Chemiluminogenic features of 10-Methyl-9-(phenoxycarbonyl)acridinium Trifluoromethanesulfonates alkyl substituted at the benzene ring in aqueous media

Article abstract of DOI:10.1021/jo1020882

10-Methyl-9-(phenoxycarbonyl)acridinium trifluoromethanesulfonates bearing alkyl substituents at the benzene ring were synthesized, purified, and identified. In the reaction with OOH^- in basic aqueous media, the cations of the compounds investigated were converted to electronically excited 10-methyl-9-acridinone, whose relaxation was accompanied by chemiluminescence (CL). The kinetic constants of CL decay, relative efficiencies of light emission, chemiluminescence quantum yields, and resistance toward alkaline hydrolysis were determined experimentally under various conditions. The mechanism of CL generation is considered on the basis of thermodynamic and kinetic parameters of the reaction steps predicted at the DFT level of theory. The chemiluminescence efficiency is the result of competition of the electrophilic center at C(9) between nucleophilic substitution by OOH ^- or OH^- and the ability of the intermediates thus formed to decompose to electronically excited 10-methyl-9-acridinone. Identification of stable and intermediate reaction products corroborated the suggested reaction scheme. The results obtained, particularly the dependency of the "usefulness" parameter, which takes into account the CL quantum yield and the susceptibility to hydrolysis, on the cavity volume of the entity removed during oxidation, form a convenient framework within which to rationally design chemiluminescent 10-methyl-9-(phenoxycarbonyl)acridinium cations.

Full text of DOI:10.1021/jo1020882

pubs.acs.org/joc
Chemiluminogenic Features of 10-Methyl-
9-(phenoxycarbonyl)acridinium Trifluoromethanesulfonates Alkyl
Substituted at the Benzene Ring in Aqueous Media
,†
Karol Krzyminski,* Agnieszka Ozog, Piotr Malecha, Alexander D. Roshal,
†
†
‡
ꢀ
_ꢀ
†
†
†
ꢀ
_
Agnieszka Wroblewska, Beata Zadykowicz, and Jerzy Bzazejowski
†
‡
Faculty of Chemistry, University of Gdansk, J. Sobieskiego 18, 80-952 Gdansk, Poland, and Institute of
ꢀ
Chemistry, Kharkiv V.N. Karazin National University, Svoboda 4, 61077 Kharkiv, Ukraine
ꢀ
karolk@chem.univ.gda.pl
Received October 29, 2010
10-Methyl-9-(phenoxycarbonyl)acridinium trifluoromethanesulfonates bearing alkyl substituents at the
benzene ring were synthesized, purified, and identified. In the reaction with OOH- in basic aqueous media,
the cations of the compounds investigated were converted to electronically excited 10-methyl-9-acridinone,
whose relaxation was accompanied by chemiluminescence (CL). The kinetic constants of CL decay, relative
efficiencies of light emission, chemiluminescence quantum yields, and resistance toward alkaline hydrolysis
were determined experimentally under various conditions. The mechanism of CL generation is considered on
the basis of thermodynamic and kinetic parameters of the reaction steps predicted at the DFT level of theory.
The chemiluminescence efficiency is the result of competition of the electrophilic center at C(9) between
nucleophilic substitution by OOH- or OH- and the ability of the intermediates thus formed to decompose to
electronically excited 10-methyl-9-acridinone. Identification of stable and intermediate reaction products
corroborated the suggested reaction scheme. The results obtained, particularly the dependency of the
“usefulness” parameter, which takes into account the CL quantum yield and the susceptibility to hydrolysis,
on the cavity volume of the entity removed during oxidation, form a convenient framework within which to
rationally design chemiluminescent 10-methyl-9-(phenoxycarbonyl)acridinium cations.
Introduction
ting species are electronically excited 9-acridinones formed
as a result of the decomposition of the addition products of
OOH- to acridinium cations.2a,3,4 This feature of acridinium
salts has prompted their use as chemiluminescent indicators
or chemiluminogenic fragments of labels in medical,5
biochemical,5e,6 chemical,7 and environmental8 analyses.
They have been employed in quantitative immunoassays of
antigens, antibodies, or hormones with detection limits at the
subattomole levels (detection limits as low as 10-17 M have
been reported).5c,e They can also be used to determine concen-
trations of highly reactive entities occurring in living matter,
such as OOH-.7,8 They have been successfully applied in
flow-injection techniques,7 cellular studies, and nucleic acid
diagnostics,9 as well as in electrochemiluminogenic assays of
Acridinium derivatives are readily oxidized with hydrogen
peroxide,1 persulfates,1b and other oxidants2 in alkaline
media leading to chemiluminescence (CL). The light-emit-
(1) (a) McCapra, F. In Chemistry of Heterocyclic Compounds; Acheson,
R. M., Ed.; Wiley: New York, 1973; Vol. 9 (Chemiluminescent Reactions of
Acridines), pp 615-630. (b) Gundermann, K. D.; McCapra, F. In Reactivity
and Structure Concepts in Organic Chemistry; Springer-Verlag: Berlin,
Heidelberg, 1987; Vol. 23 (Chemiluminescence in Organic Chemistry); pp
109-118. (c) Zomer, G.; Stavenuiter, J. F. C.; Van den Berg, R. H.; Jansen,
E. H. J. M. Pract. Spectrosc. 1991, 12, 505–521.
(2) (a) Yang, M.; Liu, C.; Hu, X.; He, P.; Fang, Y. Anal. Chim. Acta 2002,
461, 141–146. (b) Wroblewska, A.; Huta, O. M.; Patsay, I. O.; Petryshyn,
R. S.; Blazejowski, J. Anal. Chim. Acta 2004, 507, 229–236. (c) Aizawa, M.;
Ikariyama, Y.; Kobatake, E.; Ogasawara, M.; Tanaka, M. Luminescence by
reacting an acridinium ester with superoxide. United States Reissued Patent,
US RE39,047 E, Mar 28, 2006. (d) Lai, Y.; Qi, Y.; Wang, J.; Chen, G. Analyst
2009, 134, 131–137. (e) He, Y.; Zhang, H.; Chai, Y.; Cui, H. Anal. Bioanal.
Chem. 2010, DOI: 10.1007/s00216-010-4157-y.
(3) McCapra, F. Pure Appl. Chem. 1970, 24, 611–629.
(4) Rak, J.; Skurski, P.; Blazejowski, J. J. Org. Chem. 1999, 64, 3002–
3008.
1072 J. Org. Chem. 2011, 76, 1072–1085
Published on Web 01/19/2011
DOI: 10.1021/jo1020882
r
2011 American Chemical Society

