Kinetic and equilibrium studies of σ-adduct formation and nucleophilic substitution in the reactions of trinitro-activated benzenes with aliphatic amines in acetonitrile

Article abstract of DOI:10.1039/a603707k

Rate and equilibrium constants are reported for reactions in acetonitrile of butylamine, pyrrolidine and piperidine with 1,3,5-trinitrobenzene, 1, and with ethyl 2,4,6-trinitrophenyl ether, 4a, and phenyl 2,4,6-trinitrophenyl ether, 4b. Rapid nucleophilic attack at unsubstituted ring-positions may yield anionic σ-adducts via zwitterionic intermediates, while slower attack at the 1-position of 4a and 4b may lead to substitution to give 2,4,6-trinitroaniline derivatives. Base catalysis in the substitution reaction reflects rate-limiting proton transfer which may be from the zwitterionic intermediates to amine in the case of 4b, or from a substituted ammonium ion to the ethoxy leaving group in the case of 4a. Comparisons with values in DMSO indicate that values of overall equilibrium constants for adduct formation are ca. 10⁴ lower in acetonitrile, while rate constants for proton transfer are ca. 10⁴ higher. These differences may reflect strong hydrogen-bonding between DMSO and -NH⁺ protons in ammonium ions and in zwitterions. In acetonitrile homoconjugation of substituted ammonium ions with free amine is an important factor.

Full text of DOI:10.1039/a603707k

View Article Online / Journal Homepage / Table of Contents for this issue
Kinetic and equilibrium studies of ó-adduct formation and
nucleophilic substitution in the reactions of trinitro-activated
benzenes with aliphatic amines in acetonitrile
Michael R. Crampton* and Simon D. Lord
Chemistry Department, Durham University, Durham, UK DH1 3LE
Rate and equilibrium constants are reported for reactions in acetonitrile of butylamine, pyrrolidine and
piperidine with 1,3,5-trinitrobenzene, 1, and with ethyl 2,4,6-trinitrophenyl ether, 4a, and phenyl 2,4,6-
trinitrophenyl ether, 4b. Rapid nucleophilic attack at unsubstituted ring-positions may yield anionic
ó-adducts via zwitterionic intermediates, while slower attack at the 1-position of 4a and 4b may lead to
substitution to give 2,4,6-trinitroaniline derivatives. Base catalysis in the substitution reaction reflects
rate-limiting proton transfer which may be from the zwitterionic intermediates to amine in the case of 4b,
or from a substituted ammonium ion to the ethoxy leaving group in the case of 4a.
Comparisons with values in D M SO indicate that values of overall equilibrium constants for adduct
4
4
formation are ca. 10 lower in acetonitrile, while rate constants for proton transfer are ca. 10 higher.
+
These differences may reflect strong hydrogen-bonding between D M SO and ᎐N H protons in ammonium
ions and in zwitterions. In acetonitrile homoconjugation of substituted ammonium ions with free amine is
an important factor.
The reaction of 1,3,5-trinitrobenzene 1, with aliphatic amines
1
–3
in dipolar aprotic solvents yields
anionic σ-adduct by the
processes shown in Scheme 1. Kinetic studies of these reactions
Scheme 1
4
,5
have shown that in dimethyl sulfoxide (DMSO) and in mixed
6
aqueous solvents the proton transfer step may be rate-
7
3
Ϫ1
determining. In DMSO the value of k is ca. 10 dm mol
s
Am
Ϫ1
6
5
for reaction with butylamine, 10 for pyrrolidine and 10 for
piperidine. These values are considerably lower than that
expected for diffusion controlled reaction even though the pro-
ton transfers from the zwitterions, 2, to amines are thermo-
dynamically downhill. The greater acidity of compounds 2 than
of the parent ammonium ions is due to the electron withdraw-
Scheme 2
to apply for substrates, such as alkyl ethers, carrying poor leav-
9
–12
ing groups.
However, there is now strong evidence that for
substrates carrying good leaving groups, such as phenyl ethers
and phenyl sulfides, base catalysis results from rate-limiting
7
,8
ing effect of the trinitro-aromatic moiety.
The observation of general base catalysis in nucleophilic sub-
stitution reactions is also indicative of rate-determining proton
1
3,14
proton transfer from zwitterions to base.
Reaction at the 1-
2
,3
transfer. Several detailed studies have been reported
DMSO solvent and the overall mechanism is shown in Scheme
. The base-catalysed pathway may, in an analogous fashion to
in
position leading to substitution may be preceded by rapid
reversible reaction at the unsubstituted 3-position leading to
adducts 5.
2
that shown in Scheme 1, involve rate-limiting proton transfer
from the zwitterionic intermediate, 6, to base (the kAm step), or
it may involve general acid catalysis of leaving group departure
We report here kinetic and equilibrium results for reactions
in acetonitrile corresponding to those shown in Scheme 1 with
trinitrobenzene, and in Scheme 2 with ethyl, 4a, and phenyl, 4b,
ethers. The amines used were butylamine, pyrrolidine and
piperidine. Although acetonitrile has been widely used as a
solvent for the examination of base catalysis in substitution
(
the k step). The latter, the SB-GA mechanism has been shown
4
©
British Crown Copyright 1996/DERA Farnborough, Hants., UK.
1
5–18
reactions,
the substrates previously studied have been less
Published with the permission of the Controller of Her Britannic
Majesty’s Stationery Office.
activated than the trinitroaryl ethers, 4, so that intermediates
J. Chem. Soc., Perkin Trans. 2, 1997
369

View Article Online
a
Table 1 Equilibrium data for formation of 3 from 1 and pyrrolidine in acetonitrile at 25 ЊC
[
Pyrrolidine]/
[Pyrrolidinium
perchlorate]/mol dm
Ϫ3
Ϫ3
b
3
Ϫ1
c
mol dm
A (445 nm)
Kc,3 /dm mol
Kc,3
0
0
0
0
0
0
0
0
0
0
0
.01
.02
.03
.04
.06
.08
.10
.10
.20
.03
.03
0.001
0.001
0.001
0.001
0.001
0.001
0.001
—
0.028
0.120
0.267
0.46
0.71
0.88
0.96
1.08
1.09
0.153
0.087
0.26
0.31
0.36
0.46
0.52
0.65
0.74
—
0.25
0.31
0.36
0.42
0.52
0.63
0.74
—
0.002
0.004
—
0.36
0.38
0.37
0.37
a
Ϫ5
Ϫ3
b
1
2
+
1
2
2
c
0
3
Ϫ1
Concentration is 4 × 10 mol dm
. Calculated as (A/1.09 Ϫ A) (R R NH )stoich/(R R NH) . Calculated from eqn. (4) with K 0.20 dm mol
2 c,3
3
Ϫ1
and K 27 dm mol
.
h
were not observed. Our aims were: (i) to determine rate con-
stants for proton transfer reactions in acetonitrile and compare
these with corresponding values in DMSO, and (ii) to examine
the detailed mechanism of base catalysis in acetonitrile and
identify the nature of the rate determining step.
determination of the values of the extinction coefficients (see
Table 3).
