + 2.5 mmol) in 50 mL of DMF was placed
in the glass reactor equipped with a magnetic stirrer
International Journal of Chemical Kinetics DOI 10.1002/kin.20763

REACTIONS OF ARYL BENZOATES WITH POTASSIUM ARYLOXIDES IN DMF
269
International Journal of Chemical Kinetics DOI 10.1002/kin.20763

2
70
KHALFINA AND VLASOV
International Journal of Chemical Kinetics DOI 10.1002/kin.20763

REACTIONS OF ARYL BENZOATES WITH POTASSIUM ARYLOXIDES IN DMF
271
Scheme 2
Analysis
matically decreases as the substituent X changes from
EDS to EWS. For example, the ρ(X) value decreases
Samples were analyzed by GLC using an LKhM-72 gas
chromatograph equipped with a thermal conductivity
detector (4000 × 4-mm column packed with 15% of
SKTFT-803 on Chromaton-W, carrier gas helium, lin-
◦
from 2.19 to –3.35 at 25 C for the reactions of 1 and
◦
from 0.99 to –4.23 at 25 C for the reactions of 2. Such
a large change in the selectivity parameter has been in-
terpreted as a change in the RDS [2,25–28,41]. In addi-
tion, for the reactions of 1, 2 with potassium aryloxides
with an EDS, the reactivity–selectivity principle (RSP)
fails, since the more reactive system exhibits the larger
selectivity [56].
◦
ear oven temperature programming from 70 to 270 C at
the rate of 10 deg/min); the components were quantited
by the absolute calibration technique using calibration
plots and identified by adding authentic samples.
Um et al. have reported that the reactions of 1 with
◦
RESULTS AND DISCUSSION
a series of aryloxides in absolute ethanol at 25 С pro-
ceed through the stepwise mechanism in which the
formation of the symmetric tetrahedral intermediate
is the RDS [55,57]. A good Hammett correlation has
been obtained with ρ(X) = −1.96 [55]. It is evident
that the downward-curved Hammett plots obtained for
the reactions of 1 and 2 with potassium aryloxides in
DMF are contrary to the proposed mechanism.
The reactions of 5 with potassium aryloxides
in DMF proceed through the stepwise mechanism
in which the spiro-σ-complex is an intermediate
Effect of Entering Group Substituent
on Reaction Selectivity
To investigate the effect of the entering group sub-
stituent X on the reaction selectivity, we have used the
relative rate constants kX/kH(3-Me). As shown in Table I,
the relative reactivity of potassium aryloxides toward
1
decreases as the substituent X in the entering group
changes from 3-Me to stronger EDS or EWS. A similar
effect of the substituent X on the relative reactivity of
potassium aryloxides has been found for the reactions
of 2. On the other hand, the kX/kH and kX/k3-Me val-
ues for the reactions of 3 and 4, respectively, decrease
as the electron-donating ability of substituent X de-
creases. Within the corresponding temperature range,
the relative rate constant kX/kH(3-Me) changes no more
than 8 times for every substrate.
(
Scheme 2). The ρ(X) value has been found to de-
◦
crease from 0.49 to −3.33 at 25 С as the substituent
X changes from EDS to EWS [49]. The similar
downward-curved Hammett plots obtained for the re-
actions of 1 and 2 with potassium aryloxides can be
taken as an indirect evidence for the same mechanism.
For the reactions of 3 and 4 with potassium arylox-
ides with an EDS, the ρ(X) values are negative (Ta-
Good linear correlations (r ≥ 0.992) have been ob-
ble I). Since the ρ(X) value decreases from –1.24 to –
−
tained in the plots log(kX/kH(3-Me)) versus σ (Х) us-
◦
2
.20 at 25 C as the leaving group substituent Y changes
ing potassium aryloxides with an EDS (or an EWS).
The selectivity parameter ρ(X) values summarized
in Table I were calculated from the Hammett-type
equation,
from 4-Cl to H, one may conclude that the more reac-
tive system exhibits the less selectivity, i.e., the RSP
holds. Thus, these reactions can proceed through the
formation of a symmetric tetrahedral intermediate [58].
Nevertheless, a failure of the RSP is the charac-
teristic feature for the reactions of aryl benzoates 1–4
with potassium aryloxides in DMF, since for all the
reaction series the selectivity increases with increasing
temperature (Table I).
−
log(kX/kH(3−Me)) = ρ(X)σ (X) + const.
The positive ρ(X) values for the reactions of 1 and
2
with potassium aryloxides with an EDS indicate that
the kX value increases with decreasing the electron-
donating ability of the substituent X. A similar effect
of the entering group substituent has been found for the
reactions of 1 with acetophenone oximates in absolute
ethanol [55]. As shown in Table I, the ρ(X) value dra-
Similar conclusions have been drawn on the basis
of analysis of the Brønsted-type plots (Table I). It is
evident that data obtained for the reactions of 1–4 with
potassium aryloxides cannot give any significant infor-
mation regarding the reaction mechanism.
International Journal of Chemical Kinetics DOI 10.1002/kin.20763

2
72
KHALFINA AND VLASOV
Figure 2 Hammett plots for the reactions of 4-nitrophenyl, 3-nitrophenyl, 4-chlorophenyl, and phenyl benzoates (1–4) with
−
+
◦
potassium aryloxides XC6H4O K in DMF at 25 C.
◦
ρ0(Y) indicate a large charge change on the phenolic
oxygen on going from ground state to transition state
[59].
Using the relative rate constants kX/kH(3-Me) at 25 С
obtained by the competition kinetic method and the
rate constants kX obtained by the direct-measurement
◦
0
kinetic method for the seven slow reactions at 25 С,
The ρ (Y) values for the reactions 1–3 with potas-
the rate constants kX for other reactions can be cal-
culated. The rate constants kX determined in this way
are summarized in Table II. The Hammett plots for the
sium aryloxides with an EDS calculated from the
Hammett-type equation,
reactions of 1–4 with potassium aryloxides in DMF at
0
0
log kX = ρ (Y)σ (Y) + const,
◦
25 С are shown in Fig. 2.
are unusually large and range from 5.47 to 6.66 at
◦
2
5 C. As shown in Table II, a stronger EDS in arylox-
Effect of Leaving Group Substituent
on Reactivity
−
0
ide, ꢀσ (X) < 0, leads to decrease in the ρ (Y) value,
ρ (Y)<0, and the RSP holds. However, a weaker
0
ꢀ
0
As shown in Table II, for the reactions of 1–
EWS in the leaving group, ꢀσ (Y)<0, leads to de-
crease and change of the sign of ρ(X), ꢀρ(X) < 0.
Good linear correlations (r ≥ 0.989) have been ob-
3
with potassium 4-methoxy-, 4-methyl-, and 3-
methylphenoxides, the Hammett plots for variations
of the substituent Y in the leaving group exhibit much
0
−
tained in the plots ρ (Y) versus σ (Х) and ρ(X) ver-
0
0
better linear correlations with the σ (Y) constants than
sus σ (Y) for the reactions of 1–3 with aryloxides with
an EDS, the cross-interaction constant ρXY is positive
and unusually large, 5.92 at 25 C.
−
with the σ (Y).