ꢀ
Krzyminski et al.
JOCArticle
substances.2a,c Acridinium salts exhibit a higher chemilumines-
cence quantum yield in aqueous media relative to luminol
derivatives and, unlike the latter systems, do not require a
catalyst to cause the emission of light.1,5 For these reasons,
numerous commercially available chemiluminescent labels
contain a 9-(phenoxycarbonyl)acridinium fragment substi-
tuted at the benzene moiety.5e,9,10 A disadvantage of the use
of these compounds, however, is their tendency to form non-
chemiluminescing “pseudobases” in aqueous basic or neutral
media.1,11,12 In order to minimize this effect, derivatives sub-
stituted at the benzene ring have been investigated under a wide
range of analytical conditions.10 There are also other acridi-
nium derivatives substituted at C(9) with carboxamide,13
cyanide,2b and hydroxamic acid or other groups,5c whose
chemiluminogenic features have been investigated in the con-
text of possible analytical applications.
The stability of acridinium cations functionalized at C(9)
and their susceptibility to oxidation depend primarily on the
structure of the fragment removed during the formation of
electronically excited, N(10)-substituted, 9-acridinones.1,10,13
The easily removable fragments give rise to flash-type CL,
during which the efficiency of light emission is high.10c,d,12 It has
been found that CL efficiency is generally related to the pKa
values of the phenols that are the precursors of N(10)-sub-
stituted 9-(phenoxycarbonyl)acridinium cations.1,13 If these
cations are derived from phenols exhibiting a pKa < 11.97
(pKa of HOOH), they may well turn out to be efficient
chemiluminogens.3,13,14 Unfortunately, however, a higher CL
efficiency is usually accompanied by a greater susceptibility to
hydrolysis and lower stability of the compounds in aqueous
systems. On the other hand, the substituents on the acridinium
nucleus affect the wavelength of the emitted light and alter the
susceptibility of entities to hydrolysis.14,15
The mechanism of oxidation of N(10)-substituted 9-(phe-
noxycarbonyl)acridinium cations with hydrogen peroxide in
alkaline media has been discussed by several authors;1,3,5c,13
we examined it at the semiempirical level of theory.4 Com-
putations have shown that the commonly suggested reaction
pathway, involving the formation of dioxethanone as inter-
mediate and the elimination of CO2, is not the most probable
means by which light is generated. The results of our studies
suggest that light-emitting molecules of 10-methyl-9-acridi-
none are formed as a result of the elimination of the phenyl
carbonate anion, and not CO2 as other authors have sug-
gested,1,5c from the cyclic intermediate that is formed after
the initial addition of OOH- to the 10-methyl-9-(pheno-
xycarbonyl)acridinium cation and the subsequent abstrac-
tion of a proton by OH-. Analysis of the oxidation of the
9-cyano-10-methylacridinium cation revealed a similar mech-
anism of CL generation.16
This work focuses on experimental and computational
investigations of the chemiluminescent features of 10-methyl-
9-(phenoxycarbonyl)acridinium cations substituted with
various alkyl groups in the phenyl fragment (MPCAþ)
(Chart 1). By undertaking these investigations we hoped
not only to extend our knowledge of practically important
chemiluminogens but also to gain an opportunity to study
the relations between the physicochemical properties and
structure and the chemiluminogenic ability and composition
of the reacting systems, which would form a framework for
the rational design and analytical use of these acridinium
derivatives.
Results and Discussion
(5) (a) Weeks, I.; Beheshti, I.; McCapra, F.; Campbell, A. K.; Woodhead,
J. S. Clin. Chem. 1983, 29, 1474–1479. (b) Weeks, I.; Sturgess, M.; Brown,
R. C.; Woodhead, J. S. Methods Enzymol. 1986, 133, 366–387. (c) Dodeigne,
C.; Thunus, L.; Lejeune, R. Talanta 2000, 51, 415–439. (d) Kricka, L. J. Anal.
Chim. Acta 2003, 500, 279–286. (e) Sub-Attomole Labeling Kit, http://www.
assaydesigns.com/objects/catalog/product/extras/915-049.pdf (accessed Oct
1, 2010).
(6) (a) Campos, L. M.; Cavalcanti, C. L.; Lima-Filho, J. L.; Carvalho,
L. B.; Beltrao, E. I. Biomarkers 2006, 11, 480–484. (b) Scorilas, A.; Agiamarnioti,
K.; Papadopoulos, K. Clin. Chim. Acta 2005, 357, 159–167.
(7) (a) Garcia-Campana, A. M.; Baeyens, W. R. G.; Cuadros-Rodriguez,
L.; Ales Barrero, F.; Bosque-Sendra, J. M.; Gamiz-Gracia, L. Curr. Org.
Chem. 2002, 6, 1–20. (b) Garcia-Campana, A. M.; Gamiz-Gracia, L.;
Baeyens, W. R. G.; Bariero, F. A. J. Chromatogr. B 2003, 793, 49–74.
(c) Atanassova, D.; Kefalas, P.; Psillakis, E. Environ. Int. 2005, 31, 275–
280.
Characteristics of the Compounds. Electronic absorption
and emission,17 IR,18 NMR,19 thermoanalytical,20 X-ray,21,22
and computational4,17-20,22 methods were employed to de-
termine various features of the compounds investigated.
Some other properties are reported below.
There appear to be conspicuous differences between MPCAþ
retention times (Rt) (Table 1S, Supporting Information).
Cations containing voluminous alkyl substituents or that are
(16) Wroblewska, A.; Huta, O. M.; Midyanyj, S. V.; Patsay, I. O.; Rak, J.;
Blazejowski, J. J. Org. Chem. 2004, 69, 1607–1614.
(8) King, D. W.; Cooper, W. J.; Rusak, S. A.; Peake, B. M.; Kiddle, J. J.;
O’Sullivan, D. W.; Melamed, M. L.; Morgan, C. R.; Theberge, S. M. Anal.
Chem. 2007, 79, 4169–4176.
(9) (a) Fan, A.; Cao, Z.; Li, H.; Kai, M.; Lu, J. Anal. Sci. 2009, 25, 587–
597. (b) Brown, R. C.; Li, Z.; Rutter, A. J.; Mu, X.; Weeks, O. H.; Smith, K.;
Weeks, I. Org. Biomol. Chem. 2009, 7, 386–394.
(10) (a) Sato, N. Tetrahedron Lett. 1996, 37, 8519–8522. (b) Razavi, Z.;
McCapra, F. Luminescence 2000, 15, 239-244, 245–249. (c) Smith, K.; Li, Z.;
Yang, J.-J.; Weeks, I.; Woodhead, J. S. J. Photochem. Photobiol. A: Chem.
2000, 132, 181–191. (d) Smith, K.; Yang, J.-J.; Li, Z.; Weeks, I.; Woodhead,
J. S. J. Photochem. Photobiol. A: Chem. 2009, 203, 72–79.
(11) Hammond, P. W.; Wiese, W. A.; Waltrop, A. A.; Nelson, N. C.;
Arnold, L. J. J. Biolumin. Chemilumin. 1991, 6, 35–43.
(17) Krzyminski, K.; Roshal, A. D.; Niziolek, A. Spectrochim. Acta, Part
A 2008, 70, 394–402.
(18) Zadykowicz, B.; Ozog, A.; Krzyminski, K. Spectrochim. Acta, Part A
2010, 75, 1546–1551.
(19) Krzyminski, K.; Malecha, P.; Zadykowicz, B.; Wroblewska, A.;
Blazejowski, J. Spectrochim. Acta, Part A 2011, 78, 401-409.
(20) Krzyminski, K.;Malecha, P.;Storoniak,P.;Zadykowicz, B.;Blazejowski,
J. J. Therm. Anal. Cal. 2010, 100, 207–214.
(21) (a) Sikorski, A.; Krzyminski, K.; Konitz, A.; Blazejowski, J. Acta
Crystallogr., Sect. C 2005, 61, o50–o52. (b) Sikorski, A.; Krzyminski, K.;
Bialonska, A.; Lis, T.; Blazejowski, J. Acta Crystallogr., Sect. E 2006, 62,
o555–o558. (c) Sikorski, A.; Krzyminski, K.; Bialonska, A.; Lis, T.;
Blazejowski, J. Acta Crystallogr., Sect. E 2006, 62, o822–o824. (d) Sikorski,
A.; Krzyminski, K.; Malecha, P.; Lis, T.; Blazejowski, J. Acta Crystallogr.,
Sect. E 2007, 63, o4484–o4485. (e) Trzybinski, D.; Krzyminski, K.; Sikorski,
A.; Malecha, P.; Blazejowski, J. Acta Crystallogr., Sect. E 2010, 66, o826–
o827. (f) Trzybinski, D.; Krzyminski, K.; Sikorski, A.; Blazejowski, J. Acta
Crystallogr., Sect. E 2010, 66, o906–o907. (g) Trzybinski, D.; Krzyminski,
K.; Sikorski, A.; Blazejowski, J. Acta Crystallogr., Sect. E 2010, 66, o1311–
o1312. (h) Trzybinski, D.; Krzyminski, K.; Blazejowski, J. Acta Crystallogr.,
(12) Kaltenbach, M. S.; Arnold, M. A. Microchim. Acta 1992, 108, 205–
219.
(13) Adamczyk, M.; Mattingly, P. G. In Luminescence Biotechnology
Instruments and Applications (Chemiluminescent N-Sulfonylacridinum 9-
Carboxamides and Their Application in Clinical Assays); Van Dyke, K.,
Van Dyke, C., Woodfork, K., Eds.; CRC Press: Boca Raton, 2002; pp
77-103.
(14) Krzyminski, K.; Roshal, A. D.; Zadykowicz, B.; Bialk-Bielinska, A.;
Sieradzan, A. J. Phys. Chem. A 2010, 114, 10550–10562.
(15) Natrajan, A.; Sharpe, D.; Costello, J.; Jiang, Q. Anal. Biochem. 2010,
406, 204–213.
Sect. E 2010, 66, o2773–o2774. (i) Trzybinski, D.; Krzyminski, K.;
Blazejowski, J. Acta Crystallogr., Sect. E 2010, 66, o2929–o2930.
(22) Niziolek, A.; Zadykowicz, B.; Trzybinski, D.;Sikorski, A.; Krzyminski,
K.; Blazejowski, J. J. Mol. Struct. 2009, 920, 231–237.
J. Org. Chem. Vol. 76, No. 4, 2011 1073

ꢀ
Krzyminski et al.
JOCArticle
CHART 1. Canonical Structure of 10-Methyl-9-(phenoxycarbonyl)acridinium Cations Investigated (MPCAþ) with the Numbering of
Atoms Indicated
multiply substituted in the benzene ring exhibit higher tR values
in the reversed-phase system. The Rt values of selected MPCAþ
follow the trend of changes in the hydrophobicity parameter
(0.56 (Me), 1.02 (Et), 1.53 (i-Pr), and 1.98 (t-Bu));23 i.e., they
increase in the order 2 < 5 < 6 < 7. Rt values also tend to
increase linearly with increasing values of the base 10 logarithm
of the n-octanol/water partition coefficient and of the molar
refractivity computed using the Biobyte program (Figure 1S,
Supporting Information).24 The similar dependencies noted in
other systems follow the general rule that the retention time is
proportional to the hydrophobicity of the analyte.25
The MS spectra of the compounds investigated are similar.
The common signal, always appearing at m/z = 193 (Figure
2S, Supporting Information), can be attributed to the cation
of the N(10)-methylated acridinium nucleus, which is quite
resistant to fragmentation.
selected compounds (1, 2, and 9) are presented in Figure 3S
together with the fluorescence spectrum of 10-methyl-
9-acridinone (Supporting Information). In basic aqueous
media, CL ranges from 400 to 550 nm, reaching a maximum
around 450 nm, like the emission due to the oxidation of
other acridinium cations.2b,26 The emitting entity is 10-
methyl-9-acridinone, as emerges from a comparison of ex-
perimental spectra and computational predictions for this
compound.14,27 Further support for this conclusion comes
from a comparison of the CL and the fluorescence spectra
of the oxidation products of 1, 2, and 9. The spectral patterns
are almost identical in all cases. This confirms that 10-
methyl-9-acridinone is indeed the emitting entity, accounting
for the chemiluminescence in the systems investigated.
As CL and fluorescence spectra are broad and exhibit a
complex pattern, we measured the emission decay of 10-
methyl-9-acridinone dissolved in acetonitrile in order to find
out whether it originates from one or more excited forms.
Fluorescence decay appears to be monoexponential, which
implies that emission comes from only one excited form
General Features of Chemiluminescence. Stationary spec-
tra of the chemiluminescence accompanying the oxidation of
(23) Hansch, C.; Leo, A. In Substituent Constants for Correlation Analysis
in Chemistry and Biology; Wiley: New York, 1979.
(24) BioByte, Inc., Claremont, CA, 1999; available: www.biobyte.com
(accessed 1/10/2010).
(26) Cass, M. W.; Rapaport, E.; White, E. H. J. Am. Chem. Soc. 1972, 94,
3168–3175.
(25) Paschke, A.; Neitzel, P. L.; Walther, W.; Schuurmann, G. J. Chem.
Eng. Data 2004, 49, 1639–1642.
(27) Bouzyk, A.; Jozwiak, L.; Kolendo, A.Yu.; Blazejowski, J. Spectro-
chim. Acta, Part A 2003, 59, 543–558.
1074 J. Org. Chem. Vol. 76, No. 4, 2011

