A mechanistic viewpoint of the reaction between 1-fluoro-2,6-dinitrobenzene and alicyclic amines in the ethyl acetate-chloroform system

Article abstract of DOI:10.1002/poc.869

A study was made of the influence of temperature on the reactions between 1-fluoro-2,6-dinitrobenzene and each of the secondary amines pyrrolidine, piperidine and homopiperidine (hexahydro-1H-azepine), carried out in ethyl acetate-chloroform binary solvent mixtures. It involved the analysis of both global activation parameters and the corresponding Arrhenius plots from the k_A values at three amine concentrations. These values were obtained by carrying out the reactions at 5, 15, 25, 40 and 50°C over the whole range of chloroform mole fraction. The analysis of Arrhenius lines allowed us to prove the existence of isokinetic relationships, which were used as a diagnostic tool in order to infer a changeover in the nature of the rate-determining transition state as a function of solvent composition. The experimental evidence together with theoretical quantum mechanical calculations suggest that the reactions with the secondary amines explored carried out in pure ethyl acetate and ethyl acetate-chloroform solvent mixture at X_CHCl3, = 0.1 and in some cases also at X_CHCl3 = 0.3 proceed via the formation of a six-membered orientated dipolar aggregate in which the specific base-general acid (SB-GA) mechanism may take place. The reactions carried out in the remaining solvent mixtures evolve towards the classical SB-GA mechanism. Copyright

Full text of DOI:10.1002/poc.869

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2005; 18: 336–346
Published online 5 October 2004 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.869
A mechanistic viewpoint of the reaction between
1-fluoro-2,6-dinitrobenzene and alicyclic amines
in the ethyl acetate–chloroform system
P. M. Mancini,* G. G. Fortunato and L. R. Vottero
´
´
´
´
´
´
Departamento de Quımica, Area Quımica Organica, Facultad de Ingenierıa Quımica, Universidad Nacional del Litoral,
Santiago del Estero 2829, (3000) Santa Fe, Argentina
Received 16 January 2004; revised 14 July 2004; accepted 27 July 2004
ABSTRACT: A study was made of the influence of temperature on the reactions between 1-fluoro-2,6-dinitrobenzene
and each of the secondary amines pyrrolidine, piperidine and homopiperidine (hexahydro-1H-azepine), carried out in
ethyl acetate–chloroform binary solvent mixtures. It involved the analysis of both global activation parameters and the
corresponding Arrhenius plots from the kA values at three amine concentrations. These values were obtained by
carrying out the reactions at 5, 15, 25, 40 and 50 ꢀC over the whole range of chloroform mole fraction. The analysis of
Arrhenius lines allowed us to prove the existence of isokinetic relationships, which were used as a diagnostic tool in
order to infer a changeover in the nature of the rate-determining transition state as a function of solvent composition.
The experimental evidence together with theoretical quantum mechanical calculations suggest that the reactions with
the secondary amines explored carried out in pure ethyl acetate and ethyl acetate–chloroform solvent mixture at
XCHCl ¼ 0.1 and in some cases also at XCHCl ¼ 0.3 proceed via the formation of a six-membered orientated dipolar
3
3
aggregate in which the specific base–general acid (SB–GA) mechanism may take place. The reactions carried out in
the remaining solvent mixtures evolve towards the classical SB–GA mechanism. Copyright # 2004 John Wiley &
Sons, Ltd.
Supplementary electronic material for this paper is available in Wiley Interscience at http://www.interscience.
wiley.com/jpages/0894-3230/suppmat/
KEYWORDS: aromatic nucleophilic substitution; solvent effects; temperature effects; isokinetic relationships; mechanism
INTRODUCTION
nucleofuge. It was also concluded that solvation effects
are dominated by non-specific interactions and that the
order of incidence of the molecular-microscopic solvent
properties on the kA second-order rate coefficient is
dipolarity/polarizability > hydrogen-bond donor ability
(HBD) > hydrogen-bond acceptor ability (HBA).
In order to obtain a deeper insight into the kinetic
behavior of the reactions between 2,6-DNFB and the
secondary alicyclic amines PYR, PIP and HPIP, we
analyzed the temperature effect on these reactive sys-
tems. In this direction, the reactions were carried out at
five temperatures, ethyl acetate (EAc)–cosolvent CHCl3
being the binary solvent system selected. Additionally,
from Arrhenius plots obtained for a group of reactions
carried out at different temperatures with varying solvent
composition, we looked for the existence of an isokinetic
relationship. We used this extra-thermodynamic tool in
order to infer a possible change in the reaction mechan-
ism of the reactive systems explored.
We have recently studied the kinetics of nucleophilic
aromatic substitution (SNAr) reactions between 1-fluoro-
2,6-dinitrobenzene (2,6-DNFB) and each of the second-
ary amines homopiperidine (HPIP, hexahydro-1H-aze-
pine), pyrrolidine (PYR) and piperidine (PIP) in ethyl
acetate–chloroform or acetonitrile and acetonitrile–
chloroform binary solvent mixtures.1 Those studies eval-
uated the influence of the nucleophile structure and
solvent effects on the corresponding reactive systems. It
was concluded that the amine structure has a great
influence on second-order rate constants, especially on
partial rate constants related to the catalyzed step. The-
oretical quantum mechanical calculations confirmed that
the origin of these results lies in stereoelectronic effects
due to the conformational difference between the amino
moieties in the ꢀ intermediates as they release the
´
*Correspondence to: P. M. Mancini, Departamento de Quımica, Area
´
´
´
´
´
Quımica Organica, Facultad de Ingenierıa Quımica, Universidad Na-
cional del Litoral, Santiago del Estero 2829, (3000) Santa Fe, Argentina.
E-mail: pmancini@fiqus.unl.edu.ar
RESULTS AND DISCUSSION
Contract/grant sponsor: Science and Technology Secretariat, Uni-
versidad Nacional del Litoral; Contract/grant numbers: 2000-17-
151; 2002-21-152; 2002-21-153.
