Intramolecular interactions in carbenes derived during the pyrolysis of N-alkenylisoxazolones

Article abstract of DOI:10.1071/CH03154

Flash vacuum pyrolysis of ethyl 3-(3-methyl-5-oxo-2,5-dihydroisoxazole-2- yl)-3-phenylpropenoate yields 4-ethoxy-3-hydroxy-2-methyl-5-phenylpyridine in addition to the expected pyrrole. The structure assignment is based on two-dimensional NMR and computat

Full text of DOI:10.1071/CH03154

Full Paper
CSIRO PUBLISHING
www.publish.csiro.au/journals/ajc
Aust. J. Chem. 2004, 57, 455–460
Intramolecular Interactions in Carbenes Derived During the Pyrolysis
of N-Alkenylisoxazolones
Matthew Cox,A Michael Dixon,A Troy Lister,A and Rolf H. PragerA,B
A School of Chemistry, Physics and Earth Sciences, Flinders University, Adelaide SA 5001, Australia.
B Author to whom correspondence should be addressed (e-mail: rolf.prager@flinders.edu.au).
Flash vacuum pyrolysis of ethyl 3-(3-methyl-5-oxo-2,5-dihydroisoxazole-2-yl)-3-phenylpropenoate yields 4-
ethoxy-3-hydroxy-2-methyl-5-phenylpyridine in addition to the expected pyrrole.The structure assignment is based
on two-dimensional NMR and computational evidence. The pyrolysis of the corresponding dimethyl amide gives
a mixture of three 3-hydroxypyridines, in addition to the pyrrole. Evidence is presented for the formation of
intermediate1,4-oxazepinesbyasix-electronelectrocyclicreactionofacarbenewithesteroramidecarbonylgroups.
Manuscript received: 17 June 2003.
Final version: 13 October 2003.
Introduction
[1,4]-migration to form a ketenimine, followed by a [1,3]-
alkoxide migration forming a ketene which underwent
cyclization to the quinolone product (Scheme 3).[1]
In a previous paper[2] we reported that flash vacuum pyro-
lysis (FVP) of the alkenylisoxazolone 1 gave the expected
pyrrole 2 from the intermediate carbene 3 (Scheme 4) as only
the minor product, the major product being an unidentified
isomer. The present paper describes an investigation of this
isomer and related reaction products, which concludes it to
have the structure 4.
While the pyrolysis or photolysis of N-(1-alkenyl)isox-
azolones (Scheme 1) generally proceeds through a carbene
intermediate and gives pyrroles,[1,2] several alternate path-
ways were observed involving solvent capture by the triplet
state of the isoxazolone (Scheme 2),[3] rearrangement of the
initially formed carbene to a more stable one and subsequent
cyclization,[1] or more complicated pathways involving a
R1
R1
R1
Discussion
N
O
FVP of the isoxazolone 1 at 450◦C gave two products, sep-
arable by chromatography. The first was ethyl 5-methyl-
2-phenylpyrrole-3-carboxylate 2 (19%) and the second
(38%) was an isomer, to which we ascribe the structure 4.The
product 4 was weakly acidic but no longer contained an ester
group (IR data); 1H and 13C NMR spectra suggested the pres-
ence of an ethyl ether.The singlet at 7.04 ppm in the 1H NMR
spectrum, and the other substituents present, require the pres-
ence of a pyridine ring.The 13C chemical shift of the pyridine
carbon atom at 103.6 ppm is consistent only with a pyridine
unsubstituted at C-5. Plausible pathways for three possi-
ble pyridine structures, 4–6, can be envisaged (Scheme 5).
The assignment of structure for the pyridine 4 rests on
the combination of theoretical and two-dimensional NMR
evidence. The pyrolysis product 4, because of its low solu-
bility, was methylated with MeI/K2CO3, to give a single
methylation product, for which structures 7–11 could be
envisaged, based on the above. Unfortunately, both 4 and 7
were obtained as microcrystalline powders, unsuitable for
X-ray determination. The methylation product appeared to
be an O-methyl derivative, based on the 13C resonance at
60.2 ppm; N-methylpyridones would be expected to have
R2
R3
HN
N
R3
ϪCO2
OC
R3
R2
R2
Scheme 1.
