A photoinduced electron-transfer reagent for peroxyacetic acid, 4-ethylthioacetylamino-7-phenylsulfonyl-2,1,3-benzoxadiazole, based on the method for predicting the fluorescence quantum yields

Article abstract of DOI:10.1021/ac0201225

To develop new photoinduced electron-transfer (PET) reagents, we established a method for predicting the fluorescence quantum yields (Φ) of the benzofurazan compounds bearing an aliphatic substituent group having an n-electron. The PET process occurred su

Full text of DOI:10.1021/ac0201225

Anal. Chem. 2002, 74, 4089-4096
A Photoinduced Electron-Transfer Reagent for
Peroxyacetic Acid, 4-Ethylthioacetylamino-7-
phenylsulfonyl-2,1,3-benzoxadiazole, Based on the
Method for Predicting the Fluorescence Quantum
Yields
Maki Onoda, Seiichi Uchiyama, Tomofumi Santa,* and Kazuhiro Imai
Graduate School of Pharmaceutical Sciences,The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-0033, Japan
off reagent, and they are called “PET reagents”.4 In general, the
PET reagents consist of a fluorophore and a reacting group as an
electron donor and an electron acceptor in one molecule. A donor
and an acceptor are connected directly or through a spacer. These
reagents are nonfluorescent or weakly fluorescent themselves,
since the electron transfer from a donor to an acceptor in an
excited state accelerates the radiationless relaxation. After the
reaction with analytes, the electron transfer is suppressed because
the electron-donating ability of a donor decreases, restoring the
fluorescence of the fluorophore. A fluorophore and a reaction
group of most previously reported PET reagents1-3 are an acceptor
and a donor, respectively. The donor is divided into a π-donor
and an n-donor, to give a π-electron and n-electron, respectively,
to an acceptor.5-8 Diaminofluorescein (DAF-2) for nitric oxide9
and N-(9-anthrylmethyl)diethanolamine for boronic and boric
acids10 are the representative reagents with a π-donor and an
n-donor, respectively.
To develop new photoinduced electron-transfer (P ET)
reagents, we established a method for predicting the
fluorescence quantum yields (Φ) of the benzofurazan
compounds bearing an aliphatic substituent group having
an n-electron. The P ET process occurred sufficiently to
reduce the Φ values in the benzofurazan compounds
bearing an aliphatic moiety, which had a high quenching
ability. The quenching ability was estimated by the mo-
lecular orbital calculation and Stern-Volmer plotting. The
Φ values of the benzofurazan compounds could be con-
trolled by changing the quenching ability of a substituent
group. We succeeded in designing a P ET reagent for
peroxyacetic acid (P AA), 4 -ethylthioacetylamino-7 -phe-
nylsulfonyl-2 ,1 ,3 -benzoxadiazole (EP B), using the estab-
lished method for predicting the Φ values. EP B and its
oxidized derivative were separated by reversed-phase
HP LC and fluorometrically detected at 4 7 9 nm with
excitation at 3 6 2 nm. The attained detection limit for P AA
was 1 0 5 fmol (S/ N ) 3 ) and the cross-reactivity toward
hydrogen peroxide was very low, indicating EP B is a
highly sensitive and selective reagent for P AA.
To develop new PET reagents more efficiently, we intended
to establish a method for predicting the fluorescence quantum
yields (Φ) of benzofurazan compounds bearing an aromatic
substituent group as a π-donor.11 Since some fluorescent coumarin
dyes were quenched in the presence of other aromatic com-
pounds12,13 (intermolecular quenching) and a fluorescein derivative
Fluorometric detection has been widely used in many fields
of science such as analytical chemistry and biochemistry, because
of its sensitivity and selectivity. Since most analytes do not
fluoresce, their derivatization with fluorescent reagents is neces-
sary. Many fluorescent reagents have been developed, and they
are classified into two groups by their fluorescence properties.
One is a “fluorescent labeling reagent”, which is strongly fluo-
rescent itself. The other is a “fluorescent ‘on-off’ reagent”, which
is nonfluorescent or weakly fluorescent itself and reacts with the
analytes to form the fluorescent derivatives. The fluorescent on-
off reagents are generally superior because they can avoid
interference from the fluorescence of the reagents themselves.
Recently, reagents utilizing photoinduced electron transfer
(PET)1-3 have been developed as a new type of fluorescent on-
(2) de Silva, A. P.; Gunnlaugsson, T.; Rice, T. E. Analyst 1 9 9 6 , 121, 1759-
1762.
(3) Bissell, R. A.; de Silva, A. P.; Gunaratne, H. Q. N.; Lynch P. L. M.; Maguire,
G. E. M.; Sandanayake K. R. A. S. Chem. Soc. Rev. 1 9 9 2 , 21, 187-195.
(4) de Silva, A. P.; Gunaratne, H. Q. N.; Gunnlaugsson, T. Tetrahedron Lett.
1 9 9 8 , 39, 5077-5080.
(5) Jacques, P.; Haselbach, E.; Henseler, A.; Pilloud, D.; Suppan, P. J. Chem.
Soc., Faraday Trans. 1 9 9 1 , 87, 3811-3813.
(6) Jacques, P.; Burget, D. J. Photochem. Photobiol. A: Chem. 1 9 9 2 , 68, 165-
168.
(7) Jacques, P.; Allonas, X. J. Photochem. Photobiol. A: Chem. 1 9 9 4 , 78, 1-5.
(8) Jacques, P.; Allonas, X.; Burget, D.; Haselbach, E.; Muller, P. A,; Sergenton,
A. C.; Galliker, H. Phys. Chem. Chem. Phys. 1 9 9 9 , 1, 1867-1871.
(9) Kojima, H.; Nakatsubo, N.; Kikuchi, K.; Kawahara, S.; Kirino, Y.; Nagoshi,
H.; Hirata, Y.; Nagano, T. Anal. Chem. 1 9 9 8 , 70, 2446-2453.
(10) Wang, W.; Springsteen, G.; Gao, S. H.; Wang, B. H. Chem. Commun. 2 0 0 0 ,
1283-1284.
* Corresponding author. E-mail: santa@mol.f.u-tokyo.ac.jp, Tel: 81-3-5841-
4761. Fax: 81-3-5802-3339.
(1) de Silva, A. P.; Gunaratne, H. Q. N.; Gunnlaugsson, T.; Huxley, A. J. M.;
McCoy, C. P.; Rademacher, J. T.; Rice, T. E. Chem. Rev. 1 9 9 7 , 97, 1515-
1566.
(11) Uchiyama, S.; Santa, T.; Imai, K. Analyst 2 0 0 0 , 125, 1839-1845.
(12) Jones, G.; Griffin, S. F.; Chol, C. Y.; Bergmark, W. R. J. Org. Chem. 1 9 8 4 ,
49, 2705-2708.
(13) Seidel, C. A. M.; Schulz, A.; Sauer, M. H. M. J. Phys. Chem. 1 9 9 6 , 100,
5541-5553.
