Role of lactam vs. lactim tautomers in 2(1H)-pyridone catalysis of aromatic nucleophilic substitution

Article abstract of DOI:10.1039/a706533g

3-Ethylaminocarbonyl-2(1H)-pyridone 1 and 3-ethoxycarbonyl-2(1H)-pyridone 2 have been synthesised and tested as catalysts for the aromatic nucleophilic substitution of fluoride by piperidine in 2-fluoro-5-nitrobenzonitrile 3. A kinetic model which takes into account the dimerisation of the catalysts has been developed, which allows a quantitative analysis of measured data. 3-Ethylaminocarbonyl-2(1H)-pyridone 1 exists exclusively as a lactam tautomer, either monomeric or dimeric but 3-ethoxycarbonyl-2(1H)-pyridone 2 exists as a lactim monomer, whereas its dimer exists in the lactam form. Despite such differences, these two compounds exhibit similar catalytic efficiencies for the reaction studied, suggesting that lactim and lactam tautomers have comparable efficiencies in tautomeric catalysis.

Full text of DOI:10.1039/a706533g

View Online / Journal Homepage / Table of Contents for this issue
Role of lactam vs. lactim tautomers in 2(1H)-pyridone catalysis of
aromatic nucleophilic substitution
Anne Loppinet-Serani, Florence Charbonnier,* Christian Rolando† and
Ivan Huc‡
Ecole Normale Supérieure, Département de Chimie, URA 1679 du CNRS,
Processus d’Activation Moléculaire, 24 rue Lhomond, 75231 Paris Cedex 05, France
3-Ethylaminocarbonyl-2(1H)-pyridone 1 and 3-ethoxycarbonyl-2(1H)-pyridone 2 have been synthesised
and tested as catalysts for the aromatic nucleophilic substitution of fluoride by piperidine in 2-fluoro-5-
nitrobenzonitrile 3. A kinetic model which takes into account the dimerisation of the catalysts has been
developed, which allows a quantitative analysis of measured data. 3-Ethylaminocarbonyl-2(1H)-pyridone
1 exists exclusively as a lactam tautomer, either monomeric or dimeric but 3-ethoxycarbonyl-2(1H)-
pyridone 2 exists as a lactim monomer, whereas its dimer exists in the lactam form. Despite such
differences, these two compounds exhibit similar catalytic efficiencies for the reaction studied,
suggesting that lactim and lactam tautomers have comparable efficiencies in tautomeric catalysis.
Introduction
O
N
O
O
N
Since the discovery in 1952 of the unusual catalytic efficiency of
2(1H)-pyridone for 2,3,4,6-tetra-O-methyl-α--glucopyranose
anomeric epimerisation to the β-isomer,1 examples of bifunc-
tional catalytic effects have been reported for a variety of
chemical reactions.2 The attention devoted to these studies
stems from the analogies between bifunctional catalyses and
enzyme mediated processes.3 All these bifunctional catalysts are
tautomeric molecules or ions, usually both weak acids and
weak bases. They share a common ability to bind to a reaction
intermediate or transition state and to exchange simultaneously
two protons while interconverting between tautomers during
the catalytic process [reaction (1)]. However, whether these
H
O
H
O
N
H
1
2
catalytic efficiency of 1 and 2 in the substitution of 2-fluoro-
5-nitrobenzonitrile by a secondary amine was determined and
correlated to their tautomeric properties.
The name ‘2-hydroxypyridine’ will be used to designate this
compound or any derivative whatever tautomeric forms, either
monomeric or dimeric, may exist in solution.
O
O
H
H
O
O
Results and discussion
N
N
H
OMe
OMe
OMe
OMe
(1)
H
O
O
Synthesis
MeO
MeO
Amide 1 and ester 2 were prepared from 2-hydroxynicotinic
acid via the corresponding acid chloride. Because they do not
dimerise or tautomerise, the N-methyl 1a and 2a, and O-methyl
derivatives 1b and 2b were synthesised as reference compounds
for the spectroscopic studies.
Alkylation of 1 and 2 with methyl iodide in basic media
yielded mixtures of N-methyl and O-methyl derivatives that
could be separated by column chromatography. The regio-
selectivity could be directed towards either N-alkylation or
O-alkylation, respectively, by the use of K2CO3 or Ag2CO3 as
the basic reagent.5 The ratios of N-alkylated and O-alkylated
products as determined by 1H NMR analysis on the crude
product were as shown below.
