Photochemistry of 9-fluorenone oxime phenylglyoxylate: A combined TRIR, TREPR and ab initio study

Article abstract of DOI:10.1002/poc.715

The photochemistry of 9-fluorenone oxime phenylglyoxylate (1) in tetrachloromethane was investigated by a variety of time-resolved methods [STEP-SCAN TRIR and time-resolved EPR (TREPR)]. Photolysis of 1 yields the benzoyl radical, carbon dioxide and the f

Full text of DOI:10.1002/poc.715

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2004; 17: 207–214
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/poc.715
Photochemistry of 9-fluorenone oxime phenylglyoxylate:
a combined TRIR, TREPR and ab initio study
Christoph Kolano, Go¨ tz Bucher,* Hans Henning Wenk, Martin Ja¨ger, Olaf Schade and Wolfram Sander
Lehrstuhl fu¨ r Organische Chemie II, Ruhr-Universita¨t Bochum, Universita¨tsstrasse 150, 44801 Bochum, Germany
Received 5 June 2003; revised 2 September 2003; accepted 8 September 2003
ABSTRACT: The photochemistry of 9-fluorenone oxime phenylglyoxylate (1) in tetrachloromethane was investi-
epoc
gated by a variety of time-resolved methods [STEP-SCAN TRIR and time-resolved EPR (TREPR)]. Photolysis of 1
yields the benzoyl radical, carbon dioxide and the fluorene-9-iminyl radical on pulsed irradiation (ꢀ ¼ 355 nm). All
radicals have a lifetime in the range of microseconds and can be detected within the rise time of our time-resolved
equipment before undergoing recombination or reactions with the solvent. The benzoyl radical shows a strong
absorption (ꢁ~C==O ¼ 1824 cmꢀ1) in the IR spectrum. Upon purging the solution with oxygen, the benzoyl radical is
quenched to yield the benzoylperoxy radical (ꢁ~C==O ¼ 1814 cmꢀ1), which has a lifetime that is several microseconds
longer than that of the benzoyl radical. Owing to the lack of a good IR chromophore, the fluorene-9-iminyl radical is
not detected by IR spectroscopy. It can, however, be detected by TREPR spectroscopy, which allows the detection of a
1 : 1 : 1 triplet of the fluorene-9-iminyl radical at 3456 G and a 1 : 2 : 1 triplet of the benzoyl radical at 3460 G.
Copyright # 2004 John Wiley & Sons, Ltd.
Additional material for this paper is available in Wiley Interscience.
KEYWORDS: infrared spectroscopy; STEP-SCAN; TREPR; laser flash photolysis; radicals; peroxides; density functional
theory
INTRODUCTION
dispersive IR spectrometer.4,6a The reaction of the ben-
zoyl radical with molecular oxygen has been reported to
yield the benzoylperoxy radical, PhC(O)OO.4 Peroxy
radicals, and possibly their tetroxide dimers, play an
important role in atmospheric chemistry.6b The self-reac-
tion of gas-phase acylperoxy radicals, such as the acet-
ylperoxy radical,6c has been reported to lead to alkyl
radicals via acyloxy radicals. We have undertaken a study
using different time-resolved methods and a variety of
theoretical methods to investigate the photochemistry of
9-fluorenone oxime phenylglyoxylate (1) and to demon-
strate the potential of this compound to serve as a source
for the generation of acyl radicals, especially the benzoyl
radical and the benzoylperoxy radical.
It has recently been shown by our group1 that fluorenone
oxime oxalates are suitable precursors for the generation
of alkoxycarbonyl radicals. They are synthesized easily
and show excellent thermal stability. Irradiation with
ꢀ ¼ 355 nm yields carbon dioxide, the fluorene-9-iminyl
radical and an alkoxycarbonyl radical in a clean reaction.
We were able to show that the STEP-SCAN technique is a
powerful tool for the investigation of this photoreaction.