10.1021/ac0201225 CCC: $22.00 © 2002 American Chemical Society
Published on Web 07/12/2002
Analytical Chemistry, Vol. 74, No. 16, August 15, 2002 4089

having an aminobenzene moiety had a markedly small Φ value14
(intramolecular quenching), we estimated that these quenchings
were derived from the electron transfer between the aromatic
moiety (π-donor) and the fluorophore. Assuming that the mech-
anism of quenching by the intermolecular electron transfer was
identical with that of the fluorescence “off” switching by the
intramolecular electron transfer, we first investigated the inter-
molecular electron transfer between some benzofurazan com-
pounds and other aromatic compounds using Stern-Volmer plots.
As a result, fluorescence quenching of the benzofurazan com-
pounds was observed by the addition of some aromatic com-
pounds such as aniline and indole. We then synthesized the
benzofurazan compounds bearing an aromatic substituent group
to investigate the intramolecular electron transfer between the
benzofurazan skeleton and an aromatic moiety. Moreover, mo-
lecular orbital theory was adopted as an explanation of intramo-
lecular quenching by electron transfer, and we reached the
following conclusion: the Φ values of the benzofurazan com-
pounds bearing an aromatic moiety could be estimated by
investigation of the electron-donating ability of the aromatic
moiety. This ability was indicated by a HOMO energy (the
semiempirical PM3/ COSMO calculation) or a K value (the Stern-
Volmer plotting). And finally, we succeeded in developing the first
PET reagent for peroxycarboxylic acids, 4-(p-N,N-dimethylamino)-
benzylamino-7-nitro-2,1,3-benzoxadiazole (NBD-Bz-p-NMe2), by
predicting the Φ values using the K values and the HOMO
energies. Recently, it was also reported that the Φ values of the
fluorescein derivatives bearing a benzoic acid moiety could be
estimated by calculating the HOMO energies and then a PET
reagent for singlet oxygen, 9-[2-(3-carboxy-9,10-dimethyl)anthryl]-
6-hydroxy-3H-xanthen-3-one (DMAX), was developed.15
As mentioned above, the method for predicting the Φ values
of the compounds bearing a π-donor was established and new
PET reagents have been developed. We considered that the
establishment of a method for predicting the Φ values of the
compounds bearing an n-donor, which have not been yet reported,
was also valuable for developing new PET reagents. The fluores-
cence quenching rate constants for some anthracene compounds
by the addition of the aliphatic compounds having n-electrons
increased with the decrease in the ionization potential of the
aliphatic compounds5-8 (intermolecular quenching). Furthermore,
the PET process was observed on some compounds having an
aliphatic amino moiety and a fluorophore such as anthracene,16
1,8-naphthalimide,17,18 and benzofurazan19,20 (intramolecular quench-
ing). Taking these reports into consideration, the method for
predicting the Φ values of the compounds bearing a π-donor11,15
can be applied to the prediction of the Φ values of the compounds
bearing an n-donor. In this study, we first synthesized the benzo-
furazan compounds bearing an aliphatic substituent group having
an n-electron to investigate the relationship between the intramo-
lecular electron transfer and their Φ values. We then investigated
the intermolecular electron transfer by obtaining the HOMO
energies of the aliphatic moiety (the semiempirical PM3/ COSMO
calculation) and the K values (the Stern-Volmer plotting) and
tried to establish the prediction method.
The second purpose of this study is to develop a new PET
reagent for peroxyacetic acid (PAA) based on the prediction. NBD-
Bz-p-NMe2, the above-mentioned PET reagent for peroxycarboxy-
lic acids, reacted only with m-chloroperoxybenzoic acid (MCPBA)
and did not react with any other peroxycarboxylic acids.11 Among
the peroxycarboxylic acids, PAA is one of the most practical
peroxycarboxylic acids, since it has excellent antimicrobial,
bactericidal, fungicidal, virucidal, and bleaching properties21 and
is used in industrial processes, including disinfection in the food
and beverage industries and the bleaching of paper and textiles.22
Various methods, such as titration,23,24 photometry,25 electrochemi-
cal sensing,26 and chromatographic methods,22,27,28 have been used
to determine the concentration of PAA. Among these methods,
the chromatographic methods offer both high selectivity and low
detection limits. Di Furia et al. developed a gas chromatographic
method for the determination of PAA using methyl-p-tolyl sulfide
(MTS) as a precolumn derivatizing reagent.27 MTS is oxidized to
the corresponding sulfoxide, methyl-p-tolyl sulfoxide (MTSO).
This method gave accurate results, but it was time-consuming
due to the need for an extraction step with chloroform to obtain
a solution suitable for gas chromatography. This problem could
be avoided by employing HPLC.28 The detection of MTSO was
performed at 230 nm using a UV-visible detector. However, the
colored matrix components interfered with the detection of MTSO
at the low detection wavelength, so the azo dye functionalized
sulfide reagent, 2-[(3-{2-[4-amino-2-(methylsulfanyl)phenyl]-1-
diazenyl}phenyl)sulfonyl]-1-ethanol (ADS), was developed as a
new derivatization reagent.22 The HPLC method for the determi-
nation of PAA using ADS at 410 nm provided lower limits of
detection and higher selectivity toward colored matrix compo-
nents. However, to detect PAA more accurately and sensitively,
it is necessary to adopt fluorescence detection. Therefore, we
finally tried to develop a new PET reagent for PAA based on the
established method for predicting Φ values and apply it to HPLC
analysis.
EXPERIMENTAL SECTION
Materials. Water was purified using a Milli-Q reagent system
(Millipore). Acetonitrile and methanol were of HPLC grade (Kanto
Chemicals). Acecide (Saraya) was used as the disinfectant sample.
All other reagents were of guaranteed reagent grade and used
without further purification.
Apparatus. Mass spectra were measured using a Hitachi
M-1200H mass spectrometer (atmospheric pressure chemical
(14) Munkholm, C.; Parkinson, D. R.; Walt, D. R. J. Am. Chem. Soc. 1 9 9 0 , 112,
2608-2612.
(15) Tanaka, K.; Miura, T.; Umezawa, N.; Urano, Y.; Kikuchi, K.; Higuchi, T.;
Nagano, T. J. Am. Chem. Soc. 2 0 0 1 , 123, 2530-2536.
(16) Beeson, J. C.; Huston, M. E.; Pollard, D. A.; Venkatachalam, T. K.; Czarnik,
A. W. J. Fluoresc. 1 9 9 3 , 3, 65-68.
(21) Swern, D. Organic Peroxides; Wiley-Interscience: New York, 1970; p 360.
(22) Effkemann, S.; Karst, U. Analyst 1 9 9 8 , 123, 1761-1765.
(23) Grssnspan, F. P.; MacKellar, D. G. Anal. Chem. 1 9 4 8 , 20, 1061-1063.
(24) Swern, D. Org. React. 1 9 5 3 , 7, 378-433.
(17) Ramachandram, B.; Sankaran, N. B.; Karmakar, R.; Saha, S.; Samanta, A.
