Electron transfer versus proton transfer in excited states of bichromophoric aniline/olefin systems

Article abstract of DOI:10.1002/1099-0690(200207)2002:14<2317::AID-EJOC2317>3.0.CO;2-U

Photolysis of 2-allylaniline (1a) and trans-2-cinnamylaniline (2a) produced mainly the five- or the six-membered ring products 3a or 9a, respectively. Compound 1b, the N-acetyl derivative of la, preferentially underwent photo-Fries rearrangement of the anilide moiety, while - in contrast - the analogous compound 2b, derived from 2a, displayed competition between photocyclisation and double-bond isomerisation. The latter process, characteristic of the styrene chromophore, largely predominated in the case of 2c, the N-trifluoroacetyl derivative of 2a, while the allyl analogue 1c was essentially unreactive. The photochemical behaviour of the cis-cinnamyl compounds 7a and 7b was analogous to that of their trans isomers 2a and 2b, although double bond isomerisation occurred to a smaller extent. Thus, the introduction of electron-withdrawing acyl groups decreased photocyclisation. The nature of the excited states involved in the photochemistry of 1a-c, 2a-c and 7a and 7b was studied by fluorescence measurements. The most remarkable observation was the formation of intramolecular charge-transfer exciplexes in the cases of 1a, 2a, 2b, 7a and 7b. The exciplex bands of the cinnamyl compounds 2a, 2b, 7a and 7b in acetonitrile were considerably red-shifted (maxima at ca. 500 nm). A satisfactory correlation of the photochemical and photophysical data could be achieved by considering that photocyclisation took place when clear exciplex emission was observed. Alltogether, the above data strongly supported the involvement of an excited state electron-transfer mechanism in the photocyclisation of aniline/olefin bichromophoric systems. Wiley-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002.

Full text of DOI:10.1002/1099-0690(200207)2002:14<2317::AID-EJOC2317>3.0.CO;2-U

FULL PAPER
Electron Transfer versus Proton Transfer in Excited States of Bichromophoric
Aniline/Olefin Systems
Otman Benali,[a] Miguel A. Miranda,*[a] and Rosa Tormos[a]
Keywords: Electron transfer / Photochemistry / Exciplex / Photocyclisation / Fluorescence
Photolysis of 2-allylaniline (1a) and trans-2-cinnamylaniline
(2a) produced mainly the five- or the six-membered ring
products 3a or 9a, respectively. Compound 1b, the N-acetyl
photochemistry of 1a−c, 2a−c and 7a and 7b was studied by
fluorescence measurements. The most remarkable observa-
tion was the formation of intramolecular charge-transfer ex-
derivative of 1a, preferentially underwent photo-Fries re- ciplexes in the cases of 1a, 2a, 2b, 7a and 7b. The exciplex
arrangement of the anilide moiety, while − in contrast − the bands of the cinnamyl compounds 2a, 2b, 7a and 7b in ace-
analogous compound 2b, derived from 2a, displayed com-
tonitrile were considerably red-shifted (maxima at ca.
petition between photocyclisation and double-bond iso-
500 nm). A satisfactory correlation of the photochemical and
merisation. The latter process, characteristic of the styrene photophysical data could be achieved by considering that
chromophore, largely predominated in the case of 2c, the N-
trifluoroacetyl derivative of 2a, while the allyl analogue 1c
photocyclisation took place when clear exciplex emission
was observed. Alltogether, the above data strongly sup-
was essentially unreactive. The photochemical behaviour of ported the involvement of an excited state electron-transfer
the cis-cinnamyl compounds 7a and 7b was analogous to that
of their trans isomers 2a and 2b, although double bond iso-
merisation occurred to a smaller extent. Thus, the introduc-
mechanism in the photocyclisation of aniline/olefin bichrom-
ophoric systems.
tion of electron-withdrawing acyl groups decreased photo- ( Wiley-VCH Verlag GmbH, 69451 Weinheim, Germany,
cyclisation. The nature of the excited states involved in the
2002)
Introduction
The photochemistry of 2-allylanilines has attracted con-
siderable attention.[1Ϫ7] As in the case of the analogous 2-
allylphenols,[8Ϫ14] the main photoreaction is cyclisation to
give five-membered ring compounds. Interestingly, after pro-
motion to the excited singlet states, the process appears to
follow diverging mechanistic pathways for the two types of
substrates: proton transfer (PT) in the case of phenols and
electron transfer (ET) in the case of anilines (Scheme 1).
