Experimental evidence of the occurrence of intramolecular electron transfer in catalyzed 1,2-dioxetane decomposition

Article abstract of DOI:10.1021/jo1013405

The activation parameters for the thermal decomposition of 13 acridinium-substituted 1,2-dioxetanes, bearing an aromatic moiety, were determined and their chemiluminescence emission quantum yields estimated, utilizing in situ photosensitized 1,2-dioxetane generation and observation of its thermal decomposition kinetics, without isolation of these highly unstable cyclic peroxides. Decomposition rate constants show linear free-energy correlation for electron-withdrawing substituents, with a Hammett reaction constant of Φ = 1.3 ± 0.1, indicating the occurrence of an intramolecular electron transfer from the acridinium moiety to the 1,2-dioxetane ring, as postulated by the intramolecular chemically initiated electron exchange luminescence (CIEEL) mechanism. Emission quantum yield behavior can also be rationalized on the basis of the intramolecular CIEEL mechanism, additionally evidencing its occurrence in this transformation. Both relations constitute the first experimental evidence for the occurrence of the postulated intramolecular electron transfer in the catalyzed and induced decomposition of properly substituted 1,2-dioxetanes.

Full text of DOI:10.1021/jo1013405

pubs.acs.org/joc
Experimental Evidence of the Occurrence of Intramolecular Electron
Transfer in Catalyzed 1,2-Dioxetane Decomposition
Luiz Francisco M. L. Ciscato,†,‡ Fernando H. Bartoloni,‡ Dieter Weiss,† Rainer Beckert,†
and Wilhelm J. Baader*,‡
†
€
Institut fu€r Organische Chemie und Makromolekulare Chemie, Friedrich-Schiller Universitat,
‡
Humboldtstrasse 10, D-07743 Jena, Germany, and Instituto de Quımica, Departamento de Quımica
´
´
Fundamental, Universidade de Sao Paulo, Av. Prof. Lineu Prestes, 748, 05508-000, Sao Paulo, SP, Brazil
~
~
wjbaader@iq.usp.br
Received July 7, 2010
The activation parameters for the thermal decomposition of 13 acridinium-substituted 1,2-dioxetanes,
bearing an aromatic moiety, were determined and their chemiluminescence emission quantum yields
estimated, utilizing in situ photosensitized 1,2-dioxetane generation and observation of its thermal
decomposition kinetics, without isolation of these highly unstable cyclic peroxides. Decomposition rate
constants show linear free-energy correlation for electron-withdrawing substituents, with a Hammett
reaction constant of F = 1.3 ( 0.1, indicating the occurrence of an intramolecular electron transfer
from the acridinium moiety to the 1,2-dioxetane ring, as postulated by the intramolecular chemically
initiated electron exchange luminescence (CIEEL) mechanism. Emission quantum yield behavior can
also be rationalized on the basis of the intramolecular CIEEL mechanism, additionally evidencing its
occurrence in this transformation. Both relations constitute the first experimental evidence for the
occurrence of the postulated intramolecular electron transfer in the catalyzed and induced decom-
position of properly substituted 1,2-dioxetanes.
Introduction
tions, enzymes, reactive species, proteins, and nucleic acids,
among other compounds of biological interest.2,3
Chemiluminescence is defined as light emission produced
by a chemical reaction, where the electronically excited
products generated can show direct light emission or transfer
their excitation energy to suitable acceptors that are respon-
sible for the emission.1 The use of chemiluminescence
in analytical applications has been extensively researched
for decades, and there is a full spectrum of methods using 1,2-
dioxetanes able to precisely detect, in very low concentra-
The thermal, unimolecular decomposition of 1,2-dioxe-
tanes substituted with an acridinium moiety in a spiro
fashion (Scheme 1, 1) is one of the few examples of this kind
of reaction that shows high chemiluminescence quantum
yields (ΦCL), with ΦCL values of up to 0.15 einstein mol-1 4
.
This observation is in contrast with the thermal decomposi-
tion of 1,2-dioxetanes substituted with alkyl or aryl groups,
with ΦCL values typically lower than 1 ꢀ 10-4 einstein
mol-1 5
. The mechanism proposed by Lee and Singer to
(1) Baader, W. J.; Stevani, C. V.; Bastos, E. L. In The Chemistry of
Peroxides; Rappoport, Z., Ed.; Wiley: Chichester, U.K., 2006; Vol. 2,
pp 1211-1278.
(4) Lee, C.; Singer, L. A. J. Am. Chem. Soc. 1980, 102, 3823.
€
Peroxides; Ando, W., Ed.; Wiley: Chichester, U.K., 1992; pp 221-254.
(b) Adam, W.; Baader, W. J. J. Am. Chem. Soc. 1985, 107, 410. (c) Adam,
W.; In Chemical and Biological Generation of Excited States; Adam, W.,
Cilento, G., Eds.; Academic Press: New York, 1982; pp 115-152.
