Solvent-promoted chemiluminescent decomposition of a bicyclic dioxetane bearing a 4-(benzothiazol-2-yl)-3-hydroxyphenyl moiety

Article abstract of DOI:10.1016/j.tetlet.2008.04.110

An aprotic polar solvent such as N-methylpyrrolidone (NMP) promoted the thermal decomposition of bicyclic dioxetane bearing a 4-(benzothiazol-2-yl)-3-hydroxyphenyl moiety 1 without the addition of any base. This solvent-promoted decomposition (SPD) gave l

Full text of DOI:10.1016/j.tetlet.2008.04.110

Tetrahedron Letters 49 (2008) 4170–4173
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: http://www.elsevier.com/locate/tetlet
Solvent-promoted chemiluminescent decomposition of a bicyclic dioxetane
bearing a 4-(benzothiazol-2-yl)-3-hydroxyphenyl moiety
*
Masakatsu Matsumoto , Masatoshi Tanimura, Taichi Akimoto, Nobuko Watanabe, Hisako K. Ijuin
Department of Chemistry, Kanagawa University, Tsuchiya, Hiratsuka, Kanagawa 259-1293, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
An aprotic polar solvent such as N-methylpyrrolidone (NMP) promoted the thermal decomposition of
bicyclic dioxetane bearing a 4-(benzothiazol-2-yl)-3-hydroxyphenyl moiety 1 without the addition of
any base. This solvent-promoted decomposition (SPD) gave light as effectively as the base-induced
decomposition (BID) in an aprotic polar solvent. SPD caused intramolecular CT-induced chemilumines-
cence similar to BID, but, in contrast to BID, SPD proceeded through a pathway with a large negative
entropy of activation. Dioxetane 1 was also shown to give light due to excited state intramolecular proton
transfer (ESIPT) upon heating in p-xylene.
Received 13 March 2008
Revised 17 April 2008
Accepted 18 April 2008
Available online 23 April 2008
Keywords:
Dioxetane
Chemiluminescence
Solvent-promoted decomposition
ESIPT
Ó 2008 Elsevier Ltd. All rights reserved.
Dioxetanes bearing an electron donor such as oxidophenyl
anion undergo intramolecular charge-transfer (CT)-induced decom-
position to effectively give a singlet-excited carbonyl fragment that
emits light.1–4 On the other hand, uncatalyzed thermal decomposi-
tion (TD) of dioxetanes gives mainly a triplet-excited species, so
that bright light emission is hardly expected.5,6 In addition to these
two types of decomposition, we report here the aprotic polar sol-
vent-promoted decomposition (SPD) of dioxetane bearing a 4-
(benzothiazol-2-yl)-3-hydroxyphenyl moiety 1 accompanied by
the emission of bright light, and that this SPD resembled the
base-induced decomposition (BID) of 1,7,8 which proceeds through
unstable anionic dioxetane 2 to keto ester 3; however, the entropy
of activation for SPD had a large negative value, which was in sharp
contrast to BID (Scheme 1). Furthermore, we report that the uncat-
alyzed TD of 1 gave light due to excited state intramolecular proton
transfer (ESIPT) in a nonpolar solvent.9,10
green light when dissolved in DMSO at room temperature. Such
chemiluminescence was apparently caused by solvent-promoted
decomposition (SPD), although this was hardly observed for a
parent dioxetane 4 bearing a simple 3-hydroxyphenyl group. Thus,
we investigated the behavior of 1 in hot DMSO. Thermal decompo-
sition of 1 in DMSO at 100 °C exhibited the emission of green light
with maximum wavelength kmax = 497 nm and half-life t1/2 = 320 s,
although the chemiluminescence efficiency, U, was only 0.0013.
From the mixture that followed the thermal decomposition of 1,
epoxide 511 was isolated in high yield along with a trace amount
of keto ester 6 (Scheme 2). The formation of dimethyl sulfone
was also observed by 1H NMR analysis of the reaction mixture.
These results show that 1 was consumed predominantly by reduc-
tion with DMSO but not by SPD to give 6. Therefore, we next
sought to examine the SPD of 1 in an aprotic polar solvent, which
is far more resistant to oxidation than DMSO.