CO2Et
N
CO2Et 75%
O
O
CO2Et
N
O
acetone
Me
Me
Me
hn
300nm
O
CO2Et
CO2Et
N
O
Me
acetonitrile
CO2Et
N
26%
O
O
Me
Me
Scheme 2.
© CSIRO 2004
10.1071/CH03154
0004-9425/04/050455

456
M. Cox et al.
CO2Et
H
H
H
OEt
N
OEt
N
C
N
H
N
450ºC
[1,4]-shift
H
[1,3]-shift
OEt
N
C
CO2Et
H
O
ϪCO2
H
C
O
CO2Et
O
O
Scheme 3.
CO2Et
Me
CO2Et
CO2Et
Ph
O
Ph
N
Ph
OEt
Ph
N
ϩ
N
HN
OH
O
Me
Me
Me
1
3
2
4
Scheme 4.
OEt
Ph
OEt
OH
Ph
Ph
Ph
Ph
OEt
O
O
N
N
N
Me
Me
Me
4
Ph
Ph
Ph
Ph
O
OEt
OEt
C
CO2Et
Me
O
[1,2]-shift
Me
[1,5]-shift
OEt
N
O
HN
N
CH
N
N
O
O
Me
Me
C
O
CHMe
OEt
OEt
Me
1
5
Ph
O
C
O
HN
CH
N
OEt
OEt
Me
Me
6
Scheme 5.
resonances around 40 ppm. Since 4-pyridones undergo pre-
dominantly N-methylation,[4–6] structures 5 and 6, leading
to 8–11, appeared unlikely. Nonetheless, we have calculated
(Gaussian 98, restricted Hartree–Fock) the 1H and 13C NMR
spectra of the four methylated compounds 8–11: those for
7 should be very similar to those of 8 (Tables 1 and 2).
The N-methylated 4-pyridone structures 10 and 11 are
clearlyincompatiblewiththemethylationproduct, inaddition
to the 13C N-Me resonances around 40 ppm, the pyri-
done carbonyl group resonates around 180 ppm, whereas the
pyrolysis product and its methylated derivative have their
highest chemical shift around 157.9 ppm. This also rules out
structures 8 and 9 for the methylated product; their precur-
sors 5 and 6 also require pyridone carbonyl resonances. Only
Two-dimensional NMR data measured at 600 MHz
allowed the confirmation of structure 7 for the methylation
product, and hence 4 for the pyrolysis product. The pulse
sequences gHMBC and gHMQC (Varian) allowed assign-
ment of all carbon resonances and the connectivity of the
aromatic C-methyl with C-2 and C-3, the O-methyl group
with C-3, the O-ethyl group with C-4, and the positioning of
H-5 relative to C-4, C-3, C-6, and C-1ꢀ. NOESY and ROESY
spectrashowedthenuclearOverhauserinteractionofthepyri-
dine H-5 with both the O CH2 CH3 and H-2ꢀ atoms of the
phenyl group, unequivocally placing these groups.
We have used MacSpartan semi-empirical calculations to
seek to understand the reason for the differences in pyro-
lysis outcome for the cinnamate isoxazolone 1 compared
to the crotonate isoxazolone 12.[2] We suggest that while
12 is more stable in the (E)-configuration on electronic
grounds, 1 will prefer the (Z)-configuration for steric rea-
sons. It could be anticipated that the pyrolysis of (Z)- and
(E)-isoxazolones 1 would lead ultimately to the carbenes
(Z)-3 and (E)-3∗ respectively.[7] Hickson and McNab[8] have
shown that significant (E)- to (Z)-isomerization of simple
structure 7 remains consistent with the one-dimensional 13
C
NMR spectrum. While the calculated chemical shifts, using
the less accurate Hartree–Fock theory rather than the density
functional theory (DFT), do not allow for the effect of solvent
or possible intermolecular stacking, the calculated values for
8 are all uniquely in the same order as those observed for 7
and eliminate structures 9–11.