(5) (a) Adam, W.; Heil, M.; Mosandl, T.; Saha-Moller, C. R. In Organic
~
(2) (a) Garcı
´
a-Campana, A. M.; Lara, F. J. Anal. Bioanal. Chem. 2007,
387, 165–169. (b) Tsunoda, M.; Imai, K. Anal. Chim. Acta 2005, 541, 13–23.
(3) Motoyoshiya, J.; Tanaka, T.; Kuroe, M.; Nishii, Y. J. Org. Chem.
2009, 74, 1014.
6574 J. Org. Chem. 2010, 75, 6574–6580
Published on Web 09/08/2010
DOI: 10.1021/jo1013405
r
2010 American Chemical Society

Ciscato et al.
JOCArticle
FIGURE 1. Structures of acridinium-derived 1,2-dioxetanes.
SCHEME 1. Chemiluminescence from Acridinium-Derived 1,2-Dioxetanes
rationalize the chemiexcitation step in N-methylacridine-
substituted 1,2-dioxetane (NMAD) chemiluminescence,4
following the previous work of McCapra et al.,6 postulates
an intramolecular electron transfer from the lone electron
pair of the acridinic nitrogen to the peroxidic ring, followed
by O-O bond cleavage and formation of a pair of radical
ions. The back electron transfer (BET) from the radical
anion to the radical cation produces the carbonyl fragment
N-methylacridone (2, R3 = CH3) in its singlet excited state,
leading to light emission with high quantum yields
(Scheme 1). This mechanistic approach was based on the
proposal of Schuster et al. for the “chemically initiated
electron exchange luminescence” (CIEEL) mechanism,
initially formulated for the catalyzed decomposition of
diphenoyl peroxide,7 1,2-dioxetanones,8 and other cyclic and
linear peroxides.9 Later on, Schaap et al. developed a set of
thermally stable phenoxy-substituted 1,2-dioxetanes,10
whose fast decomposition can be triggered by a chemical
or enzymatic reaction leading to high ΦCL (∼0.20 einstein
mol-1) and high singlet quantum yields (ΦS ∼ 0.50 einstein
mol-1),11 presumably through an intramolecular variation
of the CIEEL sequence, as supposed for the decomposition
of NMADs.4
Despite the early proposal of Lee and Singer,4 there is still
not any conclusive experimental evidence supporting the
proposition that the CIEEL approach can be applied to
rationalize chemiluminescence from NMADs (Scheme 1)4 or
from triggered phenoxy-substituted 1,2-dioxetanes.11 One of
the reasons for this lack of experimental data may be due to
the difficult synthesis and purification of NMADs, a con-
sequence of their very low thermal stability. The most stable
1,2-dioxetane of this type yet synthesized is the adamantyl
derivate 3, with a ΔGq value for its thermal decomposition at
25 °C of 117.2 kJ mol-1, corresponding to a t1/2 value of ca.
107 s.4 This stability is considered very high, especially in
comparison to other derivatives of the same class, such as
1,2-dioxetane 4, never isolated and postulated as an inter-
mediate in the oxidation of lucigenin (ΔGq = 73.4 kJ mol-1
,
t1/2 ≈ 6 s),12 the unsubstituted derivate 5 (ΔGq = 83.3 kJ
mol-1, t1/2 ≈ 40 s),4 and the phenyl-substituted derivate 6a
(ΔGq = 82.5 kJ mol-1, t1/2 ≈ 33 s) (Figure 1).4
1,2-Dioxetane synthesis is most conveniently performed
by photooxygenation of electron-rich olefins at low
temperatures, and purification can be achieved by low-
temperature column chromatography; however, this pro-
cess is normally susceptible to low isolated yields and
contamination with decomposition products.13 Due to
these difficulties, we developed a dynamic process that
circumvents the necessity to isolate and purify the unstable
1,2-dioxetanes, allowing us to obtain kinetic data for the
thermal decomposition of those labile compounds without
their isolation.14 In this procedure, the cyclic peroxide is
generated in situ from the reaction of the appropriate olefin
(6) (a) McCapra, F.; Beheshti, I.; Burford, A.; Hann, R. A.; Zaklika,
K. A. J. Chem. Soc., Chem. Commun. 1977, 344. (b) McCapra, F. J. Chem.
Soc., Chem. Commun. 1977, 946.
(7) (a) Schuster, G. B.; Schmidt, S. P. Adv. Phys. Org. Chem. 1982, 18, 187.
(b) Schuster, G. B. Acc. Chem. Res. 1979, 12, 366. (c) Koo, J.-Y.; Schuster,
G. B. J. Am. Chem. Soc. 1978, 100, 4496. (d) Koo, J.-Y.; Schuster, G. B.
J. Am. Chem. Soc. 1977, 99, 5403.
(8) (a) Schmidt, S. P.; Schuster, G. B. J. Am. Chem. Soc. 1980, 102, 306.