Measurement of absorbance values in solutions containing
Ϫ3
0.001 mol dm amine perchlorate and varying concentrations
of amine allowed the calculation of values of Kc,3 defined in
The relative permittivity of acetonitrile is comparable with
that of DMSO; the values are 36 and 46.6, respectively. However
eqn. (2). Data are given in Table 1 for reaction with pyrrolidine
1
2
+
2
acetonitrile is a much less basic solvent and pK values for ali-
[3] [R R NH
]
Stoich
2
a
Kc,3
=
1
2
(2)
1
9,20
phatic ammonium ions are ca. 8 units larger than in DMSO.
[1] [R R NH]
Nevertheless extensive studies of acid–base behaviour have
been made in acetonitrile and the solvent is sufficiently basic to
and show that values of Kc,3 increase with increasing amine
concentration. This is attributed to stabilisation, eqn. (1), of the
2
1
completely dissociate perchloric acid. For this reason we have
2
2
pyrrolidinium cations by association with pyrrolidine, which is
used perchlorate salt in our studies. It is also known that
association of aliphatic ammonium ions with the parent amines
produces homoconjugates [eqn. (1)]; values of the equilibrium
0
c,3
in large excess. We define K in terms of the free, unassociated
cations, eqn. (3), and note that Kc,3 will approach the value of
+
+
1
2
+
1
2
1
2
1
2
1
2
k3 kAm
kϪ3 kAmH
[3] [R R NH ]
2 Free
1 2 2
R R NH + R R NH
R R NH ؒ ؒ ؒNHR R (1)
0
c,3
2
2
K
=
=
(3)
+
[1] [R R NH]
3
Ϫ1
constant K are ca. 30 dm mol . A further important differ-
h
0
0
K
as the amine concentration tends to zero. Kc,3 and K are
c,3
c,3
ence between the solvents is that DMSO is known to be a strong
hydrogen-bond acceptor while acetonitrile shows weak
hydrogen-bonding properties. Acetonitrile has previously
related to the equilibrium constant, K , for homoconjugation
by eqn. (4). A plot, not shown, of Kc,3 vs. pyrrolidine concentra-
h
2
3
been used as a solvent for obtaining NMR spectra of σ-
0
c,3
1
2
Kc,3 = K (1 + K [R R NH])
(4)
2
4–27
h
adducts.
However, the only previous quantitative meas-
urement in this solvent was a determination of the equilibrium
constant for adduct formation between 1 and piperidine.
0
3
Ϫ1
3
Ϫ1
tion gave values for K 0.20 dm mol and K 27 dm mol .
2
8
c,3
h
The data in Table 1 show that, as expected, values of Kc,3 are
virtually independent of the cation concentration in the pres-
ence of a large excess of amine. Data equivalent to that in Table
Experimental
1
for reactions with butylamine and with piperidine are
1
1
1
,3,5-Trinitrobenzene, ethyl 2,4,6-trinitrophenyl ether and
13
reported as supplementary information in Tables 11 and 12.†
Kinetic measurements were made by stopped-flow spectro-
photometry with amine in large excess of parent, 1. The equili-
bration of 1 and 3 will, in the absence of added ammonium salt,
phenyl 2,4,6-trinitrophenyl ether were available from previous
work. Amines and acetonitrile were the purest available com-
mercial specimens. Solutions of amine perchlorates were pre-
pared by neutralisation of perchloric acid in acetonitrile with
the appropriate amine. UV–VIS spectra and kinetic measure-
ments were made with Perkin-Elmer Lambda 2, or Hi-Tech SF
4
give rise to complex kinetics since it represents mixed first-
order (forward) and second-order (reverse) reactions. If suf-
ficient amine is used to ensure virtually complete conversion to
product then the forward rate term dominates and first-order
kinetics are predicted. However, with this condition rate con-
stants were too rapid for measurement. We overcame this prob-
lem for the pyrrolidine and piperidine reactions by measuring
the reverse reaction. Thus, pre-formed solutions of the adduct
were reacted with the appropriate ammonium salt and the rate
constants of the fading reactions were measured. Assuming
that the zwitterions 2 may be treated as steady-state intermedi-
3
L or Applied Photophysics SX-17 MV stopped flow spectro-
photometers at 25 ЊC. Reported rate coefficients are the means
of several determinations and are precise to ±5%. Rate con-
stants were measured under first-order conditions; hence for
reactions in buffers (amine plus amine perchlorate) the buffer
components were in large excess of the substrate concentration
Ϫ5
Ϫ3
(
1–5 × 10 mol dm ).
4
ates the general rate expression for the equilibration process is
eqn. (5). If conditions are chosen so that >90% reversion of
adduct to parent is achieved, then the reverse reaction domin-
Results and discussion
Reaction with 1,3,5-trinitrobenzene, 1
Reaction of 1 with aliphatic amines in acetonitrile resulted in
the rapid formation of a species having absorption maxima at ca.
†
Tables 11–13 have been deposited as supplementary data at the British
4
50 and 530 nm, characteristic of the σ-adducts 3. In the
Library; Suppl. No. 57173 (4 pp). For details of the British Library
Supplementary Publications scheme, see ‘Instructions for Authors
(1997)’, J. Chem. Soc., Perkin Trans. 2, 1997, issue 1.
absence of added salt virtually complete conversion to 3 could
be obtained at high amine concentrations, allowing the
3
70
J. Chem. Soc., Perkin Trans. 2, 1997

View Article Online
2
+
k3kAm[Am]
k k
Ϫ3 AmH
+
[AmH ]
With each amine this was followed by a fast but measurable
stoich
kobs
=
+
(5)
kϪ3 + kAm[Am] (kϪ3 + kAm[Am])(1 + K [Am])
reaction giving rise to species with λmax 430, 500 nm; this is
attributed to equilibration with the adducts, 7, on the substitu-
h
1
1
ates [this is represented by the final term in eqn. (5)]. The results
tion pathways. The general rate expression for this process is
in Table 2 relate to low amine concentrations where the condi-
eqn. (7), where K ≡ k /k and where Kc,3 is the equilibrium
1
1
Ϫ1
tion 1 ӷ K [Am] applies. They show that rate constants for the
h
2
fading reactions depend directly on the concentration of the
ammonium salt but are independent of the amine concentra-
tion. This indicates that kϪ3 ӷ kAm[Am] so that eqn. (6) applies.