In the course of our investigation of high-performance chemilu-
minescence compounds, we found that dioxetane 1 emitted weak
The solvent selected was N-methylpyrrolidone (NMP), which is
used as frequently as DMSO. First, the BID of 1 was examined.
:B
OH
O
O
N
N
N
O O
O O
O
O
t-Bu
t-Bu
t-Bu
S
S
S
O
O
O
Light
2
3
1
Scheme 1. Base-induced decomposition (BID).
* Corresponding author. Tel.: +81 463 59 4111; fax: +81 463 58 9684.
E-mail address: matsumo-chem@kanagawa-u.ac.jp (M. Matsumoto).
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.04.110

M. Matsumoto et al. / Tetrahedron Letters 49 (2008) 4170–4173
4171
OH
OH
N
OH
N
O O
O
O
O
O
t-Bu
t
-Bu
t
-Bu
S
S
O
O
6
5
4
Scheme 2.
When 1 was treated with tetrabutylammonium fluoride (TBAF) in
BID
(a)
NMP at 45 °C, chemiluminescence occurred effectively with k
¼
max
(g)
498 nm,
U
BID = 0.29, and rate constant kBID ¼ ln 2=tBID
¼
1=2
2:5 ꢀ 10ꢁ4 sꢁ1 (vide infra). The BID of 1 also proceeded to give
bright light in acetonitrile,8 DMF, and DMSO. These results are
summarized in Table 1, and the chemiluminescence spectra of 1
are illustrated in Figure 1.
(f)
(b)
(c)
(d)
(e)
Next, SPD was investigated in NMP. When 1 was heated with-
(g)
out any strong base in NMP at 100 °C, 1 decomposed to exclusively
give
6 accompanied by the emission of green light with
SPD
k
¼ 498 nm,
U
SPD = 0.25, and
k
SPD = 8.0 ꢀ 10ꢁ3 s. Thus, the
max
chemiluminescence spectrum coincided with that for the BID of
1, as illustrated in Figure 2, and USPD was comparable to that for
BID as shown in Table 1. These results suggested that the SPD of
1 in NMP should proceed through dioxetane bearing an oxidophen-
yl anion 2 to give anionic keto ester 3 in the excited state or a clo-
sely related species such as a solvated anion by a mechanism
350
450
wavelength / nm
550
650
Figure 2. Chemiluminescence spectra for the SPD of 1 in (a) NMP, (b) DMF, (c)
DMPU, (d) PPC, and (e) acetonitrile, and for the TD in (f) p-xylene and (g) DGM.
similar to the case of BID. This effective SPD chemiluminescence
of 1 was also observed in aprotic polar solvents, such as DMF,
N,N0-dimethylpropyleneurea (DMPU), and propylene carbonate
(PPC) at 100 °C. The results are summarized in Table 1, and the
chemiluminescence spectra are illustrated in Figure 2.
p-Xylene is a typical nonpolar solvent that is used often to
examine the uncatalyzed thermal decomposition of dioxetanes.13
Thus, we carried out the thermal decomposition (TD) of 1 in p-xy-
lene to compare the behavior with that in an aprotic polar solvent.
Table 1
Base-induced decomposition (BID) and solvent-promoted decomposition (SPD) of
dioxetane 1
Solventa
Temp (°C)
k
ðnmÞ
UCL b
t1/2 (s)
k (sꢁ1
)
CL
Mode
max
BID
BID
BID
BID
NMP
CH3CN
DMF
45
45
45
45
498
492
496
495
0.29
0.28
0.29
0.35
2800
1600
2200
1600
2.5 ꢀ 10ꢁ4
4.2 ꢀ 10ꢁ4
3.1 ꢀ 10ꢁ4
4.3 ꢀ 10ꢁ4
DMSO
SPD
SPD
SPD
SPD
SPD
NMP
DMF
DMPU
PPC
100
100
100
100
60
498
496
498
493
494
0.25
0.23
0.25
0.24
87
57
8.0 ꢀ 10ꢁ3
1.2 ꢀ 10ꢁ2
2.6 ꢀ 10ꢁ3
2.4 ꢀ 10ꢁ3
In p-xylene at 100 °C, dioxetane 1 decomposed to afford 6 with the
TD
270
290
emission of yellow light ðk
¼ 536 nmÞ, though the efficiency,
max
U
TD, was only 8 ꢀ 10ꢁ4 and the rate was very slow (rate constant
c
c
c
CH3CN
–
–
–
k
TD = 1.2 ꢀ 10ꢁ5 sꢁ1) (Table 1, Fig. 2). The small value for kTD indi-
TD
TD
p-XL
DGM
100
100
536
403,535
0.0008
60000
1.2 ꢀ 10ꢁ5
cates that 1 underwent typical uncatalyzed TD in p-xylene, as
reported for various analogs of dioxetane 1,14 in contrast to SPD.