The technique allows the detection of a variety of different
reactive acyl type intermediates with different stabilities
depending on the substituents. One of the most interesting
acyl radicals is the benzoyl radical, which plays an
important role in the initiation process of polymerizations
[e.g. photopolymerization of methyl methacrylate
(MMA)].2,3 Avariety of precursors have been synthesized
and checked by laser flash photolysis to be suitable
sources for benzoyl radicals.4,5 The most common method
to generate benzoyl radicals photochemically consists of
the irradiation of ketones, which undergo ꢂ-cleavage
(Norrish type I reaction). Such reactions have been
monitored by time-resolved IR spectroscopy using a
RESULTS AND DISCUSSION
Laser flash photolysis (LFP) of 1 in tetrachloromethane
(4.0 mM, Ar purge) at ꢀex ¼ 355 nm resulted in bleaching
of the precursor bands at ꢁ~C=O ¼ 1768 and 1691 cmꢀ1 and
the formation of a transient at ꢁ~C=O ¼ 1824 cmꢀ1 (Fig. 1).
Additionally, the strong ꢁas of CO2 at ꢁ~C=O ¼ 2336 cmꢀ1
was observed. Both the 1824 cmꢀ1 transient and ꢁas (CO2)
built up during the rise time (25 ns) of the STEP-SCAN
set-up. Furthermore, an IR band growing in at
ꢁ~C=O ¼ 1779 cmꢀ1 was detected. When the solution was
purged with oxygen prior to LFP, again the disappearance
*Correspondence to: G. Bucher, Lehrstuhl fu¨r Organische Chemie II,
¨ ¨
Ruhr-Universitat Bochum, Universitatsstrasse 150, 44801 Bochum,
Germany.
E-mail: goetz.bucher@ruhr-uni-bochum.de
Contract/grant sponsors: Deutsche Forschungsgemeinschaft; Fonds
der Chemischen Industrie.
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 207–214

=O
208
C. KOLANO ET AL.
Figure 1. Time-resolved IR difference spectra (time resolution 100 ns, spectral resolution 6 cmꢀ1) showing the photochemistry
of 1 in CCl4 purged with argon. Bands appearing on irradiation (ꢀ ¼ 355 nm) are pointing upwards; bands disappearing are
pointing downwards. Solid line, spectrum 100 ns after irradiation; dotted line, spectrum after 20 ms
of two precursor bands and the immediate formation of
the CO2 absorption at ꢁ~C=O ¼ 2335.8 cmꢀ1 were noted. A
transient species was observed at ꢁ~C=O ¼ 1814 cmꢀ1 de-
caying with a lifetime of 5.2 ꢁ 0.2 ms. With oxygen purge,
the IR absorption due to CO2 at ꢁ~C=O ¼ 2336 cmꢀ1 under-
goes a secondary growth. The lifetime of this secondary
growth (ꢃ ¼ 6.1 ꢁ 0.5 ms) does not change when the oxy-
gen concentration is varied.
The position of ꢁC==O of the benzoyl radical had been
reported before.4,5 A calculated vibrational spectrum
[UB3LYP/6–311 þ G(d,p)] and the literature value are
in good agreement with our experimental data. Decar-
boxylation of the intermediate benzoylcarbonyloxy radi-
cal 2 causes the intense initial absorption at ꢁ~C=O
(CO2)as ¼ 2335.8 cmꢀ1 The benzoylcarbonyloxy radical
2 itself has such a short lifetime that it cannot be
monitored with our set-up.
Strong IR absorptions near 1800 cmꢀ1 are character-
istic of acyl radicals.4,6a Thus, the benzoyl radical 3 is a
likely assignment for the ꢁ~C==O ¼ 1824 cmꢀ1 transient.
The first step in the reaction sequence would be cleavage
of the N—O bond of the fluorenone oxime part of
the molecule, followed by rapid decarboxylation of the
A plot of the band integrals versus the time yields
the lifetime of 3 in CCl4. The measured lifetime of
ꢃ ¼ 1.0 ꢁ 0.05 ms is in good agreement with published
experimental data.6a Quenching of the benzoyl radical 3
3
with O2 leads to the formation of the benzoylperoxy
radical 4, which has a characteristic absorption at
3
benzoylcarbonyloxy radical 2. In presence of O2 the
benzoyl radical 3 is effectively trapped and the benzoyl-
peroxy radical 4 is formed (Scheme 1).