K1kAm[Am]
kAm[Am]
kfast
=
+
2
K [Am]
c,3
1
+
1 +
+
ͩ
kϪ1 ͪͩ [AmH ]
ͪ
stoich
+
+
kobs = kAmH
+
[AmH ]
(6)
kAmH [AmH ]
+
stoich
(7)
kAm[Am]
1
+
(1 + K [Am])
Hence, at the low amine concentrations used, the proton trans-
fer step is rate determining in the equilibration of 1 and 3. We
ͩ
kϪ1
ͪ
h
6
3
Ϫ1 Ϫ1
obtain values for kAmH
dine reaction and 2.5 × 10 dm mol
reaction. We note that these values are ca. 10 times larger than
+
of 9 × 10 dm mol
s
for the pyrroli-
constant for formation of 5 from 4. The value of K is expected
c,3
6
3
Ϫ1 Ϫ1
s
for the piperidine
to show a dependence on amine concentration as indicated in
eqn. (4). Our results are only compatible with the assumption
that only free ammonium ions are active in reprotonating the
anions 7; the relation between concentrations of free and total
ammonium ions is eqn. (8). If the condition kAm[Am] ӷ kϪ1
4
4
,5
the corresponding values for reaction in DMSO. Since values
of K are known from the equilibrium measurements, values
0
c,3
0
c,3
of k3kAm/kϪ3 (≡ kAmH
Table 3.
+
K ) were calculated and are given in
+
+
In the case of the adduct formed with butylamine, rate con-
stants for both forming and fading reactions were too fast for
measurement.
[AmH ] = [AmH ]
/(1 + K [Am])
(8)
Free
stoich
h
applies, so that the proton transfer equilibrium between 6 and 7
is rapid, then eqn. (7) reduces to eqn. (9).
Reaction with ethyl 2,4,6-trinitrophenyl ether, 4a
+
The reactions with pyrrolidine, piperidine and butylamine in
acetonitrile were qualitatively similar to the corresponding reac-
k1[Am]
k k
+
[AmH ]
Ϫ1 AmH
stoich
kfast
=
+
(9)
2
K [Am]
kAm[Am](1 + K [Am])
c,3
h
1
+
1
1,29
+
tions in DMSO reported previously,
and shown in Scheme 2.
[AmH ]
stoich
With each amine a very rapid reaction with low amplitude was
observed, attributed to equilibration of the parent with 5, the
adduct at the 3-position. In acetonitrile the rate of this process
was too rapid for measurement by the stopped-flow method.
In the reactions with pyrrolidine and butylamine a third reac-
tion, kslow, was observed leading to the substitution product 8.
The spectral maxima were at 362 nm in the pyrrolidine reaction
a
Table 2 Rate data for the fading reactions of adducts 3, from pyrrolidine or piperidine, with appropriate ammonium salt at 25 ЊC
6
3
[
Pyrrolidinium
kAmH /10 dm
+
Ϫ3
Ϫ5
Ϫ3
Ϫ1
Ϫ1 Ϫ1
[
Pyrrolidine]/mol dm
perchlorate]/10 mol dm
kobs/s
mol
s
0
0
0
0
0
.001
.001
.001
.002
.002
2
3
4
2
3
160 ± 20
260 ± 20
330 ± 40
175 ± 15
255 ± 20
8
9
8
9
9
6
3
[
Piperidinium
kAmH /10 dm
+
Ϫ3
Ϫ5
Ϫ3
Ϫ1
Ϫ1 Ϫ1
[
Piperidine]/mol dm
perchlorate]/10 mol dm
kobs/s
mol
s
0
0
0
0
0
0
0
0
0
0
.001
.001
.001
.001
.001
.002
.002
.002
.002
.002
2
3
4
5
8
2
3
4
5
8
52 ± 2
75
110
2.6
2.5
2.7
2.6
2.5
2.7
2.4
2.6
2.5
2.5
130
200 ± 10
55 ± 2
73
105
125
200 ± 10
a
Ϫ4
Ϫ3
Ϫ6
Ϫ3
The concentration of 1 is 1 × 10 mol dm . The estimated concentration of adduct, 3, before reaction with salt is р5 × 10 mol dm
.
Table 3 Summary of data for reaction of 1 with amines in acetonitrile at 25 ЊC
a
0
c,3
3
Ϫ1
3
Ϫ1
3
Ϫ1 Ϫ1
6
Ϫ2 Ϫ1
Amine
3, λmax /nm
K
/dm mol
K /dm mol
kAmH
+
/dm mol
s
K3kAm/dm mol
s
h
4
Butylamine
450 (2.8 × 10 )
5
0.0025
0.20
20
27
25
—
—
4
34 (1.6 × 10 )
4
6
6
Pyrrolidine
445 (2.7 × 10 )
9 × 10
2 × 10
4
5
24 (1.6 × 10 )
b
4
6
5
Piperidine
446 (2.8 × 10 )
0.055
2.5 × 10
1.4 × 10
4
5
24 (1.6 × 10 )
a
3
Ϫ1
Ϫ1
b
28
0
c,3
ε/dm mol cm in parentheses. Previous values were K 0.058, K 24 at 20 ЊC.
h
J. Chem. Soc., Perkin Trans. 2, 1997
371

View Article Online
Ϫ1
and 340 nm in the butylamine reaction. At high amine concen-
trations the product spectra were shifted to longer wavelengths
consistent with the known formation, in rapid equilibria, of
anions by adduct formation at the 3-position and/or by loss of a
side-chain proton. At all concentrations the final spectra were
identical to those of authentic samples of product in the same
mol . The data allow the calculation of a value for k
1
3 Ϫ1 Ϫ1
(≡K1kAmk /k ) of 1250 dm mol s . Values of kslow, the
product forming reaction, go through a maximum with increas-
ing amine concentration. Use of eqn. (10) using the known
Ϫ1 Am
3
0
0
3
Ϫ1
3
Ϫ1
values of K 0.3 dm mol , and K 23 dm mol yields a value
c,1
h
3
Ϫ1 Ϫ1
for k of 190 ± 20 dm mol s .
4
reaction medium. The rate expression for product formation is
The data for reaction with piperidine in Table 5 give a nice fit
with eqn. (7) with the values given in the footnotes. They yield a
Ϫ1
0
eqn. (10), where K is defined in eqn. (11). Our results are
c,1
3
Ϫ1 Ϫ1
3
value for k of 780 dm mol s . The value of 0.025 dm mol
1
0
2
+
0
c,1
k K [Am] [AmH ]
obtained for K from absorbance data is in good accord with
4
c,1
stoich
kslow
=
0
c,1
2
+
K [Am] (1 + K [Am]) + [AmH ]
(10)
(11)
that obtained, eqn. (11), by combination of values for K1k
h
stoich
Am
and kAmH .
