Photophysical study of 5-substituted benzofurazan compounds as fluorogenic probes

Article abstract of DOI:10.1039/b202367a

Nine 5-substituted benzofurazans including the non-substituted benzofurazan were synthesized, and measurements of the fluorescence, photolysis, transient absorption, and time-resolved thermal lensing signal were performed in order to understand the relaxation processes of these compounds. The results indicated that the main relaxation process was a photoreaction from an excited singlet state and the rate of the primary bond cleavage in the excited state tended to increase in the compound in which the S₂←S₀ absorption band is located close to the S₁←S₀ absorption band. These results suggest that the reactive state might be the S₂ state, and that the interaction between the S₁ and S₂ states promotes the photoreaction. The separation of the S₁ state from the S₂ state decreases the rate of photoreaction, resulting in an increase in the Φf values. The Φf values of the 5-substituted benzofurazans in cyclohexane and acetonitrile were compared with their ΔE(S₁, S₂) (energy gap between the S₁ and S₂ states) values calculated by using a combination of AM1(EXCITED) and AM1-CAS/ CI(CI = 6) for geometric optimization and calculation of the energy levels, respectively. The ΔE(S₁, S₂) values correlated well with the Φf values in each solvent, thus enabling us to predict the fluorescence properties of the 5-substituted benzofurazans based on their chemical structures.

Full text of DOI:10.1039/b202367a

View Article Online / Journal Homepage / Table of Contents for this issue
Photophysical study of 5-substituted benzofurazan compounds as
fluorogenic probes
Seiichi Uchiyama,a Kazuyuki Takehira,b Shigeru Kohtani,c Tomofumi Santa,*a
Ryoichi Nakagaki,c Seiji Tobitab and Kazuhiro Imaia
a
Laboratory of Bio-Analytical Chemistry, Graduate School of Pharmaceutical Sciences,
The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan.
E-mail: santa@mol.f.u-tokyo.ac.jp
Department of Chemistry, Gunma University, 1-5-1 Tenjin-cho, Kiryu, Gunma 376-8515,
b
Japan
Graduate School of Natural Science and Technology, Kanazawa University, 13-1 Takaramachi,
Kanazawa 920-0934, Japan
c
Received 6th March 2002, Accepted 27th June 2002
First published as an Advance Article on the web 14th August 2002
Nine 5-substituted benzofurazans including the non-substituted benzofurazan were synthesized, and
measurements of the fluorescence, photolysis, transient absorption, and time-resolved thermal lensing signal
were performed in order to understand the relaxation processes of these compounds. The results indicated that
the main relaxation process was a photoreaction from an excited singlet state and the rate of the primary bond
cleavage in the excited state tended to increase in the compound in which the S2 S0 absorption band is located
close to the S1 S0 absorption band. These results suggest that the reactive state might be the S2 state, and that
the interaction between the S1 and S2 states promotes the photoreaction. The separation of the S1 state from the
S2 state decreases the rate of photoreaction, resulting in an increase in the Ff values. The Ff values of the 5-
substituted benzofurazans in cyclohexane and acetonitrile were compared with their DE(S1 , S2) (energy gap
between the S1 and S2 states) values calculated by using a combination of AM1(EXCITED) and AM1-CAS/
CI(CI ¼ 6) for geometric optimization and calculation of the energy levels, respectively. The DE(S1 , S2) values
correlated well with the Ff values in each solvent, thus enabling us to predict the fluorescence properties of the
5-substituted benzofurazans based on their chemical structures.
Introduction
Elucidation of the relationship between fluorescence intensity
(or fluorescence quantum yield (Ff)) and chemical structure
has been of great interest to scientists in the fields of analytical
and biological chemistry, because this relationship helps
develop useful fluorescent reagents, probes and sensors.
Recently, the ‘‘fluorogenic’’ probe for singlet oxygen, which
is non-fluorescent itself and reacts with an analyte to form a
fluorescent adduct, was developed using the relationship
between the Ff value of a fluorescein derivative and the
HOMO energy of an aromatic substituent.1 The switching of
the fluorescence ‘‘off’’ state and ‘‘on’’ state in this probe is
ascribed to the photoinduced intramolecular electron transfer.2
In general, a substituent in an organic compound signifi-
cantly influences the Ff value of the fluorophore. There have
been many reports on the effect of substituents on the Ff values
including the derivatives of coumarin,3 naphthalene,4 BOD-
IPY,5 thioxanthone,6 coelenteramine analogues,7 carbostyril,8
and imidazo[1,2-a]pyridine.9 In spite of these efforts, there
are few reports that explain the observed relationship between
the Ff value and the substituent on a fluorophore.
The benzofurazan (2,1,3-benzoxadiazole) skeleton shown in
Fig. 1 is used as a fluorophore in many fluorescent reagents
because of its high reactivity and long excitation/emission
wavelengths.10 The fluorescence intensity of the benzofurazans
is also dramatically changed by substituents.11 In a previous
report, we synthesized seventy 4,7-disubstituted benzo-
furazans and tried to elucidate the relationship between their
Fig. 1 Chemical structure of a benzofurazan (2,1,3-benzoxadiazole)
skeleton.
fluorescence intensities and chemical structures. As a result,
we found a relationship between the fluorescence intensity
and the Hammett constants (sp)12 of the substituents.11 This
relationship enabled us to predict from the substituents of a
4,7-disubstituted benzofurazan whether the compound
was fluorescent or not. Based on this relationship, new fluoro-
genic reagents, 7-phenylsulfonyl-4-(2,1,3-benzoxadiazolyl)
isocyanate (PSBD-NCO),13 7-acetylamino-4-mercapto-2,1,3-
benzoxadiazole (AABD-SH),14 and 7-methylthio-4-(2,1,3-ben-
zoxadiazolyl) isothiocyanate (MTBD-NCS)15 have been
developed for the sensitive and selective determination of alco-
hols, carboxylic acids and for peptide sequencing analysis,
respectively.
We also performed semi-empirical molecular orbital calcula-
tions on 4-substituted benzofurazans to give a more precise
prediction of the benzofurazan Ff values and found a relation-
ship between the Ff values and the S1 ꢀ T2 energy differences.16
This relationship enabled us to predict the relative fluorescence
4514
Phys. Chem. Chem. Phys., 2002, 4, 4514–4522
This journal is # The Owner Societies 2002
DOI: 10.1039/b202367a

View Article Online
intensities of the 4-substituted benzofurazans. Based on the
calculated S1 ꢀ T2 energy differences, 4-mercapto-7-
the photoirradiation with the monochromatic light at 275 nm
of the spectrofluorometer (bandpass 10 nm). At specific inter-
vals, an aliquot (5 ml) of the solution was subjected to HPLC.
