Synthesis of 2-pyridinylbenzoxazole: Mechanism for the intramolecular photosubstitution of the haloarene with the carbonyl oxygen of the amide bond in basic medium

Article abstract of DOI:10.1021/jo9909498

2-Pyridinylbenzoxazole derivatives have been synthesized by the intramolecular photosubstitution reaction of N-(2- halophenyl)pyridinecarboxamide (1 and 2) with its amide bond in basic medium. In neutral medium both intramolecular photosubstitution and photoreduction reactions occurred. In the photosubstitution reaction a singlet state of the o-haloarene is involved, whereas in the photoreduction a triplet state of the o-haloarene is involved; oxygen inhibited the photoreduction but not the photosubstitution. The relative rate studies showed that a base accelerates the photosubstitution reaction but decelerates the photoreduction. o- Iodoarenecarboxamide is more reactive than o-bromoarenecarboxamide, which in turn is more reacitive than o-chloroarenecarboxamide. UV-vis absorption change in the presence of a base showed that an imidol and/or imidolate anion is involved in the reaction. Several transient species, such as charge- transfer excited states and a cyclohexadienyl anion radical, have been identified from the photolysis of 1 and 2 in basic medium by laser flash photolysis. In neutral medium dibromide anion radical and a phenyl radical were identified in addition to the above intermediates. On the basis of the photokinetic and laser flash photolysis studies, an intramolecular photosubstitution of N-(o-halophenyl)pyridinecarboxamide with its amide bond occurs via an intramolecular S(N)(ET)Ar* mechanism to afford 2- pyridinylbenzoxazole derivative, and the photoreduction proceeds via a free radical mechanism to give N-phenylpyridinecarboxamide.

Full text of DOI:10.1021/jo9909498

8546
J . Org. Chem. 1999, 64, 8546-8556
Syn th esis of 2-P yr id in ylben zoxa zole: Mech a n ism for th e
In tr a m olecu la r P h otosu bstitu tion of th e Ha loa r en e w ith th e
Ca r bon yl Oxygen of th e Am id e Bon d in Ba sic Med iu m
Yong-Tae Park,* Chang-Hee J ung, and Kwang-Wook Kim
Department of Chemistry, Kyungpook National University, Taegu 702-701, Korea
Ho Sik Kim
Department of Chemistry, Catholic University of Taegu-Hyosung, Gyongsan 712-702, Korea
Received J une 11, 1999
2-Pyridinylbenzoxazole derivatives have been synthesized by the intramolecular photosubstitution
reaction of N-(2-halophenyl)pyridinecarboxamide (1 and 2) with its amide bond in basic medium.
In neutral medium both intramolecular photosubstitution and photoreduction reactions occurred.
In the photosubstitution reaction a singlet state of the o-haloarene is involved, whereas in the
photoreduction a triplet state of the o-haloarene is involved; oxygen inhibited the photoreduction
but not the photosubstitution. The relative rate studies showed that a base accelerates the
photosubstitution reaction but decelerates the photoreduction. o-Iodoarenecarboxamide is more
reactive than o-bromoarenecarboxamide, which in turn is more reacitive than o-chloroarenecar-
boxamide. UV-vis absorption change in the presence of a base showed that an imidol and/or
imidolate anion is involved in the reaction. Several transient species, such as charge-transfer excited
states and a cyclohexadienyl anion radical, have been identified from the photolysis of 1 and 2 in
basic medium by laser flash photolysis. In neutral medium dibromide anion radical and a phenyl
σ radical were identified in addition to the above intermediates. On the basis of the photokinetic
and laser flash photolysis studies, an intramolecular photosubstitution of N-(o-halophenyl)-
pyridinecarboxamide with its amide bond occurs via an intramolecular SN(ET)Ar* mechanism to
afford 2-pyridinylbenzoxazole derivative, and the photoreduction proceeds via a free radical
mechanism to give N-phenylpyridinecarboxamide.
In tr od u ction
Ramakrishnan and co-workers3a reported that 2-me-
thylbenzothiazoles were synthesized via intramolecular
photosubstitution of o-halothioacetanilides. Recently they3b
proposed a reasonable intramolecular electron-transfer
mechanism for the formation of benzothiazole. Bowman
and co-workers4 reported that 2-phenyl- and 2-methyl-
benzothiazole were prepared in high yield from the
photoinduced intramolecular aromatic substitution of
o-iodothioacetanilide and o-iodothiobenzanilide in the
presence of potassium t-buthoxide and acetone. Arad-
Yellin and co-workers5 reported that on UV irradiation,
2,6-dichlorocinnamic acid and its ester underwent pho-
tosubstitution to yield 5-chlorocoumarin; amide deriva-
tive yielded the corresponding imino analogues. For imino
analogue formation (N-methyl-5-chloro-2H-benzopyran-
2-imine), an intermediate formed from an addition of
carbonyl oxygen of the amide bond to the o-phenyl carbon
bearing chlorine in the 2,6-dichlorophenyl group was
assumed.
In an attempt to develop a photochemical ring forma-
tion of o-haloarene tethered to arene,1,2 we recently found
that a photoreaction of N-(2-halophenyl)pyridinecarboxa-
mide gave a 2-pyridinylbenzoaxole, a photosubstitution
product of the o-haloarene with the carbonyl oxygen of
the amide bond within the molecule. The intramolecular
photosubstitution reaction is valuable and simple for the
benzoxazole ring formation and virtually unknown.
There are few literature reports3-5 on the intramolecu-
lar photosubstitution reaction of o-haloarene with car-
bonyl oxygen and thiocarbonyl sulfur of amide and
thioamide, even though there are several reports6-8 on
the intramolecular aromatic substitution reaction of
ground-state o-haloarene with thiocarbonyl sulfur nu-
cleophile of thioamide and analogues.
(1) Park, Y.-T.; Song, N. W.; Kim, Y.-H.; Hwang, C.-G.; Kim, S. K.;
Kim. D. J . Am. Chem. Soc. 1996, 118, 11399.
(2) Park, Y.-T.; Song, N. W.; Hwang, C.-G.; Kim, K.-W.; Kim. D. J .
Am. Chem. Soc. 1997, 119, 10677.
(3) (a) Paramasivam, R.; Palaniappan, R.; Ramakrishnan, V. T. J .
Chem. Soc., Chem. Commun. 1979, 260. (b) J ayanthi, G.; Muthusamy,
S.; Paramasivam, R.; Ramakrishnan, V. T.; Ramasamt, N. K.; Rama-
murthy, P. J . Org. Chem. 1997, 62, 5766.
(4) Bowman, W. R.; Heaney, H.; Smith, P. H. G. Tetrahedron Lett.
1982, 23. 5093.
(5) Arad-Yellin, R.; Green, B. S.; Muszkat, K. A. J . Org. Chem. 1983,
48, 2578.
However, the intramolecular photosubstitution of o-
haloarene with the carbonyl oxygen of the amide bond is
rare and has not been explored mechanistically. This
paper describes the result of our investigations for the
synthesis of benzoxazole derivative and this reaction
mechanism.
(6) (a) Bunnett, J . F.; Hrutfiord, B. F. J . Am. Chem. Soc. 1961, 83,
1691. (b) Bunnett, J . F.; Flynn, R. R.; Skorcz, J . A. J . Org. Chem. 1963,
28, 1.
(7) Couture, A.; Granclaudon, P. Heterocycles 1984, 22, 1383.
(8) Flouzat, C.; Guillaumet, G. Synthesis 1990, 64.