+
+
[
[
7] [AmH ]
k1kAm
k k
The data in Table 6 for reaction with butylamine indicate
that, at the amine concentrations used, the condition
kAm[Am] ӷ kϪ1 applies. Hence eqn. (9) is applicable and a good
0
c,1
Free
2
K
=
=
Ϫ1 AmH
6] [Am]
+
3
Ϫ1 Ϫ1
compatible only with the assumption, used in the derivation of
eqn. (10), that only free (non-homoconjugated) ammonium
ions are effective in the acid catalysed expulsion of the ethoxy
leaving group from 7.
Although reaction of 4a with piperidine leads to the form-
ation of the adduct 7, the subsequent reaction was very slow
and did not lead to the expected substitution product 8. This
slow reaction was not investigated further.
fit is achieved with values of k 130 dm mol
s
and kϪ1kAmH /
+
1
Ϫ1
kAm 360 s . Combination of these values yields [eqn. (11)] a
value for K of 0.36 dm mol . Using this value and the
0
3
Ϫ1
c,1
known value for K we obtain using eqn. (10) a value of
h
3
3
Ϫ1 Ϫ1
(8.5 ± 1) × 10 dm mol
kϪ1 > 500 dm mol , so that k
s
for k . We estimate that kAm/
4
5 3 Ϫ1 Ϫ1
3
Ϫ1
+
> 1.8 × 10 dm mol s .
AmH
Reaction with phenyl 2,4,6-trinitrophenyl ether, 4b
Results for reaction with pyrrolidine are in Table 4. The
absorbance values at completion of the reaction forming
In the reactions with amines of the phenyl ether only two time
dependent processes were observed. These were an initial rapid
colour forming reaction attributed to equilibration with the 3-
adducts 5b and a slower reaction yielding the substitution
product 8. The UV–VIS spectra at completion of these reac-
tions were identical to those of authentic samples of the corres-
ponding N-substituted 2,4,6-trinitroanilines in the same reac-
adduct 7 allow the calculation of values of K and lead to a
c,1
0
c,1
3
Ϫ1
value for K of 0.3 dm mol . Values of k decrease with
fast
increasing amine concentration before rising to a maximum and
then falling. This behaviour is predicted by eqn. (7) and the
5
results are accommodated with values of K k (1 ± 0.2) × 10
1
Am
6
Ϫ2 Ϫ1
3
Ϫ1
5
3
1
dm mol s , kAm/kϪ1 80 ± 20 dm mol , k
+
(4 ± 1) × 10
tion medium. H NMR spectroscopy was also used to confirm
the identities of the reaction products. Data are given as
AmH
3
Ϫ1 Ϫ1
3
Ϫ1
0
dm mol s , K 23 dm mol and with K 0.01 ± 0.002 dm
h
c,3
a
Ϫ3
Table 4 Kinetic and equilibrium results for reaction of 4a with pyrrolidine in acetonitrile containing 0.001 mol dm pyrrolidinium perchlorate at
2
5 ЊC
e
3
g
3
[
Pyrrolidine]/
Kc,1 /dm
k /dm
4
Ϫ3
b
Ϫ1
c
Ϫ1
d
Ϫ1
f
Ϫ2 Ϫ1
Ϫ1 Ϫ1
mol dm
kfast /s
kcalc /s
A (430)
mol
kslow /10
s
mol
s
0
0
0
0
0
0
0
0
0
0
0
0
.004
.005
.006
.008
.010
.020
.030
.050
.070
.10
—
—
—
—
—
120
90
78
85
95
95
80
—
—
—
—
—
110
87
81
87
97
94
77
—
—
—
—
—
—
—
—
0.082
0.13
0.20
0.35
0.55
2.0
3.9
6.0
6.2
5.2
180
190
200
200
200
210
220
230
200
210
200
190
—
—
0.027
0.054
0.125
0.162
0.183
0.195
0.198
0.39
0.41
0.67
0.87
1.08
—
.15
.20
3.9
2.9
—
a
c
Ϫ5
Ϫ3
b
Concentration of 4a is 4 × 10 mol dm
.
Measured as a colour forming reaction; identical values were obtained at 430 and at 500 nm.
6 Ϫ2 Ϫ1 3 Ϫ1 5 3 Ϫ1 Ϫ1 3 Ϫ1
5
Calculated from eqn. (7) with K k 1.1 × 10 dm mol
s
, kAm/kϪ1 86 dm mol , kAmH
+
3.9 × 10 dm mol
s , K 27 dm mol and with
h
2
1
Am
d
e
+
Kc,3 = 0.009 (1 + 27 [Am]). At completion of reaction forming the adduct 7. Calculated as A/(0.20 Ϫ A) [AmH ]/[Am] . The data fit the expression
f
K
= 0.28 (1 + 27[Am]). Identical results were obtained from measurements of colour forming reaction at 430 nm or fading reaction at 500 nm.
Calculated from eqn. (10) with K 0.3 dm mol , and K 23 dm mol
c,1
g
0 3 Ϫ1 3 Ϫ1
.
c,1
h
a
Ϫ3
Table 5 Kinetic and equilibrium results for reaction of 4a with piperidine in acetonitrile containing 0.001 mol dm piperidinium perchlorate at
2
5 ЊC
Ϫ3
b
Ϫ1
c
Ϫ1
d
e
3
Ϫ1
[
Piperidine]/mol dm
kfast /s
kcalc /s
A (430 nm)
Kc,1 /dm mol
0
0
0
0
0
0
0
0
0
.030
.050
.070
.10
.15
.20
.25
.30
.40
106
90
77
70
75
84
87
85
69
114
89
75
70
77
85
88
85
69
—
—
0.0239
0.0478
0.0902
0.141
0.169
0.183
0.189
0.194
0.056
0.067
0.087
0.12
0.17
—
—
—
a
Ϫ5
Ϫ3
b
c
6
Ϫ2 Ϫ1
Concentration of 4a is 4 × 10 mol dm
.
Fast reaction forming 7 measured at 430 nm. Calculated from eqn. (7) with K k 6200 dm mol
s ,
1
Am
3
Ϫ1
5
3
Ϫ1 Ϫ1
3
Ϫ1
d
k
7
/kϪ1 8 dm mol , kAmH
+
2.4 × 10 dm mol
s
, K 25 dm mol and with Kc,3 0.0015 (1 + 25[Am]). At completion of reaction forming adduct
Am
e
h
+
2
.