(b) Turro, N. J.; Chow, M.-F. J. Am. Chem. Soc. 1980, 102, 5058. (c) Adam,
W.; Cueto, O. J. Am. Chem. Soc. 1979, 101, 6511. (d) Schmidt, S. P.; Schuster,
G. B. J. Am. Chem. Soc. 1978, 100, 1966. (e) Adam, W.; Simpson, A.; Yany,
F. J. Phys. Chem. 1974, 78, 2559.
(9) (a) Dixon, B. G.; Schuster, G. B. J. Am. Chem. Soc. 1981, 103, 3068.
(b) Dixon, B. G.; Schuster, G. B. J. Am. Chem. Soc. 1979, 101, 3116.
(10) (a) Zakika, K. A.; Kissel, T.; Thayer, A. L.; Burns, P. A.; Schaap,
A. P. Photochem. Photobiol. 1979, 30, 35. (b) Schaap, A. P.; Gagnon, S. D.
J. Am. Chem. Soc. 1982, 104, 3504.
(12) Lee, K. W.; Singer, L. A.; Legg, K. D. J. Org. Chem. 1976, 41, 2685.
(13) Kopecky, K. R. In Chemical and Biological Generation of Excited
States; Adam, W., Cilento, G., Eds.; Academic Press: New York, 1982;
pp 85-114.
(14) Ciscato, L. F. M. L.; Bastos, E. L.; Dieter, W.; Beckert, R.; Baader,
W. J. “Chemiluminescence-based uphill energy conversion”, manuscript in
preparation.
(11) (a) Schaap, A. P.; Handley, R. S.; Giri, B. P. Tetrahedron Lett. 1987,
28, 935. (b) Schaap, A. P.; Chen, T.-S.; Handley, R. S.; De Silva, R.; Giri,
B. P. Tetrahedron Lett. 1987, 28, 1155. (c) Schaap, A. P.; Sandinson, M. D.;
Handley, R. S. Tetrahedron Lett. 1987, 28, 1159.
J. Org. Chem. Vol. 75, No. 19, 2010 6575

Ciscato et al.
JOCArticle
SCHEME 2. Methylene Blue (MB) Sensitized Photooxygena-
tion of 9-Benzylidene-10-methylacridans, Leading to
Intermediate 1,2-Dioxetane Formation and Its Unimolecular
Decomposition, Accompanied by Chemiluminescence Emission
which were not isolated in this work due to their instability.
However, even without isolation or purification, kinetic data
from the thermal decomposition of the 1,2-dioxetanes 6a-m,
using the novel in situ generation and measurement method,
could be obtained with very high reproducibility, clean first-
order kinetics always being observed (Scheme 2, Figure 5 in
the Experimental Section).
The determination of the absolute chemiluminescence
quantum yields for the decomposition of 1,2-dioxetanes
6a-m could not be performed in a straightforward manner,
because of the difficult assessment of some parameters
intrinsic to the in situ method used. However, the relative
quantum yields could be estimated with a good degree of
confidence for all 1,2-dioxetanes, normalizing the numeric
values of the integrated emission intensity vs time plots
for each derivative with the integrated emission of 6a, the
strongest emitter in the series (Iint, Table 1). This approach is
valid if the yield of 1,2-dioxetane formation does not vary
significantly with the electronic character of the substituent
on the aromatic ring, which may be assumed to be true, as the
chemical yield of 1,2-dioxetane formation from adequate
olefins by photooxygenation is normally high (80-100%).15
Furthermore, it should be pointed out that addition of
the corresponding aldehyde to the reaction medium does
not lead to the reduction of the emission quantum yields;
proving that aldehydes, in moderate concentrations, do not
significantly quench the electronically excited state of the
N-methylacridone, conflicting with observations made in
previous literature works.3,25b
Using the method of isothermal kinetics, the activation
parameters for 1,2-dioxetanes 6a-m were determined by
linear fitting of the Eyring plots for the kobs values
obtained from the emission intensity decay curves
(Figure 2). Good linear correlations were obtained in these
plots in most of the cases, specially with 1,2-dioxetanes
with high chemiluminescence quantum yields, enabling
the precise determination of the activation parameters
for the chemiluminescent decomposition of the NMADs
6a-m (Table 1).
The studied NMADs 6a-m show similar thermal stabi-
lities at 25 °C, with ΔGq values that differ only about 2-3 kJ
mol-1 (Table 1). For the derivate 6a, a ΔGq value of 82.6 kJ
mol-1 was obtained by our in situ methodology; this value is
very similar to ΔGq = 82.5 kJ mol-1, determined previously
for the thermal decomposition of the isolated 1,2-dioxetane
6a,4 thus confirming the validity and reliability of this new
kinetic methodology. The most stable NMADs, 6d,e, are still
ca. 35 kJ mol-1 less stable than the adamantyl derivative 3
(ΔGq = 117.2 kJ mol-1).4 The least stable of them, 6j, isca. 5 kJ
mol-1 more stable than the highly unstable lucigenin-derived
1,2-dioxetane 4 (ΔGq = 73.4 kJ mol-1),12 placing the NMADs
obtained in this work in the lower half of the thermal stability
scale for these acridinium-substituted 1,2-dioxetanes.