However, the chemiluminescence for the TD of 1 was observed
at a longer wavelength than those for BID and SPD. This result
c
c
c
–
–
–
a
NMP: N-methylpyrrolidone, DMPU: N,N0-dimethylaminopropyleneurea, PPC:
propylene carbonate, p-XL: p-xylene, DGM: diethylene glycol dimethyl ether.
b
Based on a value reported for the BID of 3-adamantylidene-4-(3-tert-butyl-
was rather unexpected, since the chemiluminescence of TD due
dimethylsiloxyphenyl)-4-methoxy-1,2-dioxetane in TBAF/DMSO.12
TD
max
c
to a neutral hydroxybenzoate emitter generally shows a k
short-
Not estimated.
er than that for the corresponding BID. In fact, the chemilumines-
cence spectrum for the TD of 4 appeared at a wavelength shorter
TD
BID
than that for the BID: k
¼ 410 nm, while k
¼ 466—470 nm
max
max
for 4.8 Based on reports that 2-(benzothiazol-2-yl)phenols display
fluorescence due to ESIPT,9,10 the emission of yellow light for the
TD of 1 in p-xylene is presumably attributed to excited quinone
methide 7 (Scheme 3). This idea was supported by the fact that
(a)
methoxyphenyl-analog 8 of
1 underwent TD in p-xylene at
TD
100 °C to give weak violet light with
k
¼ 390 nm, UTD
=
(b)
max
9.2 ꢀ 10ꢁ4, and kTD = 4.3 ꢀ 10ꢁ4 sꢁ1 (Scheme 3). This result sug-
(c)
(d)
gested that the TD of 1 would show chemiluminescence with
TD
max
k
near 390 nm, if ESIPT did not occur. This expectation was
realized when the TD of 1 was carried out in diethylene glycol
dimethyl ether (diglyme) at 100 °C. As illustrated in Figure 2,
the chemiluminescence spectrum showed a peak at 403 nm in
addition to a peak at 535 nm due to ESIPT.
The effects of solvent on the thermal chemiluminescent decom-
position of 1 described above are summarized as follows. First, SPD
proceeded in an aprotic polar solvent to afford bright light with
350
450
wavelength / nm
550
650
Figure 1. Chemiluminescence spectra for the BID of 1 in (a) NMP, (b) DMF, (c)
acetonitrile, and (d) DMSO.

4172
M. Matsumoto et al. / Tetrahedron Letters 49 (2008) 4170–4173
H
H
O
O
*
*
N
N
O O
O
O
O
Δ
Δ
t
-Bu
t
-Bu
Light
Light
S
S
O
O
7
1
OMe
OMe
N
N
O O
O
t
-Bu
t
-Bu
S
S
O
O
8
9
Scheme 3. Thermal decomposition (TD) of 1 emitting light due to ESIPT and TD of 8.
SPD
SPD
k
¼ 492—498 nm similar to BID by a CT-induced decomposition
at 60–100 °C, as shown in Figure 3. On the other hand, both k
max
max
mechanism. Second, 1 underwent uncatalyzed TD to give excited 7
which emitted yellow light with k
nonpolar p-xylene. In addition to the chemiluminescence due to
ESIPT, light emission from neutral phenol 6 in the excited state
and USPD changed scarcely with the SPD reaction temperature.
These results suggested that the thermodynamic profile of the
reaction could be simply but meaningfully analyzed in terms of
pseudo-first order kinetics. Based on the thus-measured rate con-
stants, kSPDs, at 60–100 °C, the thermodynamic parameters for
the SPD of 1 in NMP were estimated from the Arrhenius plots to
TD
max
¼ 536 nm due to ESIPT in
TD
max
was observed at k
¼ 403 nm in diglyme.