Calculated as A/(0.194 Ϫ A)/[AmH ]/[Am] . The data fit the expression K = 0.025 (1 + 25[Am]).
c,1
3
72
J. Chem. Soc., Perkin Trans. 2, 1997

View Article Online
Ϫ3
Table 6 Rate data for reaction of 4a with butylamine in acetonitrile containing 0.001 mol dm butylammonium perchlorate at 25 ЊC
Ϫ3
a
Ϫ1
b
Ϫ1
c
Ϫ1
d
3
3
Ϫ1 Ϫ1
[
Butylamine]/mol dm
kfast /s
kcalc /s
kslow /s
k /10 dm mol
s
4
0
0
0
0
0
0
0
0
0
.02
.03
.05
.07
.10
.15
.20
.25
.30
14
12
11
11
14
18
23
28
32
15
11
10
11
14
18
24
28
31
1.1
1.8
2.8
3.1
2.8
2.1
1.7
1.4
1.1
9.2
8.7
8.7
9.3
9.3
8.8
8.3
8.1
8.0
a
b
3
Ϫ1 Ϫ1
Ϫ1
Colour forming reaction at 430 nm. Calculated from eqn. (9) with k 130 dm mol
s
, kϪ1kAmH /k 360 s , K = 0.0004 (1 + 20[Am]) and K
Am c,3 h
0 3 Ϫ1 3 Ϫ1
+
1
3
Ϫ1
c
d
2
0 dm mol . Identical values were obtained at 340 and at 430 nm. Calculated from eqn. (10) with K 0.36 dm mol and K 20 dm mol .
c
,
1
h
a
Ϫ3
Table 7 Kinetic and equilibrium results for reaction of 4b with pyrrolidine in acetonitrile containing 0.001 mol dm pyrrolidinium perchlorate at
2
5 ЊC
Ϫ3
b
c
3
Ϫ1
d
Ϫ1
e
Ϫ1
[
Pyrrolidine]/mol dm
A (415 nm)
Kc,3 /dm mol
ksub /s
kcalc /s
0
0
0
0
0
0
0
0
0
0
.004
.005
.006
.008
.010
.020
.030
.050
.070
.100
—
—
—
—
—
—
—
0.064
0.090
0.115
—
—
—
—
—
—
—
0.36
0.41
0.58
1.72
2.56
3.52
5.89
8.29
22.5
35.6
46.4
42.0
28.7
1.70
2.54
3.50
5.72
8.23
22.6
35.3
45.5
41.1
29.6
a
Ϫ5
Ϫ3 b
Concentration of 4b is 4 × 10 mol dm
. At completion of reaction forming 5b. Measurements in the absence of added pyrrolidinium perchlor-
c
ate give a value of 0.135 for complete conversion. Final values at completion of the substitution reaction are 0.056. Calculate A/(0.135 Ϫ A)-
+
2
d
Ϫ3
Ϫ3
[
AmH ]/[Am] . Measured at 415 nm. Colour forming reaction when [pyrrolidine] р 0.030 mol dm , fading when [pyrrolidine] > 0.030 mol dm .
5 6 Ϫ2 Ϫ1 3 Ϫ1 3 Ϫ1
e
Calculated from eqn. (12) with K1kAm 1.3 × 10 dm mol
s , kAm/kϪ1 55 dm mol and Kc,3 0.155 (1 + 27[Am]) dm mol .
supplementary information in Table 13.† Failure to observe the
intermediates, 7b, on the reaction pathway is attributed to their
rapid cleavage by loss of phenoxide. It is known that phenoxide
buffered. Fitting of the rate constants for the substitution reac-
6 Ϫ2 Ϫ1
tion with eqn. (12) yielded values of K k 5200 dm mol s ,
1
Am
3
Ϫ1
3
Ϫ1 Ϫ1
kAm/kϪ1 8 dm mol , and hence k 650 dm mol s .
1
3
1
ions are considerably better leaving groups than alkoxide ions,
explaining the contrast in behaviour with the ethyl ether 4a. A
Data in Table 9 for reaction with butylamine in the absence
of added salt yielded values of K . Rate constants for the sub-
c,3
1
3
3
similar difference has been observed in DMSO as solvent.
stitution reaction give a good fit with eqn. (13) with k 183 dm
1
Ϫ1 Ϫ1
Hence in the substitution pathway, shown in Scheme 2, form-
mol s .
ation of the adducts 7b becomes rate determining with
k4 ӷ kAmH
+
. The rate expression for substitution is eqn. (12).
Comparison of rate and equilibrium data
The results are summarised in Table 10 where they are com-
pared with corresponding values in DMSO.
2
K1kAm[Am]
ksub
=
(12)
(13)
2
kAm[Am]
K [Am]
[AmH ]
c,3
1
+
1 +
+
ͩ
ͪͩ
ͪ
Comparison of substrates
kϪ1
stoich
With each amine the most rapid reaction observed in
acetonitrile is attack at an unsubstituted ring position to give
k [Am]
1
ksub
=
2
0
K [Am]
the corresponding 3-adduct. Values of Kc,3 decrease with sub-
c,3
1
+
+
ͩ
ͪ
strate in the order 1 > 4b > 4a. There is evidence that the pres-
[
AmH ]
stoich
ence of a bulky 1-substituent forces the nitro-groups at the 2-
2
,32
Our results indicate that for reactions with pyrrolidine and
piperidine proton transfer is partially rate limiting in the form-
ation of 7b, and hence in the substitution pathway. However
with butylamine nucleophilic attack is the slow step
and 6-positions from the ring-plane, thus reducing their elec-
tron withdrawing capacity, and hence decreasing the value of
the equilibrium constant. This is counterbalanced to some
extent by the favourable electronic effects of the phenoxy and
ethoxy substituents for which the σmeta values are 0.25 and 0.10,
kAm[Am] ӷ k , so that eqn. (12) reduces to eqn. (13).
Ϫ1
3
3
Data for reaction with pyrrolidine in the presence of pyr-
rolidinium perchlorate are in Table 7. At the highest amine con-
centrations used there was appreciable initial conversion to the
-adduct, 5b, allowing calculation of values of K . Rate con-
stants for the substitution reaction give an excellent fit with eqn.
respectively. Important factors influencing the values of k
1
for
3
4,35
ipso-attack are expected
to be the electronegativity of the 1-
substituent and also its steric bulk which will affect the F strain
associated with approach of the amine. Comparisons of 4a and
4b give ratios close to unity; 0.6 for pyrrolidine, 0.7 for
butylamine and 1.2 for piperidine. Hence these factors are
evenly balanced.
3
c,3
5
6
Ϫ2 Ϫ1
3
(
12) with K k 1.3 × 10 dm mol
mol . Combination of these values yields k 2400 dm mol
s
and kAm/kϪ1 55 dm
1
Am
Ϫ1
3
Ϫ1
1
Ϫ1
s .