0
◦
The correlation of reaction rates with the σ (Y)
constants implies that direct resonance interactions of
the leaving group substituent Y and the reaction center
do not occur [3]. A similar σ dependence has been
observed for many reactions, e.g., for the reactions
of aryl benzoates with potassium ethoxide in absolute
ethanol [2,3,10,41]. Haake et al. have suggested that
in the ground state π interaction of the alcohol oxygen
with the carbonyl oxygen results in a partial positive
charge on the alcoholic oxygen. So, the large values of
It should be noted that the standard error of ρXY
strongly decreases when the plot changes from ρ(X)
0
0
0
−
versus σ (Y) to ρ (Y) versus σ (Х) (Table II). Since
the ρ(X) values are associated with the low standard
errors, so the relatively low accuracy of the determined
ρXY value indicates that for the reactions of 1–3 with
aryloxides with an EDS use the common scale of the
0
p
σ and σm constants of the substituent Y is not suffi-
ciently correct. This problem can also be the cause of
International Journal of Chemical Kinetics DOI 10.1002/kin.20763

REACTIONS OF ARYL BENZOATES WITH POTASSIUM ARYLOXIDES IN DMF
273
International Journal of Chemical Kinetics DOI 10.1002/kin.20763

2
74
KHALFINA AND VLASOV
0
the relatively large standard errors of the ρ (Y) values
Isokinetic and Compensation Effects
(
Table II).
Each reaction series has been tested in terms of the
isokinetic and compensation effects [62–65]. The ex-
istence of the isokinetic relationship (IKR) can serve as
an argument that the reaction series follow a common
mechanism [66–69].
Good linear correlations (r ≥ 0.999) have been ob-
−
tained in the plots log(kX) versus σ (Y) for the reac-
tions of 1–3 with 3-bromo- and 4-bromophenoxides.
The ρ(Y) values are 3.31 and 3.58 at 25 C for 4-bromo-
and 3-bromophenoxides, respectively. The low stan-
dard errors of ρ(Y) confirm correct use of the common
scale of the σp and σm constants of the substituent Y
for the reactions of 1–3 with aryloxides with an EWS.
As shown in Tables I and II, for the reactions of 1–
3
◦
Good linear correlations (r ≥ 0.989) obtained in
the plots of log(kX/kH(3-Me)) versus 1/T indicate that
all the reactions follow the Arrhenius temperature de-
pendence. The differences in the activation parame-
−
ꢀ
ꢀ
ꢀ
ꢀ=
and ꢀꢀS
ters ꢀꢀH = ꢀHX − ꢀH
=
−
H(3−Me)
a stronger EWS in aryloxide, ꢀσ (X) > 0 (in the
ꢀ
ꢀ
−
ꢀS − ꢀS
leaving group, ꢀσ (Y) > 0) leads to increase in ρ(Y),
ꢀ
X
H(3−Me)
type equation are summarized in Table I. As shown in
Table I, the small magnitudes of ꢀꢀH and ꢀꢀS
ρ(Y) > 0 (in ρ(X) < 0, ꢀρ(X) > 0) and the RSP
ꢀ
ꢀ=
holds in the both cases. The cross-interaction constant
ρXY has been estimated to be 1.63 at 25 C.
Data of Table II can also be analyzed using the full
equation [60],
◦
(−1.8 to 4.4 kJ/mol and −10.4 to 21.6 J/(mol·K)) be-
come larger (−24.9 to −18.1 kJ/mol and −150.0 to
−
82.5 J/(mol·K)) as the substituent X changes from
EDS to EWS.
We can conclude that each Hammett reaction series
is isokinetic, since the linear plots of log(kX/kH(3-Me))
versus 1/T intersect near one point (Table III). The
Tiso values determined from the log(kX/kH(3-Me)) − 1/T,
ρ(X) − 1/T [70], βNuc − 1/T, and log(kX/kH(3-Me))T2 −
log(kX/kH(3-Me))T1 [71] dependences differ only slightly
log (kXY/kHH) = ρXσ(X) + ρYσ(Y)
+
σXYσ(X)σ(Y) + c
where kXY is the rate constant in the presence of sub-
stituents X and Y, kHH is the rate constant when X =
Y = H, ρX, ρY, and ρXY are first-derivative sensitiv-
ity parameters and the cross-interaction constant, and
(Table III), that is more reliable evidence for the exis-
tence of IKR [72].
The statistically correct values of the compensa-
tion temperature Тcomp for isokinetic series determined
−
c is the intercept. The σ scale is used for the entering
0
group substituent X, and the σ scale is used for the
ꢀ
ꢀ=
from the slopes of the plots of ꢀꢀH versus ꢀꢀS
leaving group substituent Y. It is obvious that ρ(X) =
are also summarized in Table III. A similarity of the
Тcomp and Tiso values for each of reaction series indi-
cates that the compensation and isokinetic effects can
be synonymous [66].
0
0
−
ρX + ρXYσ (Y) and ρ (Y) = ρY + ρXYσ (X).
For the reactions of 1–3 with potassium arylox-
ides with an EDS, the overall correlation coefficient
of 0.998 (n = 9) is associated with the values of
For all isokinetic series in the current study, the Tiso
−
2.90 ± 1.02 for ρX, 7.07 ± 0.30 for ρY, 5.91 ± 1.58
ꢀ
ꢀ=
values are positive, so the δꢀH and δꢀS reaction
for ρXY, and −2.01 ± 0.19 for c. For the reactions of
ꢀ
ꢀ=
constants have the same sign. The δꢀH and δꢀS
reaction constants determined from the dependence of
1
–3 with potassium 4-bromo- and 3-bromophenoxides,
the overall correlation coefficient of 0.999 (n = 6) is
ꢀ
ꢀ=
ꢀH
associated with the values of −5.38 ± 0.37 for ρX,
stituent constants can be used for the characterization
of the electronic and steric effects of the substituents
on the reactivity [62–64,72]. Since for all the series the
Тexp > Tiso ratio holds, so the effect of the entering
group substituent X on the reactivity is controlled by
steric factors [62–65,73]. For this reason, the magni-
tude of ρ(X) increases with increasing temperature and
the RSP fails.
2
0
.94 ± 0.14 for ρY, 1.62 ± 0.44 for ρXY, and −0.37 ±
.12 for c.
Therefore, for the reactions of aryl benzoates with
potassium aryloxides in DMF, the first-derivative sen-
sitivity parameters ρX and ρY values are negative and
positive, respectively, and ones decrease as the sub-
stituent X changes from EDS to EWS.
It is important to emphasize that the accuracy of
the ρXY value calculated from the full equation is
Good linear correlations (r ≥ 0.993) have
ꢀ
ꢀ
=
been obtained in the plots ꢀꢀH and ꢀꢀS
0
less than that determined from the ρ(X) − σ (Y) or
−
ꢀ=
versus σ (Х) for each isokinetic series. The δꢀH
0
−
ρ (Y) − σ (X) dependences. We can confirm that
artificial optimizations resulting from statistical treat-
ments of multiple regressions increase the standard
error of ρXY [61].
ꢀ=
and δꢀS values determined from the slopes of the
corresponding plots are presented in Table IV. These
reaction constants have also been determined from the
slopes and intersections, respectively, of the linear plots
International Journal of Chemical Kinetics DOI 10.1002/kin.20763

REACTIONS OF ARYL BENZOATES WITH POTASSIUM ARYLOXIDES IN DMF
275
Table III Isokinetic Temperatures Тiso Determined from Four Dependences and Compensation Temperatures Тcomp
–
+
for Reactions of Aryl Benzoates PhCO2C6H4Y (1–4) with Potassium Aryloxides XC6H4O K in DMF
Tiso (К)
log(kX/kH(3-Me))T2 –
log(kX/kH(3-Me))T1
Тcomp (K)
ꢀꢀH −ꢀꢀS
а
b
c
d
ꢀ
ꢀ= e
Y
X
log(kX/kH(3-Me)) −1/T
ρ(X) −1/T
βNuc −1/T
4
3
4
-NO2
EDSf
EWS
EDSh
EWS
EDS
EDS
209
70
234
114
212
99
208
69
209
69
209
73
210
70
g
-NO2
-Cl
235
115
213
96
235
116
213
94
234
119
213
104
239
117
207
95
i
j
k
H
a
b
c
d
e
f
T
iso is the temperature of intersection of the plots log(k
X
/kH(3-Me)) − 1/T.