Tetrahedron 2 0 0 0 , 56, 7041-7044.
(18) Ramachandram, B.; Saroja, G.; Sankaran, N. B.; Samanta, A. J. Phys. Chem.
B 2 0 0 0 , 104, 11824-11832.
(25) Pinkernell, U.; Luke, H. J.; Karst, U. Analyst 1 9 9 7 , 122, 567-571.
(26) Awad, M. I.; Harnoode, C.; Tokuda, K.; Ohsaka, T. Anal. Chem. 2 0 0 1 , 73,
1839-1843.
(19) Ramachandram, B.; Samanta, A. J. Phys. Chem. A 1 9 9 8 , 102, 10579-10587.
(20) Onoda, M.; Uchiyama, S.; Santa, T.; Imai, K. Luminescence 2 0 0 2 , 17, 11-
14.
(27) Di Furia, F.; Prato, M.; Quintily, U.; Salvagno, S.; Scorrano, G. Analyst 1 9 8 4 ,
109, 985-987.
(28) Pinkernell, U.; Karst, U.; Cammann, K. Anal. Chem. 1 9 9 4 , 66, 2599-2602.
4090 Analytical Chemistry, Vol. 74, No. 16, August 15, 2002

1
ionization (APCI) system). Melting points were measured on a
Yanagimoto Micro Point apparatus and were uncorrected. Proton
nuclear magnetic resonance (1H NMR) spectra were obtained
using a JEOL LA-500 spectrometer. The J values are given in hertz.
UV-visible absorption spectra (30 µM) were measured using a
Jasco Ubest-50 spectrometer. Fluorescence spectra (50 nM-30
µM) were measured using a Hitachi F-4010 fluorescence spec-
trometer.
Determination of Fluorescence Quantum Yield (Φ). The
Φ values were determined using quinine sulfate in 0.1 M sulfuric
acid (ΦQ ) 0.55 with excitation of 366 nm) as a standard29 using
the following equation30
H)+). mp: 145-146 °C. H NMR (CD3OD): δ 3.11 (t, 2H), 3.69
(br, 2H), 6.35 (d, 1H, J ) 8.9), 8.44 (d, 1H, J ) 8.9).
4 -(2 -Acetylaminoethylamino)-7 -nitro-2 ,1 ,3 -benzoxadiaz-
ole (NBD-NH(CH2 )2 NHAc (5 )). NBD-NH(CH2)2NH2 (37 mg,
0.17 mmol) was dissolved in acetonitrile (10 mL). After the
addition of acetic anhydride (0.5 mL), the solution was stirred at
room temperature for 1 h. The reaction mixture was evaporated
to dryness under reduced pressure, and the residue was chro-
matographed on silica gel with ethyl acetate-methanol (19:1) to
afford 5 (24 mg, yield 55%) as yellow crystals. APCI-MS: m/ z
266 ((M + H)+). mp: 232-234 °C. 1H NMR (CD3OD): δ 1.84 (s,
3H), 3.42 (t, 2H), 3.55 (br, 2H), 6.33 (d, 1H, J ) 8.9), 8.44 (d, 1H,
J ) 8.9). Found: C, 45.29; H, 4.19; N, 26.39. Calcd for
C10H11N5O4: C, 45.28; H, 4.18; N, 26.41.
ΦS ) ΦQ(FS/ FQ)(AQ/ AS)(nS/ nQ)2
(1)
4 -(2 -Acetylm ethylam inoethylam ino)-7 -nitro-2 ,1 ,3 -ben-
zoxadiazole (NBD-NH(CH2 )2 NMeAc (6 )). A procedure similar
to that for 5 yielded 25% of a product as orange crystals. APCI-
MS: m/ z 280 ((M + H)+). mp: 238-240 °C. H NMR (CDCl3):
δ 2.15 (s, 3H), 3.12 (s, 3H), 3.63 (br, 2H), 3.81 (t, 2H), 6.11 (d,
1H, J ) 8.9), 8.46 (d, 1H, J ) 8.9).
where F is the area under the fluorescence spectra, A is the
absorbance, n is the refractive index of the solvent, and the
subscripts Q and S represent quinine sulfate and sample, respec-
tively.
1
Synthesis. 4-Methylamino-7-nitro-2,1,3-benzoxadiazole (NBD-
NHMe),31 4-methylthio-7-aminosulfonyl-2,1,3-benzoxadiazole (ABD-
SMe),31 4-acetylamino-7-phenylsulfonyl-2,1,3-benzoxadiazole (PSBD-
NHAc),31 4-phenylthio-7-amino-2,1,3-benzoxadiazole (PTBD-NH2),31
and 4-(N,N-dimethylethylenediamino)-7-nitro-2,1,3-benzoxadiazole
(NBD-NH(CH2)2NMe2 (1 ))19 were synthesized as previously
described.
4 -(2 -Hydroxyethylam ino)-7 -nitro-2 ,1 ,3 -benzoxadiazole
(NBD-NH(CH2 )2 OH (7 )). A procedure similar to that for 2
yielded 60% of a product as orange crystals. APCI-MS: m/ z 225
((M + H)+). mp: 153 °C. H NMR (CD3OD): δ 3.57 (br, 2H),
3.76 (t, 2H), 6.32 (d, 1H, J ) 8.9), 8.42(d, 1H, J ) 8.9). Found: C,
42.90; H, 3.71; N, 24.73. Calcd for C8H8N4O4: C, 42.86; H, 3.60;
N, 24.99.
1
4 -(N,N-Diethylethylenediam ino)-7 -nitro-2 ,1 ,3 -benzoxa-
diazole (NBD-NH(CH2 )2 NEt2 (2 )). 4-Chloro-7-nitro-2,1,3-ben-
zoxadiazole (NBD-Cl) (100 mg, 0.50 mmol) was dissolved in
acetonitrile (20 mL). After the addition of N,N-diethylethylenedi-
amine (70 µL) in acetonitrile (20 mL), the solution was stirred at
room temperature for 30 min. The reaction mixture was evapo-
rated to dryness under reduced pressure, and the residue was
chromatographed on silica gel with dichloromethane-methanol
(19:1) to afford 2 (54 mg, yield 39%) as brown crystals. APCI-MS:
4 -(2 -Methoxyethylam ino)-7 -nitro-2 ,1 ,3 -benzoxadiazole
(NBD-NH(CH2 )2 OMe (8 )). A procedure similar to that for 2
yielded 37% of a product as orange crystals. APCI-MS: m/ z 239
((M + H)+). mp: 170-172 °C. 1H NMR (CDCl3): δ 3.43 (s, 3H),
3.65 (t, 2H), 3.72 (t, 2H), 6.18 (d, 1H, J ) 8.6), 6.49 (br, 1H), 8.48
(d, 1H, J ) 8.6). Found: C, 45.27; H, 4.28; N, 23.28. Calcd for
C9H10N4O4: C, 45.38; H, 4.23; N, 23.52.
