Effects of ortho- and para-ring activation on the kinetics of S NAr reactions of 1-chloro-2-nitro- and 1-phenoxy-2-nitrobenzenes with aliphatic amines in acetonitrile

Article abstract of DOI:10.1002/ejoc.200500774

Rate constants are reported for reaction of 4-substituted 1-chloro-2,6-dinitrobenzenes 1, 6-substituted 1-chloro-2,4-dinitrobenzenes 2, and some of the corresponding 1-phenoxy derivatives, 3 and 4, with n-butylamine, pyrrolidine and pi-peridine in acetoni

Full text of DOI:10.1002/ejoc.200500774

FULL PAPER
DOI: 10.1002/ejoc.200500774
Effects of ortho- and para-Ring Activation on the Kinetics of SNAr Reactions
of 1-Chloro-2-nitro- and 1-Phenoxy-2-nitrobenzenes with Aliphatic Amines in
Acetonitrile
Michael R. Crampton,*[a] Thomas A. Emokpae,[b] Chukwuemeka Isanbor,*[b]
Andrei S. Batsanov,[a] Judith A. K. Howard,[a] and Raju Mondal[a]
Keywords: Kinetics / Nucleophilic aromatic substitution / Steric hindrance / Substituent effects
Rate constants are reported for reaction of 4-substituted 1-
chloro-2,6-dinitrobenzenes 1, 6-substituted 1-chloro-2,4-di-
nitrobenzenes 2, and some of the corresponding 1-phenoxy
derivatives, 3 and 4, with n-butylamine, pyrrolidine and pi-
peridine in acetonitrile as solvent. Values of k1, the rate con-
stant for nucleophilic attack at the 1-position, increase with
increasing ring-activation but may be reduced by steric re-
pulsion at the reaction centre which increases in the order Cl
Ͻ OPh, and n-butylamine Ͻ pyrrolidine Ϸ piperidine. ortho-
Substituents may also have adverse steric effects, and those
of the trifluoromethyl group are particularly serious. X-ray
crystal structures of phenyl 2,4-dinitro-6-trifluoromethyl-
phenyl ether and phenyl 2,6-dinitro-4-trifluoromethylphenyl
ether are reported. Base catalysis in the 1-phenoxy deriva-
tives is attributed to rate-limiting proton transfer from the
zwitterionic intermediates 6 to base. Values of rate constants
for this process decrease with increasing steric congestion at
the reaction centre and in the order n-butylamine Ͼ pyrroli-
dine Ͼ piperidine.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2006)
the expulsion of the leaving group from the deprotonated
form of Z, the SB-GA mechanism.
Introduction
The study of mechanisms and reactivity in aromatic nu-
The latter mechanism has been shown to apply to sub-
cleophilic substitution reactions has interested chemists for strates such as alkyl ethers carrying poor leaving
some years[1,2] and continues to attract attention.[3–6] The groups.[7–10] However, there is good evidence that for sub-
general mechanism for reactions involving amine nucleo- strates containing good leaving groups, such as phenyl
philes is shown in Scheme 1, and early studies were con- ethers, the RLPT mechanism applies.[11–15] The absence of
cerned with discerning the mechanism of the general base- base catalysis implies[1] that the initial nucleophilic attack,
catalysed step, kB[B]. It is now recognised that in dipolar k1 step, is rate limiting.
aprotic solvents, such as dimethyl sulfoxide (DMSO) and
Several previous studies have reported the effects of vary-
acetonitrile, the rate-limiting proton transfer may be from ing the electron-withdrawing substituents in the aromatic
the zwitterionic intermediate Z to base, the RLPT mecha- ring and/or the nature of the amine.[1,2] Electronic, steric
nism; or, it may involve general acid catalysis, by BH+, of and hydrogen-bonding effects have been identified as im-
Scheme 1.
portant factors influencing reactivity. However these studies
[a] Chemistry Department, Durham University,
have often been piecemeal in nature rather than en-
Durham, DH1 3LE, UK
compassing an extended range of compounds.[16–20] Here
[b] Chemistry Department, University of Lagos,
Lagos, Nigeria
we report kinetic studies of the reactions in acetonitrile of
a series of 4-substituted 1-chloro-2,6-dinitrobenzenes 1, 6-
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
1222
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 1222–1230

Kinetics of an Aromatic Nucleophilic Substitution Reaction
FULL PAPER
substituted 1-chloro-2,4-dinitrobenzenes 2, and some of the servation of the adduct analogous to 5. In the present work,
corresponding 1-phenoxy derivatives 3 and 4 with n-bu- reactions of the most strongly activated substrates 1c,d and
tylamine, pyrrolidine and piperidine. Our aims were to (i) 2e,f were measured in the presence of a low concentration,
examine the relative activating power of substituents at po- 0.001 moldm–3, of the perchlorate salt of the amine under
sitions ortho and para to the reaction centre, and (ii) com- investigation to remove the possibility of reaction at unsub-
pare the values of rate constants for the individual steps in stituted ring positions, while in the reactions of 3b,c,d and
the reaction pathway.
4b,e amine concentrations were Յ 0.05 moldm–3. Under
these conditions substitution proceeded smoothly to give
first-order kinetics without the observation, in spectroscopi-
cally measurable concentrations, of transient species analo-
gous to 5 or of intermediates on the substitution pathway.
Making the usual assumption that the zwitterionic ad-
duct Z may be treated as a steady-state intermediate leads,
when the amine acts both as the nucleophile and as the
catalysing base, to Equation (1).
