Hydrogen-Bonding Effects on the Fluorescence versus Electron-Transfer-Initiated Chemiluminescence Spectra of the m-Oxybenzoate Ion Derived from a Bicyclic Dioxetane

Article abstract of DOI:10.1021/jo991665q

A comparative spectral study of the fluorescence of the m-oxybenzoate anion versus the electron-transfer-initiated chemiluminescence (CIEEL) of the same anion derived from the bicyclic dioxetanes in various solvents is reported. The present study reveals that the fluorescence of this oxyanion is blue-shifted in protic versus aprotic solvents, while the CIEEL-spectral maximum is independent of the medium. The same phenomenon has been recently observed for the m-oxybenzoate ion derived from CIEEL-active spiroadamantyl dioxetanes. The reported spectral differences between the fluorescence and chemiluminescence emissions cannot be attributed to exciplex formation in the CIEEL process, but result from the differences in hydrogen-bonding effects on the photo- and chemiexcited oxyanion species. The observed solvatochromism is qualitatively rationalized in terms of the semiempirical AM1 calculations.

Full text of DOI:10.1021/jo991665q

2078
J . Org. Chem. 2000, 65, 2078-2082
Hyd r ogen -Bon d in g Effects on th e F lu or escen ce ver su s
Electr on -Tr a n sfer -In itia ted Ch em ilu m in escen ce Sp ectr a of th e
m -Oxyben zoa te Ion Der ived fr om a Bicyclic Dioxeta n e
Waldemar Adam,*,† Masakatsu Matsumoto,‡ and Alexei V. Trofimov*,†,§
Institute of Organic Chemistry, University of Wu¨rzburg, Am Hubland, D-97074 Wu¨rzburg, Germany,
Department of Material Science, Kanagawa University, Tsuchiya, Hiratsuka, Kanagawa, 259-12, J apan,
and Institute of Biochemical Physics, United Institute of Chemical Physics, Russian Academy of Sciences,
ul. Kosygina 4, Moscow 117977, Russia
Received October 25, 1999
A comparative spectral study of the fluorescence of the m-oxybenzoate anion versus the electron-
transfer-initiated chemiluminescence (CIEEL) of the same anion derived from the bicyclic dioxetanes
in various solvents is reported. The present study reveals that the fluorescence of this oxyanion is
blue-shifted in protic versus aprotic solvents, while the CIEEL-spectral maximum is independent
of the medium. The same phenomenon has been recently observed for the m-oxybenzoate ion derived
from CIEEL-active spiroadamantyl dioxetanes. The reported spectral differences between the
fluorescence and chemiluminescence emissions cannot be attributed to exciplex formation in the
CIEEL process, but result from the differences in hydrogen-bonding effects on the photo- and
chemiexcited oxyanion species. The observed solvatochromism is qualitatively rationalized in terms
of the semiempirical AM1 calculations.
Sch em e 1
In tr od u ction
The phenomenon of light emission known as chemically
initiated electron-exchange luminescence (CIEEL)1 con-
stitutes a general process that involves electron-transfer
chemistry.2,3 Soon after it had been reported for high-
energy organic peroxides,4,5 the CIEEL mechanism was
proposed for the firefly bioluminescence.6 The latter
process represents an example of intramolecular CIEEL,
which has served as a basis to develop the most effective
probes for modern chemiluminescence bioassays,7,8 with
the triggerable dioxetanes 19,10 as the CIEEL-active
species (Scheme 1). A generally accepted mechanism of
the CIEEL phenomenon still needs to be established.11
The CIEEL may be generated at will by treatment of the
dioxetanes 1 with an appropriate reagent (trigger) to
release the phenolate ion 2. The formation of 2 is followed
by the intramolecular electron transfer (ET) from the
phenolate moiety to the antibonding σ* orbital of the
peroxide bond (Scheme 1), concomitant with O-O bond
cleavage. The transitory species 3 is likely to yield a
solvent-caged ion-radical pair; the chemiexcitation is
envisaged to take place by electron back-transfer (BET)
in such a pair.12,13
An important advantage of the spiroadamantyl-sub-
stituted dioxetanes 1 for practical use is their thermal
persistence, which makes them easy to handle in bio-
analytical, clinical applications. Also the novel triggerable
bicyclic dioxetanes 6 (Scheme 2) possess remarkable
thermal persistence and appreciable CIEEL efficiency,14
and thereby qualify for chemiluminescence bioassays.