The terms (Z) and (E) here refer to the geometry of the double bond of the conjugated ester.

Intramolecular Interactions in Carbenes
457
Table 1. Comparison of calculated 1H chemical shifts for
pyridines 8–11 with those observed for 7
8
9
10
11
Methylated
product 7
H-5
7.84
9.31
7.95
7.88
7.89
8.23
4.15
4.00
1.70
2.80
6.91
9.16
7.95
7.90
7.87
8.16
4.04
4.45
1.68
2.42
6.59
8.51
8.51
8.51
8.45
8.32
3.60
4.46
2.11
2.98
6.53
8.55
8.49
8.49
8.44
8.36
3.58
4.44
2.24
2.63
7.00
7.80
7.37
7.30
7.37
7.80
3.78
4.11
1.43
2.47
H-2ꢀ
H-3ꢀ
H-4ꢀ
H-5ꢀ
H-6ꢀ
OCH3 (NCH3)
OCH2CH3
OCH2CH3
ArCH3
1.4 Å
Table 2. Comparison of calculated 13C chemical shifts for
pyridines 8–11 with those observed for 7
Fig. 1. Calculated low-energy conformation (Z)-3a, corresponding
to 13.
8
9
10
11
Methylated
product 7
C-2
C-3
C-4
C-5
161.2
145.8
163.4
116.4
161.3
141.3
130.5
131.6
132.6
130.6
129.9
55.8
169.1
107.4
168.8
100.5
161.2
141.8
133.6
131.6
132.7
130.6
129.9
50.6
145.8
144.1
177.7
119.2
159.5
141.8
134.6
133.6
134.0
132.4
133.6
40.4
159.5
113.2
183.1
118.0
158.3
141.5
134.6
133.7
134.1
132.3
133.4
39.1
152.4
142.6
157.9
103.7
153.5
139.9
126.9
128.6
128.4
128.6
126.9
60.2
unable to detect any evidence of alternative reactions of
carbene 3, such as insertion into the benzene ring, which
would lead to isoquinoline derivatives.
C-6
C-1ꢀ
In order to probe the generality of this pyrolytic rearrange-
ment leading to pyridines, the ester group in 1 was changed
to the dimethyl amide. We were aware of the observations
of Wentrup et al.,[11] that group migrations to carbenes or
electron-deficient centres were dictated by electron density,
and hence we reasoned that amide analogues of 1 might
undergo additional pyrolysis pathways, perhaps akin to those
in Scheme 6. 3-Methylisoxazolone could not be induced to
react with the piperidide or other amides of phenylpropiolic
acid under the usual conditions.Accordingly, the isoxazolone
1 was hydrolyzed to the corresponding acid, which was con-
verted to the amide 14 with dimethylamine in the presence
of 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide. Pyro-
lysis of 14 at 450◦C gave a mixture of products, of which
the four major components were separated by radial chro-
matography. The minor product (5%) was the pyrrole 15,
easily recognizable as it was the only product with nonequi-
valent N-methyl groups in the 1H NMR spectrum, typical of
amides.The most polar was the 3-hydroxypyridine 16 (18%),
whose spectral data were as expected from those of 4, its ana-
logue.Again, the infrared spectrum showed the absence of an
amide group. Two other products were isolated in significant
yields, and mass spectral evidence suggested that both were
also 3-hydroxypyridines. Structures 17 (33%) and 18 (16%)
were assigned to these compounds. Both retained the oxygen
atom of the dimethylamide but not the amino group: in 18 a
methyl group (δC 18.4 or 16.2 ppm) has been transferred from
nitrogen to the pyridine ring. We tentatively suggest these
compounds arise by the pathways shown in Scheme 7. While
[1,2]-alkyl shifts from carbon to carbon are well known in
the gas phase, resulting from cationic[12] or radical[13] inter-
mediates, migrations from nitrogen to carbon appear to be
unknown. Alternative mechanisms to those above, involving
radical opening of the epoxide, can be drawn.