4-(2-Acetoxyethylamino)-7-nitro-2,1,3-benzoxadiazole (NBD-
NH(CH2 )2 OAc (9 )). A procedure similar to that for 5 yielded
85% of a product as orange crystals. APCI-MS: m/ z 267 ((M +
1
m/ z 280 ((M + H)+). mp: 96 °C. H NMR (CD3OD): δ 0.99 (t,
6H), 2.58 (q, 4H), 2.76 (t, 2H), 3.55 (br, 2H), 6.25 (d, 1H, J ) 8.9),
8.40 (d, 1H, J ) 8.9). Found: C, 51.52; H, 6.12; N, 24.79. Calcd for
C12H17N5O3: C, 51.60; H, 6.14; N, 25.08.
1
H)+). mp: 134-136 °C. H NMR (CDCl3): δ 2.06 (s, 3H), 3.72
(q, 2H), 4.40 (t, 2H), 6.18 (d, 1H, J ) 8.6), 6.53 (br, 1H), 8.43 (d,
1H, J ) 8.6). Found: C, 45.20; H, 4.02; N, 20.80. Calcd for
C10H10N4O5: C, 45.12; H, 3.79; N, 21.05.
4 -(2 -Methylaminoethylamino)-7 -nitro-2 ,1 ,3 -benzoxadia-
zole (NBD-NH(CH2 )2 NHMe (3 )). N-Methylethylenediamine
(400 µL) was dissolved in acetonitrile (10 mL). After the addition
of NBD-Cl (200 mg, 1.0 mmol) in acetonitrile (20 mL), the solution
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-
methanol (3:1) to afford 3 (94 mg, yield 39%) as brown crystals.
APCI-MS: m/ z 238 ((M + H)+). mp: 121-122 °C. 1H NMR
(CDCl3): δ 2.48 (s, 3H), 3.01 (t, 2H), 3.50 (br, 2H), 6.13 (d, 1H,
J ) 8.6), 8.48 (d, 1H, J ) 8.6).
4 -(2 -Acetylam inoethylthio)-7 -am inosulfonyl-2 ,1 ,3 -ben-
zoxadiazole (ABD-S(CH2 )2 NHAc (1 0 )). ABD-F (50 mg, 0.23
mmol) was dissolved in a mixture of acetonitrile (2 mL) and
saturated sodium hydrogen carbonate solution (2 mL). After the
addition of 2-aminoethanethiol (19 mg) in acetonitrile (2 mL) and
saturated sodium hydrogen carbonate solution (2 mL), the solution
was stirred at room temperature for 1 h. Hydrochloric acid solution
was then added to the reaction mixture until it became pH 2-3,
the solution was evaporated to dryness under reduced pressure,
and the residue was chromatographed on silica gel with dichlo-
romethane-methanol (3:1) to afford ABD-S(CH2)2NH2. APCI-MS:
4 -Ethylenediamino-7 -nitro-2 ,1 ,3 -benzoxadiazole (NBD-
NH(CH2 )2 NH2 (4 )). A procedure similar to that for 3 yielded
64% of a product as brown crystals. APCI-MS: m/ z 224 ((M +
1
m/ z 275 ((M + H)+). H NMR (CD3OD): δ 3.01 (t, 2H), 3.36 (t,
2H), 7.37 (d, 1H, J ) 7.3), 7.87 (d, 1H, J ) 7.3). ABD-S(CH2)2NH2
(50 mg, 0.18 mmol) was dissolved in acetonitrile (20 mL). After
the addition of acetic anhydride (1 mL), the solution was stirred
at room temperature for 6 h. Acetonitrile was then evaporated
under reduced pressure, and the hydrochloric acid solution was
(29) Melhuish, W. H. N. Z. J. Sci. Technol. B 1 9 5 5 , 37, 142-149.
(30) Arimori, S.; Bell, M. L.; Oh, C. S.; Frimat, K. A.; James, T. D. J. Chem. Soc.,
Perkin Trans. 1 2 0 0 2 , 803-808.
(31) Uchiyama, S.; Santa, T.; Fukushima, T.; Homma, H.; Imai, K. J. Chem. Soc.,
Perkin Trans. 2 1 9 9 8 , 2165-2173.
Analytical Chemistry, Vol. 74, No. 16, August 15, 2002 4091

added to the residue until it became pH 2-3. The solution was
extracted with 3 × 50 mL of dichloromethane, the dichlo-
romethane was evaporated to dryness under reduced pressure,
and the residue was chromatographed on silica gel with ethyl
acetate-methanol (20:1) to afford 1 0 (24 mg, yield 41%) as yellow
on silica gel with ethyl acetate-n-hexane (1:2) to afford 1 3 (20
mg, yield 14%) as white crystals. APCI-MS: m/ z 378 ((M + H)+).
mp: 136-138 °C. 1H NMR (CDCl3): δ 1.28 (t, 3H), 2.62 (q, 2H),
3.46 (s, 2H), 7.51 (m, 2H), 7.58 (m, 1H), 8.16 (d, 2H), 8.26 (d, 1H,
J ) 7.9), 8.38 (d, 1H, J ) 7.9), 9.82 (br, 1H). Found: C, 50.86; H,
4.11; N, 11.14. Calcd for C16H15N3O4S2: C, 50.91; H, 4.01; N, 11.13.
4 -Ethylsulfinylacetylam ino-7 -phenylsulfonyl-2 ,1 ,3 -ben-
zoxadiazole (P SBD-NHCOCH2 SOEt, EP BO (1 4 )). A proce-
dure similar to that for 1 2 yielded 10% of a product as white
1
crystals. APCI-MS: m/ z 317 ((M + H)+). mp: 184 °C. H NMR
(CD3OD): δ 1.84 (s, 3H), 3.31 (t, 2H), 3.44 (t, 2H), 7.43 (d, 1H,
J ) 7.3), 7.86 (d, 1H, J ) 7.3). Found: C, 38.23; H, 4.06; N, 17.57.
Calcd for C10H12N4O4S2: C, 37.97; H, 3.82; N, 17.71.
1
crystals. APCI-MS: m/ z 394 ((M + H)+). mp: 222-223 °C. H
4 -(2 -Hydroxyethylthio)-7 -am inosulfonyl-2 ,1 ,3 -benzoxa-
diazole (ABD-S(CH2 )2 OH (1 1 )). A procedure similar to that
for ABD-S(CH2)2NH2 yielded 93% of a product as yellow crystals.
APCI-MS: m/ z 276 ((M + H)+). mp: 137-138 °C. 1H NMR (CD3-
OD): δ 3.33 (t, 2H), 3.79 (t, 2H), 7.30 (d, 1H, J ) 7.3), 7.84 (d,
1H, J ) 7.3). Found: C, 34.91; H, 3.45; N, 15.22. Calcd for
C8H9N3O4S2: C, 34.90; H, 3.30; N, 15.26.
