Kinetic studies on 2,6-lutidine catalyzed peroxyoxalate chemiluminescence in organic and aqueous medium: Evidence for general base catalysis

Article abstract of DOI:10.1016/j.jphotochem.2019.111967

The peroxyoxalate reaction, base catalyzed perhydrolysis of activated aromatic oxalate esters in the presence of chemiluminescence activators, has widespread analytical and bioanalytical applications and is one of the most efficient chemiluminescence transformations known. We report here a kinetic study on this reaction using 2,6-lutidine as catalyst in organic (1,2-dimethoxyethane) and aqueous medium. In both media, experimental conditions can be designed which lead to reproducible results important for analytical applications. Observed rate constants (determined by observing the light emission intensity as well as absorbance variation due to phenol releases) show dependence on both the 2,6-lutidine and the hydrogen peroxide concentration, indicating their participation in the rate-limiting step of the transformation. The rate constants obtained from these kinetic studies proved to be at least one order of magnitude higher in water than in 1,2-dimethoxyethane as solvent. Kinetic experiments designed to distinguish between three different types of catalysis (nucleophilic, specific base and general base catalysis) clearly indicate that the role of 2,6-lutidine in this reaction is as general base catalyst in water as well as most likely in organic medium.

Full text of DOI:10.1016/j.jphotochem.2019.111967

JournalofPhotochemistry&PhotobiologyA:Chemistry382(2019)111967
Contents lists available at ScienceDirect
Journal of Photochemistry & Photobiology A: Chemistry
journal homepage: www.elsevier.com/locate/jphotochem
Kinetic studies on 2,6-lutidine catalyzed peroxyoxalate chemiluminescence
in organic and aqueous medium: Evidence for general base catalysis
Felipe A. Augusto, Fernando H. Bartoloni1, Maidileyvis C. Cabello, Ana Paula F. dos Santos,
Wilhelm J. Baader⁎
Departamento de Química Fundamental, Instituto de Química, Universidade de São Paulo, São Paulo, Brazil
A R T I C L E I N F O
A B S T R A C T
Keywords:
The peroxyoxalate reaction, base catalyzed perhydrolysis of activated aromatic oxalate esters in the presence of
chemiluminescence activators, has widespread analytical and bioanalytical applications and is one of the most
efficient chemiluminescence transformations known. We report here a kinetic study on this reaction using 2,6-
lutidine as catalyst in organic (1,2-dimethoxyethane) and aqueous medium. In both media, experimental con-
ditions can be designed which lead to reproducible results important for analytical applications. Observed rate
constants (determined by observing the light emission intensity as well as absorbance variation due to phenol
releases) show dependence on both the 2,6-lutidine and the hydrogen peroxide concentration, indicating their
participation in the rate-limiting step of the transformation. The rate constants obtained from these kinetic
studies proved to be at least one order of magnitude higher in water than in 1,2-dimethoxyethane as solvent.
Kinetic experiments designed to distinguish between three different types of catalysis (nucleophilic, specific base
and general base catalysis) clearly indicate that the role of 2,6-lutidine in this reaction is as general base catalyst
in water as well as most likely in organic medium.
Chemiluminescence
Peroxyoxalate
Catalysis
General base catalysis
1. Introduction
development of illumination devices like light sticks and other appli-
cations [12], but its mechanism has only been studied recently [13]. 4-
The peroxyoxalate system is a classical chemiluminescence reaction
that emits light with the highest recorded efficiency [1–3], thus, being
of value for many analytical applications [4]. It is the base-catalyzed
reaction of an aromatic oxalic ester with hydrogen peroxide in the
presence of a fluorophore substance generally called activator (ACT).
Light emission occurs when the ACT interacts with the in situ generated
high-energy intermediate (HEI), usually established as being 1,2-diox-
etanedione (Scheme 1) [3,5–8].
