Reaction of 2,4-dinitrochlorobenzene with aniline. Solvent effects and molecular complex formation

Article abstract of DOI:10.1039/a903559a

The kinetics of the reaction of aniline with 2,4-dinitrochlorobenzene (2,4-DNC1B) were studied in several benzene-n-hexane mixtures at 40 °C in the presence of variable amounts of aniline. A linear dependence of the second-order rate coefficients, k_A, with [B] is observed, with a null intercept. Taking into account the non-polar character of the solvent, and the excellent nucleofugacity of chlorine, this kinetic behaviour is interpreted as evidence of the aggregation of aniline, the hydrogen-bonded dimer acting as the nucleophile. Consistent with this interpretation, when the solvent is changed to THF, a good hydrogen-bond acceptor (HBA), the k_A is no longer dependent on [B]. Electron donor-acceptor (EDA) molecular complexes of 2,4-DNC1B with benzene and with aniline, and also between aniline and the product, were clearly recognized and the stability constants of each one of these complexes were determined.

Full text of DOI:10.1039/a903559a

View Article Online / Journal Homepage / Table of Contents for this issue
Reaction of 2,4-dinitrochlorobenzene with aniline. Solvent effects
and molecular complex formation
N. Sbarbati Nudelman,a Monica Savini,b C. E. Silvana Alvaro,b Viviane Nicotrab and
Jeanette Yankelevichb
a Depto. de Química Orgánica, Facultad de Ciencias Exactas y Naturales,
Universidad de Buenos Aires, 1428 Buenos Aires, Argentina.
E-mail: nudelman@qo.fcen.uba.ar. Fax ϩ54 11 4 5763355
b Depto. de Química, Facultad de Ingeniería, Universidad Nacional del Comahue, Neuquén,
Argentina
Received (in Cambridge) 28th April 1999, Accepted 3rd June 1999
The kinetics of the reaction of aniline with 2,4-dinitrochlorobenzene (2,4-DNClB) were studied in several benzene–n-
hexane mixtures at 40 ЊC in the presence of variable amounts of aniline. A linear dependence of the second-order rate
coefficients, kA, with [B] is observed, with a null intercept. Taking into account the non-polar character of the solvent,
and the excellent nucleofugacity of chlorine, this kinetic behaviour is interpreted as evidence of the aggregation of
aniline, the hydrogen-bonded dimer acting as the nucleophile. Consistent with this interpretation, when the solvent
is changed to THF, a good hydrogen-bond acceptor (HBA), the kA is no longer dependent on [B]. Electron donor–
acceptor (EDA) molecular complexes of 2,4-DNClB with benzene and with aniline, and also between aniline and
the product, were clearly recognized and the stability constants of each one of these complexes were determined.
Table 1 Reaction of 2,4-dinitrochlorobenzene (2,4-DNClB)a with
Introduction
aniline in hexane–benzene (4:1) at 40 ЊC. Pseudo-first-order, kψ, and
second-order, kA, rate coefficientsa
The ever-increasing knowledge of the structures of aggregates
in organic solvents of low permittivity is stimulating research
into the role that the different possible aggregates play in
defining the reaction mechanisms in solution.1,2 Nucleophilic
aromatic substitutions by amines in non-polar aprotic solvents
are especially sensitive to the degree of aggregation of both the
substrate and the nucleophile, and the role of these complexes is
increasingly being recognized.3
[Aniline]/M
kψ/10Ϫ6 sϪ1
kA/10Ϫ6 MϪ1 sϪ1
0.5
1.0
1.5
2.0
0.92
4.6
9.5
1.84
4.6
6.3
8.5
17
a [2,4-DNClB] = 2 × 10Ϫ4 M.