It was important to elucidate whether SPD is mechanistically
different from BID. Thus, the kinetics of the SPD of 1 was investi-
gated in NMP as a representative solvent. SPD should be a consec-
utive reaction, which consists of the reversible formation of
dioxetane bearing an oxidophenyl anion 2 or its solvated species
that undergo CT-induced decomposition to give light. However,
the reaction proceeded following practically first-order kinetics
be free energy of activation DGà = 99.3 kJ molꢁ1
, enthalpy of
activation DHà = 82.4 kJ molꢁ1, and entropy of activation DSà =
ꢁ56.8 J Kꢁ1 molꢁ1
.
BID using TBAF in an aprotic polar solvent has been known to
proceed following pseudo-first order kinetics independent of the
concentration of TBAF when a large excess of TBAF is used. Thus,
rate constants kBID for 1 were measured using a large excess of
TBAF in NMP at five points from 35 to 55 °C. The Arrhenius plots
revealed that the activation parameters for BID were
DGà = 97.3 kJ molꢁ1, DHà = 99.8 kJ molꢁ1, and DSà = 8.2 J Kꢁ1 molꢁ1
.
If we compare these activation parameters with those for SPD,
we see that the entropy term of SPD has a large negative value in
contrast to that of the BID. These results reveal that SPD proceeded
through a transition state with considerably less disorder than that
for BID. This is presumably attributed to the participation of hydro-
gen-bonding (HB) between NMP molecule(s) and a phenolic OH of
1. Thus, an intermediary anionic dioxetane 10 was produced
through HB as an ion pair with protonated NMP molecule(s) to
cause SPD. A model scheme is illustrated in Scheme 4. On the other
hand, the intermediary anion 2 should be slightly solvated or
naked as an extreme for BID in an aprotic polar solvent such as
NMP and DMSO especially with the use of TBAF.
According to the above suggestion, SPD is presumed to proceed
effectively only in aprotic solvent with high ability of electron-pair
donor (EPD) (proton acceptor) and low ability of electron-pair
acceptor (EPA) (proton donor). This idea was consistent with the
SPD in acetonitrile, as described below. Although acetonitrile (bp
81.6 °C) is an aprotic polar solvent that is suited for BID as shown
in Table 1, it possesses lower ability of EPD and higher ability of
EPA than NMP.15–18 Upon heating in acetonitrile at 60 °C, the SPD
(a)
(b)
(c)
(a)
(d)
(e)
200 400 600 800 1000
Time / s
(b)
(d)
(e)
(c)
200
400
600
Time / s
800
100 0
of 1 also took place though very slowly and emission of light with
Figure 3. Time-course for SPD of 1 in NMP at (a) 100, (b) 90, (c) 80, (d) 70, and (e)
60 °C.
SPD
k
at 494 nm was so weak that the USPD could not be estimated.
max
Solvent
H
Solvent
Solvent
H
H
O
N
O
O
N
N
O O
O O
O
O
t
-Bu
t-Bu
t
-Bu
S
S
S
O
O
O
Light
Solvated 1
10
Scheme 4. Solvent-promoted decomposition (SPD).

M. Matsumoto et al. / Tetrahedron Letters 49 (2008) 4170–4173
4173
In conclusion, a variety of aprotic polar solvents were shown to
promote the decomposition (SPD) of bicyclic dioxetane bearing a
4-(benzothiazol-2-yl)-3-hydroxyphenyl moiety 1 to emit light as
effectively as the BID of 1 in an aprotic polar solvent. SPD caused
intramolecular CT-induced chemiluminescence, similar to BID,
though the SPD reaction proceeded through a pathway with a large
negative entropy of activation in contrast to BID. Furthermore,
uncatalyzed decomposition of dioxetane 1 was found to occur to
give light due to ESIPT in p-xylene.
(a)
(b)
(c)
(e)
(d)
Acknowledgments
350
450
550
650
The authors gratefully acknowledge financial assistance pro-
vided by Grants-in aid (Nos. 1550043 and 17550050) for Scientific
Research from the Ministry of Education, Culture, Sports, Science,
and Technology, Japan.
wavelength / nm
Figure 4. Fluorescence spectra of authentic 6 in (a) NMP, (b) acetonitrile, (c) p-
xylene, and 3 generated from 6 in situ in (d) NMP and (e) acetonitrile.
