Molecular recognition-based catalysis in nucleophilic aromatic substitution: A mechanistic study

Article abstract of DOI:10.1039/c2nj40128b

Nucleophilic aromatic substitution for nitro-activated substrates in the presence of glymes and crown ethers is reported. The kinetic study reveals the many-sided nature of the polyether-catalyzed S_NAr mechanism as well as the main features aff

Full text of DOI:10.1039/c2nj40128b

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
NJC
Cite this: New J. Chem., 2012, 36, 1519–1526
www.rsc.org/njc
PAPER
Molecular recognition-based catalysis in nucleophilic aromatic
substitution: a mechanistic studyw
a
Nuno Basilio,a Luis Garcı
´ ´
Moises Perez-Lorenzo*
´
a-Rı
´
˜
a-Gallegoa and
´
o,* Angeles Pen
b
Received (in Montpellier, France) 21st February 2012, Accepted 24th April 2012
DOI: 10.1039/c2nj40128b
Nucleophilic aromatic substitution for nitro-activated substrates in the presence of glymes and
crown ethers is reported. The kinetic study reveals the many-sided nature of the polyether-
catalyzed SNAr mechanism as well as the main features affecting the course of the reaction. Thus,
the process can be efficiently controlled by the catalyst through a molecular-recognition process,
being also strongly dependent on the electronic structure of the substrate and, therefore, giving
rise to a versatile reaction system. The kinetic model proposed herein allows quantifying the
fractions of reaction proceeding through the different routes. Information on the nature of the
host–guest interaction between the catalyst and the substrate will also be extracted.
Introduction
The ability of glymes and crown ethers to bind inorganic and
organic cations has allowed them to be widely used in homo-
geneous and heterogeneous catalysis.1,2 Generally, these species
act as catalysts either in reactions involving metal ions, salts and
Scheme 1
bases, or processes requiring substrate recognition. In this regard,
ester aminolysis in aprotic solvents is one of the most reported
of SNAr by amines is well-established as well as the factors
cases in the literature.3,4 These reactions are known to proceed
through a rate-limiting breakdown of a zwitterionic tetrahedral
intermediate, which is generated by the addition of a molecule of
amine to the ester. In this case, glymes and crown ethers act as
base catalysts through a specific host–guest interaction with the
ammonium moiety of the reaction intermediate.5–14 In order to
achieve a better understanding of these molecular-recognition
processes, the exploration of alternative reactions involving
zwitterionic intermediates appears to be the next logical step.
For this purpose, the nucleophilic aromatic substitution
(SNAr) by n-butylamine for two different and representative
nitroarenes has been studied in the presence of glymes and
crown ethers. The criterion used for selecting these two nitro-
activated substrates has been the highly different weight of
base catalysis exhibited by these compounds in their corres-
ponding SNAr reactions. It must be noted that the mechanism
determining whether base catalysis is observed or not.15–19
Scheme 1 shows the mechanism for SNAr reactions involving
primary and secondary amines.20,21
In this kinetic study, both the amine and the polyether act as
base catalysts. Given the synthetic utility of this reaction,15
SNAr processes constitute an ideal subject for a comprehensive
mechanistic analysis. Thus, this work will provide helpful
information for the elucidation of the different operating
mechanisms in this process as well as the nature of the
host–guest interaction between the catalyst and the substrate.
Results
In the present work the polyether-catalyzed SNAr reaction of
1-chloro-2,4-dinitrobenzene (CDNB) and 1-fluoro-4-nitrobenzene
(FNB) in chlorobenzene is reported. The presence of electron-
withdrawing groups in these substrates, particularly at the ortho and
para carbons, comes from the necessity of reducing the electronic
density in the aromatic ring favoring the nucleophilic attack.
All the reactions have been run in the presence of crown
ethers and their open-chain analogues, glymes (Table 1). In all
cases n-butylamine has been used as the reactant nucleophile.
