The ammonia-free partial reduction of substituted pyridinium salts

Article abstract of DOI:10.1039/b517462g

This paper reports a study into the partial reduction of N-alkylpyridinium salts together with subsequent elaboration of the intermediates thus produced. Activation of a pyridinium salt by placing an ester group at C-2, allows the addition of two electron

Full text of DOI:10.1039/b517462g

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
The ammonia-free partial reduction of substituted pyridinium salts
Timothy J. Donohoe,*a Dale J. Johnson,a Laura H. Mace,a Rhian E. Thomas,a Jessica Y. K. Chiu,a
Jason S. Rodrigues,a Richard G. Compton,b Craig E. Banks,b Peter Tomcik,b Mark J. Bamfordc and
Osamu Ichiharad
Received 8th December 2005, Accepted 30th January 2006
First published as an Advance Article on the web 16th February 2006
DOI: 10.1039/b517462g
This paper reports a study into the partial reduction of N-alkylpyridinium salts together with
subsequent elaboration of the intermediates thus produced. Activation of a pyridinium salt by placing
an ester group at C-2, allows the addition of two electrons to give a synthetically versatile enolate
intermediate which can be trapped with a variety of electrophiles. Furthermore, the presence of a
4-methoxy substituent on the pyridine nucleus enhances the stability of the enolate reaction products,
and hydrolysis in situ gives stable dihydropyridone derivatives in good yields. These versatile
compounds are prepared in just three steps from picolinic acid and can be derivatised at any position on
the ring, including nitrogen when a p-methoxybenzyl group is used as the N-activating group on the
pyridinium salt. This publication describes our exploration of the optimum reducing conditions, the
most appropriate N-alkyl protecting group, as well as the best position on the ring for the methoxy
group. Electrochemical techniques which mimic the synthetic reducing conditions are utilised and they
give clear support for our proposed mechanism of reduction in which there is a stepwise addition of
two electrons to the heterocycle, mediated by di-tert-butylbiphenyl (DBB). Moreover, there is a
correlation between the viability of reduction of a given heterocycle under synthetic conditions and its
electrochemical response; this offers the potential for use of electrochemistry in predicting the outcome
of such reactions.
regiochemistry because hard reagents, such as organolithiums,
favour attack at the 2-position4 and softer reagents, such as
Introduction
organocuprates, attack at the 4-position.5 A recent example
showing the use of Grignard reagents adding to pyridinium salts
in conjunction with stereochemical control by a chiral auxiliary
is shown in Scheme 1.6 When hydride reagents are used as
nucleophiles, regiocontrol can be difficult to achieve and requires
careful choice of both reagent and conditions in order to be
synthetically useful: a pertinent recent example of the Fowler
reduction by Ganem et al. is shown in Scheme 1 (3 → 4).7 The
use of dissolving metals in ammonia (such as the Birch reduction)
The piperidine ring is one of the central motifs in natural product
chemistry, and this heterocycle is present in a wide variety of
compounds with biological activity. Many naturally-occurring
piperidines are present in complex structures and represent chal-
lenging synthetic targets.1 The use of pyridinium salts as precursors
to piperidines is an attractive approach, because control of a
reduction or addition process allows access to dihydropyridine,
tetrahydropyridine or piperidine derivatives directly. A further
advantage is the ready availability of various substituted pyridines
via literature routes and the ease of activation of this heterocycle
as a pyridinium salt. In this regard, there are two complementary
methods of effecting the reductive transformation of pyridinium
salts: (i) direct attack of a nucleophile onto the heterocycle,2
or (ii) the use of a reducing agent which provides electrons
and forms an anion capable of reacting with electrophiles.3 The
addition of organometallic reagents to pyridinium salts (especially
acylpyridinium salts) is well documented in the literature. The
choice of nucleophilic reagent is very important with respect to
aDepartment of Chemistry, University of Oxford, Chemistry Research
Laboratory, Mansfield Road, Oxford, UK OX1 3TA. E-mail: timothy.
donohoe@chem.ox.ac.uk; Fax: (01865) 275674; Tel: (01865) 275649
bPhysical and Theoretical Chemistry Laboratory, Oxford University, South
Parks Road, Oxford, UK OX1 3QZ
cDepartment of Medicinal Chemistry, Neurology & GI Centre of Excellence
for Drug Discovery, GlaxoSmithKline Research & Development Limited,
New Frontiers Science Park, Third Avenue, Harlow, Essex, UK CM19 5AW
Scheme 1 Reagents and conditions: (i) C6H5CH2MgBr; (ii) NaBH4; (iii)
NaBH4, PhOCOCl, MeOH, −78 ◦C.
dEvotech OAI Ltd., 111 Milton Park, Abingdon, Oxon, UK OX14 4SD
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1071–1084 | 1071
©

View Article Online
as a source of electrons to effect the reduction of pyridines has
been known since 1925.8 However, this reaction often gives multi-
component mixtures, limiting its synthetic utility. Modification of
the procedure by the use of pyridinium salts in the Birch reduction
has allowed regiocontrol of the reduction to be achieved, with the
conditions favouring formation of a 1,4-dihydropyridine.9
an electrophile. In order to test this hypothesis, the synthesis
of pyridinium salt 8 was accomplished by heating iso-propyl
picolinate 7 in neat methyl iodide overnight (Scheme 3). These
harsh conditions were necessary to overcome electronic deacti-
vation by the ester group adjacent to the heterocyclic nitrogen.
Subsequent reaction of pyridinium salt 8 under standard Birch
reduction conditions (sodium, ammonia and tetrahydrofuran at
Previously, we had shown that selective reduction of the pyridine
nucleus using Birch conditions is made possible by positioning
electron-withdrawing substituents on the heterocycle (Scheme 2).10
This methodology was found to be a general one, compatible
with a wide variety of electrophilic quenches which furnished the
substituted dihydropyridines 6 in excellent yields. However, the
approach suffers from a number of drawbacks. It was found that
at least two electron-withdrawing groups were needed to properly
activate the ring towards reduction and the relative positioning
of these groups was critical for the success of this reduction
(Scheme 2). In essence, the ability of the substituted pyridine
to stabilise a dianionic intermediate (generated by the addition
of two electrons) was crucial, thus restricting the placement of
the groups (see A, Scheme 2). Also, further modification of the
products generated by this reduction reaction was found to be
limited, particularly with respect to the fixed carbon substituent
pattern that the two electron-withdrawing groups imposed. Thus,
an investigation into the development of a more versatile method
of activating the pyridine nucleus was sought.
◦
−78 C) followed by quenching of the reaction with an acid led
to the formation of a crude product that was consistent with the
1
desired dihydropyridine 9 (R = H) by H NMR spectroscopy
(Scheme 3). However, attempts at purification failed because the
product was apparently unstable. The instability of the expected
1,2-dihydropyridine species 9 has been well documented in the
literature; indeed, the addition of catalytic p-toluenesulfonic
acid to a fresh sample of crude dihydropyridine in deuterated
chloroform led to extremely rapid degradation. One potential
route for decomposition of a putative dihydropyridine would be
re-aromatisation via air oxidation. Therefore, the reduction of 8
was repeated, followed by quenching with methyl iodide as an
electrophile. However, again the crude product of the reductive
alkylation reaction 9 (R = Me) was found to be just as unstable
to purification as the protonated species 9 (R = H), indicating
an alternative pathway for product decomposition. One strategy
for the isolation of unstable dihydropyridine intermediates is to
convert them into a more stable derivative in situ.12 Thus, the crude
diene 9 (R = H) was subjected to a Diels–Alder reaction. Initial
use of maleic anhydride as a dienophile to capture 9 offered no
advantage as attempts at purification of the adduct failed, possibly
because of facile retro-cycloaddition. Instead, the crude reaction
mixture was treated with dimethyl fumarate in dichloromethane,
which furnished the Diels–Alder adduct 10 as a mixture of isomers
(stereochemistry not assigned). However, the disappointing yield
of this reaction led to the conclusion that the intermediates
generated here were again too unstable to be synthetically useful.
Scheme 2 Partial reduction of 2,5-disubstituted pyridines.
Results and discussion
Scheme 3 Reagents and conditions: (i) MeI, D; (ii) Na, NH3, THF, −78 ◦C
then isoprene then NH4Cl or MeI; (iii) dimethyl fumarate, CH2Cl2, D.
Establishing conditions for the ammonia-free Birch reduction of
pyridinium salts
Pyridinium salts have already found limited use as substrates
for the Birch reduction and shown potential for controlling the
extent of reduction. Therefore, it was hypothesised that the further
activation of pyridines (bearing one electron-withdrawing group)
as pyridinium salts would negate the need for the second electron-
withdrawing group as shown in Scheme 2.11 In the reduction of
a pyridinium salt, by the addition of two electrons, the use of
an ester group at the C-2 position may allow for the generation
of an anion stabilised as an enolate, giving an opportunity to
alkylate the heterocyclic nucleus by quenching the reaction with
Therefore, an alternative approach was sought that would
lead to the transformation (and thus stabilisation) of the 1,2-
dihydropyridine species in situ. Initial ideas, such as hydrogena-
tion, were rejected as this would greatly reduce the synthetic
potential of the reduction product by removal of the double bonds.
We were aware that Comins and co-workers had successfully
developed the use of C-4 methoxy-substituted acyl pyridinium
salts as versatile electrophiles.13 The advantage of the placement
of a methoxy group at C-4 was that it allowed a method of avoid-
ing unstable 1,2-dihydropyridines after reduction: acid catalysed
1072 | Org. Biomol. Chem., 2006, 4, 1071–1084
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
hydrolysis in situ liberated a stable dihydropyridone. Using this
idea as a starting point, an efficient, one-step synthesis of methyl
4-methoxypicolinate 12 (Scheme 4) was achieved from picolinic
acid 11 via a method originally reported by Sundberg and Jiang.14
Compound 12 was then N-alkylated by use of the same procedure
as previously described to give the N-methylpyridinium salt 13
in excellent yield.11 It was reasoned that the use of traditional
Birch reduction conditions – sodium and liquid ammonia – would
be inappropriate and that an electron transfer agent such as di-
tert-butylbiphenyl (DBB) or naphthalene in combination with an
alkali metal could be used to facilitate reduction.15 The reasoning
behind this change was three-fold: firstly, the in situ acidic work-up
required to hydrolyse the enol ether would clearly be incompatible
with a liquid ammonia solvent; secondly, previous experience
had shown that methyl esters are often prone to nucleophilic
attack by amide anions formed during the Birch reduction; and
finally, the use of ammonia-free conditions would allow the use
of sensitive electrophiles, such as methyl chloroformate, as traps
for the reduction.15 Thus, pyridinium salt 13 was subjected to
the ammonia-free reduction/hydrolysis strategy using sodium and
naphthalene to achieve reduction and then methyl iodide as an
electrophilic quench, followed by reaction with aqueous acid.