(1)
For the 1-chloro-substituted reactants 1 and 2 values of
the second-order rate constant, kA, were independent of the
amine concentration. This corresponds to the condition k2
+ kAm[Am] ϾϾ k–1, so that kA = k1. This is not unexpected
as base catalysis with chloride ion as the leaving group has
only been found in a few special cases,[22] where factors such
as the nature of the entering amine and the solvent contrib-
ute to a lowering of the ratio k2/k–1. Values of kobs, the first-
order rate constant, are available as supplementary infor-
mation in Tables S1–S6 (for supporting information see also
the footnote on the first page of this article), and the sec-
ond-order rate constants giving values of k1 are collected in
Table 1 and Table 2. Before discussing these data, results for
3 and 4 will be reported.
Results and Discussion
The reactions of parent molecules 1–4 with n-butylamine,
pyrrolidine or piperidine in acetonitrile gave the expected
products of substitution of chloride or phenoxide, respec-
tively, in Ͼ95% yield. Kinetic measurements were made
with the concentration of amine in large excess of the par-
ent concentration, ca. 1·10–4 moldm–3, and first-order ki-
netics were observed. Previous studies[11] in DMSO have
shown that substitution may be preceded by the formation,
under kinetic control, of adducts resulting from attack at
unsubstituted ring positions. An example for 3f, is shown in
Scheme 2. The values of the equilibrium constant for such
processes depend on the degree of ring activation and also
on the solvent.
Table 1. Values of rate constants, k1 [dm3 mol–1 s–1], for reaction[a]
of 1 and 2 with aliphatic amines in acetonitrile at 25 °C.
Substrate, R
n-Butylamine
Pyrrolidine
Piperidine
k1 [dm3 mol–1 s–1]
1a, 4-H
0.015(1)
3.3(1)
4.1(1)
26.1(1)
86(1)
0.0089(1)
0.50(1)
32.5(1)
86(1)
0.055(3.7)
21.6(5.5)
29.7(7.2)
233(8.9)
1000(11.6)
1.31(147)
2.85(5.7)
5690(175)
1000(11.6)
0.023(1.5)
10.8(3.3)
16.4(4.0)
106(4.1)
640(7.4)
0.52(58)
0.85(1.7)
1880(58)
640(7.4)
1b, 4-CF3
1c, 4-CO2Me
1d, 4-CN
1f, 4-NO2
2a, 6-H
Thus in acetonitrile the values for the reaction in
Scheme 2 are ca. 104 smaller than in DMSO.[11] The domi-
nant factor here is likely to be the greater ability of DMSO
than of acetonitrile to solvate the ionic reaction products.
The reaction of 4e with pyrrolidine in acetonitrile was
2b, 6-CF3
2e, ring N
2f, 6-NO2
[a] Values in parentheses are the reactivities for a particular com-
pound relative to that for n-butylamine; i.e. k1(pyrrolidine)/k1(n-
shown[21] to yield the substitution products without the ob- butylamine) and k1(piperidine)/k1(n-butylamine).
Scheme 2.
Eur. J. Org. Chem. 2006, 1222–1230
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1223

M. R. Crampton et al.
FULL PAPER
Table 2. Effects of ring-substituents on the relative reactivities, k1 kA vs. [amine], not shown, was curved. Values, in Table 3,
values, of 1 and 2 with aliphatic amines in acetonitrile at 25 °C.[a]
gave a good fit with Equation (2). That this curvature is
indicative of general base catalysis is shown by the catalysis
of the reaction by DABCO (Table 3).
Substrate, R
n-Butylamine
Pyrrolidine
Piperidine
1a, 4-H
1
220
1
393
1
470
1b, 4-CF3
1c, 4-CO2Me
1d, 4-CN
1f, 4-NO2
2a, 6-H
Table 3. Kinetic results[a] for the reactions of 4b with n-butylamine
in acetonitrile at 25 °C.
273
540
732
1.7·103
5.7·103
1
4.2·103
1.8·104
1
4.6·103
2.8·104
1
[b] [c]
[n-Butylamine] [DABCO]
kA
kA
(calcd.)
[10–3 moldm–3] [10–3 moldm–3] [dm3 mol–1 s–1] [dm3 mol–1 s–1]
2b, 6-CF3
2e, ring N
2f, 6-NO2
56
2.2
1.6
1.0
1.5
2.0
2.5
3.0
3.5
4.0
6.0
2.0
2.0
2.0
2.0
2.0
–
–
–
–
–
–
–
–
3.0
5.0
10
15
20
0.35
0.54
0.64
0.75
0.86
0.89
0.97
1.15
0.98
1.10
1.27
1.43
1.52
0.37
0.51
0.63
0.73
0.81
0.89
0.96
1.16
0.94
1.08
1.30
1.44
1.53
3.6·103
9.7·103
4.3·103
760
3.6·103
1.2·103
[a] For a given amine values are compared to the reactivity with
1a, and 2a, respectively.
The general reaction scheme for the 1-phenoxy com-
pounds is shown in Scheme 3. However, base catalysis, as
argued previously,[11–13,15] is indicative of rate-limiting pro-
ton transfer from the zwitterionic intermediates 6, in con-
trast to general acid catalysis of phenoxide expulsion.
Hence Equation (1) applies. Limiting forms, where K1 = k1/
k–1, are Equation (2), if the uncatalysed pathway can be
neglected, and Equation (3) if the condition k–1 ϾϾ k2 +
kAm[Am] applies.
[a] Measured at 345 nm. [b] Values calculated with Equation (2)
with K1kAm 450 dm6 mol–2 s–1 and kAm/k–1 220 dm3 mol–1. [c] Values
in the presence of DABCO calculated with Equation (4) with
K1kAm 450 dm6 mol–2 s–1, kAm/k–1 220 dm3 mol–1, K1kDABCO
280 dm6 mol–2 s–1 and kDABCO/k–1 140 dm3 mol–1.