Analogously to ester aminolysis, it is expected that these
polyethers act as base catalysts facilitating the decomposition
of the Meisenheimer complex during the SNAr reaction given
a Center for Research in Biological Chemistry and Molecular
Materials (CIQUS), University of Santiago de Compostela,
15782 Santiago de Compostela, Spain
b Department of Physical Chemistry, University of Vigo, 36310 Vigo,
Spain. E-mail: moisespl@uvigo.es
w Electronic supplementary information (ESI) available: Influence of
the concentration of glymes and crown ethers on kobs for CDNB and
FNB SNAr together with the corresponding linearizations. Values for
a and b terms in the presence of different polyethers. See DOI:
10.1039/c2nj40128b
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 1519–1526 1519

View Article Online
Table 1 Polyethers employed as catalysts in the kinetic study of the
CDNB and FNB SNAr reaction
Catalyst (symbol)
Structure
Diglyme (G2)
Triglyme (G3)
Tetraglyme (G4)
Pentaglyme (G5)
12-Crown-4 (12C4)
15-Crown-5 (15C5)
Fig. 2 Influence of n-butylamine concentration on kobs/[BuNH2]
(eqn (2)) for the SNAr reaction of 1-chloro-2,4-dinitrobenzene, T =
25.0 1C, [CDNB] = 1.4 ꢁ 10ꢂ4 M. (K) [G4] = 0 M; (J) [G4] = 0.11 M;
(’) [G4] = 0.25 M; (&) [G4] = 0.35 M and (m) [G4] = 0.55 M.
where kobs denotes the observed rate constant for the SNAr
reaction. This linear and quadratic dependence can be confirmed
by linearizing eqn (1) to eqn (2):
18-Crown-6 (18C6)
kobs
¼ a þ b½BuNH2ꢀ
ð2Þ
½BuNH2ꢀ
Fig. 2 shows how eqn (2) fits the experimental data (see also
Fig. S7–S12 and S22–S24, ESIw). As a result, this dual kinetic
behavior can be clearly confirmed, which suggests the possibility
of different operating mechanisms in this process.
that both processes show the same type of ammonium moiety
in their corresponding zwitterionic addition intermediate.
Under all the experimental conditions used in this study the
reaction rate for the CDNB and FNB nucleophilic aromatic
substitution shows a first-order dependence on the nitro-aromatic
compound concentration. Regarding the nucleophile, all the
reactions exhibit a linear and quadratic dependence on the
n-butylamine concentration and a remarkable catalytic effect
as the polyether concentration is increased (see Fig. 1 and also
Fig. S1–S6 and S19–S21, ESIw).
From the intercepts and slopes in Fig. 2, values for a and b
in the absence and presence of catalysts can be obtained for the
CDNB SNAr reaction. By plotting these parameters against the
catalyst concentration for every set of data, Fig. 3 is obtained.
As shown in Fig. 3, a exhibits a linear dependence on the
catalyst concentration while b remains constant despite the
variation in the polyether concentration. This behavior is
observed for all glymes and crown ethers tested in this work
(see Fig. S13–S18, ESIw).22 Accordingly, these rate constants
can be expressed as a = kA + kC[catalyst] and b = kB. Thus,
the rate equation for CDNB SNAr reaction (eqn (1)) can be
rewritten as eqn (3):
Thus, the process is governed by eqn (1).
kobs = a[BuNH2] + b[BuNH2]2
(1)
kobs =kA[BuNH2] + kB[BuNH2]2 + kC[BuNH2][catalyst]
(3)
Fig. 1 Influence of n-butylamine concentration on kobs for the SNAr
reaction of 1-chloro-2,4-dinitrobenzene, T = 25.0 1C, [CDNB] =
1.4 ꢁ 10ꢂ4 M. (K) [G4] = 0 M; (J) [G4] = 0.11 M; (’) [G4] =
0.25 M; (&) [G4] = 0.35 M and (m) [G4] = 0.55 M.
Fig. 3 Influence of G4 concentration on the a and b terms (eqn (2)) for
the SNAr reaction of 1-chloro-2,4-dinitrobenzene. (K) a and (J) b.
c
1520 New J. Chem., 2012, 36, 1519–1526
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012

View Article Online
kobs = kA[BuNH2] + kB[BuNH2]2 + kC[BuNH2][catalyst]
Table 2 Macroscopic rate constants for the CDNB SNAr by n-butyl-
amine in the presence of different polyethers, T = 25.0 1C
+ kD[BuNH2]2[catalyst]
(5)
Catalyst
104 kA/Mꢂ1 sꢂ1
103 kB/Mꢂ2 sꢂ1
103 kC/Mꢂ2 sꢂ1
Eqn (5) accounts for the experimental results found for the
CDNB and FNB nucleophilic aromatic substitution reactions.
Thus, while the CDNB SNAr reaction is governed by the rate
constants kA, kB and kC (eqn (3)), the kinetic behavior for the
FNB reaction is controlled by kB, kC and kD (eqn (4)). All
these observations point to the existence of two differentiated
operating mechanisms for the CDNB and FNB SNAr reactions
or at least, a common mechanism where the relative weight of
the different paths is markedly different.