Direct comparisons between the experimental kobs values
for the NMADs 6a-m cannot be performed, since they were
not all determined at the same temperature. Therefore, the
thermal decomposition rate constants at 25 °C for the 1,2-
dioxetanes 6a-m were calculated by using their experimental
free activation energies (k, Table 1). The calculated values
proved to be very similar to the experimental kobs values
measured at 25 °C for some of the derivates (see the
Supporting Information, Table S1).
and singlet oxygen also generated in situ by photooxygenation
sensitized by methylene blue, followed by the measurement of
the light emission produced by the 1,2-dioxetane cleavage. This
method allows the acquisition of kinetic data for the thermal
decomposition of 1,2-dioxetanes with half-lives of 30 s or less,
using just a simple commercial spectrofluorimeter.
Kinetic studies on the thermal decomposition of a series
of NMADs, obtained by photooxygenation of various sub-
stituted 9-benzylidene-10-methylacridans using this in situ
methodology, permitted the determination of activation
parameters for those 1,2-dioxetanes. The results obtained
provide substantial experimental evidence for the occurrence
of an electron transfer to the peroxidic ring in the initial
step of the 1,2-dioxetane decomposition. Therefore, these
data show, for the first time, experimental evidence for
the occurrence of an initial intramolecular electron transfer
in the chemiexcitation step of intramolecularly catalyzed 1,2-
dioxetane decomposition, as proposed in the intramolecular
version of the CIEEL mechanism, thought to operate in
NMADs as well as triggered phenoxy-substituted 1,2-dioxe-
tane decomposition.1,4,16,17,19
Results
The thermal decomposition of the NMADs 6a-m
(Figure 1) produced, as expected, two carbonyl compounds:
N-methylacridone (2, R3 = CH3) and the corresponding
substituted aromatic aldehyde. These fragments were identi-
fied by TLC analysis of the spent reaction mixture, after
irradiation with light and kinetic data acquisition.
The formation of the carbonyl compounds is, together
with the observation of chemiluminescence emission, good
evidence for the formation and identity of the 1,2-dioxetanes,
(15) (a) Zaklika, Z. K.; Kaskar, B.; Schaap, A. P. J. Am. Chem. Soc. 1980,
102, 386. (b) MacManus-Spencer, L. A.; Edhlund, B. L.; McNeill, K. J. Org.
Chem. 2006, 71, 796. (c) Bastos, E. L.; Ciscato, L. F. M. L.; Weiss, D.;
Beckert, R.; Baader, W. J. Synthesis 2006, 1781.
€
(16) (a) Nery, A. L. P.; Ropke, S.; Catalani, L. H.; Baader, W. J.
Tetrahedron Lett. 1999, 40, 2443. (b) Nery, A. L. P.; Weiss, D.; Catalani,
L. H.; Baader, W. J. Tetrahedron 2000, 56, 5317.
(17) (a) Matsumoto, M. Photochem. Photobiol. C: Photochem. Rev. 2004,
5, 27. (b) Tanimura, M.; Watanabe, N.; Ijuin, H. K.; Matsumoto, M. J. Org.
Chem. 2010, 75, 3678.
(18) (a) Trofimov, A. V.; Mielke, K.; Vasil’ev, R. F.; Adam, W. Photo-
chem. Photobiol. 1996, 63, 463. (b) Adam, W.; Bronstein, I.; Edwards, B.;
Engel, T.; Reinhardt, D.; Schneider, F. W.; Trofimov, A. V.; Vasil’ev, R. F.
J. Am. Chem. Soc. 1996, 118, 10400. (c) Adam, W.; Bronstein, I.; Trofimov,
A. V. J. Phys. Chem. A 1998, 102, 5406. (d) Nery, A. L. P.; Catalani, L. H.;
€
Ropke, S.; Nunes, G. I. P.; Baader, W. J. In Bioluminescence and Chemilu-
minescence: Perspectives for the 21st Century; Roda, A., Pazzagali, M.,
Kricka, L. J., Stanley, P. E., Eds.; Wiley: Chichester, U.K., 1999; pp 45-48.
(19) (a) Wilson, T. In Singlet Oxygen; Frimer, A. A., Ed.; CRC Press:
Boca Raton, FL, 1985; pp 37-57. (b) Catalani, L. H.; Wilson, T. J. Am.
Chem. Soc. 1989, 111, 2633.
6576 J. Org. Chem. Vol. 75, No. 19, 2010

Ciscato et al.
