Detailing the elementary stages in the oxirane ring opening reactions with carboxylic acids catalyzed by tertiary amines

Article abstract of DOI:10.1002/poc.4071

The study of the reaction systems “R₃N – proton-donor reagent (NuH) – epichlorohydrin” using a complex of various kinetic methods of investigation allowed the reaction pathway and the catalysis mechanism of the oxirane ring opening by proton-donor reagents in the presence of R₃N to be detailed. It was confirmed that the catalytic activity of R₃N consists in their quaternization by oxirane under the obligatory condition of its electrophilic assistance with the NuH reagent. The quaternization process of tertiary amines was studied using ¹H NMR and UV spectroscopy, quantum chemical modeling, and correlation analysis. It was shown that the transition states of the quaternization stage are dissociative, and their degree of looseness depends on the electrophilic activation of the oxirane ring.

Full text of DOI:10.1002/poc.4071

Received: 27 January 2020
DOI: 10.1002/poc.4071
Revised: 25 March 2020
Accepted: 27 March 2020
R E S E A R C H A R T I C L E
Detailing the elementary stages in the oxirane ring opening
reactions with carboxylic acids catalyzed by tertiary amines
1
2
3
4
Stanislav Bakhtin
| Elena Shved | Yuliia Bespalko | Tatyana Tyurina
|
5
Vitalii Palchykov
1
Faculty of Chemistry, Donetsk National
University, Donetsk, Ukraine
Abstract
2
The study of the reaction systems “R N – proton-donor reagent (NuH) – epi-
Faculty of Chemistry, Biology and
3
Biotechnology, Vasyl' Stus Donetsk
chlorohydrin” using a complex of various kinetic methods of investigation
National University, Vinnytsia, Ukraine
allowed the reaction pathway and the catalysis mechanism of the oxirane ring
3
Unit of Molecular Imaging and
opening by proton-donor reagents in the presence of R N to be detailed. It was
3
Photonics, Catholic University of Leuven,
Leuven, Belgium
confirmed that the catalytic activity of R N consists in their quaternization by
3
4
Department of Radical Reaction
oxirane under the obligatory condition of its electrophilic assistance with the
Research, L. M. Litvinenko Institute of
Physical Organic and Coal Chemistry,
Donetsk, Ukraine
NuH reagent. The quaternization process of tertiary amines was studied using
1
H NMR and UV spectroscopy, quantum chemical modeling, and correlation
5
analysis. It was shown that the transition states of the quaternization stage are
dissociative, and their degree of looseness depends on the electrophilic activa-
tion of the oxirane ring.
Institute of Chemistry and Geology, Oles
Honchar Dnipro National University,
Dnipro, Ukraine
Correspondence
K E Y W O R D S
Stanislav Bakhtin, Faculty of Chemistry,
Donetsk National University, Donetsk,
Ukraine.
nucleophilic catalysis, oxirane, reaction mechanism, transition state
Email: stanislav.g.bakhtin@gmail.com
1
| INTRODUCTION
using the ECH opening reactions in the preparation of
known biologically active compounds.
[1–4]
Oxiranes (epoxides) are attractive building synthons for the
preparation of a variety of organic compounds with different
complexity. The high activity of oxiranes is explained by the
increased strain of the three-membered ring, as well as by
the presence of reactive electrophilic centers–carbon atoms
of the С–О–С fragment. The main reactions of this class
of compounds usually occur under the action of various
nucleophilic proton-donor reagents (NuH) and allow one to
obtain various homo- and heterofunctional derivatives,
many of which are widely used in industry and everyday
life. Among oxiranes, epichlorohydrin (ECH) is one of the
most commercially important epoxides because its reactions
with NuH reagents make it possible to both build up the
carbon chain and introduce functional substituents on three
neighboring C-atoms. Scheme 1 illustrates some examples of
Therefore, a detailed study of such processes opens, on
the one hand, ways to expanding the scope of the practical
application of epoxy compounds, and on the other hand, it
provides additional information on the regularities and mech-
anism of the oxyalkylation reaction that underlies them.
Noncatalytic oxyalkylation reactions of NuH reagents
are known to have extremely low rates and are acceler-
ated in the presence of acidic or basic catalysts. Com-
+δ
+δ
pounds such as tertiary amines (R N), pyridines, and
3
tetraalkylammonium salts are often used as the latter.
Moreover, the catalytic effect of R N is explained in a
3
number of cases within the framework of the mechanism
[
5–17]
of general base catalysis,
when the amine acting as
the base activates the proton-donor reagent because of
the transfer of its proton (Equation 1):
J Phys Org Chem. 2020;e4071.
wileyonlinelibrary.com/journal/poc
© 2020 John Wiley & Sons, Ltd.
1 of 14
https://doi.org/10.1002/poc.4071

2
of 14
BAKHTIN ET AL.
SCHEME 1 The use of the
hydroxyalkylation reactions of proton donor
nucleophiles
+
δ
-δ
+
-
R N + NuH Ð R N ꢀꢀꢀH N u Ð ½R NHꢁ Nu :
ð1Þ
system “C H NMe (N,N-dimethylaniline, DMA) –
3
3
3
6
5
2
AcOH (NuH) – ECH,” a gradual irreversible consump-
[17–27]
35
In other cases,
the catalytic effect of amines is
tion of C H NMe
was revealed, which can corre-
6
5
2
explained within the framework of the nucleophilic
mechanism, where the amine exhibiting the properties of
a nucleophile interacts with a electrophilic substrate
spond only to its participation in the quaternization
reaction with ECH to form [R NCH CH(OH)
3
2
+
–
CH Cl] Nu providing the generation of the reactive
2
[28–32]
–
(oxirane)
in the trimolecular process forming an
anion of the proton-donor reagent (Nu ). Therefore,
intermediate–tetraalkylammonium salt (Equation 2):
further interest is a deeper study of the quaternization
–
The anion of proton-donor reagent (Nu ) formed
stage of R N, which explains the formation of a true
3
through the Equations 1 or 2 then interacts with the
oxirane to form products of ECH opening
catalyst in the system—the tetraalkylammonium
salt, which transfers the anion of the proton-donor
reagent.
(
NuCH CH(OH)CH Cl, Scheme 1, Equation 3):
2 2
But we adhere to the mechanism of transferring
the anion of a nucleophilic reagent by an ion pair,
In this regard, the reaction series “ArNMe2 –
33
R'COOH – ECH”, in which the nature of both the
amine and proton-donor reagent was varied, were cho-
sen as objects of the study. To solve this problem, vari-
ous methods of investigation were used: kinetic
(potentiometric acid-base titration of a proton-donor
reagent; UV spectroscopic determination of amine con-
sumption), structural (UV and NMR monitoring of the
initial reagents consumption), calculcations (the
when the quaternary ammonium salt [R NCH CH(OH)
3
2
+
–
CH Cl] Nu is formed not in an elementary stage but
2
in a complex multistage process. Confirmation of this
concept is a number of experimental facts the key of
which are (1) the decisive role of the nucleophilicity of
33
R N rather than their basicity ; (2) irreversible con-
3
[34,35]
36
sumption of R X (X = N,
P ). Thus, in the
3