G2
G3
G4
G5
12C4
15C5
18C6
6.5 0.2
7.0 0.3
6.8 0.1
7.0 0.1
6.8 0.1
6.6 0.1
6.7 0.3
2.9 0.1
2.9 0.1
3.0 0.1
3.2 0.2
3.1 0.1
3.2 0.2
2.9 0.2
1.4 0.1
3.2 0.1
4.7 0.1
5.2 0.1
2.3 0.1
3.2 0.1
4.1 0.1
Discussion
1. Polyether-catalyzed SNAr reaction of 1-chloro-2,4-
dinitrobenzene
From eqn (3), a mechanism for the CDNB SNAr reaction with
n-butylamine in the presence of polyethers can be proposed
(Scheme 2).
Thus, this reaction would proceed through an initial attack
of n-butylamine at the nucleofuge-bearing carbon atom of
CDNB to generate a zwitterionic Meisenheimer complex. The
existence of this s-complex tetrahedral intermediate (from
now on simply referred to as Tꢃ) has been widely reported
in the literature.18,19 In the absence of catalysts, this addition
intermediate may lead to the final product both spontaneously
(k1 route) and assisted by another molecule of amine (k2 route).
The latter process would take place through a stepwise mecha-
nism. As can be observed, the existence of a Tꢃ–amine complex is
proposed. This complex will be referred to as TꢂꢄA+ given that
Tꢃ is more acidic (1–2 pKa units) than the butylammonium ion
and, therefore, it is unlikely the formation of a complex between
Tꢃ and an amine molecule not involving a proton transfer. In the
presence of added polyethers, as proposed by Hogan and
Gandour8 for aminolysis reactions, the formation of a
complex between the intermediate and the catalyst (TꢃꢄC) is
suggested. Such a complex would evolve directly into the final
product through a rate-limiting decomposition (k3 route).
It must be noted that the different reaction pathways of the
mechanisms shown in Scheme 2 are accounted for in eqn (3).
On one hand, the catalyst-independent terms of this expression
show a linear and quadratic dependence on the amine concen-
tration. These terms would correspond to the spontaneous and
amine-assisted decomposition of the tetrahedral intermediate,
respectively. On the other hand, the catalyst-dependent term
Fig. 4 Influence of G4 concentration on the a and b terms (eqn (2))
for the SNAr reaction of 1-fluoro-4-nitrobenzene. (K) a and (J) b.
Table 2 shows the values for the CDNB macroscopic rate
constants. These parameters are obtained from the intercepts
and slopes in Fig. 3 and Fig. S13–S18 (ESIw).
The same procedure is followed to determine the rate
equation for the FNB SNAr reaction. By plotting a and b
values against the polyether concentration, Fig. 4 is obtained.
In Fig. 4, the FNB SNAr reaction also exhibits a linear
dependence of a on the catalyst concentration. However, in
contrast to the CDNB SNAr reaction, b does not remain constant
but increases linearly as the polyether concentration is increased
(see also Fig. S25 and S26, ESIw). Therefore, a and b for the FNB
SNAr reaction can be reformulated as a = kC[catalyst] (it must be
noted that no intercept is observed in the absence of catalysts) and
b = kB + kD[catalyst]. As a result, the rate equation for the FNB
SNAr reaction can be expressed as eqn (4):
kobs = kB[BuNH2]2 + kC[BuNH2][catalyst]
+ kD[BuNH2]2[catalyst]
(4)
Table 3 shows the values for the FNB macroscopic rate
constants. These parameters are obtained from the intercepts
and slopes in Fig. 4 and Fig. S25 and S26 (ESIw).
Taking into account these results, a more general kinetic law
for polyether-catalyzed SNAr can be proposed. This rate
equation is given by eqn (5):
Table 3 Macroscopic rate constants for the FNB SNAr by n-butyl-
amine in the presence of different polyethers, T = 25.0 1C
Catalyst
107 kB/Mꢂ1 sꢂ1
106 kC/Mꢂ2 sꢂ1
106 kD/Mꢂ2 sꢂ1
G4
G5
18C6
8.9 0.2
9.0 0.1
9.3 0.2
2.5 0.1
3.0 0.1
1.1 0.1
1.1 0.1
1.3 0.1
1.1 0.1
Scheme 2
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 1519–1526 1521

View Article Online
would be ascribed to the rate-limiting decomposition of the
TꢃꢄC complex. According to this model, the empirical constants
previously determined (kA, kB and kC) can be redefined as a
of the TꢃꢄCꢄA complex (k4 route in Scheme 3B). Concerted or
stepwise, the amine-assisted decomposition of TꢃꢄC and the
polyether-promoted breakdown of the TꢂꢄA+ intermediate must
be considered since they both are kinetically undistinguishable. It
must be also noted that the very existence of the k4 route supports
the postulated formation of the TꢃꢄC and/or TꢂꢄA+ complex,
since a concerted trimolecular process between the Meisenheimer
complex, the catalyst and the amine is highly unlikely.