The kinetics of the SNAr reactions between 2,6-DNFB
and PYR, PIP and HPIP in the EAc–CHCl3 solvent
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 336–346

REACTION MECHANISM BETWEEN 1-FLUORO-2,6-DINITROBENZENE AND ALICYCLIC AMINES
337
system was determined at 5, 15, 40 and 50 ꢀC. The results
corresponding to 25 ꢀC have been reported elsewhere.1
Each reaction was explored at different solvent com-
positions and the influence of the amine concentration
was studied in all cases. The reactions were carried out
under pseudo-first-order conditions, i.e. with a large
excess of the secondary amine. They yielded the expected
products in quantitative yield, N-(2,6-dinitrophenyl)pyr-
rolidine, N-(2,6-dinitrophenyl)piperidine and N-(2,6-di-
nitrophenyl)homopiperidine, and proved to be first order
in the corresponding substrate. Second-order rate con-
stants kA were calculated by multiplying the observed
pseudo-first-order rate constants, k’, by the appropriate
amine concentration, kA ¼ k’ [B].
plateau at high [B], indicating that the formation of
ZH is rate limiting.
In polar aprotic solvents, the specific base–general acid
(SB–GA) mechanism for the decomposition of ZH has
been well established.2 Scheme 2(a) illustrates the differ-
ent paths: the formation of ZH, a rapid deprotonation
equilibrium of this intermediate, followed by rate-limiting,
general acid-catalyzed nucleofuge departure from the
anionic ꢀ complex Zꢂ, which evolves to products. The
derived kA expression for this mechanism is
k1k2 þ k1k4BHK3½Bꢁ
kA ¼
ð3Þ
kꢂ1 þ k2 þ k4BHK3½Bꢁ
We assume that the reaction proceeds by the classical
two-step mechanism shown in Scheme 1 (where
S ¼ substrate, B ¼ amine, ZH ¼ zwitterionic intermediate
and P ¼ product). Application of the steady-state approx-
imation for the concentration of ZH, d[ZH]/d[t] ¼ 0,
yields the following equation for the second-order reac-
tion rate constant kA:
BH
where k4 is the rate coefficient for acid-catalyzed
expulsion of the leaving group and K3 is the deprotona-
tion equilibrium constant.
Kinetic results
À
Á
k1 k2 þ k3B½Bꢁ
Reactions with PYR. Table S1 (Electronic Supplemen-
tary Information, ESI; available in Wiley-Interscience)
presents the kA values for the reactions between 2,6-
DNFB and all nucleophiles performed at 5, 15, 40 and
50 ꢀC in the explored mixtures. The data in the pure
solvents are additionally presented. The corresponding
results for the reactions with PYR follow the expected
tendency: the kA values increase as the working tempera-
ture increases. As was observed at 25 ꢀC, the kinetic
results reveal a satisfactory linear dependence of the
rate on amine concentration, showing a null intercept
(Table S2 and S3, ESI). As a consequence, the reactions
are clearly base catalyzed.
kA ¼
ð1Þ
kꢂ1 þ k2 þ k3B½Bꢁ
Three main situations of interest with respect to the
reaction shown in Eqn (1) were considered:
B
(a) kꢂ1 ꢃ k2 þ k3 [B]. In this case no base catalysis
is possible, Eqn (1) simplifies to kA ¼ k1 and the
formation of the zwitterionic intermediate is rate
limiting.
(b) kꢂ1 ꢄ k2 þ k3B [B]. This situation indicates the rapid
formation of ZH followed by its rate-determining
decomposition. Eqn (1) reduces to
kA ¼ k1k2=kꢂ1 þ k1k3=kꢂ1½Bꢁ
ð2Þ
Reactions with PIP. As can be appreciated from the
data, the kA values corresponding to the reactions carried
out in pure EAc and in the EAc-rich mixtures vary
slightly with increasing temperature. Moreover, in some
mixtures the values at 40 and 50 ꢀC are smaller than those
at 25, 15 and 5 ꢀC.
which predicts base catalysis with a linear depen-
dence of kA on [B].
(c) kꢂ1 ꢅ k2 þ k3B [B]. In this situation, Eqn (1) indicates
that base catalysis is observed with a curvilinear
dependence of kA on [B]. At low [B], the plot of kA
vs [B] should be a straight line which will change to a
As in the case of the reactions with PYR, in all
instances the second-order rate constant kA varies linearly
Scheme 1
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 336–346

338
P. M. MANCINI, G. G. FORTUNATO AND L. R. VOTTERO
and from ꢂ272.9 to ꢂ192.6 J molꢂ1 Kꢂ1 for the reactions
with PYR, PIP and HPIP, respectively.
Isokinetic relationships (IKR)
In order to assess the possibility that a specific group of
reactions carried out in the different solvent mixtures may
be included in an isokinetic relationship (IKR), the
corresponding Arrhenius lines were plotted. An IKR
refers to a common point (or small area) of intersection
of Arrhenius lines.3 This would imply that the reactions
forming part of the group are similar, constituting a
‘series of reactions’, and that the reactions go through
an equivalent mechanistic pathway. The occurrence or
non-occurrence of a common intersection point of the
respective lines (even when non-linear behavior is ob-
served3) allows us to confirm that possibility. In fact, a
common point of intersection of Arrhenius lines is
mathematically equivalent to a linear relationship be-
tween activation energies and pre-exponential factors or
between enthalpy and entropy terms; thus
Scheme 2
ꢀH# ¼ ꢁꢀS# þ c
ð4Þ
with the amine concentration, exhibiting a null intercept
(Tables S4 and S5, ESI).
where ꢁ ¼ Tiso (isokinetic temperature; K) and c ¼
constant ¼ ꢀG# .