The involvement of the ET pathway in the photocyclis-
ation of 2-allylanilines has been proposed on the following
basis: a) the acidity of the NH group in the excited state
(pKa* Ͼ 12) is dramatically higher than that in the ground
state (pKa ϭ 17Ϫ22),[3,15,16] but is still insufficient to
achieve protonation of the allylic double bond, b) according
to the Weller equation,[17] the electron-transfer process
should be thermodynamically favourable,[3] and c) the high
regioselectivity towards the five-membered ring product
should be due to the distribution of the negative charge in
the intermediate radical ion pair.[2Ϫ4] Moreover, in the case
Scheme 1
of 2-allyl-N-methylaniline, the emission spectrum has been
tentatively assigned to an intramolecular charge-transfer ex-
ciplex.[4]
This mechanistic proposal is reasonable, although specu-
lative. Experimental support for it is very limited and con-
sists mainly of indirect evidence. The aim of this work was
thus to obtain solid experimental data for definite mechan-
istic assignment, in order to test the feasibility of the previ-
ous proposals. For this purpose, the parent 2-allylaniline
(1a) was acetylated to give 1b[18] and trifluoroacetylated to
give 1c. The analogous compounds 2b and 2c derived from
trans-2-cinnamylaniline (2a)[19] were also prepared.
[a]
´
´
´
Departamento de Quımica/Instituto de Tecnologıa Quımica
´
UPV-CSIC, Universidad Politecnica de Valencia,
Apartado 22012, 46071 Valencia, Spain
Fax: (internat.) ϩ 34-96/387-9349
E-mail: mmiranda@qim.upv.es
Eur. J. Org. Chem. 2002, 2317Ϫ2322  WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002 1434Ϫ193X/02/0714Ϫ2317 $ 20.00ϩ.50/0
2317

O. Benali, M. A. Miranda, R. Tormos
FULL PAPER
neither photocyclisation nor photo-Fries rearranged prod-
ucts being detected. In principle, some photorearrangement
had been anticipated; however, irradiation of the simplest
analogue, α,α,α-trifluoroacetanilide, did not result in the
formation of the corresponding ortho- or para-amino-α,α,α-
trifluoroacetophenone. The trend along the series 1aϪc is
very clear: introduction of the electron-withdrawing N-
acetyl group dramatically reduced the extent of photocyclis-
ation. This effect reached its maximum when three highly
electronegative fluorine atoms were attached to the α-car-
bonyl position. The PT mechanism hence did not come into
play even when the electron-donating ability of the aniline
moiety was substantially reduced and its acidity markedly
enhanced. Irradiation of the phenyl-substituted derivative
2a (Table 2, Entries 1Ϫ4) afforded the cis isomer 7, the in-
doline 8a,[27] the tetrahydroquinoline 9a[28] and the quinol-
ine 10.[29]
Substitution by electron-withdrawing groups at the nitro-
gen atom should result in enhanced acidity, along with de-
creased oxidizability, of the aromatic amine. This effect
should be more notable in the trifluoroacetyl derivative,
which would be expected to possess a pKa value of around
10[20,21] and a pKa* some 7 Ϯ 2 units lower.[3,15,16] This
could be enough to protonate the allylic moiety, thus fa-
vouring the PT over the ET pathway. On the other hand,
phenyl substitution of the double bond (as in 2aϪc) should
be interesting for two reasons: it could allow direct excita-
tion of the resulting styrene chromophore and might en-
hance the prospects for photocyclisation to six-membered
ring products, which are not typically formed by the ET
mechanism.[3]
As a matter of fact, it was shown that the only operating
photocyclisation mechanism within the series of com-
pounds studied was ET. Accordingly, the efficiency of the
reaction decreased in the order a Ͼ b ϾϾ c. Finally, al-
though the formation of intramolecular exciplexes has been
observed in a number of cases, no exciplexes involving
primary amine/olefin systems have, to the best of our know-
ledge, been reported previously.[4,22]
Scheme 2
Table 1. Photochemistry of compounds 1a, 1b and 1c
Results and Discussion
Product distribution (%)
Entry Sub- Condi- Conver- 1a
3a
4a
5
6
The results obtained with the allyl derivatives 1aϪb are
summarized in Scheme 2. Irradiation of an aerated benzene
solution of 1a[2,3] afforded a mixture of 2-methylindoline
(3a)[3] and 2-methylindole (4a)[23] (Table 1, Entry 1). This
confirmed the previously observed regioselectivity in the
five-membered ring compound,[2,3] based on the stabilis-
ation of spin and charge in the radical anion intermediate.