function of microscopic parameters. Thus, kA = KTk1, kB
=
KTKTAk2 and kC = KTKTCk3, where KT, KTA and KTC are the
equilibrium constants for the formation of the Tꢃ, TꢂꢄA+ and
TꢃꢄC complexes, respectively, and k1, k2 and k3 are the intrinsic
reactivity constants for the CDNB SNAr reaction.
The different reaction pathways considered in Scheme 3 are
accounted for in eqn (4). On one hand, the catalyst-independent
term shows exclusively a quadratic dependence on the amine
concentration. This term would correspond to the amine-assisted
breakdown of Tꢃ. On the other hand, the catalyst-dependent
terms show both linear and quadratic dependence on the amine
concentration. The first can be ascribed to the spontaneous
decomposition of the TꢃꢄC complex. The latter can be attributed
to the amine-promoted decomposition of the TꢃꢄC complex and/
or the polyether-promoted decomposition of the TꢂꢄA+ complex.
As pointed out in eqn (4), the FNB nucleophilic aromatic
substitution is governed by the macroscopic rate constants kB,
kC and kD. According to this kinetic model, such constants can be
redefined as a function of microscopic parameters. Thus,
kB = KTKTAk2, kC = KTKTCk3, kD ¼ KTKTCk4conc1 and/or
2. Polyether-catalyzed SNAr reaction of 1-fluoro-4-
nitrobenzene
Taking into account the kinetic law shown in eqn (4) the
following reaction mechanisms for the FNB SNAr may be
suggested (Scheme 3).
Analogously to CDNB, the FNB SNAr reaction proceeds
through a nucleophilic attack at the nucleofuge-bearing carbon
generating a tetrahedral intermediate. In the absence of catalysts,
this intermediate would lead to the final product through a stepwise
amine-promoted decomposition (k2 route). However, unlike
CDNB, the rate law for the FNB SNAr reaction shows no linear
dependence on the amine concentration in the catalyst-independent
term (eqn (4)). Therefore, a spontaneous breakdown of the addition
intermediate can be ruled out. In the presence of polyethers several
are the possibilities. On one hand, as concerted mechanisms, the
amine-assisted decomposition of the TꢃꢄC complex and/or the
polyether-promoted breakdown of the TꢂꢄA+ species are
proposed (k4 routes in Scheme 3A). On the other hand, and
assuming a stepwise nature of such pathways, the formation of
a Tꢃ–catalyst–amine species (TꢃꢄCꢄA) is postulated. This
complex would arise from the association of TꢃꢄC to a
molecule of amine and/or the binding of TꢂꢄA+ to a molecule
of polyether (see discussion in Section 3). Either way, the
reaction would proceed through a rate-limiting decomposition
kD ¼ KTKTAkc4onc2 or kD = KTKTCKTCAk4step and/or kD
=
KTKTAKTACks4tep, where KT, KTA, KTC and KTCA/TAC are the
equilibrium constants for the formation of the Tꢃ, TꢂꢄA+, Tꢃꢄ
C and TꢃꢄCꢄA complexes, respectively, and k2, k3 and k4 are
the intrinsic rate constants for the FNB SNAr reaction.
3. Nature of the polyether catalysis on SNAr reactions
In order to elucidate the nature of the observed catalysis, a
systematic comparison of the catalytic efficacies of glymes and
crown ethers is required. This analysis will allow us to identify the
best catalyst structure for the reaction as well as understand the
mechanism of the host–guest interaction between the polyether
and the substrate.