ꢁ
Reactions with HPIP. In pure EAc and in the EAc-rich
zone, this reactive system exhibits a kinetic behavior
similar to that of the reactions with PIP. For the whole
range of temperatures, the kA values vary linearly with
the amine concentration, exhibiting a null intercept in the
HBD solvent-poor mixtures (Tables S6 and S7, ESI). The
incidence of the base catalysis decreases with the CHCl3
mole fraction.
However, this mathematical equivalence does not al-
ways hold from a statistical point of view because
experimental uncertainties of the derived parameters
(e.g. EA and log A or ꢀH# and ꢀS#) can all too often
lead to an artificial linear dependence without a signifi-
cant meaning.4 Therefore, a common point of intersec-
tion allows one to infer a linear relationship between ꢀH#
and ꢀS# but the opposite does not hold, i.e. if the
corresponding plot renders a linear relationship, this
does not necessarily imply that all the reactions constitute
a single series of reactions. Krug5 suggested that the plots
of ꢀH# vs ꢀGT#hm or ꢀGT#hm vs ꢀS# were statistically
more meaningful (Thm is the harmonic mean of the
experimental temperatures). These terms are related by
the following equations:
Activation parameters. The global activation para-
meters ꢀH#, ꢀS# and ꢀG# corresponding to reactions
with PYR, PIP and HPIP, respectively, are shown in
Table 1. Table S8 (ESI) presents the respective Arrhenius
parameters. These data were obtained from the kA values
at three different amine concentrations (0.03, 0.07 and
0.15 M) for the reactions of all the reactive systems
performed in the solvent mixtures selected.
ꢀH# ¼ ð1 ꢂ ꢂÞꢀGꢁ# þ ꢂꢀG#Thm
ð5Þ
ð6Þ
From the analysis of the respective data, the following
tendencies can be observed. In the reactions with all
amines, the ꢀH# and EA values are significantly low,
particularly in the reactions with PIP and HPIP, in which
these values are almost zero or negative when they are
carried out in pure EAc and in EAc-rich mixtures. The
ꢀG#Thm ¼ ꢀG# þ ðꢁ ꢂ ThmÞꢀS#
ꢁ
where ꢂ ¼ 1/(1 ꢂ Thm/ꢁ).
ꢀH# values vary between 5.243 and 21.17 kJ molꢂ1
,
Krug based the previous proposal on the fact that the
data plotted in the enthalpy–entropy plane appear to
display a linear compensation effect that is entirely due
to the linear propagation of measurement errors. On the
ꢂ6.041 and 20.38 kJ molꢂ1 and 0.714 and 24.35 kJ molꢂ1
for the reactions with PYR, PIP and HPIP, respectively.
The activation entropies exhibit large negative values.
This trend is particularly manifested in mixtures with low
cosolvent mole fraction and with PIP and HPIP as nucleo-
philes. The corresponding values vary from ꢂ203.4 to
ꢂ169.7 J molꢂ1 Kꢂ1, from ꢂ277.8 to ꢂ202.6 J molꢂ1 Kꢂ1
other hand, the data plotted in the ꢀH#–ꢀG#
plane
Thm
display structured variations due only to chemical effects
because the measurement errors propagate in a random
manner into this plane.
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 336–346

REACTION MECHANISM BETWEEN 1-FLUORO-2,6-DINITROBENZENE AND ALICYCLIC AMINES
339
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 336–346

340
P. M. MANCINI, G. G. FORTUNATO AND L. R. VOTTERO
Additionally, the existence of an IKR must be demon-
wise, the system corresponding to [PYR] ¼ 0.15 M is
strated statistically. In this direction, Linert proposed a
simplified statistical test.3 When this method yields a
positive IKR proof, it does so at such significance level
that any IKR found by this means also holds for the non-
simplified method.4
In this direction, Arrhenius lines were plotted for the
reactions with PYR, PIP and HPIP at three amine con-
centrations varying the cosolvent molar fraction. The
ꢀH#–ꢀS# and ꢀGT#hm–ꢀS# plots are also presented.
distinguished by a break between the two groups.
[PIP] groups. Figure 2 presents the Arrhenius lines
belonging to the reactions with PIP at the explored amine
concentrations. Figures S4, S5 and S6 (ESI) present
ꢀH#–ꢀS#, ꢀH#–ꢀGT#hm and ꢀGT#hm–ꢀS# plots for
each group of reactions.
As in the previous groups, the Arrhenius lines yield
two intersection points for the three amine concentra-
tions, indicating that the reactions conform to two reac-
tion series. While the ꢀH#–ꢀS# plot yields a linear
relationship, the corresponding ꢀH#–ꢀGT#hm and
ꢀGT#hm–ꢀS# plot reveals that two groups of reactions
are involved. Linert’s simplified statistical IKR-analysis
was applied to Group 2 for each amine concentration and
to Group 1 for [PIP] ¼ 0.15 M. The respective Tiso, kiso, ꢁ
and F-distributed values are shown in Table 3. The test
allowed us to accept the existence of an IKR at the 95%
confidence level in all cases.
For the reactions corresponding to [PIP] ¼ 0.03 M and
[PIP] ¼ 0.07 M the series are constituted as follows,
Group 1: XCoS ¼ 0.0, 0.1 and Group 2: XCoS ¼ 0.1, 0.3,
0.5, 0.7, 0.9, 1.0. The one carried out in solvent
XCoS ¼ 0.1 would represent an inflection point between
one Tiso and another. The reactions corresponding to
[PIP] ¼ 0.15 M are distributed as Group 1: XCoS ¼ 0.0,
0.1, 0.3 and Group 2: XCoS ¼ 0.3, 0.5, 0.7, 0.9, 1.0. As in
the previous case, the reaction at XCoS ¼ 0.3 is indicating
a change from one series to the other. It must be pointed
out that the Tiso (46.7 ꢀC) corresponding to Group 2 is
very close to the experimental one (50 ꢀC) which coin-
cides with the fact that the reaction rate is almost
constant.