The isolation of the desired dihydropyridone 14 in a yield of
71% validated the change in reaction conditions (Scheme 4); the
dihydropyridone product 14 was completely stable at ambient tem-
perature and its structure was proven by X-ray crystallography.†
The success of this alkylation/hydrolysis protocol lends credence
to the original hypothesis that the addition of two electrons
generated an enolate anion B, which was successfully alkylated
and then hydrolysed to a stable vinylogous amide derivative
in situ.
Optimising the N-alkyl substituent in the pyridinium salt
The next issue to be addressed was the removal of the N-
methyl group from 14 because deprotection was required to allow
variation in the substituents on the nitrogen for future studies.
Initial work focused on classical electrophilic demethylation
conditions such as the use of methyl chloroformate and boron
tribromide; however, only the starting material 14 was returned,
indicating that the nitrogen lone pair was strongly conjugated into
the enone system and thus unavailable for reaction. The use of
both nucleophilic conditions, e.g. methionine in methanesulfonic
acid, and oxidative conditions, e.g. chromium trioxide or Dess–
Martin periodinane, either returned the starting material or led to
decomposition.16
This failure to deprotect the nitrogen severely limited the scope
of the reduction for its general use in organic synthesis and
so it was decided to investigate other protecting groups on the
heteroatom. The first area of investigation involved groups that
could to be cleaved under acidic conditions to yield the N-
H dihydropyridone. However, both an N-trialkylsilyl group and
an N-boron trifluoride protected version of pyridinium salt 13,
when subjected to reduction, gave no identifiable products. The
use of the corresponding N-oxide of 12 was also investigated
but again reduction failed to give a single product. Attention
was, therefore, returned to carbon-based protecting groups. One
problem with attaching a range of groups on nitrogen was the poor
reactivity of pyridine 12 towards electrophiles; for example, both
benzyl and allyl bromides proved unreactive and did not achieve
complete conversion. Attempted reaction of 12 with acylating
agents only returned the starting material; this confirms literature
reports of the relative instability of acylpyridinium species at room
temperature and especially where ortho substitution on a pyridine
provides steric hindrance.13a All this evidence demonstrated a
lack of nucleophilicity of the substrate 12 towards alkyl halides
and so a more reactive electrophile was chosen to effect alkyla-
tion. Alkyl triflates (trifluoromethanesufonates) are very reactive
electrophiles, and so a test reaction using methyl triflate and
12 was conducted. The pyridinium salt 15 was formed easily at
room temperature, and was then added immediately to a solution
of sodium naphthalenide. After alkylation with methyl iodide
and acidic hydrolysis, dihydropyridone 14 was isolated in 44%
yield. Although this yield was less than for the corresponding
reaction of iodide salt 13, as a proof of concept the use of alkyl
triflates for alkylating pyridine 12 had been demonstrated. It
was decided to test if other alkyl groups could be introduced
via the triflate route and were then tolerated by the reaction
conditions. The 1-chloroethyl and homoallyl salts 16 and 17 were
prepared (and isolated) from the respective triflates to examine the
stability of more highly functionalised groups in the ammonia-
free Birch reduction (Scheme 5). Both triflate salts underwent
smooth reduction, with yields of the two dihydropyridones 18
and 19 being consistent with previous reductions of triflates –
thus the compatibility of the ammonia-free reduction with both
alkyl chlorides and isolated alkenes has been demonstrated. The
drop in yield seen here would appear to indicate some form of
counter-ion effect having a negative impact on the reduction;
compare the reduction of 15 and 13. Bearing in mind that
the aim of further investigation was the ability to deprotect
the nitrogen, the N-homoallyl dihydropyridone 19 was reacted
Scheme 4 Reagents and conditions: (i) SOCl2, NaBr, MeOH, D; (ii) MeI;
(iii) Na, naphthalene, −78 ◦C then MeI, then H3O+.
† Crystal data for 14: C9H13NO3, M = 183.21, monoclinic, P21/n, a =
◦
˚
7.1187(2), b = 12.4355(2), c = 10.6250(3) A, b = 105.5450(8) , V =
906.17(4) A , T = 150 K, Z = 4, l = 0.101 mm−1, reflections measured
3
˚
12 098 (2169 unique), Rint = 0.030, R = 0.0372 (all data). CCDC reference
number 294438. For crystallographic data in CIF or other electronic
format see DOI: 10.1039/b517462g.
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1071–1084 | 1073
©

View Article Online
with lithium–DBB, using methyl iodide as the quench, furnished
the dihydropyridone in both cases, and in particular the PMB
protected dihydropyridone 24 was isolated in an encouraging yield
of 74%, which is comparable to the analogous result obtained
using methyl protected salt 13 (Scheme 6). It should be noted that
yields suffered if too great an excess of lithium was used, which was
attributed to reduction of the aromatic protecting group as well as
the heterocycle. With this promising result in hand, deprotection
of the dihydropyridone 24 was attempted next. Exposure of 24 to
trifluoroacetic acid (TFA), followed by heating at reflux, led to the
formation of the N-H dihydropyridone 20 in a quantitative yield
(Scheme 6).
Thus far we have established optimal conditions for the acti-
vation and subsequent partial reduction of methoxypyridinium
salts to highly functionalised and versatile dihydropyridone inter-
mediates. We have demonstrated the efficiency of the ammonia-
free Birch reduction conditions and have optimised these to the
use of LiDBB. Also, we have identified the PMB group as the
most appropriate N-protecting group in the activation of the
methoxypyridine 12 as a pyridinium iodide salt.
Scheme 5 Reagents and conditions: (i) R–OSO2CF3; (ii) Na, naphthalene,
−78 ◦C then MeI, then H3O+; (iii) OsO4, NaIO4; (iv) Et3N, D.
under Johnson–Lemieux oxidative cleavage conditions, followed
by heating in triethylamine to trigger a b-elimination, liberating
the N-H dihydropyridone 20 (Scheme 5). Although this sequence
was successful in deprotecting the nitrogen and forming 20, a poor
yield of 10% was not encouraging for future use.
Versatility of the methodology and elaboration of the
dihydropyridones
With the basic elements of the synthetic approach established, we
next investigated the utility of the intermediates obtained by this
methodology. Thus, the PMB protected salt 22 was subjected to
the ammonia-free Birch reduction, and the enolate quenched with
a variety of electrophiles, including a proton source and carbon
electrophiles comprising both alkyl and acyl species (Scheme 7).11
We have shown that reduction of a pyridinium iodide salt
is superior to the corresponding triflate salt, and so a new
series of iodides were sought to activate pyridine 12; in this
regard it was decided to evaluate both benzyl iodide and 4-
methoxybenzyl (PMB) iodide as electrophiles. The increased
leaving group aptitude of the iodide proved to be critical, as both
the benzyl and PMB protected pyridinium salts 21 and 22 were
isolated in good yields as easily handled micro-crystalline solids
(Scheme 6). During the course of these studies a second reducing
system of lithium and DBB was also tested on compound 13
and found to be superior to sodium and naphthalene in terms
of isolated yield. Electrochemical data from experiments designed
to mimic the ammonia-free reduction conditions showed that the
reduction potential of DBB (−2.8 V) is more negative than that
of naphthalene (−2.3 V), thus the radical anion is less stable and
so LiDBB is a stronger reducing agent.17 Treatment of 21 and 22
Scheme 7 Reagents and conditions: (i) Li, DBB, −78 ◦C; (ii) i-BuI;
(iii) ClCO2Me; (iv) NH4Cl; (v) X(CH2)4I (X = Cl, I); (vi) H3O+;
(vii) TFA, D.
Scheme 6 Reagents and conditions: (i) BnI or PMBI; (ii) Li, DBB, −78 ◦C
then MeI, then H3O+; (iii) TFA, D.
1074 | Org. Biomol. Chem., 2006, 4, 1071–1084
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Interesting points to note are that the yield of protonation was
independent of the source of the acid used; saturated aqueous
ammonium chloride solution, methanol or 2,6-di-tert-butylphenol
all gave yields of around 65% of the dihydropyridone 27. In
addition, the use of a mixed halide electrophile showed complete
selectivity for displacement of the better leaving group (22 → 25),
as expected. Both halide-containing dihydropyridones 25 and 26
were deprotected to test if potentially sensitive functionality could
survive the strongly acidic conditions. Pleasingly, both the N-H
dihydropyridones 30 and 31 were isolated, with yields of 91%
for the chloride and 76% for the iodide, with the lower value
for the iodide perhaps revealing some sensitivity to the reaction
conditions. The only class of electrophile that worked poorly
under the partial reduction conditions was an aldehyde.18 It is
possible that during the acidic hydrolysis conditions (and as the
reaction is warmed to room temperature) a retro-aldol reaction
was occurring, leading to decomposition in situ.
These examples show the methodology to be a viable way of
quickly accessing a synthetically useful template from a cheap and
plentiful starting material, picolinic acid, in just three steps.
The potential for further modification and elaboration of
these dihydropyridones, and their use in synthesis, can easily be
predicted from substantial precedent for the reactivity of these sys-
tems in the literature. We have performed a range of derivatisation
reactions on these templates, designed to show the flexibility that
such platforms have for the introduction of groups at any position
on the pyridone nucleus. The reactions detailed in Scheme 8 show
that this is indeed possible. For example, functionalisation at C-6
was accomplished using an organo-cuprate in conjunction with
Lewis acid activation of enone (24 → 32, 80 : 20 dr).19 The relative
stereochemistry of compound 32 (major isomer) was assigned
by NOE experiments. Conjugate addition of hydride to 24 was
also accomplished using L-selectride20 and the PMB group then
removed from tertiary amine product 33 using hydrogenolysis
conditions to give 34 (surprisingly, we found that dihydropyridone
24 was itself inert to hydrogenolysis). Next, functionalisation at
C-3 was examined by utilising the enolate derivative of pyridones
25 and 14.21 Intramolecular alkylation of 25 was accomplished by
treatment with sodamide (this base alone was capable of enolising
the dihydropyridone structure). Formation of the cis-azadecalin
ring system 35 was confirmed by X-ray crystallographic analysis
of this compound.‡ Moreover, intermolecular alkylation at C-3
was also readily accomplished when 14 was reacted with sodamide
and then methyl iodide, to furnish 36 as a single diastereoisomer.