Crown ethers are generally known to be better complexing
agents than glymes,1 and might therefore be expected to be
better SNAr catalysts as well. However, as shown in Table 2,
glyme catalysis exhibits an inverse macrocyclic effect, i.e.,
open-chain polyethers are better catalysts than the corres-
ponding macrocyclic polyethers. This behavior enables a close
analogy between SNAr and ester aminolysis reactions, since
the latter also undergo catalysis in the presence of phase
transfer agents when carried out in aprotic solvents.5,6 In this
case, catalysis arises from the binding of polyethers to the
ammonium ion part of the tetrahedral intermediate formed by
the attack of the amine at the ester carbonyl group. The binding
interaction accelerates the decomposition of Tꢃ by breaking the
stabilization of the uncomplexed zwitterionic intermediate and
as a result, facilitating the expulsion of the leaving group. In
view of the fact that the same type of ammonium moiety is
present in the addition intermediate of SNAr reactions with
amines, it may be assumed that glymes and crown ethers act in a
similar way to that found in ester aminolysis.
(a) Catalysis by glymes. A clear evidence for the analogous
kinetic behavior of ester aminolysis and SNAr reactions is
presented in Fig. 5. In this plot it is shown how the per oxygen
Scheme 3
c
1522 New J. Chem., 2012, 36, 1519–1526
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012

View Article Online
Fig. 5 Per oxygen catalytic rate constant, kC/Oxy, vs. number
of oxygens per glyme molecule. (K) Aminolysis of p-nitrophenyl
acetate by n-butylamine;13 (J) SNAr of 1-chloro-2,4-dinitrobenzene
by n-butylamine.
Fig. 6 Per oxygen catalytic rate constant, kD/Oxy, vs. number of oxygens
per glyme molecule. (K) Aminolysis of p-nitrophenyl acetate by n-butyl-
amine;13 (J) SNAr of 1-fluoro-4-nitrobenzene by n-butylamine.
acetate aminolysis the kD/Oxy ratio remains constant with an
increasing glyme length (Fig. 6). In spite of the lack of
experimental data, the FNB SNAr reaction seems to show a
similar trend for kD/Oxy than that found in ester aminolysis.
This behavior is consistent with the formation of a single
bifurcated hydrogen-bonded glyme–substrate association
(Scheme 5) since one of the two protons of the ammonium moiety
must be free to be abstracted by a base, i.e., an amine molecule.
catalytic rate constant for both reactions increases as the
number of oxygens per glyme molecule increases. The profiles
seem to level off at four oxygens, and successive oxygens
would only contribute to the catalysis in a statistical manner.
From this profile, the –(CH2OCH2)4– subunit is identified as
the optimal segment for catalysis of ester aminolysis. This
catalytic effect implies
a specific host–guest interaction
between the glyme and the tetrahedral intermediate in such a
way that four oxygens donate electron density to stabilize the
ammonium moiety of Tꢃ. Thus, the formation of a doubly
bifurcated hydrogen-bonded catalyst–substrate complex is
proposed. Since a similar catalytic profile is found for the
CDNB SNAr reaction, the existence of a four-point recognition
of the secondary ammonium ion of the zwitterionic intermediate
for this kind of processes may be also suggested. In this respect,
Scheme 4 shows a tentative transition structure where this kind
of host–guest interaction takes place. Hence, a glyme with more
than four O atoms would only use four oxygens in order to
catalyze the SNAr reaction. Further interactions between Tꢃ and
glymes have been discussed in the literature.9
(b) Catalysis by crown ethers. In Fig. 7 it is shown how the
per oxygen catalytic rate constant for the CDNB SNAr reaction
increases as the number of oxygens per crown ether molecule
increases. This catalytic profile starts to level off at five oxygen
atoms. The same tendency is observed for aminolysis of p-nitro-
phenyl acetate.
In this case, a mechanism where 12C4 binds the two
ammonium protons of the intermediate is proposed. This
interaction would consist in one simple and two bifurcated
hydrogen bonds (Scheme 6A). Thus, 12C4 would bind the
reaction intermediate through three oxygen atoms. On the other
hand, 15C5 and 18C6 would give rise to a doubly bifurcated
hydrogen-bonded polyether–substrate association (Scheme 6B
and C). Both crown ethers would make use of four oxygen
atoms in order to stabilize the zwitterionic intermediate.
According to these results, crown ether-catalyzed nucleophilic
aromatic substitution would proceed through the same mechanism
than that found for ester aminolysis.
Due to the low reaction rate of the FNB SNAr reaction
(half-life times on the order of weeks and months) only a
reduced set of experiments on this substrate is shown. However, it
must be noted that similar values were obtained for the per
oxygen catalytic rate constant, kC/Oxy, in the presence of G4 and
G5 glymes: (6.2 ꢃ 0.1) ꢁ 10ꢂ7 Mꢂ2 sꢂ1 and (6.0 ꢃ 0.2) ꢁ 10ꢂ7
M
ꢂ2 sꢂ1, respectively. These results support the idea of a doubly
4. Influence of the substrate on the polyether-catalyzed SNAr
mechanism
bifurcated hydrogen-bonded polyether–substrate interaction.