[PYR] groups. Figure 1 presents the Arrhenius lines
belonging to reactions with PYR: [PYR] ¼ 0.03 M group,
[PYR] ¼ 0.07 M group and [PYR] ¼ 0.150 M group. ꢀH#–
ꢀS#, ꢀH#–ꢀGT#hm and ꢀGT#hm–ꢀS# plots for each group
of reactions are shown in Figs S1, S2 and S3 (ESI),
respectively.
The Arrhenius lines clearly yield two intersection
points in all cases, indicating in the first instance that
the reactions are included in two different groups. The
ꢀH#–ꢀGT#hm and ꢀGT#hm–ꢀS# plots also indicate an
equivalent tendency and from the linear treatment of
the respective straight lines it was possible to estimate
the ꢁ values (Table 2) according to Eqns (3) and (4). For
the [PYR] ¼ 0.03 and 0.07 M groups, the ꢀH#–ꢀS# plots
suggest that all the reactions constitute only one series but
the ꢀH#–ꢀGT#hm and ꢀGT#hm–ꢀS# plots confirm that they
are involved in two groups.
Linert’s simplified statistical IKR analysis was applied
to Group 2 for each amine concentration. The respective
Tiso, kiso and F-distributed values are shown in Table 2.
The test allowed us to accept the existence of an IKR at
the 95% confidence level in all cases.
As a consequence, the reactions can be distributed as
Group 1, XCoS ¼ 0.0, 0.1 and Group 2, XCoS ¼ 0.1, 0.3,
0.5, 0.7, 0.9, 1.0. For the reactions corresponding to
[PYR] ¼ 0.03 and 0.07 M, that carried out in the solvent
XCoS ¼ 0.1 would represent an inflection point. Other-
[HPIP] groups. Arrhenius lines belonging to the reac-
tions with HPIP at the explored amine concentrations are
presented in Figure 3. The ꢀH#–ꢀS#, ꢀH#–ꢀGT#hm and
Figure 1. Arrhenius lines for [PYR] ¼ 0.03, 0.07 and 0.15 M
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 336–346

REACTION MECHANISM BETWEEN 1-FLUORO-2,6-DINITROBENZENE AND ALICYCLIC AMINES
341
Table 2. kiso, Tiso, F and ꢁ values corresponding to the reactions with PYR
[Amine]
(M)
Groups:
solvent
mixtures
kiso
Tiso (K)
( ꢀC)
F0.05(f1, f2)
calculated
F0.05(f1, f2)
tabulated
ꢁ (K)
(l molꢂ1 sꢂ1
)
ꢀGT#hm–ꢀS#
ꢀH#–ꢀG#
178.3
0.03
0.07
0.150
Group 1:
XCoS ¼ 0.0, 0.1
Group 2:
XCoS ¼ 0.1, 0.3, 0.5,
0.7, 0.9, 1.0
Group 1:
0.57
26.9
175.8 (ꢂ97.2)
178.3
423.5 (159.5)
0.58(5,18)
0.59(5,18)
0.065(4,15)
2.77(5,18)
2.77(5,18)
3.06(4,15)
420.4 (r ¼ 0.987)
423.0 (r ¼ 0.993)
2.13
94.0
192.7 (ꢂ80.3)
193.7
193.3
XCoS ¼ 0.0, 0.1
Group 2:
492.8 (219.8)
496.4 (r ¼ 0.980)
504.9 (r ¼ 0.993)
XCoS ¼ 0.1, 0.3, 0.5,
0.7, 0.9, 1.0
Group 1:
3.62
62.3
167.7 (ꢂ105.3)
165.4
165.9
XCoS ¼ 0.0, 0.1
Group 2:
XCoS ¼ 0.3, 0.5, 0.7,
0.9, 1.0
421.4 (148.4)
422.0 (r ¼ 0.993)
426.2 (r ¼ 0.995)
Figure 2. Arrhenius lines for [PIP] ¼ 0.03, 0.07 and 0.15 M
Table 3. kiso, Tiso, F and ꢁ values corresponding to the reactions with PIP
[Amine]
(M)
Groups:
solvent
mixtures
kiso
Tiso
(K) ( ꢀC)
F0.05(f1, f2)
calculated
F0.05 (f1, f2)
tabulated
ꢁ (K)
(l molꢂ1 sꢂ1
)
ꢀGT#hm–ꢀS#
ꢀH#–ꢀG#
0.03
0.07
0.150
Group 1:
XCoS ¼ 0.0, 0.1
Group 2:
0.104
0.099
358.9 (85.9) 0.58(5,18)
326.6 (53.6) 0.29(5,17)
2.77(5,18)
2.81(5,17)
359
359
327 (r ¼ 0.998)
327 (r ¼ 0.999)
XCoS ¼ 0.1, 0.3, 0.5,
0.7, 0.9, 1.0
Group 1:
0.197
0.205
388.0 (115)
390.5
390.2
XCoS ¼ 0.0, 0.1
Group 2:
328.8 (55.8) 0.25(5,16)
2.85 (5,16)
328.9 (r ¼ 0.997) 329.1 (r ¼ 0.998)
XCoS ¼ 0.1, 0.3, 0.5,0.7,
0.9, 1.0
Group 1:
0.617
0.362
207.6 (ꢂ68.4) 0.26(2,6)
5.14(2,6)
206.6
206.6 (r ¼ 0.999)
XCoS ¼ 0.0, 0.1, 0.3
Group 2:
XCoS ¼ 0.3, 0.5,0.7,
0.9, 1.0
319.7 (46.7) 1.56(4,12)
3.26(4,12)
319.1 (r ¼ 0.990) 319.4 (r ¼ 0.990)
J. Phys. Org. Chem. 2005; 18: 336–346
Copyright # 2004 John Wiley & Sons, Ltd.

342
P. M. MANCINI, G. G. FORTUNATO AND L. R. VOTTERO
Figure 3. Arrhenius lines for [HPIP] ¼ 0.03, 0.07 and 0.15 M
ꢀGT#hm–ꢀS# plots for the reaction at [HPIP] ¼ 0.15 M are
presented in Figure 4. The plots corresponding to the
reactions at [HPIP] ¼ 0.03 M and [HPIP] ¼ 0.07 M are
presented in Figures S7 and S8 (ESI).