We were unable to prove the relative stereochemistry of 36 beyond
all doubt and it is assigned by analogy to 35. Finally, the enamine-
like character of these dihydropyridones was harnessed (and
functionalisation at C-5 accomplished) by reaction of 24 with
potent electrophiles.22 For example, reaction with iodine gave
vinyl iodide 37 in excellent yield, ready for derivatisation by
radical reaction or transition metal coupling. Preliminary screens
in this area showed that palladium coupling of 37 could be best
accomplished with catalytic Pd/C under heterogeneous conditions
Scheme 8 Reagents and conditions: (i) CuBr·SMe2, THF; BF3·OEt2,
−78 ◦C; MeMgBr, −10 ◦C; (ii) Li(sec-Bu)3BH, THF, −78 ◦C; (iii) H2,
Pd/C, MeOH; (iv) NaNH2, NH3, −40 ◦C; (v) NaNH2, NH3, −78 ◦C,
MeI; (vi) I2, K2CO3, CH2Cl2, rt; (vii) allyl iodide, MeCN, D, 48 h;
(viii) PhB(OH)2, DME/H2O, cat. Pd/C.
in the presence of phenyl boronic acid to yield 39.23 Moreover, allyl
iodide was also capable of reacting with the enamine 24 to yield
the C-5 allyl compound 38 in reasonable yield (the site of initial
allylation, and the ensuing [3,3] sigmatropic rearrangement that
would follow from N- or O-allylation was not investigated).
Thus, the varied reactivity pattern of dihydropyridones such
as 24 has been explored and there is clearly plenty of scope to
perform significant modification to the heterocyclic skeleton post-
partial reduction.
Variation in the substitution pattern of activated pyridinium salts
Our next remit was to investigate variation in the position of the
substituents around the pyridine ring of the starting substrate.
Two criteria were applied in the choice of substrate to be reduced.
Firstly, using our knowledge of the mechanism of reduction only
certain substitution patterns (an ester at C-2 or C-4) were predicted
to stabilise the anion generated upon addition of the electrons
as an enolate. Moreover, positioning of the methoxy group meta
‡ Crystal data for 35: C19H23NO4, M = 329.40, monoclinic, P21/c, a =
◦
˚
8.0683(2), b = 10.3413(2), c = 20.1216(4) A, b = 100.6130(9) , V =
1650.16(6) A , T = 150 K, Z = 4, l = 0.093 mm−1, reflections measured
3
˚
13 884 (3656 unique), Rint = 0.030, R = 0.0397 (all data). CCDC reference
number 294439. For crystallographic data in CIF or other electronic
format see DOI: 10.1039/b517462g.
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1071–1084 | 1075
©

View Article Online
to the ester was deemed essential to allow stabilisation of the
product from hydrolysis.10c Secondly, the starting material needed
to be readily available to allow its use in synthesis, because multi-
step routes to the pyridinium starting materials would not be
attractive. Therefore, it was decided to investigate the properties
of commercially available 6-hydroxypicolinic acid 40 to test if the
position of the methoxy group was important during reduction.
Esterification and etherification of 40 were achieved in a one-pot
reaction by the addition of methyl iodide and silver carbonate to 6-
hydroxypicolinic acid (Scheme 9).24 The choice of base was crucial
to the success of the reaction, with the use of potassium carbonate
leading to methylation of the ring nitrogen instead of the hydroxy
group, which then led to pyridone formation. Attempts to form
the pyridinium salt of methyl 6-methoxypicolinate 41 using methyl
iodide led exclusively to the synthesis of the N-methyl pyridone
42 as observed previously.25 To prevent this side reaction from
occurring, a non-nucleophilic counter-ion was used and attention
was turned to the use of an alkyl triflate to alkylate 41. Although
previous work on the 4-methoxy salts 13 and 15 had shown the
iodide derivatives to be superior to their triflate counterparts, the
latter had worked in the ammonia-free Birch reduction and would
allow the exploration of a new area of pyridinium salt chemistry.
Consequently, treatment of pyridine 41 with methyl triflate led
to the quantitative formation of pyridinium salt 43 (Scheme 9).
Attempts to exchange the triflate counter-ion of 43 with iodide
always led to the formation of 42.
Scheme 10 Reagents and conditions: (i) Na, naphthalene, THF, −78 ◦C;
(ii) RX; (iii) then H3O+; (iv) LiDBB, THF, −78 ◦C; (v) LiDBB, THF,
−78 ◦C, NH4Cl(aq)
.
work-up of the reaction, to trigger the cyclisation of the amino
acid back to the desired dihydropyridone, increased the yield of the
methylated product 44 to 47% (LiDBB conditions did not improve
the yield), comparable with that obtained from the 4-methoxy
derivative 14. However, further studies showed that this yield was
the maximum that could be obtained from this system and could
not be optimised further despite much experimentation. In a move
to simplify the reaction, and to negate the need for a strongly acidic
work-up, the ammonia-free Birch reduction of pyridone 42 was
examined. It was proposed that the addition of two electrons to
42 would lead to the formation of a bis-enolate E, opening up
the potential for selective alkylation, and quenching with either
a carbon electrophile or a proton leading to dihydropyridone
formation. Thus, reduction of 42 was investigated under the
ammonia-free Birch conditions using LiDBB. Protonation of the
dianion E using saturated aqueous ammonium chloride was found
to be substantially better than for the pyridinium salt 43, with the
dihydropyridone 45 now isolated in a respectable yield of 58%.
Moreover, reductive alkylation of 42 led to the formation of the
di-alkylated product 46, again in 58% yield. However, it was found
that it was not possible to control the selectivity of alkylation with
variations in experimental conditions (equivalents of electrophile,
etc.) all producing each of the possible isomers of mono- and
di-alkylation in differing ratios. When the reaction was pushed
through to the di-alkylated dihydropyridone 46, a potentially
Scheme
(ii) MeOTf; (iii) MeI, D.
9 Reagents and conditions: (i) MeI, Ag2CO3, CHCl3, D;
Following standard ammonia-free reductive alkylation proce-
dures, using sodium–naphthalene as the source of electrons (to
allow comparison with previous work on triflate salt 15) and
methyl iodide as the electrophile, reaction of 43 gave the 6-
dihydropyridone 44 in a yield of 36% after acidic hydrolysis
of the enol ether (Scheme 10). Protonation of the enolate (C)
proved to be more problematic with a highest yield of 45 being
20% obtained using methanol as proton source. Evidence was
accumulated which suggested another reaction pathway was
operating during the reaction. Examination of the aqueous layer
from the reaction work-up using 1H NMR spectroscopic analysis
showed the presence of peaks in the vinylic region which did
not correspond to the dihydropyridone 44. It was proposed that,
during acidic hydrolysis, protonation of D could occur not only on
the enol ether but also on the nitrogen. This pathway could lead
to ring-opening and, after ester hydrolysis, to the formation of an
amino acid; basification would generate the highly water soluble
carboxylate form. The use of acetic anhydride, added during the
1076 | Org. Biomol. Chem., 2006, 4, 1071–1084
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
voltammograms using the commercially available DIGISIMTM
software package; the value reported is the average result over
all scan rates.
Cyclic voltammetry was undertaken on a selection of aromatic
heterocyclic compounds to provide further insight into the mecha-
nism of the partial reduction. The compounds examined and their
voltammograms are shown in Fig. 1.3a
useful diastereomeric ratio of 80 : 20 was obtained, although the
diastereomers were inseparable and remain unassigned. Attempts
to reproduce this methodology with the 4-pyridone 47 were
unsuccessful,26 with only trace amounts of the dihydropyridone 48
formed. Ultimately, the lack of enolate differentiation in reactivity
of the bis-enolate E limited the flexibility and generality of this
approach to synthetic applications.
Each compound investigated (except 43) gave a one-electron,
quasi-reversible wave, as evidenced by the shift in peak potential
with scan rate. The two parameters above were determined by
fitting the experimental data to electrochemical models using
DIGISIMTM. Table 1 summarises the results.
Correlation of synthetic studies with electrochemical analysis
Introduction. In order better to understand the electron transfer
processes that are taking place under synthetic reducing conditions
we were interested in developing an electrochemical analysis
that would replicate the conditions of lithium–DBB in THF at
The experimental error for the E0ꢀ values is 0.05 V and can be
attributed, in part, to the slight drift in the potential of the quasi-
reference electrode. However, the electrode appeared to behave
well during experiments and the reproducibility of results was
good. DBB was included in the study to validate the conditions
used for the electrochemical cell. Comparison of the data collected
for DBB with that found in previous studies1717 showed that the
system employed was working satisfactorily.
◦
−78 C and would allow us to measure the reduction potentials
of a variety of aromatic heterocycles.18,27 We aimed to determine
if the formation of a radical anion or dianion from aromatic
heterocycles was mechanistically reasonable. We also aimed to
obtain data regarding the rate of electron transfer to these aromatic
compounds in order to understand the role of the electron transfer
agent (typically DBB) in the ammonia-free reduction.28 In the
shorter term, we aimed to use electrochemistry as a technique
to enable us to predict not only the products of a reduction but
also the relative rate at which various compounds will be reduced
under such lithium/THF conditions; this knowledge will have
useful applications in organic synthesis.
The mechanism that we proposed for the partial reductions
indicates that these substrates undergo a two-electron reduction
to yield an enolate (e.g. B) that then reacts upon the addition
of an electrophile (Scheme 4). This implies that the radical
and anion derived from these heterocycles should both be less
reducing than the DBB radical anion (i.e. have a less negative E0ꢀ ).
Consequently, it was also expected that further electrochemical
studies of the pyridinium salts should yield two-electron reduction
waves representing the formation firstly of a radical and then of
an enolate.
A
sealed cell was used for cyclic voltammetry experi-
ments, containing THF and the supporting electrolyte, tetra-n-
butylammonium perchlorate (0.5 M). Silver wire was used as the
quasi-reference electrode, the potential of which is open to some
drift ( 10 mV); however, in most cases the electrode was found to
behave in a stable manner upon repetition of the electrochemistry
experiments. The working electrode was, in general, a 1 mm
radius platinum macroelectrode. When investigating DBB a 5 lm
microdisc was used. All experiments were carried out under
anhydrous and oxygen-free conditions at T = 200 K.
The cyclic voltammograms for three species (13, 21, 22) did
indeed show two single-electron reduction waves, both of which
were quasi-reversible. Table 2 gives the formal potentials for both
electron transfers. As expected, the reduction potentials for the
second electron transfer are less negative than that for the single
electron reduction of DBB, confirming that it is reasonable to
assume that an enolate such as B can be formed during synthetic
reductions of the pyridinium salts by LiDBB.
For each substrate, two parameters were investigated: the formal
potential E0ꢀ and the diffusion coefficient D were determined. The
formal potential is defined by eqn. (1), where ERp ed and EpOx are
the cathodic and anodic peak potentials in an electrochemically
reversible (or quasi-reversible) cyclic voltammogram.