ꢀ
Krzyminski et al.
JOCArticle
TABLE 2. Physicochemical Characteristics of Chemiluminescence
Accompanying the Reaction of MPCAþ with OOH- in Alkaline Aqueous
Media
relative CL efficiency
at various pHb (%)
12.0
10.0
8.0
compd no.
(Chart 1)
CQYa ꢀ 10-2
1
2
3
4
5
6
7
8
1.88
1.60
1.88
1.71
1.31
1.08
0.63
1.22
1.12
1.49
2.21
0.59
100.0
85.0
100.3
91.2
70.0
57.3
33.5
65.2
59.9
79.4
117.6
31.4
87.0
39.5
84.4
73.5
13.9
10.1
4.5
22.6
7.3
59.3
94.6
9.2
6.48
0.86
5.16
2.40
0.56
0.34
1.11
0.39
0.42
1.32
4.01
5.45
9
10
11
12
FIGURE 1. Time-dependent chemiluminescence intensity, in rela-
tive light units (RLU), profiles for 1, 2, and 4 (c(MPCAþ) = 2 ꢀ
10-5 M, c(H2O2) = 4.9 mM, pH = 11.6) (points), together with
curves obtained from fitting with eqs 3 and 4.
aMean values (in einstein/mol) of five replicate measurements
(c(MPCAþ) = 2 ꢀ 10-8 M, c(H2O2) = 4.9 mM, pH = 11.6) obtained
using luminol as a standard28 (the coefficient of variability varied
between 0.02 and 0.08). bValues obtained from the formula 100% ꢀ
light sum of 2-12 (pH = 8.0, 10.0 and 12.0)/light sum of 1 (pH = 11.6,
c(MCPAþ) = 2 ꢀ 10-5 M, c(H2O2) = 4.9 mM).
TABLE 1. Kinetic Data Derived from the Analysis of Time-Dependent
Chemiluminescence Profilesa
compd no.
(Chart 1)
k1
k2
tmax
C
1
2
3
4
5
6
7
8
2.49
2.69
2.22
2.41
2.90
3.25
2.90
3.05
3.81
2.37
2.46
2.71
2.82
0.50
1.0
0.62
0.62
1.1
2790
1160
2670
4280
3130
2280
161
871
303
7070
1960
110
0.119
2.22
1.02
0.068
0.041
0.004
0.095
0.006
0.893
1.75
0.93
22
0.99
16
0.74
0.62
2.1
9
10
11
12
0.004
ak1 and k2 (in s-1), tmax (in s) and C (in RLU) (eqs 3 and 4) represent
mean values from the fitting of at least five replicate time-dependent
chemiluminescence intensity profiles (Figure 1); the correlation coeffi-
cient for the fitting of an individual profile was better than 0.95.
(Figure 4S, Supporting Information). The average fluores-
cence lifetime was found to be 6.6 ꢀ 10-9s.
FIGURE 2. Dependence of k2 on H2O2 concentration for 1, 2, and
4 (c(MPCAþ) = 2 ꢀ 10-8 M, pH = 11.6) (points), together with
lines resulting from least-squares fitting.
Kinetic Features of Chemiluminescence. Chemilumines-
cence intensity always grows, reaches a maximum, and falls
in time. The fitting of time-dependent CL profiles, an
example of which is given in Figure 1, with eqs 3 and 4 for
two-step consecutive chemical processes yields the kinetic char-
acteristics shown in Table 1. Values of k1, reflecting CL growth,
vary between 2.2 and 3.9 s-1. They account for the competitive
reaction of OOH- or OH- with MPCAþ and also for the
diffusion processes involving these entities, which are difficult
to quantify in kinetic analysis. Furthermore, the relatively small
number of points harvested in the initial region of the CL-time
profiles causes uncertainties in the evaluation of k1. For these
reasons, k1 seems to be less suitable for an in-depth analysis.
Values of k2, reflecting chemiluminescence decay, are
more widely differentiated than k1 (Table 1). The lowest k2
characterizes the drop in chemiluminescence in 7 and 12, the
highest, in 1. If k2 > 1 s-1, flash-type chemiluminescence
occurs. When k2 values are low, long-lasting (glow-type)
chemiluminescence is observed.9b,10c,d The data in Table 1
indicate that there is an approximately inverse relationship
between k2 and tmax (eq 4). Low tmax values correspond to
cases of flash-type chemiluminescence. Values of C (eq 3) are
qualitatively related to the CL efficiency/chemiluminescence
quantum yield (Table 2) and can thus be considered as a
measure of the probability that MPCAþ will be converted to
light emitting species.
A set of k1, k2, tmax, and C reflects the chemiluminogenic
features and is characteristic of a given MPCAþ. CL is quite
weak in the case of 6, 7, 9, and 12, i.e., when bulky (i-Pr and
t-Bu) substituents (6 and 7) or two Me substituents (9 and 12)
are in the ortho position of the benzene ring. Values of k2 in
the series of 2, 5, 6, and 7 fall with a decrease in Taft’s size
parameter, which characterizes the steric effects of the
substituents Me, Et, i-Pr, and t-Bu in the ortho position of
the benzene ring (respectively equal to -1.24, -1.31, -1.71,
and -2.78),23 and rise with an increase in Charlton’s size
constant (respectively equal to 0.52, 0.56, 0.76, and 1.24).23
Values of k2 decrease linearly with increasing H2O2 con-
centration, which is shown for selected compounds in Figure 2.
The slopes of these dependencies, respectively, equal to 31.1,
J. Org. Chem. Vol. 76, No. 4, 2011 1075

ꢀ
Krzyminski et al.
JOCArticle
FIGURE 3. Number of photons emitted from 1 mol of MPCAþ
(n.p.) versus concentration of 1, 2, and 4 (c(H2O2) = 4.9 mM, pH =
11.6). Lines correspond to the following equations: n.p. = 1.16 ꢀ
1022 ꢀ c(MPCAþ) (1), n.p. = 9.03 ꢀ 1021 ꢀ c(MPCAþ) (2), n.p. =
1.13 ꢀ 1022 ꢀ c(MPCAþ) (4).
FIGURE 4. Stability of MPCAþ (c = 2 ꢀ 10-5 M) over time in
basic medium (c(H2O2) = 4.9 mM, pH = 8.0).
einstein/mol);28 meta-substituted (11 and 3) and unsubsti-
tuted (1) compounds have the highest values. Values of CQY
decrease with increasing size of the ortho substituent in the
benzene ring (in the order 2 > 5 > 6 > 7) or when two Me
substituents are present in the ortho position (9 and 12). If Me
is in the para position (4), there is a slight decrease in CQY
compared to the unsubstituted compound (1). A rise in pH
from 8.0 to 12.0 substantially increases CQY (Table 2). The
interrelations between CQYs are similar at various pH,
which means that structural factors affect the chemilumino-
genic ability of MPCAþ in a similar way as the basicity of the
medium changes.
1.23, and 4.56 M-1 s-1 for 1, 2, and 4, reflect the overall
processes in which MPCAþ participates and which do or do
not lead to chemiluminescence. A quantitative measure of
chemiluminescence is the light sum, the area under the
CL-time profile. Light sums/chemiluminescence quantum
yields gradually decrease with H2O2 concentration (Figures
5S and 6S, Supporting Information), which raises the pro-
blem of the choice of the appropriate value of this quantity in
chemiluminogenic assays. At pH 11.6, the value assumed in
assays using acridinium chemiluminogens,10 about half the
H2O2 molecules (pKa = 11.97), are in anionic (OOH-) form.
In all experiments, therefore, there is a large excess (several
orders of magnitude) of OOH- over MPCAþ. Lowering the
H2O2 concentration leads to an increase in the light sums, as
mentioned above. However, at H2O2 concentrations lower
than 0.5 mM, the reproducibility of the assays deteriorates.
Thus, an H2O2 concentration equal to 4.9 mM was assumed
to be optimal in our investigations.
Stability of MPCAþ in Basic Media. MPCAþ lose 8-40%
of their initial chemiluminogenic ability after 15 min of
thermostatting and 14-70% after 30 min of thermostatting
at pH = 8.0 (Figure 4). The chemiluminogenic ability of 1
falls most distinctly in time. All other cations substituted at
the benzene ring are more resistant to the influence of alka-
line media and the decrease of chemiluminogenic ability. The
resistance of MPCAþ to alkaline media increases slightly
with the size of the substituent in the ortho position, i.e., on
the order 2 < 5 < 6 < 7. The resistance is the greatest if two
Me substituents are present in the ortho position (9). Meta-
substituted derivatives (3 and 11) are only slightly more
resistant to alkaline hydrolysis than the unsubstituted com-
pound (1). Compound 12, containing two Me substituents in
the ortho position and one Me in the para position, exhibits a
lower resistance to alkaline media than 9, which has only two
Me substituents in the ortho position.
Chemiluminescence Quantum Yields. The plots of the total
number of photons emitted (light output) against MPCAþ
concentration are linear over a wide range of concentrations,
as in the case of other acridinium chemiluminogens.1c,5e This
indicates that the compounds investigated can be used as
chemiluminescent indicators. The calibration graphs for 1, 2,
and 4 are presented by way of example in Figure 3. The slopes
of these graphs are, respectively, equal 1.13 ꢀ 1022 (1), 9.61 ꢀ
1021 (2), and 1.03 ꢀ 1022 (4) and represent the total number of
photons emitted from 1 mol of MPCAþ. This information
enables one to obtain the chemiluminescence quantum yields
(CQY), given in Table 2, using luminol as a reference system.
The lowest measurable light output reflects the detection limit of
MPCAþ. The minimum detectable concentration of the com-
pounds investigated at pH = 11.6 is in the 10-10 M range.
Compounds exhibiting flash-type chemiluminescence are gen-
erally more readily detectable (lower mdc) than chemilumino-
gens showing glow-type chemiluminescence (higher mdc).
Quantum yields (Table 2) directly reflecting the tendency
The above findings may indicate the site to which MPCAþ
should be attached to investigate or assay macromolecules.
Usually, the links are via a spacer and an active group to the
para position of the benzene ring,5e although the resistance to
hydrolysis of a tracer or label should be improved if ortho
positions are used for connecting the side arm.10d
The results of our investigations concerning the stability of
MPCAþ in basic media follow the trends reported so far,10a,d
but the values of the relevant properties are generally lower in
our case (Table 2). On the other hand, the stability of MPCAþ
was investigated mostly in acidic aqueous media, the environ-
ment in which tracers are stored prior to analytical use.
The question arises as to how aqueous alkaline media
influence the stability of MPCAþ and, consequently, their
chemiluminogenic ability. It is thought that C(9) is susceptible
of MPCAþ to chemiluminesce lie in the range 0.59 ꢀ 10-2
-
2.21 ꢀ 10-2 einstein/mol (CQY for luminol =1.23 ꢀ 10-2
(28) Ando, Y.; Niwa, K.; Yamada, N.; Irie, T.; Enomoto, T.; Kubota, H.;
Ohmiya, Y.; Akiyama, H. Photochem. Photobiol. 2007, 83, 1205–1210.
1076 J. Org. Chem. Vol. 76, No. 4, 2011