OMe
OMe
features are responsible for the catalytic efficiency is still an
assumption and mechanistic studies have not yet been fully
conclusive.4
To elucidate the role of tautomerism in these reactions, we
assessed the catalytic power of lactam vs. lactim tautomers
of 2(1H)-pyridone in nucleophilic aromatic substitution. To
this end, substituted derivatives of 2(1H)-pyridone, 3-ethyl-
aminocarbonyl-2(1H)-pyridone 1 and 3-ethoxycarbonyl-2(1H)-
pyridone 2, were prepared. As shown by NMR and UV
spectroscopy, the tautomeric equilibria of 1 and 2 in CHCl3 are
considerably different although their structures are similar. In 1,
an intramolecular hydrogen bond involving the amide N᎐H
stabilises the lactam form of the heterocycle, thus allowing an
equilibrium between a lactam dimer and a lactam monomer.
Intramolecular hydrogen bonding in 2 involves the ester
carbonyl group and stabilises a lactim monomer. These two
compounds could then be considered as models of lactam and
lactim monomeric forms, respectively, of 2(1H)-pyridone. The
O
O
1a : 1b
K2CO3 89 : 11
N
H
N
H
K2CO3 22 : 78
N
O
N
O
Me
Me
1b
1a
O
O
2a : 2b
K2CO3 99 : 1
† Present address: Université des Sciences et Techniques de Lille (Lille
1), URA 351 du CNRS Chimie Organique et Macromoléculaire, UFR
de Chimie Bâtiment C3, 59655 Villeneuve d’Ascq Cedex, France.
‡ Present address: Université Louis Pasteur (Strasbourg 1), Institut Le
Bel, URA 422 du CNRS Laboratoire de Chimie Macromoléculaire, 4
rue Blaise Pascal, F-67000 Strasbourg, France.
O
H
O
H
Ag2CO3 7 : 93
N
O
N
O
Me
Me
2b
2a
J. Chem. Soc., Perkin Trans. 2, 1998
937

View Online
Fig. 1 UV spectra of amide 1 and ester 2 at 2 m in CDCl3. Top: spectral changes of 2 upon addition of MeOH from 0 to 1.6  (A), and upon
cooling from 57 to 4 ЊC (B). Bottom: spectral changes of 1 upon addition of MeOH from 0 to 1.6  (C) and upon cooling from 53 to 12 ЊC (D).
Tautomerism and dimerisation
ation. Due to its polar nature, the lactam tautomer exists pre-
dominantly as a hydrogen-bonded dimer in apolar media. On
the other hand, the less polar lactim tautomer is also found as a
monomer in apolar solvent.9 The proton exchange involved in
the tautomerism occurs within a dimer in apolar media, or
between a monomer and the solvent in protic media.10
Interconversion of tautomers, solvation, and dimerisation of
2(1H)-pyridone result in a number of equilibria (Scheme 1).6
O
H
O
H
• nS
+ nS
The co-existence of both tautomers for ester 2 in CHCl3 was
demonstrated by the presence of two absorbance maxima in
its UV spectrum. Each maximum is the sum of contributions
from absorbances of the monomers and/or dimers of each
tautomeric species.9 Comparison with UV spectra of the
N-methyl and O-methyl derivatives allowed the assignment of
lactim (300 nm) and lactam (330 nm) chromophores. Upon
addition of MeOH (Fig. 1A), the lactim absorbance decreases
and the lactam absorbance increases through the formation
of (2lactamؒMeOHn) complexes.11 The tautomerism is also
temperature-dependent (Fig. 1B). The lactam absorbance
increases upon cooling while the lactim absorbance decreases.
This is consistent with a differential stabilisation of the
dimeric lactam and monomeric lactim. The monomer–dimer
equilibrium shifts with temperature or concentration, which
results in a shift of the tautomeric equilibrium. As previously
suggested for other 3-carboxy substituted 2(1H)-pyridones,9 the
formation of an intramolecular hydrogen bond involving the
ester carbonyl group (Fig. 1) stabilizes the lactim monomer.
In contrast, the UV spectrum of amide 1 under the same
conditions consists of essentially one maximum at 320 nm.
Cooling, heating, or addition of MeOH induced minimal
changes (Figs. 1C and 1D). These results suggest that only the
lactam is present and that its UV spectrum does not change
significantly upon dimerisation or complexation with MeOH.
We propose that the relative stabilisation of the lactam is due to
intramolecular-hydrogen bonding between the lactam carbonyl
group and the amide N᎐H bond. This arrangement seemingly
N
N
Me
O
H
H
O
H
S = MeOH :
N
O
Me
O
H
O
H
2
2
N
N
H
O
O
H
O
H
H
N
N
N
O
Scheme 1 Tautomerism and dimerisation of 2-pyridones. Hetero
dimers between lactim and lactam tautomers are not to be excluded.