Resu lts a n d Discu ssion
N-(2-Halophenyl)pyridinecarboxamides and N-(phe-
nyl)pyridinecarboxamide (1 and 2 in Chart 1) were
10.1021/jo9909498 CCC: $18.00 © 1999 American Chemical Society
Published on Web 11/12/1999

Synthesis of 2-Pyridinylbenzoxazole
J . Org. Chem., Vol. 64, No. 23, 1999 8547
Ta ble 1. P r od u cts a n d Th eir Yield in th e P h otor ea ction of N-(2-Ha lop h en yl)p yr id in eca r boxa m id e
starting
compd
reaction time
(h)
recovered starting
compd (%)
product yield
(%)
reaction medium
product
1a
1b
2a
2b
1b
CH3CN (210 mL)/2 M aq NaOH (30 mL)
CH3CN (210 mL)/2 M aq NaOH (30 mL)
CH3CN (210 mL)/2 M aq NaOH (30 mL)
CH3CN (210 mL)/2 M aq NaOH (30 mL)
CH3CN (140 mL)/0.2 M aq NaOH (100 mL)
30.0
8.0
21.0
3.5
38.2
23.3
11.4
27.7
4.6
3
3
4
4
3
1d
4
5
25.3
34.4
37.3
43.1
29.2
7.6
6.0
2b
CH3CN (220 mL)/0.02 M aq Na2CO3 (20 mL)
6.5
11.1
19.5
27.3
Ch a r t 1
Sch em e 2
Ta ble 2. Qu a n tu m Yield on th e P h otosu bstitu tion of
N-(2-Ha lop h en yl)p yr id in eca r boxa m id e (1) w ith a n d
w ith ou t Oxygen a
quantum yield (φ)
substrate
with Ar
with O2
Sch em e 1
1a
1b
1c
0.002
0.025
0.060
0.002
0.025
0.055
a
In CH3CN containing 0.1 N aq NaOH (50/1) with monochro-
matic light (266 ( 3 nm, Xe lamp, 450 W).
N-(2-Bromophenyl)-N-methyl-4-pyridinecarboxamide
(1e) did not give the expected photosubstituted product
but produced photocyclized (25%) and photoreduced (5%)
products (Scheme 2). This abnormal bahavior of 1e
suggests an important imidol structure for the photo-
substitution of 1a , 1b, and 2.
A blank test for the photoreaction was performed. The
thermal reaction of 1a and 1b under the above conditions
for 2 days did not produce a cyclized and reduced product
but a small amount of a hydrolyzed product. Thus the
reactions that produced 2-arylbenzoxazole and N-(phe-
nyl)pyridinecarboxamide are not the thermal reactions
but the photoinduced subsititution and reduction reac-
tions.
prepared by reacting pyridinecarboxylic acids with the
appropriate 2-haloanilines and aniline. The preparation
details of these pyridinecarboxamides will be described
in the Experimental Section.
P r ep a r a tive Rea ction . When an acetonitrile solution
(240 mL) of N-(2-chlorophenyl)-4-pyridinecarboxamide
(1a , 1 mmol) containing 30 mL of aqueous 2 M NaOH
was irradiated by a Hg lamp (150 W, high pressure)
under nitrogen for 30 h, a photosubstituted product, 2-(4-
pyridinyl)benzoxazole (3), was obtained in 25% yield
(Scheme 1). This compound (mp, 130-131; lit. value,
130-1319) was identified by its spectral properties (NMR,
IR, MS) and elemental analysis. The photosubstitution
reaction is novel and involves replacement of the chlorine
of the chlorophenyl moiety by the carbonyl oxygen of the
amide bond within the molecule.
When bromo analogue 1b was irradiated under the
above conditions for 8 h, the same product, 2-(4-pyridi-
nyl)benzoxazole (3), was obtained in 34% yield. The
reaction was extended to N-(2-halophenyl)-3-pyridinecar-
boxamide (2). The pertinent results of the synthetic
reactions are shown in Table 1 and Scheme 1. The
photosubstitution reaction yields in the table range from
a value of 20% for 2b in acetonitrile containing Na2CO3
to 43% for 2b in acetonitrile with NaOH. This reaction
is valuable and simple for the synthesis of arylbenzox-
azole derivative.
Kin etics. To elucidate the reaction mechanism, the
quantum yields of the photosubstitution reactions and
the relative rates of the reaction on several conditions
have been studied.
The quantum yields of the photosubstitution in several
conditions have been measured and are shown in Table
2. The quantum yield in Table 2 ranges from a value of
0.002 for 1a to 0.06 for 1c. The iodoarene 1c is about 2
times more reactive than bromoarene 1b, which in turn
is about 10 times more reactive than chloroarene 1a . The
quantum yield of 1a and 1b was not changed in the
presence of oxygen as a triplet quencher, although that
of iodoarene 1c was reduced a little (8%). These results
imply that the singlet excited state of the haloarene may
be mainly involved in the photosubstitution reaction for
iodoarene 1c, a small portion of a triplet being involved
in addition to the singlet excited-state reaction.
As an example for the relative rate of the haloarene,
bromo analogue 1b (17 mg) was dissolved in acetonitrile
In weak bases such as acetonitrile containing Na2CO3
or dillute NaOH (last two entries in Table 1), a photore-
duced product, N-(phenyl)pyridinecarboxamide (1d , 5),
was formed in addition to the photosubstitution products.
The photoreduction of haloarene is well known in neutral
medium.10
(10) Bunce, N. J . In CRC Handbook of Organic Photochemistry and
Photobiology; Horspool, W. H., Song, P.-S., Eds.; CRC Press: New York,
1995; p 1181.
(9) Hisano, T.; Ichkawa, M.; Tsumoto, K.; Tasaki, M. Chem. Pharm.
Bull. 1982, 30, 2996.

8548 J . Org. Chem., Vol. 64, No. 23, 1999
Park et al.
Ta ble 3. Rela tive Ra tes of th e F or m a tion s of
2-P yr id in ylben zoxa zole a n d
N-P h en ylp yr id in eca r boxa m id e in th e P h otor ea ction of
N-(2-Ha lop h en yl)p yr id in eca r boxa m id e w ith a Br oa d
Sp ectr u m of Ligh t fr om a Hg La m p (150 W)
This effect of the leaving halide anions on the subsititu-
tion rate is consistent with the quantum yields of 1 as
described above. It seems likely that the substitution
reaction is governed by the leaving trend of the halide
anion from a reaction intermediate. In other words the
rate-determining step of the reaction is the leaving step
of the halide anion of an intermediate. Moreover, if the
SN2Ar* mechanism is operative, a phenolic-type product,
which could be formed from the intermolecular substitu-
tion reaction of the haloarene with hydroxide ion, could
not be detected in the reaction condition. Thus these
results suggest that the intramolecular SN(ET)Ar* mech-
anism is more favorable than SN2Ar*.
relative rate
entry substra
solvent
atm photosub photoredb
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
1b
1b
1b
1b
1c
1b
1b
1b
2a
2a
2a
2b
1b
1b
1b
CH3CN
CH3CN
N2
O2
N2
O2
N2
N2
N2
N2
N2
O2
N2
N2
N2
2.2
2.2
5.1
5.1
15.0
1.0
0.3
<0.01
0.1
0.1
1.6
6.0
1.8
2.1
4.6
3.8
<0.01
CH3CN/2 M NaOHb
CH3CN/2 M NaOHb
CH3CN/2 M NaOHb
benzene
1.2, (nd)d
<0.01
12.5
0.4
0.7
0.7
0.05
<0.01
0.06
0.06
4.2
cyclohexane
MeOH
CH3CN
CH3CN
CH3CN/2 M NaOHb
CH3CN/2 M NaOHb
CH3CN/NEt3 (2.1 mM)
CH3CN/MMAc (5.6 mM) N2
2.9
nd
CH3CN/2 M NaOH
5.0 × 10-4 M isoprene
CH3CN/2 M NaOH
0.05 M K2S2O8
N2
The reactivities of the haloarene toward the photore-
duction were not very clear because of the low reaction
rates in basic medium (entries 11 and 12), although iodo
analogue 1c was somewhat reactive (entry 5). Probably
the heavy atom effect of the iodo group increases the
population of the triplet state, which is involved in the
reduction, by enhancing intersystem crossing. The phenyl
σ radical is possible by a homolytic cleavage of the phenyl-
iodine bond from the triplet state even in basic medium
and may then cause an increasing photoreduction reac-
tion rate.