NMR (CDCl3): δ 1.36 (t, 3H), 2.92 (q, 2H), 3.93 (s, 2H), 7.51 (m,
2H), 7.58 (m, 1H), 8.17 (d, 2H), 8.25 (d, 1H, J ) 7.9), 8.35 (d, 1H,
J ) 7.9).
Computational Methods. We adopted the semiempirical
PM3/ COSMO method as described in our previous report.11 The
keywords, PM3, EF, PRECISE, and EPS ) 37.5, were used for
all calculations. The PM3/ COSMO calculations were carried out
using the program MOPAC 2000 in the WinMOPAC ver. 3.0
package (Fujitsu) with a DynaBook DB65C/ 4RC computer
(Toshiba).
Stern-Volmer P lots. The method described in our previous
report11 was adopted and some modifications were made. To the
acetonitrile solution (2.4 mL) of the benzofurazan compounds
(NBD-NHMe (100 nM), ABD-SMe (1 µM) or PSBD-NHAc (200
nM)), a series of acetonitrile (100 µL) or acetonitrile solutions of
aliphatic compounds (100 µL) were added 12 times and the
fluorescence intensity of this mixture was defined as I0 (control)
or I (test), respectively. The excitation wavelengths were 459 (for
NBD-NHMe), 384 (for ABD-SMe), and 368 nm (for PSBD-NHAc).
The emission wavelengths were 522 (for NBD-NHMe), 510 (for
ABD-SMe), and 471 nm (for PSBD-NHAc). According to the
Stern-Volmer equation (eq 2), the K value was calculated from
4 -(2 -N-Oxydiethylam inoethylam ino)-7 -nitro-2 ,1 ,3 -ben-
zoxadiazole (NBD-NH(CH2 )2 NEt2 O (1 2 )). NBD-NH(CH2)2-
NEt2 (80 mg, 0.29 mmol) was dissolved in chloroform (5 mL).
After the addition of 70% MCPBA (75 mg) in chloroform (5 mL),
the solution was stirred at room temperature for 15 min. The
reaction mixture was evaporated to dryness under reduced
pressure, and the residue was chromatographed on silica gel with
dichloromethane-methanol (3:1) to afford 1 2 (68 mg, yield 80%)
as brown crystals. APCI-MS: m/ z 280 ((M - O + H)+). mp: 86-
87 °C. 1H NMR (CD3OD): δ 1.26 (t, 6H), 3.27-3.40 (m, 4H), 3.55
(t, 2H), 4.80 (br, 2H), 6.21 (d, 1H, J ) 8.9), 8.27 (d, 1H, J ) 8.9).
4 -Ethylthioacetylamino-7 -phenylsulfonyl-2 ,1 ,3 -benzoxa-
diazole (P SBD-NHCOCH2 SEt, EP B (1 3 )). PTBD-NH2 (100
mg, 0.41 mmol) was dissolved in dichloromethane (10 mL). After
the addition of bromoacetyl chloride (100 µL) in dichloromethane
(5 mL) and triethylamine (50 µL) in dichloromethane (5 mL), the
solution was stirred at room temperature for 45 min. The reaction
mixture was evaporated to dryness under reduced pressure, and
the residue was chromatographed on silica gel with dichlo-
romethane-n-hexane (2:1) to afford PTBD-NHCOCH2Br (126 mg,
yield 84%). PTBD-NHCOCH2Br (64 mg, 0.18 mmol) was dissolved
in chloroform (2 mL). After the addition of 70% MCPBA (100 mg)
in chloroform (5 mL), the solution was stirred at room temperature
for 2 h. The chloroform was then evaporated to dryness under
reduced pressure, and a saturated sodium hydrogen carbonate
solution (50 mL) was added to the residue. The solution was
extracted with 3 × 50 mL of dichloromethane, the dichlo-
romethane was evaporated to dryness under reduced pressure,
and the residue was chromatographed on silica gel with ethyl
acetate-n-hexane (2:1) to afford PSBD-NHCOCH2Br (50 mg, yield
71%) as white crystals. APCI-MS: m/ z 395 ((M - H)-). 1H NMR
(CDCl3): δ 4.09 (s, 2H), 7.52 (m, 2H), 7.59 (m, 1H), 8.17 (m, 2H),
8.27 (d, 1H, J ) 7.9), 8.37 (d, 1H, J ) 7.9), 9.13 (br, 1H). Found:
C, 42.65; H, 2.73; N, 10.50. Calcd for C14H10BrN3O4S: C, 42.44; H,
2.54; N, 10.61. PSBD-NHCOCH2Br (150 mg, 0.38 mmol) was
dissolved in N,N-dimethylformamide (DMF) (1 mL). After the
addition of ethanethiol (50 µL) and potassium carbonate (100 mg),
the solution was stirred at room temperature for 3 h. Hydrochloric
acid solution was then added to the reaction mixture until it
became pH 2-3. The solution was extracted with 3 × 50 mL of
dichloromethane, the dichloromethane was evaporated to dryness
under reduced pressure, and the residue was chromatographed
I0/ I ) 1 + K[Q]
(2)
[Q] (the concentration of aliphatic compounds) and the obtained
I0/ I by least-squares analysis. All the least-squares analyses were
carried out using Microsoft EXCEL 2000.
High-P erformance Liquid Chromatography. The high-
performance liquid chromatograph consisted of a Hitachi L-6300
pump, a Hitachi L-1080 fluorescence detector, and a Hitachi D-2500
integrator. The separation for the derivative was studied on an
analytical column, TSKgel ODS-80Ts (150 × 4.6 mm i.d., 5 µm)
(TOSOH). The eluent for the derivative was acetonitrile-water
(2:3). The eluate was monitored with fluorescence detection
(excitation at 362 nm, emission at 479 nm).
Time Course of the Reaction of 1 3 with P AA. 1 3 (100 µM)
was reacted with PAA (10 µM) in acetonitrile at room temperature
or 50 °C for 10, 30, or 60 min. An aliquot (5 µL) of each reaction
mixture was subjected to HPLC. The reaction yields were
determined by comparison with the peak area of authentic 1 4 .
Calibration Curve for the Derivative of 1 3 with P AA. 1 3
(100 µM) was reacted with PAA (0.020, 0.078, 0.31, 1.25, 5, and
20 µM) in acetonitrile at 50 °C for 30 min. An aliquot (5 µL) of
each reaction mixture was subjected to HPLC.
Reaction Yields of the Derivative of 1 3 with Other P er-
oxycarboxylic Acids, Hydrogen P eroxide, and Hydroperox-
ides. With each MCPBA, peroxybenzoic acid, hydrogen peroxide,
4092 Analytical Chemistry, Vol. 74, No. 16, August 15, 2002

Table 1. Fluorescence Characteristics of the Synthesized Benzofurazan Compounds
in benzene
in acetonitrile
in methanol
ꢀ
ꢀ
ꢀ
λab
(nm)
(104 M-1
λem
(nm)
λab
(nm)
(104 M-1
λem
(nm)
λab
(nm)
(104 M-1
λem
(nm)
no.