The reactions with pyrrolidine gave evidence for base ca-
talysis. Plots of values of kA vs. amine concentration, not
shown, were, except for 4b, curved and gave good fits with
Equation (2). Specimen results are in Table 4 and other data
are in supplementary Table S8. For 4b the plot was linear
through the origin indicating that the value of kAm/k–1 is
(2)
kA = K1kAm[Am] + K1k2
(3)
Reactions with n-butylamine of 3a–d and 4a,e gave val- relatively low (Ͻ 5 dm3 mol–1).
ues of kobs which increased linearly with the amine concen-
The progression from pyrrolidine to piperidine resulted
tration, indicating that nucleophilic attack, the k1 step, is in lower values for the parameter kAm/k–1, so that for 3a–d
rate limiting. Conditions used are given in Table S7. How- and 4b Equation (3) was applicable. Linear plots of kA vs.
ever, for 4b carrying an ortho-CF3 substituent the plot of [piperidine] yielded values of K1kAm. Only in the case of 3a
Scheme 3.
1224
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 1222–1230

Kinetics of an Aromatic Nucleophilic Substitution Reaction
FULL PAPER
Table 4. Kinetic data[a] for reactions of 3 and 4 with pyrrolidine in acetonitrile at 25 °C.
3b
3c
4a
[Pyrrolidine]
kA
kcalcd.
kA
kcalcd.
kA
kcalcd.
[10–3 moldm–3]
[dm3 mol–1 s–1]
[dm3 mol–1 s–1]
[dm3 mol–1 s–1]
1.0
1.5
2.0
2.5
3.0
4.0
5.0
8.0
10
0.12
0.18
0.23
0.29
0.33
–
0.55
0.88
1.06
–
0.12
0.17
0.23
0.29
0.34
–
0.56
0.87
1.07
–
0.35
–
0.70
–
0.99
1.2
1.5
–
2.9
5.5
6.8
0.34
–
0.67
–
0.99
1.3
1.6
–
2.9
5.0
6.6
0.025
–
0.041
–
0.059
0.075
0.087
–
0.16
0.23
0.27
0.024
–
0.045
–
0.064
0.081
0.096
–
0.15
0.22
0.25
20
30
–
–
[a] Values of kcalcd. were calculated using Equation (2) with the values collected in Table 5.
Table 5. Summary of results[a] for reaction of 3 and 4 with aliphatic amines in acetonitrile at 25 °C.
Substrate, R
n-Butylamine
Pyrrolidine
Piperidine
3a 4-H
k1 [dm3 mol–1 s–1]
0.047(1)
–
–
–
5.6(1)
–
–
9.3(1)
–
–
45(1)
–
0.024(0.51)
3.5
0.083
–
12(2.1)
10
120
17.5(1.9)
20
350
450(10)
7
3200
0.37(76)
70
–
–
kAm/k–1 [dm3 mol–1]
K1kAm [dm6 mol–2 s–1]
K1k2 [dm3 mol–1 s–1]
k1 [dm3 mol–1 s–1]
0.0011
6·10–5
–
–
3.3
–
–
10.2
–
–
112
0.16(33)
5.0
0.80
–
3b 4-CF3
3c 4-CO2Me
3d 4-CN
kAm/k–1 [dm3 mol–1]
K1kAm [dm6 mol–2 s–1]
k1 [dm3 mol–1 s–1]
kAm/k–1 [dm3 mol–1]
KK1kAm [dm6 mol–2 s–1]
k1 [dm3 mol–1 s–1]
kAm/k–1 [dm3 mol–1]
K1kAm [dm6 mol–2 s–1]
k1 [dm3 mol–1 s–1]
–
4a 6-H
4.9·10–3(1)
kAm/k–1 [dm3 mol–1]
K1kAm [dm6 mol–2 s–1]
k1 [dm3 mol–1 s–1]
26
Ͼ2(Ͼ1)
Ͻ5
4b 6-CF3
4e ring N[b]
2.0(1)
220
450
0.95(1)
–
–
183(1)
–
–
kAm/k–1 [dm3 mol–1]
K1kAm [dm6 mol–2 s–1]
k1 [dm3 mol–1 s–1]
–
11.3
0.30
24(25)
100
2440
650(3.6)
8
85(90)
1300
1.1·105
2400(13)
55
kAm/k–1 [dm3 mol–1]
K1kAm [dm6 mol–2 s–1]
k1 [dm3 mol–1 s–1]
[c]
4f 6-NO2
kAm/k–1 [dm3 mol–1]
K1kAm [dm6 mol–2 s–1]
1.3·105
5200
[a] Values in parentheses are the reactivities for a given compound relative to that for n-butylamine; i.e. k1(pyrrolidine)/k1(n-butylamine)
and k1(piperidine)/k1(n-butylamine). [b] Values for reaction with pyrrolidine from ref.[21]. [c] Values from ref.[13]
.
did the plot have an intercept indistinguishable from zero, disorder which we have rationalised as two orientations in
indicating that here the uncatalysed decomposition of the a 9:1 ratio. In each molecule, the two nitro groups and the
intermediate, k2 step, contributed to the reaction flux. Val- PhO group are inclined to the plane of the tetra-substituted
ues of kA are available in Table S9 and results are collected ring A in a propeller-like fashion. The sense of the twist is
in Table 5. For 4a and 4e, carrying no bulky substituent at opposite in the two independent molecules which are thus,
the 6-position, values of kAm/k–1 were larger so that the roughly, mirror images of each other. The O(1) atom is
data in Table 6 gave a good fit to Equation (2).
slightly tilted out of the ring A plane, to the opposite side
X-ray Crystal Structures: Our results for 4b indicate un- from the phenyl group B.
usual effects which may be due to the presence of the ortho-
Compound 4b crystallises in a non-centrosymmetric, al-
trifluoromethyl group. Hence the X-ray structure of 4b was beit not chiral, space group Pc, with two molecules per
determined and compared with that for 3b where the CF3 asymmetric unit, shown in Figure 3. Interestingly, these
group is in the para position. The structure of 4f, which has molecules are related by a pseudo-inversion centre (½ ¼
been briefly reported previously,[6] is shown in Figure 1.