ꢀ
Krzyminski et al.
JOCArticle
to nucleophilic substitution simultaneously by OOH- (the
form of H2O2 existing in alkaline media) and OH- (pH >
7.0).1b,4,13,16 In the former case, an intermediate is formed
that is subsequently converted to electronically excited light
emitting 10-methyl-9-acridinone. In the latter case, a so-
called “pseudobase” is formed,1,4,13 which is not converted
to a light emitting species. The introduction of MPCAþ to
alkaline media would thus bring about the formation of
the relevant “pseudobases”, which cannot be converted to the
products of the addition of OOH- to MPCAþ because the
activation barrier for this process is quite high.2b For these
reasons, MPCAþ and H2O2 are mixed in acidic conditions,
and chemiluminescence is triggered by the addition of the
aqueous base solution. In such a situation, OH- and OOH-
appear simultaneously and compete for the cation according
to the rules of probability of encountering reacting entities,
since both processes proceed in the absence of activation
barriers.4,16 Therefore, the loss of the chemiluminogenic
ability of MPCAþ in alkaline media is caused predominantly
by “pseudobase” formation and its subsequent transforma-
tion to nonexcited molecules of 10-methyl-9-acridinone.1a,5c
Another possible reason for the loss of chemiluminogenic
ability is the partial alkaline hydrolysis of MPCAþ, resulting
from the attack of OH- and/or OOH- on the carbonyl C(15)
atom (Chart 1), leading to the production of nonchemilumi-
nogenic 9-carboxy-10-methylacridinium acid.9b,13,14
Origin of Chemiluminescence. According to the mecha-
nisms of the oxidation of 10-methylacridinium derivatives by
H2O2 in alkaline media put forward in the literature,1,4,5c,13
the primary steps are determined mainly by the location of
the electrophilic center in the cation. As indicated by frontier
orbital theory,29 the LUMO distribution in an electrophilic
species (cations) determines the molecular centers sensitive
to nucleophilic attack of OOH- or OH-. The LCAO coeffi-
cients of the pz MPCAþ’s LUMO is ca. 10 times higher at the
endocyclic C(9) than at the carbonyl group C(15) atom
(0.325 and 0.04, respectively; average for 1, 4, 7, and 12).19
This implies that C(9) rather than the C(15) atom of the
carbonyl group is the site of the primary nucleophilic attack
of the above-mentioned anions, despite the fact that the
electronic charge deficiency at C(9) is lower than at C(15)
(Mulliken relative partial charges are 0.09 and 0.51, respec-
tively; average for 1, 4, 7, and 12).19
preferences, which seem to be similar for steps I-IV. In this
way, a certain amount of intermediate 16, identified
experimentally,13,14 is formed, which is the precursor of the
light-emitting species. The possible simultaneous occurrence
of steps I-IV lessens, however, the probability of the for-
mation of 16 and consequently the efficiency of CL. More-
over, 23 would be difficult to convert to 16 by SN2
substitution with OOH- and likewise 16 to 23 by SN2
substitution with OH-, owing to the expected relatively high
activation barrier to such processes.2b Thus, once 16 is
produced, it reacts immediately with OH- to form the cyclic
intermediate 17, which then decomposes to the energy-rich
10-methyl-9-acridinone (19) and phenyl carbonate anion
(20) (Table 3). The energy released in step VI (Table 3)
exceeds that necessary to cause electronic excitation of 10-
methyl-9-acridinone (88.2 kcal/mol).27 Hence, the excited
10-methyl-9-acridinone molecules emit light, which is man-
ifested as CL. Electronically excited molecules of 10-methyl-
9-acridinone could also be formed via steps III, XIII, and VI.
The latter pathway is, however, less probable than that via
steps I, V, and VI, since the thermodynamic force (Δr,298Go)
for step III is lower than that for step I (Table 3) and the
probability of nucleophilic substitution of OOH- at the
carbonyl C(15) atom is much lower than at C(9), as indicated
by the values of the pz LUMO. It has long been assumed that
the formation of electronically excited 10-methyl-9-acridi-
none is accompanied by the release of phenoxy anions from
17 and the subsequent elimination of CO2 from the inter-
mediate thus formed (10-methyl-10H-spiro[acridine-9,30-
[1,2]dioxetane]-40-one (31) (Figure 6S, Supporting Informa-
tion)).1,5c We found previously, however, that such processes
need to overcome quite high activation barriers.4 The bar-
riers predicted for the latter process, 31 f 19 þ 22, are
respectively (Δa,298Ho (gaseous phase), Δa,298Go (gaseous
phase), Δa,298Go (aqueous phase)) equal to (in kcal/mol)
14.3, 14.9, and 15.9. They thus remain relatively high, which
suggests that the formation of electrically excited molecules
of 10-methyl-9-acridinone proceeds via the elimination of
phenyl carbonate anions rather than the consecutive release
of phenoxy anions and CO2. Phenyl carbonate anions do not
decompose instantaneously to CO2 and phenoxy anions in
alkaline media in reactions leading to chemiluminescence (step
VII) (Table 3). In acidic media, however, in which reaction
products were identified, the former ions can be decarboxy-
lated, as will be shown later, to form the corresponding
phenols, which were identified together with the ones produced
via steps IX, XI, or XIV (27 and 30 have been regarded as
byproducts of the oxidation of 10-methyl-9-(phenoxycarbonyl)-
acridinium cations in alkaline media).1,9b,14,26 The evidence for
the above mechanism would undoubtedly be strengthened by
the identification of phenyl carbonate anions. We have been
tackling this problem.
Another factor important for the generation of chemilu-
minescence is that H2O2 reacts spontaneously with OH-
yielding OOH-.1b,4,13,16
Taking into account the mechanisms of the chemilumines-
cence of 10-methylacridinium cations proposed in the
literature,1,5c and following our findings so far,4,16 we con-
sidered only certain processes, which are shown in Scheme 1.
These processes account for the chemiluminescence (steps I,
V, and VI), the formation of the so-called “pseudobase” (step
II), and the appearance of phenoxy anions (steps VII; II, VIII
and IX or II, X and XI; and IV and XIV) and 10-methyla-
cridinium-9-carboxylate anions (steps IV and XIV) among
the reaction products. All of the primary steps I-IV proceed
spontaneously since they are reactions between ions (Table 3).
Which intermediate (16, 23, 28, or 29) occurs depends on the
thermodynamic factors preferred in step II and on the kinetic
Reactions between ionic species (steps I-IV) and proton
abstraction by OH- (steps V, VIII, XIII, and XIV) exhibit
low activation barriers, if any. Thus, the processes in which
such barriers can be expected are steps VI and IX-XII. The
predicted barriers for step VI are moderate, which means
that elimination of phenyl carbonate anions and the forma-
tion of electronically excited 10-methyl-9-acridinone mole-
cules is achieved under the experimental conditions applied
(Table 4). This also confirms the predicted quite high rate
constants and low reaction completion times. It should,
(29) Fleming, I. Frontier Orbitals and Organic Chemical Reactions; Wiley:
London, 1976.
J. Org. Chem. Vol. 76, No. 4, 2011 1077

ꢀ
Krzyminski et al.
JOCArticle
SCHEME 1. Pathways of the Reactions of MPCAþ with OOH- and OH-
1078 J. Org. Chem. Vol. 76, No. 4, 2011