Separation of the photoreaction mixture was performed using
an analytical column, TOSOH TSKgel ODS-80Ts (250 ꢁ 4.6
mm; id, 5 mm) and isocratic eluent (acetonitirile–water) at a
flow rate of 1.0 mL minꢀ1. The eluate was monitored by UV
detection (275 nm). When S and S0 are the peak areas of the
benzofurazans which are included in the aliquot before and
after photoirradiation, respectively, and e275 nm is the molecu-
lar extinction coefficient at 275 nm, the ratio S0/S varied
linearly with the photoirradiation time t (min) according to
the following expression:
methylthio-2,1,3-benzoxadiazole (MTBD-SH), which is
a
more sensitive fluorogenic reagent than AABD-SH, was
designed and synthesized.17 This relationship suggested that
the main non-radiative relaxation process of the benzofurazan
compounds was intersystem crossing.
In this article, we have selected nine 5-substituted benzo-
furazans, including the non-substituted benzofurazan, to eluci-
date the relationship between the Ff values and the calculated
energies of the electronically excited states. The absorption and
fluorescence characteristics of these compounds were obtained.
Measurements of the steady-state photolysis, transient absorp-
tion, and time-resolved thermal lensing were performed to
understand the photophysical and photochemical properties
of the 5-substituted benzofurazans. In addition, the energies
of the ground and excited states were obtained using semi-
empirical molecular orbital calculations. By comparing the
experimental results with the computational results, we discuss
the relationship for the prediction of the 5-substituted benzo-
furazan Ff values.
S0
¼ 1 ꢀ ae275 nmt
ð1Þ
S
The relative photodecomposition efficiency a value was calcu-
lated from the t value and the observed S0/S by least-squares
analyses. All least-squares analyses were carried out with
Microsoft EXCEL 98. The Fd values of the 5-substituted ben-
zofurazans were obtained by comparing their a values with
that of non-substituted benzofurazan.
The transient absorption spectra were obtained with an
USP-554 system from Unisoku which can give a transient
absorption spectrum with a one-shot laser pulse. A XeCl exci-
mer laser (308 nm) (LAMBDA PHYSIK, LEXTRA 50, pulse
width 17 ns) or the third harmonics (355 nm) of a Nd3+: YAG
laser (Spectra Physics, GCR-130, pulse width 6 ns) was used as
light source for the laser flash photolysis. For the time-resolved
thermal lensing (TRTL) measurements,19 the XeCl excimer
laser or the third harmonics of a Nd3+ : YAG laser was used
as the excitation source. A He–Ne laser (Uniphase, 1103P;
633 nm) was also used as a probe beam for the TRTL measure-
ments. The output signals were converted to a voltage using a
digital storage oscilloscope (Tektronix, TDS-744) connected to
a personal computer (NEC, PC-9821Ap). The absorbance of
the sample solution was adjusted to ca. 0.10 at the excitation
Experimental
Apparatus
Melting points were measured using a Yanagimoto Micro
Melting Point Apparatus and were uncorrected. Proton
nuclear magnetic resonance (1H-NMR) spectra were obtained
on a JEOL LA-500 spectrometer using tetramethylsilane as the
internal standard in CDCl3 (abbreviations used: s ¼ singlet,
d ¼ doublet and m ¼ multiplet). The J values are given in
Hz. Mass spectra were measured using a Hitachi M-1200H
mass spectrometer (atmospheric pressure chemical ionization
(APCI) system).
The absorption and fluorescence spectra were measured
using a JASCO U-best 50 and a Hitachi F-4010 spectrofluo-
rometer, respectively, at room temperature. Each emission
spectrum was obtained by excitation at the maximum absorp-
tion wavelength. The fluorescence quantum yield (Ff) was
determined using quinine sulfate in 0.1 M sulfuric acid
(Ff ¼ 0.55, excitation wavelength: 355 nm, at room tempera-
ture)18 as the standard. The fluorescence lifetime (tf) measure-
wavelength. A benzophenone–acetonitrile solution (Fisc
¼
1.0)20 was used to confirm the optimized experimental condi-
tions.
Synthesis
2,1,3-Benzoxadiazole (5-H),21 5-chloro-2,1,3-benzoxadiazole
(5-Cl),21 5-methoxy-2,1,3-benzoxadiazole (5-OMe),21 5-
methyl-2,1,3-benzoxadiazole (5-Me)21 and 5-nitro-2,1,3-ben-
zoxadiazole (5-NO2)22 were synthesized and purified as pre-
viously reported.
ments were carried out using
a nitrogen laser (LTB
Lasertechnik Berlin, 337 nm, fwhm 350 ps) as the excitation
source. Fluorescence decay signals at the wavelength selected
by a monochromator (Ritsu MC-20L) were detected by a
photomultiplier tube (Hamamatsu H-3284). Signals from the
photomultiplier were recorded on a digital storage oscilloscope
(Tektronix TDS 520) interfaced to a personal computer
(NEC, PC9801). The fluorescence decay curves thus obtained
were analyzed by the deconvolution method involving lineari-
zation of the single exponential function and least-squares
fitting. The detection limit of the method is estimated to be
100 ps.
The steady-state photolysis study was performed using an
extra-high-pressure mercury lamp (USHIO, 250 W) as the light
source with cut-off filters. Acetonitrile was used as the solvent
and the absorbance of each sample solution was adjusted to ca.
0.60 at the maximum absorption wavelength. The photoirra-
diation was carried out on a vigorously stirred sample solution
in a quartz cell. The primary photodecomposition yields (Fd)
of 5-substituted benzofurazans were determined by using the
Hitachi F-4010 spectrofluorometer and high performance
liquid chromatography (HPLC). The HPLC consisted of a
Hitachi L-6200 pump, a Hitachi L-4000H UV detector, and
a Hitachi D-2500 integrator. Each benzofurazan sample was
dissolved in acetonitrile and its absorbance was adjusted to
3.00 at 275 nm. The sample solution (3.2 ml) was stirred during
5-Amino-2,1,3-benzoxadiazole(5-NH2).
5-Nitro-2,1,3-ben-
zoxadiazole (300 mg, 1.8 mmol) was dissolved in a mixture
of dichloromethane (10 ml), conc. hydrochloric acid (4.0 ml)
and methanol (20 ml). After the addition of iron powder
(700 mg), the mixture was vigorously stirred at room tempera-
ture for 30 min. The reaction mixture was poured into water
(100 ml) and extracted with dichloromethane (100 ml, 3 times).