C-2ꢀ
C-3ꢀ
C-4ꢀ
C-5ꢀ
C-6ꢀ
OCH3 (NCH3)
OCH2CH3
OCH2CH3
ArCH3
62.3
16.5
21.2
58.0
15.5
9.9
63.2
23.4
20.3
66.8
22.7
17.0
64.0
14.6
19.2
alkenes, including those with conjugating phenyl and nitrile
groups, occurs above 650◦C under FVP conditions, but we
feel such isomerization would be minimal at the low tem-
peratures (450◦C) used herein. AM1 calculations suggest
that carbene (E)-3 will collapse spontaneously to the cyclic
precursor of pyrrole 2, but the carbene (Z)-3 has its lowest
energy state in a conformation which allows neither the cin-
namate double bond nor the ester carbonyl group to interact
with the carbenoid centre. However, simple rotations about
the 3,4 and 5,6 single bonds leads to an apparent conforma-
tion of (Z)-3a (see Fig. 1), which is only 48 kJ mol−1 higher
in energy, and has the carbonyl oxygen atom lying 1.40 Å
above the carbene centre, the angle C-1 O C-6 being 114◦,
which indeed corresponds to the oxazepine 13, a precursor
known to readily form 3-hydroxypyridines under mild ther-
mal conditions,[9,10] as shown in Scheme 6. Whereas reaction
of 3-methylisoxazol-5(4H)-one with ethyl butynoate gives
a single product 12, which gives the corresponding pyrrole
in 93% yield on pyrolysis,[1] reaction with ethyl phenylpro-
piolate gives a mixture of the esters (E)-1 and (Z)-1 (ratio
1 : 3.5), which cannot be separated by chromatography and
are possibly the precursors of the pyrrole and pyridine prod-
ucts respectively. It is interesting to note that we have been

458
M. Cox et al.
CO2Et
Me
CO2Et
OEt
Ph
Ph
5
Ph
R
OEt
6
3
2
N
N
O
O
N
N
O
1
Me
Me
Me
O
1 Rϭ Ph
(E)-3
(Z )-3
13
12 Rϭ Me
Scheme 6.
Me
H
Ph
Ph
Ph
Ph
Me
OH
NMe2
NMe2
N
[1,2]-shift
N -Me
ϪCH2ϭ NH
14
O
H
N
O
N
N
N
OH
Me
Me
Me
Me
18
H
[1,2]-shift
Me
N
H
Ph
NMe2
OH
Ph
Ph
ϪCH2ϭ NMe
N
N
N
H
O
OH
Me
Me
Me
16
17
Scheme 7.
O
O
O
withinbondingdistanceofthecarbenoidcentre.Theresulting
oxazepines rearrange to the observed 3-hydroxypyridines.
RЈ
R
Ph
Me
Ph
Me
O
N
H
OH
Me
Ph
O
Experimental
O
18
19
20 Rϭ RЈ ϭ H
21 Rϭ RЈ ϭ Br
All solvents used were freshly distilled and dried according to the
methods of Perrin et al.[20] Melting points were determined on a Reichert
hot stage microscope and are uncorrected. 1H (300 MHz) and 13C
(75.5 MHz) NMR measurements were recorded on a GeminiVarian 300
spectrometer in deuteriochloroform with tetramethylsilane as internal
standards, unless otherwise stated. 2D NMR spectra were determined
on a Varian Unity 600 MHz spectrometer, using the pulse sequences
gHMBC, gHMQC, gCOSY, ROESY, andTOCSY. Infrared spectra were
recorded on a Perkin–Elmer 1600 FT-infrared spectrophotometer, using
fused sodium chloride cells, measured as Nujol mulls or films. Mass
spectra and high-resolution mass spectra were recorded on a Kratos
MS25RF spectrometer. Microanalyses were performed by Chemical &
Micro Analytical Services, Melbourne. Flash vacuum pyrolyses were
carried out by slowly subliming the substrate through a silica tube
(400 mm × 25 mm), packed with silica chips, and heated to the quoted
temperature under reduced pressure. The products were collected in a
liquid nitrogen cold trap. The NMR calculations were carried out on a
Sun 64-SVR4-Unix system, using Gaussian 98, Rev. A.11.3.