The E0ꢀ values for both the first and second electron reductions
gave a relative order for the thermodynamic ease of reduction of
the pyridinium salts, with 21 being the easiest to reduce and 13
ꢀ
ꢁ
EpRed + EpOx
ꢀ
E0
=
(1)
Table 1 Electrochemical data determined by DIGISIMTM
2
Here, E0ꢀ is the formal potential of the redox couple, a measure
of the thermodynamic ‘ease’ of reduction of the substrate, F is the
Faraday constant, n is the number of electrons transferred and A
is the electrode area The diffusion coefficient D is a measure of
the rate at which the molecule can diffuse through the solvent and
is a function of several variables including the effective size of the
species and the solvent viscosity. The coefficient D appears in eqn.
(2) and can be inferred from the gradient of a plot of the peak
current density, ip, against the square root of the scan rate, m, for
an electrochemically reversible process. Cbulk is the concentration
0ꢀ
Compound
106 D/cm2 s−1
E /V vs. Ag
13
2.4
3.1
0.75
—
1.7
3.5
−0.63
−0.50
−0.58
−1.03
−1.66
−2.80
21
22
43
42
DBB
Table 2 First and second formal potentials
ꢀ
ꢀ
Compound
E0 /V vs. Ag (1st e−)
E0 /V vs. Ag (2nd e−)
of the bulk solution.
1
ꢂ
ꢃ
/
2
F3
RT
1
1
3
/ /
2 2
/
2
13
−0.63
−0.50
−0.58
−2.80
−1.49
−1.29
−1.38
—
ip = 0.4463Cbulk
m AD n
(2)
21
22
DBB
Eqn. (1) and (2) were used to generate initial estimates for our
parameters which were then established by simulation of the cyclic
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1071–1084 | 1077
©

View Article Online
Fig. 1 Cyclic voltammograms of electron transfer agents, pyridinium salts and pyridones.
being the most difficult (21 > 22 > 13). However, the difference
in half-wave potentials for the three compounds is relatively small
and so an explanation for this order may be quite subtle.
is clearly different from the pyridinium salts and should give a
dianion D upon addition of two electrons.
As the synthetic results for both 42 and 43 indicate a two-
electron reduction mechanism (albeit with poor yields), it was
expected that a two-electron reduction wave would be observed
electrochemically at potentials more positive than DBB, Table 3.
The results of the cyclic voltammetry in Table 3 are in line with
these expectations. Note the absence of a back peak on the cyclic
voltammogram for 43 which indicated that both electron transfers
are irreversible. This indicated that both of the reduced species are
involved in homogeneous reactions in situ. Two further, smaller,
Investigations then moved on to 42 and 43, both of which have
been shown to be relatively poor substrates for the ammonia-free
partial reduction. The observation that pyridinium salt 43 does
not give satisfactory yields under synthetic reducing conditions is
particularly surprising given its similarity to substrate 13, the 4-
methoxy analogue. As outlined earlier, Scheme 10, the mechanism
for the reduction of 43 is expected to be almost identical to that
of compound 13, forming anion C in solution. The pyridine 42
1078 | Org. Biomol. Chem., 2006, 4, 1071–1084
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Table 3 First and second formal potentials
performed on commercially available pre-coated plates (Merck
silica Kieselgel 60F254).
ꢀ
ꢀ
Compound
E0 /V vs. Ag (1st e−)
E0 /V vs. Ag (2nd e−)
Proton and carbon NMR spectra were recorded on Bruker DPX
400 Fourier transform spectrometer using an internal deuterium
lock. Chemical shifts are quoted in parts per million (ppm)
downfield of tetramethylsilane. Coupling constants J are quoted in
Hz and are not rationalised. Carbon NMR spectra were recorded
with broad band proton decoupling and NMR assignments
were made possible using Attached Proton Test (APT) and
HMQC spectra where appropriate. Infrared spectra were recorded
on a Perkin-Elmer 1600 FTIR spectrophotometer. Positive ion
chemical ionisation (CI) mass spectra and accurate mass data
were recorded on a Micromass GCT instrument connected to an
Agilent 6890 Series GC system.
43
−1.03
−1.66
−2.80
−2.10
−2.00
—
42
DBB
reduction peaks can also be seen in the voltammogram, possibly
from the products of reactions the reduced species underwent.
Compound 42 has a quasi-reversible first electron reduction
but the second electron reduction is irreversible, again indicating
reaction of the second reduced species in situ. For compounds 42
and 43 it is clear that neither give clean electrochemical responses
when compared to their isomeric analogues (13, 22). Moreover,
both of them are significantly more difficult to reduce than the
4-methoxy analogues (compare 13: E0ꢀ = −0.63, −1.49 with 43:
E0ꢀ = −1.03, −2.10 V). Not only does cyclic voltammetry show
that 42 and 43 are relatively difficult to reduce but it also hints
that the reduced species, once formed, are capable of undergoing
reaction in situ. Thus, there is a clear correlation between the
appearance of the voltammograms, and the cleanliness with which
the compounds are reduced under synthetic conditions.
Electrochemical instrumentation and procedure
The electrochemical measurements were made using a computer-
controlled l-Autolab system (EcoChemie, Utrecht, Netherlands)
potentiostat. The electrochemical cell was an airtight three-
electrode arrangement. The counter electrode was a large area
bright platinum wire with a pseudo silver wire as the reference
electrode. The working electrodes were either a 5 lm (radius) or
a 1 mm (diameter) platinum disc electrode. The working electrode
was polished on soft lapping pads (Kemet, UK) with 1.0 lm
sized aqueous alumina slurry before rinsing in de-ionised water
with a resistivity not less than 18 MXcm from an Elgastat filter
system (Vivendi, Bucks., UK). Before carrying out electrochemical
experiments the working electrode’s radius (for the microelectrode)
was electrochemically calibrated. The working electrode and all
components of the cell were rigorously dried before use. Tetra-
n-butylammonium perchlorate (ca. 1.36 g), the supporting elec-
trolyte, and the substrate (0.04 mmol) were dissolved in anhydrous
THF (40 cm3). The low temperature experiments were conducted
in a dry ice–acetone bath (200 K). Before taking measurements
the system was degassed thoroughly by bubbling impurity-free
nitrogen gas (BOC Gases, UK) into the solution to remove any
dissolved oxygen for at least 20 min.
Conclusions
This paper describes our studies on the partial reduction of
electron-deficient pyridinum salts. By judicious choice of pyri-
dinium substitution pattern, and also the introduction of a p-
methoxybenzyl group on the heterocyclic nitrogen, we were able
to effect smooth partial reduction. Subsequent alkylation and
hydrolysis, all in situ, furnished a range of dihydropyridones
with the potential for the introduction of a variety of groups
at the C-2 position. The rich functionality contained within the
dihydropyridone ring systems affords highly versatile intermedi-
ates and it is worth noting that these valuable compounds are
prepared from picolinic acid in just three steps. We have also
described the electrochemical analysis of a range of pyridinium
heterocycles utilising conditions that mimic those attained under
synthetic conditions. This study has provided support for a
proposed mechanism of reduction involving the addition of two
electrons to form an enolate, all mediated by electron transfer
from LiDBB. Pyridinium salts with various substitution patterns
were examined both synthetically and electrochemically, and a
correlation was found between the yields obtained during synthetic
transformations and the appearance of the cyclic voltammogram.
This could have significant ramifications for the future use of
electrochemistry as a predictive technique for the outcome of
partial reduction reactions.
General procedures
(A) Sodium–naphthalene. Sodium (5 equiv.) and naphtha-
lene (5 equiv.) were added to THF (40 mL) at rt and sonicated
for 4 h forming a dark green solution under an atmosphere of
argon. The reaction was cooled to −78 ◦C and the pyridinium salt
was added either as a solid or in a solution of THF (10 mL). The
◦
reaction was stirred for 1 h at −78 C and then the electrophile
(5 equiv.) was added dropwise. After a further one hour at −78 ◦C,
saturated NH4Cl solution (10 mL) was added and stirred for
15 min, followed by HCl (10 mL, 3 M) and the reaction was allowed
to warm to rt over 16 h. The aqueous layer was washed with Et2O
(3 × 50 mL), basified with saturated Na2CO3 solution until the pH
was >9 and extracted with CH2Cl2 (6 × 25 mL). The combined
organic extracts were dried (MgSO4) and concentrated in vacuo.
Experimental
General experimental
All solvents were distilled before use. Tetrahydrofuran was
freshly distilled from sodium-benzophenone ketyl radical whilst
dichloromethane was freshly distilled from calcium hydride. All
non-aqueous reactions were carried out under argon using oven-
dried glassware.
(B) Lithium–di-tert-butylbiphenyl (LiDBB). Sliced lithium
wire (4 equiv.) and DBB (4 equiv.) were added to a Schlenk tube
containing glass anti-bumping granules under an atmosphere of
argon. The mixture was stirred vigorously for 2–4 h until the the
lithium had been reduced to a fine powder. THF (30 mL) was
Flash column chromatography was carried out using Merck
Kieselgel 60 (230–400 mesh). Thin layer chromatography was
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1071–1084 | 1079
©

View Article Online
added and the resultant deep blue-green solution was cooled to
−78 ◦C and stirred for 30 min. The pyridinium salt (1 equiv.) was
added as a solid and the reaction stirred for a further 15 min.
1,2-Dibromoethane (filtered through K2CO3 and MgSO4) was
added dropwise until the blue-green colour was quenched. The
electrophile (4 equiv.) was added dropwise and the reaction stirred
for 3 h before NH4Cl (5 mL, saturated solution) was added,
followed by stirring for a further 15 min. HCl (10 mL, 1 M)
was added and the reaction was warmed to rt and stirred for
16 h. The reaction was extracted with Et2O (3 × 50 mL) and
the combined organic extracts were washed with brine (50 mL,
saturated solution), dried (MgSO4) and concentrated in vacuo.
(5 mL) and stirred for 16 h at room temperature. The solution was
concentrated in vacuo and the crude residue was purified via flash
column chromatography (SiO2, 5 : 95 MeOH–CH2Cl2) to furnish
the pyridinium salt 15 (2.03 g, 100%) as a white solid. Mp 69–71 ◦C
(CH2Cl2); mmax (KBr)/cm−1: 3078, 1637; dH (400 MHz, CDCl3) 8.87
(1 H, d, J 7.2, CH), 7.84 (1 H, d, J 3.1, CH), 7.60 (1 H, dd, J 7.2
and 3.1, CH), 4.40 (3 H, s, NCH3), 4.16 (3 H, s, OCH3), 4.05 (3 H, s,
OCH3); dC (100 MHz, CDCl3) 171.9, 159.8, 151.5, 143.5, 117.6,
114.5, 58.8, 54.8, 47.4; MS (ESI) 182 (M + H+, 100%); HRMS
(ESI) found 182.0826, C9H12NO3 (M+) requires 182.0817.