Regarding the polyether-promoted decomposition of the TꢂꢄA+
complex (and/or the amine-promoted breakdown of the TꢃꢄC
complex) an analogy between nucleophilic aromatic substitution
and ester aminolysis can be also drawn. Thus, for p-nitrophenyl
The observation of polyether catalysis for the CDNB and
FNB nucleophilic aromatic substitution reactions points to a
rate-limiting decomposition of the SNAr Meisenheimer complex.
Scheme 4
Scheme 5
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 1519–1526 1523

View Article Online
On one hand, this attractive interaction makes the ortho-nitro
derivatives being more reactive than the para-isomers in their
reaction with amines. On the other hand, the hydrogen bond
decreases the energy of the reaction intermediate promoting
the spontaneous collapse (Scheme 7A). Additional evidence in
favor of this hydrogen bond stabilization was provided by the
same authors.25 On studying the o- and p-fluoronitrobenzene
SNAr with amines, an o : p ratio of 444 for the reaction rates in
toluene is obtained. The magnitude of this value indicates the
higher stability of the intermediate s-complex in the ortho-
isomer. In addition, possible steric effects due to the presence
of the nitro group at the ortho carbon can be ruled out. In the
absence of catalysts, the much higher reactivity of CDNB than
that found for FNB supports these statements. Accordingly,
the presence of a poorer leaving group in FNB and the lack of
a nitro group at the ortho carbon of its aromatic ring
(Scheme 7B) would justify the absence of a non-supported
decomposition of Tꢃ for the FNB SNAr reaction.
Fig. 7 Per oxygen catalytic rate constant, kC/Oxy, vs. number of
oxygens per crown ether molecule. (K) Aminolysis of p-nitrophenyl
acetate by n-butylamine;13 (J) SNAr of 1-chloro-2,4-dinitrobenzene
by n-butylamine.
Even though glymes and crown ethers are much weaker bases
than amines, they exhibit a similar ability to act as base
catalysts despite their pKa differences (see kB and kC values
in Table 2). In this regard, it has been reported that for ester
aminolysis in aprotic solvents the catalytic abilities of oxygen
bases are enhanced relative to those of nitrogen bases.4 Thus,
the catalytic efficacy of the bases would be correlated with
their hydrogen-bonding capacity rather than their aqueous
basicity. This behavior can be extrapolated to SNAr reactions
where the same type of polyether–ammonium moiety inter-
action can take place.
According to Schemes 2 and 3, another major difference
between the CDNB and FNB SNAr mechanisms can be
observed, which is the formation and breakdown of the Tꢃꢄ
CꢄA rate-limiting structure found in the FNB SNAr reaction.
The absence of this route in the CDNB SNAr reaction can be
considered as expected taking into account the different
experimental evidence found previously. In this regard, by
using eqn (3) and (4) and data shown in Tables 2 and 3 the
fraction of polyether-catalyzed SNAr reaction can be calcu-
lated (Fig. 8). Thus, taking T = 25.0 1C, [BuNH2] = 0.1 M,
[G4] = 0.1 M and [(O2N)xAr–L] = 1.5 ꢁ 10ꢂ4 M as reaction
conditions, it can be estimated that only 30% of the CDNB
SNAr reaction is catalyzed by the glyme intervention. Under
the same conditions, 75% of the FNB SNAr reaction occurs
with the involvement of polyethers. These results show how
variable the significance of the polyether catalysis depending
on the substrate structure can be. Hence, the remarkable
stabilization of Tꢃ in the CDNB SNAr reaction compared
to FNB favors the route where no polyether is involved, while
the high energy barrier for the spontaneous decomposition
of Tꢃ in the FNB SNAr reaction inhibits this process,
promoting the reaction pathways where glymes and crown
ethers catalyze the formation of the final product. Therefore,
the absence of a route in the CDNB SNAr reaction where Tꢃꢄ
CꢄA is formed and evolves into the final product can
be justified on the basis of the existence of a lower energy
reaction pathway. This may decrease the need for the poly-
ether catalyst favoring alternative and probably less energy
demanding pathways.