Both calculation and experiment have shown the standard
enthalpies of formation of 2(1H)-pyridone and 2-hydroxy-
pyridine to be similar.7 In solution, the lactam–lactim ratio
depends on the polarity of the solvent. The lactim tautomer is
a better hydrogen bond donor and acceptor and predominates
as a solvated monomer in polar or protic media.8 In apolar
solvents, the tautomeric equilibrium is influenced by dimeris-
938
J. Chem. Soc., Perkin Trans. 2, 1998

View Online
Fig. 2 1H NMR chemical shifts (CDCl3, 23 ЊC) of H4 (᭡), H5 (᭿) and
H6 (᭹) signals for amide 1 and H4 (᭝), H5 (ᮀ) and H6 (᭺) for ester 2 at
different concentrations. For comparison, the concentration independ-
ent shifts of the same signals for the N-methyl and O-methyl derivatives
are represented above and below respectively.
Fig. 3 Observed 1H NMR (CDCl3, 23 ЊC) chemical shift of the signals
of the protomeric protons for amide 1 (ᮀ) and ester 2 (᭺). Data cal-
culated for a simple dimerisation isotherm with a dimerisation constant
of 280 Ϫ1 for 1, and of 1335 Ϫ1 for 2 are shown as solid lines.
an intramolecular hydrogen bond between the lactam carbonyl
group and the amide N᎐H bond of amide 1: (i) the chemical
shift of the amide N᎐H in CDCl3 is relatively far downfield
(9.55 ppm) and concentration-independent; (ii) the deshielding
effect of the carbonyl group in position 3 on the H4 signal
is stronger for amide 1 (δH4 8.6) than for ester 2 (δH4 8.2)
(Scheme 2).
Chemical shifts of protomeric protons of amide 1 and ester 2
are also concentration-dependent. The observed data fit a bind-
ing isotherm calculated for a simple dimerisation equilibrium,12
which results in a sigmoidal curve when the concentration is
represented on a logarithmic scale (Fig. 3).
The best fit was obtained for a dimerisation constant Kdim
of 280 Ϫ1 (estimated error margin 10%) for amide 1, and
of 1335 Ϫ1 ( 15%) for ester 2, defined as follows: 2 [mono-
mer]Total = [dimer]Total, Kdim = [dimer]Total/[monomer]Total2. The
discrepancy between these values may come from repulsive
secondary interactions13 between the lactam N᎐H bond and the
amide N᎐H bond in dimers of 1. Also, the fact that the lactam
oxygen of 1 partially satisfies its electrostatic demand in an
intramolecular hydrogen bond might decrease its ability to
form a second hydrogen bond with its remaining lone pair.
In conclusion, equilibria represented in Scheme 2 are consist-
ent with the observed tautomeric and dimerisation properties
of amide 1 and ester 2. The next step is to evaluate how these
differences affect the catalytic power of these compounds in
aromatic nucleophilic substitutions.
forms despite the possibility of a hydrogen bond between
the lactim hydroxy group and the amide carbonyl.
Significant changes in the H NMR spectra of 1 and 2 were
1
observed upon varying concentration or temperature (shifts of
up to 0.6 ppm for aromatic protons and up to 4 ppm for proto-
meric protons). These changes were not observed for methyl-
ated compounds 1a, 1b, 2a and 2b. As shown in Fig. 2, at a
concentration of 1 mol lϪ1, the aromatic signals of ester 2
almost match those of its N-methyl analogue 2a. These signals
shift toward those of the O-methyl derivative 2b upon dilution.
Again, these results support the hypothesis of a monomer–
dimer equilibrium in which the lactam tautomer resides pre-
dominantly in dimers and the lactim tautomer in monomers
[equilibria (2) and (3)]. Dilution causes dissociation of dimers
and therefore a shift in the protomeric equilibrium toward the
lactim tautomer.
On the other hand, at no concentration does the NMR spec-
trum of 1 match that of its O-methyl derivative. Upon dilution,
H5 and H6 signals shift upfield for amide 1 whilst they shift
downfield for ester 2 (Fig. 2).