When the solution was purged with argon, the yield of 4a
showed a marked decrease (Table 1, Entry 2). It is thus reas-
onable to assume that the indole 4a was a secondary prod-
uct arising from irradiation of the indoline 3a in the pres-
ence of air. This was confirmed by a control experiment, in
which 3a was submitted to irradiation under the conditions
given in Table 1, Entries 1 and 2. The result was the forma-
tion of 4a as the only product, with a higher yield under
aerobic conditions (17%) than when the solution was
purged with argon (8%). In the case of the acetyl derivative
1b, the level of conversion was much lower and more com-
plex mixtures were obtained (Table 1, Entries 3 and 4). In
strate tions[a] sion
1
2
3
4
5
1a
1b
1c
A
B
A
B
A
59
62
29
34
Ϫ
Ϫ
Ϫ
3
4
Ϫ
79
94
Ϫ
5
21
6
3
traces
Ϫ
Ϫ
Ϫ
60
59
Ϫ
Ϫ
Ϫ
34
32
Ϫ
Ϫ
[a]
Irradiation for 50 min in benzene under air (A) or under argon
(B) with quartz-filtered light from a medium-pressure Hg lamp.
When the solution was purged with argon, the yield of
addition to 3a (under argon) or 4a (under air), the C-acyl- the last product once more showed a marked decrease.
ation products 5 and 6 were also obtained. As the major Again, control experiments showed that irradiation of 9a
photoproducts 5 and 6 still contained the intact allyl side under air resulted in aromatisation to 10, with trace
chain, photo-Fries rearrangement of the acetanilide amounts of an intermediate dihydroquinoline being detect-
moiety[24Ϫ26] appeared to be the dominant process. The last able by GC-MS.[30] As was to be expected, this process was
Entry in Table 1 shows that the trifluoroacetyl derivative 1c much less efficient when the solution was purged with ar-
was essentially unreactive under the irradiation conditions, gon. The 8a/9a ratio decreased with increasing solvent po-
2318
Eur. J. Org. Chem. 2002, 2317Ϫ2322

Excited States of Bichromophoric Aniline/Olefin Systems
FULL PAPER
the fluorescence spectra of these compounds were recorded
in both nonpolar (hexane) and polar (acetonitrile) solvents.
The results are summarized in Table 3, together with the
data corresponding to the reference compounds containing
the isolated chromophores. Thus, o-toluidine (Entries 1 and
2) in hexane displayed an emission spectrum with a max-
imum at 319 nm, which in acetonitrile was red-shifted to
331 nm. The N-acetyl derivative of o-toluidine had similar
spectra (Entries 3 and 4), but the fluorescence was substan-
tially weaker (the quantum yields dropping from ΦF ഠ 0.2
to ca. 10Ϫ3). In the case of 2,2,2-trifluoro-N-(2-methyl-
phenyl)acetamide there was essentially no emission; how-
ever, the residual fluorescence bands were found in the same
wavelength range (Entries 5 and 6). On the other hand, β-
methylstyrene (Entries 7 and 8) showed a relatively intense
emission band at 308 nm (ΦF ഠ 0.3), which was insensitive
to solvent effects.[22,31] Remarkably, the spectrum of 2-allyl-
aniline (1a) showed a maximum at 347 nm in hexane (Entry
9). Hence, introduction of the allyl substituent had resulted
in a 28 nm shift of the emission maximum.