The results for reaction with piperidine are in Table 8. The
presence of piperidinium perchlorate largely inhibited the for-
mation of the 3-adduct, 5b. Absorbance measurements in the
absence of added salt allowed the estimation of values of Kc,3
although these are subject to error since the solutions are not
Comparison of amines
2
0,22
The pK values in acetonitrile
are butylamine 18.26, piperi-
a
dine 18.92 and pyrrolidine 19.58. There is little steric hindrance
to attack at the 3-position and values of K reflect the basicities
of the amines; values are between four and ten times larger for
0
c,3
J. Chem. Soc., Perkin Trans. 2, 1997
373

View Article Online
Table 8 Kinetic and equilibrium data for reaction of 4b with piperidine in acetonitrile at 25 ЊC
[
Piperidine]/
[Piperidinium
perchlorate]/mol dm
Ϫ3
Ϫ3
a
b
3
Ϫ1
c
Ϫ1
d
Ϫ1
mol dm
A (430 nm)
Kc,3 /dm mol
ksub /s
kcalc /s
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.020
.030
.050
.070
.10
—
—
—
—
0.069
0.096
0.132
0.152
0.17
—
—
—
—
—
0.024
0.027
0.032
0.038
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
.01
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.47
0.73
1.03
1.40
1.78
3.81
9.20
14.4
19.0
0.48
0.74
1.04
1.39
1.77
3.69
8.60
13.8
19.5
.0125
.015
.0175
.020
.030
.050
.07
—
—
—
—
.10
—
a
b
+
2
+
c
At completion of reaction forming 5b. Calculated as A/(0.18 Ϫ A)[AmH ]/[Am] , where [AmH ] = [5b]. Colour forming reaction at 353 nm.
d
6
Ϫ2 Ϫ1
3
Ϫ1
3
Ϫ1
Calculated from eqn. (12) with K1kAm 5200 dm mol
s , kAm/kϪ1 8 dm mol and Kc,3 0.016 (1 + 25 [Am]) dm mol .
a
Table 9 Kinetic and equilibrium results for reaction of 4b with butylamine in acetonitrile at 25 ЊC
c
Ϫ3
[
Butylamine]/
[Butylammonium
perchlorate]/mol dm
Kc,33 /10
Ϫ3
Ϫ3
b
Ϫ1
d
Ϫ1
e
Ϫ1
mol dm
A (430 nm)
dm mol
ksub /s
kcalc /s
0
0
0
0
0
0
0
0
0
.02
.04
.07
.50
.01
.02
.05
.07
.10
—
—
—
—
0.001
0.001
0.001
0.001
0.001
0.019
0.044
0.109
0.181
—
—
—
—
—
1.2
2.0
2.8
—
—
—
—
—
—
—
—
—
—
1.81
3.65
9.20
12.7
17.5
—
—
—
—
1.83
3.66
9.10
12.6
17.6
a
Ϫ5
Ϫ3
b
c
+
2
Concentration of 4b is 4 × 10 mol dm
.
At completion of reaction forming 5b. Calculated as A/(0.18 Ϫ A)[AmH ]/[Am] , where
+
d
e
3
Ϫ1 Ϫ1
Ϫ4
3
[
AmH ] = [5b]. Colour forming reaction at 418 nm. Calculated from eqn. (13) with k 183 dm mol
s
and Kc,3 9.4 × 10 (1 + 20[Am]) dm
1
Ϫ1
mol
.
1
3
reaction with pyrrolidine than with piperidine, and between
four and twenty times larger for reaction with piperidine than
that observed in DMSO. The phenoxide ion is a much better
31
leaving group than ethoxide so that k > k
+
and formation
4
AmH
with butylamines. Values of k also reflect the basicities of the
of the intermediates 7b becomes rate determining in the substi-
tution process. The reaction with butylamine is first order in
amine indicating that nucleophilic attack is rate limiting,
1
amines. However amine basicity is not the major factor in
0
c,1
determining values of K
which decrease in the order
butylamine > pyrrolidine > piperidine. This order is likely to
kAm[Am] ӷ k . However the reactions with pyrrolidine and
Ϫ1
reflect the increasing steric congestion in the 1-adducts, 7a,
piperidine show base catalysis, so that proton transfer (from
zwitterions 6b to amine) is rate-limiting or partially so,
kϪ1 у kAm[Am]. This is probably due to the lower values of kAm
expected for reaction with the secondary amines resulting from
steric hindrance to proton transfer. Values of kϪ1 for the sec-
ondary amines are also likely to be higher due to steric strain in
the zwitterions 6b. An analogous situation is observed in the
formation of the adducts 7a from the ethyl ether; the rate-
limiting step changes from nucleophilic attack to proton trans-
fer as the amine is changed from primary to secondary.
which contain two bulky groups at the 1-position. It is however
0
noteworthy that for each amine values of K are larger than
c,1
0
c,3
values of K ; the ratios are 900 for butylamine, 31 for pyrroli-
dine and 17 for piperidine. The thermodynamic preference for
the 1-adducts can be attributed to the polar effect of the alkoxy
group and to the relief of strain present in the parent when the
alkoxy group is twisted from the ring-plane. The kinetic prefer-
ence for attack at the 3-position and thermodynamic preference
for the 1-adducts may be described as K3T1 in Buncel’s
3
6
nomenclature.
Solvent effects
Mechanism of substitution
It is particularly interesting to compare data in acetonitrile with
those observed in DMSO (Table 10). Values of the overall equi-
In the reactions of the ethyl ether, 4a, the intermediates, 7a, on
the substitution pathway are spectroscopically observable.
Reprotonation of the intermediates is more rapid than leaving
0
0
librium constants for adduct formation, K and K , are con-
c,3
c,1
siderably higher in DMSO than in acetonitrile. The major fac-
tor here is probably the greater ability of DMSO than
group departure, kAmH
+
> k , so that general acid catalysed loss
4
1
9,20
of the ethoxy group becomes rate determining. The substitu-
acetonitrile to solvate charged species.
The ratios of values
in the two solvents are ca. 10 for reactions involving
2
,9
5
tion conforms to the SB-GA mechanism. The results in Table
0showthat k (butylamine) > (pyrrolidine)reflectingthegreater
4
4
1
butylamine, ca. 3 × 10 with piperidine and ca. 10 with pyrroli-
4
acidity of the butylammonium ion. Steric factors are also likely
to be important so that the approach of the butylammonium
ion to 7a is less hindered than approach of the pyrrolidinium
dine. DMSO is known to be a particularly good hydrogen-bond
2
3
acceptor and this trend may reflect the greater relative stabil-
isation, through hydrogen-bonding interactions, of the primary
ammonium ions than of the secondary ammonium ions in this
solvent. The need for stabilisation of the substituted ammo-
nium ions in acetonitrile is evidenced by the observation of
their homoconjugation with the parent amines. This interaction
is not observed in DMSO.
ion. In fact the k step involving the piperidinium ion is so slow
4
that alternative reaction pathways dominate.