BAKHTIN ET AL.
3 of 14
quantum chemical modeling of the behavior of system
components – FIREFLY 7.1.G package), and correlation
analysis (establishing the relationship between changes
in the structures of the starting reagents and the
corresponding transition states [TSs]).
2.2 | Kinetic measurements
2.2.1 | Kinetic measurement procedure I
(consumption of proton-donor reagent
R'COOH)
The determination of the rate of the total reaction of
ECH and R'COOH (NuH, Scheme 1) with the subsequent
calculation of the observed and catalytic rate constants
was based on the determination of the decrease in the
concentration of the acid reagent (potentiometric acid-
base titration). The solution of R'COOH of the
corresponding concentration in 2 mL of ECH was intro-
duced into one of the two compartments of the kinetic
2
2
| EXPERIMENTAL SECTION
.1 | Synthesis and purification of
compounds
The commercially available ECH was dried during a
day over sodium hydroxide granules and then distilled
at atmospheric pressure collecting a fraction with a
boiling point of 115–116 C; n 1.438 (lit. b.p. 115.5 C;
n 1.438 ).
D
flask, and the solution of ArNMe in 1 mL of ECH was
2
added to the other. The flask was placed in the thermo-
stat at the required temperature. After thermostating (10
min), the solutions were quickly mixed, and the flask was
placed in the thermostat again. The initial time of the
reaction was taken to be the mixing time of the solutions.
After a necessary period of time, the reaction was stopped
by adding 10 mL of a mixture of isopropyl alcohol and
ꢂ
ꢂ
D
37
The commercially available glacial acetic acid con-
tains impurities of carbonyl compounds that were
removed by refluxing in the presence of 5% KMnO4
solution for 1 h after which the acid was distilled over
ꢂ
ꢂ
P O to remove the residual water; b.p. 117–118 С; n
water (1:1 by volume) cooled to 0–5 C with rapid stirring
2
5
D
ꢂ
37
1
.372 (lit. b.p. 118 С; nD 1.3715 ). Ethoxyacetic acid
(dilution and cooling method). The contents of the flask
were quantitatively transferred with distilled water to the
titration cell. The current concentration of the proton-
donor reagent was determined by potentiometric acid-
base titration with a 0.1 M NaOH solution.
was synthesized by the interaction of chloroacetic acid
with sodium ethylate ; Н NMR (400 MHz, CDCl ): δ
38 1
3
1
.25 (t, J = 9.1 Hz, 3H, CH ), 3.62 (q, J = 9.2 Hz, 2H,
3
CH ), 4.15 (s, 2H, CH ), 10.25 (s, 1H, СООН).
2
2
2
,2-Dimethylpropanoic (pivalic) acid was synthesized
according to the published methodology starting from
2
.2.2 | Kinetic measurement procedure
39 1
acetone ; Н NMR (CDCl ): δ 1.24 (s, 9Н, 3CH ), 11.6
3
3
II (consumption of ArNMe2)
(
s, 1Н, СООН).
In order to remove the impurities of primary and sec-
ondary amines from N,N-dimethylaniline (DMA), it was
The determination of the rate of the ArNMe quater-
2
nization stage and the subsequent calculation of the rate
constants for the formation of the tetraalkylammonium
salt were based on the determination of the concentra-
tion of the free base by UV spectroscopy. For this, at cer-
tain time intervals, the reaction mixture obtained
similarly to the procedure I, after cooling in the ice bath
(but without adding isopropyl alcohol) was diluted with
a 10-fold excess of ECH, an aliquot was taken into a qua-
rtz cuvette (l = 1 cm) and placed in an SF-2000 spectro-
photometer. The comparison solution was that of
carboxylic acid in ECH previously diluted with a 10-fold
37
refluxed for 5 h with excess of Ac O. In this case, pri-
2
mary and secondary amines are acetylated and lose their
basic character. Then most of the anhydride was distilled
off, and the residue was dissolved in concentrated
hydrochloric acid on cooling, and insoluble organic
impurities were removed by extraction with a small
amount of ether. To isolate DMA, treatment of its hydro-
chloride with a concentrated solution of NaOH was car-
ried out with further distillation of the amine under
ꢂ
reduced pressure (b.p. 105–107 C/20 mm); n 1.555 (lit.
D
37
nD 1.556 ). 4-Methoxy-N,N-dimethylaniline was obtai-
excess of ECH. The consumption rate of ArNMe was
2
ned by methylation of p-anisidine with dimethyl sul-
monitored by measuring the optical density of the reac-
tion mixture at the optimal wavelength based on previ-
ously constructed calibration curves.
40 1
fate ; Н NMR (CDCl ): δ 2.90 (s, 6Н, 2СН ), 3.78 (s, 3Н,
3
3
ОСН ), 6.86 (m, 4Н, C H ). 3-Nitro-N,N-dimethylaniline
3
6
4
was synthesized by methylation of m-nitroaniline with
2
.3 | Procedure of quantum chemical
41 1
methyl iodide in the presence of sodium hydride ; Н
NMR (CDCl ): δ 3.02 (s, 6Н, 2СН ), 6.94 (d, J = 7.9 Hz,
modeling
3
3
1
Н), 7.34 (t, J = 7.8 Hz, 1Н), 7.47 (s, 1Н), 7.55 (d, J = 7.8
Quantum chemical calculations were performed using
4
2
Hz, 1Н).
the PC FIREFLY 7.1.G software package.
The