These results are consistent with UV data and support the
absence of the lactim tautomer for 3-ethylaminocarbonyl-
2(1H)-pyridone 1 and a simple equilibrium between lactam
monomers and lactam dimers. While the large shifts of H5 and
H6 signals for the 3-ethoxycarbonyl-2(1H)-pyridone 2 are best
explained by a lactim–lactam interconversion, the smaller shifts
observed for amide 1 simply reflect the deshielding effect of the
proximal carbonyl group in lactam dimers (Scheme 2). The 1H
NMR analysis also provides data supporting the formation of
Catalysis
The efficiency of pyridones for catalysing aromatic nucleophilic
substitutions was first demonstrated by Pietra and Vitali using
2,4-dintrofluorobenzene and piperidine as reagents.2d As shown
in Scheme 3, the proposed mechanism involves complexation of
the zwitterionic intermediate by pyridone. This complex, where
the oxo group and the tautomeric hydrogen of pyridone are
hydrogen bonded respectively to the ammonium proton and to
the electronegative fluorine atom, gets transformed into prod-
ucts through a cyclic transition state, by concerted departure of
ammonium proton and fluoride anion from the intermediate.
This process leaves the catalyst as a tautomeric hydroxy form.
We studied the catalytic efficiency of amide 1, ester 2 and
2(1H)-pyridone itself in the substitution of fluoride by piper-
idine in 2-fluoro-5-nitrobenzonitrile 3. The cyano compound
was preferred over the trifluoromethyl analogue whose reaction
with piperidine is very slow.14 Also the nitro derivative was
avoided to prevent complications due to intramolecular hydro-
gen bonding between the nitro group and the ammonium
proton in the intermediate.15
O
N
N
O
N
N
H
H
O
H
O
H
H
N
N
(2)
2
H
H
O
H
1
O
O
O
N
O
O
N
H
H
O
O
H
H
O
N
O
(3)
2
2
O
Since the tetrahedral intermediate may collapse to form the
products or the starting reagents (Scheme 3),16 the apparent rate
constant can be expressed by eqn. (4), where k1 and kϪ1 are the
Scheme 2 Dimerisation and tautomerism of amide 1 and ester 2.
Arrows represent some of the deshielding effects observed in the 1H
NMR spectra.
J. Chem. Soc., Perkin Trans. 2, 1998
939

View Online
Table 1 Apparent second order rate constants kapp for the piperidino-substitution of 2-fluoro-5-nitrobenzonitrile 3 in CHCl3 at 25 ЊC in presence of
2(1H)-pyridone, 3-ethylaminocarbonyl-2(1H)-pyridone 1 or ethoxycarbonyl-2(1H)-pyridone 2
[Pyridone]a/mol lϪ1
kapp/l molϪ1 sϪ1
[1]b/mol lϪ1
kapp/l molϪ1 sϪ1
[2]b/mol lϪ1
kapp/l molϪ1 sϪ1
5.70 × 10Ϫ5
1.14 × 10Ϫ4
5.70 × 10Ϫ4
1.14 × 10Ϫ3
5.70 × 10Ϫ3
1.14 × 10Ϫ2
4.52 × 10Ϫ2
8.33 × 10Ϫ2
1.13 × 10Ϫ1
4.03 × 10Ϫ3
3.74 × 10Ϫ3
4.99 × 10Ϫ3
5.43 × 10Ϫ3
7.70 × 10Ϫ3
9.05 × 10Ϫ3
1.69 × 10Ϫ2
2.35 × 10Ϫ2
2.64 × 10Ϫ2
1.01 × 10Ϫ5
1.01 × 10Ϫ4
5.05 × 10Ϫ4
1.01 × 10Ϫ3
5.05 × 10Ϫ3
1.01 × 10Ϫ2
5.05 × 10Ϫ2
1.01 × 10Ϫ1
4.22 × 10Ϫ3
4.16 × 10Ϫ3
4.48 × 10Ϫ3
4.68 × 10Ϫ3
5.86 × 10Ϫ3
6.72 × 10Ϫ3
1.09 × 10Ϫ2
1.40 × 10Ϫ2
1.28 × 10Ϫ5
1.28 × 10Ϫ4
5.13 × 10Ϫ4
1.03 × 10Ϫ3
2.26 × 10Ϫ3
5.13 × 10Ϫ3
1.03 × 10Ϫ2
4.25 × 10Ϫ3
4.29 × 10Ϫ3
4.56 × 10Ϫ3
4.72 × 10Ϫ3
5.04 × 10Ϫ3
5.59 × 10Ϫ3
6.08 × 10Ϫ3
a [Piperidine] = 1.62 × 10Ϫ3 mol lϪ1; [2-fluoro-5-nitrobenzonitrile] = 8.93 × 10Ϫ5 mol lϪ1
benzonitrile] = 9.22 × 10Ϫ5 mol lϪ1
.
b [Piperidine] = 1.64 × 10Ϫ3 mol lϪ1; [2-fluoro-5-nitro-
.