In nonpolar solvents such as benzene and cyclohexane,
the substitution rates of 1b were quite lower than that
in acetonitrile, and the reduction rate was moderately
lower than that in acetonitrile (entries 6 and 7). Probably
an imidolate anion and/or imidol is not efficiently popu-
lated in this medium for the substitution and reduction.
In the protic solvent methanol, the substitution reaction
of 1b sparingly occurred, and its reduction reaction
occurred weakly (entry 8). Methanol probably inhibits the
substitution by blocking the formation of the imidolate
anion of 1b by its protonation.
It is noteworthy that the reduction product was not
detected when a methanol solution filter was used before
the window of the reaction cuvette (entry 3). This
suggests that the reduction path is blocked by filtering
out the shorter wavelength from the broad spectrum of
light. In other words, a locally excited state of the
pyridinecarboxamide, probably the excited state of the
halo phenyl moiety produced by the excitiation by the
shorter wavelength, undergoes reduction reaction via a
homolytic cleavage of phenyl-halogen bond (vide infra).
3-Pyridinylcarboxamide 2b is more reactive than 4-py-
ridinylcarboxamide 1b (entries 3 and 12) for the photo-
substitution, indicating the importance of the electronic
effect of nitrogen position in the pyridine moiety; the
nitrogen on the 3-pyridine moiety of 2b withdraws
electrons more effectively to give a more acidic imidol,
to provide a more anionic form of the imidol, and
eventually to produce more substitution product com-
pared to the reaction of 1b.
16
1b
N2
1.5
nd
The concentration of the substrates used is 4.1 × 10-3 M.
a
d
b CH3CN/2 M NaOH ) 13/2. c Methyl methacrylate. Not detected
when methanol solution filter is used before the cuvette.
(15 mL), and the solution was purged with nitrogen (or
oxygen) for 10 min and irradiated with a Hg lamp (150
W, high pressure) for 1.5 h. The reaction mixture was
analyzed by GC (FID), and formation of the substituted
products and the reduced product was monitored
(Table 3).
In acetonitrile the rate of photoreduction of 1b is faster
than that for photosubstitution (entry 1 in Table 3). In
the presence of oxygen the substitution reaction rate of
1b was again not affected, whereas the reduction rate
was reduced extensively (entries 1 and 2). This held true
for the photoreactions of 2a in acetonitrile and of 1b in
acetonitrile containing NaOH (entries 9, 10, 3, and 4).
These results are in line with the oxygen effect on the
quantum yield as described above. Thus, in the photo-
substitution reaction a singlet excited state of the
haloarene is involved, whereas in the photoreduction a
triplet state is involved.
The presence of a base such as NaOH led to an
acceleration for the photosubstitution by more than twice
and to a deceleration for the photoreduction by more than
3 times (entries 1 and 3). Similar trends were observed
for the photosubstitution (entries 9 and 11) of 2a and 2b.
One possible explanation is that an anionic nucleophile
of carbonyl oxygen of the amide bond in basic medium
intramolecularly displaces the halogen of the haloarene
in the singlet excited state (SN2Ar*).11 Another possibility
is that an anionic radical species photoinduced from
anionic carbonyl oxygen of the imidol in basic medium
(vide infra) intramolecularly substitutes the halogen of
the haloarene (SN(ET)Ar*).11
When the photosubstitution reactivities of the iodoare-
necarboxamide and bromoarenecarboxamide are com-
pared, the iodo analogue, which has a better leaving
group, is 3 times more reactive than the bromo analogue
(entries 3 and 5). A similar trend was observed for the
photosubstitution rate of 2a and 2b (entries 11 and 12).
In the presence of triethylamine, which is known as a
-
good electron donor for stimulated SR N1Ar reactions,12
the photosubstitution rate is reduced a little, whereas
(11) Cornelisse, J . In CRC Handbook of Organic Photochemistry and
Photobiology; Horspool, W. H., Song, P.-S., Eds; CRC Press: New York,
1995; p 250.
(12) Beugelmans, R. In CRC Handbook of Organic Photochemistry
and Photobiology; Horspool, W. H.; Song, P.-S., Eds.; CRC Press: New
York, 1995; p 1200.

Synthesis of 2-Pyridinylbenzoxazole
J . Org. Chem., Vol. 64, No. 23, 1999 8549
F igu r e 1. UV-vis absorption spectra of 1b in acetonitrile (1.2 × 10-4 M 1b, 2 mL, dotted line) and in acetonitrile containing
NaOH (1.2 × 10-4 M 1b, 2 mL; 1.0 M NaOH, 0.1 mL; thin solid line). The thick solid line was obtained by subtraction of the
dotted line from the thin solid line.
the photoreduction rate is enhanced somehow (entry 13).
This result implies that the SR N1Ar mechanism is not
present in the photosubstitution reaction but may par-
tially be in the reduction reaction.
In the presence of radical scavengers such as methyl
methacrylate or isoprene, the photosubstitution reaction
is retarded a little, whereas the photoreduction reaction
is retarded extensively (entry 14 and 15). This implies
that photoreduction obviously proceeds via a radical-
mediated reaction.
and/or light absorption, because the species was not seen
in the UV-vis absorption of 1b in methanol containing
NaOH and the similar species was not seen in the UV-
vis absorption of 1e in acetonitrile containing NaOH (see
the Supporting Information). Moreover, the species with
λmax at 242 and 325 nm was not seen in the above
conditions in the UV-vis absorption of 1e, which has a
methyl group on the nitrogen of the amide bond and thus
is not possible for the formation of the imidol structurally
in acetonitrile containing NaOH (not shown). The same
UV-vis absorption behavior of 1a in the basic medium
as that of 1b was observed (λmax 266 nm, not shown). The
lifetime of the imidol form from 1a by irradiation by a
Xe lamp was measured. The lifetime was 2.84 h (Table
4). In fact, an imidol formation from the photoinduced
proton-transfer tautomerization of benzanilide has been
reported.14
The fluorescence spectra of 2b in acetonitrile with and
without NaOH were obtained at 200 ns delay after laser
flash (power 30 mJ ) and are shown in Figure 2. The
luminescence spectra consists of two bands around 337
and 400 nm, with onset at 319 and 350 nm, respectively
(solid line; the sharp line at 532 nm is from the laser
source). In the presence of NaOH a new broad fluores-
cence spectra around 475 nm (dotted line) appeared in
addition to the two peaks in acetonitrile only. The inset
of Figure 2 showed the fluorescence spectra of 2b
measured 50 ns after the laser flash (80 mJ ). The thick
solid line (lowest one) in the inset shows a diffused
fluorescence spectra in these conditions. It seems that
the efficient formation of a band around 400 nm by high
power laser (80 mJ ) blocks the observation of the band
at about 337 nm (vide infra). The thin solid line exhibits
that just the scanning (irradiation) increased the fluo-
rescence intensity around 400 nm. The dotted lines show
that the addition of NaOH increased the fluorescence
-
The presence of K2S2O8, which is known as a solvated
+
electron acceptor for a stimulated SR N1Ar* reaction,13
did not improve the photoreactions, indicating that the
mechanism is other than SR N1Ar* (entry 16).
+
La ser F la sh P h otolysis. Before the laser flash pho-
tolysis study, the UV-vis absorption behavior of the
halophenylpyridinecarboxamide was studied. The λmax
values of all N-(2-halophenyl)pyridinecarboxamides in
acetonitrile are from 264 to 267 nm, whereas that of
N-phenylpyridinecarboxamide (1d , no o-substituent) is
270 nm and that of N-(2-bromophenyl)-N-methylpyridi-
necarboxamide (1e) is 254 (sh) nm. Thus, the energy state
of the singlet excited state of an arene with an o-halogen
substituent is higher than that of an arene without an
o-substituent (see Experimental Section for data).