Scheme 8.
Unfortunately we are unable to complete this study by
the unambiguous synthesis of structure 4, but wish to report
some preliminary investigations. Reaction of benzaldehyde
with the dianion of acetylacetone gave the hydroxyketone
18,[14–16] Scheme 8, which could be oxidized to the tri-
ketone 19 with MnO2, but reaction with ammonia failed to
give satisfactory yields of the pyridone 20, which we hoped
to convert to both 7 and 8. However, the procedure of
Koshima and coworkers,[17] in which acetylacetone dianion
is treated with benzonitrile, proceeded to 20 directly. We
hoped to achieve monobromination of this compound at C-3,
but several mild reagents, including KBr3,[18] gave only the
dibromide 21, as might have been expected from literature
precedent.[19]
FVP of Ethyl 3-(3-methyl-5-oxo-2,5-dihydroisoxazol-2-yl)-
3-phenylpropenoate 1
Pyrolysis (450◦C, 150◦C, 0.01 mmHg) of 1 (65 mg, 0.23 mmol) gave a
pale brown oil (50 mg), which was purified by radial chromatography
(10% ethyl acetate/light petroleum) to give two fractions.
Conclusions
The first fraction was ethyl 5-methyl-2-phenylpyrrole-3-carboxylate
2, which was obtained as a colourless oil (10 mg, 19%). (Found: M+
229.1102. C14H15NO2 requires M+ 229.1103). vmax/cm−1 1672. δH
8.13 (1H, bs, NH), 7.55–7.59 (2H, m), 7.29–7.41 (3H, m), 6.40 (1H, dq,
J 0.82, 3), 4.19 (2H, q, J 7.0), 2.28 (3H, d, J 0.82), 1.24 (3H, t, J 7.0).
δC 165.0, 135.9, 132.3, 128.8, 128.0, 127.9, 127.6, 112.2, 109.6, 59.5,
14.3, 12.7. m/z 229 (M+, 90%), 201 (23), 184 (100), 170 (3), 156 (13),
141 (3), 128 (10), 104 (7), 77 (10).
Flash vacuum pyrolysis of derivatives of N-carboxyethenyl
isoxazolones leads to carbenes which undergo cyclization
forming either pyrroles or pyridines, depending on the con-
figuration of the alkene. Carbenes from (E)-alkenes give
mainly pyrrole derivatives, while those from (Z)-alkenes lead
to conformations with the ester or amide carbonyl groups

Intramolecular Interactions in Carbenes
459
The second fraction was 4-ethoxy-3-hydroxy-2-methyl-6-phenyl-
pyridine 4 (20 mg, 38%), colourless microcrystals, mp 130–132◦C.
(Found: M+ 229.1092. C14H15NO2 requires M+ 229.1103). vmax/cm−1
3110, 1473, 1078. δH 7.84–7.87 (2H, m), 7.29–7.44 (3H, m), 7.04 (1H,
s), 4.21 (2H, q, J 7.0), 2.56 (3H, s), 1.48 (3H, t, J 7.0), (OH unsighted).
δC 152.1, 149.6, 144.4, 139.6, 139.3, 128.6, 128.2, 126.8, 102.5, 64.7,
18.2, 14.6. m/z 229 (M+, 100%), 214 (6), 201 (56), 185 (17), 172 (29),
156 (5), 131 (9), 115 (6), 102 (10), 77 (9).
The second fraction, tentatively assigned as 3-hydroxy-2,3-dimethyl-
6-phenylpyridine 18 (15 mg, 16%), was obtained as a pale yellow oil.