Table 2. Photochemistry of compounds 2aϪc and 7a and 7b
Product distribution (%)
Entry Sub- Condi- Conver- 7 (or 2) 8a 9a 9b 10
strate tions[a] sion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
2a
2b
2c
7a
7b
A
B
C
D
A
B
89
90
93
91
78
70
89
91
63
61
50
59
96
91
100
98
67
86
87
82
8
8
3
11 67
24 61
Ϫ
Ϫ
Ϫ
Ϫ
43
14
7
27
8
6
2
2
3
Ϫ
68
83
6
6
45
48
24
22
100
100
100
100
2
6
Ϫ
Ϫ
16
3
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
12 38
C
Ϫ
55 21
24 47
D
A
B
7
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
15
5
58
13
8
C
D
A
B
10 73
18 71
C[b]
D[b]
A
B
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
42
87
14 62
37 54
26 59 15
39 55
6
C
D
Ϫ
2
A similar band was observed in acetonitrile, although the
shift associated with the introduction of the allyl substituent
was only 19 nm (Entry 10). The same type of effect had
already been observed in the case of the N-methyl analogue
of 1a and attributed to formation of an intramolecular exci-
plex.[4,22] Interestingly, exciplex formation appeared to occur
to a far lesser extent in the case of the N-acetyl derivative
1b, as shown by comparison of the data shown in Table 3,
Entries 3 and 11. The same was true for N-(2-allylphenyl)-
2,2,2-trifluoroacetamide (1c) (compare Entries 5 and 13), for
which no change associated with intramolecular exciplex
formation was detected. Parallel behaviour was observed for
the cinnamyl analogues 2aϪc. The aniline 2a was charac-
terised by a single exciplex band (340 nm) in hexane, or by
two bands (aniline singlet and exciplex) in acetonitrile
(Table 3, Entries 15 and 16). The longer-wavelength band,
at 505 nm, was remarkably red-shifted from its position in
hexane. The changes associated with exciplex formation are
shown in Figure 1, in which the fluorescence spectra of 2a
in hexane and in acetonitrile are compared with those of o-
toluidine as reference compound. N-Acetylation to give 2b
did not prevent formation of the intramolecular exciplex,
which appeared at 338 nm (hexane) or 500 nm (acetonitrile)
(Table 3, Entries 17 and 18). This contrasted with the beha-
viour of 1b, in which emission occurred mainly from the
anilide-like singlet. Again, the very weak residual emission
from the trifluoroacetyl derivative 2c was just a combination
of those of the isolated anilide and styrene chromophores
(Entries 19 and 20). The cis compounds 7a and 7b had
fluorescence spectra very similar to those of their trans iso-
mers 2a and 2b (Table 3, Entries 21Ϫ24), although the in-
tensities of the exciplex bands were lower. In general, the
emissions of 1aϪc, 2aϪc and 7a and 7b were very weak (ΦF
values as low as 10Ϫ2 to 10Ϫ3, or even less). This was espe-
cially the case for the acetyl and trifluoroacetyl derivatives,
as was to be expected in view of the trends observed for the
isolated o-toluidine chromophore (see above).
4
[a]
Irradiation for 40 min with quartz-filtered light from a medium-
pressure Hg lamp in benzene under air (A) or under argon (B) and
in acetonitrile under air (C) or under argon (D). [b] An acetonitrile
addition product was detected by GC-MS when 7a was irradiated
in acetonitrile.
larity. The main differences with respect to the results ob-
tained with the parent 2-allylaniline (1a) were the higher
level of conversion, the occurrence of trans to cis isomeris-
ation and the regioselectivity of the photocyclisation to-
wards the six-membered ring product. Clearly, conjugation
with the phenyl ring was strongly influencing the relative
stabilities of the benzylic radical and anion centres, thus
controlling regioselectivity. This could indicate a dominant
role for the styrene chromophore. As observed in the allyl
series, N-acetylation to give 2b resulted in a decrease in the
photocyclisation efficiency (Table 2, Entries 5Ϫ8). It was re-
markable that photo-Fries rearrangement of 2b, a photo-
chemical process characteristic of the anilide chromophore,
did not take place. This contrasted with the results obtained
with 1b and supported the above statement regarding the
key role of the styrene excited states. As a general comment,
the trifluoroacetyl derivative 2c did not undergo photo-
cyclisation, nor photorearrangement (Table 2, Entries
9Ϫ12). However, introduction of the electron-withdrawing
acyl groups did not prevent trans to cis isomerisation, which
increased in the case of 2b and became the only observed
process with 2c. Again, the photocyclisation efficiency de-
creased in the order a Ͼ b ϾϾ c, consistently once more
with the process occurring exclusively by the ET mechan-
ism. The photochemical behaviour of the cis-cinnamyl com-
pounds 7a and 7b (Table 2, Entries 13Ϫ20) was similar to
that of 2a and 2b, but with double-bond isomerisation oc-
curring to a smaller extent.