4
of 14
BAKHTIN ET AL.
construction and visualization of the structures of
model objects were implemented using the
respect to the acid reagent (z) varies depending signifi-
cantly on the reaction conditions.
4
3
ChemCraft. The components of the system were cal-
culated in vacuum using the density functional
method (DFT) and the standard basis set 6–31 + G**.
DFT calculations were performed using the B3LYP
hybrid density functional. At the end of the optimiza-
tion of the geometry of the structures, the calculation
of rotational constants, normal vibrations frequencies,
and thermodynamic functions was performed. Belong-
ing the found TSs for the corresponding reaction was
proved by the IRC descents into the valleys of reagents
and products.
Based on the obtained experimental data (kinetic
measurement procedure I), it was found that the kinetic
curves for the consumption of acetic acid (CAcOH, М)
obtained at various initial concentrations of tertiary
amines are linear (r ≥ 0.99) in the coordinates CAcOH – t
(Figure
1
for the reaction in the presence of
3
| RESULTS AND DISCUSSION
The differential equation of the rate of the total reaction
Scheme 1) is described by the Equation 4:
(
dCAcOH
À
Á
y
z
r = −
= knon + kcatCx
C
C
ECH AcOH
ð4Þ
cat
dt
y
z
=
kobsC
C
,
ECH AcOH
where knon and kcat are the rate constants of non-
catalytic and catalytic reactions, respectively; kobs is the
observed reaction rate constant. The analysis of the lit-
erature shows that the order of reaction with respect to
the oxirane for most of these reaction series is constant
and equals to 1 (y = 1), while the order of reaction with
FIGURE 2 The dependence of the observed rate constants for
the reaction of ECH with AcOH on the concentration of
ꢂ
0
6 4 2
4-MeOC H NMe at 60 C; C AcOH = 0.2 M; r = 0.999
FIGURE 1 Kinetic curves for the reaction of acetic acid
FIGURE 3 UV spectra of the reaction mixture
0
ꢂ
ꢂ
(
C
AcOH = 0.2 М) with ECH at 60 C in the presence of
“4-MeOC
6
H
4
NMe
2
0
– AcOH – ECH” (70 C) at various time points;
0
–3
4
-MeOC NMe catalyst
6
H
4
2
C
AcOH = 0.3 M; C amine = 5.03ꢀ10
M

BAKHTIN ET AL.
5 of 14
4
-MeOC H NMe ). This fact corresponds to a zeroth
para-position with respect to the reaction center. Thus,
the revealed series of changes in kcat indicates the same
effect of both basicity and nucleophilicity on the catalytic
activity of amines and does not allow us to differentiate
which property of amines is decisive for explaining their
behavior in the hydroxyalkylation of ECH.
In this regard, for further study of the potential inter-
actions (acid-base, nucleophile-electrophile) that are pos-
sible in the reaction system and for detailing the
elementary stages and TSs in the reaction pathway, the
key aspect is the need to monitor the behavior of
R3N. One of the most informative methods for studying
interactions in complex multicomponent systems is NMR
spectroscopy. At the same time, for the case of catalytic
6
4
2
order of reaction with respect to AcOH.
Thus, basing on the established order of the reaction
with respect to AcOH (z = 0) and taking into account the
excess of the oxirane (pseudo-order for ECH, y = 1), the
observed rate constants were calculated using the Equa-
tion 5:
0
C
−CAcOH
AcOH
kobs
=
,
ð5Þ
0
C
t
ECH
0
where C
M); t is the reaction time.
is the initial concentration of oxirane (≈12.5
ECH
To establish the order with respect to the catalyst (ter-
tiary amine), a comparison of the observed rate constants
of the reaction of ECH with acetic acid and the
concentration of the catalyst in accordance with
Equation (6) was made:
reactions, when the amount of the catalyst (R N) is much
3
less than the concentration of reagents (reaction under
study: CECH > > CAcOH > > CR3N), it is impossible to
carry out the indicated monitoring of the amine behavior
by NMR.
Bypassing this limitation is possible using the UV
spectroscopy method for the systems containing aromatic
tertiary amines. The structures of substituted N,N-
kobs = knon + kcatCcat:
ð6Þ
The presence of a linear correlation (Figure 2 for
4
-MeOC H NMe as example) indicates the first order of
dimethylanilines (ArNMe ) have characteristic features:
6
4
2
2
reaction with respect to the catalyst (x = 1).
(1) conjugation of the lone pair of electrons on the N
atom with the benzene ring; (2) the disappearance of con-
In accordance with Equation 6, the catalytic rate con-
−
1 −1
stants (k , M s ) for amines were calculated: (1.02 ±
jugation during the reaction of ArNMe with various
cat
4
2
-
−4
+
+
+
0
.03)ꢀ10 – 4-MeOC H NMe ; (0.584 ± 0.032)ꢀ10
–
electrophiles Е (H —amine acts as the base, R —
amine acts as the nucleophile) because of the participa-
tion of this electrons pair in the formation of the covalent
N–E bond:
6
4
2
35
−4
C H NMe ; (0.104 ± 0.007)ꢀ10 – 3-O NC H NMe .
6
5
2
2
6
4
2
Moreover, the established series of catalytic activity of
amines, 4-MeOC H NMe (pKa = 5.86) > C H NMe
2
6
4
2
6
5
(
pK = 5.07) > 3-O NC H NMe (pK = 2.67), is consis-
The behavior of the nitrogen-containing nucleophile
was monitored by this method for amines C H NMe ,
4-MeOC H NMe , and 3-O NC H NMe . Indeed, in the
6 4 2 2 6 4 2
a
2
6
4
2
a
tent with that predicted from the basicity of amines
pK ); that is, electron-donating substituents in the ben-
6
5
2
(
a
zene ring increase the catalytic activity of ArNMe , and
UV spectra of the reaction system “ArNMe – R'COOH –
2
2
electron-withdrawing substituents decrease it.
ECH” (Figure 3) for all studied amines a gradual decrease
in the intensity of the long-wavelength absorption band
corresponding to forbidden transitions due to the conju-
It should be noted that the nucleophilicity of the stud-
ied tertiary amines determined jointly by the electronic
and steric effects at the N atom changes in the same
order as their basicity. This follows from the Hammett
equation when the change in steric effect can be
neglected if the substituents are varied in the meta- or
gation of the group: NMe with the benzene ring was
2
observed.
The established fact allows the reliable measurements
of the concentration of free nucleophile ArNMe2 in