The absorption spectra of 1b in acetonitrile (dotted
line) and acetonitrile containing NaOH (thin solid line)
are shown in Figure 1. The prototype absorption spectra
of 1b in acetonitrile (λmax ) 267 nm) changed to a
structureless type of absorption spectra by the addition
of NaOH. If the prototype absorption was subtracted from
the structureless absorption, two new peaks around 242
and 325 nm were obtained (thick solid line). The species
at about 325 and 242 nm is believed to be an intramo-
lecular charge-transfer excited state from the imidolate
anion of 1b formed from the amide by action of the base
(13) Bunce, N. J .; Bergsma, J . P.; Schmidt, J . L. J . Chem. Soc.,
Perkin Trans. 1981, 2, 713.
(14) Tang, G.-Q.; MacInnis, J .; Kasha, M. J . Am. Chem. Soc. 1987,
109, 2531.

8550 J . Org. Chem., Vol. 64, No. 23, 1999
Park et al.
F igu r e 2. Fluorescence spectra of 2b 200 ns after laser flash (30 mJ ) in acetonitrile without NaOH (solid curve) and with NaOH
(dotted line). Inset: fluorescence spectra of 2b (36 mg/500 mL AN) 50 ns after laser flash (80 mJ ) on several conditions: ) first
scan; s second scan; - - - with 0.5 mL of 4 M NaOH; - - - - - with 1.0 mL of 4 M NaOH; - ‚ ‚ -, with 1.5 mL of 4 M NaOH.
Ta ble 4. Tr a n sien t Sp ecies fr om th e P h otolysis of
1 a n d 2
from the locally singlet excited state of the keto amide
(LE) and the intramolecular charge-transfer excited state
of the imidol (CTH), respectively, and in the base the
diffused luminescence peak around 475 nm originated
from the intramolecular charge-transfer excited state of
the enolate ion (CT).
It should be noted that the variation of the fluorescence
intensity on the basicity of the medium could be seen at
50 ns delay after laser flash but not at 200 ns delay
(Figure 3, Table 4). Thus the lifetime of the charge
transfer species is around 50 ns, less than 200 ns. It
seems likely that in high laser intensity (80 mJ ) the
process of the strong luminescence species (CTH) to a
deprotonated species (CT) by deprotonation with NaOH
was the major one. Thus, with the deprotonation process
of the excited state, the pKa value of the charge-transfer
excited state from the imidol could be measured. The pKa
intensity around 475 nm with an isosbestic point at about
440 nm. It seems likely that the fluorescence intensity
around 475 nm increases by simultaneous expenditure
of the band around 400 nm in the basic medium, even
though the irradiation (scanning) disrupts the intensity
around 400 nm. Figure 3 shows the fluorescence spectra
of 1b in acetonitrile with and without NaOH (at 50 ns
delay, laser power 80 mJ ). Again a scan (irradiation of
laser) increased the intensity of the luminescence at
about 410 nm, the addition of NaOH shifted the λmax of
the fluorescence (415 nm) to that of fluorescence at 451
nm with an isosbestic point at 434 nm, and the band
around 475 nm increased by expenditure of the band
around 375 nm. It is assumed that, in Figure 2, in
acetonitrile the small luminescence peak around 337 nm
and the strong luminescence peak around 400 nm arise
values for the excited imidol 2b and 1b were roughly
estimated to be 9 and 10, respectively, by measuring the
fluorescence change in several pH ranges (Figures 2 and
3). The excited state is extremely acidic compared with
that of the ground state of benzanilide (pKa ) 16.5)15 and
thus probably provides the intramolecular charge-
transfer species (CT), which undergoes an intramolecular
photosubsitution reaction via an anionic radical inter-
mediate [SN(ET)Ar* reaction].
Transient absorption spectra were obtained from the
laser flash photolysis of 1b in argon-saturated acetoni-
trile containing NaOH as shown in Figure 4. There is a
transient absorption peak at about 393 nm at 10 µs delay
(15) Homer, R. B.; J ohnson, C. D. In The Chemistry of Amides; Patai,
S., Ed.; Interscience: London, 1970; p 187.

Synthesis of 2-Pyridinylbenzoxazole
J . Org. Chem., Vol. 64, No. 23, 1999 8551
F igu r e 3. Fluorescence of 1b 50 ns after laser flash (80 mJ ) in acetonitrile (36 mg/500 mL AN): ) first scan; s second scan;
- - - with 0.5 mL of 4 M NaOH; - - - - - with 1.0 mL of 4 M NaOH; -‚ - ‚ -with 1.5 mL of 4 M NaOH; - ‚ ‚ - ‚ ‚ with 2.5 mL
of 4 M NaOH.
F igu r e 4. Transient absorption spectra obtained upon 266 nm excitation of 1b in acetonitrile (1 L) containing 1 mL of 4 M
NaOH 10 µs ()) and 1 ms (s) after laser pulse. Inset: temporal profile of 393 nm transient (- - - -) and with mercuric acetate (s).
after laser flash. The region between 280 and 320 nm is
governed by the luminescence of 1b. At 1 ms delay after
laser flash the absorption band at about 393 nm disap-
peared, whereas a new band at about 285 nm appeared.
The temporal profile of the transient at about 393 nm
from the photolysis of 1b is shown in the inset of Figure
4. The species decays in a first-order process, and its
lifetime is 39 µs. The species was not quenched by radical
scavengers isoprene or oxygen but was quenched weakly
by the presence of mercuric acetate (τ ) 27 µs), indicating
an anionic radical species (thin curve in inset of Figure
4). Thus, the transient species that has a lifetime of 39
µs is assigned to a cyclohexadienyl anion radical (CHAR),
which produces the pyridinylbenzoxazole (Table 4). This
assignment was strengthened by the observation that
dibromide anion radical could not be seen in the laser
flash photolysis of 1b in acetonitrile containing NaOH.
Furthermore, the 393-transient was not seen in the laser
flash photolysis of the unsubstituted N-phenyl-4-pyridi-
necarboxamide (1d ) and N-(2-bromophenyl)-N-methyl-

8552 J . Org. Chem., Vol. 64, No. 23, 1999
Park et al.
F igu r e 5. Dependence of the 393 nm transient concentration (10 µs after laser pulse) on the laser pulse energy in the photolysis
of acetonitrile of 1b (1.4 × 10-4 M) under argon; the slope in the log-log plot is 1.8.
4-pyridinecarboxamide (1e), which do not afford the
photosubstituted product (both transient absorption spec-
tra shown as Supporting Information).
were reported by Grossweiner and Matheson;17 λmax of
the diiodide anion radical is 370 nm in aqueous solution.
In acetonitrile only, transient absorption spectra from
the photolysis of 1b were obtained and are shown in
Figure 6. At 10 µs delay after laser flash the weak
diffused absorption band in the region between 340 and
420 nm was seen, whereas a luminescence disturbed the
region between 280 and 320 nm. It seems likely that the
diffused band consists of two bands peaking at 360 and
393 nm. At 1 ms delay a certain band in the region
between 280 and 320 nm (the same band in the basic
medium) was observed.
A temporal profile of the band at 360 nm decayed in
the first order with a lifetime of 33 µs. Although the
lifetime (33 µs) in acetonitrile is longer than that of
dibromide anion radical in water (10 µs),1 the band is
assigned to dibromide anion radical because its formation
increased in the presence of tetraethylammonium bro-
mide and its lifetime remained (33 µs, right inset of
Figure 6, Table 4).
Figure 5 shows the dependence of the cyclohexadienyl
anion radical intermediate (CHAR, 393-nm transient
species) from 1b in the basic medium on the photolysis
laser pulse energy. The slope of the log-log plot for the
intermediate versus laser pulse energy was obtained to
be 1.8. This result indicates that the intermediate
(CHAR) was formed mainly by a biphotonic process.
The band from 285 to 315 nm in Figure 4 is quenched
by the presence of a radical scavenger, N-tert-butyl-R-
phenylnitrone. Thus it is assigned to phenyl σ radical
from 1b. It is believed that this is related with reduction
product (vide infra).