The mechanism by which SNAr by amines in protic solvents
occurs is well established, but the mechanism in aprotic solvents
is still controversial.4 Due to the dipolar character of the
reactants and the zwitterionic structure of the reaction inter-
mediate, in a solvent of low permittivity several types of
“molecular complexes” could, in principle, be formed and, in
fact, most of them have been observed, each one under special
reaction conditions (throughout this work, as in previous
papers, the generic term “molecular complex” (MC) would
involve any type of non-covalent interactions, e.g. charge-
transfer, hydrogen-bonding, electron donor–acceptor (EDA)
complexes, etc.). One of the most fundamental pieces of
evidence for the role of MCs in the mechanism of SNAr by
amines is the observation of a third order in amine, which was
reported for the first time in 1981.5 This feature was inter-
preted as being due to the dimer of the amine (a hydrogen-
bonded molecular complex) acting as the true nucleophile
and additional observations such as negative entropies of
activation, inverse solvent effects, effects of hydrogen-bond
acceptor (HBA) additives, etc. supported this interpretation.6,7
Further research on SNAr by amines in non-polar aprotic
solvents reported the observation of EDA complexes between
the nitro-activated substrates and benzene (which was used as
the solvent);8,9 of “mono-heteroconjugates” between the nucleo-
phile and a HBA additive;10,11 of a MC between the nitro-acti-
vated substrate and the nucleophile;12–14 of a MC between the
nucleophile and any other tertiary non-nucleophilic amine;7,15–17
and, recently, a charge-transfer complex between 1,3,5-trinitro-
benzene and indole-3-carboxylate has been isolated.18
The present paper describes a kinetic and spectroscopic
study of the reactions between 2,4-dinitrochlorobenzene (2,4-
DNClB) and aniline in aromatic pure solvents and in binary
mixtures of any of these solvents with n-hexane. Formation of
various types of electron donor–acceptor complexes and also
of hydrogen bonded aggregates was clearly recognized and the
stability constants of several of those MCs were determined.
The observed results afford new evidence concerning the critical
role that complexation plays in the kinetic behaviour of these
reactions.
Results and discussion
Reactions in aromatic solvent and n-hexane–aromatic solvent
mixtures. Molecular complex formation
The kinetics of the reaction of aniline with 2,4-dinitrochloro-
benzene were studied in benzene and in a binary solvent n-
hexane–benzene (Hx–Bz) at 40 ЊC in the presence of variable
amounts of the nucleophile under pseudo-first order con-
ditions. The pseudo-first, kψ and the second order rate co-
efficients, kA, of the reaction of 2,4-dinitrochlorobenzene
with aniline in Hx:Bz (4:1) are shown in Table 1. The plot
of kA vs. [B] exhibits a linear response to the nucleophile con-
centration, B.
A linear dependence on [B] is characteristic of most base-
catalyzed reactions, it has been widely observed in SNAr with
J. Chem. Soc., Perkin Trans. 2, 1999, 1627–1630
1627

View Article Online
Fig. 1 Overall second-order rate coefficients, kA, for the reaction
of 2,4-dinitrochlorobenzene with aniline in hexane–benzene (4:1) at
40 ЊC.
Fig. 2 “Excess” optical densities of the reaction mixtures of 2,4-
dinitrochlorobenzene with aniline at 450 nm, extrapolated at t = 0 as a
function of aniline concentration. ᭿ hexane–benzene (4:1), ᭹ benzene.