6
of 14
BAKHTIN ET AL.
reaction systems to be conducted on the basis of the
intensity of the corresponding long-wavelength absorp-
tion band (Figure 3).
The validity of this assumption was proved by check-
ing the fulfillment of the Bouguer–Lambert–Beer law
(
Figure 4 for the system “4-MeOC H NMe – AcOH –
6
4
2
ECH”) at the starting time obtained at various initial con-
centrations of the tertiary amine.
The short-wavelength intense absorption (λ ≈ 250
nm, Figure 3), corresponding to the B-band due to the
allowed electronic transitions of the benzene ring, is not
suitable for monitoring the concentration of free amine.
This is due to the fact that both the starting compound
+
ArNMe and the product of its reaction with E contrib-
2
ute to this absorption.
Typical kinetic curves for ArNMe consumption in
2
the reaction systems with different initial concentrations
of nitrogen-containing nucleophile obtained by electronic
spectroscopy are presented in Figure 5.
FIGURE 5 Kinetic curves for 4-MeOC H NMe consumption
at different initial concentrations of amine for the quaternization
6
4
2
stage of the reaction mixture “4-MeOC
6
H
4
2
NMe – AcOH – ECH”
ꢂ
(
70 C)
The irreversible rather slow consumption of a free
amine can only correspond to its quaternization by the
oxirane molecule according to Equation 2, because the
process of tertiary amines protonation (Equation 1), as a
special case of the so-called barrier-free interactions, is a
logr0
0
= 0:988ꢀlogC
−1:65
quater
amine
ð8Þ
ðr = 0:998Þ:
2
3
−1 −144
fairly quick reaction (k = 10 – 10 M
s
).
The reaction order with respect to ArNMe at the
Thus, the ArNMe consumption step is the S 2 type reac-
2
2
N
stage of its quaternization by ECH was determined by the
Van't Hoff method comparing the initial consumption
tion because its rate is directly proportional to the con-
centration of the nucleophilic reagent in the first degree
rate of ArNMe (the slope of the kinetic curves, Figure 5)
(rSN2 = kCNuCE, rSN1 = kC ). The quaternization stage of
2
E
with its initial concentration and turned out to be ≈1
the tertiary amine leads to the formation of the anion of
the proton-donor reagent (R'COO in the case of NuH =
–
(0.988, Equation 8, correlation coefficient 0.998).
R'COOH).
The reaction systems “ArNMe2 – AcOH – ECH”
where the amine contains both electron donating and
electron withdrawing substituents in the benzene ring
were studied with the kinetic measurement procedure I
(kcat, monitoring the concentration of R'COOH) and II
(
kquater, monitoring the concentration of ArNMe ) in
2
–
order to confirm the fact that the generation of R'COO
in the reaction system is ensured precisely by ArNMe2
quaternization and to determine the effect of this quater-
nization on the rate of the total reaction. Using the corre-
lation analysis method, the obtained values of kquater and
kcat are compared with the basicity parameters according
to the Brønsted Equation 9
0
logk = logk + βpK ,
ð9Þ
a
and the electronic influence of substituents according to
the Hammett Equation 10:
FIGURE 4 The dependence of the optical density (A310) on
the concentration of 4-MeOC NMe in the reaction mixture
4-MeOC NMe – AcOH – ECH” (diluted with ECH) at the
initial time point; ε310 = 1130 M− cm ; r = 0.998
6
H
4
2
“
6
H
4
2
0
1
−1
logk = logk + ρσ:
ð10Þ