(Found: [M + H]+ 200.1084. C13H14NO requires [M + H]+ 200.1075).
vmax/cm−1 1605, 1536. δH 7.84–7.87 (2H, m), 7.34–7.44 (5H, m, ArH
and H-5), 4.55 (1H, bs, OH), 2.59 (3H, s), 2.33 (3H, s). δC 149.1, 148.0,
144.1, 138.0, 134.9, 128.6, 128.3, 126.8, 121.7, 18.3, 16.1. m/z (GC-
MS) 200 ([M + H]+, 100%), 170 (7), 129 (4).
The third fraction was N,N-dimethyl-5-methyl-2-phenylpyrrole-
3-carboxamide 15 (5 mg, 5%), which was obtained as a brown oil.
(Found: M+ 228.1254. C14H16N2O requires M+ 228.1263). vmax/cm−1
1656, 1599, 1449. δH 8.33 (1H, bs, NH), 7.28–7.40 (5H, m), 5.99 (1H,
dq, J 0.82, 2.75), 3.02 (3H, bs), 2.72 (3H, bs), 2.28 (3H, d, J 0.82).
δC 169.4, 132.4, 129.0, 128.8, 128.4, 126.6, 125.5, 111.8, 108.1, 38.6,
34.9, 12.8. m/z 228 (M+, 38%), 184 (100), 156 (9), 128 (10), 105 (22),
72 (21), 43 (6).
Methylation of 4
Methyl iodide (84 mg, 0.6 mmol) was added to a solution containing 4
(68 mg, 0.3 mmol) and potassium carbonate (41 mg, 0.3 mmol) in
acetone (5 mL), and the mixture was stirred at room temperature for
3d. The solvent was evaporated, and the residue extracted with ethyl
acetate (20 mL). The product was purified by radial chromatography
(10% EtOAc/light petroleum), to give 4-ethoxy-3-methoxy-2-methyl-
6-phenylpyridine 7 (53 mg, 74%) as a white powder, mp 39◦C. (Found:
M+ 243.1270. C15H17NO2 requires M+ 243.1259). vmax/cm−1 1578,
1469, 1343, 1245, 1206, 1113, 1076. δH (600 MHz) 7.81 (2H, dd, J 5.9,
3.0), 7.36 (2H, dt, J 5.9, 3), 7.27 (1H, dt, J 6, 3), 7.00 (1H, s), 4.11 (2H,
q, J 7.1), 3.78 (3H, s), 2.47 (3H, s), 1.43 (3H, t, J 7.1). δC 157.9, 153.5,
152.4, 142.6, 139.9, 128.6, 128.4, 126.9, 103.7, 64.0, 60.2, 19.2, 14.6.
The fourth fraction was 3-hydroxy-4-dimethylamino-2-methyl-
6-phenylpyridine 16 (16 mg, 18%), obtained as a brown oil. (Found:
M+ 228.1262. C14H16N2O requires M+ 228.1263). vmax/cm−1 1478,
1092. δH 7.81–7.87 (2H, m), 7.29–7.46 (4H, m, ArH and OH), 7.11
(1H, s), 2.84 (6H, s), 2.56 (3H, s). δC 148.8, 148.6, 144.9, 143.5, 138.9,
128.8, 128.5, 126.9, 109.3, 43.6, 18.6. m/z 228 (M+, 100%), 213 (77),
199 (15), 185 (33), 170 (6), 144 (10), 128 (13), 115 (8), 102 (13), 83
(10), 59 (6), 43 (5).
N,N-Dimethyl-3-(3-methyl-5-oxo-2,5-dihydroisoxazol-2-yl)-
3-phenylpropenamide 14
Bromination of 2-Methyl-6-phenylpyridin-4(1H)-one 20
A solution of bromine (52 mg, 0.29 mmol) in 1 M aqueous KBr (2 mL)
was added over 5 min to a stirred suspension of 2-methyl-6-phenyl-
4-pyridone 20[17] (50 mg, 0.29 mmol) in 1 M KBr (2 mL). After 24 h
the solid precipitate was collected and recrystallized from acetonitrile
to give 3,5-dibromo-2-methyl-6-phenylpyridin-4(1H)-one 21 as small
white crystals (70 mg), mp 290–292◦C. (Found: C 42.4, H 2.7, N 4.3%.