Similar transient absorption spectra were obtained
from the laser flash photolysis studies of 1a and 1c under
basic conditions (shown as Supporting Information).
However, for iodo analogue 1c, there was an additional
transient absorption band around 370 nm whose lifetime
was 1.8 ms. The transient was not formed at all in the
presence of N-tert-butyl-R-phenylnitrone, probably a
precursor for the transient formation being quenched
completely. The transient formation was lowered in the
presence of oxygen, probably because the precursor, the
triplet state of 1c, for the transient formation was
quenched by oxygen (shown as Supporting Information).
The transient was quenched by addition of o-xylene to
yield a complex between an iodine atom and o-xylene
(λmax ) 580 nm, τ580 ) 43 µs) which is known.16 It seems
likely that the transient at 370 nm corresponds to
diiodide anion radical, which could be formed by reaction
of iodine radical with iodide ion: the iodine radical was
formed by the homolytic cleavage of the phenyl-iodine
bond from the excited state, and the iodide ion was
formed from the preceding photoreaction (Table 4). The
identification of diiodide anion radical and its kinetics
The behavior of the 393-nm band from this neutral
solvent is the same as that observed in acetonitrile
containing NaOH. The lifetime is somewhat shorter (21
µs) than that in the base.
The lifetime of the band from 280 to 320 nm was 4 ms.
Oscilloscope traces show (left inset of Figure 6) a match-
ing decay of the band (300 nm) and a band formation of
N-tert-butyl-R-phenylnitrone adduct (300 nm), indicating
the band at 300 nm to be radical species. The property
of the band of the 300-nm transient is similar to the
phenyl σ radical from the photolysis of N-(o-halophenyl)-
pyridinium salt.1 On the basis of the above observations
of its radical property and of the presence of dibromide
anion radical in the medium, the band between 280 and
320 is assigned to phenyl σ radical formed from homolytic
cleavage of the phenyl-bromine bond in the triplet state,
probably a locally excited state of the amide. This phenyl
(16) Strong, R. L.; Rand, S. J .; Britt, J . A. J . Am. Chem. Soc. 1960,
82, 5053.
(17) Grossweiner, L. I.; Matheson, M. S. J . Phys. Chem. 1957, 61,
1089.

Synthesis of 2-Pyridinylbenzoxazole
J . Org. Chem., Vol. 64, No. 23, 1999 8553
F igu r e 6. Transient absorption spectra obtained upon 266 nm excitation of 1b in acetonitrile (1.4 × 10-4 M) 10 µs ()) and 1 ms
(s) after laser pulse. Left inset: solid line, decay curve of 300 nm transient; dotted line, the formation of an adduct of N-tert-
butyl-R-phenylnitrone. Right inset: solid line, decay curve of 360 nm-transient; dotted line, the same decay curve in the presence
of tetraethylammonium bromide.
Sch em e 3
2-(pyridinyl)benzoxazole (3) by ejection of the halide
anion (right path in the Scheme 3). This mechanism also
explains why no intermolecular hydroxide anion-substi-
tution product is observed in the haloarene and why
radical savengers cannot quench the anion radical. A
similar intramolecular electron-transfer mechanism has
been proposed for the benzothiazole formation in the
photoreaction of o-halothioacetanilide.3b
In neutral conditions the intramolecular charge-
transfer excited state of imidol (CTH) is involved in the
substitution. Deprotonation of CTH produces charge-
transfer excited state (CT), which in turn eventually
yields the 2-pyridinylbenzoxazole (middle path in Scheme
3). Either the locally excited state or the triplet imidol is
involved in the reduction reaction via a homolytic cleav-
age of the phenyl-halogen bond of the haloarene to
produce a halogen radical (eventually dihalide anion
radical) and a phenyl σ radical, and the subsequent
hydrogen atom abstraction of the phenyl radical from the
environment affords the reduced product (left path in
Scheme 3).
σ radical from 1b is a reaction intermediate for the
photoreduction, whereas a cyclohexadienyl anion radical
is a reaction intermediate for the photosubstitution.
If the intramolecular SN(ET)Ar* mechanism is involved
in the substitution, deuterium-exchanged product can be
seen in 2-pyridinylbenzoxazole when the reaction is
performed in deuterium oxide. A small amount (deute-
rium content 5%) of deuterated 2-(4-pyridinyl)benzox-
azole was indeed observed in the GC-MS spectra. If the
proposed mechanisms are operated in the photoreaction
of the o-halophenylpyridinecarboxamide in the basic
medium, a triplet photosensitizer such as benzophenone
will accelerate the reduction reaction but not the substi-
tution. Indeed, the experiment showed that photoreduc-
tion of 1b occurred in the presence of benzophenone, but
photosubstitution of 1b did not occur upon only excitation
of benzophenone by a monochromatic light (360 nm).
These results confirmed the proposed SN(ET)Ar* mech-
anism and free radical mechanism for the photosubsti-
tution and photoreduction reactions, respectively.
Mech a n ism . Steady state and laser flash studies
indicate that the intramolecular SN(ET)Ar* mechanism
is involved in the photosubstitution, whereas a free
phenyl σ radical is involved in the photoreduction (Scheme
3). The keto form of the amide 1 is in equilibrium with
the imidol I and the imidolate anion (IA) by its (their)
absorption of light and/or the action of the base NaOH.
In the presence of NaOH the imidolate anion and/or the
imidol is excited by absorption of light to populate the
intramolecular charge-transfer excited state (CT and/or
CTH), which produces a cyclohexadienyl anion radical
(CHAR) by the intramolecular addition of the oxygen
radical of the imidol to the very close anionic halophenyl
moiety in the charge-transfer excited state (with depro-
tonation step in case of CTH) and, in turn, yields the

8554 J . Org. Chem., Vol. 64, No. 23, 1999
Park et al.
ecarboxamide was performed by using the J ohnstone method.18
N-(2-Bromophenyl)-4-pyridinecarboxamide (1.39 g, 5 mmol)
was dissolved in acetone (30 mL). To the above solution at 50
°C were added potassium hydroxide (1.12 g) and methyl iodide
(1.13 g, 8 mmol). The reaction mixture was refluxed for 5 min,
and then the excess methyl iodide and acetone were stripped
off. The reaction mixture was analyzed on silica gel column
(70-230 mesh) with acetone/n-hexane (2/5) as eluent. The solid
(Rf ) 0.25) was recrystallized from n-hexane/diethyl ether:
yield 1.10 g (73%); mp 101 °C; UV (λmax in acetonitrile), 250
nm (sh) (ꢀ ) 7.0 × 103 L/mol cm); IR (gas phase) 3075, 2943,
Exp er im en ta l Section
Ma ter ia ls. All reagents (nicotinic acid, isonicotinic acid,
2-chloroaniline, 2-bromoaniline, 2-iodoaniline, methyl iodide,
deuterium oxide, methyl methacrylate, benzophenone, aceto-
phenone, and isoprene) and solvents used were obtained from
Aldrich.
Melting points were measured on a capillary melting point
apparatus, Thomas Hoover, or digital scanning calorimeter
(DSC). UV-vis absorption spectra were recorded on a Varian
Cary 3E spectrophotometer or a Hewlett-Packard 8452A diode
array spectrophotometer. Infrared spectra were recorded on
a Nicolet Magna 550 FT-IR (Mattson Galax series 7020)
spectrophotometer or a Hewlett-Packard IRD equipped 5890
series II gas chromatograph. A gas chromatograph with NPD
(6890 series Hewlett-Packard) was used. NMR spectra were
recorded on a Brucker Ac 200 (200 MHz) operating at 200 MHz
1
1690 cm-1; H NMR (200 MHz, CDCl3) δ 8.40 (d, J ) 5.6 Hz,
2H), 7.50 (d, J ) 7.9 Hz, 1H), 7.17-7.02 (m, 5H), 3.35 (s, 3H);
MS (EI) m/z (rel intensity) 292 (0.2, M+ + 2), 290 (0.2, M+),
211 (100). Anal. Calcd for C13H11ON2Br: C, 53.65; H, 3.81; N,
9.62. Found: C, 53.47; H, 3.88; N, 9.62.
N-(2-Ch lor op h en yl)-3-p yr id in eca r boxa m id e (2a ): yield
1.36 g (66%); mp 82 °C; UV (λmax in acetonitrile), 264 nm (ꢀ )
1.0 × 104 L/mol cm); IR (gas phase) 3440, 3084, 1712 cm-1; IR
1
for proton studies. The H NMR spectra were referenced with
respect to TMS. The mass spectra were obtained from a
Hewlett-Packard 5989A mass spectrometer with a Hewlett-
Packard 5890 series II gas chromatograph. Elemental analysis
was performed on an elemental analyzer, Carlo Erba CHNS-O
E.A. 1180.