ꢀ
Krzyminski et al.
JOCArticle
however, be noted that Gibbs’ free energies of activation
(Table 4) do not account for the trends revealed by analysis
of chemiluminescence quantum yields (Table 2); 1 and 2
exhibit relatively high Gibbs’ free energy activation barriers
and relatively high CQY, and 9 exhibits a low activation
barrier and low CQY, but the expected tendency should be
the opposite. Therefore, DFT computations suggest that fac-
tors other than kinetic ones may determine the efficiency
of chemiluminescence. The formation of “pseudobase” (23)
initiates nonchemiluminescent pathways of MPCAþ deple-
tion. According to our predictions, the activation barriers to
steps X and XI, i.e., the substitution at OOH- of the carbonyl
C atom of 23 and the release of the phenoxy anion from the
entity thus formed, appear to be less than 0.1 kcal/mol each.
Activation barriers do, however, exist for the final steps (IX
and XII) of the dark transformation of MPCAþ to 10-methyl-
9-acridinone (Table 4), but the energy released in the latter two
processes is too small (Table 3) to electronically excite the
above product. The activation barriers to steps IX and XII,
responsible for the dark transformation of MPCAþ to 10-
methyl-9-acridinone, are higher than that of step VI; the
former steps should thus be slower than the latter one. Never-
theless, they may occur, and when they do, they cause irrever-
sible loss of the chemiluminogenic ability of MPCAþ.
Scheme 1 depicts chemical processes which may or may
not lead to chemiluminescence. It does not show, however,
how the released energy can be converted to electronic
excitation. There are numerous reports suggesting that the
decomposition of acridinium 1,2-dioxetanes is accompanied
by intramolecular chemically initiated electron exchange
luminescence (CIEEL) arising from the electronically excited
10-methyl-9-acridinones formed in the process.30 There are
grounds for believing that such a mechanism may also
account for the chemiluminescence generated during the
oxidation of 10-methyl-9-(phenoxycarbonyl)acridinium ca-
tions in alkaline media.
TABLE 3. Thermodynamic Data (in kcal/mol) of the Elementary Steps
of the Reactions of 1, 2, and 9 with OOH- and OH- a
gaseous phase
aqueous phase
step no. (Scheme 1) compd no.
Δ
r,298H° r,298G°
Δ
Δ
r,298G°
I
1
2
9
1
2
9
1
2
9
1
2
9
1
2
9
1
2
9
1
2
9
1
2
9
1
2
9
1
2
9
1
2
9
-166.6 -153.6
-167.6 -155.0
-167.9 -155.3
-186.1 -175.6
-193.6 -182.7
-194.0 -183.9
-139.1 -125.3
-133.6 -120.4
-132.7 -118.9
-157.9 -147.1
-158.7 -148.6
-157.0 -146.7
-51.5
-51.6
-53.3
-73.3
-74.4
-74.4
-33.1
-26.7
-27.8
-55.2
-55.9
-48.3
-37.7
-38.3
-35.6
-105.2
-104.4
-105.6
14.1
13.9
14.8
-32.1
-27.0
-28.4
-6.1
II
III
IV
V
-75.2
-74.9
-72.9
-75.6
-75.3
-72.7
VI
VII
VIII
IX
X
-94.1 -108.3
-94.4 -108.6
-95.4 -110.2
14.4
13.3
7.0
5.9
11.3
3.3
-72.9
-66.2
-67.4
3.4
2.2
2.4
-67.0
-59.5
-54.0
16.1
-73.1
-67.4
-67.3
-20.8
-22.1
-22.9
-53.0
-45.8
-38.4
1.2
According to the proposed chemiluminogenic pathway
(steps I, V, and VI in Scheme 1), the efficiency of emission
should rise with increasing OOH- concentration, since the
latter entity takes part in bimolecular step I. However, the
dependencies shown in Figures 5S and 6S (Supporting
Information) demonstrate the reverse trend, namely, the
decrease in chemiluminescence efficiency with increasing
OOH- concentration. A possible explanation for this find-
ing is that as the concentration of the latter entity increases,
so does the contribution of the parallel dark pathway
(depicted by steps II, X, XI, and XII in Scheme 1) in the
total consumption of the chemiluminogenic precursor (13).
Such an effect could also explain the drop in k2 values with
increasing OOH- concentration (Figure 2).
-7.1
-4.0
-24.1
-20.8
-15.2
1.2
XI
14.0
7.5
-1.5
-9.5
2.0
-1.9
XII
XIII
-84.4 -103.0
-102.6 -103.9
-108.8 -109.9
-108.2 -109.0
-87.2
-56.2
-62.9
-61.0
-58.2
-57.6
-64.5
1
2
9
1
2
9
XIV
-84.3
-85.6
-98.8
-99.8
-88.7 -103.6
aΔr,298H° and Δr,298G°, respectively, represent the enthalpy and
Gibbs’ free energy (gaseous phase) or free energy (aqueous phase) of
the reaction corresponding to a given step number at temperature
298.15 K and standard pressure.
The structures of selected MPCAþ and the entities shown
in Scheme 1 and mentioned in the text are presented in the
Supporting Information (Figure 6S) to give an idea of the
TABLE 4. Kinetic Characteristics of the Elementary Steps of the Reaction of 1, 2, and 9 with OOH- and OH- a
gaseous phase
aqueous phase
step no. (Scheme 1)
VI
compd no.
Δ
a,298H°
Δ
a,298G°
298k° (298τ°99
)
Δ
a,298G°
1
2
9
1
2
9
11.8
11.7
8.9
10.3
10.0
11.0
28.4
11.7
11.4
8.4
7.8
7.9
4.4 ꢀ 104 (1.0 ꢀ 10-4
)
)
)
)
7.9
7.9
2.9
14.6
12.7
17.1
25.2
2.9 ꢀ 104 (1.6 ꢀ 10-4
3.8 ꢀ 106 (1.2 ꢀ 10-6
1.3 ꢀ 107 (3.7 ꢀ 10-7
1.1 ꢀ 107 (4.4 ꢀ 10-7
)
IX
)
7.8
30.2
1.2 ꢀ 107 (3.8 ꢀ 10-7
XII
4.2 ꢀ 10-10 (1.1 ꢀ 1010
)
aΔa,298H° and Δa,298G° (both in kcal/mol), respectively, represent the enthalpy and Gibbs’ free energy (gaseous phase) or free energy (aqueous phase)
of activation at temperature 298.15 K and standard pressure. 298k° (in s-1) and 298τ°99 (in s) (eqs 5 and 6), respectively, denote the rate constant and the
time after which the reaction is 99% complete.
J. Org. Chem. Vol. 76, No. 4, 2011 1079

ꢀ
Krzyminski et al.
JOCArticle
molecular transformations taking place along the reaction
pathway. The Cartesian coordinates, total energies, and the
latter quantity plus the zero point energy, thermal energy/
enthalpy, and entropy at 298.15 K and standard pressure, for
all the entities appearing in Scheme 1 and the text, and
corresponding to stationary (lowest energy and transition
state) structures, are given in the Supporting Information.
Products of Processes Leading to Chemiluminescence and
Side Processes. Attempts were made to identify some stable
and intermediate products of the processes depicted in
Scheme 1. Stable products, i.e., 10-methyl-9-acridinone,
9-carboxy-10-methylacridinium acid, and substituted phe-
nols, were found in the acidified reactant mixture after
triggering reactions leading to chemiluminescence in 1, 2,
and 9. These entities were identified using HPLC with UV
absorption detection by adding relevant standards to the
reactant mixtures and comparing the retention times of
products and standards. The results of selected HPLC
analyses are given in Figure 8S and Table 2S (Supporting
Information), demonstrating some quantitative data for the
identified reaction products. The dominant product was 10-
methyl-9-acridinone, whose amount, relative to the initial
amount of MPCAþ, was 85-96 mol % in postreaction
mixtures in the cases of 1 and 2. The amount of 9-carboxy-
10-methylacridinium acid accounts for the depletion of 4-
15 mol % of MPCAþ. This product is most probably formed
as a result of the hydrolysis of MPCAþ (steps IV and XIV in
Scheme 1). Lastly, the amount of the corresponding phenols
accounts for roughly 100% of MPCAþ depleted in the case
of 1 and 2. Since the HPLC signals of 10-methyl-9-acridinone
and phenol in the case of 9 overlapped, it was not possible to
estimate the amounts of these products relative to the sub-
strate. The above data demonstrate that MPCAþ undergoes
hydrolysis to nonchemiluminescent 9-carboxy-10-methyla-
cridinium acid and that the extent of this process grows with
time under the analytical conditions applied. Further, 10-
methyl-9-acridinone appears in only somewhat smaller
amounts in relation to MPCAþ. Since chemiluminescence
quantum yields are of the order of a few percent, this product
must be formed mainly via nonchemiluminescent chemical
and physical pathways, among which is the transformation
of “pseudobase” (23) to nonexcited 10-methyl-9-acridinone
(steps II, VIII-XII in Scheme 1). Lastly, since phenols and
depleted MPCAþ were found in roughly equimolar amounts,
it seems that they are the products of the chemiluminescent and
nonchemiluminescent pathways of the conversion of MPCAþ
to 10-methyl-9-acridinone.
and -4.8 (R2 = CH3, R6 = H); -7.6, -16.0, and -5.0 (R2 =
CH3, R6 = CH3)). The method used is thus unable to show that
phenyl carbonate anions are released in the last step of the
reactions leading to the chemiluminescence of MPCAþ.
The lack of MPCAþ signals in HPLC analyses and the
balance of reactants indicate that all of the substrate is
converted to products. Since only a small portion of MPCAþ
is converted to light-emitting 10-methyl-9-acridinone, the
main part must take part in nonchemiluminescent pathways,
among which the one initiated by the attack of OH- on the
electrophilic center of MPCAþ at C(9) leading to the forma-
tion of “pseudobase” (23) (step II in Scheme 1) seems to
dominate. Two pathways are postulated to account for the
dark transformation of 23 to 10-methyl-9-acridinone: one
via steps VIII and IX and the other one via steps X, XI, and
XII. McCapra found that the dark transformation of
MPCAþ took place efficiently in the presence of oxidants
and postulated 27 as an intermediate.1a,5c We also noted
that in the absence of oxidants, the “pseudobase” could be
converted to MPCAþ, but not in their presence, i.e., under
the conditions of our chemiluminescence experiments. Steps
X, XI, and XII in Scheme 1 offer reasonable support for
these findings. On the other hand, the results of computa-
tions indicate that 23 may well be transformed to nonexcited
10-methyl-9-acridinone via steps VIII and IX, especially as
the latter is kinetically preferred to step XII, since it needs to
overcome a lower activation barrier. In this situation, both
the above pathways may account for the dark transforma-
tion of MPCAþ.
In the search for further evidence in support of the
assumed reaction scheme, we attempted to detect species
23 (“pseudobase”) by carrying out the spectrophotometric
acid-base titration of 1, 2, and 9. From the results obtained
(Figure 9S, Supporting Information), pK values reflecting
the equilibrium MPCAþ þ OH- T MPCAOH, equal
to -4.73, -4.74, and -4.62 for 1, 2, and 9, respectively, were
calculated. Spectrophotometric data reveal that cations of 1,
2, and 9 absorb around 260 and 360 nm in acidic media.
When the pH increases from 2 to 12 a new maximum appears
at 283 nm, which can be ascribed to the “pseudobase”. Two
isosbestic points characterize the titration profiles, one at
268 nm and the other at 335 nm. At pH = 11.6, at which the
chemiluminescence experiments were done, almost all
MPCAþ should be in the “pseudobase” form under equilib-
rium conditions. This means that upon alkalization of the
reaction mixture (final step of experiments), the instanta-
neously occurring OOH- and OH- compete for the electro-
philic center at C(9) of MPCAþ. This competition is one of
the crucial factors influencing chemiluminescence quantum
yields.
The identification of the corresponding phenols does not
solve the question whether they are released upon conversion
of MPCAþ to 10-methyl-9-acridinone or 9-carboxy-10-methy-
lacridinium acid, or during the decomposition of the phenyl
carbonate anions postulated in the chemiluminescence mech-
anism (steps I, V, and VI in Scheme 1). The latter anions
are converted to the corresponding phenyl carbonates under
the acidic conditions in which the HPLC analyses were per-
formed and decompose readily to phenols and CO2 (calculated
Another pathway competing with chemiluminescence is
hydrolysis of MPCAþ, the extent of which for 1, 2, and 9 is
presented in Figure 5. Compound 9, exhibiting glow-type
chemiluminescence, is much more resistant to hydrolysis
than 1 and 2. Two stages of hydrolysis are seen in the case
of 1 and 2 (1 exhibits flash type chemiluminescence), the first,
lasting ca. 7 and 25 s, respectively, and the second proceeding
several times more slowly. As chemiluminescence in 1 and 2
lasts from 3 to 30 s, respectively, it must be influenced by
hydrolysis. The data obtained suggest that from several to a
dozen or so percent of MPCAþ are depleted in such a
process. Hydrolysis of MPCAþ is initiated by the attack of
Δ
Δ
r,298Ho (gaseous phase), Δr,298Go (gaseous phase) and
r,298Go (aqueous phase) are, respectively, equal to (in kcal/
mol) -7.1, -15.3, and -4.3 (R2 = H, R6 = H); -7.4, -15.6,
(30) Ciscato, L. F. M. L.; Bartoloni, F. H.; Weiss, D.; Beckert, R.;
Baader, W. J. J. Org. Chem. 2010, 75, 6574–6580.
1080 J. Org. Chem. Vol. 76, No. 4, 2011