In order to ascertain the nature of the excited states in-
volved in the photoreactions of 1aϪc, 2aϪc and 7a and 7b,
Eur. J. Org. Chem. 2002, 2317Ϫ2322
2319

O. Benali, M. A. Miranda, R. Tormos
FULL PAPER
Table 3. Fluorescence data for the isolated chromophores and for 1aϪc and 2aϪc
Entry
Compound
Solvent[a]
λ (nm)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
o-toluidine
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
319
331
323
340
325
347
308
307
347
350
335
339
320
351
340
N-(2-methyphenyl)acetamide
2,2,2-trifluoro-N-(2-methyphenyl)acetamide
β-methylstyrene
1a
1b
1c
2a
2b
2c
7a
7b
348, 505
310 (sh), 338
313 (sh), 360, 500
310, 319
311, 319, 357 (sh)
338
345, 460Ϫ500
310 (sh), 333, 342
352, 460Ϫ500
[a]
Solvents used were hexane (A) or acetonitrile (B).
with the dominant formation of the photo-Fries rearrange-
ment products upon irradiation of 1b, dominated by the
‘‘isolated’’ anilide moiety. Isomerisation of the double bond
in the cinnamyl series is obviously related to the styrene
chromophore. In this context, it appears relevant that such
a reaction competes with photocyclisation, becoming the
major process in the case of 2c. Indeed, this compound,
which does not display exciplex emission, shows a weak
fluorescence maximum corresponding to the styrene chro-
mophore at 310 nm. Shoulders at this wavelength are also
observed in the spectra of 2b and 7b, in which double bond
isomerisation occurs as a competing process.
Experimental Section
General: IR spectra were obtained with a Nicolet FT-IR 720 or
with a Jasco FT/IR-460 Plus; ν˜max (cmϪ1) is given for selected ab-
sorption bands. H NMR spectra were measured in CDCl3 with a
300-MHz Varian Gemini 300, chemical shifts are reported in δ
(ppm) values, with TMS as internal standard. Mass spectra were
obtained by electron impact with a HewlettϪPackard 6980 series
instrument; the m/z values and the relative intensities (%) are indic-
ated for selected peaks. High-resolution mass spectra were con-
ducted with a VG Autospect instrument. Combustion analyses
Figure 1. Fluorescence spectra of o-toluidine in hexane (······) or
acetonitrile (———) and trans-2-cinnamylaniline (2a) in hexane
(- - - ) or acetonitrile (———)
1
Conclusion
Satisfactory correlation of the photophysical and photo-
chemical data can be achieved through the following inter-
pretation. Photocyclisation takes place to a significant ex-
tent when clear exciplex emission is observed. Exciplexes of
this type are known to have charge transfer natures,[22]
which agrees well with the involvement of an ET mechan-
ism, as stated above. This is compatible with the efficient
´
´
were performed at the Instituto de Tecnologıa Quımica of the CSIC
in Valencia. An Edinburgh Analytical Instruments Mod FS900 was
used to record the fluorescence spectra.
Preparation of the Substrates: Compounds 1a and 2a were prepared
by Claisen rearrangement of N-allylaniline and N-cinnamylaniline,
respectively, with ZnCl2, as described in the literature.[17]
Acetylation of 1a (1 g) or 2a (0.5 g) with acetyl chloride by stand-
photocyclisation of the exciplex-forming 2b, in comparison ard procedures afforded 1b (1.3 g) and 2b (0.6 g) in nearly quantit-
2320 Eur. J. Org. Chem. 2002, 2317Ϫ2322

Excited States of Bichromophoric Aniline/Olefin Systems
FULL PAPER
ative yields. The trifluoroacetyl derivatives 1c and 2c were obtained CH), 6.65Ϫ6.80 (m, 3 H, CHϭCHPh, 4,6-ArH), 7.22Ϫ7.38 (m, 7
by condensation of 1a (1 g) or 2a (0.5 g) with equimolar amounts
H, ArH) ppm. MS: m/z ϭ 209 (67) [Mϩ], 132 (21), 118 (100), 91
of trifluoroacetic acid and dicyclohexylcarbodiimide, with 4-(di-
(23), 77 (16). Exact mass calcd. for C15H15N 209.1205, found
methylamino)pyridine as catalyst and dichloromethane as solvent. 209.1214.