JOCArticle
TABLE 1. Activation Parameters, Thermal Decomposition Rate Constants (k), and Integrated Normalized Chemiluminescence Quantum Yields (Iint) for
the NMADs 6a-m
1,2-dioxetane
R
σ
ΔHq (kJ mol-1
)
ΔSq (J K-1 mol-1
)
ΔGq (kJ mol-1 a
)
102k (s-1 b
)
Iint (au)c
6b
6c
6d
6e
6a
6f
6g
6h
6i
p-N(CH3)2
p-OH
p-OCH3
p-CH3
H
m-OH
m-OCH3
p-Br
-0.83
-0.37
-0.27
-0.17
0.00
þ0.12
þ0.12
þ0.23
þ0.37
þ0.56
þ0.71
þ0.66
þ0.78
61 ( 3
-63 ( 4
-101 ( 2
79 ( 5
81 ( 3
7.5 ( 0.5
3.6 ( 0.1
0.03
0.17
0.48
0.69
1.00
0.06
0.66
0.31
0.17
0.05
0.01
0.04
0.02
51.1 ( 0.8
74.1 ( 0.8
77.5 ( 0.4
76.2 ( 0.4
70.4 ( 0.8
76.6 ( 0.8
72.4 ( 0.4
74.5 ( 0.8
57.8 ( 0.4
71 ( 4
-29.3 ( 0.4
-18.5 ( 0.1
-21.5 ( 0.1
-32.3 ( 0.5
-17.8 ( 0.2
-28.9 ( 0.2
-19.6 ( 0.2
-68.2 ( 0.2
-25 ( 3
83 ( 1
1.90 ( 0.02
1.80 ( 0.01
2.09 ( 0.01
5.85 ( 0.07
2.76 ( 0.03
3.91 ( 0.02
5.14 ( 0.03
12.9 ( 0.1
10.8 ( 0.7
8.4 ( 0.1
83.0 ( 0.4
82.6 ( 0.4
80 ( 1
82 ( 1
81.1 ( 0.4
80.4 ( 0.5
78.1 ( 0.5
79.0 ( 5.0
79 ( 1
m-Cl
6j
6k
6l
m-CN
m-NO2
p-CN
62.8 ( 0.4
80.0 ( 0.4
-54.8 ( 0.8
-3.5 ( 0.2
6m
p-NO2
81.0 ( 0.5
4.00 ( 0.02
aAt 298 K. bCalculated from the ΔGq values, at 298 K. cEstimated from the experimental emission intensity decay curves obtained at 298 K, using
numerical integration and normalized with respect to the integrated emission of 6a.
(Scheme 3, path B).1 In contrast to this observation, studies
on the solvent cavity effect on the quantum yields for these
transformations pointed to the occurrence of an intermole-
cular BET step (Scheme 3, path A).20 Apart from this still
unresolved question, it has to be mentioned that there is still
not any direct experimental evidence on the reaction steps
postulated, although these appear to be reasonable on the
basis of the intermolecular CIEEL mechanism.1,11,16,17
Using the aforementioned mechanistic outlines, the
chemiluminescent decomposition of the NMADs 6a-m
can also be rationalized. The first, rate-limiting step is an
endothermic electron transfer (ET) from the lone pair of the
nitrogen atom of the acridinium moiety to the peroxidic ring
(Scheme 4). This intramolecular ET (ETintra) is supposed to
occur simultaneously to the cleavage of the O-O bond,
thereby avoiding the highly favorable exothermic back
electron transfer, which would reset the process to the initial
FIGURE 2. Eyring plots for the determination of activation para-
state. Two biradical species can be formed in this process
meters in the unimolecular decomposition of the NMADs 6a (H),
(7a,b), differing only in the position of the negative charge
6c (p-OH), 6i (m-Cl) and 6l (p-CN).
and the unpaired electrons. It should be noted that structures
7a,b are distinct species; they do not represent resonance
structures, since there is no conjugation between the two oxy-
Discussion
gen atoms (Scheme 4). However, conjugation does occurs in the
transition state of the C-C bond cleavage (Scheme 5) of the next
reaction step. Consequently, independently of the intermediate
formed initially, 7a or 7b, C-C bond cleavage can result in the
formation of both of the following species, an acridinium radical
cation/aldehyde radical anion pair or an acridinium biradical
zwitterion; the latter is accompanied by the formation of a
neutral benzaldehyde derivative (Scheme 4).
If the acridinium biradical zwitterion is generated, an
intramolecular back electron transfer (BETintra) may directly
form the N-methylacridone in the singlet excited state,
probably in high quantum yields. Alternatively, C-C bond
cleavage and BET can occur as a concerted process, as the
acridinium biradical zwitterion might be considered a cano-
nical structure of the N-methylacridone in the singlet excited
state (Scheme 4).1 The other possibility is an intermolecular
back electron transfer (BETinter), from the aldehyde radical
anion to the acridinium radical cation, a pathway that prob-
ably leads to low yields for excited state formation, because
Mechanistic Aspects of NMAD Chemiluminescence. The
decomposition mechanism of the NMADs 6a-m can be for-
mulated in analogy to the intramolecular CIEEL mecha-
nism proposed for the decomposition of “silyl-triggered”
3-phenoxyl-substituted 1,2-dioxetanes.16-18 The reaction is
initiated by deprotection of the silyl group and formation of
a phenolate anion, followed by an intramolecular electron
transfer (ET) from the phenolate to the cyclic peroxide,
leading to the formation of a biradical anion (Scheme 3).