ꢀ
Krzyminski et al.
JOCArticle
FIGURE 7. Chemiluminescence quantum yields (CQY) (Table 3)
against the cavity volume (the volume of the hydration layer) of the
respective phenyl carbonate anions (20 in Scheme 1). The numbers
indicate the entities listed in Chart 1; the line corresponds to the
equation: CQY = -0.0110 ꢀ (cavity volume) þ 3.80.
FIGURE 5. Extent of hydrolysis of 1, 2, and 9 (c = 8.3 ꢀ 10-4 M) at
pH = 12.0. The lines correspond to the following equations: (1)
ln(cx/c0) = 0.261(t) - 3.55 (first stage), ln(cx/c0) = 0.0108(t) - 1.57
(second stage); (2) ln(cx/c0) = 0.0545(t) - 3.18 (first stage), ln(cx/c0) =
0.0145(t) - 2.12 (second stage); and (9) ln(cx/c0) = 0.0046(t) -
4.41; cx and c0 denote the actual concentration of 9-carboxy-10-
methylacridinium acid and the initial concentration of MPCAþ,
respectively.
FIGURE 8. Usefulness (U) (Figure 6) against the cavity volume
(the volume of the hydration layer) of the respective phenyl carbo-
nate anions (20 in Scheme 1). Upper graph (empty squares), values
corresponding to 15 min of thermostatting; lower graph (filled
squares), values corresponding to 30 min of thermostatting. The
numbers indicate the compounds investigated (Chart 1); the curves
correspond to the equations: U = -0.00770 ꢀ (cavity volume)2 þ
3.15 ꢀ (cavity volume) - 240.9 (upper graph) and U = -0.00737 ꢀ
(cavity volume)2 þ 3.26 ꢀ (cavity volume) - 301.4 (lower graph).
relatively high and at the same time they do not lose their
chemiluminogenic ability in alkaline media. Compunds 5and 6,
which exhibit a reasonable CL efficiency and preserve their
chemiluminogenic ability in alkaline media quite well, are also
interesting. The CL efficiency of nonsubstituted MPCAþ (1) is
quite high, but the compound loses its chemiluminogenic ability
in alkaline media fairly quickly. Compounds 7 and 12, i.e.,
MPCAþ substituted with a bulky t-Bu group or three Me
groups in the benzene ring, cannot be recommended as useful
chemiluminogens since they have the lowest U values.
Chemiluminogenic Ability against Structural Features. At-
tempts were made to discover possible relationships between
chemiluminescent characteristics, chemiluminescence quan-
tum yields (Table 2) and usefulness (Figure 6), and factors
reflecting the structural or physicochemical features of
MPCAþ. After consideration of a number of dependencies
we noted a falling trend in the plot of the chemiluminescence
quantum yield versus the cavity volume (the volume of the
hydration layer) of the phenyl carbonate fragment (20 in
Scheme 1) removed during the formation of electroni-
cally excited 10-methyl-9-acridinone (step VI in Scheme 1)
(Figure 7). On the other hand, the values of the usefulness
after 15 and 30 min of thermostatting against cavity volume
FIGURE 6. Usefulness (U) of MPCAþ, the product of residual CL
intensity after 15 or 30 min of thermostatting (percent of initial light
sum, Figure 4) and relative CL efficiency (percent for pH = 12.0,
Table 2), normalized, in %, to the highest value.
OH- on the C(15) atom of the carboxy group (step IV in
Scheme 1). This process competes with steps I, II, and III
leading to chemiluminescence and to the dark transforma-
tion of the cation to nonemitting 10-methyl-9-acridinone.
It is thus possible that this competition has the main
influence on the formation 9-carboxy-10-methylacridinium
acid.
The Choice of Optimal Chemiluminescent Derivatives.
The usefulness of MPCAþ (U) was analyzed (Figure 6) by
comparing the products of their relative chemiluminescence
efficiency (measured at pH = 12.0) (Table 2) and residual CL
intensity after thermostating in alkaline media (pH = 8.0)
(Figure 4). At pH = 8.0, chemiluminescent labeling of macro-
molecules is usually carried out, whereas at about pH = 12
chemiluminescence is initiated.1,5 The most promising chemi-
luminescent indicators or fragments of chemiluminescent labels
are 2, 8, and 9, i.e., MPCAþ with one or two Me groups in the
ortho position of the benzene ring, since their CL efficiency is
J. Org. Chem. Vol. 76, No. 4, 2011 1081

ꢀ
Krzyminski et al.
JOCArticle
can be roughly approximated with the polynomial type
dependencies shown in Figure 8. These relationships can
tentatively be used to design MPCAþ in the context of
their utility as chemiluminogens. They indicate that optimal
the phenyl carbonate fragment removed upon oxidation can
be roughly fitted with polynomial relationships, which can
be tentatively used for designing MPCAþ in the context of
their ability to chemiluminesce.
values of the volume of the hydration layer of the leaving group
in the case of an MPCAþ range between 200 and 260 A .
˚
This paper combines various aspects of the chemilumines-
cence of 10-methyl-9-(phenoxycarbonyl)acridinium salts,
which can be used as indicators of fragments of chemilumi-
nescence labels. The finding that alkyl substituents in the
phenyl fragment affect the chemiluminogenic and physico-
chemical features of this group of compounds has prompted
the continuation of investigations into the influence of
substituents with different electronic properties on these
features. It is also interesting to know whether and how
substituents at C(9) other than phenoxycarbonyl impart a
chemiluminogenic ability to the relevant acridinium cations.
These are problems that we are currently working on.
3
Conclusions
The synthesis of 10-methyl-9-(phenoxycarbonyl)acridi-
nium trifluoromethanesulfonates substituted with one or
more alkyl groups at the benzene ring provided a unique
opportunity for studying the influence of the structure of
these entities on their physicochemical and chemilumino-
genic features.
The basic chemiluminescent characteristics of the com-
pounds were determined and analyzed from the point of view
of their chemiluminogenic ability and susceptibility to parti-
cipation in side reactions in alkaline media.
Experimental Section
Syntheses, Purification, and Analyses. Phenyl acridine-9-
carboxylates (PAC) were synthesized using a modification of the
method described by Sato,10a which enabled compounds to be
obtained with a relatively high yield.17-21 Dehydrated acridine-
9-carboxylic acid (2.5 g) was converted into 9-chlorocarbonyl
acridine hydrochloride by refluxing it with a 5-fold molar excess
of thionyl chloride (yield 92%). Then, 0.15-0.2 g portions of the
synthesized compound were stirred with a stoichiometric quan-
tity of phenol or appropriate alkylphenol (Chart 1) in dichloro-
methane (5 mL) in the presence of N,N-diethylethanamine
(0.4 mL) and catalytic amounts of N,N-dimethyl-4-pyridina-
mine (room temperature, 15-25 h, anhydrous conditions). The
crude PAC (yield 60-90%) were washed with HCl, NaHCO3,
and NaCl solutions and purified by gravitational chromatogra-
phy using a SiO2 column (Silica Gel 60, Merck) and cyclohex-
ane/ethyl acetate as the mobile phase (3/2 v/v). The purity of the
PAC was checked by thin-layer chromatography (fluorescence
detection) and their identity confirmed by elemental analysis
(Table 1S, Supporting Information). Complete NMR data for
all precursors of the compounds investigated are reported in ref
19, and thermoanalytic data including their melting points are
listed in ref 20. The structures of the precursors of compounds 2,
5, and 8 were determined by the X-ray method.21a,b
An effective methylating agent, namely, methyl trifluoro-
methanesulfonate, was used to quaternarize the PAC.31 The
PAC (120 mg) were thus stirred with a 5-fold molar excess
of methyl trifluoromethanesulfonate in dry dichloromethane
(3.5 mL) (Ar atmosphere, room temperature, 3-4 h). Crude 10-
methyl-9-(phenoxycarbonyl)acridinium trifluoromethanesulfo-
nates were dissolved in a small amount of ethanol, filtered, and
precipitated with a 25 v/v excess of diethyl ether (yield 85-92%).
The purity of the compounds (1-12) was checked by HPLC (the
relative areas under the main signal were >99% in all cases)
and their identity confirmed by mass spectrometry (Table 1S,
Supporting Information). Complete NMR data for all the
compounds investigated are reported in ref 19, and their melting
points are provided in ref 20. The structures of most of the salts
(1, 2, 4-7, 9, and 12) were determined by X-ray methods.21c-i,22
Spectrophotometric Titrations. Selected compounds (1, 2, and
9) were dissolved in 10-3 M HCl to obtain a final concentra-
tion of MPCAþ of 4 ꢀ 10-4 M. Aliquots of the latter solution
(1.9 mL) were titrated at ambient temperature against sodium
hydroxide solution (9.0 ꢀ 10-3 M) in steps of 0.004 mL to a final
pH of 12.0, using a calibrated 0.5 mL Hamilton syringe and
combined pH microelectrode calibrated prior to measurements
The suggested and computationally supported mechanism
of chemiluminescence generation accounts for the experi-
mental observations and reveals the pathways by which
MPCAþ are converted to light-emitting species and the side
reactions during which they are converted to nonchemilu-
minescing entities. This provides a convenient framework
within which to consider the chemiluminescence of acridi-
nium cations.
Chemiluminescence quantum yields are proportional to
the numbers of 10-methyl-9-(phenoxycarbonyl)acridinium
cations over a wide range of concentrations (from 10-10
to 10-4 M). Detection limits are at the 10-10 M level. This
information forms the basis for the analytical application of
the compounds investigated.
Cations containing one or two Me groups in the ortho
position of the benzene ring appeared to be the most
promising chemiluminescent indicators or fragments of che-
miluminescent labels. Meta- and para-substituted deriva-
tives exhibit relatively high chemiluminescence quantum
yields but are poorly resistant to alkaline media, which
makes them less attractive than ortho-substituted deriva-
tives. If a bulky t-Bu group or three Me groups are present
in the benzene ring, the relevant cations are only weakly
chemiluminescent.
The chemiluminescence quantum yields of MPCAþ eval-
uated using luminol as a standard range from 0.6 ꢀ 10-2 to
2.2 ꢀ 10-2 einstein/mol; the latter value is higher than the one
for luminol. Moreover, the chemiluminescence of MPCAþ is
initiated without a catalyst, but the latter is needed to trigger
the chemiluminescence of luminol. In the case of MPCAþ a
better signal-to-noise ratio is also obtained. This indicates
that MPCAþ appear to be a suitable chemiluminogens.
Chemiluminescence quantum yields decrease with the
cavity volume (the volume of the hydration layer) of the
phenyl carbonate fragment, removed during the formation
of light emitting 10-methyl-9-acridinone molecules. This
dependency can be tentatively used for selecting MPCAþ
with respect to their chemiluminescence output.
The usefulness parameter, simultaneously reflecting the
chemiluminogenic ability and resistance of the compounds
to hydrolysis, was introduced in response to the analytical
applications of MPCAþ. The dependencies of this quantity
on the cavity volume (the volume of the hydration layer) of
(31) Alder, R. W. Chem. Ind. 1973, 983–986.
1082 J. Org. Chem. Vol. 76, No. 4, 2011