ꢁ~C ¼ 1814 cmꢀ1, decaying with a measured lifetime
of ꢃ ¼ 5.2 ꢁ 0.2 ms (Fig. 2).
Scheme 1
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 207–214

PHOTOCHEMISTRY OF 9-FLUORENONE OXIME PHENYLGLYOXYLATE
209
demonstrates that both the benzoyl radical 3 and the
iminyl radical 5 are formed very rapidly upon LFP of 1.
Our results presented so far agree well with what has
previously been known about the chemistry of the ben-
zoyl radical. Noteworthy, however, is the secondary
build-up of IR absorption due to carbon dioxide. The
formation of CO2 from benzoylperoxy radicals 4 had
been observed before by Hancock-Chen and Scaiano,
who could follow this reaction only indirectly by mon-
itoring the formation of phenylperoxy radicals.14 Their
reaction mechanism involves the dimerization of 4 to a
highly labile dibenzoyl tetroxide (6), which subse-
quently decays yielding 2 equiv. of benzoyloxy radicals
(7). The latter rapidly decarboxylate, yielding phenyl
radicals (8), which in turn react with oxygen to yield
phenylperoxy radicals (9) as observed by Hancock-Chen
and Scaiano. The reaction mechanism is shown in
Scheme 2.
Figure 2. A three-dimensional representation of the time-
resolved difference STEP-SCAN FTIR spectrum of 1 in
oxygen-saturated CCl4. The transient absorption is observed
at 1814 cmꢀ1. Carbon dioxide is observed at 2336 cmꢀ1
,
Calculations [B3LYP/6–311 þ G(d,p)] on dibenzoyl
—
whereas the two disappearing bands at 1768 and
tetroxide (6) predict two similarly intense C O bands
—
1691 cmꢀ 1 belong to the starting material
at ꢁ~C==O ¼ 1792 and 1823 cmꢀ1. In our TRIR experi-
ments, no transient double peak was observed in this
region of the spectrum. Hence our experiments do not
allow us to conclude whether 6 is indeed involved in the
decay of 4.
In a further quenching experiment we used bromotri-
chloromethane as an effective scavenger for the benzoyl
radical 3. The lifetime of 3 in the presence of 3 mM
CCl3Br is ꢃ ¼ 460 ꢁ 50 ns. From this value and the life-
time of 3 in neat CCl4, the rate constant for the reaction
of 3 with BrCCl3 can be estimated to be k ¼
3.7 ꢂ 108 l molꢀ1 sꢀ1, which is higher than the published
value of k ¼ (2.2 ꢁ 0.1) ꢂ 108 l molꢀ 1 sꢀ 1.4 Concomitant
with the decay of 3, we observed a growth at
ꢁ~C==O ¼ 1780 cmꢀ1 (ꢃ ¼ 450 ꢁ 50 ns), which is assigned
to the product, benzoyl bromide. Evaluation of the
growth kinetics of benzoyl bromide gave a similar value
of k ¼ 3.8 ꢂ 108 l molꢀ1 sꢀ1. Experimental data and cal-
culated IR modes are summarized in Table 1.
The photochemistry of 1 was also investigated by
time-resolved EPR (TREPR) spectroscopy. LFP
(ꢀex ¼ 355 nm) of a 5 mM solution of 1 in CCl4 (1 atm
Ar) led to the detection of two sets of transient signals,
both of which evolve during the laser pulse. The 1:2:1
triplet centred at 3460.3 G (Jmeta ¼ 1.3 G) is attributed to
the benzoyl radical,7–12 whereas the 1:1:1 triplet centred
at 3455.7 G (JN ¼ 9.7 G) is assigned to the 9-fluorenonei-
minyl radical 5 (Fig. 3).13 Hence TREPR clearly
Molecules containing several heteroatoms connected
in a chain are notoriously difficult to describe by theore-
tical methods.15a,b Owing to the size of the molecule,
calculations on 6 and its reactions were limited to
standard density functional theory (DFT). In order to
check for the reliability of the method chosen, we also
studied the model system diformyl tetroxide (10), using
both DFT and MP4(SDQ) ab initio theory. The structure
of 10, optimized at the MP4(SDQ)/6–31 þ G(d) level of
theory, consists of a chain of four oxygen atoms with an
O—O—O—O dihedral angle of 82.2 ꢃ. The formyl
groups are oriented such that the vectors of the carbonyl
dipole moments point in nearly opposite directions. As a
result the two formyloxy groups show opposing confor-
mations of the C—O bonds (one Z and one E).