In all cases, Arrhenius plots clearly yield two intersec-
tion points indicating the occurrence of two reaction
series. This is confirmed by the ꢀH#–ꢀGT#hm and
ꢀGT#hm–ꢀS# plots and by the statistical test. Table 4
presents ꢁ values obtained from these plots and Tiso, kiso
and F-distributed values graphically obtained from the
Arrhenius lines and by means of the respective test. The
reactions at [HPIP] ¼ 0.03 M are involved in the following
groups: Group 1, XCoS ¼ 0.0, 0.1, 0.3 and Group 2,
XCoS ¼ 0.3, 0.5, 0.7, 0.9, 1.0, the reaction at XCoS ¼ 0.3
indicating a change from one series to the other. Reac-
tions of Group 1 exhibit a Tiso value (35.7 ꢀC) almost
equivalent to the experimental value, indicating that the
reaction rate at that temperature is not influenced by the
solvent. Group 2 reactions present a Tiso value (14.2 ꢀC)
which also belongs to the experimental temperature range,
with the following consequences: (i) the kA second-order
rate coefficients corresponding to the reactions carried
out at 15 ꢀC exhibit little change as the CHCl3 mole
fraction increases; (ii) at that temperature, the solvent
effect is reversed: the reaction rates at 25, 40 and 50 ꢀC
increase with increase in CHCl3 mole fraction whereas
those at 5 ꢀC do the opposite.
With respect to the reactions corresponding to
[HPIP] ¼ 0.07 and 0.15 M, they are distributed in the
same groups as the previous amine concentration:
Group 1, XCoS ¼ 0.0, 0.1, 0.3 and Group 2, XCoS ¼ 0.3,
0.5, 0.7, 0.9, 1.0, the reaction at XCoS ¼ 0.3 also indicat-
ing a change from one series to the other. The Tiso values
corresponding to Group 1 from [HPIP] ¼ 0.07 M and
0.15 M belong to the experimental range. For the
[HPIP] ¼ 0.07 M group, Tiso is ꢇ25 ꢀC so that reaction
rates at this temperature are not influenced to a great
extent by the change in solvent composition. Otherwise,
the reaction rates at 5 and 15 ꢀC decrease with increase
in CHCl3 mole fraction whereas the reactions carried
out at 40 and 50 ꢀC exhibit an increase in reaction rates.
For the [HPIP] ¼ 0.15 M group, Tiso is 37 ꢀC. Therefore, it
can be observed that below this temperature, reaction
rates decrease with increase in CHCl3 mole fraction and
above that temperature they do the opposite, although
slightly.
The above results suggest the occurrence of a change in
the nature of the mechanistic pathway for all the amines
explored as a function of solvent composition.
Figure 4. ꢀH#–ꢀS#, ꢀH#–ꢀG# and ꢀG#–ꢀS# plots for the
kA [HPIP] ¼ 0.15 M group
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 336–346

REACTION MECHANISM BETWEEN 1-FLUORO-2,6-DINITROBENZENE AND ALICYCLIC AMINES
343
Table 4. kiso, Tiso, F and ꢁ values corresponding to the reactions with HPIP
[Amine]
(M)
Groups:
solvent
mixtures
kiso
Tiso
(K) ( ꢀC)
F0.05(f1, f2)
calculated
F0.05 (f1, f2)
tabulated
ꢁ (K)
(l molꢂ1 sꢂ1
)
ꢀGT#hm–ꢀS#
ꢀH#–ꢀG#
0.03
0.07
0.150
Group 1:
XCoS ¼ 0.0, 0.1, 0.3
Group 2:
0.027
0.020
308.7 (35.7)
0.34(2,6)
5.14(2,6)
309.7 (r ¼ 0.996)
287.8 (r ¼ 0.973)
309.7 (r ¼ 0.996)
287.2 (r ¼ 0.971)
287.2 (14.2) 0.03(4,15)
3.06(4,15)
XCoS ¼ 0.3, 0.5, 0.7,
0.9, 1.0
Group 1:
0.070
0.150
299.0 (26)
1.18(2,7)
4.74(2,7)
298.0 (r ¼ 0.958)
337.7 (r ¼ 0.998)
298.0 (r ¼ 0.958)
337.8 (r ¼ 0.999)
XCoS ¼ 0.0, 0.1, 0.3
Group 2:
337.8 (64.8) 0.05(4,15)
3.06(4,15)
XCoS ¼ 0.3, 0.5, 0.7,
0.9, 1.0
Group 1:
0.165
0.027
310.0 (37)
1.58(2,7)
0.34(2,6)
4.74(2,7)
5.14(2,6)
309.6 (r ¼ 0.985)
309.7 (r ¼ 0.996)
309.7 (r ¼ 0.986)
309.7 (r ¼ 0.996)
XCoS ¼ 0.0, 0.1, 0.3
Group 2:
XCoS ¼ 0.3, 0.5, 0.7,
0.9, 1.0
308.7 (35.7)
Mechanistic aspects
ate is formed straightforwardly in the addition step
through the dimer of the amine. This intermediate is in
mobile equilibrium with the classical intermediate and
either of them can react to form ultimate products, by
spontaneous or base-catalyzed decomposition. A simpli-
fied reacting scheme, where only attack of the dimer is
considered, is shown in Scheme 2(b). The dimer of the
nucleophile (B:B) attacks the substrate, S, forming the
intermediate, SB2, and a third molecule of amine assists
the decomposition step. The derived expression for kA in
this simplified reacting scheme is
In pure EAc and in its richest mixtures, the reactions are
highly catalyzed over the whole range of amine concen-
tration at all temperatures and the zwitterionic intermedi-
ate decomposition can be considered as the slow step.