½), as shown in Figure 3 (d). Although this operation ap-
Compound 3b crystallises in a chiral space group plies only to the approximate positions of the molecule as
P212121. The asymmetric unit comprises two molecules, a whole, rather than individual atomic positions, it produces
shown in Figure 2; in both the CF3 groups show rotational pseudo-systematic absences characteristic of the space
Eur. J. Org. Chem. 2006, 1222–1230
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1225

M. R. Crampton et al.
FULL PAPER
Table 6. Kinetic results[a] for reaction of 4a and 4e with piperidine superpositions of nitro and CF3 substituents are observed
in acetonitrile at 25 °C.
at both C(2) and C(6). With 85% probability, the nitro
group is attached to C(2) and the CF3 group to C(6), with
15% vice versa. The phenyl group B is tilted more towards
the ortho-nitro group and away from the CF3 group. Se-
lected bond lengths and angles are listed in Table 7.
4a
4e
[piperidine]/ kA
kcalcd.
kA
kcalcd.
10–3
[dm3 mol–1 s–1]
[dm3 mol–1 s–1]
moldm–3
1.0
3.0
4.0
5.0
8.0
10
–
–
–
–
2.2
6.1
–
8.1
–
2.2
5.6
–
8.1
–
0.0031
–
0.0059
–
0.0031
–
0.0061
–
12
15
20
40
60
80
100
0.0088
–
0.015
0.026
0.037
0.047
0.055
0.0089
–
0.014
0.026
0.036
0.045
0.053
12.1
14.8
16.4
–
–
–
12.2
14.6
16.3
–
–
–
–
–
[a] Values calculated using Equation (2) with the values given in
Table 5.
Figure 3. (a) Ordered molecule I in the structure of 4b; (b) major
and (c) minor orientations of the disordered molecule II; (d) over-
lap of molecules I and II by the pseudo-inversion symmetry.
Figure 1. Molecular structure of 4f. Henceforth the atomic dis-
placement ellipsoids are drawn at the 50%% probability level.
Molecular geometry can be compared with those of re-
cently studied sterically less encumbered analogues, di-
phenyl ether 8,[23] and p-nitrophenyl phenyl ether 9.[24] In
two different polymorphs of 8, the dihedral angle between
the benzene rings is similar, 87.6 and 88.4° (cf. 67.7° pre-
dicted by ab initio calculations[23]), in 9 it is reduced to
63.2°. The C–O–C angles, varying from 117.9 to 119.7°,
agree with our results. In 9 the C–O bond to the nitro-
substituted ring is shorter than the other [1.378(2) vs.
1.396(2) Å], which can be explained by electron-with-
drawing effect of NO2. However, similar difference was ob-
served in 8, viz. 1.381(2) vs. 1.392(2) Å in the monoclinic
form, 1.379(2) vs. 1.394(2) in the orthorhombic. Rather, the
C–O distance correlates with the C–C–O–C torsion angle,
which is close to 90° for the former ring (88.0 and 87.2°,
Figure 2. Independent molecules I (left) and II in the structure of
3b, projected on the plane of ring A. Minor (10%) orientations of
the CF3 groups are not shown.
group P21/c: reflections 0k0 with odd k are present but respectively) and close to 0 for the latter (3.6 and 4.8°).
weaker than the average by a factor of 10. One of the inde- Thus, shorter C–O distance reflects the conjugation of the
pendent molecules is disordered between two conforma- oxygen lone pair with the ring. Nevertheless, in 3b, 4b and
tions, which differ by a rotation of the entire C6H2(NO2)2- 4f the difference between C(1)–O(1) and C(7)–O(1) distance
CF3 moiety by 180° around the O(1)···N(2) vector, so that is much larger than in 8 and 9 and show no correlation with
1226
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 1222–1230

Kinetics of an Aromatic Nucleophilic Substitution Reaction
Table 7. Selected bond lengths and angles.
FULL PAPER
3b, I
3b, II
4b, I
4b, II
4f[b]
O(1)–C(1) [Å]
O(1)–C(7) [Å]
1.357(2)
1.410(2)
117.7(2)
1.357(3)
1.409(2)
118.7(2)
1.369(4)
1.411(4)
117.9(2)
1.364(4)
1.411(4)
117.7(2)
1.349(3)
1.407(2)
120.1(2)
C(1)–O(1)–C(7) [°]
dihedral angles [°]
ring A/2-nitro group
ring A/6-nitro group
ring A/4-nitro group
ring A/ring B
38.1
53.1
–
77.3
2.7
37.1
42.3
–
74.7
3.8
27.2
–
6.6
75.6
0.9
29.1
42[a]
10.9
89.7
1.9
46.1
46.3
15.4
66.0
4.3
O(1)–C(1) tilt from plane A [°]
torsion angles [°]
C(2)–C(1)–O(1)–C(7)
C(1)–O(1)–C(7)–C(12)
–60.1(3)
–32.4(3)
62.9(3)
23.3(3)
–79.7(4)
6.5(4)
–69.0(4)
–37.0(4)
–58.3(3)
–17.0(3)
[a] Minor position of the disordered group. [b] Ref.[6]
.
the torsion angles, indicative of the strong p-π conjugation creases. The pyrrolidine/piperidine reactivity ratio for all the
with the strongly electron withdrawing dinitro-substituted compounds in Table 1 is fairly constant, varying only be-
ring.
tween 1.6 to 3.3, indicating that here piperidine is not steri-
The steric interactions in 3b, 4b and 4f are accommo- cally disadvantaged relative to pyrrolidine.
dated by rotation of the ortho-nitro group(s) from the ring
The values in Table 5 show that for compounds 3 and 4,
plane and by a twist between the planes of the two aromatic carrying a phenoxy group at the position of nucleophilic
rings. In each structure there is evidence of steric crowding attack, values of the pyrrolidine/n-butylamine ratio are gen-
around the 1-position, which is at least as severe in 4b as in erally lower than those for correspondingly activated 1 and
3b and 4f.