References and notes
A critical difference was seen in the fluorescence of authentic
emitter keto ester 6 when dissolved in NMP, acetonitrile, and p-xy-
lene. The fluorescence spectra of oxido anion 3 prepared from 6 in
situ coincided with the corresponding chemiluminescence spectra
for the BID of 1 in both NMP and acetonitrile (Fig. 4). On the other
hand, the fluorescence of 6 showed a peak at kmax = 497 nm
together with a peak at kmax = 403 nm due to neutral 6 in NMP
without any additive base, while the fluorescence of 6 in acetoni-
trile (kmax = 527 nm) resembled that in p-xylene (kmax = 530 nm)
rather than that in NMP (Fig. 4). These results show that consider-
able amount of oxido anion 3 existed together with neutral 6 in hot
NMP, though little 3 exists in acetonitrile. These and the SPD
results suggest that acetonitrile molecule(s) was capable of
abstracting a phenolic proton of 1 to cause SPD, but the resulting
excited species was quenched or the chemiexcitation process did
not function effectively by backward protonation from acetonitrile
molecule(s) or its protonated species. Anyway, SPD apparently
related closely to characteristics as electron-pair donor (EPD)/elec-
tron-pair acceptor (EPA) and/or hydrogen-bond donor (HBD)/
hydrogen-bond acceptor (HBA) for aprotic polar solvents, as illus-
trated typically by Gutmann’s DN (donor number) and AN (accep-
tor number).15–18 However, simple correlation between these
parameters and chemiluminescence properties especially kSPD
could not be observed at present.
1. Beck, S.; Köster, H. Anal. Chem. 1990, 62, 2258–2270.
2. Adam, W.; Reihardt, D.; Saha-Möller, C. R. Analyst 1996, 121, 1527–1531.
3. Matsumoto, M. J. Photochem. Photobiol. C: Photochem. Rev. 2004, 5, 27–53.
4. Matsumoto, M.; Watanabe, N. Bull. Chem. Soc. Jpn. 2005, 78, 1899–1920.
5. Adam, W. In The Chemistry of Peroxide; Patai, S., Ed.; Wiley: New York, 1983; p
829.
6. Adam, W. In Small Ring Heterocycles; Hassner, A., Ed.; Wiley: New York, 1986; p
351.
7. The BID of 1 has been reported to show chemiluminescence effectively in
acetonitrile and even in water.8
8. Matsumoto, M.; Akimoto, T.; Matsumoto, Y.; Watanabe, N. Tetrahedron Lett.
2005, 46, 6075–6078.
9. Ikegami, M.; Arai, T. J. Chem. Soc., Perkin Trans. 2002, 2, 1296–1301. See also
references cited therein.
10. Rodembusch, F. S.; Campo, L. F.; Stefani, V.; Rigacci, A. J. Mater. Chem. 2005, 15,
1537–1541. See also references cited therein.
11. Authentic epoxide 5 was exclusively produced by reduction of 1 with triphenyl
phosphine.
12. Trofimov, A. V.; Mielke, K.; Vasil’ev, R. F.; Adam, W. Photochem. Photobiol. 1996,
63, 463–467.
13. Baumstark, A. L. In The 1,2-Dioxetane Ring System: Preparation, Thermolysis, and
Insertion Reactions in Singlet O2; Frimer, A. A., Ed.; CRC: Florida, 1985; Vol. II, pp
1–35.
14. Matsumoto, M.; Murakami, H.; Watanabe, N. Chem. Commun. 1998, 2319–
2320.
15. Gutmann, V. The Donor–Acceptor Approach to Molecular Interaction; Plenum
Press: New York, 1978.
16. Reichardt, C. Solvent Effects in Organic Chemistry; Chemie: Weinheim, 1987.
17. Chastrette, M. Tetrahedron 1979, 35, 1441–1448.
18. Maria, P.-C.; Gal, J.-F.; de Franceschi, J.; Fargin, E. J. Am. Chem. Soc. 1987, 109,
483–492.