The yields of 1c and 2c were 75% (1.3 g) and 70% (0.5 g), respect-
ively.
cis-2-Cinnamylacetanilide (7b): FTIR: ν˜ ϭ 3263, 1662, 1529, 1297,
756 cmϪ1 1H NMR: δ ϭ 1.69 (s, 3 H, COCH3), 3.67 (d, J ϭ
.
7.6 Hz, 2 H, CH2), 5.70 (dt, J1 ϭ 11.5 Hz, J2 ϭ 7.6 Hz, 1 H,
CH2CHϭCH), 6.75 (d, J ϭ 11.5 Hz, 1 H, CHϭCHPh), 6.80 (br.
s, 1 H, NH), 7.10Ϫ7.41 (m, 8 H, ArH), 7.82 (d, J ϭ 7.6 Hz, 1 H,
6-ArH) ppm. MS: m/z ϭ 251 (47) [Mϩ], 208 (88), 160 (33), 130
(33), 118 (100), 91 (55). Exact mass calcd. for C17H17NO 251.1310,
found 251.1311.
Irradiation Procedure: Solutions of 1 or 2 (5 m⁾ in the appropriate
solvent were placed in quartz tubes surrounding a centrally posi-
tioned quartz cooling jacket containing a 125-W medium-pressure
Hg lamp and irradiated for the indicated times. The course of the
reaction was monitored by GC, GC-MS and 1H NMR; the degrees
of conversion and the product distributions were determined by
use of appropriate standards. Isolation and purification were per-
formed by conventional column chromatography on Merck silica
gel 60 (0.063Ϫ0.200 mm) with hexane/ethyl acetate or hexane/
dichloromethane as eluent, or by use of isocratic HPLC equipment
provided with a semipreparative Microporasil column, with hex-
ane/ethyl acetate as eluent. Compounds 1a, 1b, 3a, 4a, 2a, 8a, 9a,
9b and 10 are known;[3,18,19,23,27Ϫ29,32] their structures were assigned
by comparison with authentic samples.
N-(cis-2-Cinnamylphenyl)-2,2,2-trifluoroacetamide (7c): FTIR: ν˜ ϭ
3297, 1724, 1593, 1535, 1161 cmϪ1 1H NMR: δ ϭ 3.71 (d, J ϭ
.
7.0 Hz, 2 H, CH2), 5.68 (dt, J1 ϭ 11.4 Hz, J2 ϭ 7.0 Hz, 1 H,
CH2CHϭCH), 6.71 (d, J ϭ 11.4 Hz, 1 H, CHϭCHPh), 7.15Ϫ7.41
(m, 8 H, ArH), 7.79 (br. s, 1 H, NH), 7.90 (d, J ϭ 7.8 Hz, 1 H,
ArH) ppm. MS: m/z ϭ 305 (43) [Mϩ], 236 (15), 214 (100), 208
(33), 132 (31), 115 (40), 91 (46). Exact mass calcd. for C17H14F3NO
305.1027, found 305.1017.
Spectroscopic Data of the New Compounds
N-(2-Allylphenyl)-2,2,2-trifluoroacetamide (1c): FTIR: ν˜ ϭ 3282,
1708, 1542, 1168 cmϪ1. H NMR: δ ϭ 3.41 (d, J ϭ 7.8 Hz, 2 H,
1
Acknowledgments
CH2), 5.11Ϫ5.25 (m, 2 H, CHϭCH2), 5.85Ϫ6.00 (m, 1 H, CHϭ
CH2), 7.10Ϫ7.40 (m, 3 H, ArH), 7.84 (d, J ϭ 7.9 Hz, 1 H, 6-ArH),
8.20 (s, 1 H, NH) ppm. MS: m/z ϭ 229 (54) [Mϩ], 160 (79), 132
(100), 117 (65). C11H10F3NO (229.07): calcd. C 57.26, H 4.72, N
6.18; found C 57.64, H 4.40, N 6.11.