2. For example, for 3a and 1a the ratios are 0.51 and 3.7,
respectively. These reductions may be attributed to increases
in the difficulty of approach to the 1-position by the amine
nucleophiles in 3 and 4, carrying a phenoxy group, com-
Comparisons
Rate Constants, k1, for Nucleophilic Attack: The results pared with corresponding attack on 1 and 2, carrying chlo-
in Table 1 involving nucleophilic attack at a ring-carbon rine. Since the steric requirements of the secondary amine
carrying chlorine show a reactivity order for k1 of pyrroli- are higher than those of the primary amine, the effect is
dine Ͼ piperidine Ͼ n-butylamine. This is the order com- more serious for pyrrolidine (and piperidine) than for n-
monly found in nucleophilic substitution reactions[1,2] and butylamine.
reflects the relative basicities of the amines in acetonitrile;
It is also instructive to compare k1 values for similarly
pKa values,[25] for the protonated amines are: pyrrolidine activated chloro and phenoxy compounds, in terms of the
19.58, piperidine 18.92, n-butylamine 18.26. The superior ratio of k1(Cl) to k1(OPh), which we have called “Cl/OPh”
reactivity of the secondary amines has also been attrib- in Table 8. For n-butylamine, where steric effects are likely
uted[8] to favourable ion-induced dipole interactions in the to be small, values of Cl/OPh are, with two exceptions, in
transition state between the partially positively charged ni- the range 0.25–0.6, indicating rather faster attack at the
trogen moiety and the polarisable alkyl substituents at- phenoxy-substituted carbon atom than at the chloro-substi-
tached to it.
tuted carbon atom. For pyrrolidine, the values are higher
For compounds 2a and 2e the pyrrolidine/n-butylamine (generally ca. 2). This is attributable to the greater steric
reactivity ratios are 147 and 175, respectively. These com- requirements of phenoxy than of chloro and of pyrrolidine
pounds carry hydrogen or ring-nitrogen at the 6-position so than of n-butylamine groups. The two notable exceptions
that steric interaction at the reaction centre with the incom-
ing nucleophile will be minimised. We note that a pyrroli-
dine/n-butylamine ratio of ca. 100 was found[3] for reaction
with 2-bromo-5-nitro-3-substituted thiophenes, where steric
effects should also be small. However, for compounds 1a–
f, carrying a nitro group at the 6-position, and for 2b, with
a 6-CF3 group, the pyrrolidine/n-butylamine ratio decreases
dramatically to values in the range 3.7–11.6. This decrease
is likely to reflect steric hindrance to approach of the sec-
ondary amine to the reaction centre. We note also that the
ratio increases consistently from 3.7 for 1a, the least acti-
vated compound, to 11.6 for 1f, the most activated. This
increase may reflect a shift to a somewhat “earlier” transi-
tion state, involving rather less steric interaction, as the
Table 8. Relative reactivities[a] of similarly activated 1-chloro- and
1-phenoxy compounds in their reactions with n-butylamine and
with pyrrolidine in acetonitrile.
Cl/OPh
n-butylamine
Substrates, R
pyrrolidine
1a/3a, 4-H
0.3
0.6
0.4
0.6
1.8
0.25
34
2.3
1.8
1.7
0.5
3.5
–
1b/3b, 4-CF3
1c/3c, 4-CO2Me
1d/3d, 4-CN
2a/4a, 6-H
2b/4b, 6-CF3
2e/4e, 6 ring N
2f/4f, 6-NO2
67
0.4
0.5
thermodynamic stability of the zwitterionic intermediate in- [a] Ratios compare k1 values, k1(Cl)/k1(OPh).
Eur. J. Org. Chem. 2006, 1222–1230
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1227

M. R. Crampton et al.
FULL PAPER
are compounds 2a and 4a and 2e and 4e which lack a bulky
The effects of ring activation in the series 4a–f are com-
substituent at the 6-position. Here the 1-chloro derivatives plicated by the reduced reactivity of 4a and 4e due to the
are more reactive than the 1-phenoxy derivatives. This can ground-state stabilisation referred to earlier.
be reasonably ascribed to reduced reactivity of the phenoxy
It is also worth considering the relative activating effects
derivatives due to ground-state stabilisation involving reso- of groups ortho and para to the reaction centre. Results re-
nance interaction as shown in 10 and 11. Such stabilisation ported[2] for methoxide addition to 1-cyano-3,5-dinitroben-
has previously been demonstrated in σ-adduct-forming re- zene and 3,5-dinitro-1-(trifluoromethyl)benzene show that
actions of ring-activated alkyl ethers.[2,26]
there is kinetic preference for reaction at the 4-position to
give adducts 12 and 13, respectively, while the isomeric ad-
ducts formed by addition at the 2-position have greater
thermodynamic stabilities. Hence, in the absence of steric
effects, reactions with nucleophiles of 1a may be expected
to be faster than those of 2a. The results in Table 1 show
that this is observed for reaction with n-butylamine.
A similar but less pronounced effect has been noted in
the corresponding phenyl ethers.[21,27] Because this interac-
tion involves p-π overlap between oxygen and the electron-
deficient ring, it will be greatly reduced, or eliminated, by
the presence of a bulky 6-substituent which increases the
steric congestion around the 1-position.
However, for secondary amines steric effects become im-
Effects of Ring-Activation: For compounds 1a–f the steric portant and this order is reversed. Nevertheless, for all three
situation at the reaction centre is similar, and the data in amines the reactivity of 1b is higher than that of 2b, i.e.