Gen er a l P r oced u r e for t h e P r ep a r a t ion of N-(2-
Ha lop h en yl)p yr id in eca r boxa m id e (1 a n d 2). The mixture
of isonicotinic acid (2.46 g, 0.02 mole) and thionyl chloride (20
mL) was refluxed for 2 h. After the excess thionyl chloride was
stripped off, pyridine (20 mL) was added to the mixture. After
2-chloroaniline (3.18 g, 0.025 mole) was added to the above
mixture in an ice-water bath, the reaction mixture was stirred
in the ice-water bath for 30 min and then at room tempera-
ture for 1 night. When water (200 mL) was added to the
reaction mixture at 0 °C, a white solid appeared. The solid
was dried at 50-55 °C and recrystallized from diethyl ether/
n-hexane.
N-(2-Ch lor op h en yl)-4-p yr id in eca r boxa m id e (1a ): yield
1.47 g (63%); mp 134 °C; UV (λmax in acetonitrile) 266 nm (ꢀ )
8.9 × 103 L/mol cm); IR (gas phase) 3438, 3079, 1715 cm-1;1H
NMR (200 MHz, CDCl3) δ 8.80 (dd, J ) 4.4, 1.6 Hz, 2H), 8.55
(br s, 1H), 8.45 (dd, J ) 8.2, 1.5 Hz, 1H), 7.74 (dd, J ) 4.4, 1.6
Hz, 2H), 7.42 (dd, J ) 7.9, 1.5 Hz, 1H), 7.33 (dt, J ) 7.9, 1.5
Hz, 1H), 7.12 (dd, J ) 7.8, 1.5 Hz, 1H); MS (EI) m/z (rel
intensity) 234 (3, M+ + 2), 232 (9, M+), 197 (84, M+ - Cl), 106
(100). Anal. Calcd for C12H9ON2Cl: C, 61.95; H, 3.90; N, 12.04.
Found: C, 61.68; H, 3.65; N, 12.15.
N-(2-Br om op h en yl)-4-p yr id in eca r boxa m id e (1b): yield
2.20 g (79%); mp 132 °C; UV (λmax in acetonitrile), 266 nm (ꢀ
) 8.6 × 103 L/mol cm); IR (gas phase) 3425, 3079, 1715 cm-1;1H
NMR (200 MHz, CDCl3) δ 8.81 (dd, J ) 4.4, 1.7 Hz, 2H), 8.56
(br s, 1H), 8.44 (dd, J ) 8.2, 1.5 Hz, 1H), 7.75 (dd, J ) 4.4, 1.7
Hz, 2H), 7.58 (dd, J ) 8.0, 1.5 Hz, 1H), 7.37 (dt, J ) 7.8, 1.5
Hz, 1H), 7.05 (dt, J ) 7.8, 1.5 Hz, 1H); MS (EI) m/z (rel
intensity) 278 (5, M+ + 2), 276 (5, M+), 197 (100, M+ - Br).
Anal. Calcd for C12H9ON2Br: C, 52.01; H, 3.27; N, 10.11.
Found: C, 52.07; H, 2.92; N, 10.36.
1
(KBr) 1653 cm-1; H NMR (200 MHz, CDCl3) δ 9.16 (d, J )
2.1 Hz, 1H), 8.80 (dd, J ) 4.8, 1.5 Hz, 1H), 8.51-8.46 (m, 2H),
8.24 (dt, J ) 8.0, 2.0 Hz, 1H), 7.50-7.28 (m, 3H), 7.12 (dt, J )
7.8, 1.5 Hz, 1H); MS (EI) m/z (rel intensity) 234 (4, M+ + 2),
232 (11, M+), 197 (74), 106 (100). Anal. Calcd for C12H9ON2Cl:
C, 61.95; H, 3.90; N, 12.04. Found: C, 61.63; H, 3.60; N, 12.22.
N-(2-Br om op h en yl)-3-p yr id in eca r boxa m id e (2b): yield
2.15 g (78%); mp 90.5 °C; UV (λmax in acetonitrile), 264 nm
(ꢀ ) 9.6 × 103 L/mol cm); IR (gas phase) 3427, 3083, 1712 cm-1
;
IR (KBr) 1663 cm-1; 1H NMR (200 MHz, CDCl3) δ 9.17 (d, J )
1.9 Hz, 1H), 8.78 (dd, J ) 4.8, 1.6 Hz, 1H), 8.53 (br s, 1H),
8.43 (dd, J ) 8.2, 1.6 Hz, 1H), 8.23 (dt, J ) 2.0, 8.0 Hz, 1H),
7.57 (dd, J ) 7.9, 1.5 Hz, 1H), 7.48-7.31 (m, 2H), 7.03 (dt, J
) 7.8, 1.5 Hz, 1H); MS (EI) m/z (rel intensity) 278 (7, M+ + 2),
276 (7, M+), 197 (93), 106 (100). Anal. Calcd for C12H9ON2Br:
C, 52.01; H, 3.27; N, 10.11; Found: C, 51.96; H, 2.93; N, 10.23.
P r ep a r a tive P h otor ea ction . P h otor ea ction of N-2-
(Ch lor oph en yl)-4-pyr idin ecar boxam ide (1a). Gen er al P r o-
ced u r e. To a large (300 mL) quartz immersion well photolysis
unit with provision for circulation of nitrogen were added 210
mL of acetonitrile, 30 mL of 2 N aqueous NaOH, and 1 mmol
of N-2-(chlorophenyl)-4-pyridinecarboxamide (1a ). With nitro-
gen circulation, the solution was irradiated with a 150 W
mercury lamp (high pressure) at 100 V at 4 °C for 30 h. From
time to time solids (products) precipitated on the sleeve of the
photocell were removed for better passage of light. The
resulting two-phase mixture was separated, and the water
layer was extracted with ethyl acetate. The acetonitrile and
ethyl acetate portion were together evaporated and analyzed
by column chromatography (2.523 cm, silica gel 70-230/230-
400 mesh ) 3/1, THF/cyclohexane ) 1/9). The eluents, THF/
n-hexane ) 15/135, 40/160, and 75/175 (v/v), were used in the
order given. The column fractions corresponding to Rf 0.26 and
0.49 on TLC were collected and identified as the starting
material (38%) and 2-(4-pyridinyl)benzoxazole (3, 25%),
respectively. This compound (mp 130-131; lit. value9 130-
131 °C) was identified by its spectral properties (NMR, IR, MS)
as well as elemental analysis.
N-(2-Iodoph en yl)-4-pyr idin ecar boxam ide (1c): yield 4.99
g (77%); mp 122-123 °C; UV (λmax in acetonitrile), 267 nm (ꢀ
) 9.5 × 103 L/mol cm); IR (KBr) 3252, 3028, 1653 cm-1; H
1
2-(4-P yr id in yl)ben zoxa zole (3): UV (λmax in acetonitrile),
299.6 nm, 312 (sh); IR (gas phase) 3049, 1669 cm-1; IR (KBr)
NMR (200 MHz, CDCl3) δ 8.84 (d, J ) 6.0 Hz, 2H), 8.41 (dd,
J ) 8.2, 1.5 Hz, 1H), 8.30 (br s, 1H), 7.82 (dd, J ) 8.1, 1.5 Hz,
1H), 7.78 (dd, J ) 4.4, 1.5 Hz, 2H), 7.41 (dt, J ) 8.1, 1.5 Hz,
1H), 6.92 (dt, J ) 7.5, 1.5 Hz, 1H); MS (EI) m/z (rel intensity)
324 (4, M+), 197 (68, M+ - I). Anal. Calcd for C12H9ON2I: C,
44.47; H, 2.80; N, 8.64. Found: C, 44.39; H, 2.83; N, 8.59.