The results of our calculations on the formation and
reactions of 10 are summarized in Scheme 3. Formation
of 10 from (Z )-formylperoxy [(Z )-11] and (E)-formyl-
peroxy [(E )-11] is calculated to be exothermic by
16–19 kcal molꢀ1 (1 kcal ¼ 4.184 kJ). The reaction is
—
Table 1. Lifetime, experimental and calculated C O stretching vibrations of the transient species
—
Species
Lifetime ꢃa (ms)
Lifetime ꢃ(ms)
ꢁC==O expa (cmꢀ1
)
ꢁC==O exp (cmꢀ1
)
ꢁC==O calcdb
2
3
—
1.0 ꢁ 0.02
—
—
1824
—
1733
1882
1.76
18286
18185
18206
—
4
CO2
5.2 ꢁ 0.2
—
—
1814
2336
1870
—
—
a
In CCl4, ambient temperature.
UB3LYP/6–311þg(d,p), unscaled.
b
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 207–214

210
C. KOLANO ET AL.
thus predicted to be more exothermic than the dimeriza-
tion of alkylperoxy radicals such as methylperoxy,
for which ꢀH of the dimerization reaction had
been calculated by DFT (B3LYP/6–31G**) to be
ꢀ9.34 kcal molꢀ1 15c
.
Cleavage of 10 can also occur at
the O1—O2 bond, which yields the formyloxy radical 12
and the formyltrioxy radical 13. Trioxy radical 13 is only
found to be a minimum when MP4(SDQ) theory is used,
whereas attempted optimization using B3LYP theory
results in fragmentation into 12 and oxygen. Overall,
the fragmentation of 10 into 2 equiv. of 12 þ O2 (3ꢁgꢀ) is
predicted to be weakly endothermic [MP4(SDQ)] or
exothermic (B3LYP). Direct oxygen transfer from (Z)-
11 to another molecule of (Z)-11, yielding a (13 þ 12)
triplet radical pair, can also possibly occur. On the triplet
surface, this reaction is predicted to have a significant
activation enthalpy of ꢀH # ¼ 35.1 kcal molꢀ1 (B3LYP).
The results of our calculations on 6 and its formation
and fragmentation reactions are summarized in Scheme 4.
Figure 3. Time-resolved EPR spectrum of the doublet
transient species formed upon LFP (ꢀex ¼ 355 nm) of 1 in
CCl4
Scheme 2
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 207–214

PHOTOCHEMISTRY OF 9-FLUORENONE OXIME PHENYLGLYOXYLATE
211
Scheme 3
Owing to the size of the molecule, MP4(SDQ) ab initio
theory could no longer be applied here, and the calcula-
tions are limited to DFT (B3LYP).
Dibenzoyl tetroxide (6) shows a helical structure, as
does diformyl tetroxide (10). Unlike in 10, the dihedral
of ꢀH # ¼ 34.6 kcal molꢀ1. The TS for O-transfer from
(Z)-4 to (Z)-4 is calculated to have an rO2—O3 of the
newly formed benzoyltrioxy fragment of 137 pm, which
is shorter than rO2—O3 of 6. This indicates that along the
reaction coordinate of the O-transfer reaction the disso-
ciation of 14 does not proceed to any significant degree.
Hence the TS for O-transfer from (Z)-4 to (Z)-4 is not
able to profit from the energy gain connected with the
fragmentation of 14 into 7 plus O2 (3ꢁgꢀ). As the same
argument probably applies for the fragmentation of 6
into 7 plus 14, this reaction can also be expected to
have a significant activation enthalpy of the order of
25 kcal molꢀ1 (cf. Scheme 3, MP4 results). It is noted that
we have been unable to locate a transition structure for a
direct fragmentation of 6 (10) into 2 equiv. of 7 (12) plus
O2 (3ꢁgꢀ), in which both O—O bonds would be broken
simultaneously. The results of our calculations on the
system 6–4 are summarized in Fig. 4.