Major differences between the CDNB and FNB SNAr
reactions can be observed by analyzing mechanisms shown
in Schemes 2 and 3. Thus, the CDNB SNAr reaction shows a
pathway where the Tꢃ intermediate evolves spontaneously
into the final product (k1 route). Conversely, the assistance
of an amine molecule to the zwitterionic intermediate becomes
essential for the FNB SNAr reaction. The ability of the
tetrahedral intermediate of CDNB to decompose without
further support may be due to two different factors. First,
the better leaving group present in CDNB compared to FNB
will decrease the energy of the reaction intermediate facilitating
the spontaneous breakdown. Second, and based on previous
works,23,24 the ability of the Meisenheimer complex of CDNB to
collapse without the participation of the polyether may also be
attributed to the presence of the nitro group at the ortho carbon
of the aromatic ring. In this respect, Nudelman et al. reported a
detailed semiempirical study on s-complexes derived from the
reaction of fluorobenzene, o- and p-fluoronitrobenzene with
ammonia. Their results show the importance of the hydrogen
bond between the ortho-nitro group and the ammonium proton.
Scheme 6
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
1524 New J. Chem., 2012, 36, 1519–1526

View Article Online
Scheme 7
At this point, combining Schemes 2 and 3, a general
mechanism for the polyether-catalyzed SNAr reaction with
amines in aprotic solvents can be obtained (Scheme 8). Both
CDNB and FNB will contribute to the overall reaction
mechanism.
Scheme 8
SNAr reaction shows a spontaneous decomposition of Tꢃ
(k1 route), the formation of a TꢃꢄCꢄA rate-limiting structure
and its subsequent collapse (k4 route) is exclusive of the
FNB SNAr reaction. Besides, an additional route involving a
reaction between the substrate and an amine–catalyst precomplex
could be suggested. Such preassociation would modify the charge
distribution of the amine giving rise to a more reactive nucleophile.
However, this possibility may be ruled out in view of previous
works.10,11 In this regard, when formation of Tꢃ in ester aminolysis
becomes rate-limiting, the reaction rate is not affected by the
presence of added-crown ethers. This absence of catalysis points
to the participation of a non-precomplexed nucleophile in the
reaction process. Hence, an association between Tꢃ and CꢄA to
yield TꢃꢄCꢄA may also be excluded.
On one hand, CDNB and FNB show common reaction
pathways through which SNAr takes place. These routes are
the amine-promoted breakdown of Tꢃ (k2 route) and the
spontaneous decomposition of TꢃꢄC (k3 route). On the other
hand, each substrate shows a distinctive reaction pathway to
give rise to the final product. Thus, while only the CDNB
The obtained results show the critical influence of the
electronic structure of the substrate on the SNAr reaction
mechanism. Accordingly, depending on the nature of the
substrate the amine : catalyst ratio can be altered in order to
increase in a controlled manner the overall reaction rate by
modifying the balance among the mentioned operating
mechanisms. The kinetic model proposed herein also allows
quantifying the fraction of reaction proceeding through the
different reaction routes.
Conclusions
Nucleophilic aromatic substitution for nitro-activated
substrates in the presence of glymes and crown ethers has
been reported. This kinetic study reveals the multipathway
character of the polyether-catalyzed SNAr reaction. In this
regard, the substrate structure defines the nature of the
reaction mechanism giving rise to a versatile process. Thus,
depending on the electronic properties of the nitro-activated
compound different reaction routes have been found. On the
other hand, it has been observed that the formation of the final
product can be efficiently accelerated either by the presence of
glymes or crown ethers, being identified the –(CH2OCH2)4–
subunit as the optimal segment for catalysis.
Experimental section
All chemicals were of the highest commercially available
purity and were used as supplied. Kinetic experiments were
conducted in chlorobenzene at 25 1C. All rates were measured
using a 8453 Agilent diode array UV-Vis spectrophotometer
monitoring the formation of the final product at 350 and 375 nm
Fig. 8 Fraction of polyether-catalyzed SNAr reaction calculated
through eqn (3) and data shown in Table 2, and eqn (4) and data
shown in Table 3 for the CDNB SNAr (A) and FNB SNAr (B)
reactions, respectively. Data range: [BuNH2] = 0 to 0.5 M; [G4] =
0 to 0.5 M.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012
New J. Chem., 2012, 36, 1519–1526 1525

View Article Online
for the CDNB and FNB SNAr reactions, respectively. Typical
nitro-activated substrate concentrations were [(O2N)xAr–L] =
(1.3–1.5) ꢁ 10ꢂ4 M. Amine, glyme and crown ether concentrations
were always in large excess over the nitroaromatic compound,
ensuring pseudo-first-order conditions. The absorbance–time data
always fitted the first-order integrated equation satisfactorily and
the observed rate constant, kobs, was found to be reproducible
within a precision of about 3% or better. As expected, under
all the experimental conditions the corresponding amino-
substituted product was observed.