By contrast in reactions of the phenyl ether, 4b, there was
no evidence for the accumulation of intermediates, 7b, on the
substitution pathway. The situation is qualitatively similar to
3
74
J. Chem. Soc., Perkin Trans. 2, 1997

View Article Online
a
Table 10 Summary of kinetic and equilibrium data in acetonitrile (MeCN) and comparison with values in DMSO
Reaction with butylamine
MeCN
0.0025
DMSO
1000
MeCN
0.0004
0.36
130
>1.8 × 10
8500
DMSO
15
5 × 10
250
>50
8.3
MeCN
0.00094
DMSO
210
0
3
Ϫ1
Ϫ1
K
K
/dm mol
c,3
0
3
4
/dm mol
c,1
3
Ϫ1 Ϫ1
k /dm mol
s
183
410
1
b
3
Ϫ1 Ϫ1
5
k AmH
+
/dm mol
s
3
Ϫ1 Ϫ1
k /dm mol
s
4
Reaction with pyrrolidine
0
3
Ϫ1
K
/dm mol
0.20
3500
1 × 10
3000
0.009
70
0.155
1300
c,3
6
Ϫ2 Ϫ1
6
7
K3kAm/dm mol Ϫ1s
2 × 106
c
3
Ϫ1
k
K
+
/dm mol
s
9 × 10
AmH
0
3
Ϫ1
/dm mol
0.28
1300
86
3.9 × 10
200
2000
4000
30
60
0.25
c,1
3
Ϫ1 Ϫ1
k /dm mol
s
2400
55
10000
20
1
3
Ϫ1
kAm/k dm mol
k AmH
Ϫ1
b
3
Ϫ1 Ϫ1
5
+
/dm mol
s
3
Ϫ1 Ϫ1
k /dm mol
s
4
Reaction with piperidine
0
3
Ϫ1
K
/dm mol
0.055
2140
6 × 10
280
0.0015
27
0.016
400
c,3
6
Ϫ2 Ϫ1
5
5
K3kAm/dm mol Ϫ1s
1.4 × 106
c
3
Ϫ1
k
K
+
/dm mol
s
2.5 × 10
AmH
0
3
Ϫ1
/dm mol
0.025
780
8
600
1800
3
c,1
3
Ϫ1 Ϫ1
k /dm mol
s
650
8
5000
0.6
1
3
Ϫ1
kAm/k dm mol
k AmH
Ϫ1
b
3
Ϫ1 Ϫ1
5
+
/dm mol
s
2.4 × 10
9
a
All values are at 25 ЊC. Data in DMSO are from refs. 4, 5, 11, 13, 29. The subscript 3 indicates reaction at an unsubstituted ring position, and the
b
c
subscript 1 reaction at the 1-position of 4a or 4b. Statistical corrections have not been applied. For reaction at the 1-position, see Scheme 2. For
reaction at the unsubstituted 3-position, see Scheme 1.
The results in acetonitrile lead to the direct determination
of values of kAmH , the rate constants for protonation of
factor of 500 for kAm/k
+
in acetonitrile yields values for
Ϫ1 Ϫ1
AmH
3
9
+
kAm of (1 to ca. 5) × 10 dm mol
stituted ring position and (1 to ca. 2) × 10 dm mol
reaction is at a substituted ring position. The former value
approaches the diffusion limit (2 × 10 dm mol s ) which is
consistent with proton transfers that are in the thermo-
dynamically favoured direction. There is likely to be some steric
hindrance when reaction involves attack at a 1-substituted pos-
ition, accounting for the somewhat lower values observed in
formation of 7a.
s
for reaction at an unsub-
8 3 Ϫ1 Ϫ1
anionic adducts by substituted ammonium ions. The values
s
when
4
are ca. 10 larger in acetonitrile than in DMSO, when reaction
1
0
3
Ϫ1 Ϫ1
is both at an unsubstituted and at a substituted ring position.
There are however no major differences between the two
solvents in values of k , the rate constant for nucleophilic
1
attack by the amine at the 1-position, or in values of the
ratio kAm/k . The former values are lower by a factor of ca.
Ϫ1
2
–4 in acetonitrile, while the latter ratios are larger by a
factor of ca. 3–10. Although comparisons are not possible
from experimental data we would, similarly, not expect large
differences between the two solvents in values of k3 or of
the ratio kAm/kϪ3 involving reaction at unsubstituted ring
positions.
In fact, the value of 500 for kAm/kAmH may be somewhat too
+
large in acetonitrile, since the reaction involves formation of a
free substituted ammonium ion from a zwitterion. Nevertheless
the conclusion is that both values of kAmH and kAm are con-
siderably lower in DMSO than in acetonitrile. It has previously
been reported that values of rate constants for thermodynamic-
ally favourable proton transfers in DMSO are often well below
that expected for diffusion-controlled reaction. Steric hindrance
is thought to be an important factor contributing to this
+
0
0
The decrease in values of K (≡ k1kAm/k k
+
) and K
c,1
Ϫ1 AmH
c,3
(
≡ k3kAm/k k
+
) on transfer from DMSO to acetonitrile are
Ϫ3 AmH
largely explicable in terms of corresponding increases in values
of kAmH . However, it is interesting to speculate on the solv-
+
6
,37–39
ent dependence of kAm, the rate constant for proton transfer
reduction.
However, this factor should similarly apply to
from zwitterionic adducts to amine. It has been shown that
reactions in acetonitrile. The lower values for rate constants for
proton transfer observed in DMSO probably reflect the strong
hydrogen-bonding between DMSO and the proton to be trans-
ferred. Hydrogen bonding in the zwitterions, depicted by 9,
reduces values of kAm and in the ammonium ions, depicted by
ratios kAm/kϪ1 (and kAm/k ) have little solvent dependence.
Ϫ3
Nevertheless, values of both kAm and kϪ1 (kAm and k ) may be
Ϫ3
4
0
much larger in acetonitrile than in DMSO.
The first step in adduct formation, Scheme 1, is zwitterion
formation and this involves production of charges. Hence it is
reasonable to expect that the equilibrium constant, K (≡ k /
10, reduces values of kAmH .
+
It is worth noting that solvent effects on rate constants for
proton transfer from phenols to the anthracene radical anion
1
1
kϪ1), for this process should be strongly solvent dependent,
4
1,42
whereas the equilibrium constant for the second step (≡ kAm
kAmH
argued previously that in DMSO the ratio kAm/kAmH
/
have been reported.
It was suggested that only the free
+
) should show little dependence on solvent. It has been
phenol, in equilibrium with phenol hydrogen-bonded to the
solvent, is able to effect proton transfer. Faster reaction in
acetonitrile than in DMSO reflects the weaker interaction with
the former solvent. A similar situation may apply in our systems
with proton transfer occurring only from non hydrogen-bonded
ammonium ions and zwitterions.