* To whom correspondence should be addressed. Prof. Waldemar
Adam: Tel +49 931 8885340/339. Fax: +49-931-8884756. e-mail:
adam@chemie.uni-wuerzburg.de. Internet: http://www-organik.chemie.
uni-wuerzburg.de.
† University of Wu¨rzburg.
‡ Kanagawa University.
§ Russian Academy of Sciences.
(1) Schuster, G. B.; Horn, K. A. Chemically Initiated Electron-
Exchange Luminescence. In Chemical and Biological Generation of
Excited States; Adam, W., Cilento, G., Eds.; Academic Press: London,
1982; pp 229-247.
(2) (a) Bard, A. J .; Faulkner, L. R. Electrochemical Methods; J ohn
Wiley & Sons: New York, 1980; pp 621-629. (b) Faulkner, L. R. Int.
Rev. Sci.: Phys. Chem. Ser. Two 1976, 9, 213-263. (c) Hercules, D.
M. Acc. Chem. Res. 1969, 2, 301-307.
(3) (a) Weller, A.; Zachariasse, K. J . Chem. Phys. 1967, 46, 4984-
4985. (b) Weller, A.; Zachariasse, K. Chem. Phys. Lett. 1971, 10, 197-
200. (c) Weller, A.; Zachariasse, K. Chem. Phys. Lett. 1971, 10, 424-
427.
(4) Koo, J .-Y.; Schuster, G. B. J . Am. Chem. Soc. 1978, 100, 4496-
4503.
(5) Adam, W.; Cueto, O. J . Am. Chem. Soc. 1979, 101, 6511-6115.
(6) Koo, J .-Y.; Schmidt, S. P.; Schuster, G. B. Proc. Natl. Acad. Sci.
U.S.A. 1978, 75, 30-33.
(7) Adam, W.; Reinhardt, D.; Saha-Mo¨ller, C. R. Analyst 1996, 121,
1527-1531.
(8) Beck, S.; Ko¨ster, H. Anal. Chem. 1990, 62, 2258-2270.
(9) (a) Bronstein, I.; Edwards, B.; Voyta, J . C. J . Biolumin. Chemi-
lumin. 1988, 2, 186. (b) Edwards, B.; Sparks, A.; Voyta, J . C.;
Bronstein, I. J . Biolumin. Chemilumin. 1990, 5, 1-4. (c) Bronstein,
I.; Edwards, B.; Voyta, J . C. J . Biolumin. Chemilumin. 1989, 4, 99-
111. (d) Edwards, B.; Sparks, A.; Voyta, J . C.; Strong, R.; Murphy, O.;
Bronstein, I. J . Org. Chem. 1990, 55, 6225-6229.
(10) (a) Schaap, A. P.; Chen, T.-S.; Handley, R. S.; DeSilva, R.; Giri,
B. P. Tetrahedron Lett. 1987, 28, 1155-1158. (b) Schaap, A. P.;
Handley, R. S.; Giri, B. P. Tetrahedron Lett. 1987, 28, 935-938.
(11) Wilson, T. Photochem. Photobiol. 1995, 62, 601-606.
(12) Adam, W.; Bronstein, I.; Trofimov, A. V. J . Phys. Chem. A 1998,
102, 5406-5414.
(13) Adam, W.; Bronstein, I.; Trofimov, A. V.; Vasil’ev, R. F. J . Am.
Chem. Soc. 1999, 121, 958-961.