BAKHTIN ET AL.
7 of 14
The kinetic and correlation characteristics of the reac-
molecular. To describe the nucleophilic opening of
tion system “ArNMe – AcOH – ECH” are presented in
oxirane under the action of ArNMe , it is necessary to
2
2
Table 1.
determine which of the electrophiles present in the reac-
The following facts are noteworthy. Firstly, the nega-
tive sign of the reaction series constant ρ indicates the
development of a positive charge on the reaction
center—the nitrogen atom—in the TS for the consump-
tion system—free oxirane (ECH, stages k , k and k ,
1
−1
2
0
Scheme 2) or bound by the proton donor (ECHꢀHOCOR ,
0
0
2
stages K and k )—reacts with the tertiary amine.
1
To model the preferred nucleophile–electrophile
interaction, quantum chemical calculations of the quater-
nization stage of C H NMe were carried out for the
tion of ArNMe in the quaternization stage during the
2
interaction of the amine with oxirane. Secondly, the
values of the corresponding constants for the reaction
series—β (the sensitivity to the base properties of
6
5
2
paths I (Figure 6, TS1, electrophile—ECH) and II (Figure
6, TS2, electrophile—ECHꢀHOAc). The structural charac-
teristics of the found TSs are presented in the Table 2.
The resulting structures and changes that occur when
the nature of the electrophile (free/activated oxirane) var-
ies in TS1 and TS2 for SN reactions (including the
ArNMe ) and ρ (the sensitivity to the electronic effects in
2
the Ar structure)—corresponding to the opening of the
ECH ring by both ArNMe (kquater) and the anion of the
2
−
proton donor AcO (k ) differ slightly (β
= 0.37,
quater
cat
βcat = 0.31; ρquater = –1.2, ρcat = –1.0, Table 1). This indi-
cates that the amine quaternization stage (kquater) deter-
ArNMe quaternization with the oxirane) can be conve-
2
niently described using the two-dimensional reaction
4
6
mines the observed oxirane ring opening rate (k ) and is
coordinate—the More O'Ferrall–Jencks diagram
cat
a composite reaction of the overall process mechanism.
On the other hand, the quaternization of the amine
represented by Equation 2 cannot be an elementary stage
in the mechanism of the total reaction because it is tri-
(Figure 7). Along the axes, the diagram reflects the
dynamics of key processes that occur during the nucleo-
philic substitution at a saturated carbon atom: top-down
movement—the formation of the bond between the
TABLE 1
2
Kinetic and correlation characteristics of the reaction system “ArNMe – AcOH – ECH” (σ—Hammett constant of the
substituent)
4
5
quaterꢀ10 (70 C), M s−1
4
ꢂ
−1
catꢀ10 (60 C), M s−1
4
ꢂ
−1
ArNMe
2
pK
a
σ
k
k
5
5
2
.86
.07
.67
–0.268
19.7
1.02 ± 0.03
0
6.11
0.584 ± 0.032
0.104 ± 0.007
0.71
1.07
Correlation parameters
β
ρ
quater = 0.37 (r = 0.984)
quater = –1.2 (r = 0.989)
β
ρ
cat = 0.31 (r = 0.999)
cat = –1.0 (r = 0.999)
SCHEME 2 Possible ways of
quaternization of tertiary amine with oxirane

8
of 14
BAKHTIN ET AL.
FIGURE 6 Structures of the transition states TS1 (path I) and TS2 (path II, electrophilic assistance by acetic acid) calculated in the
B3LYP/6-31 + G** approximation for the gas phase
6¼
TABLE 2
TSs geometric parameters (bond length l, angle ∠, bond order n ) and imaginary frequency values (iν) in the vibrational
spectrum of transition states for the paths I (TS1) and II (TS2)
ꢂ
6¼
l, Å
∠,
n
1
1
2
2
1
1
1
1
2
1
1
2
1
2
1
1
2
1
1
1
2
1
2
1
2
1
TS
1
-iν, cm−
N C
C O
C O
C C
O C C
94.7
C C N
125.7
O C N
164.6
O C C N
−177.7
C O
C N
264
1.780
2.045
2.119
1.942
1.339
1.390
1.537
1.488
0.294
0.409
0.467
0.325
2
411
84.8
119.9
164.4
−173.5
FIGURE 7 Two-dimensional
reaction coordinate of the ECH opening
6 5 2
by C H NMe molecule (Scheme 3)
upon varying the nature of the
electrophilic component (free/activated
oxirane)
6¼
nucleophile and the electrophilic center (N–C bond order
group in the oxirane (C–O bond order n C2O1). The upper
left (A) and lower right (D) angles are the valleys of the
starting reagents and the quaternization product,
6¼
n
C2N1, Table 2), horizontally left to right—bond break-
ing between the electrophilic center and the leaving

BAKHTIN ET AL.
9 of 14
respectively, with the diagonal AD describing the path of
Then, taking into account the Hammond postulate and
the Thornton rule, the sum of the perpendicular and par-
allel to AD vectors really describes the observed resulting
displacement TS1 ! TS2 (Figure 7), which fits the
results of quantum chemical modeling.
the synchronous S 2 mechanism (an analog of the one-
N
dimensional coordinate). The two remaining angles show
the position of the intermediates for the limiting variants
of the nucleophilic substitution reaction: The angle B is
for the S 1 mechanism (includes carbocation, path
The calculated activation parameters for TS1 and TS2
are shown in Table 3.
N
ABD), and the angle C is for the S N mechanism
A
(
includes a hypothetical intermediate with a five-
According to the data obtained, in the case of the
DMA nucleophile, path II (electrophilic assistance with
coordinated carbon atom, path ACD).
The diagram shows the positions of the TSs for the
quaternization of DMA by a molecule of free ECH (TS1)
and protonated with acetic acid (TS2) according to the
results of quantum chemical modeling (Table 2). It fol-
lows from the diagram that both TSs are dissociative
because the degree of C–O bond cleavage prevails over
the degree of C–N bond formation.
the acid reagent CH COOH (TS2)) is characterized by a
3
significant decrease in the activation energy (E ) and by
a
an increase in the rate constant for the opening of
oxirane (k), that is, is preferred over the path I.
To experimentally establish the preferred path of the
quaternization reaction, the effect of varying the initial
concentration of the acid reagent on its rate was studied
(Figure 8).
To predict the displacement of the TS in the More
O'Ferrall–Jencks two-dimensional reaction coordinates,
it is convenient to be guided by the Hammond postu-
late (the shift of the TS along the main diagonal of
the reaction AD occurs to an angle that increases in
energy or from an angle whose free energy decreases)
and the Thornton rule (the TS shift occurs perpendicu-
lar to the diagonal AD to the angle whose energy
The obtained kinetic curves of the consumption of
0
ArNMe2 show that the initial reaction rate (r quater)
increases with rising the initial concentration of
acetic acid.
A comparison of the initial consumption rate of
0
ArNMe (logr quater) with the initial acid concentration
2
0
(logr quater) shows (Equation 11) that the reaction order
46
decreases or from the angle with increased energy).
with respect to AcOH is ≈1 (the tangent of the straight
The observed total TS shift is described by the sum of
the displacement vectors parallel and perpendicular to
the AD diagonal.
Analyzing the positions of the TSs, it can be seen that
path II, which involves the activation of the oxirane by
the proton-donor reagent, is characterized by an earlier
and more dissociative TS (TS2). Paths I and II differ in
the structure of particles containing an opened oxirane
ring (angles B and D, Scheme 3).
line) at the stage of quaternization:
logr0
= 1:003ꢀlogС
0
−3:21
quater
AcOH
ð11Þ
ðr = 0:995Þ:
Therefore, the experimental kinetic equation for the stage
of opening the oxirane ring by the amine is described by
the Equation 12:
From Scheme 3, it is obvious that the free energies of
the angles B and D in the case of path II will be much
lower than for I, because the neutral alcohol molecule is
thermodynamically more stable than the corresponding
rquater = kexpCamineCR0COOH:
ð12Þ
To establish the correspondence of the obtained kinetic
equation to paths I and/or II, Scheme 2 was analyzed
−
alcoholate, and ROH is a better leaving group than RO .
SCHEME 3 Particle structures in the
diagram for non-activated (path I) and activated
(
path 2) oxirane opening
6
6
6¼
TABLE 3
а
Standard activation thermodynamic parameters (ΔH , ΔS , ΔG ), activation energy (Е ), and reaction rate constants (k)
calculated in the B3LYP/6-31 + G** approximation for gas phase
6
6
6
k, s−1
TS
1
Path
ΔH , kJ/mol
ΔS , kJ/molꢀK
ΔG , kJ/mol
а
Е , kJ/mol
−
15
6
I
145.7
78.70
–40.21
157.7
105.5
142.9
83.96
1.47ꢀ10
2.08ꢀ10
−
2
II
–89.71