1-(2-Chloroethyl)-4-methoxy-2-methoxycarbonylpyridinium
trifluoromethanesulfonate 16
2-iso-Propoxycarbonyl-1-methyl-pyridinium iodide 8
A solution of pyridine 12 (820 mg, 4.90 mmol) and chloroethyl
trifluoromethanesulfonate (1.03 g, 4.84 mmol) in CH2Cl2 (10
ml) was stirred at room temperature for 3 h. The solution was
concentrated in vacuo and purified by gradient flash column
chromatography (SiO2, 1 : 9 acetone–CH2Cl2 to acetone) to isolate
the product 16 (835 mg, 44%) as an oil which was thoroughly dried
under high vacuum prior to further use. mmax (KBr)/cm−1: 1636;
dH (400 MHz, CDCl3) 9.00 (1 H, d, J 7.2, CH), 7.89 (1 H, d,
J 3.4, CH), 7.68 (1 H, dd, J 7.3 and 3.2, CH), 5.23 (2 H, t, J 5.5,
ClCH2CH2N), 4.22 (3 H, s, CH3), 4.08 (3 H, s, CH3), 4.04 (2 H,
t, J 5.5, ClCH2CH2N); dC (100 MHz, CDCl3) 172.5, 160.1, 151.7,
143.6, 118.1, 114.5, 59.1, 59.0, 55.1, 43.5; HRMS (ESI) found
230.0584, C10H13NO3Cl (M+) requires 230.0584.
A solution of pyridine 7 (2.35 g, 14.3 mmol) in methyl iodide
(10 ml, 160 mmol) was heated at reflux for 16 h. Et2O (30 ml) was
added, and the resulting suspension filtered to give compound
8 (3.87 g, 88%) as a yellow solid which was washed with Et2O
(2 × 10 ml) and dried under vacuum. Mp 120–123 ◦C (Et2O); mmax
(nujol)/cm−1: 1735, 1621, 1457, 1283; dH (400 MHz, CDCl3) 9.82
(1 H, d, J 6.0, CH), 8.86 (1 H, dd, J 7.9 and 7.9, CH), 8.53 (1 H,
dd, J 7.9 and 1.2, CH), 8.44 (1 H, ddd, J 7.6, 6.4 and 1.2, CH),
5.34 [1 H, sp, J 6.3, CH(CH3)2], 4.74 (3 H, s, NCH3), 1.43 [6 H,
d, J 6.3, CH(CH3)2]; dC (100 MHz, CDCl3) 158.6, 150.2, 147.3,
142.5, 131.4, 130.2, 73.9, 50.0, 21.8; HRMS (ESI) found 180.1023,
C10H14NO2 (M+) requires 180.1025.
2-Methyl-2-aza-bicyclo[2.2.2]oct-7-ene-3,5,6-tricarboxylic acid
3-iso-propyl ester 5,6-dimethyl ester 10
1-But-3-enyl-4-methoxy-2-methoxycarbonylpyridinium
trifluoromethanesulfonate 17
A solution of dihydropyridine 9 (216 mg, 1.20 mmol) and dimethyl
fumarate (203 mg, 1.41 mmol) in CH2Cl2 (10 ml) was heated at
55 ◦C for 3 h. The solvent was removed in vacuo and the crude
residue purified by gradient flash column chromatography (SiO2,
petroleum ether to 5 : 95 acetone–petroleum ether) to produce the
Diels–Alder adduct 10 (59 mg, 15%, 2 : 1 dr) as a yellow oil. mmax
(thin film)/cm−1: 2982, 2933, 1732, 1436, 1374, 1200; dH (400 MHz,
CDCl3) (minor isomer): 6.73–6.69 (1 H, m, C5-H), 6.04 (1 H, t,
J 7.3, C4-H), 5.01 [1 H, sp, J 6.4, CH(CH3)2], 4.01 (1 H, br. d,
J 6.6, 6-H), 3.77 (3 H, s, CO2CH3), 3.74–3.64 (1 H, m, 7-H), 3.70
(3 H, s, CO2CH3), 3.53–3.48 (1 H, m, 3-H), 3.14 (1 H, d, J 1.4,
2-H), 2.89 (1 H, dd, J 5.2 and 1.7, 8-H), 2.50 (3 H, s, NCH3), 1.24
(6 H, d, J 6.3, 2 × CHCH3); (major isomer) 6.52–6.48 (1 H, m,
C5-H), 6.31 (1 H, t, J 7.1, C4-H), 4.98 [1 H, sp, J 6.3, CH(CH3)2],
3.80 (1 H, br. dd, app. J 6.0 and 2.2, 6-H), 3.78 (3 H, s, CO2CH3),
3.74–3.64 (1 H, m, 7-H), 3.68 (3 H, s, CO2CH3), 3.41–3.37 (1 H, m,
3-H), 3.09 (1 H, dd, J 6.0 and 2.9, 8-H), 2.86 (1 H, d, J 1.8, 2-H),
2.61 (3 H, s, NCH3) 1.23 (3 H, d, J 6.4, CHCH3), 1.22 (3 H, d,
J 6.4, CHCH3); dC (100 MHz, CDCl3) (minor isomer): 173.7,
173.3, 172.3, 138.6, 127.9, 68.5, 65.4, 56.9, 52.5, 52.5, 46.0, 43.5,
41.7, 36.0, 22.0; (major isomer): 173.7, 173.3, 172.3, 135.3, 131.3,
68.6, 61.7, 56.5, 52.7, 52.5, 44.2, 41.2, 38.7, 35.9, 22.0; HRMS
(ESI) found 326.1605, C16H24NO6 (M + H+) requires 326.1604.
Triflic anhydride (0.90 mL, 6.5 mmol) was added to a solution of
homoallyl alcohol (0.43 mL, 5.0 mmol) and Hu¨nig’s base (0.95 mL,
5.5 mmol) in toluene (5 mL) at 0 ◦C. The mixture was stirred for
16 h at room temperature, and the resultant suspension was filtered
to give a brown solution to which pyridine 12 (0.678 g, 4.06 mmol)
was added. The mixture was stirred for 16 h and concentrated in
vacuo to give the crude product as a brown oil which was taken
up into CH2Cl2 (30 mL) and extracted with distilled H2O (6 ×
40 mL). The aqueous extracts were combined, concentrated in
vacuo and thoroughly dried under high vacuum over phosphorus
pentoxide to give pyridinium salt 17 (1.49 g, 99%) as an oil. mmax
(KBr)/cm−1: 1743 and 1636; dH (400 MHz, CDCl3) 8.95 (1 H, d,
J 7.2, 6-H), 7.85 (1 H, d, J 3.0, 3-H), 7.65 (1 H, dd J 7.2 and
=
3.1, 5-H), 5.77 (1 H, ddt, J 17.0 10.0 and 7.0, CH CH2), 5.09
=
(1 H, dd, J 10.2 and 0.4, CH CHAHB), 5.00 (1 H, dd, J 17.2 and
=
0.8, CH CHAHB), 4.87 (2 H, t, J 7.0, NCH2), 4.19 (3 H, s, CH3),
4.09 (3 H, s, CH3), 2.66 (2 H, dt, J 7.0 and 7.0, NCH2CH2); dC
(100 MHz, CDCl3) 171.9, 160.0, 151.0, 143.3, 131.9, 120.5, 117.8,
114.7, 58.9, 58.3, 55.0 and 35.7; HRMS (ESI) found 222.1131,
C12H16NO3 (M+) requires 222.1130.
(2-RS)-1-(2-Chloroethyl)-2-methyl-5,6-dihydropyridone-2-
carboxylic acid methyl ester 18
Sodium (112 mg, 4.90 mmol), naphthalene (776 mg, 6.06 mmol)
and pyridinium salt 16 (365 mg, 0.96 mmol) were subjected to
general procedure A using methyl iodide (0.5 mL, 8.03 mmol)
as the electrophile. The residue was purified by reverse-phase
chromatography (0.5 M NaCl) to give the product 18 (95.1 mg,
4-Methoxy-2-methoxycarbonyl-1-methylpyridinium
trifluoromethanesulfonate 15
Methyl trifluoromethanesulfonate (0.75 mL, 6.6 mmol) was added
to a stirred solution of pyridine 12 (1.02 g, 6.13 mmol) in CH2Cl2
1080 | Org. Biomol. Chem., 2006, 4, 1071–1084
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
43%) as a yellow oil. mmax (thin film)/cm−1: 2956, 1737, 1640, 1585,
1436, 1373 and 1272; dH (400 MHz, CDCl3) 7.07 (1 H, d, J 7.8,
CDCl3) 7.39–7.27 (5 H, m, Ph), 7.00 (1 H, d, J 7.8, NCH), 5.01
=
(1 H, d, J 7.8, NCH CH), 4.55 (1 H, d, J 15.8, NCHAHB), 4.42
=
=
NCH CH), 5.05 (1 H, dd, J 7.8 and 1.1, NCH CH), 3.78–3.60
(3 H, m, CH2Cl and NCHAHBCH2Cl), 3.74 (3 H, s, CO2CH3),
3.57–3.50 (1 H, m, NCHAHBCH2Cl), 2.89 (1 H, d, J 16.3,
(1 H, d, J 15.8, NCHAHB), 3.72 (3 H, s, CO2CH3), 2.92 (1 H, d,
=
=
J 16.4, C OCHAHB), 2.62 (1 H, d, J 16.4, C OCHAHB), 1.50
(3 H, s, CCH3); dC (100 MHz, CDCl3) 190.1, 173.3, 154.7, 137.4,
129.0, 128.0, 127.4, 100.0, 65.8, 55.1, 53.0, 47.0, 22.7; m/z (ESI)
=
=
C OCHAHB), 2.61 (1 H, dd, J 16.3 and 1.1, C OCHAHB) and
1.59 (3 H, s, CH3); dC (100 MHz, CDCl3) 190.3, 173.4, 154.8, 101.0,
65.7, 53.3, 52.8, 47.2, 43.4 and 23.0; HRMS (ESI) found 232.0743,
C10H15NO3Cl (M + H+) requires 232.0740.
318 (M + MeCN + NH4 , 52%), 260 (M + H+, 100); HRMS (ESI)
found 260.1282, C15H18NO3 (M + H+) requires 260.1287.