R
cm-1
)
Φ
cm-1
)
Φ
cm-1
)
Φ
NBD-NH(CH2)2R
1
2
3
4
5
6
7
8
NMe2
NEt2
NHMe
NH2
NHAc
NMeAc
OH
OMe
OAc
451
453
451
453
453
450
452
449
446
448
1.69
1.76
1.56
1.75
1.50
1.51
1.59
1.55
1.57
1.55
511
509
512
512
511
510
510
510
505
508
0.0091
0.0084
0.087
0.23
0.26
0.43
0.33
0.37
0.31
0.41
462
464
462
461
461
461
460
459
457
459
2.06
2.09
1.85
2.06
1.85
2.13
1.92
1.95
1.93
1.85
525
526
524
536
525
526
528
525
523
522
0.00038
0.00017
0.0018
0.014
0.30
0.32
0.29
0.30
0.32
461
464
458
458
461
460
464
463
456
461
2.00
2.04
1.69
2.02
1.93
1.78
2.17
1.97
1.93
1.80
522
520
522
524
528
528
531
531
527
529
0.0051
0.0089
0.046
0.078
0.17
0.16
0.14
0.14
0.17
9
cf. NBD-NHMe
0.39
0.16
ABD-S(CH2)2R
1 0
1 1
NHAc
OH
386
387
387
0.75
0.79
0.84
484
499
493
0.058
0.21
0.33
385
386
384
0.89
0.87
0.88
501
508
510
0.016
0.083
0.097
384
384
384
0.79
0.86
0.86
505
508
509
0.017
0.028
0.017
cf. ABD-SMe
compounds, the benzofurazan skeleton and an aliphatic moiety
having an n-electron (R) are connected through the short
methylene spacer. A previous paper reported that the PET process
occurred more efficiently as the spacer length became shorter.20
NBD-NHMe and ABD-SMe were also synthesized as the com-
pounds without an aliphatic moiety having an n-electron (R). The
fluorescence characteristics of the synthesized compounds in
benzene, acetonitrile, and methanol are shown in Table 1. The
absorption characteristics and the maximum emission wave-
lengths of NBD-NH(CH2)2R and ABD-S(CH2)2R are similar to
those of NBD-NHMe and ABD-SMe, respectively. However, the
Φ values of the compounds bearing a tertiary or secondary amino
moiety in an aliphatic subsituent group (1 -3 ) are very small
compared with those of NBD-NHMe in all the solvents. These
results indicated that the intramolecular electron transfer (the PET
process) occurred between the excited benzofurazan skeleton and
the aliphatic moiety (R) in compounds having small Φ values,
and the tertiary and secondary amino moiety would work as an
excellent n-donor. NBD-NH(CH2)2NH2 (4 ) is weakly fluorescent
in acetonitrile and methanol, whereas it fluoresces in benzene.
These results could be explained by a previous report33 that the
PET process occurred more efficiently in polar solvents than in
nonpolar solvents.
Method for P redicting the Φ Values of the Benzofurazan
Compounds Bearing an Aliphatic Substituent Group Having
an n-Electron. First, the molecular orbital calculation was carried
out to obtain the HOMO energies of the aliphatic compounds
having an n-electron (RCH3 or RC2H5) in acetonitrile. The
compounds, RCH3 or RC2H5, corresponded to the aliphatic moiety
(R) of the synthesized NBD compounds (1 -9 ) (Table 2). For
example, triethylamine (Et3N) was chosen as the aliphatic
compound corresponding to the aliphatic moiety of NBD-NH-
(CH2)2NEt2 (2 ), and similarly, N-methylacetamide (CH3NHAc)
was chosen corresponding to that of NBD-NH(CH2)2NHAc (5 ).
The calculated HOMO energy of RCH3 or RC2H5 denotes the
Figure 1. Structures of the benzofurazan compounds used in this
study.
tert-butyl hydroperoxide, or cumene hydroperoxide (20 µM), 1 3
(100 µM) was reacted in acetonitrile at 50 °C for 30 min. An aliquot
(5 µL) of each reaction mixture was subjected to HPLC. The
reaction yields were determined by comparison with the peak area
of authentic 1 4 .
Analysis of P AA in Disinfectant Sample. The concentration
of PAA in disinfectant sample was determined. The disinfectant
samples were diluted with acetonitrile to an appropriate concentra-
tion for analysis. The diluted disinfectant sample was reacted with
1 3 (100 µM) at 50 °C for 30 min, and an aliquot (5 µL) of reaction
mixture was subjected to HPLC. The result was compared with
that of two-step titration method.24 For the accuracy and precision
study, the diluted disinfectant sample was added to acetonitrile
containing 1 and 10 µM PAA and reacted with 1 3 (100 µM) at 50
°C for 30 min. An aliquot (5 µL) of each reaction mixture was
subjected to HPLC.
RESULTS AND DISCUSSION
Fluorescence Characteristics of the Benzofurazan Com-
pounds Bearing an Aliphatic Substituent Group Having an
n-Electron. Nine NBD compounds (NBD-NH(CH2)2R, 1 -9 ) and
two ABD compounds (ABD-S(CH2)2R, 1 0 , 1 1 ) were synthesized
for this study (Figure 1). The benzofurazan compounds having
an NBD-NH- moiety and ABD-S- moiety were selected as the
fluorophore, since they had relatively large Φ values.32 Compounds
1 -1 1 have an n-electron in the aliphatic moiety (R). In these
(32) Uchiyama, S.; Santa, T.; Imai, K. J. Chem. Soc., Perkin Trans. 2 1 9 9 9 , 2525-
2532.
(33) Bissell, R. A.; de Silva, A. P.; Fernando, W. T. M. L.; Patuwathavithana, S.
T.; Samarasinghe, T. K. S. D. Tetrahedron Lett. 1 9 9 1 , 32, 425-428.
Analytical Chemistry, Vol. 74, No. 16, August 15, 2002 4093

Table 2. HOMO Energy of the Aliphatic Compound (RCH3 or RC2H5) Corresponding to the Aliphatic Moiety (R) of the
NBD Compound
HOMO energy of the corresponding
aliphatic compd
Φ value of the NBD-NH(CH2)2R
corresponding aliphatic compd
HOMO (eV)
no.
R
Φ
triethylamine (Et3N)
trimethylamine (Me3N)
dimethylamine (Me2NH)
methylamine (CH3NH2)
dimethyl ether (Me2O)
-9.657
-9.753
2
1
3
4
8
7
5
6
9
NEt2
0.00017
0.00038
0.0018
0.014
0.30
0.29
0.30
0.32
0.32
NMe2
NHMe
NH2
OMe
OH
NHAc
NMeAc
OAc
-9.827
-9.977
-11.066
-11.380
-11.429
-11.509
-11.857
methanol (CH3OH)
N-methylacetamide (CH3NHAc)
N,N-dimethylacetamide (Me2NAc)
acetic acid methyl ester (CH3OAc)
Table 3. K Value of the Aliphatic Compound (RCH3 or RC2H5) Corresponding to the Aliphatic Moiety (R) of the
Benzofurazan Compound
K value of the corresponding aliphatic compd
Φ value of the compdc
c
K
fluorophore
added aliphatic compd
triethylamine (Et3N)
dimethylamine (Me2NH)
trimethylamine (Me3N)
methylamine (CH3NH2)
N-methylacetamide (CH3NHAc)
N,N-dimethylacetamide (Me2NAc)
methanol (CH3OH)
diethyl ether (Et2O)
ethyl acetate (EtOAc)
N-methylacetamide (CH3NHAc)
methanol (CH3OH)
(mM)a
rb
(M-1
)
no.