10.1021/jo991665q CCC: $19.00 © 2000 American Chemical Society
Published on Web 03/09/2000

Hydrogen-Bonding Effects on Fluorescence vs CIEEL Spectra
J . Org. Chem., Vol. 65, No. 7, 2000 2079
Sch em e 2
spectral maxima (466 nm) in all the solvents we have
used also renders improbable exciplex formation in the
CIEEL process shown in Scheme 1. Nevertheless, a more
rigorous scrutiny of exciplex involvement appears neces-
sary. For this purpose, we have chosen in the present
spectral study the bicyclic dioxetanes 6 (Scheme 2) as
precursors to the CIEEL emitter, namely the m-oxyben-
zoate ion 9, which is akin to the m-oxybenzoate ion 4
produced from the adamantyl-substituted dioxetanes 1
(Scheme 1). The incentive and the relevance of such a
study are the following: First, the CIEEL emitters 4 and
9 possess the same chromophore, i.e, the m-oxybenzoate-
ion moiety, and one may expect similar spectral charac-
teristics and solvatochromism of the fluorescence for both
anions 4 and 9, provided that for the latter no intramo-
lecular exciplex intervenes. Second, contrary to the
m-oxybenzoate ion 4, which is formed from the dioxetanes
1 along with the ketone fragment 5 as separate species
(Scheme 1), the m-oxybenzoate ion 9 is a single reaction
product in the CIEEL-triggering process of the dioxetanes
6 (Scheme 2); the latter fact should facilitate exciplex
formation in view of intramolecularity and, moreover,
escape from the solvent cage is prevented.
Indeed, the only reported example of an exciplex
formed in the CIEEL cleavage of dioxetanes, namely from
a bicyclic indole-derived dioxetane, is of the intramolecu-
lar type, whose spectral emission was strongly dependent
on solvents.16 Thus, the bicyclic dioxetanes 6 may serve
as a probe for intramolecular exciplex involvement in the
triggered CIEEL generation, by elucidation of chemilu-
minescence and fluorescence spectral properties of the
CIEEL emitter 9 in protic and aprotic solvents. Such an
investigation of the solvatochromic effects on the m-
oxybenzoate ion 9 in protic versus aprotic media is of
particular relevance since the bicyclic dioxetane 6, from
which the CIEEL emitter is released by triggering,
should be useful for chemiluminescence bioassays.
Presently we report the results of the solvatochromic
effects on the CIEEL of the dioxetanes 6 and the
fluorescence of the authentic CIEEL emitter 9 (Scheme
2) and compare these spectral data with those previously
obtained for the CIEEL of the dioxetanes 1 and the
fluorescence of CIEEL emitter 4 (Scheme 1) in the same
solvents. The experimental observations are qualitatively
rationalized in terms of semiempirical (AM1) calcula-
tions.
The choice of the trigger and the reaction medium,
which depends on the nature of the protective group (X),
is of prime importance for the rational design of efficient
CIEEL systems. For that reason, a detailed knowledge
on the medium effects for the CIEEL generation is
required. Since the chemiluminescence bioassays are
conducted in aqueous media, the elucidation of the
hydrogen-bonding effects on the CIEEL emission is
particularly relevant.