+
(2-RS)-2-(4-Iodobutyl)-1-(4-methoxybenzyl)-5,6-
(2-RS)-1-But-3-enyl-2-methyl-5,6-dihydropyridone-2-carboxylic
dihydropyridone-2-carboxylic acid methyl ester 26
acid methyl ester 19
Lithium (30 mg, 4.3 mmol), DBB (1.07 g, 4.04 mmol) and
pyridinium salt 22 (414 mg, 1.00 mmol) were subjected to general
procedure B using 1,4-diiodobutane (0.55 mL, 4.17 mmol) as
the electrophile. The crude residue was purified via flash column
chromatography (SiO2, 500 mL 1 : 99 Et2O–petroleum ether
followed by 1 : 9 acetone–Et2O) to furnish the dihydropyridone 26
(281 mg, 62%) as a yellow oil. mmax (thin film)/cm−1: 2593, 1737,
1581, 1514, 1460, 1252, 732; dH (400 MHz, CDCl3) 7.23 (2 H, d,
J 8.8, ArH), 6.96 (1 H, d, J 7.8, NCH), 6.91 (2 H, d, J 8.8, ArH),
Sodium (457 mg, 19.9 mmol), naphthalene (3.16 g, 24.7 mmol)
and pyridinium salt 17 (1.60 g, 2.69 mmol) were subjected to
general procedure A using methyl iodide (2.5 mL, 40 mmol) as
the electrophile. The crude residue was purified by flash column
chromatography (Et3N-doped SiO2, 75 : 24.5 : 1 EtOAc–petroleum
ether–Et3N) to give the product 19 (0.359 g, 37%) as a brown
oil. mmax (thin film)/cm−1: 2954, 1735, 1641, 1585 and 1270; dH
=
(400 MHz, CDCl3) 7.03 (1 H, d, J 7.7, NCH CH), 5.80–5.70
=
=
=
(1 H, m, NCH CH), 5.14–5.12 (1 H, m, CH CHAHB), 5.09
4.97 (1 H, d, J 7.8, NCH CH), 4.48 (1 H, d, J 14.8, NCHAHB),
=
=
(1 H, s, CH CHAHB), 4.98 (1 H, dd, J 7.8 and 0.8, NCH CH),
4.36 (1 H, d, J 14.8, NCHAHB), 3.81 (3 H, s, OCH3), 3.72
3.71 (3 H, s, CO2CH3), 3.39 (1 H, qn, J 7.4, NCHAHB), 3.26 (1 H,
(3 H, s, CO2CH3), 3.14–3.06 (2 H, m, CH2I), 2.87 (1 H, d, J 16.3,
=
=
=
qn, J 7.4, NCHAHB), 2.86 (1 H, dd, J 16.5 and 0.8, C OCHAHB),
C OCHAHB), 2.70 (1 H, d, J 16.3, C OCHAHB), 2.00–1.86
(2 H, m, CH2), 1.84–1.65 (2 H, m, CH2), 1.55–1.39 (2 H, m,
CH2); dC (100 MHz, CDCl3) 190.2, 172.9, 159.5, 154.2, 129.5,
128.3, 114.4, 99.8, 68.8, 55.3, 53.7, 52.9, 43.5, 33.4, 33.1, 24.8,
5.8; HRMS (ESI) found 458.0830, C19H25NO4I (M + H+) requires
458.0828.
=
2.55 (1 H, d, J 16.5, C OCHAHB), 2.36 (2 H, q, J 7.4, NCH2CH2),
1.56 (3 H, s, CH3); dC (100 MHz, CDCl3) 190.1, 173.5, 154.6, 134.1,
118.1, 99.5, 65.7, 53.1, 50.9, 47.1, 35.4, 22.6; HRMS (ESI) found
224.1284, C12H18NO3 (M + H+) requires 224.1287.
1-Benzyl-4-methoxy-2-methoxycarbonylpyridinium iodide 21
(2-RS)-2-(4-Chlorobutyl)-1-H-5,6-dihydropyridone-2-carboxylic
acid methyl ester 30
Pyridine 12 (632 mg, 3.78 mmol) was added to benzyl iodide
(5.00 g, 22.9 mmol) at 30 ◦C under an atmosphere of argon. The
reaction was wrapped in foil and stirred for 48 h. THF (20 mL)
was added to the reaction and the precipitate was isolated by
filtration and washed with further portions of THF (2 × 10 mL).
The resulting solid was dried under high vacuum to furnish the
pyridinium salt 21 (1.04 g, 72%) as a yellow powder. Mp 105–
106 ◦C (THF); mmax (KBr)/cm−1: 3031, 1743, 1626, 1570, 1332; dH
(400 MHz, CDCl3) 9.83 (1 H, d, J 7.2, NCH), 7.90 [1 H, d, J 3.0,
Dihydropyridone 25 (26 mg, 0.07 mmol) in trifluoroacetic acid
(5 mL) was heated at reflux for 16 h. The reaction was concentrated
in vacuo andthe residue was dissolved inCH2Cl2 (5 mL). Potassium
carbonate (46.9 mg, 0.34 mmol) was added and the reaction mix-
ture stirred for 16 h before being filtered and concentrated in vacuo.
The crude residue was purified via flash column chromatography
(SiO2, 1 : 9 acetone–Et2O) to furnish the product 30 (16 mg, 91%)
as a pale yellow oil. mmax (thin film)/cm−1: 3237, 2955, 1738, 1579,
1439, 1238, 1019; dH (400 MHz, CDCl3) 7.19–7.14 (1 H, m, NCH),
=
C(OCH3)CH], 7.78 (1 H, dd, J 7.2 and 3.0, NCH CH), 7.37–
7.30 (5 H, m, Ph), 6.19 (2 H, s, NCH2), 4.25 (3 H, s, OCH3), 3.96
(3 H, s, OCH3); dC (100 MHz, CDCl3) 171.7, 159.9, 150.5, 143.8,
133.0, 129.5, 129.3, 128.5, 117.7, 114.9, 61.0, 59.8, 55.0; m/z (ESI)
258 (M+, 100%); HRMS (ESI) found 258.1132, C15H16NO3 (M+)
requires 258.1130.
=
5.59 (1 H, br. s, NH), 5.03 (1 H, dd, J 7.6 and 1.1, NCH CH),
3.79 (3 H, s, CO2CH3), 3.52 (2 H, t, J 6.4, CH2Cl), 2.88 (1 H, d,
=
=
J 16.3, C OCHAHB), 2.61 (1 H, d, J 16.3, C OCHAHB), 2.06–
1.97 (1 H, m, CHH), 1.80–1.69 (3 H, m, CHH and CH2), 1.52–
1.32 (2 H, m, CH2); dC (100 MHz, CDCl3) 190.5, 173.3, 149.0,
99.6, 63.2, 53.1, 44.3, 43.5, 35.1, 31.9, 21.0; HRMS (ESI) found
246.0890, C11H17NO3Cl (M + H+) requires 246.0897.
(2-RS)-1-Benzyl-2-methyl-5,6-dihydropyridone-2-carboxylic acid
methyl ester 23
Lithium (19 mg, 2.7 mmol), DBB (536 g, 2.02 mmol) and
pyridinium salt 21 (188 mg, 0.49 mmol) were subjected to
general procedure B using methyl iodide (0.13 mL, 2.02 mmol)
as the electrophile. The reaction was stirred for 5 min before the
addition of 1,2-dibromoethane and the crude residue was purified
via flash column chromatography (SiO2, 500 mL 1 : 99 Et2O–
petroleum ether followed by 1 : 9 acetone–CH2Cl2) to furnish the
dihydropyridone 23 (60 mg, 47%) as a yellow oil. mmax (film)/cm−1:
2994, 2952, 1739, 1641, 1578, 1451, 1372, 1271, 1212; dH (400 MHz,
(2-RS)-1-H-2-(4-iodobutyl)-5,6-dihydropyridone-2-carboxylic
acid methyl ester 31
Dihydropyridone 26 (358 mg, 0.78 mmol) was treated by the same
method as reported for dihydropyridone 25 and furnished the
product 31 (199 mg, 76%) as a yellow oil. mmax (thin film)/cm−1:
3229, 2952, 1738, 1580, 1437, 1237, 731; dH (400 MHz, CDCl3)
7.20–7.15 (1 H, m, NCH), 5.72 (1 H, br. s, NH), 5.02 (1 H, dd,
=
J 7.5 and 1.0, NCH CH), 3.78 (3 H, s, CO2CH3), 3.16 (2 H,
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1071–1084 | 1081
©

View Article Online
=
t, J 6.8, CH2I), 2.87 (1 H, d, J 16.3, C OCHAHB), 2.60 (1 H,
mixture warmed to room temperature to allow the evaporation of
the ammonia, basified (pH > 9) with saturated Na2CO3 solution
and extracted with CH2Cl2 (3 × 40 mL). The organic extracts
were combined, dried (MgSO4) and concentrated in vacuo to give
the crude product, which was purified by gradient flash column
chromatography (basic Al2O3, CH2Cl2 to 5 : 95 acetone–CH2Cl2)
to give the product 36 (70.9 mg, 69%) as an off-white solid. Mp
93–96 ◦C; mmax (thin film)/cm−1: 2980, 2953, 1738, 1644 and 1591;
=
d, J 16.3, C OCHAHB), 2.03–1.94 (1 H, m, CHH), 1.82–1.67
(3 H, m, CHH and CH2), 1.46–1.27 (2 H, m, CH2); dC (100 MHz,
CDCl3) 190.6, 173.3, 149.4, 99.4, 63.2, 53.1, 43.5, 34.7, 32.6, 24.6,
6.0; HRMS (ESI) found 338.0253, C11H17NO3I (M + H+) requires
338.0253.
(2-RS)-2-Methyl-1-(4-methoxybenzyl)-tetrahydropyridone-2-
=
dH (400 MHz, CDCl3) 6.95 (1 H, d, J 7.5, NCH CH), 4.89 (1 H,
carboxylic acid methyl ester 33
=
d, J 7.5, NCH CH), 3.74 (3 H, s, CO2CH3), 3.06 (3 H, s, NCH3),
L-Selectride (0.34 mL, 1 M in THF, 0.34 mmol) was added drop-
wise to a stirred solution of dihydropyridone 24 (50 mg, 0.17 mmol)
in THF (3 mL) at −78 ◦C. The reaction was stirred for 1 h before
the addition of MeOH (2 mL) and H2O (5 mL). The aqueous layer
was extracted with Et2O (3 × 10 mL) and the combined organic
layers where dried (MgSO4) and concentrated in vacuo. The crude
residue was purified via flash column chromatography (SiO2, 2 : 3
Et2O-petroleum ether) to yield the tetrahydropyridone 33 (34 mg,
71%) as an oil. mmax (thin film)/cm−1: 2956, 2836, 1727, 1612,
1512, 1245; dH (400 MHz, CDCl3) 7.28 (2 H, d, J 8.6, ArH), 6.88
(2 H, d, J 8.6, ArH), 4.11 (1 H, d, J 13.9, NCHAHB), 3.81 (3 H, s
OCH3), 3.74 (3 H, s, CO2CH3), 3.24 (1 H, d, J 13.9, NCHAHB),
2.92 (1 H, ddd, J 14.6, 6.4 and 3.6, NCHAHBCH2), 2.72 (1 H, dd,
=
2.77 (1 H, q, J 7.3, C OCHCH3), 1.43 (3 H, s, CH3) and 1.07
[3 H, d, J 7.3, C OCHCH3)]; dC (100 MHz, CDCl3) 194.7, 173.6,
154.6, 97.8, 68.3, 53.0, 47.0, 39.4, 17.1 and 11.4; HRMS (ESI)
found 198.1137, C10H16NO3 (M + H+) requires 198.1130.