T
iso is calculated from the equation ρ = const(1– Tiso /T).
T
iso is calculated from the equation βNuc = const(1– Tiso /T).
iso is calculated from the equations log(k
T2 = a + b log(k
comp is calculated from the equation ꢀꢀH = const. + ТcompꢀꢀS .
T
Т
X
/kH(3-Me)
)
X
ꢀ=
/kH(3-Me)
)
T1 and Tiso = T
2 1 2 1
T (b – 1)/(bT – T ).
ꢀ
Tiso = 209 K s = 2; ρ(Х) = (7.35 ± 0.07) − (1529 ± 20)/Т (r = 0.999, s = 0.011, n = 4); βNuc = −(1.63 ± 0.02) + (341 ± 6)/Т (r =
0
.999, s = 0.003, n = 4); log(k
Х
/k /k T1 for T
H
)
T2 = (0.068 ± 0.004) + (1.75 ± 0.03); log(k
Х
H
)
1
= 258 K, T = 313 K (r = 0.999, s = 0.006,
2
ꢀ=
3
ꢀ=
n = 3); ꢀꢀH = (0.36 ± 0.04) + (0.210 ± 0.003) × 10 ; ꢀꢀS (r = 0.999, s = 0.063, n = 3).
g
Tiso = 70 K, s = 2; ρ(X) = −(4.38 ± 0.12) + (302 ± 34)/Т (r = 0.988, s = 0.018, n = 4); βNuc = (0.96 ± 0.04) – (66 ± 11)/Т (r =
0
.973, s = 0.006, n = 4); log(k
Х
/k
H
)
T2 = (0.973 ± 0.006) + (1.07 ± 0.01); log(k
Х
/k
H
)
T1 for T
1
= 258 K, T = 313 K (r = 0.999, s = 0.006,
2
ꢀ=
3
ꢀ=
n = 3); ꢀꢀH = – (14.45 ± 0.09) + (0.070 ± 0.001) × 10 ; ꢀꢀS (r = 0.999, s = 0.037, n = 3).
h
Tiso = 234 K, s = 2; ρ(X) = (4.59 ± 0.12) − (1078 ± 34)/T (r = 0.999, s = 0.018, n = 4); βNuc = −(1.02 ± 0.03) + (240 ± 8)/T (r =
0
.999, s = 0.004, n = 4); log(k
Х
/k
H
)
T2 = −(0.275 ± 0.016) + (2.75 ± 0.08); log(k
Х
/k
H
)
T1 for T
1
= 258 K, T = 313 K (r = 0.999, s = 0.005,
2
ꢀ=
3
ꢀ=
n = 3); ꢀꢀH = – (0.75 ± 0.02) + (0.239 ± 0.002) × 10 ; ꢀꢀS (r = 0.999, s = 0.019, n = 3).
i
T
iso = 114 K, s = 1; ρ(X) = −(6.96 ± 0.28) + (802 ± 79)/Т (r = 0.990, s = 0.042, n = 4); βNuc = (1.56 ± 0.06) – (181 ± 16)/Т (r =
0
.992, s = 0.008, n = 4); log(k
Х
/k
H
)
T2 = −(0.484 ± 0.007) + (1.15 ± 0.01); log(k
Х
/k
H
)
T1 for T
1
= 258 K, T = 313 K (r = 0.999, s = 0.003,
2
ꢀ=
3
ꢀ=
n = 3); ꢀꢀH = – (7.50 ± 0.12) + (0.117 ± 0.001) × 10 ; ꢀꢀS (r = 0.999, s = 0.023, n = 3).
j
T
iso = 212 K, s = 3; ρ(X) = −(4.30 ± 0.09) + (916 ± 27)/Т (r = 0.999, s = 0.012, n = 4); βNuc = (0.94 ± 0.04) – (200 ± 11)/Т (r =
0
.997, s = 0.005, n = 4); log(k
Х
/k
H
)
T2 = − (0.077 ± 0.005) + (1.59 ± 0.02); log(k
Х
/k
H
)
T1 for T
1
= 273 K, T = 328 K (r = 0.999, s = 0.003,
2
ꢀ=
3
ꢀ=
n = 3); ꢀꢀH = – (0.49 ± 0.01) + (0.207 ± 0.001) × 10 ; ꢀꢀS (r = 0.999, s = 0.011, n = 3).
k
Tiso = 99 K, s = 4; ρ(Х) = −(3.25 ± 0.04) + (313 ± 13)/Т (r = 0.998, s = 0.004, n = 4); βNuc = (0.72 ± 0.01) − (68 ± 2)/Т (r = 0.999,
ꢀ
s = 0.001, n = 4); log(k
Х
/k
H
)
T2 = 1.07; log(k
Х
/k
H
)
T1 for T
1
= 298 K, T
2
= 343 K (r = 1.000, s = 0.001, n = 3); ꢀꢀH = (0.01 ± 0.02) +
3
ꢀ=
(
0.095 ± 0.002) × 10 ; ꢀꢀS (r = 0.999, s = 0.019, n = 3).
ρ(X) versus 1/T. The results obtained by independent
ways are close, so a validity of the δꢀH and δꢀS
values are confirmed. Table IV also contains the previ-
the carbonyl carbon to yield a tetrahedral intermedi-
ate T, followed by the intramolecular cyclization; (2)
the intramolecular cyclization of aryl benzoate, gen-
erating a zwitterionic Meisenheimer complex, which
eventually reacts with aryloxide; (3) the addition of
aryloxide and the intramolecular cyclization proceed
concertedly. One can suggest that the first and latter
mechanisms are the most plausible.
ꢀ
ꢀ=
ously published data for the reactions of 5 with potas-
ꢀ=
sium aryloxides in DMF [49]. The negative δꢀH and
ꢀ=
δꢀS values obtained for all the reaction series except
those of 1 and 2 with potassium aryloxides with an
ꢀ=
EDS indicate that the activation parameters ꢀH and
ꢀ=
0
ꢀ
S
decrease with decreasing the nucleophilicity of
The large ρ (Y) values for the reaction series of 1–3
aryloxides.
with aryloxides with an EDS indicate that the spiro-σ-
complex formation should be referred to the nucle-
ophilic aromatic substitution, i.e., the intramolecular
cyclization of T is the RDS (Scheme 2). As evidence,
Mechanism of Reactions of Aryl Benzoates
with Potassium Aryloxides in DMF
0
the magnitude of ρ (Y) for the reactions of 4-Y-2,6-
In the generally accepted stepwise mechanism of acyl
transfer reactions, a nucleophile attacks at the carbonyl
carbon pushing the π bond electrons up onto the oxy-
gen to give a tetrahedral intermediate [74]. Three re-
action mechanisms can be proposed for the spiro-σ-
complex formation: (1) the addition of aryloxide at
dinitrochlorobenzene (Y = NO , CN, CF ) with ben-
2
3
zylamines in MeCN are in the range of 6.25−6.43 [75].