The organic layer was dried over anhydrous sodium sulfate
and concentrated in vacuo. The residue was chromatographed
on silica gel with dichloromethane to afford 5-NH2 (110 mg,
45%) as a yellow powder. mp: 129 ꢂC. dH (CDCl3) 7.65 (1H,
d, J 9.5), 6.89 (1H, d, J 9.5), 6.60 (1H, s), 4.22 (2H, br). Found:
C, 53.53; H, 3.85; N, 31.12. Calc. for C6H5N3O: C, 53.33; H,
3.73; N, 31.10%; APCI-MS: m/z 136 [(M + H)+].
5-Acetylamino-2,1,3-benzoxadiazole(5-NHAc). 5-NH2 (20
mg, 0.15 mmol) was dissolved in a mixture of pyridine (1 ml)
and acetic anhydride (2 ml). The mixture was stirred at
100 ꢂC for 1 h and concentrated in vacuo. The residue was
chromatographed on silica gel with dichloromethane–n-hexane
Phys. Chem. Chem. Phys., 2002, 4, 4514–4522
4515

View Article Online
(2 : 1) to afford 5-NHAc (14 mg, 53%) as a white powder. mp:
154–155 ꢂC. dH (CDCl3) 8.31 (1H, s), 7.78 (1H, d, J 9.5), 7.45
(1H, br), 7.37 (1H, d, J 9.5), 2.25 (3H, s). Found: C, 54.12; H,
4.06; N, 23.45. Calc. for C8H7N3O2 : C, 54.24; H, 3.98; N,
23.72%; APCI-MS: m/z 178 [(M + H)+].
C.I. ¼ 4 (or 5, 6) 1SCF for the CAS calculation or ZINDO
for the ZINDO calculation).
Results and discussion
Fluorescence characteristics
5-Fluoro-2,1,3-benzoxadiazole(5-F). 2-Nitro-4-fluoroaniline
(1 g, 6.4 mmol) was dissolved in ethanol (30 ml) containing
potassium hydroxide (0.40 g). After the addition of sodium
hypochlorite solution (20 ml) at 0 ꢂC, the mixture was stirred
at room temperature for 20 min. The reaction mixture was
poured into water (100 ml) and extracted with ethyl acetate
(100 ml, 3 times). The organic layer was dried over anhydrous
sodium sulfate and concentrated in vacuo. The residue was
chromatographed on silica gel with dichloromethane–n-hexane
(1 : 1) to afford 5-fluoro-2,1,3-benzoxadiazole N-oxide (0.83 g,
84%) as a yellow powder. mp: 49–50 ꢂC. dH (CDCl3) 7.43 (1H,
br), 7.02–7.13 (2H, br). Found: C, 46.64; H, 2.24; N, 18.11.
Calc. for C6H3FN2O2 : C, 46.64; H, 1.96; N, 18.18%. 5-
Fluoro-2,1,3-benzoxadiazole N-oxide (0.83 g, 5.4 mmol) was
dissolved in chlorobenzene (8 ml). After the addition of triphe-
nylphosphine (1.2 g), the mixture was stirred at room tempera-
ture for 20 min. The reaction mixture was evaporated to
dryness under reduced pressure and the residue was chromato-
graphed on silica gel with dichloromethane–n-hexane (1 : 2) to
afford 5-F (0.14 g, 19%) as a yellow oil. dH (CDCl3) 7.83 (1H,
m), 7.34 (1H, d, J 7.0), 7.20 (1H, m).
Nine 5-substituted benzofurazans which include the non-
substituted benzofurazan, 5-H, 5-NMe2 , 5-NH2 , 5-SMe,
5-OMe, 5-NHAc, 5-NO2 , 5-Cl and 5-Me, were used in this
study. Table 1 shows the maximum absorption and emission
wavelengths, Ff and tf values of the 5-substituted benzofura-
zans. The absorption and fluorescence spectra of these com-
pounds in cyclohexane and acetonitrile are shown in Fig. 2,
where the fluorescence signal is almost absent for 5-NO2 ,
5-Me, 5-Cl, and 5-H. It can be seen that the fluorescence char-
acteristics (wavelength and intensity) of these compounds are
significantly affected by the substituent at the 5-position of
the benzofurazan skeleton as in the case of the 4-substituted
compounds.16 The Stokes shifts for the fluorescence of
5-NMe2 , 5-NH2 , 5-SMe, 5-NHAc, and 5-OMe are remark-
ably increased in polar acetonitrile compared to those in non-
polar cyclohexane, suggesting that the fluorescent state has an
intramolecular charge transfer character. It is noteworthy here
that larger Ff values are seen for the compounds of which the
first transition band is separated from the second transition
band having vibrational fine structure, while the compounds
(5-NO2 , 5-Me, 5-Cl and 5-H), in which the two absorption
bands overlap each other, are non-fluorescent (see Fig. 2 and
Table 1). According to the results of the molecular orbital cal-
culations described later, the first and second absorption bands
of these compounds can be assigned to the S1 S0 and S2 S0
transitions. These results suggested that the interaction of
the S1 and S2 states influences the non-radiative relaxation
processes of the 5-substituted benzofurazans.
5-N,N-Dimethylamino-2,1,3-benzoxadiazole (5-NMe2). 5-F
(21 mg, 0.15 mmol) was dissolved in 6 ml of acetonitrile. After
the addition of dimethylamine solution (50% in water, 10 ml),
the mixture was stirred at room temperature for 3 days. The
reaction mixture was evaporated to dryness under reduced
pressure and the residue was chromatographed on silica gel
with dichloromethane to afford 5-NMe2 (17 mg, 69%) as a yel-
low powder. mp: 97–98 ꢂC (lit.23 mp: 97–98 ꢂC). dH (CDCl3)
7.59 (1H, d, J 10.5), 7.21 (1H, d, J 10.5), 6.35 (1H, s), 3.03
(6H, s). Found: C, 58.75; H, 5.68; N, 25.46. Calc. for
C8H9N3O: C, 58.89; H, 5.56; N, 25.75%; APCI-MS: m/z 164
[(M + H)+].