=
(2-RS)-1-(Methoxybenzyl)-2-methyl-5-phenyl-5,6-
dihydropyridone-2-carboxylic acid methyl ester 39
Phenylboronic acid (136 mg, 1.12 mmol) and Na2CO3 (151 mg,
1.42 mmol) were added to a stirred solution of dihydropyridone 37
(230 mg, 0.56 mmol) and palladium/carbon (69 mg, 10% loading)
in 1,2-dimethoxyethane–H2O (5 mL, 1 : 1) at room temperature.
R
The reaction was stirred for 18 h and then filtered through Celiteꢀ
=
J 14.7 and 1.8, C OCHAHB), 2.58 (1 H, ddd, J 14.6, 9.9 and 4.8,
before being diluted with brine (5 mL). The resulting mixture was
extracted with Et2O (3 × 10 mL) and the combined organic layers
where dried (MgSO4) and concentrated in vacuo. The crude residue
was purified via flash column chromatography (SiO2, 7 : 3 Et2O–
petroleum ether) to yield the dihydropyridone 39 (184 mg, 90%)
as an oil. mmax (thin film)/cm−1: 2998, 2953, 1738, 1641, 1594, 1513,
1367, 1249; dH (400 MHz, CDCl3) 7.36–7.33 (2 H, m, ArH), 7.30–
7.25 (5 H, m, ArH), 7.16 (1 H, tt, J 7.3 and 1.6, NCH), 6.93 (2 H,
d, J 8.8, ArH), 4.56 (1 H, d, J 15.4, NCHAHB), 4.46 (1 H, d, J 15.4,
NCHAHB), 3.82 (3 H, s, OCH3), 3.75 (3 H, s, CO2CH3), 3.09 (1 H,
=
NCHAHBCH2), 2.43 (1 H, d, J 14.7, C OCHAHB), 2.34 (2 H,
m, NCH2CH2), 1.53 (3 H, s, CH3); dC (100 MHz, CDCl3) 207.1,
172.7, 158.8, 131.5, 129.4, 113.8, 65.0, 55.3, 53.6, 51.7, 50.8, 46.2,
40.7, 23.6; HRMS (ESI) found 314.1363, C16H21NO4Na (M +
Na+) requires 314.1363.
(2-RS)-1-H-2-Methyltetrahydropyridone-2-carboxylic acid methyl
ester 34
Palladium on carbon (5 mg, 20% loading) was added to a stirred
solution of tetrahydropyridone 33 (30 mg, 0.10 mmol) in MeOH
(5 mL) under an atmosphere of hydrogen. The reaction was
stirred for 2 h before filtration through Celiteꢀ. The mixture was
concentrated in vacuo and the crude residue was purified via flash
column chromatography (SiO2, 3 : 2 acetone–petroleum ether) to
furnish the tetrahydropyridone 34 (9 mg, 50%) as an oil. mmax (thin
film)/cm−1: 3426, 2981, 2852, 1715, 1645, 1455, 1439, 1365, 1228,
1145; dH (400 MHz, CDCl3) 3.73 (3 H, s, CO2CH3), 3.22 (1 H, ddd,
J 12.4, 6.7 and 3.4, NCHAHBCH2), 2.92 (1 H, ddd, J 12.4, 10.7 and
4.2, NCHAHBCH2), 2.81 (1 H, dd, J 14.7 and 1.9, C OCHAHB),
2.45 (1 H, ddd, 14.8, 10.7 and 6.7, NCH2CHAHB), 2.37–2.29 (2 H,
m, C OCHAHB and NCH2CHAHB), 2.01 (1 H, br. s, NH), 1.41
(3 H, s, CH3); dC (100 MHz, CDCl3) 206.9, 175.3, 62.0, 52.4, 49.7,
41.9, 40.78, 26.5; HRMS (ESI) found 194.0790, C8H13NO3Na
(M + Na+) requires 194.0788.
=
=
d, J 16.2, C OCHAHB), 2.79 (1 H, d, J 16.2, C OCHAHB), 1.59
(3 H, s, CH3); dC (100 MHz, CDCl3) 187.3, 173.3, 159.4, 153.3,
135.6, 129.0, 128.9, 128.1, 127.8, 125.9, 114.4, 112.4, 66.0, 55.3,
54.3, 53.0, 47.6, 22.4; HRMS (ESI) found 366.1698, C22H24NO4
(M + H+) requires 366.1705.
R
6-Methoxypyridine-2-carboxylic acid methyl ester 41
A suspension of 6-hydroxypicolinic acid 40 (2.53 g, 18.2 mmol),
silver carbonate (15.0 g, 54.6 mmol) and methyl iodide (9 mL) in
chloroform (100 mL) was heated at reflux for 18 h in the dark. The
reaction mixture was filtered through a silica pad and the filtrate
was concentrated in vacuo. Purification of the residue via flash
column chromatography (SiO2, 5 : 95 EtOAc–petroleum ether)
gave the product 41 (2.75 g, 91%) as colourless cubes, whose data
were consistent with those reported in the literature.29
=
=
1,2,3-Trimethyl-5,6-dihydropyridone-2-carboxylic acid methyl
ester 36
1-Methyl-6-oxo-1,6-dihydropyridine-2-carboxylic acid methyl
ester 42
Iron(III) chloride (5 mg) was added to a solution of sodium (30.0 g,
1.3 mmol) in ammonia (ca. 30 mL) and stirred until the dark
blue colour faded to leave a grey suspension. Dihydropyridone 14
(99.5 mg, 0.54 mmol) was added in THF (10 mL). The solution
was stirred at −78 ◦C for 30 min before the addition of methyl
iodide (0.16 mL, 2.6 mmol) after which the mixture was stirred for
a further 1 h. Saturated NH4Cl solution (10 mL) was added and the
A solution of methoxypyridine 41 (4.00 g, 23.9 mmol) in methyl
iodide (7.44 mL, 120 mmol) was heated in a sealed tube at 200 ◦C
for 18 h. The methyl iodide was evaporated under reduced pressure
and the residue was purified by column chromatography (SiO2, 1 :
1 EtOAc–petroleum ether) to give pyridone 42 (3.65 g, 21.8 mmol,
91%) as colourless cubes. The data were consistent with those
reported in the literature.24
1082 | Org. Biomol. Chem., 2006, 4, 1071–1084
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
6-Methoxy-2-carboxylic acid methyl ester-N-methylpyridinium
(2-RS)-1-Methyl-3,4-dihydropyridone-2-carboxylic acid methyl
trifluoromethanesulfonate 43
ester 45
A solution of pyridine 41 (1.89 g, 11.3 mmol) and methyl trifluo-
romethanesulfonate (1.47 mL, 13.0 mmol) in CH2Cl2 (50 mL) was
stirred at room temperature for 18 h. The solvent was removed
in vacuo and the resulting crude residue was purified via flash
column chromatography (SiO2, 3 : 97 MeOH–CH2Cl2) to furnish
the pyridinium salt 43 (3.73 g, 100%) as colourless needles. Mp
54–55 ◦C (CHCl3–Et2O); Anal. calc. for C10H12F3NO6S: C, 36.26;
H, 3.65; N, 4.23; S, 9.68. Found: C, 36.11; H, 3.60; N, 4.09; S,
9.60%; mmax (KBr)/cm−1: 1745, 1628, 1590, 1511, 1438, 1349, 1271
and 1031; dH (400 MHz, CDCl3) 8.52 (1 H, dd, J 8.9, 7.9 Hz,
4-H), 7.92–7.97 (m, 2 H, 3-H and 5-H), 4.46 (s, 3 H, OCH3), 4.23
(s, 3 H, NCH3) and 4.08 (s, 3 H, OCH3); dC (100 MHz, CDCl3)
162.0, 160.2, 147.6, 140.8, 121.4, 115.2, 60.5, 54.5, 38.5; m/z (ESI)
182 (M+, 100%); HRMS (CI) found 182.0819, C9H12NO3 (M+)
requires 182.0817.
Pyridone 42 (167 mg, 1.00 mmol) was reduced using the general
procedure B. After addition of the 1,2-dibromoethane, saturated
NH4Cl solution (5 mL) was added. On work-up and purification
by column chromatography (SiO2, acetone–CH2Cl2, 20 : 80)
dihydropyridone 45 (98 mg, 0.58 mmol, 58%) was obtained as
colourless needles. Spectroscopic data were consistent with those
reported earlier.
1,2,5-Trimethyl-3,4-dihydropyridone-2-carboxylic acid methyl
ester 46
Lithium (28 mg, 4.0 mmol), DBB (1.05 g, 3.95 mmol) and pyridone
42 (175 mg, 1.05 mmol) were subjected to general procedure B
using methyl iodide (0.36 mL, 5.8 mmol) as the electrophile. The
crude residue was purified by flash column chromatography (SiO2,
70 : 30 Et2O–petroleum ether to Et2O) to yield dihydropyridone 46
(121 mg, 58%, 4 : 1 dr) as a pale yellow oil. The diastereoisomers
were inseparable and the data are quoted for the mixture. mmax (thin
film)/cm−1: 2955, 1741, 1650; dH (400 MHz, CDCl3) 5.80 (1 H, m,
4-H), 5.56 (1 H, d, J 10, 3-H), 3.74 (3 H, s, OCH3), 2.93–3.03 (1 H,
m, CHCH3), 2.89 (3 H, m, NCH3), 1.62 (3 H, m, C2–CH3), 1.33
(3 H, d, J 7.4, CHCH3); dC (100 MHz, CDCl3) 172.2, 171.5, 130.2 +
130.0, 125.0 + 124.9, 66.5 + 66.3, 53.0 + 53.0, 35.7 + 35.2, 30.6 +
30.5, 24.3 + 24.0, 19.4 + 19.1; MS (CI) 198 (M + H+, 100%),
138 (M − CO2Me, 38%); HRMS (CI) found 198.1130, C10H16NO3
(M + H+) requires 198.1123.