The most commonly used base catalyst in the peroxyoxalate reac-
tion is imidazole (IMI-H) [9]. Its catalytic effect on the peroxyoxalate
system is well established, acting as a nucleophilic and base catalyst in
different reaction steps (Scheme 1) [9,10]. Nonetheless, IMI-H may also
decompose the generated HEI in a bimolecular fashion, reducing the
chemiluminescence yields [11]; since this system possess an overall
high efficiency, light emission is still easily observable. Apart from IMI-
H, only few catalysts appear to work adequately in for the chemilu-
minescence reaction. Sodium salicylate has been used as base catalyst
in the early work from the Rauhut group, mainly directed to the
N,N-Dimethylaminopyridine and some other pyridine derivatives were
pointed as efficient nucleophilic catalysts, leading to an intense and fast
light emission; high intensity emissions were obtained on using a
mixture of 1,2,4-triazole and 1,2,2,6,6-pentamethylpiperidine as cata-
lysts, in a ‘base induced nucleophilic catalysis’ mechanism [14]. This
catalyst system was applied in capillary electrophoresis, allowing the
detection of low amounts of fluorophores [15].
Despite working well for analytical applications, nucleophilic cata-
lysis increases the number of intermediates and reaction steps, com-
plicating the rationalization of the underlying mechanism in such sys-
tems. The use of non-nucleophilic catalysts potentially simplifies the
transformation and allows the observation of usually faster reaction
steps, like the generation of the HEI.
Kinetic studies on the hydrolysis of phenyl acetates containing
electron-withdrawing groups catalyzed by pyridine derivatives showed
that using 2,4- and 2,6-lutidines (i.e., 2,4- and 2,6-dimethylpyridines)
leads to remarkably lower rate constants than other methylated pyr-
idine derivatives [16], which was explained by the steric hindrance of
⁎ Corresponding author. Av. Prof. Lineu Prestes, 748, Butantã, 05508-000, Brazil.
E-mail address: wjbaader@iq.usp.br (W.J. Baader).
1 Current address: Centro de Ciências Naturais e Humanas, Universidade Federal do ABC, Santo André, Brazil.
https://doi.org/10.1016/j.jphotochem.2019.111967
Received 22 March 2019; Received in revised form 2 July 2019; Accepted 7 July 2019
Availableonline12July2019
1010-6030/©2019ElsevierB.V.Allrightsreserved.

F.A. Augusto, et al.
JournalofPhotochemistry&PhotobiologyA:Chemistry382(2019)111967
Scheme 1. Simplified reaction sequence for the peroxyoxalate
reaction with nucleophilic and base catalysis by imidazole (IMI-
H). [3].
the methyl groups, preventing these compounds from acting as nu-
cleophilic catalysts. In another example of steric hindrance caused by
methyl groups, it has been shown that 2-methylpyridine is a hundred
fold less efficient than pyridine for nucleophilic catalysis in the hy-
drolysis of aryl acetates [17]. Finally, a series of 2,6-lutidines was
studied in aqueous acid buffers (Lut-H+) for the dissociation of metal
oligopeptide complexes, with the conclusion that the methyl groups
suppressed the general acid catalysis by the Lut-H+ species [18].
In the present work, we study the kinetics of the peroxyoxalate re-
action with 2,6-lutidine (Lut) as non-nucleophilic base catalyst, which
should allow the direct kinetic observation of the nucleophilic attack
from H2O2 to the oxalic ester. In the absence of reaction intermediates
from nucleophilic substitution by the base catalyst, the analysis of ac-
quired kinetic data is potentially simpler. In line with recent studies
from our group in different solvents [19,20–22], particularly water
[23,24], this system will be studied in both organic and aqueous media,
to compare the kinetics with that of previously studied reactions, and to
create a starting point for new analytical applications of this system.