Table 2 Reaction of 2,4-dinitrochlorobenzene, 2,4-DNClB,a with
aniline in binary solvents hexane–benzene at 40 ЊC. Second-order rate
coefficients kA/10Ϫ6 sϪ1 MϪ1, at different aniline concentrations
explained by the specific solvent effect, which solvates prefer-
entially the aromatic reactant, by virtue of its electron donor
properties, through EDA complex formation. The formation
of EDA complexes has been earlier considered either as a
side-reaction21 or as a pre-equilibrium to the formation of the
σ-intermediate.8,22 The nitro-substituted aromatic substrate
is a typical π-electron acceptor and benzene can be a good
π-electron donor. The stability constant between benzene and
2,4-dinitrochlorobenzene, KD, has been reported to be
0.076 0.02.23
The lack of additivity in the absorbances of the reaction
mixtures extrapolated at zero time, observed in the present
work, suggests formation of EDA complexes also between the
aromatic substrate and the amine. Fig. 2 shows the values of
absorbances vs. [aniline] at t = 0. To examine the possible
formation of an EDA complex prior to substitution, the stab-
ility constants, KS, between 2,4-DNClB and aniline were cal-
culated by a method previously described9,24 and the values are
0.066 0.010 in n-hexane–benzene (4:1) and 0.035 0.01 in
pure benzene. In view of the observed result, it is proposed that,
for reactions in a solvent mixture in which a complexing agent is
present (i.e. benzene), the substrate may undergo two parallel
reactions: a) proceeding from the free aromatic substrate, or b)
through an EDA complex with the donor solvent, D; a com-
petition between the nucleophile (monomer and dimer) and the
donor solvent as complexing agents may be established in the
solvent mixtures. This situation is depicted in Scheme 1 where
[Aniline]
% Hexane
0.5 M
1.0 M
2.0 M
0
20
40
50
60
80
0.36
0.66
0.88
0.86
1.08
1.84
0.71
0.85
1.50
1.80
2.40
4.60
1.25
1.75
2.50
3.25
4.35
8.50
a [2,4-DNClB] = 2 × 10Ϫ4 M.
amines when the second step is rate determining, the second
order reaction rate coefficients obey eqn. (1).3,4,19
k1k
2 ϩ
kϪ1
k1k3 [B]
(1)
kA =
kϪ1
The intercept represents the uncatalysed reaction and the
slope measures the base catalysis. In some cases, with very poor
nucleofugues in aprotic solvents, only the base catalysed
mechanism is observed.3 Nevertheless, when chlorine is the
nucleofugue, usually formation of the zwitterionic inter-
mediate, k1, is the rds and no base catalysis is observed, but
instead a spontaneous decomposition of the intermediate to
products is seen.4
KS
In the present case, taking into account the non-polar char-
acter of n-hexane (π* = Ϫ0.088),20 aggregation of the nucleo-
phile is expected and a mechanism in which the nucleophile is
the dimer of the amine is proposed. In the plot of the second
order rate coefficient, kA, as a function of [aniline], (Fig. 1) a null
intercept is found. A kinetic law consistent with the observed
results is eqn. (2), where K = [B:B]/[Bo]2 represents the amine
auto-association. The null intercept observed with this good
nucleofugue, together with the linear dependence of kA on [B],
clearly indicates that eqn. (2) applies to the present case.
2,4-DNClB
+
B:B
[2,4-DNClB•B:B]
KD + D (donor solvent)
kU
kC
[2,4-DNClB•D]
Product
B:B
Scheme 1
kU and kC represent the pseudo-first order rate coefficients for
the free substrate (uncomplexed) and the complexed substrate,
respectively, and D represents the complexing solvent. A similar
scheme, but with a monomer nucleophile, has been reported by
Silber and co-workers for the reaction of dinitrobenzene with
n-butylamine.25 Complexation with the monomer nucleophile is
also possible under the present reaction conditions but, since
reaction with the dimer is faster, it is omitted in Scheme 1 for
the sake of simplicity.
An inhibitory effect of the donor solvent is observed when-
ever the substrate–solvent complex constant is greater than the
substrate–nucleophile stability constant. Under the experi-
mental conditions the absorbances at “infinite” reaction time
are higher than expected from standard solutions of N-(2,4-
dinitrophenyl)aniline and aniline, assuming a quantitative
k1k2 K [B]
(2)
kA =
kϪ1
The reaction was then carried out in other Hx–Bz mixtures
at three aniline concentrations, and the results are shown in
Table 2. The reactions are less sensitive to [B] in the presence of
aromatic solvent, and the null intercept is again observed. Only
when the hexane content is high (80%) are the rates comparable
to pure hexane. The plot of kA vs. % hexane (not shown)
exhibits an upward curvature which strongly increases with [B].