(2-RS)-1,2-Dimethyl-3,4-dihydropyridone-2-carboxylic acid
methyl ester 44
Sodium (173 mg, 7.5 mmol), naphthalene (960 mg, 7.50 mmol)
and pyridinium salt 43 (490 mg, 1.48 mmol) were subjected to
general procedure A using methyl iodide (0.62 mL, 7.5 mmol) as
the electrophile. Removal of the aqueous solvent in vacuo gave a
pale yellow crude residue which was added to an acetic anhydride–
pyridine solution (1 : 1, 1.5 mL). The reaction was stirred for 20 h
and concentrated in vacuo. NaOH (1 M, 10 mL) was added and
the reaction mixture extracted with CH2Cl2 (6 × 20 mL). The
combined organic extracts were dried (MgSO4), concentrated in
vacuo and purified via flash column chromatography (SiO2, 5 : 95
acetone–CH2Cl2) to furnish dihydropyridone 44 (128 mg, 47%)
as a yellow solid. Mp 93–96 ◦C (acetone–petroleum ether); mmax
(KBr)/cm−1: 1738 and 1651; dH (400 MHz, CDCl3) 5.86 (1 H,
References
1 P. D. Bailey, P. A. Millwood and P. D. Smith, Chem. Commun., 1998,
633; P. A. Weintraub, J. S. Sabol, J. M. Kane and D. R. Borcherding,
Tetrahedron, 2003, 59, 2953.
2 For examples, see: M.-L. Bennasar, J.-M. Jime´nez, B. Vidal, B. A. Sufi
and J. Bosch, J. Org. Chem., 1999, 64, 9605; J. Pabel, G. Ho¨fner and
K. T. Wanner, Bioorg. Med. Chem. Lett., 2000, 10, 1377; A. P. Krapcho
and D. J. Waterhouse, Heterocycles, 1999, 51, 737; G. R. Heintzelman,
W.-K. Fang, S. P. Keen, G. A. Wallace and S. M. Weinreb, J. Am. Chem.
Soc., 2001, 123, 8851.
3 (a) Y. Kita, M. Maekawa, Y. Yamaski and I. Nishiguchi, Tetrahedron,
2001, 57, 2095; (b) J. MacTavish, G. R. Procter and J. Redpath, J. Chem.
Soc., Perkin Trans. 1, 1996, 2545.
=
ddd, J 10.1, 3.6 and 3.5, CH2CH CH), 5.61 (ddd, 1 H, J 10.1, 2.2
=
and 2.0, CH2CH CH), 3.72 (s, 3 H, CO2CH3), 3.00–3.04 (m, 2 H,
CH2), 2.92 (s, 3 H, NCH3) and 1.62 (s, 3 H, CH3); dC (100 MHz,
CDCl3) 172.0, 167.7, 126.3, 123.9, 66.4, 53.0, 31.4, 30.3 and 23.9;
m/z (CI) 184 (M + H+, 100%), 124 (M − CO2Me, 65%); HRMS
(CI) found 184.0975, C9H14NO3 (M + H+) requires 184.0974.
4 R. Francis, W. Davis and J. Wiesner, J. Org. Chem., 1974, 39, 59.
5 E. Piers and M. Soucy, Can. J. Chem., 1974, 52, 3563.
6 Y. Ge´nisson, C. Marazano and C. D. Bhupesh, J. Org. Chem., 1993, 53,
2052.
(2-RS)-1-Methyl-3,4-dihydropyridone-2-carboxylic acid methyl
ester 45
7 G. Zhao, U. C. Deo, U. C and B. Ganem, Org. Lett., 2001, 3, 201.
8 B. D. Shaw, J. Chem. Soc., Trans., 1925, 27, 215; B. D. Shaw, J. Chem.
Soc., 1937, 300; A. J. Birch, J. Chem. Soc., 1947, 1270; S. J. Danishefsky
and R. Cavanaugh, J. Am. Chem. Soc., 1968, 90, 520; S. J. Danishefsky,
P. Cain and A. Nagel, J. Am. Chem. Soc., 1975, 97, 380; J. P. Kutney, P.
Grice, K. Piotrowska, S. J. Rettig, J. Szykula, J. Trotter and L. Van Chu,
Helv. Chim. Acta, 1983, 66, 1820; A. J. Birch and E. A. Karakhamov,
J. Chem. Soc., Chem. Commun., 1975, 480.
Sodium (115 mg, 5.0 mmol), naphthalene (641 mg, 5.0 mmol) and
pyridinium salt 43 (336 mg, 1.02 mmol) were subjected to general
procedure A using methanol (2 mL) as the electrophile. The crude
residue was purified via flash column chromatography (SiO2, 5 :
95 acetone–CH2Cl2) to furnish dihydropyridone 45 (34 mg, 20%)
as a pale yellow oil. Anal. calc. for C8H11NO3: C, 54.60; H, 6.45;
N, 7.82. Found: C, 54.49; H, 6.43; N, 7.81%; mmax (thin film)/cm−1:
2955, 1746 and 1652; dH (400 MHz, CDCl3) 5.94 (1 H, dddd, J 9.9,
9 A. J. de Konig, P. H. M. Budzelaar, L. Brandsma, M. J. A. de Bie and
J. Boersma, Tetrahedron Lett., 1980, 21, 2105.
10 (a) T. J. Donohoe, A. J. McRiner and P. Sheldrake, Org. Lett., 2000, 2,
3861; (b) T. J. Donohoe, A. J. McRiner, M. Helliwell and P. Sheldrake,
J. Chem. Soc., Perkin Trans. 1, 2001, 1435; (c) T. J. Donohoe, L. H.
Mace, M. Helliwell and O. Ichihara, Synlett, 2002, 331.
11 T. J. Donohoe, D. J. Johnson, L. H. Mace, O. Ichihara and M. J.
Bamford, Org. Lett., 2005, 7, 453.
=
4.1, 3.0 and 1.5, CH2CH CH), 5.82 (1 H, dddd, J 9.9, 4.8, 2.3
=
and 1.5, CH2CH CH), 4.58 (1 H, dddd, J 4.8, 3.5, 3.5 and 1.5,
2-H), 3.78 (3 H, s, CO2CH3), 3.01–3.05 (2 H, m, CH2) and 2.98
(3 H, s, CH3); dC (100 MHz, CDCl3) 169.7, 167.7, 126.4, 119.6,
63.8, 52.8, 33.7, 32.0; m/z (CI) 170 (M + H+, 100%); HRMS (CI)
found 170.0822, C8H12NO3 (M + H+) required 170.0817.
12 F.-A. Kunng, J.-M. Gu, S. Chao, Y. Chen and P. S. Mariano, J. Org.
Chem., 1983, 48, 4262.
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1071–1084 | 1083
©

View Article Online
13 (a) D. L. Comins and J. D. Brown, Tetrahedron Lett., 1986, 27, 4549;
for recent refs., see: (b) D. L. Comins, M. O. Killpack, E. Despagnet and
E. Zeller, Heterocycles, 2002, 58, 505 (c) J. T. Kuethe and D. L. Comins,
J. Org. Chem., 2004, 69, 5219; (d) J. T. Kuethe and D. L. Comins, J. Org.
Chem., 2004, 69, 2863.
14 R. J. Sundberg and S. Jiang, Org. Prep. Proced. Int., 1997, 29, 117; see
also: M. Y. Essawi and P. S. Portoghese, J. Heterocycl. Chem., 1983, 20,
477.
19 J. D. Brown, M. A. Foley and D. L. Comins, J. Am. Chem. Soc., 1988,
110, 7445; see also: J. F. Liu and C. H. Heathcock, J. Org. Chem., 1999,
64, 8263; M. Weymann, M. Schultz-Kukula and H. Kunz, Tetrahedron
Lett., 1998, 39, 7835.
20 D. L. Comins and A. L. Williams, Tetrahedron Lett., 2000, 41, 2839.
21 For reviews, see: D. L. Comins, J. Heterocycl. Chem., 1999, 36, 1491;
S. P. Joseph and D. L. Comins, Curr. Opin. Drug Discovery Dev., 2002,
5, 870.
15 T. J. Donohoe and D. House, J. Org. Chem., 2002, 67, 5015.
16 J. R. Ferguson, K. W. Lumbard, F. Scheinmann, A. V. Stachul-
ski, P. Stjernlo¨f and S. Sundell, Tetrahedron Lett., 1995, 36, 8867;
N. G. Kundu, R. P. Hertzberg and S. J. Hannon, Tetrahedron Lett.,
1980, 21, 1109; D. G. Melillo, R. D. Larson, D. J. Mathre, W. F.
Shukis, A. W. Wood and J. R. Colleluor, J. Org. Chem., 1987,
52, 5143; A. Cave´, C. Kan-Fan, P. Potier, J. Le Men and M.-M.
Janot, Tetrahedron, 1967, 23, 4691; K. C. Nicolaou, T. Montagnon,
P. S. Baran and Y.-L. Zhong, J. Am. Chem. Soc., 2002, 124,
2245.
22 For examples, see: A. I. Meyers and T. R. Elworthy, J. Org. Chem.,
1992, 57, 4732; D. L. Comins, J. T. Kuethe, T. M. Miller, F. C. Fevrier
and C. A. Brooks, J. Org. Chem., 2005, 70, 5221.
23 F.-X. Felpin, J. Org. Chem., 2005, 70, 8575.
24 M. Adamczyk, S. R. Akireddy and R. E. Reddy, Tetrahedron:
Asymmetry, 2001, 12, 2385.
25 T. Kametani, Y. Kigawa, T. Takahashi, H. Nemoto and K. Fukumoto,
Chem. Pharm. Bull., 1976, 24, 1870.
26 P. Meghani and J. A. Joule, J. Chem. Soc., Perkin Trans. 1, 1988, 1.
27 C. E. Banks, R. G. Evans, J. S. Rodrigues, P. G. Turner, T. J. Donohoe
and R. G. Compton, Tetrahedron, 2005, 61, 2365.
17 T. J. Donohoe, R. G. Compton, D. J. Johnson and J. D. Wadhawan,
Tetrahedron, 2004, 60, 5945.
28 N. L. Holy, Chem. Rev., 1974, 74, 243.
18 T. J. Donohoe, D. House and K. W. Ace, Org. Biomol. Chem., 2003, 1,
3749.
29 D. Campbell, E. Chan, S. Y. Chooi, L. W. Deady and R. A. Shanks,
Aust. J. Chem., 1971, 24, 377.
1084 | Org. Biomol. Chem., 2006, 4, 1071–1084
This journal is
The Royal Society of Chemistry 2006
©