—
angle DOCꢄ ꢄ ꢄCO which is defined by the two C O bonds
—
is no longer close to 180 ꢃ {6, DOCꢄ ꢄ ꢄCO ¼ 40.2 ꢃ; 10,
DOCꢄ ꢄ ꢄCO ¼ 167.3 ꢃ [MP4(SDQ)]}. Hence, in 6, the or-
ientation of the vectors of the dipole moments of the two
carbonyl moieties is closer to parallel than to antiparallel.
The chemistry of 6 should be similar to that of 10. The
fragmentation of 6 into 2 equiv. of benzoyloxy radical 7
plus O2 (3ꢁgꢀ) is predicted to be more exothermic
[B3LYP/6–311 þ G(d,p): ꢀH ¼ ꢀ20.2 kcal molꢀ1] than
the corresponding fragmentation of 10. This is probably
due to stabilization of 7 by the phenyl substituent.
However, the oxygen transfer reaction from (Z)-4 to
another molecule of (Z)-4, yielding 7 plus 14, which
should then decay into O2 (3ꢁgꢀ) plus a second molecule
of 7, is predicted to have a significant activation enthalpy
Bimolecular self-quenching of two molecules of ben-
zoylperoxy radicals 4 could thus give rise to tetroxide 6 in
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 207–214

212
C. KOLANO ET AL.
Scheme 4
a mildly exothermic, yet entropically unfavourable reac-
tion. For the dissociation of the related di-tert-butyl
tetroxide into two molecules of tert-butylperoxyl, Hon-
eywill and Mile16 recently reported ꢀH ¼ 7.9 kcal molꢀ1
and ꢀS ¼ 33.3 cal molꢀ1 Kꢀ1. If we assume a similar
reaction entropy for the cleavage of 6, a ꢀG of around
0 kcal molꢀ1 results for the dissociation of 6 at ambient
temperature. Hence the room temperature dimerization of
4 could (but does not have to) play a role in the chemistry
of 4. If the reaction pathway leading from 4 to 7 is to run
via tetroxide 6, the reaction of 6 leading to 7 has to
compete with the re-dissociation of 6 into 2 equiv. of 4. In
our calculations, however, we have been unable to find
such a monomolecular reaction of 6. Bimolecular reac-
tions of 6 therefore cannot be excluded. Direct oxygen
transfer from 4 to 4, yielding 2 equiv. of 7 plus O2
(3ꢁgꢀ), represents an alternative mechanism, which
does not involve 6. This reaction, however, is predicted to
be impeded by a significant activation enthalpy.
CONCLUSION
We have shown by two independent time-resolved meth-
ods that 9-fluorenone oxime phenylglyoxylate is a very
good precursor for the generation of benzoyl radicals.
Photolysis of 1 leads in good yields and in a clean
photoreaction to the benzoyl radical 3, which plays an
important role in a variety of synthetic applications, such
as photochemical-initiated polymerization reactions. By
the combined use of time-resolved FTIR spectroscopy,
DFT and ab initio calculations we were able to gain new
3
insights into the reaction of 3 in the presence of O2
leading to secondary growth of carbon dioxide.
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 207–214

PHOTOCHEMISTRY OF 9-FLUORENONE OXIME PHENYLGLYOXYLATE
213
Figure 4. Graphic representation of relevant stationary points on the C14H10O6 surface
EXPERIMENTAL
flow-rate of 50 ml minꢀ1 The photoreaction was initiated
by the third harmonic (355 nm) of an Nd:YAG laser
(Spectra-Physics Quanta-Ray LAB 130) with a repetition
rate of 10 Hz. The laser beam was widened by an
f ¼ 50 mm Suprasil lens. Three hundred time traces
(10 ms, 1000 points; averaged over 50 laser pulses)
between 3442 and 3442 G (0.1 G resolution) were re-
corded with a LeCroy 9450 digital oscilloscope triggered
by a Thorlabs DET210 photodetector (1 ns rise time,
200–1100 nm spectral range).