10 of 14
BAKHTIN ET AL.
Given the principle of stationarity, the concentration of
unstable intermediate alcoholate could be found from
Equations 17 and 18:
dCalcoholate
=
k1CECHCamine −k−1Calcoholate ð17Þ
dt
−
k2Calcoholate
C
0
= 0,
R COOH
k1CECHCamine
Calcoholate
=
:
ð18Þ
k−1 + k C
0
R COOH
2
Substituting Equation 18 (for k > > k ) into (16), the
2
-1
Equation 19 is obtained:
rquater = k CECHCamine,
ð19Þ
1
which corresponds to the zeroth order of the reaction
with respect to the proton-donor reagent and contradicts
the experimental results.
6 4 2
FIGURE 8 Kinetic curves for 4-MeOC H NMe consumption
at different initial concentrations of the acid reagent of the reaction
ꢂ
mixture “4-MeOC
6
H
4
NMe
2
– AcOH – ECH” (70 C)
Alternatively, the answer to the question about the
experimentally feasible reaction path was obtained by
comparing the dependence of the initial rate of the DMA
quaternization step (logkquater) on the acidity of the
using approximate kinetic methods. For the path II,
when the tertiary amine attacks the protonated ECH, the
proton-donor reagent (pK ). For this, the reaction systems
a
rate of the quaternization stage (ArNMe consumption)
“C H NMe – trimethylacetic acid (pK = 5.03) – ECH”
2
6
5
2
a
is described by the Equation 13:
and “C H NMe – ethoxyacetic acid (pK = 3.60) – ECH”
6
5
2
a
with various parameters of pK of the reagent were stud-
ied additionally by means of electronic spectroscopy
a
0
2
rquater = k CECHꢀHOCORCamine:
ð13Þ
(
Kinetic measurement procedure II). In accordance with
the Brønsted equation for acids—Equation 20,
The concentration of the oxirane activated by the
acid, according to the principle of quasi-equilibrium con-
centrations, is described by the Equation 14:
0
logk = logk −αpK ,
a
ð20Þ
it has been established that the rate constants of DMA
quaternization by ECH in the “C NMe – R'COOH –
ECH” systems are well correlated with the acidity param-
0
1
C
ECHꢀHOCOR0
= K CECHC
0
:
R COOH
ð14Þ
6
H
5
2
eters of the reagent (Equation 21):
Given the Equations 13 and 14, the expression for the
rate of quaternization takes the form:
logkquater = 1:0−0:51pKa
ð21Þ
ðr = 0:995,N = 3Þ:
0
0
rquater = k K CECHC Camine,
0
R COOH
ð15Þ
2
1
This suggests that acids with both electron-donating and
electron-withdrawing substituents act by a single mecha-
nism and confirm the participation of the proton-donor
reagent in the quaternization stage. An increase in the
which is in full agreement with the experimental results
Equation 12), because it corresponds to the first order of
the reaction with respect to both the proton donor and
(
amine, and kexp = k' K' CECH.
acidic properties of carboxylic acid leads to an increase in
2
1
0
For path I, when the amine attacks nonactivated
ECH, the quaternization rate is described by the
Equation 16:
the rate of quaternization, similar to rising C
in the
AcOH
reaction system “4-MeOC H NMe – AcOH – ECH”
6
4
2
(Figure 8). The slope of the straight line (α = 0.51, Equa-
+
tion 21) indicates a significant degree of H transfer in
the TS of the quaternization stage, because, otherwise, a
linear correlation either would not be observed or would
rquater = k2CalcoholateC
0
:
ð16Þ
R COOH

BAKHTIN ET AL.
11 of 14
be characterized by low sensitivity to the proton donor
acidity (α < 0.1 ).
The results of UV spectroscopic studies and correla-
tion analysis of the quaternization stage of tertiary
leaving group than to the nucleophilicity of ArNMe (α >
2
46
β). This is in favor of dissociative TS. Thus, the conclu-
sion about the dissociative nature of the TS of the amine
quaternization stage obtained by experimental methods
is completely consistent with the data of quantum chemi-
cal modeling (Table 2, Figure 7).
amines in the reaction series “ArNMe – AcOH – ECH”
2
(
β = 0.37, Table 1) and “C H NMe – R'COOH – ECH”
6
5
2
(
α = 0.51, Equation 21) indicate the greater sensitivity of
UV spectroscopy as noted above is not suitable for
confirming the occurrence of the quaternization stage
in the case of aliphatic tertiary amines (there is no
the quaternization rate constant to the strength of the
acid reagent R'COOH and, therefore, to the nature of the
1
FIGURE 9
3 3
H NMR spectrum of the mixture of ECH and Et N (1:2, CDCl , 400 MHz)
1
FIGURE 10
H NMR
spectrum of the mixture of ECH,
Et N, and AcOH at the initial
time point (CDCl , 400 MHz)
3
3