Financial support by the Spanish MCYT (Grant No. BQU 2001-
2725) and the Generalitat Valenciana (Grant GV01-272) is grate-
fully acknowledged.
trans-2-Cinnamylacetanilide (2b): FTIR: ν˜ ϭ 3263, 1660, 1585,
[1]
1
1525, 1445, 1296, 754 cmϪ1. H NMR: δ ϭ 2.13 (s, 3 H, COCH3),
N. Paillous, A. Lattes, Tetrahedron 1971, 52, 4945Ϫ4948.
[2]
U. Koch-Pomeranz, H. J. Hansen, Helv. Chim. Acta 1975, 58,
3.55 (d, J ϭ 6.0 Hz, 2 H, CH2), 6.30Ϫ6.52 (m, 2 H, CHϭCH),
7.10Ϫ7.34 (m, 8 H, ArH), 7.82 (d, J ϭ 7.6 Hz, 1 H, 6-ArH) ppm.
MS: m/z ϭ 251 (45) [Mϩ], 208 (86), 160 (32), 130 (33), 118 (100), 91
(57). Exact mass calcd. for C17H17NO 251.1310, found 251.1313.
178Ϫ182.
[3]
U. Koch-Pomeranz, H. J. Hansen, Helv. Chim. Acta 1977, 60,
768Ϫ797.
S. Jolidon, H. J. Hansen, Helv. Chim. Acta 1979, 62,
2581Ϫ2611.
[4]
N-(trans-2-Cinnamylphenyl)-2,2,2-trifluoroacetamide (2c): FTIR:
[5]
B. Scholl, H. J. Hansen, Helv. Chim. Acta 1980, 63, 1823Ϫ1832.
1
ν˜ ϭ 3298, 1724, 1597, 1535, 1160 cmϪ1. H NMR: δ ϭ 3.52Ϫ3.60
[6]
B. Scholl, H. J. Hansen, Chimia 1981, 35, 49Ϫ51.
(m, 2 H, CH2), 6.20Ϫ6.65 (m, 2 H, CHϭCH), 7.10Ϫ7.41 (m, 8 H,
ArH), 7.90 (d, J ϭ 7.8 Hz, 1 H, 6-ArH), 8.15 (s, 1 H, NH) ppm.
MS: m/z ϭ 305 (58) [Mϩ], 236 (20), 214 (100), 208 (35), 132 (33),
115 (40), 91 (48). Exact mass calcd. for C17H14F3NO 305.1027,
found 305.1020.
[7]
B. Scholl, S. Jolidon, H. J. Hansen, Helv. Chim. Acta 1986,
69, 184Ϫ194.
[8]
´
G. Frater, H. Schmid, Helv. Chim. Acta 1967, 50, 255Ϫ262.
[9]
W. M. Horspool, P. L. Pauson, J. Chem. Soc., Chem. Commun.
1967, 195Ϫ196.
[10]
[11]
A. Shani, R. Mechoulam, Tetrahedron 1971, 27, 601Ϫ606.
Y. L. Chow, X. -M. Zhou, T. J. Gaitan, Z. -Z. Wu, J. Am.
Chem. Soc. 1989, 111, 3813Ϫ3818.
3-Allyl-2-aminoacetophenone (5): FTIR: ν˜ ϭ 3493, 3326, 1648,
1609, 1560, 1428, 1248 cmϪ1 1H NMR: δ ϭ 2.59 (s, 3 H, CH3),
.
[12]
[13]
[14]
3.30 (d, J ϭ 6.0 Hz, 2 H, CH2), 5.11 (m, 2 H, CHϭCH2),
5.85Ϫ6.00 (m, 1 H, CHϭCH2), 6.49 (s, 2 H, NH2), 6.64 (t, J ϭ
7.3 Hz, 1 H, 5-ArH), 7.21 (d, J ϭ 7.1 Hz, 1 H, 4-ArH), 7.67 (d,
J ϭ 7.1 Hz, 1 H, 6-ArH) ppm. MS: m/z ϭ 175 (100) [Mϩ], 160
(61), 132 (56), 117 (29). Exact mass calcd. for C11H13NO 175.0997,
found 175.0996.
M. A. Miranda, R. Tormos, J. Org. Chem. 1993, 58,
3304Ϫ3307.
´
´
M. C. Jimenez, F. Marquez, M. A. Miranda, R. Tormos, J.
Org. Chem. 1994, 59, 197Ϫ202.