1
3045, 1613 cm-1; H NMR (200 MHz, CDCl3) δ 8.81 (dd, J )
4.5, 1.6 Hz, 2H), 8.07 (dd, J ) 4.5, 1.6 Hz, 2H), 7.84-7.80 (m,
1H), 7.64-7.60 (m, 1H), 7.46-7.36 (m, 2H); MS (EI) m/z (rel
intensity) 196 (100, M+). Anal. Calcd for C12H8ON: C, 73.46;
H, 4.11; N, 14.28. Found: C, 73.51; H, 4.08; N, 14.33.
P h otor ea ction of N-(2-Br om op h en yl)-4-p yr id in eca r -
boxa m id e (1b). The photoreaction of 1b was carried out for
8 h as in the case of 1a . The reaction mixture was separated
on silica gel column with THF/n-hexane ) 1/7. The product
was the same as the product from 1a .
N-P h en yl-4-pyr idin ecar boxam ide (1d): yield 3.16 g (80%);
mp 165 °C; UV (λmax in acetonitrile), 270 nm (ꢀ ) 1.0 × 104
L/mol cm); IR (KBr) 3340, 3043, 1655 cm-1; 1H NMR (200 MHz,
CD3OD) δ 8.71 (dd, J ) 4.5, 1.7 Hz, 2H), 7.87 (dd, J ) 4.5, 1.7
Hz, 2H), 7.69 (dt, J ) 7.2, 1.4 Hz, 2H), 7.41-7.31 (m, 2H),
7.16 (tt, J ) 7.4, 1.4 Hz, 1H); MS (EI) m/z (rel intensity) 198
(54, M+), 106 (100). Anal. Calcd for C12H10ON2: C, 72.71; H,
5.08; N, 14.13. Found: C, 72.77; H, 4.82; N, 14.16.
P h otor ea ction of N-(2-Ch lor op h en yl)-3-p yr id in eca r -
boxa m id e (2a ). The photoreaction of 2a was carried out for
N -(2-B r o m o p h e n y l)-N -m e t h y l-4-p y r id in e c a r b o x -
a m id e (1e). N-Methylation of N-(2-bromophenyl)-4-pyridin-
(18) J ohnstone, R. A. W.; Payling, D. W.; Thomas, C. J . Chem. Soc.
(C). 1969, 2223.

Synthesis of 2-Pyridinylbenzoxazole
J . Org. Chem., Vol. 64, No. 23, 1999 8555
21 h as in the case of 1a . The product separated by using
column chromatograph was 2-(3-pyridinyl)benzoxazole (4):
yield 37%; mp 110 (lit.19 108 °C); UV (λmax in acetonitrile), 299.4
nm, 311 (sh); IR (gas phase) 3087, 1619 cm-1; IR (KBr) 3058,
1613 cm-1;1H NMR (200 MHz, CDCl3) δ 9.47 (dd, J ) 2.1, 0.6
Hz, 1H), 8.75 (dd, J ) 4.8, 1.7 Hz, 1H), 8.49 (dt, J ) 8.0, 2.0
Hz, 1H), 7.83-7.74 (m, 1H), 7.65-7.55 (m, 1H), 7.49-7.33 (m,
3H); MS (EI) m/z (rel intensity) 196 (100, M+). Anal. Calcd for
Rela tive Ra te. Gen er a l P r oced u r e. N-(2-Bromophenyl)-
4-pyridinecarboxamide (1b , 17 mg) was dissolved in aceto-
nitrile/2 M NaOH (15 mL; 13/12, v/v), purged with nitrogen
(or oxygen), and irradiated with a Hg lamp (150 W, high
pressure) for 1.5 h. The reaction mixtures were analyzed by
GC (FID). The relative area of the peak (retention time, 7.3
min) on several conditions such as the presence of inhibitors
were compared and are shown in Table 3.
C
12H8ON: C, 73.46; H, 4.11; N, 14.28; Found. C, 73.38; H, 4.07;
N, 14.40.
P h otor ea ction of N-(2-Br om op h en yl)-3-p yr id in eca r -
UV Absor p tion . An acetonitrile solution of 1b (1.2 × 10-4
M, 2 mL) was placed in a UV cuvette, and the absorption
spectrum was taken by the UV spectrometer (dotted line in
Figure 1). The λmax of the absorption was 266 nm. Then, an
acetonitrile solution of 1b (1.2 × 10-4 M, 2 mL) containing
NaOH (1.2 × 10-4 M, 2 mL) was placed in the UV cuvette,
and the absorption spectrum was taken (thin solid line). The
spectrum was diffused into longer wavelength. When the
prototype absorption (dotted line) was subtracted from the
diffused absorption spectra, two bands peaking at 245 and 325
nm appeared. This behavior was not seen for 1b in methanol
containing NaOH. Furthermore, this behavior was not ob-
served for 1e, which has a methyl group on the nitrogen of
the amide bond in acetonitrile containing NaOH.
La ser F la sh P h otolysis. A detailed description of the
experimental setup can be found elsewhere.1 The fourth
harmonic (266 nm) output from a Q-switched Nd:YAG laser
(Spectron SL852G-30) was used as an excitation source. The
time duration of the excitation pulse was ca. 6 ns, and the
pulse energy was typically 30 or 80 mJ . A cw Xe lamp (Oriel
model 6259, 300 W) was used as a probe light source for
transient absorption measurement. The spectral resolution
was obtained by using a SP-275l monochromator (J AMS 27
model 82-49702) after the probe light passed through the
sample solution. A boxcar signal averager (Standard Research
system SR 250) was used in recording the transient signals.
The temporal profile of the transient absorption signal was
monitored by a 500 MHz digital storage oscilloscope (DSO,
Tektronix TDS620B). Sample solutions were prepared by
dissolving the reactants in acetonitrile with or without tet-
ramethylammonium chloride, and the concentration of the
solution was adjusted to be 0.5-1.5 in the absorbance at 266
nm. The experiment of the temporal profile with DSO was
performed in the same concentration of 1b in acetonitrile. The
sample solution was circulated from a bottle (3 L in volume)
to the fluorometer quartz cuvette of 10 mm in pathlength (flow
type, Helma QS 1.0) to reduce the effect of the accumulation
of the product and dissociation of the reactant in the photolysis
cell.
F lu or escen ce Mea su r em en t. Gen er a l P r oced u r e. The
fluorescence spectrum was obtained from the laser flash
photolysis of an acetonitrile solution of the pyridinecarboxa-
mide 2b; the flow-type UV cuvette, which is connected to the
1 L reservoir of the acetonitrile solution of 2b, was illuminated
by 266 nm laser light (30 or 80 mJ ), and the fluorescence
produced from the UV cuvette after 200 ns delay after the laser
flash was detected with a photomultiplier tube and quantified
with the appropriate electronic devices (thick solid line in
Figure 2). To study the effect of base on the excited states of
the pyridinecarboxamides, the fluorescences of 1 and 2 were
also measured in the presence of NaOH (thin solid line) or in
different concentrations of NaOH (Figures 2 and 3).
La ser In ten sity Dep en d en ce. From the argon-saturated
acetonitrile solution of 1b (40 mg in 1 L of acetonitrile)
containing 10 mL of 0.5 N NaOH, the absorption intensity at
393 nm, where a transient appeared, was monitored with the
boxcar averager at 10 µs delay after laser flash (266 nm). The
absorption intensity of the 393-nm transient versus laser
power were plotted and are shown in Figure 5.
boxa m id e (2b). The photoreaction was performed for 3.5 h
as in the case of 1a . The same product as that from 2a , 2-(3-
pyridinyl)benzoxazole, was obtained in 43% yield.