0
A similar ρ (Y) value has been obtained for the reac-
tions of 4-Y-2-nitrochlorobenzene (Y = OMe, Me, H,
Cl, CO Me, COMe, NO ) with 4-methylthiophenoxide
2
2
◦
in DMF at 25 C [76]. The large positive ρ also is
XY
International Journal of Chemical Kinetics DOI 10.1002/kin.20763

2
76
KHALFINA AND VLASOV
ꢀ
ꢀ=
Table IV Reaction Constants δꢀH and δꢀS for Reactions of Aryl Benzoates PhCO2C6H4Y (1–4) and
2
−
+
,4-Dinitrophenyl Benzoate (5) with Potassium Aryloxides XC6H4O K in DMF
X
δꢀH (kJ/mol)a
ꢀ
δꢀS (J/(mol·K))
ꢀ
b
Tiso (К)c
Y
EDS
EWS
EDS
EWS
EDS
EWS
e
140.5d,k (140.7)
e
4
3
4
-NO2
-NO2
-Cl
29.5d,k (29.3)
−5.8 (−5.8)
−83.8 (−83.9)
209
234
212
99
70
114
f,k
g
f,k
g
21.0 (20.6)
−15.5 (−15.4)
88.0 (87.9)
−132.9 (−133.3)
h,k
h,k
−17.0 (−17.5)
−82.0 (−82.3)
i
i
H
2
−6.0 (−6.0)
−63.0 (−62.2)
,4-(NO2)2 j
−21.0
−36.9
−61.5
−185.1
345
215
aδꢀH is the slope of the plot ꢀꢀH versus. σ (X). Data in the brackets are obtained from the dependence ρ − 1/T (Table III).
ꢀ
=
ꢀ
−
bδꢀS is the slope of the plot ꢀꢀS versus. σ (X). Data in the brackets are obtained from the dependence ρ − 1/T (Table III).
ꢀ
=
ꢀ
−
c
Data are taken from Table III.
ꢀ
d
ꢀ=
−
ꢀ
−
ꢀH = (6.1 ± 0.2) + (29.5 ± 0.7)σ (X) (r = 0.999, s = 0.122, n = 3); ꢀꢀS = (27.4 ± 0.4) + (140.5 ± 2.0)σ (X) (r = 0.999,
s = 0.286, n = 3).
ꢀH = −(19.1 ± 0.2) − (5.8 ± 0.3)σ (X) (r = 0.999, s = 0.147, n = 3); ꢀꢀS = −(67.0 ± 1.9) − (83.8 ± 2.7)σ (X) (r = 0.999,
s = 1.585, n = 3).
e
ꢀ=
−
ꢀ
−
ꢀ
f
ꢀ=
−
ꢀ
−
ꢀ
ꢀH = (5.9 ± 0.0) + (21.0 ± 0.0)σ (X) (r = 1, s = 0, n = 3); ꢀꢀS = (27.7 ± 0.1) + (88.0 ± 0.6)σ (X) (r = 0.999, s = 0.082,
n = 3).
g
ꢀ=
−
ꢀ
−
ꢀ
ꢀH = −(15.1 ± 0.0) − (15.5 ± 0.1)σ (X) (r = 0.999, s = 0.015, n = 3); ꢀꢀS = −(65.3 ± 0.1) − (132.9 ± 0.4)σ (X) (r = 1,
s = 0.062, n = 3).
ꢀH = −(1.7 ± 0.2) − (17.0 ± 1.2)σ (X) (r = 0.998, s = 0.163, n = 3); ꢀꢀS = −(5.6 ± 1.0) − (82.0 ± 5.2)σ (X) (r = 0.998,
s = 0.735, n = 3).
h
ꢀ=
−
ꢀ
−
ꢀ
i
ꢀ=
−
ꢀ
−
ꢀ
ꢀH = −(0.5 ± 0.1) − (6.0 ± 0.6)σ (X) (r = 0.995, s = 0.082, n = 3); ꢀꢀS = −(4.8 ± 1.4) − (63.0 ± 7.5)σ (X) (r = 0.993,
s = 1.061, n = 3).
j
Data are taken from [49].
k
ꢀ
=
0
ꢀ
0
δꢀH = (−39.9 ± 1.0) + (85.1 ± 1.5)σ (Y) (r = 0.999, s = 0.617, n = 3); δꢀS = (−190.8 ± 9.7) + (399.4 ± 15.0)σ (Y) (r = 0.999,
s = 6.172, n = 3).
in line with the rate-limiting formation of a Meisen-
heimer complex. The cross-interaction constant ρXY
for the reactions of 4-Y-2,6-dinitrochlorobenzene (Y
can be explained by the significant difference between
the substituent effect on the solvation of the neutral
substrate and that of the anionic transition state. How-
ever, for the anionic T and TS (transition state) this
difference in DMF should not be essential, and so the
=
NO2, CN, CF3) with pyridines in MeCN is 3.95 at
5 C [77].
◦
2
ꢀ
ꢀ
ꢀ=
According to Scheme 2, the rate constant kX is ef-
fective and contains the equilibrium constant K1 of
the first fast step and the rate constant k2 of the sec-
solvent effect on δꢀH and δꢀS (as well as δꢀH
2
2
ꢀ=
and δꢀS ) is insignificant.
Generally speaking, the linear dependence of
δꢀH (i.e., ꢀH ) on the σ constant of the enter-
ꢀ
0
ꢀ
ꢀ
ꢀ
ꢀ
−
ond RDS. Then, ꢀH = ꢀH + ꢀH and ꢀS
=
1
2
2
0
1
ꢀ
=
0
1
ꢀ
0
1
ꢀ
ing group substituent X indicates that the TS formation
can be considered in the terms of the electrostatic bond-
ing [78] between the aryloxide oxygen (the O-reaction
center) and the carbon at the ipso position in the leaving
group (the C-reaction center), which is accompanied
by an entropy decrease due to forming 1,3-dioxetane
ring.
Williams et al. studied in detail associations be-
tween two or more molecules where the electrostatic
interactions play an important role [78]. The mono-
tonic dependence of the enthalpic benefit on the en-
tropic cost is typical of associations, and it can be the
cause of the enthalpy–entropy compensation [78]. We
employed this type of monotonicity to interpret the
compensation effect observed for the studied reactions
ꢀ
<
ꢀ
S + ꢀS , where ꢀH , ꢀH > 0 and ꢀS , ꢀS
2
2
2
0 [63–65]. We believe that T is relatively stable, so
0
1
0
1
H and ꢀS are insignificantly dependent of the sub-
stituents, and, hence, the reaction constants δꢀH and
δꢀS obtained for reaction series of 1–3 with arylox-
ides with an EDS are a good approximation to δꢀH
ꢀ=
ꢀ=
ꢀ
2
ꢀ=
and δꢀS .
2
ꢀ
ꢀ
As a rule, the reaction constants δꢀH and δꢀS
are formally divided into internal and external terms,
which are related to the reaction and the solvation,
respectively [63–65]. The external term influences
ꢀ=
markedly the value of δꢀH , and it determines the
ꢀ=
value of δꢀS for the reaction series of substituted
carboxylic esters with an anionic nucleophile in wa-
ter or solvent mixtures containing water [63–65]. It
International Journal of Chemical Kinetics DOI 10.1002/kin.20763

REACTIONS OF ARYL BENZOATES WITH POTASSIUM ARYLOXIDES IN DMF
277
ꢀ
ꢀ=
to unsubstituted nucleophile, the ratio δꢀH /δꢀS
ꢀ
ꢀ=
i
ꢀ=
are Tcomp. Then, ꢀH (X) = ꢀH + ꢀH0 + δꢀH
·
2
−
ꢀ
ꢀ=
i
ꢀ
−
σ (Х) and ꢀS (X) = ꢀS + ꢀS0 + δꢀS · σ (X),
2
ꢀ
ꢀ=
i
where ꢀH and ꢀS are the terms of the activation
i
parameters, which do not depend on the electrostatic
bonding. Obviously, van’t Hoff (Arrhenius) lines de-
ꢀ
scribing the kinetics of this reaction series, ꢀG (X) =
2
ꢀ
ꢀ=
ꢀ
H (X) − TꢀS (X) (logkX = logA(X) − Ea(X)/RT),
2
2
should intersect at the temperature Tcomp, and then,
the compensation and isokinetic effects can be
synonymous.