Table 2 Non-linear least-square regression analysis of measured apparent second order rate constants (kapp) using the eqn. (6)
(k1k2/kϪ1 s2)/
(k1k3/kϪ1 s3)/
a
Catalyst
Kdim s1
10Ϫ3 l molϪ1 sϪ1 b
l2 molϪ2 sϪ1 b
(k3/k2)/l molϪ1
2(1H)-Pyridone
Amide 1
Ester 2
2880 115
280 12
1335 217
3.23 0.37
4.12 0.03
4.23 0.03
5.27 0.37
0.781 0.04
1.06 0.16
1631 270
190 10
250 30
a Kdim values were measured by NMR dilution experiments; s1: standard deviation of Kdim as determined by the curve fitting program. b s2: standard
deviation of k1k2/kϪ1; s3: standard deviation of k1k3/kϪ1; s2 and s3 are the sum of the standard deviation determined by the curve fitting program, and
of the deviation due to the error on Kdim
.
F
H
N
NC
+
NO2
k–1
k1
3
+
NH
H
F
N
k2
+ HF
NC
–
NC
k3[Cat]
NO2
NO2
Fig. 4 Apparent second order rate constants as a function of the total
concentration of 3-ethylaminocarbonyl-2(1H)-pyridone 1. Data fitted
according to the proposed model are displayed as a solid line.
+
NH
O
H
N
F
H
–
NC
The variations of kapp as a function of the concentration in
monomeric catalyst can also be estimated, the concentration in
monomers being expressed as a function of the total concen-
tration in catalyst [Cat] and the dimerisation equilibrium
constant Kdim, [Cat] = [monomeric Cat] ϩ 2 [dimeric Cat] and
Kdim = [dimeric Cat]/[monomeric Cat]2 gives eqn. (6).
NO2
Scheme 3 Mechanism of a nucleophilic aromatic substitution. The
breakdown of the intermediate into products can either take place
directly (k2) or be catalysed by the pyridone (k3[Cat]).
¹
²
[monomeric Cat] = [(1 ϩ 8Kdim[Cat]) Ϫ 1]/4Kdim (6)
kapp = k1(k2 ϩ k3[Cat])/(kϪ1 ϩ k2 ϩ k3[Cat])
(4)
The plot thus obtained is a straight line. Thus, the measured
kapp values were fitted to eqn. (7) by non-linear least square
rate constants relative to the formation of the zwitterionic
intermediate, respectively for forward and backward reactions,
and k2 and k3 are the rate constants relative to the formation of
products, respectively, via the non-catalysed and the catalysed
routes.
kapp = k1k2/kϪ1 ϩ [monomeric Cat] k1k3/kϪ1
(7)
regression analysis, using the Kdim values obtained from the
analysis of the data measured by NMR spectroscopy in dilu-
tion experiments. The calculated values thus obtained for k1k2/
In chloroform kϪ1 ӷ k2 ϩ k3[Cat], the second step is then
rate-limiting, and the apparent rate constant is a linear function
of the concentration of catalysts17 [eqn. (5)].
k
Ϫ1, k1k3/kϪ1 and Kdim are reported in Table 2. The reasonable
homogeneity of the values calculated for k1k2/kϪ1 (uncatalysed
pathway) indicate that although simple, the model gives a reli-
able description of the reaction. The deviation observed in the
variations of kapp with the total concentration of catalyst can
thus be explained by the dimerisation of the catalyst that takes
place at higher concentrations, leaving pyridones as associated
kapp = k1k2/kϪ1 ϩ [Cat]k1k3/kϪ1
(5)
In Table 1 are reported the measured kapp values at different
concentrations of 1, 2 and 2(1H)-pyridone. These values do
increase with the total concentration of catalyst, but less than
predicted by the linear relationship mentioned above (Fig. 4).
940
J. Chem. Soc., Perkin Trans. 2, 1998

View Online
species less available for catalysis. Pyridone dimers do not par-
ticipate in catalysis; only monomers are effective.
ments without further purification. 2(1H)-Pyridone (Aldrich
Co.) was sublimed before kinetic measurements. 2-Fluoro-5-
nitrobenzonitrile,19 5-nitro-2-(1-piperidino)benzonitrile20 and
4-nitro-1-(1-piperidino)-2-trifluoromethylbenzene20 were pre-
pared and purified according to reported methods.
The efficiency of the catalysts tested is best expressed by k3/
k2. It varies from 200–300-fold accelerations for 1 and 2 and to
1600-fold for 2(1H)-pyridone. The small difference between the
catalytic efficiencies of 1 and 2 sharply contrasts with their
largely different behaviours toward dimerisation and tauto-
merism noted from the NMR and UV studies. Ester 2 has
a better hydrogen bonding ability (a larger Kdim value) than 1,
but this does not significantly enhance its catalytic power.