The different behavior of the reactions in the presence of
an aromatic solvent in comparison with pure hexane can be
1628
J. Chem. Soc., Perkin Trans. 2, 1999, 1627–1630

View Article Online
Table 3 “Excess” in the absorbances of the mixture of N-(dinitro-
phenyl)aniline, DNPA,a with aniline in benzene and in binary solvents
hexane–benzene (Hx–Bz). Stability constants of the complexes, Ks, and
extinction coefficients, ε (in brackets)
Absorbance
[Aniline]
Hx–Bz (4:1)
Hx–Bz (1:1)
Bz
0.00
0.000
0.051
0.085
0.113
0.143
0.173
0.192
0.224
0.242
0.000
0.015
0.033
0.057
0.078
0.099
0.116
0.136
0.150
0.000
0.011
0.039
0.041
0.060
0.080
0.085
0.099
0.110
0.248
0.500
0.748
1.000
1.248
1.500
1.748
2.000
Fig. 4 Solvent effects on the second-order rate coefficients, kA, for the
reaction of 2,4-dinitrochlorobenzene with 2 M aniline in THF–toluene
as a function of ET(30) values (see ref. 26).
Ks
0.257 0.02 (3553) 0.130 0.03 (3500) 0.109 0.01 (3090)
a [DNPA] = 2 × 10Ϫ4 M, wavelength = 450 nm.
Fig. 5 “Excess” optical densities of the reaction mixtures of 2,4-
dinitrochlorobenzene with aniline at 450 nm, extrapolated at t = 0, as a
function of [aniline] in: ᭺ THF, ᭿ THF–toluene (1:1) and ᭡ toluene at
40 ЊC.
Fig. 3 Pseudo-first-order, kψ, and overall second-order, kA, rate co-
efficients for the reaction of 2,4-dinitrochlorobenzene with aniline in
tetrahydrofurane at 40 ЊC. ᭹ kψ (sϪ1), ᮀ kA (sϪ1 MϪ1).
In THF–toluene mixtures an increase in the rate was
observed as a function of the THF content in the mixture.
The plot of kA vs. ET(30)26 (Fig. 4) is not linear, the upward
curvature is consistent with an effect of THF greater than that
expected from its polarity effect. Cooperation in the decom-
position of the zwitterionic intermediate due to their known
HBA properties is superimposed to the polarity effect. Consist-
ent with this interpretation is the linear dependence of the
second order rate coefficient with [THF]; this dependence
indicates the significance of a [THF] term in the kinetic law
related to the contribution of this HBA solvent in the second
step of the reaction.
The absorbances of the reaction mixtures (measured at
λ = 450 nm, where the reactants are transparent) extrapolated
at t = 0 show a linear increase with the [aniline] in the three
solvents studied (THF, THF–toluene (1:1) and toluene) (see
Fig. 5). The null intercept and the excellent linear corre-
spondence in the three solvent studied are also consistent
with the formation of EDA complexes prior to the aromatic
nucleophilic substitution.
conversion of 2,4-dinitrochlorobenzene to product (see data
in Table 3). An almost linear dependence (not shown) was
observed between the “excess” in absorbances of the product
and the [aniline] indicating formation of MC. The stability con-
stants, KS, between N-(2,4-dinitrophenyl)aniline and aniline
were determined in Hx–Bz (4:1), (1:1) and pure benzene. The
data gathered in Table 3 show that the Ks values are higher than
those with DNClB and increase with the hexane proportion in
the solvent, conditions that favour EDA complex formation.