We believe that the TS motion along the reaction
coordinate in the direction of the spiro-σ-complex (a
normal Hammond effect) leads to increase the elec-
trostatic bonding between the reaction centers, and,
hence, increase the slope of the corresponding tan-
gent line segment. Then, the isokinetic temperature
Tiso as well as the compensation temperature Tcomp
can be interpreted in the terms of the transition state
position along the reaction coordinate: the larger the
corresponding value, the later the transition state.
It should be emphasized that the existence of the
isokinetic effect is unlikely in the case of the strong
substituent effect on the electrostatic bonding. How-
ever, the compensation effect can be operative because
a statistical criterion to judge its existence is less cor-
rect. As a rule, the transition state structure and, hence,
the reacting bond do not vary significantly with vari-
ation of the substituents in the nucleophile [2], and
such a reaction series is isokinetic. At the same time,
the Tiso values obtained for reaction series of 1–3 with
potassium aryloxides with an EDS are close, 209, 234,
and 212 K, respectively, and it means that the TS struc-
ture remains essentially constant despite changing the
leaving group substituent Y from 4-NO2 to 3-NO2 and
Figure 3 An enthalpy–entropy diagram for the reacting
bond. The general form of the curves a–c is defined by
the monotonic dependence of the enthalpic benefit on the
entropic cost in a simple exothermic bimolecular associa-
tion [78]. The upper limit determines a covalent bond. Series
of tangent line segments approximate the curve a. The elec-
trostatic bonding in the point (ꢀH0, ꢀS0) is the same as in TS.
Three points on this tangent line segment refer to variation
of the entering group substituents X1, X2, and X3, and the
coordinates of these points can be obtained from the follow-
ꢀ
−
ing equations: ꢀH(X) = ꢀH0 + δꢀH ·σ (Х) and ꢀS(X)
ꢀ
=
−
=
ꢀS0 + δꢀS ·σ (Х), where ꢀH0 and ꢀS0 are related
ꢀ
ꢀ=
to the unsubstituted nucleophile, the ratio δꢀH /δꢀS are
Tcomp. The curvature of the curve depends on the tightness
of the TS structure; the curves b and c refer to tighter and
looser structures of TS, respectively.
(Fig. 3). According to the general form of the curve
a, every point on it has a different slope. The slope
changes all along the curve a, and so it can be taken
as a measure of the electrostatic bonding. The original
curve a can be approximated by series of tangent line
segments. It is very important that the slope of a point
on the curve a is the slope of the tangent line segment
at this point. One of these segments is tangent at the
point in which the electrostatic bonding is the same
as in TS. As shown in Fig. 3, ꢀH0 and ꢀS0 are the
coordinates of the corresponding point.
ꢀ
ꢀ=
4-Cl (Table IV). The δꢀH and δꢀS values for
the corresponding series range from 29.5 to 21 and
–
17 kJ/mol and from 140.5 to 88 and –82 J/(mol·K),
respectively, i.e., they strongly decrease with reducing
the electron-withdrawing ability of the substituent Y.
ꢀ
ꢀ=
The positive δꢀH and δꢀS values obtained for
the reactions of 1 and 2 with potassium aryloxides
with an EDS are typical of the nucleophilic substitu-
tion reactions in which the entering group substituent
is varied. For the reactions of 2-chloro-3-nitro- and
In line with this model, the existence of the excellent
compensation effect for reaction series implies that a
variation of TS with the substituents X1, X2, and X3
can be well approximated by a tangent line segment at
the point (ꢀH0, ꢀS0) whose slope is equal to Tcomp.
For these substituents, the terms of the activation pa-
rameters referring to the electrostatic bonding are lin-
2-chloro-5-nitro-pyridines with aryloxides in MeOH
proceeded through the formation of a Meisenheimer
complex the reaction constants are also positive [79].
ꢀ
ꢀ=
On the other hand, the negative δꢀH and δꢀS
values for the corresponding reaction series of 3 can at-
tribute only to the effect of the substituent X on the elec-
trophilic center. These results are in a good agreement
with the TS structure in which the entering group sub-
stituent X oppositely influences the O- and C-reaction
−
early dependent on the σ (X) substituent constants,
ꢀ
−
i.e., ꢀH(X) = ꢀH0 +δꢀH ·σ (Х) and ꢀS(X) =
ꢀ=
−
ꢀ
S0 + δꢀS ·σ (Х), where ꢀH0 and ꢀS0 are related
International Journal of Chemical Kinetics DOI 10.1002/kin.20763

2
78
KHALFINA AND VLASOV
centers. Good linear correlations (r ≥ 0.999) have been
decrease. However, the Tiso value has been found to sig-
nificantly increase when the substrate changes from 4
to 1–3. It means that the TS motion can be discussed
in terms of two vectors, one along the reaction coordi-
nate and one perpendicular to it. Since the zwitterionic
intermediate shown in Fig. 4a can be considered as an
acylium cation, so the upper left corner should become
much less stable than the lower right corner with the
EWS in the leaving group. Such a destabilization of
the zwitterionic intermediate should cause the TS mo-
tion perpendicular to the reaction coordinate toward
the lower right corner (an anti-Hammond effect [83])
increasing the tightness of the TS structure [84]. We
believe that the tightness of the TS structure determines
the curvature of the dependence ꢀH – ꢀS for the react-
ing bond just as the strength of the electrostatic bonding
determines the slope of the tangent line segment: the
tighter the TS, the larger the curvature (curves a-b in
Fig. 3). The fact that the Tiso value (i.e., the slope of the
tangent line segment) increases despite decreasing the
CAr–O bond order means that in this case the change in
the tightness of the TS structure is dominant (Fig. 3).
It is important to emphasize that the Tiso value
changes depending on normal and anti-Hammond ef-
fects, and hence, incorrect use of the isokinetic tem-
perature Tiso to interpret the reaction mechanism can
lead to false conclusions.
ꢀ
ꢀ
0
obtained in the plots δꢀH and δꢀS versus σ (Y)
Table IV). The negative intercepts of the correspond-
(
ing plots, −39.9 kJ/mol and −190.8 J/(mol·K), indicate
that the substituent sensitivity of the C-reaction center
is larger than that of the O-reaction center, and it is
in line with the RSP. On the other hand, the positive
slopes of the plots, 85.1 kJ/mol and 399.4 J/(mol·K),
mean that an EWS in the leaving group increasing the
positive charge on the C-reaction center decreases its
ꢀ=
substituent sensitivity. It is interesting that the δꢀH
ꢀ=
0
and δꢀS values are close to zero when σ (Y) ∼ 0.47.