ꢀ
Krzyminski et al.
JOCArticle
against at least five buffers. After the addition of gradually
increasing amounts of base, the UV-vis spectra from 250 to 500
nm were recorded. By plotting absorbance at nine selected
wavelengths against pH, sigmoid-type curves with one inflec-
tion point were obtained, from which the pK of the reaction
MPCAþ þ OH- f MPCΑΟH (corresponding to the formation
of 23 (Scheme 1)) were determined using the Hender-
son-Hasselbach algorithm32 implemented in the Spectral Data
Lab software.33
Fluorometric Measurements. Chemiluminescence spectra in
the 400-600 nm range at room temperature were recorded after
mixing 1 mL of a 4 ꢀ 10-5 M solution of MPCAþ (1, 2, or 9) in
0.001 M HCl with 0.5 mL of a 0.06% solution of H2O2 in 0.1 M
HNO3 in a spectrofluorimeter cuvette and adding, with contin-
uous stirring, 0.5 mL of 0.2 M NaOH (all aqueous solutions). On
completion of the reaction leading to chemiluminescence, the
postreaction mixtures were left for 1 h, and fluorescence spectra
were recorded at excitation wavelengths 254, 260, and 285 nm. For
comparison, fluorescence spectra of 10-methyl-9-acridinone
(2 ꢀ 10-5 M) were recorded under the conditions described above.
The fluorescence decay at 10-methyl-9-acridinone was re-
corded by a time-correlated single photon counting tech-
nique using Ludox as reference material.34 The decay curve
was fitted to the equation I(t) = ΣiRi exp(-t/τi), where I is the
number of counts in a given time (t), R is the scaling factor
(amplitude), and τ is the lifetime (the reciprocal of the unim-
olecular rate constant), using an iterative convolution procedure
based on the least-squares method.
Chemiluminescence Investigations. To record time-dependent
profiles of chemiluminescence, MPCAþ was dissolved in HP-
LC-grade acetonitrile to a concentration of 0.005 M (stock
solution). Portions of this solution were diluted with HCl
(0.001 M) to obtain a concentration of 4 ꢀ 10-8 M. Aliquots
(200 μL) of this solution (five samples for each compound) were
placed in the 96-well white polystyrene plate of the chemilumin-
ometer. Next, 100 μL portions of an H2O2 solution (0.06% w/w)
in 0.1 M HNO3 were added to each well, and the plate
was shaken for 10 s. Chemiluminescence was triggered by
adding 100 μL of NaOH solution (0.2 M) to each well (the final
concentration of MPCAþ was 2 ꢀ 10-8 M). The CL intensity-
time profiles, expressed in relative light units (RLU), were
measured at a photomultiplier voltage of 1000 mV over a period
of time corresponding to a 99% drop in its maximum value. The
influence of oxidant concentration on chemiluminescence out-
put was investigated by adding 100 mL portions of an H2O2
solution (concentration 0.01-0.90%) in 0.1 M HNO3. The
CL-time profiles were also measured at 500 mV, using initial
dependencies are represented by calibration graphs, which were
generated for selected compounds (1, 2, and 4).
To determine the absolute chemiluminescence efficiency, the
chemiluminometer was calibrated using luminol as luminogenic
standard (the spectral characteristics of luminol (range of emis-
sion 380-600 nm, λmax = 435 nm) and MPCAþ (range of
emission 400-600 nm, λmax = 455 nm) are comparable).28 In
the calibration experiments, the concentration of luminol in
0.1 M K2CO3 was 1.13 ꢀ 10-6 M, the concentration of H2O2 was
2 ꢀ 10-3 M, and the concentration of horseradish peroxidase
was 2 ꢀ 10-7 M (final pH = 11.6). Knowing the chemilumi-
nescence quantum yield (1.23 ꢀ 10-2 einstein/mol)28 and the
light sums of luminol (five replicate measurements at 99% com-
pletion of the reactions leading to light emission), the chemilu-
minescence quantum yields of MPCAþ (φCL) were derived from
the light sums obtained in the above manner using eq 1
SX cST
ꢀ
φCL
¼
ꢀ φST
ð1Þ
SST cX
where φST is the chemiluminescence quantum yield of luminol,
X is the area under the CL-time profile of MPCAþ, SST is the
S
area under the CL-time profile of luminol, cX is the concentra-
tion of MPCAþ, and cST is the concentration of luminol. After
collection of the light sums for the standard (five replicate
measurements, 99% completion of reaction, PMT voltage =
1000 mV), the light sums for MPCAþ were obtained. The
chemiluminescence intensities of 1-12, expressed as the total
number of photons emitted (NCL) from a known number of
molecules, were calculated using eq 2
SX cST
ꢀ
NCL
¼
ꢀ NCL,ST
ð2Þ
SST cX
where NCL is the total number of photons emitted by MPCAþ,
SX is the area under the CL-time profile of MPCAþ, SST is the
area under the CL-time profile for luminol, cX is the concen-
tration of MPCAþ, cST is the concentration of luminol, and NCL,ST
is the total number of photons emitted from 2.09
ꢀ
1014
molecules of luminol.28
To investigate the stability of MPCAþ in basic media, the
stock solutions of the compounds were diluted with buffer of pH
= 8.0 to give a final concentration of 4 ꢀ 10-5 M. The solutions
were thermostated at 298 K for 15 min, and 200 μL aliquots of
them (five samples of each compound) were transferred to the
96-well white polystyrene plate of the chemiluminometer. Sub-
sequently, 100 μL of 0.06% w/w H2O2 in 0.125 M HNO3 was
added to each well (to reach pH = 6.0), and the plate was shaken
for 10 s. To trigger chemiluminescence, 100 μL of NaOH (0.3 M)
was added to each well (final concentration of MPCAþ = 2 ꢀ
10-5 M, final pH = 11.6). The light sums were measured as
described above. The stability of MPCAþ, expressed as residual
CL intensity, was obtained from the following equation:
MPCAþ solutions in 0.001 M HCl of concentration 4 ꢀ 10-5
(the final concentration of MPCAþ was 2 ꢀ 10-5 M).
M
To measure the chemiluminescence output (calibration
curves), acetonitrile solutions of MPCAþ (0.005 M) were diluted
with 0.001 M HCl to obtain a series of solutions with concen-
trations from 4 ꢀ 10-8 to 4 ꢀ 10-5 M. Chemiluminescence was
initiated as described above. The areas under the CL-time
profiles were determined using chemiluminometer software
(five samples for each compound). These areas, expressed in
integral RLU, represent the relative light sums. These values are
linearly dependent on the concentration of MPCAþ.1c,5e These
stability ¼ 100% ꢀ light sumðt ¼ 15 min, t ¼ 30 minÞ=
light sumðt ¼ 0 minÞ
To measure the relative CL efficiency at various pH, 50 μL
aliquots of MPCAþ (4 ꢀ 10-5 M, five samples of each compound)
were distributed over the 96-well white polystyrene plate of the
chemiluminometer. Next, a 50 μL portion of 0.12% w/w H2O2 in
0.01 M HNO3 was added to each well, and the plate was shaken for
10 s. Chemiluminescence was triggered by adding 300 μLofeachof
the following buffers, pH = 8.0, 10.0, and 12.0, after which the
(32) (a) Albert, A.; Briggs, J. M. In The Determination of Ionisable
Constants; Chapman and Hall: London, 1971. (b) Ossowski, T.; Goulart,
M. O. F.; de Abreu, F. C.; Sant’Ana, A. E. G.; Miranda, P. R. B.; de Oliveira
Costa, C.; Liwo, A.; Falkowski, P.; Zarzeczanska, D. J. Braz. Chem. Soc.
2008, 19, 175–183.
(33) Doroshenko, A. O. Spectral Data Lab software, Institute of Chem-
istry at Kharkiv National University, Kharkiv, Ukraine, 2003.
(34) O’Connor, D. V.; Philips, D. Time-Correlated Single Photon Count-
ing; Academic Press: London, 1984.
(35) (a) Atkins’ Physical Chemistry, 7th ed.; Atkins, P. W., de Paula, J.,
Eds.; Oxford University Press: Oxford, 2006. (b) Hadd, A. G.; Seeber, A.;
Birks, J. W. J. Org. Chem. 2000, 65, 2675–2683.
J. Org. Chem. Vol. 76, No. 4, 2011 1083