´
S. Gil, M. C. Jimenez, M. A. Miranda, R. Tormos, Eur. J. Org.
Chem., in press.
[15]
[16]
Th. Förster, Z. Elektrochem. 1950, 54, 531Ϫ535.
D. D. Roserbrook, W. W. Brandt, J. Phys. Chem. 1966, 70,
3857Ϫ3862.
D. Rehm, A. Weller, Ber. Bunsenges. Phys. Chem. 1969, 73,
834Ϫ839.
L. S. Hegedus, G. F. Allen, J. Bozell, E. L. Waterman, J. Am.
Chem. Soc. 1978, 100, 5800Ϫ5807.
3-Allyl-4-aminoacetophenone (6): FTIR: ν˜ ϭ 3467, 3359, 3237,
1665, 1593, 1357 cmϪ1. 1H NMR: δ ϭ 2.50 (s, 3 H, COCH3), 3.31
(d, J ϭ 6.1 Hz, 2 H, CH2), 4.22 (s, 2 H, NH2), 5.10 (m, 2 H, CHϭ
CH2), 5.86Ϫ6.00 (m, 1 H, CHϭCH2), 6.65 (d, J ϭ 8.8 Hz, 1 H, 5-
ArH), 7.70 (m, 2 H, 2,6-ArH) ppm. MS: m/z ϭ 175 (100) [Mϩ], 160
(72), 148 (6), 132 (12), 117 (17). Exact mass calcd. for C11H13NO
175.0997, found 175.0996.
[17]
[18]
[19]
[20]
[21]
[22]
C. D. Hurd, W. W. Jenkins, J. Org. Chem. 1957, 22, 1418Ϫ1423.
P. M. Mader, J. Am. Chem. Soc. 1965, 87, 3191Ϫ3195.
U. Meresaar, Acta Chem. Scand. 1974, 28, 656Ϫ660.
F. D. Lewis, J. M. Wagner-Brennan, A. M. Miller, Can. J.
Chem. 1999, 77, 595Ϫ604.
cis-2-Cinnamylaniline (7a): FTIR: ν˜ ϭ 3456, 3375, 1620, 1497
1
cmϪ1. H NMR: δ ϭ 3.54 (d, J ϭ 7.4 Hz, 2 H, CH2), 3.90 (br. s,
2 H, NH2), 5.77 (dt, J1 ϭ 11.5 Hz, J2 ϭ 7.4 Hz, 1 H, CH2CHϭ
L. J. Dolby, M. R. Rodia, J. Org. Chem. 1970, 35, 1493Ϫ1496.
[23]
Eur. J. Org. Chem. 2002, 2317Ϫ2322
2321

O. Benali, M. A. Miranda, R. Tormos
FULL PAPER
[24]
[30]
V. I. Stenberg, Org. Photochem. 1967, 1, 127Ϫ153.
S. W. Goldstein, P. J. Dambek, Synthesis 1989, 3, 221Ϫ222.
The m/z ratios and the relative intensities (%) for 2-phenyl-1,2-
dihydroquinoline were: 207 (100), 206 (71), 178 (11), 130 (67),
118 (7), 102 (12), 77 (10).
F. D. Lewis, D. M. Bassani, R. A. Caldwell, D. J. Unett, J. Am.
Chem. Soc. 1994, 116, 10477Ϫ10485.
[25]
´
D. Bellus, Adv. Photochem. 1971, 8, 109Ϫ159.
[26]
M. A. Miranda, Photo-Fries Reactions and Related Processes,
in Handbook of Organic Photochemistry and Photobiology (Ed.:
W. M. Horspool, P. S. Song), CRC Press, Boca Raton, 1995,
p. 570Ϫ578.
[31]
[32]
[27]
[28]
[29]
H. Ahlbrecht, C. Schmitt, Synthesis 1994, 9, 983Ϫ988.
C. J. Moody, M. R. Pitts, Synlett 1998, 1029Ϫ1030.
M. R. Pitts, J. R. Harrison, C. J. Moody, J. Chem. Soc., Perkin
Trans. 1 2001, 955Ϫ977.
M. Protiva, V. Rericha, M. Borovicka, J. O. Jilek, Coll. Czech.
Chem. Commun. 1950, 15, 532Ϫ540.
Received January 24, 2002
[O02035]
2322
Eur. J. Org. Chem. 2002, 2317Ϫ2322