The Tiso value for reactions of 5 with potassium
aryloxides with an EDS has been determined to be 345
K. One can suggest that TS becomes later as the sub-
stituent Y changes the position in the leaving group
from para (or meta) to ortho. At the same time, the
ꢀ
ꢀ=
negative δꢀH and δꢀS values, –21 kJ/mol and –
6
1.5 J/(mol·K), obtained for the reaction series of 5
with potassium aryloxides with an EDS indicate the
small positive charge on the C-reaction center, i.e., the
decreased CAr–O bond order. The δꢀH and δꢀS
ꢀ
ꢀ=
values, –6.0 kJ/mol and –63.0 J/(mol ·K), obtained for
the reaction series of the unsubstituted substrate (4)
are greater than those for the corresponding reactions
of the substrate with a weak EWS (3). and it sup-
ports the increased CAr–O bond order. The lower Tiso
value, 99 K, obtained for the reactions of 4 with potas-
sium aryloxides with an EDS is also in line with this
assumption.
To explain such changes in the TS structure, the
one-dimensional reaction coordinate is not sufficient.
Therefore, we employed two-dimensional reaction co-
ordinate diagram, also known as the More O’Ferrall–
Jencks diagram [80,81]. In two-dimensional reaction
coordinate diagram shown in Fig. 4a, the reactants are
depicted in the lower left corner, the spiro-σ-complex
in the upper right corner, the symmetric tetrahedral
intermediate in the lower right corner, and the zwitte-
rionic intermediate in the upper left corner. The CAr–O
bond formation proceeds along the y axis and the C–O
bond formation proceeds along the x axis. The reaction
pathway representing a stepwise process traverses near
the edges of the More O’Ferrall–Jencks diagram. The
transition state TS for the rate-determining intramolec-
ular cyclization of the intermediate T is located at a
saddle point on the right-hand limb of pathway.
Using the corresponding approach for the reaction
series of 5 with aryloxides with an EDS, we can con-
clude that if the nitro group is added at the ortho po-
sition in the leaving group, the transition state TS will
move along and perpendicular to the reaction coor-
dinate toward the symmetric tetrahedral intermediate
(Fig. 4).
We believe that variation of the entering group sub-
stituent X also affects the TS structure. Since an EDS
in the entering group destabilizes the spiro-σ-complex
and stabilizes the symmetric tetrahedral intermediate,
but an EWS stabilizes the first and destabilizes the lat-
ter, so the lower right corner becomes less stable than
the upper right corner when the substituent X changes
from EDS to EWS. This effect move TS along the
reaction coordinate toward lower right corner, i.e., TS
becomes earlier (Fig. 4a). In contrast to variation of the
leaving group substituent Y, one of the entering group
substituent X has a little effect the upper left corner,
and hence, the TS motion along the reaction coordi-
nate is dominant. In line with this conclusion, the Tiso
value decreases as the substituent X changes from EDS
to EWS, from 209 to 70 K for 1, from 234 to 114 K
for 2, and from 345 to 215 K for 5 (Table IV). The
We believe that the EWS in the leaving group stabi-
lizes the spiro-σ-complex and destabilizes the symmet-
ric tetrahedral intermediate [82], so TS moves along the
reaction coordinate toward the lower right corner de-
creasing the CAr–O bond order (a normal Hammond
effect). One can suggest that if the TS motion along the
reaction coordinate were dominant the Tiso value would
ꢀ
ꢀ=
negative δꢀH and δꢀS values for reactions of 1, 2
with potassium aryloxides with an EWS, –5.8 kJ/mol
and –83.8 J/(mol·K) for 1 and –15.5 kJ/mol and
International Journal of Chemical Kinetics DOI 10.1002/kin.20763

REACTIONS OF ARYL BENZOATES WITH POTASSIUM ARYLOXIDES IN DMF
279
Figure 4 More O’Ferrall–Jencks diagrams for the four-step mechanism of the reactions of aryl benzoates with potassium
aryloxides in DMF: the left diagram (a) for the stepwise spiro-σ-complex formation; the right diagram (b) for the stepwise
spiro-σ-complex breakdown. In the left diagram, the positions of TS for the reactions of 1–5 with potassium aryloxides with
da ni s cE uDs sS i oa nn .d) 1–3, 5 with potassium aryloxides with an EWS are represented by points (•) and (o), respectively. (See the text for
–
132.9 J/(mol·K) for 2, support a decrease in the CAr–O
substituent Y changes from 4-NO2 to 3-NO2. How-
ever, the magnitude of the meta effect is much smaller
than that of the ortho effect. The weaker meta effect
can be observed for the reaction of aryl benzoates with
potassium aryloxides with an EDS, i.e., one becomes
apparent for the earlier TS.
Thus, for all isokinetic series of the reactions of
aryl benzoates with potassium aryloxides in DMF, the
RDS is the spiro-σ-complex formation. It is evident
that the mechanism of the spiro-σ-complex break-
down formally refers to a [2+2] retro-cycloaddition,
and hence it should proceed through a zwitterionic in-
termediate [89]. Such an intermediate will be formed
only if the C–O bond in the 1,3-dioxetane cycle breaks.
The departure of the leaving group from this zwitteri-
onic intermediate leads to products.
As shown in Fig. 4b, in two-dimensional reac-
tion coordinate diagram of the spiro-σ-complex break-
down, the products are depicted in the lower right cor-
ner, the spiro-σ-complex in the upper left, and the
symmetric tetrahedral intermediate in the lower left. In
the upper right corner of this system, there is the zwit-
terionic intermediate. The reaction pathway traversing
near the edges of the More O’Ferrall–Jencks diagram
represents a stepwise process involving the zwitteri-
onic intermediate.
ꢀ
ꢀ=
bond order. The large negative δꢀH and δꢀS val-
ues for reactions of 5 with potassium aryloxides with
an EWS, –36.9 kJ/mol and –185.1 J/(mol·K), also in-
dicate the earlier TS (Table IV). A direct resonance
interaction of the substituent Y with the C-reaction
center in TS for the reactions of 1–3 with potassium
aryloxides with an EWS implies that a small negative
charge is delocalized in the leaving group. The small
ρ(Y) and ρXY values for the corresponding reactions
series support this conclusion.
It is interesting to note that the ρ(Y) and ρXY val-
ues for these reactions are similar to that for the re-
actions of aminolysis and pyridinolysis of aryl esters
and their thiol, thiono, and dithio derivatives in MeCN
proceeding through the rate-limiting breakdown of a
zwitterionic tetrahedral intermediate [85–88].
ꢀ
ꢀ
−
Thus, broken ꢀꢀH and ꢀꢀS versus σ (X) plots
for the substrates 1, 2, 5 can be interpreted in the terms
of change in the TS position along the reaction co-
ordinate. In the protic solvent, it refers to changes
in the solvation with the electronic effect of the sub-
stituents [63–65].
Employing the proposed approach for the reactions
of 1 and 5 with potassium aryloxides with an EWS, one
can conclude that TS moves along and perpendicular
to the reaction coordinate toward the symmetric tetra-
hedral intermediate as the leaving group substituent Y
changes from 4-NO2 to 2,4-(NO2)2 (Table IV). Sur-
prisingly, the same conclusion can be drawn when the
Therefore, the mechanism proposed for the reac-
tions of aryl benzoates with potassium aryloxides in
DMF includes four steps. We believe that the delo-
calization of the negative charge in the leaving group
International Journal of Chemical Kinetics DOI 10.1002/kin.20763

2
80
KHALFINA AND VLASOV
stabilizes TS compared to the transition states of the
formation and breakdown of the symmetric tetrahedral
intermediate, respectively, and so, the four-step mech-
anism is preferred. However, in a protic solvent the
stabilizing solvation of TS should be smaller than that
of the corresponding transition states of the formation
and breakdown of the symmetric tetrahedral interme-
diate, and then, the AN + DN mechanism can compete
with the four-step mechanism.
dicated that the reactions of aryl benzoates with potas-
sium aryloxides in DMF proceed via the four-step
mechanism involving (1) the nucleophilic attack of ary-
loxide at the carbonyl carbon generating the tetrahedral
intermediate T, (2) the intramolecular nucleophilic at-
tack of the alkyloxide at the ipso position in the leaving
group leading to formation of the spiro-σ-complex, (3)
the spiro ring opening with yielding the zwitterionic in-
termediate, and (4) the departure of the leaving group.