10 A. B. Maude and A. Williams, J. Chem. Soc., Perkin Trans. 2,
1995, 691–696.
11 A. B. Maude and A. Williams, J. Chem. Soc., Perkin Trans. 2,
1997, 179–183.
12 N. Basilio, L. Garcı
Lorenzo, Chem. Commun., 2005, 3817–3819.
13 N. Basilio, L. Garcıa-Rıo, J. C. Mejuto and M. Pe
J. Org. Chem., 2006, 71, 4280–4285.
´
a-Rı
´
o, J. R. Leis, J. C. Mejuto and M. Pe
´
rez-
´
´
´
rez-Lorenzo,
14 H. J. Koh, K. L. Han and I. Lee, J. Org. Chem., 1999, 64,
4783–4789.
15 F. Terrier, Nucleophilic Aromatic Displacement, Wiley, New York,
1991.
16 M. R. Crampton, T. A. Emokpae, C. Isanbor, A. S. Batsanov,
J. A. K. Howard and R. Mondal, Eur. J. Org. Chem., 2006,
1222–1230.
17 M. Barra, H. R. de Rossi and E. B. de Vargas, J. Org. Chem., 1987,
52, 5004–5008.
Acknowledgements
M.P.-L. acknowledges the financial support from the Isidro
Parga Pondal Program (Xunta de Galicia, Spain). This work
18 J. F. Bunnett, Quart. Rev. (London), 1958, 12, 1–16.
19 C. F. Bernasconi, MTP Int. Rev. Sci.: Org. Chem., Ser. One, 1973,
3, 33–63.
was funded by the Spanish Ministerio de Ciencia y Tecnologıa
´
20 It must be noted that only the sX adduct is depicted in this scheme.
In this regard, in halonitroarenes contrary to esters, amines as well
as other nucleophiles can be added in few positions to yield the
corresponding sH adducts.21 Since a hydride anion cannot depart
in a spontaneous reaction, whereas this addition is a reversible
process, dissociation of the initially formed sH adducts followed by
slower addition in the halogen position results in the formation of
the sX adducts that finally provide the products of the SNAr. Thus,
substitution of hydrogen will be the primary fast process whereas
the conventional SNAr of halogen will be a slower secondary
process which, therefore, will be subjected to polyether catalysis.
21 M. Ma˛kosza, Chem. Soc. Rev., 2010, 39, 2855–2868 and references
therein.
(Project CTQ2011-22436) and Xunta de Galicia (PGIDIT07-
PXIB209041PR, PGIDIT10-PXIB209113PR and 2007/085).
References
1 G. W. Gokel and O. Murillo, Compr. Supramol. Chem., 1996, 1,
1–33.
2 J. S. Bradshaw, R. M. Izatt, A. V. Bordunov, C. Y. Zhu and
J. K. Hathaway, Compr. Supramol. Chem., 1996, 1, 35–95.
3 F. M. Menger and J. H. Smith, J. Am. Chem. Soc., 1972, 94,
3824–3829.
4 C.-W. Su and J. W. Watson, J. Am. Chem. Soc., 1974, 96,
1854–1857.
5 R. D. Gandour, D. A. Walker, A. Nayak and G. R. Newkome,
J. Am. Chem. Soc., 1978, 100, 3608–3609.
6 J. C. Hogan and R. D. Gandour, J. Am. Chem. Soc., 1980, 102,
2865–2866.
22 According to these experimental results, medium effects on the rate
constant due to the increasing polyether concentration can be ruled
out. A change in the polarity of chlorobenzene would be evidenced
through an upward curvature in the plot of a against polyether
concentration.
7 H. Zipse, L. H. Wang and K. N. Houk, Liebigs Ann., 1996,
1511–1522.
8 J. C. Hogan and R. D. Gandour, J. Org. Chem., 1991, 56,
2821–2826.
9 J. C. Hogan and R. D. Gandour, J. Org. Chem., 1992, 57, 55–61.
23 C. F. Bernasconi and R. H. de Rossi, J. Org. Chem., 1974, 41, 44.
24 N. S. Nudelman and P. MacCormack, Tetrahedron, 1984, 40,
4227–4235.
25 N. S. Nudelman and S. Cerdeira, J. Chem. Soc., Perkin Trans. 2,
1986, 695–698.
c
1526 New J. Chem., 2012, 36, 1519–1526
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2012