The reactions are characterized as exhibiting low activa-
tion energies and enthalpies and for the reactions with
PIP and HPIP they are practically zero and in some cases
negative. The reactions are also characterized by large
negative entropies of activation. The kA values remain
nearly constant and in some instances they decrease
slightly with increasing temperature.
With respect to the above results it can be considered
that: (i) total third-order reactions are characterized by
low activation energies, large negative entropies of acti-
vation and rates exhibiting little change with increasing
temperature;6 (ii) these parameter magnitudes were found
in the mechanism involving the formation of EDA-type
complexes;7 (iii) they characterize reactions that are
second order in amine, involving a six-membered cyclic
transition state with two molecules of amine, with kA
values varying slightly as the reaction temperature is
increased;8,9 (iv) null activation energies, negative ꢀH#
and large negative ꢀS# were also found in reaction
mechanisms third order in amine: dimer mechanism,10
cyclic transition state involving eight-membered rings.11
Moreover, in these mechanisms the reaction rates were
found to increase slightly with increasing temperature
and in some instances reaction rates decrease with in-
creasing temperature.
2
k1k2K½Bꢁ þ k1k3K½Bꢁ
kA ¼
ð7Þ
kꢂ1 þ k2 þ k3½Bꢁ
Plots of kA vs [B] exhibit an upward curvature and when
ꢂ1 ꢄ k2 þ k3 [B] the previous equation can be simplified
to
k
kA=½Bꢁ ¼ k1k2=kꢂ1 þ k1k3=kꢂ1½Bꢁ
ð8Þ
which predicts a linear dependence of kA/[B] vs [B].
When kꢂ1ꢅ k2 þ k3 [B], at high [B], Eqn (7) can be
transformed into the equation kA/[B] ¼ k1K, which is
responsible for the plateau observed in some cases in
the plot of kA/[B] vs [B].10
In reference to our reactive systems, reactions with all
the nucleophiles for the explored amine concentrations
do not exhibit the characteristic upward curvature over
the whole range of temperatures. Otherwise, the plots of
kA/[B] vs [B] do not afford straight lines.
The dimer nucleophile mechanism was proposed by
Nudelman and Palleros.10 This mechanism exhibits a
third-order dependence on the amine involving the attack
of the dimer of the nucleophile superimposed on the
classical reaction with the monomer. A cyclic intermedi-
The cyclic transition state involving eight-membered
rings mechanism was proposed by Banjoko and Ezeani.11a
This mechanism also involves a quadratic dependence of
kA on [B] and the formation of a cyclic intermediate. The
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 336–346

344
P. M. MANCINI, G. G. FORTUNATO AND L. R. VOTTERO
authors found that in non-polar aprotic solvents nucleo-
philic aromatic substitution involving primary amines
and substrates with fairly poor leaving groups will
proceed predominantly by a catalyzed path through a
cyclic transition state involving an eight-membered ring.
This structure is formed through a network of the inter-
hydrogen bonding between the zwitterionic intermediate
and two amine molecules. A simplified scheme of this
mechanism is shown in Scheme 2(c). The first step
involves the formation of ZH. This intermediate reacts
with two amine molecules to yield the eight-membered
cyclic intermediate [SB3], which evolves to products. The
resulting expression for kA is
Figure 5. Six-membered cyclic intermediate structure
orientated dipolar aggregate between the zwitterionic
intermediate and the nucleophile and (ii) when the
nucleophile is a secondary amine, the breaking of the
hydrogen bond with the ortho-nitro group will occur
within the aggregate followed by proton exchange and
the electrophilically catalyzed breaking of the C—F
bond, i.e. it can be considered that the SB–GA mechan-
ism takes place within the aggregate. Notwithstanding
what the evolution to products is like, whether it proceed
in a concerted or non-concerted way, the basic steps for
this mechanism are illustrated in Scheme 2(d). The first
step involves the formation of ZH. This intermediate
reacts with another amine molecule to yield the six-
membered cyclic intermediate [SB2], which evolves to
products. The proposed structure for this intermediate is
shown in Fig. 5. This mechanism8 generally exhibits total
third-order reactions showing a linear relationship be-
tween kA values and initial amine concentration with
negligible intercepts. The resulting kA expression would
be kA ¼ k1k2/kꢂ1 þ k1K3k4/kꢂ1 [B].
2
k1k2½Bꢁ þ k1k3½Bꢁ
kA ¼
ð9Þ
kꢂ1 þ k2 þ k3½Bꢁ
In reactions of amines with poor nucleofuges, the
second step is usually rate determining so kꢂ1 ꢄ
k2 þ k3 [B]2 and the above expression simplifies to
2
kA ¼ k1k2=kꢂ1 þ k1k3=kꢂ1½Bꢁ
ð10Þ
showing a linear dependence between kA and the square
of the nucleophile concentration. The intercept and the
slope represent the non-catalytic and the catalytic rate
coefficient, respectively. Banjoko and Ezeani11a found
that whereas plots of kA vs amine concentration render
straight lines with negative intercepts, plots of kA vs the
square of the nucleophile concentration give rise to
straight lines with positive intercepts. For reactions
involving good leaving groups, the authors observed
that the mechanism proceeds through a six-membered
cyclic intermediate formed by a similar inter-hydrogen
bonding between the zwitterionic intermediate and only
one amine molecule.
The experimental evidence emerging from the present
reactive systems which strongly support this latter me-
chanism is the following:
ꢈ The reactions show a second-order dependence on
amine concentration and total third-order, exhibiting
a linear relationship between the kA values and the
initial amine concentration with negligible intercepts.
ꢈ The decomposition of the zwitterionic intermediate is
the rate-determining step. The low activation enthal-
pies and the apparent absence of activation energy
indicate that the reaction occurs stepwise; the rate-
determining step must be preceded by at least one fast
equilibrium, whereby the expected increase in rate for
the slow step with increasing temperature would be
compensated by a shift of the preceding equilibrium
toward the reagents. The negative activation enthalpies
require the exothermic formation of a weak complex
preceding the rate-determining step.