The second step is the RDS, i.e., the reactions of aryl
benzoates with potassium aryloxides in DMF, refer to
a nucleophilic aromatic substitution. The origin of the
breaks in the Hammett plots can be explained by the
TS motion along the reaction coordinate according to a
Hammond effect. The reaction mechanism can change
from four steps to AN + DN when the solvent changes
from DMF to protic ones.
Um et al. have suggested that the reactions of 1
with aryloxides in absolute ethanol and DMSO–H2O
◦
mixture at 25 С proceed through the stepwise mech-
anism with the formation of the symmetric tetrahe-
dral intermediate [55,57]. Comparison of the rate con-
stants shows that the reactions of 1 with potassium
aryloxides in DMF are much faster than that in ab-
solute ethanol or DMSO–H2O mixture. For example,
the rate constant k4-CN for the reaction of 1 with potas-
sium 4-cyanophenoxide in DMF was determined to be
−
1
−1
◦
1.19 M
s
at 25 С and it is ∼700 times faster than
that in the absolute ethanol [55]. We can conclude that
the reactants and tetrahedral intermediate are strongly
stabilized as a solvent changes from DMF (aprotic) to
absolute ethanol or DMSO–H2O mixture (protic), so
the reaction mechanism changes and the rate constants
decrease.
BIBLIOGRAPHY
1. Bourne, N.; Chrystiuk, E.; Davis, A. M.; Williams, A. J
Am Chem Soc 1988, 110, 1890–1895.
2
. Buncel, E.; Um, I. H.; Hoz, S. J Am Chem Soc 1989,
11, 971–975.
. Pregal, M.; Dunn, E. J.; Buncel, E. J Am Chem Soc
991, 113, 3545–3550.
1
3
It should be emphasized that the proposed mech-
anism is realized provided that the pKa values of the
entering aryloxides are larger than that of the leaving
aryloxides. In line with the principle of the microscopic
reversibility [90,91], the mechanism of the reverse re-
actions also includes four steps: (1) a nucleophilic at-
tack of the carbonyl oxygen of aryl benzoates at the
ipso position in potassium aryloxide, followed by (2)
the intramolecular cyclization leading to formation of
the spiro-σ-complex, (3) the spiro ring opening with
yielding the tetrahedral intermediate T and then, (4)
the departure of the leaving group. One can suggest
that the reactions of 3 with potassium aryloxides with
an EWS proceed according to this mechanism. Since
good linear correlations (r ≥ 0.999) have been obtained
1
4. Hengge, A. J Am Chem Soc 1992, 114, 6575.
5. Williams, A. Adv Phys Org Chem 1992, 27, 1–55.
6. Stefanidis, D.; Cho, S.; Dhe-Paganon, S.; Jencks, W. P.
J Am Chem Soc 1993, 115, 1650–1656.
7. Hengge, A.; Hess, R. A. J Am Chem Soc 1994, 116,
11256–11257.
8
. Hengge, A.; Eadens, W. A.; Elsing, H. J Am Chem Soc
994, 116, 5045–5049.
. Williams, A. Chem Soc Rev 1994, 23, 93–100.
1
9
1
1
1
0. Tarkka, R. M.; Buncel, E. J Am Chem Soc 1995, 117,
503–1507.
1. Colthurst, M. J.; Williams, A. J Chem Soc, Perkin Trans
1997, 1493–1497.
2. Fernandez, M. A.; Rossi, R. H. J Org Chem 1999, 64,
000–6004.
1
2
6
−
in the plots log(kX) versus σ (Y) for the reactions of
13. Williams, A. Concerted Organic and Bio-organic Mech-
anisms; CRC Press: Boca Raton, FL, 2000; Ch. 4, pp.
43–46.
1–3 with 3-bromo- and 4-bromophenoxides, so for the
corresponding reactions of 3 the third step is the RDS.
1
1
1
4. Lee, H. W.; Yun, Y. S.; Lee, B. S.; Koh, H. J.; Lee, I. J
Chem Soc, Perkin Trans 2 2000, 2302–2305.
5. Andres, G. O.; Granados, A. M.; Rossi, R. H. J Org
Chem 2001, 66, 7653–7657.
6. Um, I. H.; Han, H. J.; Ahn, J. A.; Kang, S.; Buncel, E. J
Org Chem 2002, 67, 8475–8480.
CONCLUSIONS
We have shown in the present study that the approach
based on analysis of the effect of the entering group
substituent on the activation parameters gives signifi-
cant information regarding the transition state structure
of the RDS and its position on the two-dimensional
reaction coordinate diagram. Our result in general in-
1
1
7. Hengge, A. Acc Chem Res 2002, 35, 105–112.
8. Castro, E. A.; Pavez, P.; Santos, J. G. J Org Chem 2001,
66, 3129–3132.
19. Castro, E. A.; Pavez, P.; Santos, J. G. J Org Chem 2003,
68, 3640–3545.
International Journal of Chemical Kinetics DOI 10.1002/kin.20763

REACTIONS OF ARYL BENZOATES WITH POTASSIUM ARYLOXIDES IN DMF
281
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
0. Castro, E. A.; Arellano, D.; Pavez, P.; Santos, J. G. J Org
Chem 2003, 68, 9034–9039.
1. Castro, E. A.; Andujar, M.; Toro, A.; Santos, J. G. J Org
Chem 2003, 68, 3608–3613.
48. Khalfina, I. A.; Vlasov, V. M. Russ J Org Chem 2008,
44, 1619–1626.
49. Khalfina, I. A.; Vlasov, V. M. Russ J Org Chem 2011,
47, 845–854.
50. Vogel, A. T. Practical Organic Chemistry; Longman’s
Green: London, 1962.
51. Kornblum, N.; Lurie, A. P. J Am Chem Soc 1959, 81,
2705–2715.
52. Seo, J. A.; Kim, S. I.; Hong, Y. J.; Um, I. H. Bull Korean
Chem Soc 2010, 31, 303–308.
53. Hansch, C.; Leo, A.; Taft, R. W. Chem Rev 1991, 91,
165–195.
54. Kepka, C.; Penhoat, M.; Barbry, D.; Rolando, Ch.
In 13th International Electronic Conference on Syn-
thetic Organic Chemistry (ECSOC-13), November 1–
30, 2009; a039:1–11.
55. Um, I. H.; Oh, S. J.; Kwon D. S. Bull Korean Chem Soc
1996, 17, 802–807.
2. Castro, E. A.; Aguayo, R.; Santos, J. G. J Org Chem
2
003, 68, 8157–8161.
3. Lee, I.; Lee, H. W.; Yu, Y. K. Bull Korean Chem Soc
003, 24, 993–998.
4. Castro, E. A.; Aliaga, M.; Santos, J. G. J Org Chem
004, 69, 6711–6714.
5. Oh, H. K.; Park, J. E.; Sung, D. D.; Lee, I. J Org Chem
004, 69, 9285–9288.
6. Oh, H. K.; Park, J. E.; Sung, D. D.; Lee, I. J Org Chem
004, 69, 3150–3153.
7. Oh, H. K.; Ha, J. S.; Sung, D. D.; Lee, I. J Org Chem
004, 69, 8219–8223.