Non-radiative relaxation processes of 5-substituted
benzofurazans
1. Steady-state photolysis. It has been reported that the
photoirradiation of non-substituted benzofurazan in n-hexane
or water leads to photochemical reaction to give cis-cis-1-
cyano-4-N-oxycyano-buta-1,3-diene with quantum yield 0.48
in water.24 In n-hexane, cis-cis-1-cyano-4-N-oxycyano-buta-
1,3-diene isomerized to cis-cis-1-cyano-4-isocyano-buta-1,3-
diene. Therefore, we first performed photolysis experiments
using the nine 5-substituted benzofurazans. The irradiation
was carried out at wavelengths longer than each S1 S0
absorption maximum. All the 5-substituted benzofurazans in
cyclohexane were unstable to the photoirradiation. Represen-
tative results of the photolysis in cyclohexane are shown in
Fig. 3. The non-substituted compound 5-H seems to give cis-
cis-1-cyano-4-isocyano-buta-1,3-diene via cis-cis-1-cyano-4-N-
oxycyano-buta-1,3-diene by photoirradiation in cyclohexane
as previously reported, because the change in the absorption
spectrum of 5-H in cyclohexane was similar to that in n-hexane
shown in a previous report.24 The absorption spectrum of 5-
SMe changes by gaining an isosbestic point on photoirradia-
tion as in the case of 5-H. On the other hand, the changes in
the absorption spectra of the other compounds such as 5-
NMe2 did not give an isosbestic point, suggesting that the
photolysis of these compounds produces more than one com-
pound. We also investigated the effects of dissolved oxygen on
the photolysis of 5-H and 5-SMe but found no remarkable
effects. Mukai and Nitta25 have reported that no sensitizing
effect for the primary bond cleavage of 5-H was observed when
5-H was irradiated with light of 365 nm in the presence of
benzophenone, xanthone, or phenanthrene. These results indi-
cate that photochemical decomposition is one of the main
5-Methylthio-2,1,3-benzoxadiazole(5-SMe). 5-F (70 mg, 0.51
mmol) was dissolved in 5 ml of acetonitrile. After the addition
of methyl mercaptan sodium salt solution (15% in water, 0.30
ml), the mixture was stirred at room temperature for 30 min.
The reaction mixture was evaporated to dryness under reduced
pressure and the residue was chromatographed on silica gel
with dichloromethane to afford 5-SMe (68 mg, 81%) as a white
powder. mp: 99–100 ꢂC. dH (CDCl3) 7.62 (1H, d, J 9.5), 7.22
(1H, s), 7.10 (1H, d, J 9.5), 2.51 (3H, s). Found: C, 50.54; H,
3.82; N, 16.86. Calc. for C7H6N2OS: C, 50.59; H, 3.64; N,
16.86%, APCI-MS: m/z 167 [(M + H)+].
Computational methods
All the semi-empirical molecular orbital calculations were car-
ried out using the program MOPAC2000 in WinMOPAC
ver.3.0 package (Fujitsu) or the program ZINDO in Gaussian
98 (Gaussian, Inc.) with a Gateway 7–800 (CPU; PENTIUM
III 800 MHz, RAM; 768 MB) The geometries of the 5-substi-
tuted benzofurazan compounds in the first excited singlet state
were first completely optimized (keywords PRECISE
EXCITED or PRECISE C.I. ¼ 4 (or 5, 6) ROOT ¼ 2 SING-
LET) by the eigenvector following routine (keyword EF). For
the optimization of the geometry of 5-NHAc, the keyword
MMOK was also used to correct the increase in the barrier
to rotation of the amide linkage. The CAS (Compete Active
Space) or ZINDO calculations were then performed to obtain
the electronic energies of the optimized geometries (keyword
4516
Phys. Chem. Chem. Phys., 2002, 4, 4514–4522

View Article Online
Table 1 Fluorescence characteristics of 5-substituted benzofurazans; maximum absorption wavelength (lab), maximum emission wavelength
(lem), fluorescence quantum yield (Ff) and fluorescence lifetime (tf) at 293 K. The fluroescence characteristic of the nonsubstituted benzofurazan
compound (5-H) was reported previously16
5-NMe2
ab/nm
5-NH2
ab/nm
No.
Solvent
l
lem/nm
Ff
tf/ns
l
lem/nm
Ff
tf/ns
1
2
n-Hexane
Cyclohexane
Benzene
373
376
398
395
401
397
401
397
397
413
452
453
481
496
498
509
517
521
523
553
0.21
0.25
6.8
9.0
349
352
365
369
364
372
374
377
377
367
422
423
454
467
462
484
487
506
508
525
0.096
0.13
0.16
2.9
3.9
9.6
11
3
0.19
0.19
15.0
15.0
19.0
15.0
12.0
4.9
4
Ethyl acetate
Dichloromethane
Acetone
0.16
0.17
5
0.25
0.15
9.2
12
6
0.12
0.10
7
Acetonitrile
Ethanol
0.11
11
12
8
0.040
0.019
0.0016
0.094
0.086
0.012
9
10
Methanol
Water
2.4
0.5
10
3.0
5-SMe
5-OMe
ab/nm
No.
lab/nm
lem/nm
Ff
tf/ns
l
lem/nm
Ff
tf/ns
1
2
345
346
350
348
352
351
351
350
349
352
392
395
440
451
450
465
469
477
487
512
0.014
0.017
0.016
0.023
0.018
0.040
0.050
0.019
0.014
0.0043
0.30
0.30
316
319
323
318
320
334
321
319
320
325
365
372
392
391
394
401
403
407
412
430
0.0018
0.0026
0.0065
0.0068
0.0065
0.012
0.1
0.1
0.2
0.2
0.2
0.4
0.4
0.4
0.5
1.2
3
0.1 (93%), 2.3 (7%)
0.1 (80%), 2.8 (20%)
0.1 (92%), 3.3 (8%)
4
5
6
3.2
3.9
1.9
1.4
1.0
7
0.010
8
0.0083
0.011
0.018
9
10
5-NHAc
5-NO2
5-Cl
5-Me
No.
l
ab/nm
lem/nm
Ff
tf/ns
l
ab/nm
Ff
lab/nm
Ff
lab/nm
Ff
1
2
329
329
336
334
333
338
332
332
332
329
383
0.0036
0.0043
0.016
0.019
0.017
0.021
0.025
0.027
0.033
0.040
0.2
0.2
0.5
0.7
0.5
1.0
1.1
1.4
1.6
2.9
308
321
315
313
314
–
–
–
–
–
–
–
–
–
–
–
300
300
305
301
304
–
–
–
–
–
–
–
–
–
–
–
300
303
304
304
306
–
–
–
–
–
–
–
–
–
–
–
380
411
411
415
418
421
434
440
455
3
4
5
6
7
313
313
316
318
302
304
303
305
304
306
303
311
8
9
10
relaxation processes of the 5-substituted benzofurazans in the
excited singlet state.
none in acetonitrile was utilized as an actinometer. The Fp
value of 5-H was determined by using the following equation:
eBTP
ep ꢀ es
DAp
2. Transient absorption spectra. The transient absorption
spectra of the 5-substituted benzofurazans in cyclohexane were
obtained in order to estimate the photoproduct formation
yield (Fp) and to investigate whether the nine 5-substituted
benzofurazans show T–T absorption. Fig. 4 shows the transi-
ent absorption spectra obtained at 100 ns after the laser photo-
lysis of 5-NMe2 , 5-SMe and 5-H. The positive transient
absorption bands seen for 5-NMe2 and 5-SMe (Fig. 4(a) and
(b)) were assigned to their Tn T1 absorption based on the tri-
plet photosensitization experiments using acetone as sensitizer,
although the absorption band at around 300 nm includes a
contribution from the photoproduct absorption. The transient
absorption spectrum of 5-H in cyclohexane does not show any
T–T absorption band in the 300–550 nm wavelength range
(Fig. 4(c)). We performed the Fp measurements for 5-H in
cyclohexane and acetonitrile by monitoring the photoproduct
absorption at 290 nm. The Tn T1 absorption of benzophe-
Fp ¼
ꢃ
ð2Þ
DABTP
where eBTP, ep , and es are the molecular extinction coefficients of
triplet benzophenone at 530 nm, photoreaction product of 5-H
at 290 nm, and 5-H at 290 nm, DAp and DABTP are the absor-
bance changes obtained by the transient absorption spectra
of 5-H at 280 nm and benzophenone at 530 nm, respectively.