The EDA complexes between the reaction products and aniline
do not affect the reaction mechanism, although they must be
included in the rate coefficient calculation (eqn. (3)).
k1k2 K Ks [B]
(3)
kA =
kϪ1
Reactions in electron-donor solvents
Likewise, the reactions of 2,4-dinitrochlorobenzene with aniline
were carried out in tetrahydrofuran (THF) and in binary
solvent mixtures toluene–THF at 40 ЊC. The reactions in THF
show a peculiar kinetic behavior in which dependence of the
second-order rate coefficients, kA, on the amine concentration is
not observed. The results are shown in Fig. 3. It can be
observed that kA slightly diminishes on increasing [aniline] in
the range 0.1–0.5 M and then remains almost constant. The
pseudo-first order rate coefficients are linearly related to
[aniline] in the whole range studied. These results can be
interpreted as competition between THF and aniline to be the
proton acceptor at low [aniline] up to a value with higher THF
contents after which the rate is insensitive to [B].
Experimental
Reagents and solvents
Aniline (Fluka) was kept overnight over potassium hydroxide,
distilled over zinc powder and then over sodium; both distil-
lations were carried out under nitrogen at reduced pressure.
2,4-Dinitrochlorobenzene (Sigma) was crystallized twice from
absolute ethanol (mp 52–53 ЊC, lit. 52–53 ЊC).16 Toluene,
benzene and hexane were kept over sodium wire for several
days, and distilled twice over sodium. Tetrahydrofuran was
treated successively with sodium sulfite, sodium hydroxide and
J. Chem. Soc., Perkin Trans. 2, 1999, 1627–1630
1629

View Article Online
sodium wire and then distilled, the fraction of bp 64 ЊC was
collected. N-(2,4-Dinitrophenyl)aniline (mp 155–156 ЊC, lit.
155–156 ЊC)27 was prepared from 2,4-dinitrochorobenzene and
aniline following the procedure reported for N-(2,4-dinitro-
phenyl)-2-methoxyaniline16 Solvents were stored in special
vessels which allow delivery without air contamination.
References
1 D. H. Williams and M. S. Westwell, Chem. Soc. Rev., 1998, 27, 57.
2 P. Hobza and R. Zahradník, Intermolecular Complexes, Elsevier,
Amsterdam, 1988, Ch. I, p. 25.
3 N. S. Nudelman, in The Chemistry of Amino, Nitroso, Nitro
and Related Groups, 1996, Ch. 26, Ed. S. Patai, J. Wiley & Sons,
London.
4 F. Terrier, Nucleophilic Aromatic Displacement: The influence of
the nitro group, in Organic Nitro Chemistry Series, Ed. H. Ferrer,
VCH Publishers, New York, 1991.
5 N. S. Nudelman and D. Palleros, Acta Sud. Am. Quim., 1981, 1, 125.
6 N. S. Nudelman, J. Phys. Org. Chem., 1989, 2, 1.
7 N. S. Nudelman, C. E. S. Alvaro and J. S. Yankelevich, J. Chem.
Soc., Perkin Trans. 2, 1997, 2125 and references therein.
8 E. Durantini, L. Zingaretti, J. D. Anunziata and J. J. Silber, J. Phys.
Org. Chem., 1992, 5, 557.
9 J. D. Anunziata, J. O. Singh and J. J. Silber, Can. J. Chem., 1981, 38,
505.
10 J. Hirst, J. Phys. Org. Chem., 1995, 7, 68 and references therein.
11 D. Palleros and N. S. Nudelman, J. Chem. Soc., Perkin Trans. 2,
1985, 479.
12 L. Forlani, J. Chem. Res (S), 1984, 260.
13 R. Cattana, J. O. Singh, J. D. Anunziata and J. J. Silber, J. Chem.
Soc., Perkin Trans. 2, 1987, 79.
14 L. Forlani, M. Sintoni and P. E. Todesco, J. Chem. Res. (S), 1986,
66, 344.