Table 2 show that for each amine values of k1 increase regu- reaction is faster at the position between the nitro groups
larly with increasing electron withdrawal of the 4-substitu- than at a position ortho to nitro and trifluoromethyl groups.
ent. Hammett plots, not shown, vs. σ– are reasonably linear This again bears witness to the large steric effect of the CF3
with slopes, ρ, of ca. 3.5 indicating that bond formation is group.
fairly well developed in the transition state for nucleophilic
Incidence of Base Catalysis: The observation of base ca-
attack. However, there is some evidence for downward cur- talysis in the 1-phenoxy compounds 3 and 4 depends on the
vature in the Hammett plots as electron-withdrawal by the value of the ratio kAm/k–1 and becomes more likely as the
4-substitutents increases, and this may be indicative (as re- value of the ratio decreases. There is good evidence[13,14,30]
ferred to earlier) to a shift towards an “earlier” transition that with strongly activated substrates such as 3 and 4 the
state. The same trends are followed by the 2,6-dinitro-1- zwitterionic intermediates 6 are more acidic than the corre-
phenoxy compounds 3a–d and 4f where again the steric sit- sponding ammonium ion, RRЈH2N+, so that the proton-
uation at the 1-position is constant.
transfer process, kAm, will be in the thermodynamically
For compounds 2a–f the bulk of the 6-substituent as well “downhill” direction. Hence values of kAm may approach
as its activating effect clearly has to be considered. The re- the diffusion limit, but are known to be strongly influenced
sults in Table 2 show that for each amine the reactivity or- by steric factors. Thus values are considerably decreased by
der is ring-N Ͼ CF3 Ͼ H. However, for n-butylamine the steric congestion around the 1-position[9,11,13,14,31] and have
6-NO2 group is more activating than the ring-nitrogen, been found to decrease in the order n-butylamine Ͼ pyrroli-
while this order is reversed for the secondary amines. This dine Ͼ piperidine.[11,13,30] It must be noted that steric fac-
may be attributed to the steric effect of the 6-NO2 group, tors on kAm differ from those involved in nucleophilic at-
which is more serious in reactions with the secondary tack at the 1-position, k1, and relate to hindrance of the
amines. The data for 2b indicate that the 6-CF3 group has approach of an amine molecule to the zwitterionic interme-
a serious steric effect for the secondary amines. Thus with diates 6 to allow proton transfer to occur.
pyrrolidine and piperidine 2b is only slightly more reactive
Interestingly, in the present work the only reaction in-
than 2a. The steric effects of the CF3 group in nucleophilic volving n-butylamine where base catalysis is observed is
substitutions have been noted previously,[20] and its size has with 4b. This is further evidence for the large “steric” effect
been estimated to be comparable to that of an isopropyl of the ortho-trifluoromethyl substituent, probably involving
group.[28] Recent calculations[29] have shown that these ef- electrostatic repulsion by the CF3 group of the amine base
fects may derive in part from electrostatic repulsions be- catalyst.
tween the local negative charge on the trifluoromethyl
The reactions of all the phenoxy compounds with pyr-
group and the incoming nucleophile. Although the X-ray rolidine are base-catalysed. The results in Table 5 show that
structure of 4b does not indicate any particularly large ef- for compounds 3a–d and 4f, where the steric situation
fects in the parent molecule the kinetic results suggest that around the 1-position is similar, values of kAm/k–1 tend to
sterically the effect of an ortho-CF3 group on nucleophilic increase with increasing electron-withdrawal by the 4-sub-
attack is greater than that of a nitro group.
stituent. Since values of kAm are likely to be unchanged,
1228
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 1222–1230

Kinetics of an Aromatic Nucleophilic Substitution Reaction
FULL PAPER
these increases may be attributed to decreases in values of (c) Ground-state stabilisation – as shown in 8 and 9 may
k–1 as the zwitterionic intermediates 6 become more ther- decrease reactivity.
modynamically stable. The low value of kAm/k–1 for 4b
The incidence of base catalysis depends on values of the
again reflects the large “steric” effects of the 6-CF3 substit- ratio kAm/k–1. Values of kAm decrease in the order n-bu-
uent. The relatively high values of kAm/k–1 for 4a and 4e, tylamine Ͼ pyrrolidine Ͼ piperidine, and with increasing
carrying hydrogen or ring-nitrogen at the 6-position, respec- steric congestion in the zwitterionic adducts, 6; e.g. ortho-
tively, can be attributed to a combination of increases in substitution decreases kAm. Steric congestion in the zwitter-
value of kAm and decreases in value of k–1 as the steric ions may also increase k–1 values.
strain decreases.
Although previous studies[1,2] have considered these and
Work with related systems[9,11,30,32] has shown that values other factors, we have succeeded in making quantitative
of kAm are likely to be lower by a factor of ca. ten for piperi- measurements using a range of substrates and amines.
dine than pyrrolidine. For compounds 3a–d and 4b these
decreases result in the condition k–1 ϾϾ kAm[Am] so that
Experimental Section
Equation (3) applies and proton transfer is fully rate-de-
termining. For the most activated compound 4f the ex-
pected reduction in value of k–1 allows a value for kAm/k–1
to be measured, and for 4a and 4e, where steric hindrance
is reduced, values are also measurable.
The 1-chloro compounds 1 and 2 were, apart from 1c, the purest
available commercial samples. Compound 1c was prepared by es-
terification of the corresponding benzoic acid with methanol; m.p.
102 °C, (ref.[33] 106°). 1c: calcd. C 36.9, H 1.93, N 10.7; found C
1
The only system where the uncatalysed decomposition 36.5, H 1.93, N 10.3. H NMR (200 MHz, CD3CN): δ = 8.63 (s, 2
H), 3.95 (s, 3 H) ppm. The 1-phenoxy compounds 3b–d and 4b
were prepared by reaction at 45 °C for two hours of the appropriate
1-chloro compound (1 equiv.) with potassium hydroxide (1 equiv.)
in an excess of phenol in aqueous ethanol. On completion water
was added and the solid formed was recrystallised from ethanol.