For all the nucleophiles explored in the present work
and over the whole range of amine concentrations, kA
values do not exhibit a linear dependence with [B]2
whereas plots of kA vs [B] do not afford lines with
definite negative intercepts. These results and those
corresponding to the evaluation of the dimer mechanism
allow us to reject a quadratic dependence of kA on [B].
The six-membered cyclic transition state mechanism:
the occurrence of a six-membered cyclic transition state
was first proposed by Capon and Rees.8a As mentioned
above, this transition state includes two molecules of
amine which are joined through hydrogen bonds. The
second molecule of amine may act as a proton donor to
the leaving group as well as a proton acceptor from the
positively charged nitrogen atom of the zwitterionic
intermediate, the reaction to product taking place in a
concerted manner. Ayediran et al.,9a considering the
strength and the range of electrostatic forces in solvents
of low ionizing power, proposed modifications for the
base-catalyzed path: (i) the initial formation of a suitably
ꢈ The reactions exhibit large negative values of the
activation entropies, which corresponds to the forma-
tion of that aggregate or complex. This kind of me-
chanism was found for reactions carried out in solvents
with low ionizing power, with low dipolarity/polariz-
ability values, such as isooctane and hexane, and also
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 336–346

REACTION MECHANISM BETWEEN 1-FLUORO-2,6-DINITROBENZENE AND ALICYCLIC AMINES
345
in solvents characterized by moderate values such as
benzene and ethyl acetate. In this work, we extended
the operability of this mechanism to EAc–CHCl3
binary solvent mixtures in the EAc-rich zone.
the amine (DF–H). These values correspond to four
different optimized structures, obtained by perturbing
the system and allowing it to evolve to the respective
geometry.
ꢈ The extent of the influence of increasing temperature
on reaction rates is related to the stability of that weak
complex. The more stable the complex, the greater its
equilibrium constant of formation. Therefore, an in-
crease in temperature leads to an enhancement shift
towards reactants and in this way to a decrease in the
reaction rate. In connection with our experimental
results, we assume that the corresponding complex
with PYR would be less stable in comparison with
those with PIP and HPIP, taking into account the fact
that reactions with the former amine are less affected
by the increasing temperature.
The results allow us to make the following remarks:
ꢈ The negative ꢀꢀH# values indicate that the formation
of these complexes is exothermic and that they con-
stitute stable structures.
ꢈ The complex with PYR is the structure with the highest
energy and is less stable than the corresponding com-
plexes with PIP and HPIP, thus verifying what we
expressed in the paragraphs above.
ꢈ The magnitudes of the DN–H and DF–H inter-atomic
distances suggest that the interaction forces coming
into play are electrostatic rather than hydrogen-bond
type.13 This result favors the formation of a suitably
orientated dipolar aggregate rather than an inter-
hydrogen bonding structure.
Six-membered intermediate complex: theoretical
calculations. Semi-empirical molecular orbital calcula-
tions were performed in order to obtain a relative stability
order of the proposed intermediate complex. In this
direction, the complex intermediate structures for PYR,
PIP and HPIP were optimized at the AM1 level.12 In
order to perform these calculations, the corresponding
results obtained in previous work1b for the structures of
each amine and for each ZH were taken into account.
Figures S9, S10 and S11 (ESI) show the optimized
structures which represent a local minimum in the
potential energy surface for the complexes with PYR,
PIP and HPIP, respectively. Table 5 presents the total
energy (ET), the heat of formation (ꢀH#), the heat of
formation with respect to reactants (ꢀꢀH#) of the
respective complexes and the inter-atomic distances be-
tween the ammonium hydrogen of ZH and the nitrogen
atom of the entering amine (DN–H) and between the
fluorine atom of ZH and the hydrogen (amino group) of
CONCLUSIONS
Based on the overall evaluation presented in this paper,
we can conclude the following: (i) the analysis of
Arrhenius lines allowed us to prove the existence of
isokinetic relationships, which were used as a diagnostic
tool in order to infer a changeover in the nature of the
rate-determining transition state as a function of solvent
composition; (ii) the reactions with the explored second-
ary amines carried out in pure EAc and EAc–CHCl3
solvent mixtures at XCoS ¼ 0.1, and in some cases also at
XCoS ¼ 0.3, proceed via the formation of a six-membered
orientated dipolar aggregate in which the SB–GA me-
chanism may take place; (iii) the reactions carried out in
the remaining solvent mixtures evolve towards the clas-
sical SB–GA mechanism.
Table 5. Total energy, heats of formation and inter-atomic distances at the AM1 level corresponding to the different
complexesa
ET (kJ molꢂ1
)
ꢀH# (kJ molꢂ1
)
ꢀꢀH# (kJ molꢂ1
)
˚
DN–H (A)
˚
DF–H (A)
Complex
2,6-DNFB þ PYR
1
ꢂ450766.1
ꢂ450766.1
ꢂ450766.1
ꢂ450766.1
ꢂ51.59
ꢂ51.60
ꢂ51.60
51.60
ꢂ90.29
ꢂ90.29
ꢂ90.33
ꢂ90.33
2.74734
2.71905
2.71585
2.71917
2.34827
2.38004
2.37614
2.37768
2
3
4
2,6-DNFB þ PIP
1
2
3
4
ꢂ480888.5
ꢂ480888.5
ꢂ480888.5
ꢂ480888.5
ꢂ145.1
ꢂ145.0
ꢂ145.0
ꢂ145.1
ꢂ136.9
ꢂ136.7
ꢂ136.7
ꢂ136.9
2.58408
2.61162
2.62010
2.58744
2.40092
2.48852
2.49330
2.38766
2,6-DNFB þ HPIP
1
2
3
4
ꢂ510944.0
ꢂ510943.6
ꢂ510944.0
ꢂ510944.0
ꢂ171.3
ꢂ171.1
ꢂ171.3
ꢂ171.3
ꢂ149.3
ꢂ149.1
ꢂ149.3
ꢂ149.3
2.74540
2.800868
2.75368
2.76642
2.41476
2.35458
2.40743
2.40271
a ET ¼ total energy; ꢀH# ¼ heat of formation; ꢀꢀH# ¼ heat of formation with respect to reactants; DN–H and DF–H ¼ inter-atomic distances.