8. Um, I.-H.; Lee, J. Y.; Kim, H. T.; Bae, S. K. J Org Chem
004, 69, 2436–2441.
2
2
2
2
2
2
9. Um, I. H.; Kim, K. H.; Park, H. R.; Fujio, M.; Tsuno, Y.
J Org Chem 2004, 69, 3937–3942.
0. Um, I.-H.; Lee, J. Y.; Lee, H. W.; Nagano, Y.; Fujio, M.;
Tsuno, Y. J Org Chem 2005, 70, 4980–4987.
56. Pross, A. Adv Phys Org Chem 1977, 14, 69–126.
57. Kwon, D. S.; Park, J. Y.; Um, I. H. Bull Korean Chem
Soc 1994, 15, 860–864.
58. Lee, I.; Lee, B. S.; Koh, H. J.; Chang, B. D. Bull Korean
Chem Soc 1995, 16, 277–281.
1. Um, I.-H.; Hong, J. Y.; Seok, J. A. J Org Chem 2005,
7
0, 1438–1444.
2. Castro, E. A.; Gazitua, M.; Santos, J. G. J Org Chem
005, 70, 8088–8092.
59. Haake, P. J Am Chem Soc 1973, 95, 8088–8096.
60. Lee, I. Chem Soc Rev 1990, 19, 317–333.
61. Dubois, J.-E.; Ruasse, M.-F.; Argile, A. J Am Chem Soc
1984, 106, 4840–4845.
2
3. Castro, E. A.; Aguayo, R.; Bessolo, J.; Santos, J. G. J
Org Chem 2005, 70, 7788–7791.
4. Castro, E. A.; Aguayo, R.; Bessolo, J.; Santos, J. G. J
Org Chem 2005, 70, 3530–3536.
¨
62. Ruff, F.; Farkas, O. J Org Chem 2006, 71, 3409–3416.
63. Ruff, F. Int Electron J Mol Des 2004, 3, 474–498.
64. Ruff, F. J Mol Struct (Theochem) 2003, 625, 111–120.
65. Ruff, F. J Mol Struct (Theochem) 2002, 617, 31–45.
66. Liu, L.; Guo, Q. X. Chem Rev 2001, 101, 673–695.
67. Linert, W.; Jameson, R. F. Chem Soc Rev 1989, 18,
477–505.
68. Linert, W. Chem Soc Rev 1994, 23, 429–438.
69. Exner, O. Chem Commun 2000, 1655–1656.
70. Hepler, L. G. Can J Chem 1971, 49, 2803–2807.
71. Exner, O. Collect Czech Chem Commun 1964, 29,
1094–1113.
72. Pal’m, V. A. Osnovy Kolichestvennoi Teorii Organich-
eskikh Reaktsii (Fundamentals of Quantitative Theory
of Organic Reactions), 2nd ed.; Khimiya: Leningrad,
Russia, 1977.
73. Leffler, J. E.; Grunwald, E. Rate and Equilibria of Or-
ganic Reactions as Treated by Statistical, Thermody-
namic, and Extrathermodynamic Methods; Wiley: New
York, 1963.
74. Lee, I. Adv Phys Org Chem 1992, 27, 57–117.
75. Kim, Y. S.; Choi, H.; Yang, K.; Park, J. K.; Koo, I. S.
Bull Korean Chem Soc 2010, 31, 3279–3282.
76. Carniti, P.; Beltrame, P.; Cabiddu, S. J Chem Soc, Perkin
Trans 2 1973, 1430–1433.
77. Sung, R. Y.; Choi, H.; Lee, J. P.; Park, J. K.; Yang, K.;
Koo, I. S. Bull Korean Chem Soc 2009, 30, 1579–1582.
78. Williams, D. H.; Westwell, M. S. Chem Soc Rev 1998,
27, 57–63.
5. Um, I.-H.; Lee, J. Y.; Bae, S. Y.; Buncel, E. Can J Chem
2005, 83, 1365–1371.
6. Um, I.-H.; Kim, E. Y.; Park, H.-R.; Jeon, S.-E. J Org
Chem 2006, 71, 2302–2306.
7. Um, I.-H.; Hwang, S. J.; Baek, M. H.; Park, H.-R. J Org
Chem 2006, 71, 9191–9197.
8. Um, I.-H.; Lee, J. Y.; Ko, S. H.; Bae, S. K. J Org Chem
2006, 71, 5800–5803.
9. Castro, E. A.; Echevarria, G. R.; Opazo, A.; Robert, P.;
Santos, J. G. J Phys Org Chem 2006, 19, 129–135.
0. Um, I.-H.; Lee, J. Y.; Lee, H.W; Fujio, M.; Tsuno, Y.
Org Biomol Chem 2006, 4, 2979–2985.
1. Um, I.-H.; Hwang, S. J.; Buncel, E. J Org Chem 2006,
71, 915–920.
2. Um, I.-H.; Park, Y.-M.; Fujio, M.; Mishima, M.; Tsuno,
Y. J Org Chem 2007, 72, 4816–4821.
3. Castro, E. A.; Ramos, M.; Santos, J. G. J Org Chem
2009, 74, 6374–6377.
4. Castro, E. A.; Aliaga, M.; Campod o´ nico, P. R.; Cepeda,
M.; Contreras, R.; Santos, J. G. J Org Chem 2009, 74,
9173–9179.
4
5. Castro, E. A.; Milan, D.; Aguayo, R.; Campod o´ nico,
P. R.; Santos, J. G. Int J Chem Kinet 2011, 44, 687–
6
93.
6. O s´ kina, I. A.; Vlasov, V. M. Org Chem 2009, 45, 523–
27.
7. O s´ kina, I. A.; Vlasov, V. M. Russ J Org Chem 2010, 46,
26–330.
4
4
5
79. El-Bardan, A. A. J Phys Org Chem 1999, 12, 347–353.
80. More O’Ferrall, R. A. J Chem Soc B 1970, 274–277.
3
International Journal of Chemical Kinetics DOI 10.1002/kin.20763

2
82
KHALFINA AND VLASOV
8
8
1. Jencks, W. P. Chem Rev 1972, 72, 705–718.
2. Castro, E. A.; Leandro, L.; Millan, P.; Santos, J. G. J
Org Chem 1999, 64, 1953–1957.
87. Oh, H. K.; Kim, S. K.; Lee, H. W.; Lee, I. J Chem Soc,
Perkin Trans 2 2001, 1753–1757.
88. Oh, H. K.; Kim, S. K.; Cho, I. H.; Lee, H. W.;
Lee, I. J Chem Soc, Perkin Trans 2 2000, 2306–
2310.
89. Huisgen, R. Pure Appl Chem 1980, 52, 2283–2302.
90. Hine, J. Physical Organic Chemistry, 2nd ed.;
McGraw-Hill: New York, 1962; p. 69.
91. Gould, E. S. Mechanism and Structure in Organic Chem-
istry; Holt, Rinehart and Winston: New York, 1959;
p. 319.
8
8
8
8
3. Buncel, E.; Wilson, H.; Chuaqui, C. J Am Chem Soc
1
982, 104, 4896–4900.
4. Arnett, E. M.; Reich, R. J Am Chem Soc 1980, 102,
892–5902.
5. Koh, H. J.; Kang, S. J.; Kim Ch. J.; Lee, H. W.; Lee, I.
Bull Korean Chem Soc 2003, 24, 925–930.
5
6. Oh, H. K.; Lee, j. Y.; Lee, H. W.; Lee, I. New J Chem
2002, 26, 473–476.
International Journal of Chemical Kinetics DOI 10.1002/kin.20763