Furthermore, the fact that 1 is exclusively lactam, whilst 2 is
predominantly lactim, in the monomeric form does not affect
the catalysis either. Each catalytic step induces interconversion
of tautomers, but all the steps involved in the interconversion
between tautomeric forms are faster than the aromatic nucleo-
philic substitution and the more stable tautomer is quickly
regenerated.10 Unless hydrogen bonding ability and tauto-
merism compensate for each other, it appears that these factors
are not essential for catalytic efficiency.
Both 3-ethylaminocarbonyl-2(1H)-pyridone 1 and 3-ethoxy-
carbonyl-2(1H)-pyridone 2 are both less effective than 2(1H)-
pyridone. A possible explanation may be that intramolecular
hydrogen bonding in monomers for both amide 1 and ester 2
involves the lactam (or lactim) function. If the additional stabil-
isation of one tautomer by hydrogen bonding results in slower
tautomer interconversion, it may also decrease the rate of pro-
ton exchange with the ammonium intermediate in the catalytic
process. The proton exchange could also be dependent upon the
acid–base properties of 2(1H)-pyridones. Regardless of which
tautomer is stabilised, the electron-withdrawing substituent in
position 3 for amide 1 and ester 2 tends to increase their basicity
and decrease their acidity. Indeed, the measured pKa values
(10.5 and 10.7, respectively) are about one unit lower than that
of 2(1H)-pyridone (11.62).18
NMR Spectroscopy
Assignment of resonances in the 1H NMR spectra of com-
pounds 1, 1a, 1b and 2, 2a, 2b in CDCl3 were based on the
following: (i) H5 signal is always at a higher field than H4 and
H6 resonances and exhibits a characteristic coupling pattern;
(ii) H4 has a higher coupling constant with H5 than with H6
(7.5 vs. 4.9 Hz for O-methyl derivatives 1b and 2b),21 and its
signal is most often at a lower field because its deshielding effect
of the carbonyl in position 3 (2b is the only exception); (iii)
N-methyl and O-methyl derivatives can be distinguished from
the shift of the methyl group signal (4.1 ppm for O᎐Me and 3.6
for N᎐Me).
For dilution experiments, concentrated solutions of 1, 2 or
2(1H)-pyridone (1 ml, in the molar range) were prepared, a 500
µl aliquot was transferred to an NMR tube and replaced by the
same volume of pure solvent. The procedure was repeated
down to submillimolar concentrations. Spectra were recorded
with increasing acquisition times for diluted samples. Non-
linear least square regression curve fitting was performed using
the Axum software from MathSoft (Bagshot, UK) for deter-
mining association constants.
Kinetic measurements
The kinetics were followed spectrophotometrically at λanal = 370
nm. At this wavelength, the molar extinction coefficient of 2-
fluoro-5-nitrobenzonitrile is εanal = 15 700 l molϪ1 cmϪ1. Con-
centrations of substrate, amine and catalyst are reported in the
tables. The kinetic runs were recorded during one hour in a
thermostatted cell at 25 ЊC. The kapp values were calculated for
each concentration. Least-square analyses performed with the
Axum software from MathSoft company (Bagshot, UK)
allowed catalytic efficiencies to be calculated.
Conclusions
In conclusion, we have developed a simple kinetic model for a
reliable evaluation of the catalytic efficiency 2(1H)-pyridones in
aromatic nucleophilic substitutions. Correlation of the catalytic
potential of amide 1 and ester 2 with their physical properties
enabled us to assess the role of different factors in tautomeric
catalysis: dominant tautomer, hydrogen bonding ability, tauto-
mer interconversion rate, and acid–base properties. Our results
suggest that the first two properties do not account for the
catalytic power of the 2(1H)-pyridones studied. Further work
using 2(1H)-pyridones with various substituents is now in pro-
gress. It should provide more information and help unravel the
critical criteria for large rate accelerations.
Syntheses
3-Ethylaminocarbonyl-2(1H)-pyridone 1. A mixture of 2-
hydroxynicotinic acid (1 g, 7.2 mmol) and SOCl2 (6 ml) was
heated to reflux during 15 min. Excess thionyl chloride was then
removed and the residual solid was dried under vacuum. The
acid chloride was dissolved in anhydrous THF (10 ml), and
added dropwise to a solution of EtNH2 (4 ml) in anhydrous
THF (10 ml) at Ϫ78 ЊC. The mixture was then stirred at room
temperature during 12 h. Volatiles were carefully evaporated
and the residue was suspended in aqueous saturated NaHCO3.