Recently, we have reported a detailed comparative
spectral study of the hydrogen-bonding effects on the
CIEEL process of the spiroadamantyl-substituted dioxe-
tanes 1 (Scheme 1) and the fluorescence emission of the
methyl m-oxybenzoate ion (4) in protic versus aprotic
media.12 Comparison of the CIEEL spectra of the dioxe-
tanes 1, which have been triggered in protic (H2O, D2O,
and MeOH) and aprotic (MeCN, DMSO) solvents
(Scheme 1), with the fluorescence spectra of the m-
oxybenzoate ion 4, the authentic CIEEL emitter, has
revealed the following intriguing facts: The fluorescence
of the 4 species is blue-shifted (∆λmaxfl ) 51 nm) in protic
versus aprotic solvents, while the CIEEL-spectral maxima
CIEEL
(λmax
) 466 nm) are independent of the reaction
medium and the type of trigger!12 In other words, the
fluorescence of 4 is blue-shifted in protic media relative
to the CIEEL emission, whereas in aprotic solvents the
CIEEL and the fluorescence spectra coincide. These
divergent spectral observations have been rationalized
in terms of different hydrogen-bonding effects on the
photo- and chemiexcited m-oxybenzoate ion 4.12
It was supposed that the second dioxetane cleavage
fragment, namely adamantanone (5), does not play any
significant role in the spectral difference between the
chemiluminescence (CIEEL) and fluorescence emissions
observed in protic media.12 However, two possible reasons
for the involvement of the adamantanone (5) may be
considered, which may cause the above-mentioned spec-
tral shift for the photoexcited emitter 4: (i) The adaman-
tanone (5) fragment, generated in the immediate prox-
imity of the CIEEL emitter 4 (Scheme 1), may protect
the latter from hydrogen bonding through aggregation
in the solvent cage; (ii) the oxybenzoate ion 4 may form
an exciplex with the adamantanone (5). In our previous
analysis of the CIEEL-spectral behavior,12 we had only
examined the latter possibility. Such exciplex involve-
ment was ruled out, since the observed shift between the
CIEEL and the fluorescence spectra did not correlate
with the solvent polarity as would be expected for an
exciplex,15 but rather it depended on whether the solvent
is protic or not.12 Moreover, the coincidence of the CIEEL-
Resu lts a n d Discu ssion
Comparison of the CIEEL spectra of the dioxetanes
6a ,b (Scheme 2), triggered in protic (H2O, D2O, and
MeOH) and aprotic (MeCN, DMSO) solvents, with the
fluorescence spectra of the m-oxybenzoate ion 9, the
authentic CIEEL emitter (Figure 1), has revealed the
same phenomenon as we have observed before for dioxe-
tanes 1: The fluorescence (Figure 1a; Table 1) of the
fl
emitter 9 species is blue-shifted (∆λmax ≈ 50 nm) in protic
versus aprotic media, while the CIEEL-spectral maxima
CIEEL
(λmax
) 467 nm) remain the same (Figure 1; Table
1) in all the solvents. In other words, the solvatochromic
behavior of the oxyanion 9 matches that which we have
observed previously for the anion 4.12
Besides the qualitatively similar spectra for both
oxyanions 4 and 9, also the quantitative data (Table 1)
(14) Matsumoto, M.; Watanabe, N.; Kasuga, N. C.; Hamada, F.;
Tadokoro, K. Tetrahedron Lett. 1997, 38, 2863-2866.
(15) Kavarnos, G. J .; Turro, N. J . Chem. Rev. 1986, 86, 401-449.
(16) Nakamura, H.; Goto, T. Photochem. Photobiol. 1979, 30, 27-
33.

2080 J . Org. Chem., Vol. 65, No. 7, 2000
Adam et al.
F igu r e 1. Normalized CIEEL and fluorescence spectra in protic (a) and aprotic (b) media. (a) CIEEL emissions (at the right) of
the NaOH-triggered (pH 12.6) hydroxy-substituted dioxetane 6b (3.0 × 10-3 M) in H2O, D2O and MeOH at room temperature (ca.
20 °C) and the fluorescence emission (at the left, λex ) 330 nm) of the authentic m-oxybenzoate ion ([9] ) 4.5 × 10-5 M) under the
same conditions. (b) CIEEL emissions in the fluoride-ion-triggered ([n-Bu4NF] ) 6.3 × 10-4 M) decomposition of the dioxetane 6a
(1.1 × 10-4 M) at room temperature (ca. 20 °C) in MeCN and DMSO and the fluorescence emission (λex ) 330 nm) of the
m-oxybenzoate ([9] ) 1.2 × 10-5 M) under the same conditions.