Fig. 2. Dependence of the chemiluminescence quantum yields (ΦCL) on the
[Lut], in the presence of H2O2 (100 mmol L–1), in DME and water (Table S1).
measurements on monitoring the release of both 2,4,6-trichlorophenol
(TCP) residues from TCPO by measuring the absorbance increase at
296 nm (in DME) or 312 nm (in water), the maximum absorption wa-
velengths for TCP in DME or for its phenolate in water. The time pro-
files for absorption measurements consist in first order exponential
increase (Fig. S2), which, after fitting with a saturation equation, pro-
vide observed absorption rate constants (kAB).
2. Results
Kinetic chemiluminescence measurements with the peroxyoxalate
system were conducted in 1,2-dimethoxyethane (DME) or water as
solvents, using bis(2,4,6-trichlorophenyl) oxalate (TCPO) and 2,6-luti-
dine (Lut) as oxalic ester and catalyst, respectively. 2,5-
Diphenyloxazole (PPO) was used as ACT in DME and fluorescein (FLU)
in water. In DME, with TCPO (0.1 mmol L–1), Lut (50 mmol L–1), H2O2
(100 mmol L–1), and PPO (10 mmol L–1), smooth first order exponential
decay curves were obtained as emission time profiles (Fig. S1). These
concentrations were defined as the standard conditions for our system,
while the initial [TCPO] was kept at 0.1 mmol L–1 throughout the ex-
periments. Mathematical fitting of the emission time profile with a first
order exponential decay equation provided values for the observed
emission rate constants (kEM) and the initial emission intensity (I0).
Mean values and standard deviations were obtained for assays con-
ducted as three replicas.
The kAB show a linear increase with the [Lut] in the low lutidine
concentration range (< 200 mmol L−1) in both solvents, leading to
bimolecular rate constants one order of magnitude higher in water as
compared to DME (Fig. 3, Table S2). When the reaction kinetics are
performed in DME at high [Lut] (< 1 mol L−1), a saturation behavior is
observed for higher [Lut], with maximum rate constant values of up to
0.15
0.03 s−1, from both emission and absorption measurements
(Fig. 4).
It should still be mentioned here that the reaction of TCPO with Lut
in DME in the absence of hydrogen peroxide ([H2O2] =0.00 mol L−1
)
does not lead to a considerable liberation of 2,4,6-trichlorophenol
(TCP) during the time of the experiments in the presence of the per-
oxide (Fig. S3). The small increase in the absorption during the ex-
periment might be due to hydrolysis caused by minimum amounts of
water present in the solvent.
A linear dependence of kEM with [H2O2] is observed in both sol-
vents, with a [Lut] of 50 mmol L–1 for peroxide concentrations of up to
1.0 mol L−1. The bimolecular rate constant is more than one order of
The kEM show a linear dependence with the [Lut], up to 0.2 mol L–1
in both solvents (Fig. 1, Table S1). The chemiluminescence quantum
yields (ΦCL) of the reaction decrease with increasing [Lut] within all the
studied concentration range in both conditions (Fig. 2). Interestingly,
the quantum yields reported here are higher in water than in DME,
however, these cannot be directly compared, as different activators
have been used in the two solvents, PPO in DME and FLU in water.
The reaction kinetics can also be followed by absorption
Fig. 1. Dependence of the observed emission rate constants (kEM) with [Lut], in the presence of H2O2 (100 mmol L–1), in DME and water (Table S1). Linear fit using
kEM = a + b [Lut]: DME: a = 0.00 s–1 (fixed); b = 0.189
0.002 L mol–1 s–1; water: a = 0.07
0.01 s–1; b = 2.23
0.09 L mol–1 s–1
.
2

F.A. Augusto, et al.
JournalofPhotochemistry&PhotobiologyA:Chemistry382(2019)111967
Fig. 3. Dependence of the observed absorption rate con-
stants (kAB) with [Lut] in the low concentration range, in
the presence of H2O2 (100 mmol L–1), in DME and water
(Table S2). Linear fit using kAB = a + b [Lut]: DME:
a = 0.00 s–1
water: a = 0.131
(fixed);
b = 0.167
0.002 L mol–1 s–1
0.1 L mol–1 s–1
;
.