Data are given in Table 9. The following compounds were available
from previous work, 3a and 4a,[11] 4e,[21] 4f.[15] Amines and acetoni-
pathway makes a measurable contribution is for reaction
of 4a with piperidine. The k2 step is likely[1,2] to involve
intramolecular proton transfer within the zwitterionic ad-
duct 6 coupled with movement of electrons away from the
anionic ring. Because 4a is the substrate with lowest ring
activation such movement of electrons should be easier than
for other substrates. This coupled with the relatively low
value expected for kAm for reactions involving piperidine
allows competition of the uncatalysed with the catalysed
pathway.
An additional factor to be considered is intramolecular
hydrogen bonding in the zwitterionic intermediates, 6, be-
tween an N–H proton and an ortho-nitro group. The pres-
ence of such hydrogen bonding, affecting the proton to be
transferred, has been used[1,2] as an argument to explain
differences in reactivity between primary and secondary
amines. In the case of primary amines there will always be
one non-hydrogen bonded proton available for transfer thus
reducing the susceptibility to base catalysis. However, pre-
vious studies[15,30] have not found evidence for such hydro-
gen bonding. Our results for the systems studied in the pres-
ent work are adequately explained in terms of steric factors
which result in decreases in value of kAm in the series n-
butylamine, pyrrolidine, piperidine.
1
trile were the purest available commercial samples. H NMR spec-
tra were measured with a Varian Mercury 200-MHz instrument.
Kinetic measurements were made spectrophotometrically at the ab-
sorption maxima of the products using Perkin–Elmer Lambda 2 or
Shimadzu UV PC spectrometers or an Applied Photophysics SX-
17 MV stopped-flow spectrometer. Rate constants were measured
at 25 °C under first-order conditions with substrate concentrations
of 1·10–4 moldm–3 and were evaluated by standard methods. Values
are precise to 3%.
Table 9. Analytical data for 3b–d and 4b.
% C, H, N (found)
M.p. [°C]
% C, H, N (calcd.) 1H NMR shifts[a]
3b
3c
83
47.4, 2.12, 8.4
47.5, 2.15, 8.5
53.2, 3.16, 8.9
52.8, 3.17, 8.8
8.59 (s, 2 H), 7.37 (t, 2 H)
7.18 (t, 1 H), 6.98 (d, 2 H)
8.83 (s, 2 H), 7.35 (t, 2 H)
7.15 (t, 1 H), 6.99 (d, 2 H),
3.95 (s, 3 H)
105
3d
4b
160
68
54.7, 2.44, 14.7
54.7, 2.47, 14.7
47.1, 2.10, 8.6
47.5, 2.15, 8.5
9.08 (s, 2 H), 7.35 (t, 2 H)
7.15 (t, 1 H), 7.02 (d, 2 H)
8.99 (s, 1 H), 8.85 (s, 1 H)
7.38 (t, 2 H), 7.19 (t, 1 H),
6.95 (d, 2 H)
Conclusions
[a] Spin coupling, J = 7–8 Hz, was observed between ortho-hydro-
gen atoms. 3b and 4b were in CD3CN and 3c and 3d in [D6]DMSO.
Our study has identified the following as major factors
in determining values of k1 for nucleophilic attack for the
systems studied in acetonitrile:
(a) Ring activation – values increase with increasing electron
withdrawal by ring substituents. However substituents, no-
tably the CF3 group, at ortho positions can have deleterious
steric effects.
(b) Steric effects at the reaction centre – values decrease
with increasing congestion at the 1-position; effects increase
in the order Cl Ͻ OPh and n-butylamine Ͻ pyrrolidine Ϸ
piperidine.
X-ray Crystallography: Light yellow crystals of 3b and 4b were
grown from ethanol at room temperature. X-ray experiments were
carried out on Bruker SMART 3-circle diffractometers with CCD
area detectors 1 K (for 4b and 4f) or 6 K (for 3b), using graphite-
¯
monochromated Mo-Kα radiation (λ = 0.71073 Å) and Cryostream
(Oxford Cryosystems) open-flow N2 gas cryostats. The structures
were solved by direct methods and refined, by full-matrix least-
squares against F2 of all reflections, using SHELXL programs.[34]
Non-hydrogen atoms were refined with anisotropic displacement
parameters, except the minor positions of the disordered atoms
Eur. J. Org. Chem. 2006, 1222–1230
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1229

M. R. Crampton et al.
FULL PAPER
[5] F. Terrier, M. Mokhtari, R. Goumont, J.-C. Halle, E. Buncel,
Org. Biomol. Chem. 2003, 1, 1757–1763.
[6] M. R. Crampton, T. A. Emokpae, J. A. K. Howard, C. Isanbor,
R. Mondal, J. Phys. Org. Chem. 2004, 17, 65–70.
[7] J. A. Orvik, J. F. Bunnett, J. Am. Chem. Soc. 1970, 92, 2417–
2427.
[8] J. F. Bunnett, S. Sekiguchi, L. A. Smith, J. Am. Chem. Soc.
1981, 103, 4865–4871.
[9] M. R. Crampton, P. Routledge, J. Chem. Soc., Perkin Trans. 2
1984, 573–581.
[10] Y. Hasegawa, J. Chem. Soc., Perkin Trans. 2 1985, 87–92.
[11] R. A. Chamberlin, M. R. Crampton, J. Chem. Soc., Perkin
Trans. 2 1995, 1831–1838.
[12] R. A. Chamberlin, M. R. Crampton, J. Chem. Soc., Perkin
Trans. 2 1994, 425–432.
[13] M. R. Crampton, S. D. Lord, J. Chem. Soc., Perkin Trans. 2
1997, 369–376.