Ta ble 1. CIEEL (λm a xCIEEL), F lu or escen ce (λm a xfl), a n d
Absor p tion (λm a xa bs) Ma xim a of th e Oxya n ion 9 a n d
Sp ectr a l Sh ifts of th e F lu or escen ce (∆λm a xfl) a n d th e
Absor p tion (∆λm a xa bs) Ma xim a in a Va r iety of Solven ts
Rela tive to Wa ter a n d th e F lu or escen ce Qu a n tu m
Yield s (Φfl)
CIEEL
fl
fl
abs
abs
λmax
(nm)
λmax ∆λmax λmax
(nm) (nm) (nm)
∆λmax
b,c
solvent
(nm)
Φfl
D2O
H2O
467
467
416
416
418
467
-
-
312
-
-
0.110 ( 0.010d
0.066 ( 0.004d
0.046 ( 0.003d
0.240 ( 0.010e
0.320 ( 0.010e
312
315
366
389
MeOH 467
MeCN 467
2
3
51
51
54
77
DMSO 467 [469b] 467
a
b
Reference 14. Measured versus quinine bisulfate as the
fluorescence standard ([QBS] ) 10-5 M) in 1 N H2SO4. c Each value
d
is an average of at least four measurements. NaOH was used
as base (pH 12.6). e [n-Bu4NF] ) 6.3 × 10-4 M.
match well. Consequently, only the common m-oxyben-
zoate chromophore of the anions 4 and 9 is responsible
for the solvatochromism in protic versus aprotic solvents.
This spectral agreement not only rules out exciplex
formation in the CIEEL process of the dioxetanes 6
(Scheme 2), but provides experimental evidence for our
supposition12 that adamantanone (5) is also not involved
in the CIEEL process of the dioxetanes 1 (Scheme 1),
neither through aggregation with the CIEEL emitter 4
nor by exciplex formation with the latter species. More-
over, the coincidence of the CIEEL-spectral maxima in
protic and aprotic media, contrary to the fluorescence
emission, suggests that hydrogen bonding does not oper-
ate in the CIEEL emitters 4 and 9, generated cor-
respondingly from the dioxetanes 1b and 6b by triggered
chemiexcitation (Schemes 1 and 2).
F igu r e 2. Pertinent MOs for the electronic excitation of the
crossed-conjugated oxyanions 4 versus 9, as calculated by AM1
method.
in protic versus aprotic media. Both m-oxybenzoate ions
4 and 9 belong to the crossed-conjugated category (Figure
2) and thereby reveal a strong hypsochromic shift in
protic solvents.
Recently, we have established12 that the solvatochromic
effect on aromatic oxyanions caused by hydrogen bonding
in protic solvents depends on the anion structure: The
anions with an extended-conjugated substitution pattern
are insensitive to solvatochromic effects, while the crossed-
conjugated17 anions are subject to considerable solvato-
chromism, as manifested by a strong spectral blue shifts
To rationalize qualitatively the observed solvato-
chromism, it is essential to consider the MO’s involved
in the electronic excitation of the anions of interest.12 In
Figure 2 are displayed the pertinent MO’s for the
excitation of the m-oxybenzoate anions 4 and 9, as
calculated by the semiempirical AM1 method. As one can
see from Figure 2, the orbital coefficients for both 4 and
9 look very similar, except that for the anion 4 the π f
π* excitation constitutes a transition between HOMO and
LUMO, while for the anion 9 it is between HOMO and
LUMO+1. Noteworthy is the pronounced difference
(17) For discussion on extended and crossed conjugation, see: (a)
March, J . Advanced Organic Chemistry, 4th ed.; J ohn Wiley & Sons:
New York, 1992. (b) Phelan, N. F.; Orchin, M. J . Chem. Educ. 1968,
45, 633-637.

Hydrogen-Bonding Effects on Fluorescence vs CIEEL Spectra
J . Org. Chem., Vol. 65, No. 7, 2000 2081
F igu r e 3. Schematic representation of the hydrogen-bonding effects on the photo- and chemiexcited oxyanions 4 and 9.
between the HOMO and the LUMO (LUMO+1) coef-
ficients centered on the phenolate oxygen atom.