The Fp values in cyclohexane and acetonitrile were determined
to be 0.91 and 0.36, indicating that the decomposition of the
benzofurazan skeleton to the corresponding buta-1,3-diene
was the main relaxation process from the excited singlet state
of 5-H. In cyclohexane, 5-NH2 and 5-NO2 gave a T–T absorp-
tion band, but no T–T absorption band was observed for 5-
NHAc, 5-OMe, 5-Me, and 5-Cl.
The primary photodecomposition yield (Fd) of benzofura-
zans was determined by steady-state photolysis experiments
Phys. Chem. Chem. Phys., 2002, 4, 4514–4522
4517

View Article Online
Fig. 3 Change in absorption spectra of 5-NMe2 , 5-SMe and 5-H in
deaerated cyclohexane upon photoirradiation (5-NMe2 ; l > 390 nm,
5-SMe; l > 350 nm, 5-H; l > 290 nm) at 293 K.
formation of the triplet state can be confirmed by the presence
of the slow rise component (Uslow) in the TRTL signals. As
shown in Fig. 5, the slow rise component is seen for 5-NMe2
(Uslow/Utotal ¼ 0.112) and 5-SMe (Uslow/Utotal ¼ 0.356), indi-
cating the formation of the triplet state, whereas the only fast
component is found in 5-H which shows the absence of triplet
formation. 5-NH2 and 5-NO2 also gave slow components
in their TRTL signals, consistent with the results of the transi-
ent absorption measurements. On the other hand, a slow
Fig. 2 Absorption (---) and fluorescence (ꢃ ꢃ ꢃꢃ ꢃ ꢃ) spectra of nine 5-
substituted benzofurazans in (a) cyclohexane (CH) and (b) acetonitrile
(ACN).
at 275 nm combined with HPLC. Assuming that the Fd value
of 5-H is unity, those for 5-NO2 , 5-OMe, 5-SMe, and 5-NMe2
were obtained to be <0.01, 0.41, 0.07, and 0.57, respectively.
For 5-H, 5-Me, and 5-Cl, which give similar absorption spec-
tra (see Fig. 2), almost the same spectral changes were
observed in the photolysis experiments, showing that these
three compounds have large Fd values. The photodecomposi-
tion yields of 5-NH2 and 5-NHAc were estimated to be similar
to those of 5-NMe2 and 5-OMe, respectively.
3. Time-resolved thermal lensing (TRTL) signals. Since T–T
absorption was found in 5-NMe2 , 5-NH2 , 5-SMe, and 5-NO2
in cyclohexane and intersystem crossing seems to occur in
these compounds from the S1 state, we made TRTL measure-
ments in cyclohexane to confirm the occurrence of intersystem
crossing. In the TRTL measurements, the intensity ratio of the
total signal (Utotal) to that of the slow rise component (Uslow
is related to the quantum yields of intersystem crossing (Fisc
by the following equation:
)
)
Uslow
F
iscET
¼
ð3Þ
Utotal Eex ꢀ Ff hESi ꢀ FrDH
where Eex , ET , and hESi are the excitation energy, the 0–0
transition energy for the T1 state, and the mean energy dissi-
pated by fluorescence of the 5-substituted benzofurazans,
respectively, and DH is the total heat of formation of the
photodecomposition products. According to eqn. (3), the
Fig. 4 Transient absorption spectra of 5-NMe2 , 5-SMe and 5-H
observed at 100 ns after laser irradiation (5-NMe2 and 5-SMe; 355
nm, 5-H; 308 nm) in cyclohexane at 293 K.
4518
Phys. Chem. Chem. Phys., 2002, 4, 4514–4522

View Article Online
fluorescent compounds (5-NMe2 and 5-NH2) which have Ff
values greater than 0.1; (II) weakly-fluorescent compounds
(5-SMe, 5-NHAc and 5-OMe) in which a relatively fast inter-
system crossing (5-SMe) or decomposition (5-NHAc and
5-OMe) is involved; and (III) non-fluorescent compounds
(5-NO2 , 5-Me, 5-Cl, and 5-H) in which very fast intersystem
crossing (5-NO2) or decomposition (5-Me, 5-Cl, and 5-H)
takes place. Considering the above classification of the 5-sub-
stituted benzofurazans, the main relaxation process of the
5-substituted benzofurazans is the photoreaction from the
excited singlet state and for 5-SMe and 5-NO2 the occurrence
of fast intersystem crossing is involved in addition to the
photoreaction. The increased rate of intersystem crossing in
5-SMe and 5-NO2 can be ascribed to the characteristic effects
of the methylthio and nitro26 groups.
As described in earlier sections, the Ff values increase for the
compounds in which the S2 S0 absorption band is separated
from the S1 S0 absorption band, and decomposition
occurred by photoirradiation at around 280 nm. For 5-OMe,
the fluorescence from excitation to the S2 state was signifi-
cantly weaker than that from excitation to the S1 state. These
results suggest that the S2 state of the 5-substituted benzofur-
azans is the reactive state. Therefore, the energy difference
between the S1 and S2 states (DE(S1 , S2)) reflects the photode-
composition efficiency from the S2 state by excitation at the
maximum S1 S0 absorption wavelength. For instance, the
non-fluorescent property of 5-H, 5-Cl and 5-Me is due to very
rapid photoreaction from the excited singlet state because the
S2 state is extremely close to the S1 state in these compounds
(see Fig. 2). In contrast, the S2 level of 5-NMe2 and 5-NH2
is far from the S1 level, resulting in the decreased photoreac-
tion rate and the occurrence of fluorescence and intersystem
crossing.
Fig. 5 Time evolution of the time-resolved thermal lensing signals of
5-NMe2 , 5-SMe and 5-H in deaerated cyclohexane excited at 355 nm
(5-NMe2 and 5-SMe) and 308 nm (5-H) at 293 K.
component was not detected in 5-NHAc, 5-OMe, 5-Me and
5-Cl which did not show any T–T absorption.