[14] C. F. Bernasconi, M. C. Muller, P. Schmid, J. Org. Chem. 1979,
44, 3189–3196.
which were refined in isotropic approximation. All H atoms were
“riding” in idealised positions. The absolute structures of 3b and
4f could not be determined in the absence of atoms with substantial
anomalous scattering, therefore all ∆fЈЈ were set to 0 and all Friedel
equivalents merged. The crystal data and experimental details are
listed in Table 10. CCDC-285790 (for 3b), -285791 (for 4b) and
-285694 (for 4f) contain the supplementary crystallographic data
for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
Table 10. Crystal data.
Compound
3b
4b
4f
Formula
M
Temp. [K]
Crystal system
Space group
a [Å]
C13H7F3N2O5 C13H7F3N2O5 C12H7N3O2
328.21
120
328.21
120
305.21
120
orthorhombic monoclinic
P212121 (#19) Pc (#7)
orthorhombic
P212121 (#19)
7.6597(15)
9.6982(19)
17.107(3)
90
[15] C. Isanbor, T. A. Emokpae, M. R. Crampton, J. Chem. Soc.,
Perkin Trans. 2 2002, 2019–2024.
[16] B. Capon, N. B. Chapman, J. Chem. Soc. 1957, 600–609.
[17] C. F. Bernasconi, R. H. de Rossi, J. Org. Chem. 1976, 41, 44–
49.
5.6276(6)
13.238(1)
35.711(2)
90
2660.4(4)
8
15.779(3)
5.5901(9)
15.736(3)
108.80(1)
1313.9(4)
4
b [Å]
c [Å]
β, [°]
[18] J. Kavalek, V. Sterba, Collect. Czech. Chem. Commun. 1973,
V [Å3]
1270.8(4)
4
38, 884–891.
Z
[19] T. A. Emokpae, P. U. Uwakwe, J. Hirst, J. Chem. Soc., Perkin
Trans. 2 1993, 125–132.
[20] R. E. Akpojivi, T. A. Emokpae, J. Hirst, J. Chem. Soc., Perkin
Trans. 2 1994, 443–449.
[21] M. R. Crampton, T. A. Emokpae, J. A. K. Howard, C. Isanbor,
R. Mondal, Org. Biomol. Chem. 2003, 1, 1004–1011.
[22] R. H. de Rossi, R. A. Rossi, F. N. R. Gimenez, J. Org. Chem.
1976, 41, 3163–3166.
[23] A. R. Choudhury, K. Islam, M. T. Kirchner, G. Metta, T. N.
Guru Row, J. Am. Chem. Soc. 2004, 126, 12274–12275.
[24] C. Glidewell, J. N. Low, J. M. S. Skakle, J. L. Wardell, Acta
Crystallogr. Sect. C: Cryst. Struct. Commun. 2005, C61, 185–
187.
Dcalcd. [g/cm3]
1.249
0.21
1.659
0.16
1.595
0.14
µ [mm–1
]
Refls. measured
Refls. unique[a]
Refls. with I Ն 2σ(I)[a]
Rint
30794
3531
3338
0.031
0.031
0.077
14213
3044
2875
0.043
0.043
14483
1690
1530
0.046
0.034
0.074
R[I Ն 2σ(I)]
wR(F2), all data
0.103
[a] Friedel equivalents merged.
Supporting Information (see footnote on the first page of this arti-
cle): The Supporting Information in Tables S1 to S9 contains kin-
etic data as detailed in the main text.
[25] J. F. Coetzee, Prog. Phys. Org. Chem. 1967, 4, 45–92.
[26] C. F. Bernasconi, J. Am. Chem. Soc. 1970, 92, 4682–4688.
[27] M. R. Crampton, L. M. Pearce, L. C. Rabbitt, J. Chem. Soc.,
Perkin Trans. 2 2002, 257–261.
[28] T. Nagai, G. Nishioka, M. Koyama, A. Ando, T. Miki, I. Kum-
adaki, J. Fluorine Chem. 1992, 57, 229–238.
[29] T. Katagiri, S. Yamaji, M. Handa, M. Irie, K. Uneyama, Chem.
Commun. 2001, 2054–2055.
[30] M. R. Crampton, B. Gibson, J. Chem. Soc., Perkin Trans. 2
1981, 533–539.
Acknowledgments
We thank the Royal Society, London, for financial assistance to
allow C. I. to visit Durham for study leave.
[31] M. R. Crampton, P. J. Routledge, P. Golding, J. Chem. Soc.,
Perkin Trans. 2 1984, 329–336.
[32] M. R. Crampton, C. Greenhalgh, J. Chem. Soc., Perkin Trans.
2 1983, 1175–1178.
[33] J. T. Manka, F. Guo, J. Huang, H. Yin, J. M. Farrar, M. Sien-
kowska, V. Benin, P. Kaszynski, J. Org. Chem. 2003, 68, 9574–
9588.
[1] C. F. Bernasconi, MTP Int. Rev. Sci. Org. Chem. Ser. 1 1973,
3, 33–63.
[2] F. Terrier, Nucleophilic Aromatic Displacement, VCH, New
York, 1991.
[3] a) G. Consiglio, V. Frenna, S. Guernelli, G. Macaluso, D. Spi-
nelli, J. Chem. Soc., Perkin Trans. 2 2002, 965–970; b) G. Consi-
glio, V. Frenna, S. Guernelli, G. Macaluso, D. Spinelli, J. Chem.
Soc., Perkin Trans. 2 2002, 971–995.
[4] A. El-Bardan, G. M. El-Subruiti, F. El-Zahraa, M. El-Hegazy,
E. A. Hamed, Int. J. Chem. Kinet. 2002, 34, 645–650.
[34] SHELXTL v. 6.12, Bruker AXS, Madison, Wisconsin, USA,
2001.
Received: October 6, 2005
Published Online: December 22, 2005
1230
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 1222–1230