Sch em e 3
What is the reason for the different hydrogen-bonding
effects on the photo- and chemiexcited m-oxybenzoates
4 and 9? Negligible LUMO (LUMO+1) versus significant
HOMO coefficients on the phenolate oxygen atom of 4
and 9 (Figure 2) imply that the electron density on this
oxygen atom is dramatically reduced on π f π* excita-
tion, with the consequence that the hydrogen bonding at
this site is significantly diminished in the excited versus
the ground state (cf. Figure 3 and Scheme 3). The greater
stabilization of the ground versus excited state by
hydrogen bonding accounts for the strong blue shift of
the absorption, which is the optical transition between
the relaxed ground and the Franck-Condon excited
states (S0 + hνabs f S1FC) (Figure 3). Conversely, the blue
shift of the fluorescence emission, i.e., the transition
between the relaxed excited state and the Franck-
Condon ground state (S1 f S0FC + hνfl), suggests stronger
stabilization of the S0FC versus S1 state in protic solvents
by hydrogen bonding (Figure 3, Scheme 3). Appreciable
hydrogen bonding in the S0FC state prior to its relaxation
to S0 means that during photoexcitation, despite the low
electron density on the O- site, enough solvent molecules
are still associated with the singlet-excited anions 4 and
9 to be effective for weak hydrogen bonding (Scheme 3).
The following reason may be suggested for this “memory
effect”:12 The strong hydrogen bonding with the O- site
in the ground state of the 4 and 9 anions effectively
localizes the charge so that in the photoexcited state more
electron density (higher LUMO or LUMO+1 coefficients)
is retained on the O- site compared to that calculated
for vacuum (Figure 2). This is not the case for the
chemiexcited 4 and 9 anions formed in the triggering of
the dioxetanes 1 and 6. Indeed, the precursors to the
chemiexcited CIEEL emitters 4 and 9, namely the
diradical species 3 and 8 formed on ET from the O- site
to the peroxide bond (Schemes 1 and 2), are negligibly
hydrogen-bonded at their phenoxyl-radical site (Scheme
3). The lack of appreciable hydrogen bonding on this
radical oxygen atom prior to chemiexcitation means that
the CIEEL emitters 4 and 9, formed directly in the S1
state, are also not subject to hydrogen bonding on the

2082 J . Org. Chem., Vol. 65, No. 7, 2000
Adam et al.
O- site. Consequently, in the CIEEL process both S1 and
S0FC states are negligibly stabilized by hydrogen bonding
and, contrary to photoexcitation, no spectral shift in the
CIEEL emission is observed in protic media.
derived from the dioxetanes 1 (Scheme 1), the CIEEL-
spectral maximum of the dioxetane phenolate 7 obtained
from the dioxetanes 6 (Scheme 2) is also independent of
the medium. This appears to be a general phenomenon
for the phenolate-initiated CIEEL processes.
In conclusion, since the CIEEL spectra of both dioxe-
tane phenolates 2 and 7, released on triggering from the
dioxetanes 1 and 6, match well in all the solvents used,
this observation rules out exciplex formation in the
CIEEL process of the bicyclic dioxetane 6 (Scheme 2) and
justifies our previous supposition that in the CIEEL
process of the dioxetanes 1 (Scheme 1), the adaman-
tanone fragment (5) is in no way involved, neither in
aprotic nor in protic media.12 Therefore, the spectral
difference between the CIEEL and fluorescence emissions
in protic solvents is rationalized in terms of different
hydrogen-bonding effects on photo- and chemiexcited
m-oxybenzoate-ion fragment, the common chromophore
for the 4 and 9 emitters. Finally, as for the phenolate 2
Ack n ow led gm en t . Generous funding by the
Deutsche Forschungsgemeinschaft (Sonderforschungs-
bereich 172 “Moleculare Mechanismen kanzerogener
Prima¨rvera¨nderungen”) is gratefully appreciated. A.V.T.
thanks also the Russian Foundation for Basic Research
(grant N 99-03-32121). We are grateful to A.-M. Krause
for a valuable technical assistance in the synthetic work
and Dr. H. Ihmels for helpful discussions.
Su p p or tin g In for m a tion Ava ila ble: Experimental sec-
tion. This material is available free of charge from the Internet
at http://pubs.acs.org.
J O991665Q