The relaxation processes of the nine 5-substituted benzofur-
azans obtained by measurements of the fluorescence, photoly-
sis, transient absorption and TRTL signals are summarized in
Table 2. The quantum yield for the non-radiative deactivation
(Fnr) from the S1 state is estimated by Fnr ¼ 1 ꢀ Ff in Table 2,
which includes decomposition, intersystem crossing and inter-
nal conversion. Although the Ff and tf values of each com-
pound are influenced by solvent polarity, the higher Ff
values were observed in the compounds showing smaller knr
values in cyclohexane and acetonitrile, respectively. Therefore
the elucidation of non-radiative relaxation properties seems to
be especially important for the prediction of 5-substituted ben-
zofurazans. The nine 5-substituted benzofurazans are classified
into the following groups by their relaxation properties: (I)
Relationship between fluorescence quantum yields and calculated
S1 - S2 energies
Since DE(S1 , S2) values seem to influence the Ff values of the
5-substituted benzofurazans, we tried to relate their Ff values
to their energy differences obtained by using semi-empirical
molecular orbital calculations. First, we calculated the energy
levels corresponding to the optimized S1 structure for each
5-subsituted benzofurazan. The reliability of the calculation
methods was examined by comparing the experimental
S1 ! S0 emission energies evaluated by the energy of the fluor-
escence maximum in cyclohexane with the calculated S1 ! S0
Table 2 Relaxation properties of 5-substituted benzofurazans in cyclohexane (CH) and acetonitrile (ACN); fluorescence quantum yield (Ff), emis-
sion energy (nem), fluorescence lifetime (tf), yield of non-radiative relaxation (Fnr), radiative rate constant (kf) and non-radiative rate constant (knf)
Compound
5-NMe2
Solvent
n
em/cmꢀ1
tf/ns
Ff
Uslow/Utotal
Fnr
kf/107 sꢀ1
knr/107 sꢀ1
CH
ACN
CH
22 100
19 300
23 600
20 500
25 300
21 300
26 300
23 800
26 900
24 800
–
9.0
12
3.9
11
0.3
3.9
0.2
1.1
0.1
0.4
0
0.25
0.11
0.13
0.10
0.017
0.050
0.0043
0.025
0.0026
0.010
0
0.112
0
0.0721
0.75
2.8
0.92
3.3
0.91
5.7
1.3
2.2
2.3
2.6
2.5
–
8.4
7.4
0.89
0.87
0.90
0.98
0.95
>0.99
0.98
>0.99
0.99
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
5-NH2
5-SMe
5-NHAc
5-OMe
5-NO2
5-Me
22
8.2
330
25
>500
88
>990
250
–
ACN
CH
0
0.356
0.295
ACN
CH
0
0.184
ACN
CH
0
0.0950
ACN
CH
0.335
0.303
ACN
CH
–
0
0
–
–
–
0
0
0
0
0
0
0
0
–
–
ACN
CH
–
0
0
–
–
5-Cl
–
0
0
–
–
ACN
CH
ACN
–
0
0
–
–
5-H
–
0
0
–
–
–
0
0
–
–
Phys. Chem. Chem. Phys., 2002, 4, 4514–4522
4519

View Article Online
Table 3 The correlation coefficient (r) between the experimental S1 ! S0 emission energies and calculated S1 ! S0 transition energies of the fluor-
escent 5-substituted benzofurazans (5-NMe2 , 5-NH2 , 5-SMe, 5-NHAc, 5-OMe)
Geometric optimization
Calculation of energy levels
r
Geometric optimization
Calculation of energy levels
r
AM1-CAS/CI(CI ¼ 4)
AM1-CAS/CI(CI ¼ 4)
AM1-CAS/CI(CI ¼ 5)
AM1(EXCITED)
AM1-CAS/CI(CI ¼ 4)
ZINDO(CI ¼ 20)
0.970
0.810
0.729
0.817
0.852
0.978
PM3-CAS/CI(CI ¼ 4)
PM3-CAS/CI(CI ¼ 4)
PM3-CAS/CI(CI ¼ 5)
PM3(EXCITED)
PM3-CAS/CI(CI ¼ 4)
ZINDO(CI ¼ 20)
0.677
0.533
0.674
0.820
0.643
0.740
AM1-CAS/CI(CI ¼ 5)
ZINDO(CI ¼ 20)
PM3-CAS/CI(CI ¼ 5)
ZINDO(CI ¼ 20)
AM1(EXCITED)
AM1(EXCITED)
AM1-CAS/CI(CI ¼ 5)
AM1-CAS/CI(CI ¼ 6)
PM3(EXCITED)
PM3(EXCITED)
PM3-CAS/CI(CI ¼ 5)
PM3-CAS/CI(CI ¼ 6)
transition energies of the five fluorescent compounds, 5-NMe2 ,
5-NH2 , 5-SMe, 5-OMe, 5-NHAc. The calculation methods
tested in this study and the correlation coefficients between
the experimental and calculated S1 ! S0 energies are shown
in Table 3. As seen in Table 3, the combination of the
AM1(EXCITED) and AM1-CAS/CI(CI ¼ 6) methods for
geometric optimization and calculation of the energy levels,
respectively, gives the best correlation coefficients
(r ¼ 0.978). Although the geometric optimization at the
excited state was generally less precise than that at the ground
state, these methods enable us to precisely estimate the energy
levels of the excited 5-substituted benzofurazans. Therefore,
we used the AM1(EXCITED) and AM1-CAS/CI(CI ¼ 6)
methods.
The calculated singlet and triplet energy levels, and the
energy differences, DE(S1 , S2), of the nine 5-substituted benzo-
furazans are summarized in Table 4. In general, for the com-
pounds which have a carbonyl group or N-hetero atom in a
fluorophore such as coumarin27 and acridine,28 fast intersys-
tem crossing from S1(p,p*) to Tn(n,p*) (or from S1(n,p*) to
Tn(p,p*)) is often involved. Since the 5-substituted benzofura-
zans in Table 4 also include nitrogen and oxygen atoms in the
benzofurazan skeleton and the substituent, it is expected that
singlet or triplet (n,p*) states exist near the fluorescent state.