STEP-SCAN FTIR. For time-resolved measurements we
used the set-up described previously.1,17 Excitation of
the sample was carried out by the third harmonic
(355 nm) of an Nd:YAG laser (Spectra-Physics Quanta-
Ray LAB 130) with a repetition rate of 10 Hz. To avoid
thermal effects and shockwaves, the laser energy was
attenuated to between 10 and 18 mJ per pulse by an
external attenuator.
All measurements were carried out with 6 cmꢀ1 reso-
lution in a spectral range between 0 and 3160 cmꢀ1
,
Calculations. Calculations were performed with the
Gaussian 98 suite of programs.18 Vibrational spectra
were calculated using the B3LYP19 functional or
MP4(SDQ)20,21 with 6–311 þ G(d,p) or 6–31 þ G(d,p)
basis sets. The calculated vibrational frequencies given
are unscaled.
resulting in an interferogram containing 1060 points.
The signal was averaged 20 times per sampling position.
In all experiments a time range of 20 ms was recorded
with a resolution of 25 ns. After FT, the average of 20
successive spectra was calculated to achieve a better
signal-to-noise ratio, resulting in an effective time resolu-
tion of 500 ns. If necessary, this figure can be improved
by averaging a smaller number of spectra. The sample
was dissolved (between 1.0 and 1.5 g lꢀ1, depending on
the layer thickness of the cell) in carbon tetrachloride
9-Fluorenone oxime phenylglyoxylate (1). 9-Fluorenone
oxime (3.5 g, 17.9 mmol) was suspended in dry dichlor-
omethane (170 ml). A solution of phenylglyoxylyl chlor-
ide (3.9 g, 23.1 mmol) in dry dichloromethane (10 ml)
was added dropwise. After a few minutes, the colour of
the solution changed and after an additional 1 h of stirring
the solution became clear. The solution was stirred for a
further 12 h at room temperature. Removal of the solvent
yielded a residue, which was recrystallized from hexane
(ꢅ1500 ml). This yielded 5.1 g (87%) of pure 1. IR
(KBr): ꢁ~C=O ¼ 1753, 1676, 1597, 1579, 1453, 1320,
1228, 1204, 1150, 1105, 1002, 950, 880, 843, 816, 785,
750, 729, 687, 677, 646 cmꢀ1 1H NMR (200 MHz,
DMSO-d6): ꢄ 7.35 (td, 1H), 7.41 (td, 1H), 7.62 (m, 5H),
¨
(Riedel-de Haen, spectroscopic grade) and purged with
argon for at least 45 min.
TREPR. The TREPR spectra were measured with a Bruker
Elexsys E500 spectrometer at 9.66 GHz microwave fre-
quency and 0.002 W microwave power. The ER4118 split
ring resonator was equipped with a 1 mm inner diameter
quartz tube connected to an Ismatec IPC dispensing
pump with PTFE tubing. An argon-purged 5 mM solution
of 1 in CCl4 was pumped through the quartz tube at a
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 207–214

214
C. KOLANO ET AL.
Moortgat G, Veyret B, Lesclaux R. J. Phys. Chem. 1989; 93:
2362–2368.
7. Davies AG, Sutcliffe R. J. Chem. Soc., Perkin Trans. 2, 1980;
819–824.
8. Huggenberger C, Lipscher J, Fischer H. J. Phys. Chem. 1980; 84:
3467–3474.
9. Krusic PJ, Rettig TA. J. Am. Chem. Soc. 1970; 92: 722–724.
10. Paul H, Fischer H. Helv. Chim. Acta 1973; 56: 1575–1594.
11. Bennett JE, Mile B. Trans. Faraday Soc. 1971; 67: 1587–1597.
12. Atkins PW, Dobbs AJ, McLauchlan KA. J. Chem. Soc., Faraday
Trans. 2, 1975; 71: 1269–1274.
7.79 (d, 1H), 7.88 (t, 2H), 8.12 (d, 2H), 8.25 (d, 1H) ppm.