15 J. Hirst, G. N. Onuoha and I. Onyido, J. Chem. Soc., Perkin Trans. 2,
1988, 971.
16 (a) N. S. Nudelman, M. Marder and A. Gurevich, J. Chem. Soc.,
Perkin Trans. 2, 1993, 229; (b) N. S. Nudelman and D. R. Palleros,
J. Org. Chem., 1983, 48, 1613.
17 L. Forlani, The Chemistry of Amino, Nitroso, Nitro and Related
Groups, 1996, Ch. 10, Ed. S. Patai, J. Wiley & Sons, London.
18 P. Sepulcri, R. Goumont, J.-C. Hallé, E. Buncel and F. Terrier,
Chem. Commun., 1997, 789.
19 C. Bernasconi and R. H. Rossi, J. Org. Chem., 1976, 41, 44.
20 M. Kamlet, J. Abboud, M. Abraham and R. Taft, J. Org. Chem.,
1983, 48, 2877.
21 E. Buncel, A. Norris and K. Russel, Rev. Chem. Soc., 1962, 22,
123.
22 A. K. Colter and S. H. Hui, J. Org. Chem., 1968, 33, 1935.
23 L. Forlani and C. Cimarelli, J. Phys. Org. Chem., 1989, 2, 653.
24 J. O. Singh, J. D. Anunziata and J. J. Silber, Can. J. Chem., 1985, 63,
903.
25 S. Chiachiera, R. Cattana, J. O. Singh, J. D. Anunziata and J. J.
Silber, J. Phys. Org. Chem., 1989, 2, 631.
Ancillary spectrophotometric measurements
UV-VIS spectra of the substrate, the product, and of different
mixtures of both compounds with aniline at several concen-
trations were recorded in a Shimadzu spectrophotometer.
UV spectra of control solutions of the substitution product
in toluene containing different amounts of the amine were
recorded, and the extinction coefficients determined at λ = 450
nm where the reagents are transparent under these conditions.
All the solutions were found to obey Beer’s law.
Determination of EDA complex formation constants, KS. A
UV spectrophotometric method previously described was
used.24,25 The solutions, prepared in the concentration range
0.5–2.0 M, were completely stable. The absorbances of the
mixtures were measured at 450 nm. The increases in the
absorbances of the mixtures at 450 nm with respect to the sum
of the isolated components alone (“excess” in the absorbances)
are interpreted as being due to the formation of MC; KS were
calculated in every case by an iterative procedure developed by
Silber and co-workers.24,25
Kinetic procedures
Kinetic runs were performed by the methods previously
reported28 following the appearance of the reaction product at
λ = 450 nm. In all cases pseudo-first order kinetics were
observed. The pseudo-first order rate coefficients, kψ, were
obtained by the least-squares method as the slope of the corre-
lation ln (A∞ Ϫ At)/A∞ against time, where A∞ is the optical
density of the reaction mixture measured at “infinity” (more
than ten half-lives); the second order rate coefficients, kA, were
obtained by dividing kψ by the amine concentrations. Rate
coefficients were reproducible to 2%. No corrections for
expansion coefficients were applied to the concentration values.
26 P. M. E. Mancini, A. Terenzani, M. G. Gasparri and L. R. Vottero,
J. Phys. Org. Chem., 1995, 8, 617.
27 L. Forlani, Gazz. Chim. Ital., 1982, 112, 205.
28 N. S. Nudelman, M. Savino, V. Nicotra and J. S. Yankelevich,
Anales Asoc. Quim. Argent., 1997, 85, 231.
Acknowledgements
The authors are deeply indebted to Professor J. Silber (Univ.
Nacional de Rio Cuarto, Argentine) for her generous assistance
with the iterative procedure. Financial support from the
University of Comahue and from the National Research
Council (CONICET) from Argentine is gratefully acknow-
ledged.
Paper 9/03559A
1630
J. Chem. Soc., Perkin Trans. 2, 1999, 1627–1630