However, the singlet (n,p*) and triplet (n,p*) states are not
obtained in four compounds, 5-NMe2 , 5-NH2 , 5-SMe, 5-
OMe, as shown in Table 4. Although the (n,p*) state is
obtained for 5-NHAc, 5-NO2 , 5-Me, 5-Cl and 5-H, the singlet
(n,p*) and triplet (n,p*) states are too high to interact with the
S1 state of these compounds.
compared with those of the S0 and S1 states. These results also
suggested that the S2 state of 5-substituted benzofurazans is
the photoreactive state. Fig. 7(a) displays the relationship
between the Ff values of the 5-substituted benzofurazans in
cyclohexane and their calculated DE(S1 , S2) values. As
expected, the Ff values of the 5-substituted benzofurazans
are well correlated to their DE(S1 , S2) values, i.e., the order
of the Ff values, 5-NMe2 (0.25) > 5-NH2 (0.13) > 5-SMe
(0.017) > 5-NHAc (0.0043) > 5-OMe (0.0026) > 5-NO2 , 5-Me,
5-Cl, 5-H (0) was in good agreement with the order of the
DE(S1 , S2) values, 5-NMe2 (0.952 eV) > 5-NH2 (0.857) > 5-
SMe (0.703) > 5-NHAc (0.589) > 5-OMe (0.504) > 5-NO2
(0.405) > 5-Me (0.183) > 5-Cl (0.179) > 5-H (0.090). These
results indicate that calculation of the DE(S1 , S2) values using
the combination of the AM1(EXCITED) and AM1-CAS/
CI(CI ¼ 6) methods is the best way to precisely predict the
Ff values of the 5-substituted benzofurazans in cyclohexane.
Fig. 7(b) shows the relationship between the Ff values of the
5-substituted benzofurazans in acetonitrile and their calculated
DE(S1 , S2) values. Similarly to what happens in cyclohexane,
the Ff values of the 5-substituted benzofurazans in acetonitrile
were also well related to the calculated DE(S1 , S2) values.
These results suggested that the photoreaction from the S1
state was the main non-radiative relaxation process, even in
a polar solvent, indicating that calculation of the DE(S1 , S2)
values using the combination of the AM1(EXCITED) and
AM1-CAS/CI(CI ¼ 6) methods is also the best way to predict
the Ff values of the 5-substituted benzofurazans in a polar
solvent. As described above, we succeeded in elucidating the
method for the prediction of the Ff values of the 5-substituted
benzofurazans. Therefore, the Ff values of all the mono-substi-
tuted benzofurazans can be predicted using semi-empirical
molecular orbital calculations based on the relationships
obtained in the previous study for the 4-substitued benzofura-
zans10 or the present study for the 5-substituted benzofura-
zans.
Based on the photolysis, transient absorption and the TRTL
measurement results, the main non-radiative relaxation pro-
cess of the 5-substituted benzofurazans was found to be the
photoreaction from the excited singlet state and the photoreac-
tion rate was related to the energy difference, the DE(S1 , S2)
values. In Fig. 6, we show the bond order of 5-OMe in the
S0 , S1 , and S2 states obtained by the AM1-CAS/CI method
as the representative result of calculation. As shown in Fig.
6, the order of the N–O bond between the 1- and 2-positions
in the benzofurazan skeleton decreased in the S2 state as
Finally, we compared the present results for the 5-substi-
tuted benzofurazans with the previous results for the 4-substi-
tuted benzofurazans.10 First, the most precise method for
the calculation of the energy levels is different between the
Table 4 Energy levels and energy difference between S1 and S2 states DE(S1 , S2) of 5-substituted benzofurazans obtained by using AM1(EX-
CITED) and AM1-CAS/CI(CI ¼ 6) methods for geometric optimization and calculation of energy levels, respectively
Compound
T1/eVa
T2/eVa
S1/eVa
S2/eVa
Tn(n,p*)/eV
Sn(n,p*)/eV
DE(S1 , S2)/eV
b
b
b
b
b
b
5-NMe2
5-NH2
5-SMe
5-NHAc
5-OMe
5-NO2
5-Me
1.635
1.489
1.879
1.820
1.627
1.905
1.829
1.847
1.853
2.664
2.760
2.846
2.964
3.023
3.244
3.180
3.159
3.268
2.692
2.747
2.997
3.147
3.208
3.463
3.513
3.518
3.629
3.644
3.604
3.700
3.736
3.712
3.868
3.695
3.696
3.719
0.952
0.857
0.703
0.589
0.504
0.405
0.183
0.179
0.090
5.343
b
5.412
b
4.377
5.028
5.494
4.954
5.241
5.163
5.567
5.095
5-Cl
5-H
a
b
Relative to the S0 state. Not obtained.
4520
Phys. Chem. Chem. Phys., 2002, 4, 4514–4522

View Article Online
Conclusion
In this study, the relationship between the Ff values of the 5-
substituted benzofurazans and their DE(S1 , S2) values calcu-
lated using the semi-empirical AM1 method was obtained.
The measurements of the fluorescence, photolysis, transient
absorption, and TRTL signal indicated that the main relaxa-
tion process of the nine 5-substituted benzofurazans was
photoreaction from an excited singlet state and the Ff values
of the 5-substituted benzofurazans increased in the compound
in which the S1 state was separated from the S2 state. There-
fore, we selected the combination of the AM1(EXCITED)
and AM1-CAS/CI(CI ¼ 6) methods as the most precise
method for calculating the energy levels, and succeeded in
relating the Ff values of the 5-substituted benzofurazans in
cyclohexane and acetonitrile to the calculated DE(S1 , S2)
values. In the near future, highly sensitive fluorogenic reagents
with the benzofurazan skeleton will be developed based on the
present relationship. We will also try to elucidate the relation-
ship between the Ff values and calculated energies for disubsti-
tuted benzofurazans.
Fig. 6 Bond order of 5-OMe in the S0 , S1 , and S2 states calculated
by the AM1-CAS/CI(CI ¼ 6) method.
5-substituted benzofurazans (AM1(EXCITED) and AM1-
CAS/CI(CI ¼ 6) methods for geometric optimization and
the calculation of energy levels, respectively) and the 4-substi-
tuted benzofurazans (PM3 and PM3-CAS/CI(CI ¼ 5) meth-
ods). Although the unification of the calculation methods for
the 4- and 5-substituted benzofurazans enables us to directly
compare the Ff values of the benzofurazans substituted at
the different position, it was impossible in this study. For
instance, the combination of the AM1(EXCITED) and
AM1-CAS/CI(CI ¼ 6) methods gave poor correlation coeffi-
cients (r ¼ 0.875) for the 4-substituted benzofurazans, whereas
this combination gave the best result for 5-substitued benzo-
furazans. Second, the energy difference related to the Ff values
is different between the 5-substituted benzofurazans (DE(S1 ,
S2)) and 4-substituted benzofurazans (DE(S1 , T2)). This differ-
ence seems to be ascribed to the fact that the main non-radia-
tive relaxation process competing with the fluorescence is
different between the 5- and 4-substituted benzofurazans.
References
1
2
K. Tanaka, T. Miura, N. Umezawa, Y. Urano, K. Kikuchi,
T. Higuchi and T. Nagano, J. Am. Chem. Soc., 2001, 123, 2530.
(a) A. P. de Silva, H. Q. N. Gunarathe, T. Gunnlaugsson, A. J. M.