ꢀ
Krzyminski et al.
JOCArticle
light sums were measured as described above. The relative CL
efficiency was calculated from the formula
density functional theory (DFT) level37 using gradient tech-
niques38 and the 6-31G** basis set.39 The calculations were
carried out with the B3LYP functional, in which Becke’s non-
local exchange40 and the Lee-Yang-Parr correlation func-
tionals41 were applied. After completion of each optimization,
the Hessian (second derivatives of the energy as a function of the
nuclear coordinates) was calculated to assess whether stationary
structures had been obtained.37,39b The harmonic vibrational
frequencies were then derived from the numerical values of these
second derivatives and used to obtain the Gibbs’ free energy
contributions at 298.15 K and standard pressure with the aid of
a built-in computational program of statistical thermodynamics
routines.42 The solvent effect was included in the single-point
DFT calculations utilizing the polarized continuum model (PCM)
(UAHF radii were used to obtain the molecular cavity).43 With
this approach, the cavity volumes (the volumes of hydration
layers) of the phenyl carbonate fragments (20 in Scheme 1) were
obtained. Calculations were carried out using the Gaussian 03
program package.44
relative CL efficiency ¼ 100%ꢀlight sum ðpH ¼ 8:0, 10:0, 12:0Þ=
light sum of 1 ðpH ¼ 12:0Þ
Kinetic Analysis of Time-Dependent Profiles of Chemilumines-
cence. The formalism of the kinetics of the two-step consecu-
tive chemical process was adopted to describe the CL-time
profiles.35 It was assumed in this approach that chemilumines-
cence is generated by the process
k1
k2
MPCAþ f intermediate f A
ꢁ
where A* stands for the electronically excited light emitting
species. Using the relevant kinetic equations35 and assuming
that the intensity of emission is proportional to the concentra-
tion of A*, one obtains the expression
k1
CL intensity ¼ C
½expð - k1tÞ - expð - k2tÞ
ð3Þ
The enthalpies (Δr,298H°) and Gibbs’ free energies (free
energies in the case of DFT(PCM)) (Δr,298G°) of the reactions
(r), as well as the enthalpies (Δa,298H°) and Gibbs’ free energies
(free energies in the case of DFT(PCM)) (Δa,298G°) of activation
(a) indicated in Scheme 1 were calculated by following the
basic rules of thermodynamics.35a The rate constants (298k°)
for the gaseous phase reactions were obtained by applying the
equation
k1 þ k2
in which C is a constant (expressed in RLU), t represents the
time from the triggering of chemiluminescence, and tmax denotes
the time corresponding to the maximum CL intensity:
1
k2
ð4Þ
tmax
¼
ln
k2 - k1 k1
RT
298k°
¼
exp½ - Δa,298G°=ðRTÞꢂ
ð5Þ
The values of C, k1, and k2 were calculated by fitting CL-time
profiles simultaneously to the two above-mentioned equations
using the nonlinear least-squares method implemented in the
Origin 7.5 software.36
Nh
resulting from transition state theory, and the reaction comple-
tion time (298τ99) from the formula
MPCAþ Conversion Products. To investigate the chemical
changes accompanying the oxidation of MPCAþ (1, 2 and 9),
20 μL aliquots of stock solutions were placed in Eppendorf vials
and mixed first with 50 μL of H2O2 solution (0.06% w/w) in
0.1 M HNO3 and then with 50 μL 0.2 M NaOH (the final
concentration of MPCAþ was ca. 8.3 ꢀ 10-4 M). On completion
of emission, the reactant mixtures were acidified with the
addition of a drop of a 1/1 v/v trifluoroacetic acid/acetonitrile
mixture and subjected to HPLC analysis (spectrophotometric
detection at 268 nm). The products, 10-methyl-9-acridinone, the
9-carboxy-10-methylacridinium cation and phenol (in the case
of 1), 2-methylphenyl (in the case of 2), or 2,6-dimethylphenol
(in the case of 9), were identified and assayed quantita-
tively using standard procedures. The chemical conversions of
MPCAþ (1, 2, or 9) in oxidant-free conditions were investigated
in a similar way. Thus, 10 μL aliquots of the stock solutions were
placed in Eppendorf vials and diluted with 100 μL 0.1 M HNO3.
The dark processes were initiated by the addition of 100 μL
0.2 M NaOH to the above mixtures (the final concentration of
MPCAþ was ca. 2.4 ꢀ 10-4 M). They were terminated at
selected time intervals after initiation of the reactions (2 s to 40
min) by the addition of 20 μL of the 1/1 trifluoroacetic acid/
acetonitrile mixture. The reaction mixtures obtained were sub-
jected to HPLC analysis in order to identify the products and
assay them quantitatively, as described above. The quantitative
data obtained were used to analyze the kinetics of the dark
processes occurring.
298τ ¼ ln 100=298k°
99
ð6Þ
where R, T, N, and h denote the gas constant, tempera-
ture (298.15 K), Avogadro number, and Planck’s constant,
respectively.4,16,35a
Acknowledgment. This study was financed by the State
Funds for Scientific Research through grant No. 4 T09A 123 23
(contract no. 0674/T09/2002/23) of the Polish State Committee
for Scientific Research (for the period 2002-2005) and grant
no. N204 123 32/3143 (contract no. 3143/H03/2007/32) of the
Polish Ministry of Research and Higher Education (for the
period 2007-2010), as well as EU Structural Funds in Poland
(contract no. ZPORR/2.22/II/2.6/ARP/U/2/05 to A.O.). The
purchase of the Fluoroskan Ascent FL fluoroluminometer was
possible as a result of financial support from the Foundation for
Polish Science - IMMUNO’99 program (contract no. 61/99).
We thank Dr M. Gole-biewski for recording the MS spectra.
DFT calculations were carried out on the computers of the Tri-
City Academic Network Computer Centre (TASK) in Gdansk
(Poland).
(39) (a) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213–
222. (b) Hehre, W. J.; Radom, L.; Schleyer, P. v. R.; Pople, J. A. Ab Initio
Molecular Orbital Theory; Wiley: New York, 1986.
(40) (a) Becke, A. D. Phys. Rev. A 1988, 38, 3098–3100. (b) Becke, A. D.
J. Chem. Phys. 1993, 98, 1372-1377, 5648–5652.
Quantum Chemistry Computations. Unconstrained geometry
optimizations of isolated molecules were carried out at the
(41) Lee, C; Yang, W; Parr, R. G. Phys. Rev. B 1988, 37, 785–789.
(42) Dewar, M. J. S.; Ford, G. P. J. Am. Chem. Soc. 1977, 99, 7822–7829.
(43) (a) Tomasi, J.; Persico, M. Chem. Rev. 1994, 94, 2027–2094.
(b) Barone, V.; Cossi, M.; Mennucci, B; Tomasi, J. J. Chem. Phys. 1997,
107, 3210–3221.
(36) Origin 7.5, Origin Lab Co., Northampton, MA, 1991-2004.
(37) Density Functional Methods in Chemistry; Labanowski, J. K., Andzelm,
J. W., Eds; Springer-Verlag: New York, 1991.
(38) Modern Electronic Structure Theory: Geometry Optimization on
Potential Energy Surfaces; Schlegel, H. B., Ed.; World Scientific Publishing:
Singapore, 1994.
(44) Frisch, M. J. et al. Gaussian 03, revision C.02; Gaussian, Inc.:
Wallingford, CT, 2004.
1084 J. Org. Chem. Vol. 76, No. 4, 2011

ꢀ
Krzyminski et al.
JOCArticle
Supporting Information Available: Elemental analyses of
1-12; MS data (molecular ions); retention factors (TLC);
retention times (HPLC); composition of post-CL reaction mix-
tures (HPLC); retention time vs partition coefficient relation-
ships; FAB mass spectrum of 1; stationary fluorescence spectra
of light-emitting mixtures (1, 2, and 9); stationary fluorescence
spectra of post-CL mixtures (1, 2, and 9) together with the
spectrum of 10-methyl-9-acridinone; fluorescence decay spec-
trum of 10-methyl-9-acridinone; chemiluminescence intensity
profiles (1) at various H2O2 concentrations; relationships
between CQY (1, 2, and 4) and H2O2 concentration; DFT-
optimized geometries of entities involved in the reaction
mechanism of MPCAþ (1, 2, and 9) with OOH- and OH- or
appearing in the text; chromatograms of post-CL reaction
mixtures (2); UV-vis spectra of spectrophotometric titrations
(1); Cartesian coordinates and values of thermodynamic char-
acteristics of all the molecules investigated. This material is
available free of charge via the Internet at http://pubs.acs.org.
J. Org. Chem. Vol. 76, No. 4, 2011 1085