12 of 14
BAKHTIN ET AL.
1
FIGURE 11
H NMR spectrum of
the mixture of ECH, Et
3
N, and AcOH
ꢂ
after heating for 2 h at 60 C (CDCl
3
,
400 MHz)
1
characteristic chromophore). For this purpose, H NMR
the products of the ECH opening under the action of
AcO (chlorohydrin ester of acetic acid, Scheme 1 and
−
spectroscopy was used to analyze the “Et N – ECH”
3
(
Figure 9) and “ECH – AcOH – Et N” mixtures
Equation 3 (NuCH CH(OH)CH Cl, Nu = AcO)). Sec-
3
2
2
(Figures 10 and 11) with comparable concentrations of
ondly, a new triplet is formed at δ 1.34 ppm. Obvi-
reacting substances at different time intervals. The
assignment of signals from diastereotopic protons of
ECH was carried out according to the spectral
ously, the source of the indicated triplet in this range
+
of δ can be only the methyl group in the CH -CH -N
3
2
fragment. Increasing the intensity of the new triplet
over time, which was observed experimentally, can be
47
database.
In the case of the system “Et N – ECH” (Figure 9),
explained only by the gradual quaternization of Et N
3
3
the observed δ values of the triplet from the CH3
group (H(6), 1.02 ppm) and the quadruplet from the
CH group of Et N (H(5), 2.51 ppm) practically coin-
and the formation of tetraalkylammonium carboxylate,
which was initially absent in the reaction system.
Finally, the complication of the character and the
appearance of new signals in the region of 3–5 ppm, a
gradual decrease in the intensity of the multiplets of
the initial oxirane (signals H(1), H(2), H(3)), is due to
the opening of the ECH ring under the influence of
2
3
cide with the literature data for a solution of Et N in
3
47
CDCl3 (0.97 and 2.43 ppm, respectively) and, there-
fore, belong to the free base molecule. When acetic
acid is added to the mixture at the initial time point
(
Figure 10), the position of the ECH signals remains
nucleophiles present in the reaction system: Et N
3
unchanged, while the CH3 and CH2 signals of the
triethylamine protons are shifted to the weaker field
by 0.2 and 0.5 ppm, respectively (signals H(6 ) and
(initial) and AcO− (formed as a result of Et N
3
quaternization).
Finally, in order to elucidate the role of the acid
reagent on the interaction of ECH with amine, the
0
0
H(5 )). Such deshielding can be explained by the bind-
ing of the tertiary amine with the acid reagent
two-component system “Et N – ECH,” which does not
3
1
(Equation 1), which leads to the appearance of a posi-
contain acetic acid, was studied by H NMR method.
ꢂ
tive charge on the N atom. It is noteworthy that,
because of the development of a partial negative
Heating this mixture at 60 C (1 h) practically does not
lead to any changes in the spectrum where the signals
of the initial amine and oxirane remained unchanged,
in contrast to the results shown in Figure 11, where
charge on the O atom, the singlet of the CH group of
3
acetic acid (H(7), δ 1.97 ppm), on the contrary, slightly
shifts to a stronger field (in the acetic acid itself δ CH
is 2.1 ppm ).
Et N and ECH were consumed simultaneously accom-
3
3
47
panied by the accumulation of reactions products.
Thus, the presence of a proton donor is a prerequisite
for the possibility of nucleophilic opening of the
oxirane ring by aliphatic tertiary amines (as in the
case of ArNMe2).
Heating the mixture “Et N – AcOH – ECH” at
3
ꢂ
6
0 С for 2 h (Figure 11) leads to three key changes in
1
the H NMR spectrum of the initial reaction system.
Firstly, a singlet appears at 2.1 ppm, corresponding to