P h otor ea ction of N-(2-Br om op h en yl)-4-p yr id in eca r -
boxa m id e (1b) in Dilu te Na OH. The procedure is the same
as for 1a . The separation of the reaction mixture gave
2-pyridinylbenzoxazole (3, 29%) and N-phenyl-4-pyridinecar-
boxamide (1d , 8%). The idendification of 1d has been done by
comparison with the authentic sample.
P h otor ea ction of N-(2-Br om op h en yl)-3-p yr id in eca r -
boxa m id e (2b) in Wea k Ba se. The acetonitrile solution of
2b (2b, 278 mg; acetonitrile, 220 mL) containing 20 mL of
aqueous Na2CO3 (55 mg) solution was irradiated for 6.5 h as
in the case of 1a . The resulting mixture was analyzed on the
silica column with n-pentane/ethyl acetate [150/30, 80/20, 90/
30, 250/100, 140/160, 50/50, 0/200 (v/v) in the order given] as
eluents to yield 2-(3-pyridinyl)benzoxazole (4, 20%), N-phenyl-
3-pyridinecarboxamide (5, 27%), and starting material (11%).
2-(3-P yr id in yl)ben zoxa zole (4): yield 54 mg (20%); mp
110 °C (lit.19 108 °C).
N-P h en yl-3-pyr idin ecar boxam ide (5): yield 54 mg (27%);
mp 118.5-119 (lit. 119-120 °C20 and 116.8-117.2 °C21); IR
(gas phase) 3462, 3071, 1712 cm-1; IR (KBr) 3351, 3055, 1653
1
cm-1; H NMR (200 MHz, CDCl3) δ 9.03 (d, J ) 2.1 Hz, 1H),
8.96 (br s. 1H), 8.65 (dd, J ) 4.8, 1.4 Hz, 1H), 8.14 (dt, J )
1.9, 7.9 Hz, 1H), 7.61 (d, J ) 7.7, 2H), 7.35-7.27 (m, 3H), 7.14
(dt, J ) 0.8, 7.3 Hz, 1H); MS (EI) m/z (rel intensity) 198 (44,
M+), 197 (11, M+ - H), 106 (100). Anal. Calcd for C12H10ON2:
C, 72.71; H, 5.08; N, 14.13. Found: C, 72.86; H, 5.15; N, 14.32.
P h otor ea ction of N-(2-Br om op h en yl)-N-m eth yl-4-p y-
r id in eca r boxa m id e (1e). An acetonitrile solution of 1e (278
mg) was irradiated for 10 h as in the case of 1a . A separation
of the resulting reaction mixture on column chromatograph
with THF/n-hexane (70/280, 60/140) gave two products,
6-(methyl)benzo[c][2,6]naphthyridin-5-(6H)-one (50 mg, 25%)
and N-methyl-N-phenyl-4-pyridinecarboxamide (5%).
6-(Meth yl)ben zo[c][2,6]n a p h th yr id in -5-(6H)-on e: mp
1
89.5 °C; IR (gas phase) 3074, 2954, 1683 cm-1; H NMR (200
MHz, CDCl3) δ 9.64 (s, 1H), 8.79 (d, J ) 5.2 Hz, 1H), 8.35 (dd,
J ) 8.0, 1.1 Hz, 1H), 8.25 (d, J ) 5.2 Hz, 1H), 7.60 (dt, J )
1.3, 7.9, 1H), 7.45-7.33 (m, 2H), 3.79 (s, 3H); MS (EI) m/z (rel
intensity) 211 (16, M++H), 210 (100, M+). Anal. Calcd for
C
13H10ON2: C, 74.28; H, 4.79; N, 13.32. Found: C, 73.95; H,
4.81; N, 13.15.
Mea su r em en t of Qu a n tu m Yield . The procedure of light
intensity measurement was described in the preceding paper.2
An acetonitrile solution of N-(2-halophenyl)-4-pyridinecar-
boxamide (1) (1.0 × 10-3 M, 2.5 mL) containing 0.1 M NaOH
(50/1, v/v) was deaerated with nitrogen or oxygen for 30 min
and irradiated for 30 min with monochromatic light (266 ( 3
nm, Xe lamp, 450 W). The reaction mixture was analyzed by
GC. There were two peaks whose retention times were 7.3 and
11.8 min (in the case of 1b). The peak with retention time 7.3
was the 2-(4-pyridinyl)benzoxazole, and the peak with reten-
tion time 11.8 was the starting material. The quantitative
measurement was determined by standard curve of authentic
sample.
Tr a n sien t Absor p tion . The concentration (absorbance 1)
of the pyridinecarboxamide 1b (1d or 1e) was prepared by
dissolving the amide 1b, 1d , or 1e (about 40 mg) in acetonitrile
(1 L) containing 1 mL of 4 M NaOH. The solution was
circulated through a flow-type UV cuvette, which was con-
nected to a 2 L reservoir, and deaerated with argon. Transient
absorption spectra and their temporal profiles were obtained
at 10 µs and 1 ms delay after the laser flash (80 mJ ) of the
(19) Sener, E.; Turgut, H.; Yalcin, I.; O¨ ren, I.; Tu¨rker, L.; Celebi,
N.; Akin, A. Int. J . Pharm. 1994, 110, 109.
(20) Baxter, J . N.; Cymerman-Craig, J . J . Chem. Soc. 1953, 1490.
(21) Badgett, C. O.; Provost, R. C., J r.; Ogg, C. L.; Woodward, C. F.
J . Am. Chem. Soc. 1945, 67, 1135.

8556 J . Org. Chem., Vol. 64, No. 23, 1999
Park et al.
solution (Figures 4 and 6), with the appropriate electronic
devices such as the boxcar averager and a 500 MHz digital
storage oscilloscope.2
and reduced product formation was analyzed by GC (FID) as
described in relative rate studies. The reduced product was
produced, but the substituted product not formed (spectra not
shown).
Deu ter ia tion a n d P h otosen sitiza tion . N-(2-Bromophe-
nyl)pyridinecarboxamide (1b) was dissolved in CD3CN (2 mL,
1.8 × 10-3 M) containing 0.1 mL of D2O and irradiated with a
Xe lamp (450 W, 10 cm path of MeOH solution filter) for 40
min. The reaction mixture was analyzed by GC-MS. Deute-
riation content in the product was calculated by examining
the increment in mass of the molecular ion.22 Deuterium was
incorporated in 5% yield in 2-pyridinylbenzoxazole (spectra not
shown).
Ack n ow led gm en t. This research was supported by
the Basic Science Research Program, Ministry of Edu-
cation, 1999 (Project 1998-015-D00174, Y.-T.P.) and the
Research Project of Korea Research Institute of Stan-
dards and Science, 1999 (Project 8).
An acetonitrile solution (10 mL) of 1b (1.2 × 10-4 M) and
benzophenone (1.7 × 10-3 M) containing 0.1 mL of 1.0 M
NaOH was placed in a quartz cell (volume 10 mL), deaerated
with nitrogen, and irradiated with a monochromatic light (360
( 3 nm) from a Xe lamp (450 W) for 15 min. The substituted
Su p p or tin g In for m a tion Ava ila ble: Copies of UV ab-
sorption spectra of 1b in methanol and 1e in acetonitrile,
transient absorption spectra from the laser flash photolysis
of 1a , 1c, 1d , and 1e in acetonirile containing NaOH, and
temporal profile of a 370-nm transient from 1c in the presence
of oxygen. This material is available free of charge via the
Internet at http://pubs.acs.org.
(22) Budzikiewicz, H.; Djerassi, C.; Williams, D. H. Structure
Elucidation of Natural Products by Mass Spectrometry, Vol. 1: Alka-
loids; 1964; p 34.
J O9909498