This biradical anion undergoes C-C bond cleavage, forming
either a radical pair (Scheme 3, path A) or another biradical
anion (Scheme 3, path B). Consequently, annihilation of
those radical or biradical species may occur by inter- or intra-
molecular back electron transfer (BET) processes, which can
lead to the formation of the corresponding phenolate anion
in its singlet excited state, responsible for the observed light
emission.16-18
The high singlet quantum yields (ΦS ≈ 1.0 einstein mol-1 16
)
obtained in the induced decomposition of “silyl-triggered”
3-phenoxyl-substituted 1,2-dioxetanes, in contrast to the
much lower yields observed in intermolecular CIEEL systems
(ΦS ≈ 10-4 einstein mol-1),19 indicate that the whole CIEEL
sequence should occur in a totally intramolecular fashion
(20) (a) Adam, W.; Bronstein, I.; Trofimov, A. V.; Vasil’ev, R. F. J. Am.
Chem. Soc. 1999, 121, 958. (b) Adam, W.; Trofimov, A. V. J. Org. Chem.
2000, 65, 6474. (c) Adam, W.; Matsumoto, M.; Trofimov, A. V. J. Am. Chem.
Soc. 2000, 122, 8631.
J. Org. Chem. Vol. 75, No. 19, 2010 6577

Ciscato et al.
JOCArticle
SCHEME 3. Intramolecular CIEEL Mechanism in the Induced Decomposition of “Silyl-Triggered” 3-Phenoxyl-Substituted
1,2-Dioxetanesa
aTBDMS = tert-butyldimethylsilyl.
SCHEME 4. Possible Chemiexcitation Pathways for the Decomposition of the NMADs
Activation Parameters for NMADs Decomposition. The
SCHEME 5. Structure of the Transition State for the C-C
activation parameters determined for the decomposition of
1,2-dioxetanes 6 show that activation enthalpy and entropy
values do not vary in a systematic way with the electronic
properties of the substituents (Table 1). Activation entropies
are negative in all cases, which might be due to the necessity
of a specific conformation of the acridinium moiety for
the occurrence on the intramolecular electron transfer.4,25b
However, the free activation energies, although showing
little variation, appear to vary systematically for electron-
withdrawing substituents, with generally lower values for
more strongly electron attracting substituents (Table 1).
However, also the good electron donor substituted deriva-
tives 6b,c show ΔGq values lower than that for the unsub-
stituted derivative 6a. The rate constant values calculated
from the activation parameters at 25 °C show variations of
almost 1 order of magnitude, being lowest for 6a (X = H)
and showing the highest values for good electron acceptors
as well as donating substituted derivatives (Table 1).
Bond Cleavage Step
of its intermolecular nature.1 The low chemiexcitation quan-
tum yields of CIEEL systems with an initial intermolecular
ET step (ΦS ≈ 10-4 einstein mol-1 19
)
yields observed for systems where an initial intramolecular
and the much higher
ET process is supposed to occur (ΦS ≈ 1.0 einstein mol-1 1,16
)
suggest that the decomposition of NMADs, showing high
emission quantum yields, should occur by an entirely intra-
molecular process, including the back electron transfer step
(Scheme 4, BETintra).
6578 J. Org. Chem. Vol. 75, No. 19, 2010

Ciscato et al.
JOCArticle
FIGURE 3. Linear free-energy correlation of the rate constants
(kX/kH) of the emission intensity decay in the decomposition of the
NMADs 6a-m with the Hammett substituent constants (σ). Values
for the substituents p-N(CH3)2, p-OH, p-OCH3, m-OH, m-CN, and
p-NO2 were not included in the correlation.
FIGURE 4. Correlation of the normalized quantum yields of light
emission (Iint) for the NMADs 6a-m and the Hammett substituent
constants (σ).
However, if it is assumed that both sets of structures (the
radical ion pair and the acridinium biradical zwitterion) can
be formed after C-C bond cleavage in a “competitive”
manner, it is possible to explain the decrease of the chemi-
luminescence quantum yields by the stabilization effect
of the aldehyde radical anion for more strongly electron
withdrawing substituents. This stabilization effect would
decrease the portion of the reaction occurring through the
completely intramolecular, high excitation quantum yield,
pathway (via an acridinium biradical zwitterion). Therefore,
the observation that the chemiluminescence quantum yields
decrease with more strongly electron withdrawing substitu-
ents can be rationalized by the competition of the inter- and
intramolecular BET steps, supposing that only the BETintra
leads to efficient excited state formation (Scheme 4).
Chemiexcitation Pathways in NMADs Chemilumines-
cence. The experimental facts can also be rationalized on the
basis of the exclusive occurrence of the intermolecular BET.