R
Φ
NBD-NHMe
NBD-NHMe
NBD-NHMe
NBD-NHMe
NBD-NHMe
NBD-NHMe
NBD-NHMe
NBD-NHMe
NBD-NHMe
ABD-SMe
20
20
20
0.999
0.998
0.996
0.992
0.989
0.979
d
98.0
74.4
35.9
23.8
0.2
0.1
d
d
d
1.9
0.3
2
3
1
4
5
6
7
8
NEt2
NHMe
NMe2
NH2
NHAc
NMeAc
OH
OMe
OAc
NHAc
OH
0.00017
0.0018
0.00038
0.014
0.30
0.32
0.29
0.30
0.32
50
1000
1000
1000
1000
1000
1000
1000
d
d
9
1 0
1 1
0.999
0.995
0.013
0.083
ABD-SMe
a
Concentration of the added aliphatic compound. b Correlation coefficient. c NBD-NH(CH2)2R for compounds 1 -9 ; ABD-S(CH2)2R for 1 0 and
1 1 . d Quenching was not observed.
electron-donating ability of the aliphatic moiety (R). As shown in
Table 2, the Φ values of the NBD compounds (1 -9 ) bearing an
aliphatic moiety (R), which has a high electron-donating ability
(i.e., the calculated HOMO energy of RCH3 or RC2H5 is large),
are greatly reduced, similar to those of the NBD compounds
bearing an aromatic moiety.11 Therefore, the Φ values of the
benzofurazan compounds bearing an aliphatic moiety having an
n-electron (R) are predictable by calculating the HOMO energies
of the corresponding aliphatic compounds (RCH3 or RC2H5).
Next, the Stern-Volmer plotting was carried out to obtain the
K values of the aliphatic compounds having an n-electron (RCH3
or RC2H5) in acetonitrile corresponding to the aliphatic moiety
(R) of the synthesized compounds (1 -1 1 ) (Table 3). The
compound with the larger K value has a higher quenching ability.
As shown in Table 3, the Φ values of compounds 1 -1 1 bearing
an aliphatic moiety (R), which has a high quenching ability (i.e.,
the K value of RCH3 or RC2H5 is large), are reduced. These results
indicated that the mechanism of the intramolecular electron
transfer between the benzofurazan skeleton and an aliphatic
moiety (R) was identical with that of the intermolecular electron
transfer between the corresponding benzofurazan compound and
aliphatic compound (RCH3 or RC2H5), and the Φ values of the
benzofurazan compounds bearing an aliphatic moiety having an
n-electron (R) are also predictable by obtaining the K values of
the corresponding aliphatic compounds (RCH3 or RC2H5).
Design of New P ET Reagents for P AA and Application to
HP LC Analysis. The above results enable us to design the PET
reagents by obtaining the HOMO energies using the molecular
orbital calculations and the K values with Stern-Volmer plotting.
First, an aliphatic moiety (R), which has a high quenching ability
(i.e., RCH3 or RC2H5 has a large HOMO energy or a large K value),
must be selected as a reacting moiety of the PET reagent. We
should then confirm that an aliphatic moiety corresponding to the
reacted moiety has a low quenching ability (i.e., a small HOMO
energy or a small K value). For example, methylamine (RCH3, R
) NH2) has a large HOMO energy (-9.977 eV; see Table 2) and
a large K value for NBD-NHMe (K ) 23.8; see Table 3) and the
acetylated derivative of methylamine, N-methylacetamide, has a
small HOMO energy (-11.429 eV, Table 2) and a small K value
for NBD-NHMe (K ) 0.2, Table 3), suggesting that NBD-NH-
(CH2)2NH2 can be a PET reagent for carboxylic acids such as
acetic acid. Actually, 4 (Φ ) 0.014) was weakly fluorescent and
its derivative with acetic acid, 5 (Φ ) 0.30), was strongly
fluorescent in acetonitrile as indicated in Table 1.
In the same way, we designed a new PET reagent for PAA.
11
We first tried to improve the previously reported NBD-Bz-p-NMe2
with respect to its reactivity to peroxycarboxylic acids. NBD-Bz-
p-NMe2 has a N,N-dialkylaniline moiety as a reacting moiety, but
it reacted only with MCPBA and did not react with the less
reactive peroxycarboxylic acids. Since a trialkylamine moiety
seemed to be more reactive than an N,N-dialkylaniline moiety,
NBD-NH(CH2)2NEt2 (2 ) was chosen, expecting to be oxidized
by PAA to the corresponding amine N-oxide, NBD-NH(CH2)2-
NEt2O (1 2 ). The calculated HOMO energy of triethylamine (as
4094 Analytical Chemistry, Vol. 74, No. 16, August 15, 2002

Figure 2. Synthesis of EPB (13) and its derivative, EPBO (14).
a reacting moiety, -9.657 eV, Table 2) was large, whereas that of
triethylamine N-oxide (as a reacted moiety, -11.356 eV) was small.
Therefore, it was estimated that NBD-NH(CH2)2NEt2 (2 ) would
be a PET reagent for PAA. As indicated in Table 1, 2 did not
fluoresce (Φ ) 0.00017) in acetonitrile. 1 2 was then synthesized
to obtain the fluorescence property. As expected, 1 2 fluoresced
(Φ ) 0.051, λab ) 474 nm, λem ) 534 nm) in acetonitrile and the
Φ value of 1 2 was 294 times greater than that of 2 . The
fluorescence properties of 2 and 1 2 in water were also obtained
to apply the reagent, 2 , to the reversed-phase HPLC analysis,
resulting in the fact that the Φ value of the derivative, 1 2 (Φ )
0.021, λab ) 469 nm, λem ) 531 nm), was smaller than that of the
reagent, 2 (Φ ) 0.026, λab ) 463 nm, λem ) 528 nm). Considering
the pKa value of triethylamine (pKa ) 10.7), the electron transfer
from the n-electron of the -NEt2 moiety to the benzofurazan
skeleton would be suppressed because the -NEt2 moiety would
be protonated in water. These results indicated that this reagent
was not suitable for reversed-phase HPLC analysis.