C12H9Br2NO requires C 42.0, H 2.6, N 4.1%). δH (CDCl3/TFA) 7.64–
7.74 (5H, m), 2.88 (3H, s). δC (CDCl3/TFA) 166.0, 154.5. 154.0, 132.5,
130.0, 129.8, 128.9, 110.0, 109.5, 20.5.
Bromination as above under more dilute conditions, or with NBS in
chloroform, or with bromine in various solvents, all gave predominantly
the 3,5-dibromo compound above, with small quantities of the desired
mono brominated species, as determined by the pyridine H-3 and H-5
resonances at 7.1 and 6.8 ppm.
Isoxazolone 1 (0.825 g, 3 mmol) was stirred with sodium hydroxide
(0.121 g, 3 mmol) in ethanol (9.2 mL) and water (9.2 mL) for 2 h at
0◦C. The reaction mixture was diluted with water (50 mL) and extracted
with ethyl acetate (50 mL). The aqueous phase was acidified with conc.
HCl to pH 3, and extracted with ethyl acetate (2 × 30 mL) to give
the carboxylic acid as a dark orange oil (0.609 g, 82%). (Found: M+
245.0686. C13H11NO4 requires M+ 245.0688). vmax/cm−1 1777, 1719,
1628, 1577, 1449, 1278. δH 8.84 (1H, bs), 7.38–7.52 (5H, m), 6.26 (1H,
s), 5.17 (1H, q, J 0.82), 2.03 (3H, d, J 0.82). δC 170.4, 167.5, 161.5,
145.7, 133.9, 132.1, 129.5, 127.8, 115.1, 89.3, 13.1. m/z 245 (M+, 42%),
227 (17), 201 (33), 184 (19), 172 (30), 147 (100), 115 (28), 103 (62),
77 (68), 69 (76), 51 (20), 44 (40).
The carboxylic acid above (0.609 g, 2.5 mmol), dimethylamine
hydrochloride (0.223 g, 2.73 mmol), pyridine (0.216 g, 2.73 mmol),
and 1-(3-(dimethylamino)propyl)-3-ethylcarbodiimide hydrochloride
(0.768 g, 3.73 mmol) were stirred in dichloromethane (30 mL) and DMF
(3 mL) for 3 h at 0◦C and then at room temperature for 16 h. The solvent
was evaporated, and the residue was redissolved in ethyl acetate (50 mL)
and then washed with 1 M HCl (2 × 20 mL) and water (20 mL). The
organic layer was dried (Na2SO4) and evaporated to give an orange oil
(0.493 g), whichwaspurifiedbyradialchromatography(ethylacetate)to
give the title compound as an orange/brown glass (0.15 g, 23%). (Found:
Acknowledgments
The authors are grateful to the Australian Research Grants
Committee for support of this work.
References
M
+ 272.1164. C15H16N2O3 requires M+ 272.1161). vmax/cm−1 1760,
[1] M. Cox, F. Heidarizadeh, R. H. Prager, Aust. J. Chem. 2000,
53, 665. doi:10.1071/CH00098
1732, 1640, 1396. δH 7.36–7.42 (5H, m), 6.48 (1H, s), 5.12 (1H, q,
J 0.82), 3.07 (3H, s), 2.97 (3H, s), 1.99 (3H, d, J 0.82). δC 170.1, 164.7,
163.3, 139.4, 134.0, 130.4, 129.1, 126.4, 121.7, 89.2, 38.0, 34.6, 12.7.
m/z 272 (M+, 8%), 255 (2), 228 (100), 213 (10), 200 (5), 184 (29), 169
(11), 156 (7), 144 (35), 129 (12), 105 (12), 82 (13), 72 (41).