Huxley, C. P. McCoy, J. T. Rademacher and T. E. Rice, Chem.
Rev., 1997, 97, 1515; (b) A. P. de Silva, D. B. Fox, T. S. Moody
and S. M. Weir, Pure Appl. Chem., 2001, 73, 503; (c) A. P. de
Silva, D. B. Fox, T. S. Moody and S. M. Weir, Trends Biotechnol.,
2001, 19, 29; (d ) A. P. de Silva, D. B. Fox, A. J. M. Huxley and
T. S. Moody, Coord. Chem. Rev., 2000, 205, 41; (e) A. P. de Silva,
T. Gunnlaugsson and T. E. Rice, Analyst, 1996, 121, 1759.
(a) C. E. J. Wheelock, J. Am. Chem. Soc., 1959, 81, 1348; (b) W. R.
Sherman and E. Robins, Anal. Chem., 1968, 40, 803; (c) A. Taka-
date, T. Masuda, C. Murata, T. Tanaka, M. Irikura and S. Goya,
Anal. Sci., 1995, 11, 97; (d ) A. Takadate, T. Masuda, C. Murata,
A. Isobe, T. Shinohara, M. Irikura and S. Goya, Anal. Sci., 1997,
13, 753; (e) A. Takadate, T. Masuda, C. Murata, M. Shibuya and
A. Isobe, Chem. Pharm. Bull., 2000, 48, 256; ( f ) H. Takechi,
Y. Oda, N. Nishizono, K. Oda and M. Machida, Chem. Pharm.
Bull., 2000, 48, 1702.
3
4
5
E. M. Kosower, Acc. Chem. Res., 1982, 15, 259.
(a) M. Shah, K. Thangaraj, M.-L. Soong, L. T. Wolford, J. H.
Boyer, I. R. Politzer and T. G. Pavlopoulos, Heteroatom Chem.,
1990, 1, 389; (b) J. H. Boyer, A. M. Haag, G. Sathyamoorthi,
M.-L. Soong and K. Thangaraj, Heteroatom Chem., 1993, 4, 39.
M. G. Neumann, M. H. Gehlen, M. V. Encinas, N. S. Allen,
T. Corrales, C. Peinado and F. Catalina, J. Chem. Soc., Faraday
Trans., 1997, 93, 1517.
6
7
8
9
R. Saito, T. Hirano, H. Niwa and M. Ohashi, J. Chem. Soc.,
Perkin Trans. 2, 1997, 1711.
W. M. F. Fabian, K. S. Niederreiter, G. Uray and W. Stadlbauer,
J. Mol. Struct., 1999, 477, 209.
H. Tomoda, T. Hirano, S. Saito, T. Mutai and K. Araki, Bull.
Chem. Soc. Jpn., 1999, 72, 1327.
10 S. Uchiyama, T. Santa, N. Okiyama, T. Fukushima and K. Imai,
Biomed. Chromatogr., 2001, 15, 295.
11 S. Uchiyama, T. Santa, T. Fukushima, H. Homma and K. Imai,
J. Chem. Soc., Perkin Trans. 2, 1998, 2165.
12 C. Hansch, A. Leo and R. W. Taft, Chem. Rev., 1991, 91, 165.
13 S. Uchiyama, T. Santa, S. Suzuki, H. Yokosu and K. Imai, Anal.
Chem., 1999, 71, 5367.
14 T. Santa, T. Okamoto, S. Uchiyama, K. Mitsuhashi and K. Imai,
Analyst, 1999, 124, 1689.
15 A. Toriba, K. Adzuma, T. Santa and K. Imai, Anal. Chem., 2000,
72, 732.
16 S. Uchiyama, T. Santa, N. Okiyama, K. Aduma and K. Imai,
J. Chem. Soc., Perkin Trans. 2, 2000, 1199.
17 S. Uchiyama, T. Santa and K. Imai, Anal. Chem., 2001, 73, 2165.
18 A. N. Fletcher, Photochem. Photobiol., 1969, 9, 439.
19 K. Suzuki, H. Tanabe, S. Tobita and H. Shizuka, J. Phys. Chem.
A, 1997, 101, 4496.
20 (a) K. Sandros, Acta Chem. Scand., 1969, 23, 2815; (b) S. K.
Chattopadhyay, C. V. Kumar and P. K. Das, J. Photochem.,
1985, 30, 81.
Fig. 7 Relationships between the Ff values of the 5-substituted ben-
zofurazans in (a) cyclohexane (CH) and (b) acetonitrile (ACN) and
their DE(S1 , S2) values calculated using the combination of the
AM1(EXCITED) and AM1-CAS/CI(CI ¼ 6) methods.
Phys. Chem. Chem. Phys., 2002, 4, 4514–4522
4521

View Article Online
27 (a) W. W. Mantulin and P.-S. Song, J. Am. Chem. Soc., 1973, 95,
21 R. J. Gaughran, J. P. Picard and J. V. R. Kaufman, J. Am. Chem.
Soc., 1954, 76, 2233.
22 F. B. Mallory and S. P. Varimbi, J. Org. Chem., 1963, 28, 1656.
23 A. J. Boulton, P. B. Ghosh and A. R. Katritzky, J. Chem. Soc. C,
1966, 971.
24 W. Heinzelmann and P. Gilgen, Helv. Chim. Acta, 1976, 59, 2727.
25 T. Mukai and M. Nitta, Chem. Commun., 1970, 1192.
26 (a) A. K. Chibisov and H. Go¨rner, J. Phys. Chem. A, 1997, 101,
4305; (b) A. K. Chibisov and H. Go¨rner, J. Photochem. Photobiol.
A, 1997, 105, 261.
5122; (b) J. S. S. deMelo, R. S. Becker and A. L. Mac¸anita,
J. Phys. Chem., 1994, 98, 6054; (c) J. S. deMelo, R. S. Becker,
F. Elisei and A. L. Mac¸anita, J. Chem. Phys., 1997, 107, 6062;
(d ) F. Elisei, G. G. Aloisi, F. Dall’Acqua, L. Latterini, F. Masetti
and P. Rodighiero, Photochem. Photobiol., 1998, 68, 164.
28 (a) A. Kellmann, J. Phys. Chem., 1977, 81, 1195; (b) K. Kasama,
K. Kikuchi, S. Yamamoto, K. Uji-ie, Y. Nishida and H. Koku-
bun, J. Phys. Chem., 1981, 85, 1291; (c) K. Kasama, K. Kikuchi,
Y. Nishida and H. Kokubun, J. Phys. Chem., 1981, 85, 4148.
4522
Phys. Chem. Chem. Phys., 2002, 4, 4514–4522