Next, the sulfide moiety was chosen as a reacting moiety with
PAA, since the reacting moiety of the conventional derivatization
reagents for PAA consisted of a sulfide moiety.22,27,28 First, the
NBD compounds bearing a sulfide moiety (R ) SEt), NBD-NH-
(CH2)2SEt, was chosen as a reagent, expecting to be oxidized by
PAA to the corresponding sulfoxide, NBD-NH(CH2)2SOEt. As the
HOMO energy of methyl ethyl sulfoxide (as a reacted moiety,
-9.826 eV) was smaller than that of methyl ethyl sulfide (RCH3,
R ) SEt) (as a reacting moiety, -9.751 eV), it was estimated that
the Φ value of NBD-NH(CH2)2SOEt would be larger than that of
NBD-NH(CH2)2SEt. However, considering the K value of methyl
ethyl sulfide for NBD-NHMe (K ) 1.8), NBD-NH(CH2)2SEt would
be fluorescent, so NBD-NHMe was not suitable for the fluoro-
phore of the PET reagent having a sulfide moiety. The PSBD
compound bearing a sulfide moiety (R ) SEt), 1 3 (see Figure
2), was then chosen, since PSBD-NHAc also had a large Φ value
(Φ ) 0.45 in acetonitrile, measured in this study). From the K
value of methyl ethyl sulfide (RCH3, R ) SEt) for PSBD-NHAc
(K ) 60.1) and that of dimethyl sulfoxide (RCH3, R ) SOMe, easily
obtained, in place of R ) SOEt) for PSBD-NHAc (K ) 8.5), it was
expected that the reagent, 1 3 , would not fluoresce and its
derivative, 1 4 , would fluoresce. Finally, 1 3 and 1 4 were synthe-
sized (Figure 2), and their fluorescence characteristics in aceto-
nitrile are shown in Table 4. As expected, the Φ value of 1 4 was
Table 4. Fluorescence Characteristics of 13 (reagent)
and 14 (derivative)
λab
λem
solvent
compd
(nm) (nm)
Φ
ratioa
254
acetonitrile
1 3
1 4
1 3
EPB
EPBO
EPB
367
365
367
468
466
483
0.0014
0.34
0.00049
acetonitrile-
water (2:3)
53
1 4
EPBO
362
479
0.026
a
The ratio of the Φ value of EPBO (1 4 ) to the Φ value of EPB
(1 3 ).
254 times greater than that of 1 3 in acetonitrile. The fluorescence
characteristics in water could not be obtained due to the poor
solubility of 1 3 and 1 4 , but considering the pKa value of the
dialkyl sulfide (pKa ) -5.3), the PET process occurring in 1 3
would not be suppressed even in water. Therefore, 1 3 seemed
to have the appropriate fluorescence property of a PET reagent
for PAA. This is the first report showing that the PET process
occurs between a fluorophore and an aliphatic sulfide moiety.
The reactivity of 1 3 as a derivatization reagent for PAA was
then tested using reversed-phase HPLC. The peak for 1 4 was
fluorometrically detected at ∼6 min using acetonitrile-water (2:
3) as the eluent, as shown in Figure 3. The fluorescence
characteristics of 13 and 14 in this eluent are also shown in Table
4. In the HPLC eluent, the Φ value of 1 4 is 53 times greater than
that of 1 3 . The time course study of the derivatization of PAA
(10 µM) with 1 3 (100 µM) was performed at room temperature
or at 50 °C. The reaction was completed at 50 °C within 30 min in
acetonitrile to give a fluorescent derivative of 1 4 (yield, 99.8%),
whereas the reaction was not completed at room temperature for
60 min (yield: 84.3%). Thus, this condition (50 °C, 30 min) was
adopted for the derivatization of 1 3 with PAA. The chromatogram
of the reaction mixture of 13 (100 µM) and PAA (20 µM) is shown
in Figure 3. The peak of the reagent, 1 3 , (b) is much smaller
than that of the derivative, 1 4 , (a), though excess reagent was
included in the reaction mixture. The calibration curve for PAA
(0.020-20 µM) with 1 3 (100 µM) was linear over the range from
390 fmol (78 nM) to 100 pmol (20 µM) (r > 0.999). The detection
limit (S/ N ) 3) of the derivative of the developed 1 3 was 105
fmol (21 nM), which was lower than that of the conventional
reagent.22,28 The reaction yields of the derivative of 1 3 (100 µM)
Analytical Chemistry, Vol. 74, No. 16, August 15, 2002 4095

Table 5. Precision and Accuracy in the Determination
of PAA in Disinfectant Sample
added amounts (µM)
0
1
10
Intraday (n ) 5)
2.20
5.4
found (µM)
RSD (%)
3.29
5.0
12.23
1.5
accuracy (%)
109.6
100.3
Interday (n ) 5)
2.21
5.8
found (µM)
RSD (%)
3.13
4.3
11.98
2.5
accuracy (%)
92.1
97.7
5. Both intraday and interday accuracy and precision were
satisfactory for the analysis of PAA in real samples. From these
results, 1 3 was indicated to be an excellent reagent for PAA with
high sensitivity and selectivity.
CONCLUSIONS
The Φ values of the benzofurazan compounds bearing an
aliphatic substituent group having an n-electron became predict-
able based on the molecular orbital calculations and the Stern-
Volmer plotting. It is expected that various new PET reagents
can be developed using this method for prediction. The fluorescent
reagents have been used not only for HPLC analysis but also for
bioimaging, so this method can be widely applicable to the
development of the fluorescent reagents, such as derivatization
reagents, sensors, probes, and so on. We succeeded in developing
a highly sensitive and selective PET reagent for PAA, 1 3 , based
on this method. In the future, the simultaneous determination of
peroxycarboxylic acids by postcolumn derivatization seems to be
possible because of the fluorescent on-off property of this reagent.
Figure 3. Chromatogram of PAA derivatized with excess 13: (a)
14, 100 pmol, (b) excess 13, 400 pmol; column, TSKgel ODS-80Ts
(150 × 4.6 mm, i.d. 5 µm); eluent, acetonitrile-water (2:3); flow rate,
1.0 mL min-1; detection, excitation at 362 nm, emission at 479 nm.
with MCPBA, peroxybenzoic acid, hydrogen peroxide, tert-butyl
hydroperoxide, and cumene hydroperoxide (20 µM) under the
same conditions were 100.1%, 0.057%, 0.096%, 0.019%, and 0.005%,
respectively, so the cross-reactivity of 1 3 toward hydrogen
peroxide was negligible. The cross-reactivity of the reagent toward
hydrogen peroxide must be remarkably low,22 since PAA solutions
always contain significant amounts of hydrogen peroxide.
The proposed method was applied to the determination of PAA
in disinfectant sample. The determined concentration of PAA in
disinfectant sample by this method was 0.94 ( 0.036 M (mean (
SD, n ) 5). This result was comparable to the result obtained by
the popular two-step titration method24 (0.96 ( 0.055 M; mean (
SD, n ) 5). The accuracy and precision data were shown in Table
Received for review February 25, 2002. Accepted June 3,
2002.
AC0201225
4096 Analytical Chemistry, Vol. 74, No. 16, August 15, 2002