BAKHTIN ET AL.
13 of 14
4
| CONCLUSIONS
[16] J. Lubczak, R. Lubczak, D. Naróg, Open J Phys Chem 2018,
8
, 67.
[
[
17] V. F. Shvets, O. A. Tyukova, Zh Org Khim 1971, 7, 1847.
18] J. Hetflejs, F. Mares, V. Bazant, Collect Czech Chem Commun
Detailing the nucleophile–electrophile interactions in the
systems “R N – NuH – oxirane” was carried out with
3
1969, 34, 3098.
experimental and theoretical methods. It was found that
in the reaction under investigation, the activation of the
oxirane by the proton-donor reagent is carried out; the
tertiary amine is consumed in the quaternization stage
reacting with the activated oxirane according to the SN2
mechanism with the formation of tetraalkylammonium
carboxylate. The salt thus formed serves as a true catalyst
for the subsequent stage of the oxirane ring opening by
NuH leading to the formation of chlorohydrin esters. The
quaternization of tertiary amines under reaction condi-
[
[
19] M. F. Sorokin, L. G. Shode, Zh Org Khim 1968, 4, 666.
20] C. Maitre, F. Ganachaud, O. Ferreira, J. F. Lutz, Y. Paintoux,
P. Hemery, Macromolecules 2000, 33, 7730.
[21] H. Gerard, S. Götz, Y. Pellegrini, A. Castanet, A. Mortreux,
Appl Catal A 1998, 170, 297.
[
[
[
22] J. Chlebicki, J. Wegrzynska, I. Maliszewska, M. Oswiecimska,
J Surfactants Deter 2005, 8, 227.
23] F. D. Bobbink, D. Vasilyev, M. Hulla, S. Chamam, F. Menoud,
G. Laurenczy, P. J. Dyson, ACS Catal 2018, 8, 2589.
24] V. Calo, A. Nacci, A. Monopoli, A. Fanizzi, Org. Lett. 2002, 4,
2561.
1
tions was first studied by H NMR and UV spectroscopy.
[25] Y. Tsutsumi, K. Yamakawa, M. Yoshida, T. Ema, T. Sakai,
Using the methods of quantum chemical modeling and
correlation analysis, it was shown for the first time that
the TSs of the quaternization stage are dissociative, their
degree of looseness increasing upon the activation of the
oxirane ring by the NuH reagent.
Org. Lett. 2010, 12, 5728.
26] X. Yan, R. Fu, F. Liu, Y. Pan, X. Ding, G. He, Ind Eng Chem
[
Res 2018, 57, 3195.
[
[
[
[
[
27] J. G. Pritchard, F. A. Long, J. Am. Chem. Soc. 1957, 79, 2365.
28] C. G. Swain, C. B. Scott, J. Am. Chem. Soc. 1953, 75, 141.
29] F. A. Long, J. G. Pritchard, J. Am. Chem. Soc. 1956, 78, 2663.
30] J. G. Pritchard, F. A. Long, J. Am. Chem. Soc. 1956, 78, 2667.
31] J. G. Pritchard, F. A. Long, J. Am. Chem. Soc. 1956, 78, 6008.
ORCID
Stanislav Bakhtin https://orcid.org/0000-0003-3401-
[32] F. A. Long, J. G. Pritchard, F. E. Stafford, J. Am. Chem. Soc.
1957, 79, 2362.
8
770
[
[
[
[
[
[
33] S. Bakhtin, Y. Bespalko, E. Shved, React Kinet Mech Cat 2016,
19, 139.
34] S. G. Bakhtin, E. N. Shved, Y. N. Bespal'ko, Kinet Catal 2016,
7, 47.
35] S. Bakhtin, E. Shved, Y. Bespal'ko, J Phys Org Chem 2017, 30,
e3717.
36] S. Bakhtin, E. Shved, Y. Bespal'ko, Y. Stepanova, Prog React
Kinet Mec 2018, 43, 121.
37] D. D. Perrin, W. L. F. Amarego, Purification of Laboratory
Chemicals, Pergamon Press, Oxford 1988.
38] V. K. Ahluwalia, P. Bhagat, R. Aggarwal, R. Chandra, Interme-
diates For Organic Synthesis, I. K. International Publishing
House, New Delhi 2005.
39] A. I. Vogel, A Text Book of Practical Organic Chemistry,
Longman, London 1974.
40] A. Ito, Z. Fang, M. K. Brennaman, T. J. Meyer, Phys. Chem.
Chem. Phys. 2014, 16, 4880.
41] M. H. Kim, S. H. Kim, S. K. Ku, C. H. Park, B. Y. Joe,
K. W. Chun, H. C. Lee, US Patent Application 2011,
1
REFERENCES
[
1] A. Robin, F. Brown, N. Bahamontes-Rosa, B. Wu, E. Beitz,
J. F. Kun, S. L. Flitsch, J. Med. Chem. 2007, 50, 4243.
5
[
2] S. P. Pathare, K. G. Akamanchi, Tetrahedron Lett. 2013, 54,
6
455.
[
[
3] B. S. Reddy, US Patent Application 2011, 13(084), 848.
4] R. L. Thurmond, M. P. Beavers, H. Cai, S. P. Meduna,
D. J. Gustin, S. Sun, J. P. Edwards, J. Med. Chem. 2004, 47,
4
799.
5] W. J. Blank, Z. A. He, M. Picci, J Coat Technol Res 2002,
4, 33.
6] M. Rafizadeh, H. Ghasemi, V. Haddadi-Asl, Chinese J Polym
Sci 2006, 24, 599.
[
[
[
[
[
7
[
[
[
7] N. Pal, A. Srivastava, J. S. P. Rai, Polym Plast Technol 2003,
4
2, 105.
8] M. Kucharski, R. Lubczak, J. Chem. Technol. Biotechnol. 1998,
2, 117.
9] I. Zarzyka-Niemiec, J. Lubczak, Int. J. Chem. Kinet. 2003,
5, 73.
7
13/128, 030.
3
[
[
[
42] A. A. Granovsky, Firefly version 7.1.G, http://classic.chem.
msu.su/gran/firefly/index.html.
[10] Y. Tanaka, A. Okada, M. Suzuki, Can. J. Chem. 1970, 48, 3258.
[11] Y. Tanaka, H. Takeuchi, Tetrahedron 1968, 24, 6433.
[12] H. Kakiuchi, Y. Tanaka, J. Org. Chem. 1966, 31, 1559.
[13] M. F. Sorokin, E. L. Gershanova, Kinet Katal 1967, 8, 512.
[14] Y. Ishii, S. Sakai, T. Sugiyama, J Jpn Petrol Inst 1963, 5, 44.
[15] J. Lubczak, Open J. Phys. Chem. 2012, 2, 97.
43] G.
chemcraftprog.com.
44] M. S. Klebanov, F. Y. Kir'yazev, I. M. Shologon, Kinet Katal
984, 25, 1004.
A.
Andrienko,
Chemcraft,
https://www.
1

14 of 14
BAKHTIN ET AL.
[
45] A. Williams, Free Energy Relationships in Organic and Bio-
Organic Chemistry, Royal Society of Chemistry, Cambridge
How to cite this article: Bakhtin S, Shved E,
2
003.
Bespalko Y, Tyurina T, Palchykov V. Detailing the
elementary stages in the oxirane ring opening
reactions with carboxylic acids catalyzed by tertiary
amines. J Phys Org Chem. 2020;e4071. https://doi.
org/10.1002/poc.4071
[
46] E. V. Anslyn, D. A. Dougherty, Modern Physical Organic
Chemistry, University science books, Sausalito 2006.
47] Spectral Database For Organic Compounds, https://sdbs.db.
aist.go.jp/sdbs/cgi-bin/cre_index.cgi.
[