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 336–346

346
P. M. MANCINI, G. G. FORTUNATO AND L. R. VOTTERO
2. (a) Orvick JA, Bunnett JF. J. Am. Chem. Soc. 1970; 92:
EXPERIMENTAL
2417–2427; (b) Bunnett JF, Sekiguchi S, Smith LA. J. Am.
Chem. Soc. 1981; 103: 4865–4871; (c) Nudelman NS,
Mancini PME, Martinez RD, Vottero LR. J. Chem. Soc.,
Perkin Trans 2 1987; 951–954; (d) Sekiguchi S, Hosokawa M,
Suzuki T, Sato M. J. Chem. Soc., Perkin Trans 2 1993; 1111–
1118.
2,6-DNFB was synthesized as reported previously.14
PYR, PIP, HPIP and the corresponding solvents were
purified as usual.15 The solvents were kept over 4 A
˚
molecular sieves and stored in special vessels that allow
delivery without air contamination. All binary mixtures
were prepared prior to use. ꢀH# and ꢀS# were calculated
by means of the Eyring equation. The ꢀG# values were
calculated from ꢀH# and ꢀS# values at 25 ꢀC. Thm was
close to 25 ꢀC so that this temperature was selected as the
harmonic mean temperature.
The kinetics of the reactions were studied by UV–
visible spectrophotometry. A Perkin-Elmer Model 124
spectrophotometer was used, equipped with a data acqui-
sition system. Molecular orbital calculations were carried
out using the HyperChem 5.11 system of programs.
Computational details are presented in the Electronic
Supplementary Information.
3. Linert W. Chem. Soc. Rev. 1994; 429–438.
4. Linert W. Inorg. Chim. Acta 1988; 141: 233–242.
5. Krug R. Ind. Eng. Chem. Fundam. 1980; 19: 50–59.
6. (a) Bunnett J. In Techniques of Chemistry, vol. 6, Lewis ES (ed).
Wiley: New York, 1974 Chapt. VIII; (b) Bernasconi C, Zollinger
H. Helv. Chim. Acta 1966; 49: 103–111.
7. (a) Forlani L. J. Chem. Res. (S) 1984; 66: 260–261; (b) Forlani L,
Todesco PE. J. Chem. Soc., Perkin Trans. 2 1988; 1585–1589; (c)
Chiacchiera SM, Singh JO, Anunziata JD, Silber JJ. J. Chem.
Soc., Perkin Trans. 2 1988; 1585–1589; (d) Durantini E,
Zingaretti JD, Anunziata JD, Silber JJ. J. Phys. Org. Chem.
1992; 5: 557–565.
8. (a) Capon B, Rees W. Annu. Rep. Prog. Chem. 1963; 247–293; (b)
Bitter B, Zollinger H. Helv. Chim. Acta 1961; 44: 812–823; (c)
Illuminatti G, La Torre F, Liggieri G, Sleiter G, Stegel F. J. Am.
Chem. Soc. 1975; 97: 1851–1854.
9. (a) Ayediran T, Bamkole O, Hirst J, Onyido I. J. Chem. Soc.,
Perkin Trans. 2 1977; 597–600; (b) Bamkole O, Hirst J, Onyido I.
J. Chem. Soc., Perkin Trans. 2 1982; 889–893.
10. (a) Nudelman NS, Palleros D. J. Org. Chem. 1989; 2: 1607–1612;
(b) Nudelman NS. In The Chemistry of Amino, Nitroso, Nitro
and Related groups, Patai S (ed). Wiley: Chichester, 1996;
Chapt. 26.
Acknowledgements
11. (a) Banjoko O, Ezeani J. J. Chem. Soc., Perkin Trans. 2 1982;
1357–1360; (b) Banjoko O, Ezeani J. J. Chem. Soc., Perkin Trans.
2 1986; 531–536; (c) Banjoko O, Bayeroju YA. J. Chem. Soc.,
Perkin Trans. 2 1988; 1853–1857.
12. (a) Dewar MJS, Zoebisch EG, Healy EF, Stewart JJP. J. Am.
Chem. Soc. 1985; 107: 3902–3909; (b) Dewar MJS, Dieter KM.
J. Am. Chem. Soc. 1886; 108: 8075–8086; (c) Stewart JJP.
J. Comput. Aided. Mol. Des. 1990; 4: 1–105.
We are indebted to the Universidad Nacional del Litoral
´
cial support from the Science and Technology Secretar-
iat, UNL, CAI þ D Program (Projects 2000-17-151,
2002-21-152 and 2002-21-153).
(UNL), Republica Argentina. This work received finan-
13. Jeffrey GA. An Introduction to Hydrogen Bonding. Oxford Uni-
versity Press: Oxford, 1997; 11–16.
14. Parker RE, Read TO. J. Chem. Soc. 1962; 3149–3153.
REFERENCES
´
15. (a) Mancini PM, Terenzani AJ, Adam C, Perez A, Vottero LR. J.
Phys. Org. Chem. 1999; 12: 713–724; (b) Mancini PM, Terenzani
1. (a) Mancini PM, Fortunato GG, Adam C, Vottero LR, Terenzani
AJ. J. Phys. Org. Chem. 2002; 15: 258–269; (b) Mancini PM,
Fortunato GG, Vottero LR. J. Phys. Org. Chem. 2004; 17: 138–
147.
´
AJ, Adam C, Perez A, Vottero LR. J. Phys. Org. Chem. 1999; 12:
207–220; (c) Mancini PM, Adam C, Perez A, Vottero LR. J. Phys.
´
Org. Chem. 2000; 13: 221–231.
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2005; 18: 336–346