13C NMR (50 MHz, DMSO-d6): ꢄ 121.20, 122.75,
128.86, 128.91, 129.19, 129.32, 129.94, 130.46, 132.23,
132.77, 132.83, 133.85, 135.53, 141.05, 142.15, 159.37,
161.24, 185.03 ppm. EI MS: m/z 327, 179, 151, 105, 77,
51, 44, 28. C21H13NO3 (327.339): calcd C 77.06, H 4.00,
N 4.28; found C 76.17, H 4.22, N 4.08%.
13. Forrester AR, Gill M, Meyer CJ, Sadd JS, Thomson RH. J. Chem.
Soc., Perkin Trans. 1, 1979; 606–611.
14. Hancock-Chen T, Scaiano JC. Photochem. Photobiol. 1998; 67:
174–178.
Acknowledgements
15. (a) Kraka E, He Y, Cremer D. J. Phys. Chem. A 2001; 105: 3269–
3276; (b) Kraka E, Cremer D, Koller J, Plesnicar B. J. Am. Chem.
This work was financially supported by the Deutsche
Forschungsgemeinschaft and the Fonds der Chemischen
Industrie.
´
Soc. 2002; 124: 8462–8470; (c) Bohr F, Henon E, Garcıa I, Castro
M. Int. J. Quantum Chem. 1999; 75: 671–682.
16. Honeywill JD, Mile B. J. Chem. Soc., Perkin Trans. 2, 2002;
569–575.
17. Kolano C, Sander W. Eur. J. Org. Chem. 2003; 1074–1079.
18. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA,
Cheeseman JR, Zakrzewski VG, Montgomery JA Jr, Stratmann
RE, Burant JC, Dapprich S, Millam JM, Daniels AD, Kudin KN,
Strain MC, Farkas O, Tomasi J, Barone V, Cossi M, Cammi R,
Mennucci B, Pomelli C, Adamo C, Clifford S, Ochterski J,
Petersson GA, Ayala PY, Cui Q, Morokuma K, Salvador P,
Dannenberg JJ, Malick DK, Rabuck AD, Raghavachari K,
Foresman JB, Cioslowski J, Ortiz JV, Baboul AG, Stefanov BB,
Liu G, Liashenko A, Piskorz P, Komaromi I, Gomperts R, Martin
RL, Fox DJ, Keith T, Al-Laham MA, Peng CY, Nanayakkara A,
Challacombe M, Gill PMW, Johnson B, Chen W, Wong MW,
Andres JL, Gonzalez C, Head-Gordon M, Replogle ES, Pople JA.
Gaussian 98, Revision A.11.1. Gaussian: Pittsburgh, PA, 2001.
19. Becke AD. J. Chem. Phys. 1993; 98: 5648–5652.
20. Trucks GW, Watts JD, Salter EA, Bartlett RJ. Chem. Phys. Lett.
1988; 153: 490–495.
REFERENCES
1. Bucher G, Halupka M, Kolano C, Schade O, Sander W. Eur. J.
Org. Chem. 2001; 545–552.
2. Ajayaghosh A, Das S, George MV. J. Polym. Sci., Polym. Chem.
1993; 31: 653–659.
3. Allen NS, Marin MC, Edge M, Davies DW, Garrett J, Jones F,
Navaratnam S, Parsons BJ. J. Photochem. Photobiol. A Chem.
1999; 126: 135–149.
4. Brown CE, Neville AG, Rayner DM, Ingold KU, Lusztyk J. Aust.
J. Chem. 1995; 48: 363–379.
5. Colley CS, Grills DC, Besley NA, Jockusch S, Matousek P, Parker
AW, Towrie M, Turro NJ, Gill PMW, George MW. J. Am. Chem.
Soc. 2002; 124: 14952–14958.
6. (a) Neville AG, Brown CE, Rayner DM, Lusztyk J, Ingold KU.
J. Am. Chem. Soc. 1991; 113: 1869–1870; (b) Wallington TJ,
Dagaut P, Kurylo MJ. Chem. Rev. 1992; 92: 667–710; (c)
21. Trucks GW, Salter EA, Sosa C, Bartlett RJ. Chem. Phys. Lett.
1988; 147: 359–366.
Copyright # 2004 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2004; 17: 207–214