In this case, the efficiency of excited state formation will
be determined by the energy liberated in the BET step from
the aldehyde radical anion to the acridinium radical cation,
as discussed elsewhere for the peroxyoxalate reaction.21 As
electron-withdrawing substituents stabilize the benzaldehyde
derivative, less energy will be available for chemiexcitation in
the decomposition of NMADs bearing these substituents,
explaining the experimentally observed lower quantum yields
(Figure 4). In contrast, if the only BET process operating
would be the intramolecular one, the observed quantum yield
behavior can be also rationalized on the basis of energy
considerations. The principal intermediate 7b (and also 7a,
although to a lesser extent) will have a lower energy content
for the derivatives bearing electron-withdrawing substituents.
Therefore, less energy will be liberated in the combined C-C
bond cleavage and intramolecular BET processes, the lower
quantum yields being due to energy reasons.21 In this argu-
ment, C-C bond cleavage and BET are formulated as con-
certed processes, as the acridinium biradical zwitterion is
considered a canonical structure of the N-methylacridone in
the singlet excited state.1
Linear Free-Energy Correlations for NMADs Chemilumi-
nescence. The linear free-energy correlation of the rate con-
stants (kX/kH) for the thermal decomposition of the NMADs
with the Hammett substituent constants (σ) shows a good
linear dependence for moderately electron withdrawing sub-
stituents, withF = 1.3(0.1 (Figure 3). The positivesignofthe
parameter F indicates the formation of a partial negative
charge in the transition state of the rate-determining step:
namely, the intramolecular electron transfer from the acridine
nitrogen to the dioxetane ring (ETintra, Scheme 4). Addition-
ally, a F value close to 1.0 indicates a transition state structure
similar to that of the benzoic acid anion, implying that the
most probable structure of the intermediate, formed after
initial electron transfer and O-O bond cleavage, is of type 7b
(Scheme 4). This can be implied since the distance of the
negative charge to the stabilizing substituent, in this case, is
similar to that in the corresponding conjugated base of
benzoic acid. The result obtained in the linear free-energy
correlation with electron-withdrawing substituted derivatives
constitutes the first experimental evidence for the occurrence
of an intramolecular electron transfer in the catalyzed decom-
position of properly substituted 1,2-dioxetanes, such as the
NMADs studied in the present work (Figure 3).
Although the results discussed above indicate the prefer-
ential formation of 7b in the initial electron transfer step, any
of these radical species can lead to both of the possible C-C
bond cleavage products; therefore, these results do not give
insight into this part of the chemiexcitation mechanism.
The correlation of the normalized chemiluminescence quan-
tum yields (Iint) with the Hammett substituent constants (σ)
shows that, for NMADs bearing electron-withdrawing sub-
stituents, the quantum yield decreases with an increase in the
electron-accepting ability of the substituent (Figure 4). As-
suming efficient excited state formation in the intramolecu-
lar process (BETintra), justified by the fact that the acridinium
biradical zwitterion can be considered as a charge-separated
structure of the electronically excited N-methylacridone, the
variation of the electron acceptor quality of the substituent is
not expected to exert any influence on the efficiency of
excited state formation, as there are always the same species
participating in the back electron transfer step (Scheme 4).
(21) (a) Stevani, C. V.; Silva, S. M.; Baader, W. J. Eur. J. Org. Chem. 2000,
4037. (b) Silva, S. M.; Wagner, K.; Weiss, D.; Beckert, R.; Stevani, C. V.;
Baader, W. J. Luminescence 2002, 17, 362.
J. Org. Chem. Vol. 75, No. 19, 2010 6579

Ciscato et al.
JOCArticle
Conclusions
This work provides, for the first time, unequivocal experi-
mental evidence for the occurrence of an intramolecular electron
transfer from a electron-donor group (N-methylacridinium)
to the peroxidic ring, in the intramolecularly catalyzed decom-
position of 1,2-dioxetanes. This experimental evidence can be
extended to the analogous case of phenoxyl-substituted 1,2-
dioxetanes, widely utilized in bioanalytical applications, making
it of general importance for an understanding of the mechanisms
of efficient electronic excited state formation. Furthermore, the
influence of the electronic properties of substituents on the
emission quantum yields of the derivatives could be rationalized
on the basis of the intramolecular CIEEL mechanism, validating
its application to this case.
FIGURE 5. Example of an experimental emission intensity decay
curve obtained from in situ generation of a NMAD by methylene
blue sensitized photooxygenation of a substituted 9-benzylidene-
10-methylacridan.
Experimental Section
In conclusion from this part, it can be stated that the
experimentally observed tendency of the singlet excitation
quantum yields can be explained on the basis of the intra-
molecular CIEEL mechanism, being involved an intra- or
intermolecular BET. In turn, this fact also contributes to the
validation of the proposed CIEEL sequence for the decom-
position of the NMADs.
Acetonitrile (VWR, 99.9%) was distilled twice from CaH2
under an inert atmosphere before use. The preparation of
9-diethylphosphono-N-methylacridan and the 9-benzylidene-
10-methylacridans substituted with various aryl groups, synthe-
sized by the Horner-Wadsworth-Emmons (HWE) method, is
described elsewhere.3,25 The detailed procedures and analytical
data for all compounds are given in the Supporting Information.