The product was extracted with CH2Cl2 (3 × 10 ml). The
organic layers were combined, washed with water, dried over
MgSO4 and evaporated to dryness. The product was recrystal-
lised from acetone–pentane (970 mg, 81%), mp 158 ЊC; δH(250
MHz, CDCl3) 9.56 (1H, br, s), 8.67 (1H, dd, Ja 7.2, Jb 2.2), 7.59
(1H, dd, Ja 6.6, Jb 2.2), 6.58 (1H, dd, Ja = Jb 6.9), 3.51 (2H, dq,
Ja = Jb 7.3), 1.27 (3H, t, J 7.3); δC(250 MHz, [2H6]DMSO)
163.39, 162.65, 144.13, 139.54, 120.88, 106.56, 33.74, 15.26
(C8H10N2O2 requires: C, 57.48; H, 5.43; N, 8.38. Found: C,
57.21; H, 5.65; N, 8.41%); νmax(KBr)/cmϪ1 3165, 3077, 3017,
2970, 2936, 2881, 1667, 1548, 1470.
3-Ethoxycarbonyl-2(1H)-pyridone 2. Ethanol was carefully
added to the acid chloride prepared as above and the mixture
was stirred at room temperature during 12 h. Solvent was evap-
orated to dryness. The off-white crystalline residue was sub-
limed, and recrystallised from acetone–hexane to yield the ester
as white needles (840 mg, 70%), mp 139 ЊC; δH(250 MHz,
[2H6]DMSO) 8.02 (1H, dd, Ja 7.1, Jb 2.3), 7.64 (1H, dd, Ja 6.7,
Jb 2.3), 6.25 (1H, dd, Ja = Jb 6.8), 4.18 (2H, q, J 7.0), 1.24 (3H,
t, J 7.0); δC(250 MHz, [2H6]DMSO) 165.02, 159.55, 145.42,
141.23, 120.66, 104.65, 60.53, 14.52 (C8H9NO3 requires: C,
Experimental
General
1H NMR measurements were performed on a Bruker AC-250
spectrometer, and a Varian VXR-500 spectrometer. NMR
characterisation of 1 and 2 was performed in [2H6]DMSO as
the signals are not concentration dependent in this solvent.
Chemical shifts are reported in parts per million (δ) relative to
the residual solvent peak. Coupling constants are given in
Hertz. IR measurements were performed using a Nicolet
Impact 400D instrument. Melting points were obtained on a
Büchi 510 apparatus and are uncorrected. Absorbance meas-
urements were performed on a lambda 2 Perkin-Elmer UV–VIS
spectrometer controlled by the PECSS software package. Kin-
etic measurements were performed on a DU 7400 Beckmann
UV–VIS spectrometer.
Tetrahydrofuran (THF) was distilled from sodium benzo-
phenone–ketyl anion radical, CH2Cl2 was distilled from CaH2
and hexane from P2O5. CDCl3 was stored over molecular sieves.
HPLC grade CHCl3 (Aldrich Co.) was used for kinetic experi-
J. Chem. Soc., Perkin Trans. 2, 1998
941

View Online
3 W. P. Jencks, Catalysis in Chemistry and Enzymology, Dover
Publications, Inc., New York, 1987, 200.
4 (a) K.-Å. Engdahl, H. Bivehed, P. Ahlberg and W. H. Saunders Jr.,
J. Am. Chem. Soc., 1983, 105, 4767; (b) M. Kusuya, A. Noguchi and
T. Okuda, Bull. Chem. Soc. Jpn., 1984, 57, 3461.
5 For a survey of alkylation methods, see: D. L. Comins and
G. Jianhua, Tetrahedron, 1994, 35, 2819 and references cited therein.
6 M. Gallant, M. T. P. Viet and J. D. Wuest, J. Am. Chem. Soc., 1991,
113, 721.
7 (a) R. S. Brown, A. Tse and J. C. Veredas, J. Am. Chem. Soc., 1980,
102, 1174; (b) A. Sygula, J. Chem. Res. (S), 1989, 56.
8 P. Beak, J. B. Covington and J. M. White, J. Org. Chem., 1980, 45,
1347.
9 P. Beak, J. B. Covington, S. G. Smith, J. M. White and J. M. Zeigler,
J. Org. Chem., 1980, 45, 1354.
10 (a) M. Chevrier, J. Guillerez and J.-E. Dubois, J. Chem. Soc., Perkin
Trans. 2, 1983, 979; (b) O. Bensaude, M. Chevrier and J.-E. Dubois,
J. Am. Chem. Soc., 1978, 100, 7055.
11 T. Kitagawa, K. Matsumoto and E. Hirai, Chem. Pharm. Bull.,
1978, 26, 1415.