[2] M. Cox, R. H. Prager, D. M. Riessen, Arkivoc 2001, 2, Part 5.
[3] K. H. Ang, M. Cox, W. D. Lawrance, R. Prager, J. A. Smith,
W. Staker, Aust. J. Chem. 2003, 56, 101. doi:10.1071/CH03140
[4] J. L. Janin, J. Chiki, M. Legraverend, C. Huel, E. Bisagni, Tetra-
hedron 1999, 51, 12 797. doi:10.1016/S0040-4020(99)00755-3
[5] L. F. Lee, Y. L. Sing, J. Org. Chem. 1990, 55, 380.
[6] A. Basak, M. K. Nayak, A. K. Chakraborti, Tetrahedron Lett.
1998, 39, 4993.
FVP of 14
Pyrolysis (450◦C, 175◦C, 0.1 mmHg) of 14 (120 mg, 0.44 mmol) gave
a dark brown oil (100 mg, 83%), which was purified by radial chro-
matography (beginning at 30% ethyl acetate/light petroleum), to give
four fractions.
[7] D. W. Jeffery, T. Lister, R. H. Prager, Aust. J. Chem. 2004, 57,
491. doi:10.1071/CH03139
[8] C. L. Hickson, H. McNab, J. Chem. Res. 1989, 176.
[9] J. Kurita, K. Iwata, T. Tsuchiya, J. Chem. Soc., Chem. Commun.
1986, 1188.
[10] J. Kurita, K. Iwata, T. Tsuchiya, Chem. Pharm. Bull. Jpn. 1987,
35, 3166.
[11] A. Ben Cheikh, J. Chuche, N. Manisse, J. C. Pommelet,
K. Netsch, P. Lorencak, C. Wentrup, J. Org. Chem. 1991,
56, 970.
The first fraction was tentatively assigned as 3-hydroxy-2-methyl-
6-phenylpyridine 17 (30 mg 33%) as white needles, mp 45◦C
(Found: [M + H]+ 186.0919. C12H12NO requires [M + H]+ 186.0918).
vmax/cm−1 2917, 1667, 1633, 1584, 1557, 1499. δH (CDCl3/TFA) 7.93
(1H, d, J 8.79), 7.70 (1H, d, J 8.79), 7.50–7.61 (5H, m), 2.67 (3H,
s). δC (CDCl3/TFA) 153.3, 144.4, 143.3, 131.7, 131.6, 130.2, 129.7,
127.1, 123.9, 14.3. m/z (GC-MS) 186 ([M + H]+, 100%), 156 (19),
115 (15), 89 (5).

460
M. Cox et al.
[12] A. Filippi, G. Roselli, G. Renzi, F. Grandinetti, M. Speranza,
Chem Eur. J. 2003, 9, 2072. doi:10.1002/CHEM.200204281
[13] J. J. Gajewski, J. Am. Chem. Soc. 1970, 92, 3688.
[14] J. R. Peterson, T. J. Winter, C. P. Miller, Syn. Commun. 1988,
18, 949.
[17] C. Koshima, M. Yamomoto, S. Kobayashi, N. Sugiyama, Bull.
Chem. Soc. Jpn. 1969, 42, 2389.
[18] O. S. Tee, M. Paventi, J. Am. Chem. Soc. 1982, 104, 4142.
[19] M. A. Elkaschef, M. H. Nosseir, J. Am. Chem. Soc. 1960, 82,
4344.
[15] S. N. Huckin, L. Weiler, Tetrahedron Lett. 1971, 4835.
doi:10.1016/S0040-4039(01)97628-2
[20] D. D. Perrin, W. L. F. Armarego, D. R. Perrin, Purification of
Laboratory Chemicals, 2nd edn 1980 (Pergamon: Oxford).
[16] B. Lygo, Tetrahedron 1995, 51, 12 859. doi:10.1016/0040-
4020(95)00820-X