4
,6
+
has a
value of ca. 500 reflecting the greater acidity of the zwitterions
than of the parent ammonium ions. Since values of kAmH are
+
much larger in acetonitrile than in DMSO, this implies that
values of kAm will also be much larger in acetonitrile. Using the
J. Chem. Soc., Perkin Trans. 2, 1997
375

View Article Online
1
1
2 Y. Hasegawa, J. Chem. Soc., Perkin Trans. 2, 1985, 87.
3 R. Chamberlin and M. R. Crampton, J. Chem. Soc., Perkin Trans. 2,
995, 1831.
4 R. Chamberlin and M. R. Crampton, J. Chem. Soc., Perkin Trans. 2,
994, 425.
5 D. Ayediran, T. O. Bamkole, J. Hirst and I. Onyido, J. Chem. Soc.,
Perkin Trans. 2, 1977, 597; 1977, 1580.
6 T. O. Bamkole, J. Hirst and I. Onyido, J. Chem. Soc., Perkin Trans. 2,
1
1
1
1
1
1
1
1
1
979, 1317.
7 T. A. Emokpae, P. U. Uwakwe and J. Hirst, J. Chem. Soc., Perkin
Trans. 2, 1993, 125.
8 T. O. Bamkole, J. Hirst and I. Onyido, J. Chem. Soc., Perkin Trans. 2,
1
982, 889.
9 J. F. Coetzee and C. D. Ritchie, Solute Solvent Interactions, Marcel
Dekker, New York, 1969.
2
2
0 J. F. Coetzee, Progr. Phys. Org. Chem., 1967, 4, 45.
1 I. M. Kolthoff, S. Bruckenstein and M. K. Chantooni, J. Am. Chem.
Soc., 1961, 83, 3927; J. M. Kolthoff and M. K. Chantooni, 1965, 87,
4
428.
22 J. F. Coetzee and G. R. Padmanabhan, J. Am. Chem. Soc., 1965, 87,
005.
Since ratios kAm/kϪ1 are only slightly higher in acetonitrile
than in DMSO, the implication of higher kAm values is that kϪ1
values are also higher in acetonitrile. Hence the expected reduc-
tion in values of K reflects slightly smaller values of k with
5
2
3 M. J. Kamlet and R. W. Taft, J. Am. Chem. Soc., 1976, 98, 377;
L. Joris, J. Mitsky and R. W. Taft, J. Am. Chem. Soc., 1972, 94, 3438.
4 M. R. Crampton and V. Gold, J. Chem. Soc., 1964, 4293.
1
1
2
much higher values of k . The ratio kAm/k affects the
Ϫ1
Ϫ1
25 R. Foster and R. K. Mackie, Tetrahedron, 1965, 21, 3363.
26 P. Caveng and H. Zollinger, Helv. Chim. Acta, 1967, 50, 861.
27 E. Buncel and R. A. Manderville, J. Am. Chem. Soc., 1993, 115,
susceptibility of substitution reaction to base catalysis. The
similar values obtained for this ratio in acetonitrile and in
1
5–18
8
985; E. Buncel and R. A. Manderville, J. Phys. Org. Chem., 1993,
DMSO imply that, as observed,
reactions should generally
6
, 71.
show a rather similar susceptibility to base catalysis in the two
solvents.
2
2
3
3
3
8 G. Briegleb, W. Liptay and M. Cantner, Az. Phys. Chem. N.F., 1960,
6, 55.
9 R. Chamberlin, M. R. Crampton and I. Robotham, J. Chem. Res. (S),
994, 408; (M), 1994, 2232.
0 R. Chamberlin and M. R. Crampton, J. Chem. Res. (S), 1993 106;
M), 1993, 811.
1 C. F. Bernasconi and M. C. Muller, J. Am. Chem. Soc., 1978, 100,
536.
2
1
Acknowledgements
(
This work was carried out with the support of the Defence
Research Agency. We thank Dr R. Millar for helpful
discussions.
5
2 C. Grammacioli, R. Destro and M. J. Simonetta, J. Chem. Soc.,
Chem. Commun., 1967, 331; Acta Crystallogr., Sect. B, 1968, 24,
1
29.
References
3
3 D. H. McDaniel and H. C. Brown, J. Org. Chem., 1958, 23, 420.
1
2
3
M. R. Crampton and V. Gold, Chem. Commun., 1965, 549; J. Chem.
Soc. B, 1967, 23.
F. Terrier, Nucleophilic Aromatic Displacement, VCH, New York,
34 G. Baldini, G. Doddi, G. Illuminati and F. Stegel, J. Org. Chem.,
1976, 41, 2153.
35 R. Chamberlin, M. R. Crampton and R. L. Knight, J. Chem. Res. (S),
1993, 444; (M), 1993, 2986.
36 E. Buncel, J. M. Dust, A. Jonczyk, R. A. Manderville and I. Onyido,
J. Am. Chem. Soc., 1992, 114, 5610; R. A. Manderville and
E. Buncel, J. Am. Chem. Soc., 1993, 115, 8985.
1
991.
E. Buncel, M. R. Crampton, M. J. Strauss and F. Terrier, Electron
Deficient Aromatic- and Heteroaromatic-base Interactions, Elsevier,
Amsterdam, 1984.
4
5
6
M. R. Crampton and B. Gibson, J. Chem. Soc., Perkin Trans. 2,
37 C. F. Bernasconi and D. J. Carre, J. Am. Chem. Soc., 1979, 101,
2698.
38 C. F. Bernasconi, J. P. Fox and S. Fornarini, J. Am. Chem. Soc., 1980,
102, 2810.
39 C. D. Ritchie, J. Am. Chem. Soc., 1983, 105, 3573.
40 C. D. Ritchie and S. Lu, J. Am. Chem. Soc., 1989, 111, 8542.
41 M. F. Nielsen, Acta Chem. Scand., 1992, 46, 533.
42 M. F. Nielsen and O. Hammerich, Acta Chem. Scand., 1992, 46, 883.
1
981, 533.
M. R. Crampton and C. Greenhalgh, J. Chem. Soc., Perkin Trans. 2,
983, 1175.
C. F. Bernasconi, M. C. Muller and P. Schmid, J. Org. Chem., 1979,
4, 3189.
1
4
7
8
9
M. R. Crampton, J. Chem. Soc., 1971, 2112.
E. Buncel and J. G. K. Webb, Can. J. Chem., 1974, 52, 630.
J. A. Orvik and J. F. Bunnett, J. Am. Chem. Soc., 1970, 92, 2417.
1
0 J. F. Bunnett, S. Sekiguchi and L. A. Smith, J. Am. Chem. Soc., 1981,
03, 4865.
1 M. R. Crampton and P. Routledge, J. Chem. Soc., Perkin Trans. 2,
Paper 6/03707K
Received 28th May 1996
Accepted 17th July 1996
1
1
1
984, 573.
3
76
J. Chem. Soc., Perkin Trans. 2, 1997