12 C. S. Wilcox, Frontiers in Supramolecular Organic Chemistry and
Photochemistry, ed. H.-J. Schneider and H. Dürr, VCH Weinheim,
New York, 1991.
13 W. L. Jorgensen and J. Pranata, J. Am. Chem. Soc., 1990, 112, 2008.
14 R. E. Akpojivi, T. A. Emokpae and J. Hirst, J. Chem. Soc., Perkin
Trans. 2, 1994, 443.
15 C. F. Bernasconi and R. H. de Rossi, J. Org. Chem., 1976, 41, 44.
16 N. S. Nudelmann, P. M. E. Mancini, R. D. Martinez and L. R.
Vottero, J. Chem. Soc., Perkin Trans. 2, 1987, 951.
17 J. F. Bunnett and R. H. Garst, J. Am. Chem. Soc., 1965, 87, 3875.
18 A. Albert and J. N. Phillips, J. Chem. Soc., 1956, 1294.
19 J. F. K. Wilshire, Aust. J. Chem., 1967, 20, 1663.
20 J. D. Hepworth and P. Jones, Synthesis, 1974, 874.
21 H. Günter, NMR Spectroscopy, Wiley, New York, 1987, 113.
57.48; H, 5.43; N, 8.38. Found: C, 57.53; H, 5.40; N, 8.34%);
νmax(KBr)/cmϪ1 3197, 3087, 2971, 1721, 1665, 1596, 1556, 1480,
1430.
General methylation procedure. A mixture of the starting 2-
hydroxypyridine (1 mmol) and K2CO3 (or Ag2CO3) (0.5 g) was
suspended in hexane (20 ml). Methyl iodide (2 ml) was added
and the mixture was stirred at room temperature during 12 h,
then heated to reflux during 1 h. After cooling, CH2Cl2 (20 ml)
was added and the mixture was stirred vigorously. Remaining
solids were filtered and the filtrate was evaporated. The residue
was chromatographed on silica gel, eluting with Et2O to yield
the desired compound.
3-Ethylaminocarbonyl-1-methyl-2(1H)-pyridone 1a.—δH(250
MHz, CDCl3) 9.74 (1H, br s), 8.55 (1H, dd, Ja 7.2, Jb 2.2), 7.52
(1H, dd, Ja 6.6, Jb 2.2), 6.42 (1H, dd, Ja = Jb 6.9), 3.65 (3H, s),
3.51 (2H, dq, Ja = Jb 7.3), 1.25 (3H, t, J 7.3).
3-Ethylaminocarbonyl-2-methoxypyridine 1b.—δH(250 MHz,
CDCl3) 8.51 (1H, dd, Ja 7.5, Jb 2.0), 8.24 (1H, dd, Ja 4.9, Jb
2.1), 7.89 (1H, br s), 7.04 (1H, dd, Ja 7.5, Jb 4.9), 4.09 (3H, s),
3.48 (2H, dq, Ja = Jb 7.3), 1.25 (3H, t, J 7.3).
3-Ethoxycarbonyl-1-methyl-2(1H)-pyridone
2a.—δH(250
MHz, CDCl3) 8.13 (1H, dd, Ja 7.2, Jb 2.2), 7.55 (1H, dd, Ja 6.6,
Jb 2.2), 6.22 (1H, dd, Ja = Jb 6.9), 4.36 (2H, q, J 7.2), 3.59 (3H,
s), 1.38 (3H, t, J 7.2).
3-Ethoxycarbonyl-2-methoxypyridine 2b.—δH(250 MHz,
CDCl3) 8.30 (1H, dd, Ja 4.9, Jb 2.0), 8.15 (1H, dd, Ja 7.5, Jb
2.0), 6.94 (1H, dd, Ja 7.5, Jb 4.9), 4.36 (2H, q, J 7.2), 4.04 (3H,
s), 1.38 (3H, t, J 7.2).
References
1 C. G. Swain and J. F. Brown Jr., J. Am. Chem. Soc., 1952, 74, 2538.
2 (a) P. R. Rony, J. Am. Chem. Soc., 1969, 91, 6090 and references
cited therein; (b) H. T. Openshaw and N. Whittaker, J. Chem. Soc., C,
1969; (c) C.-W. Su and J. W. Watson, J. Am. Chem. Soc., 1974, 96,
1854; (d) F. Pietra and D. Vitali, Tetrahedron Lett., 1966, 46, 5701.
Paper 7/06533G
Received 8th September 1997
Accepted 7th January 1998
942
J. Chem. Soc., Perkin Trans. 2, 1998