The kinetic measurements were performed in a cuvette con-
taining 2 mL of a methylene blue (14.4 μmol L-1, Merck,
>95%) solution and the olefin (97.5 μmol L-1) in acetonitrile.
This solvent has two important characteristics related to
the in situ singlet oxygen generation: high oxygen solubility
(1.9 mmol L-1 at 293 K and 0.21 atm of O2)26a and long singlet
oxygen lifetime (30 μs).26b Air-equilibrated solutions ([O2] ≈
1.9 mmol L-1) were utilized in all the experiments performed.
The cuvette was placed in the cell holder of the spectrofluorimeter
with magnetic stirring and allowed to reach the desired tempera-
ture. The solution was then irradiated with light of wavelength
655 nmfor 100s, after which the excitation shutter was closed and
the emission shutter opened to initiate the measurement of the
integrated (whole spectral range) light emission (Figure 5). The
emission intensity was always shown to be maximum at acquisi-
tion time 0 and showed a first-order decay, which was followed
From the Hammett plot, it can be observed that two
NMADs clearly do not fit into the linear free-energy correla-
tion: the electron-donor-substituted derivatives 6b (p-N-
(CH3)2) and 6c (p-OH) (Figure 3). This behavior can be
rationalized by the fact that in these cases the aromatic group
itself constitutes a very good electron donor and, therefore,
this group can transfer an electron to the dioxetane ring.22
This parallel reaction is expected to lead to an increase in the
observed decomposition rate constant for 6b,c. This effect
may also contribute to the slightly higher rate constants
obtained for the methoxy- and methyl-substituted deriva-
tives 6d,e (Figure 4). Nevertheless, the rate constant value for
the m-OH derivate 6f, considerably higher than would be
expected from its σ value (þ0.12), cannot be explained by the
occurrence of this parallel reaction. However, it is known
that also m-phenoxy-substituted 1,2-dioxetanes can undergo
decomposition from an intramolecular electron transfer
from the phenolate to the dioxetanic ring;11b,20b a corre-
sponding effect may be taking place in the present case.
This direct electron transfer from electron-donor substit-
uents to the dioxetane ring suggested above can also explain
the decrease in the chemiluminescence quantum yields for
NMADs containing such groups (Figure 4). This process will
afford radical species of the substituted benzaldehyde
which,23 by an intra- or intermolecular back electron trans-
fer, will lead to the production of the singlet excited state of
the aldehyde derivative. Since the fluorescence quantum
yields of substituted benzaldehyde derivatives are extremely
low (ca. 10-5),24 the expected chemiluminescence quantum
yields would also be low for those NMADs bearing electron-
donor substituents, which is in agreement with the experi-
mental observations.
during at least 4 half-lives. The observed rate constants (kobs
)
were obtained by fitting the intensity versus time curves using a
monoexponential decay function. In all cases, rate constants were
shown to be independent of the irradiation time, as expected for
the case of unimolecular 1,2-dioxetane decomposition.
~
ꢀ
Acknowledgment. We thank the Fundac-ao de Amparo a
Pesquisa do Estado de Sao Paulo (FAPESP) and Arbeitsge-
~
meinschaft industrieller Forschungsvereinigungen “Otto von
Guericke” e.V. (AiF) for financial support. L.F.M.L.C. thanks
the Deutscher Akademischer Austauschdienst (DAAD) and
Coordenadoria de Aperfeic-oamento de Pessoal de Ensino Super-
ior (CAPES) for a doctoral fellowship (BEX 1132/04-0). F.H.B.
thanks the FAPESP for a doctoral fellowship (2005/58320-4).
Supporting Information Available: Text and a table giving the
detailed synthesis of the substituted 9-benzylidene-10-methylacri-
dan and observed rate constants at different temperatures for the
N-methylacridan-substituted 1,2-dioxetanes 6a-m. This material
is available free of charge via the Internet at http://pubs.acs.org.
(22) Zaklika, K. A.; Thayer, A. L.; Schaap, A. P J. Am. Chem. Soc. 1978,
100, 4916.
(23) Oliveira, M. A.; Baader, W. J. In Chemistry, Biology and Applica-
tions: Proceedings of the 14th International Symposium on Bioluminescence
and Chemiluminescence; Szalay, A. A., Hill, P. J., Kricka, L. J., Stanley, P. E.,
Eds.; World Scientific: Singapore, 2006; pp 231-234.
(25) (a) Redmore, D. J. Org. Chem. 1969, 34, 1420. (b) Perkizas, G.;
Nikokavouras, J. Monatsh. Chem. 1983, 114, 3.
(26) (a) Murov, S. L.; Carmichael, I.; Hug, G. L. In Handbook of
Photochemistry, 2nd ed.,; Marcel Dekker: New York, 1993; pp 289-293.
(b) Merkel, P. B.; Kearns, D. R. J. Am. Chem. Soc. 1972, 94, 7244.
(24) Crowell, E. P.; Varsel, C. J. Anal. Chem. 1963, 35, 189.
6580 J. Org. Chem. Vol. 75, No. 19, 2010