0.008 s–1; b = 2.5
magnitude higher in water than in DME (Fig. 5, Table S3). The ΦCL
show a slight increase with increasing peroxide concentrations in DME;
for a 40-fold concentration increase the quantum yields increase about
4 times (Fig. 6, left). In similar reaction conditions in water, the ΦCL
show a small and not systematic variation with the [H2O2] (Table S3);
however, on using a higher [Lut] the behavior of the quantum yields in
water showed to be similar to that in DME, increasing slightly with
increasing peroxide concentrations (Fig. 6, right), where a 50-fold in-
crease in the peroxide concentration leads to an increase of about 5
times in the quantum yields (Table S4). Also with the higher [Lut] the
rate constants show a linear dependence on the peroxide concentration
with an intercept of about 0.4 s–1, corresponding to the Lut catalyzed
hydrolysis of TCPO and a slope of around 3.5 L mol–1 s–1, corresponding
to the reaction of TCPO with H2O2, also catalyzed by Lut (Fig. S4).
Absorption experiments were performed only in DME and, fol-
lowing the previously observed trend, a linear dependence of the kAB
values with [H2O2] is observed, with a slope of 0.083 L mol–1 s–1, si-
(50 mmol L–1), in the presence of an activator, whose nature and con-
centration does not influence the reaction kinetics, although being es-
sential for the observation of light emission and determining its in-
tensity [25].
The discussion on the reaction mechanism is based mainly on ki-
netic evidences obtained in DME and water, using [Lut] and [H2O2] up
to 0.2 and 1.0 mol L–1, respectively. The observed rate constants for
kinetics acquired through emission (kEM) and absorption (kAB) experi-
ments show linear dependencies with both [Lut] and [H2O2], indicating
that both these reactants are involved in the rate-determining step. This
observation excludes preliminarily the occurrence of a nucleophilic
catalysis by Lut, as observed before in the imidazole catalyzed perox-
yoxalate reaction of TCPO [9,10]. Additionally, we show here that no
reaction occurs between TCPO and Lut in the absence of hydrogen
peroxide in anhydrous DME. Therefore, the reaction mechanism does
not involve nucleophilic catalysis by 2,6-lutidine, in agreement with
former reports in the literature on catalysis in acyl transfer reactions
[16–18].
The second mechanistic possibility involves specific base catalysis
(SBC) by Lut of the hydrogen peroxide attack to the oxalic ester, which
would happen through a fast pre-equilibrium between Lut and H2O2,
forming the highly nucleophilic hydroperoxide anion, HO2–. To test the
hypothesis of occurrence of a SBC by Lut, we calculated the [HO2–],
from the equilibrium constant (Keq), estimated from the pKa values for
Lut-H+ (6.72) [26] and H2O2 (11.6) [27] in water (Scheme 2, Tables S7
and S8).
milar to that observed in the emission experiments (0.095 L mol–1 s–1
(Fig. 7, Table S5).
)
In order to verify the occurrence of a general or specific base cat-
alysis by 2,6-lutidine, experiments were performed in aqueous media
where the pH value was kept constant (pH = 8.2), while the [Lut]
varied. In these conditions, the kEM values show a linear dependence on
[Lut], with an intercept of 0.014 s−1, corresponding to specific catalysis
at this pH value, and a slope of about 0.2 L mol-1 s-1, corresponding to
the contribution of general base catalysis by Lut (Fig. 8, Table S6).
The correlation of the observed rate constants with the HO2– con-
centration (varied by using different [Lut] at constant [H2O2], as well as
different [H2O2] at constant [Lut]) for emission and absorption ex-
periments in DME shows three distinct regions: initial, linear, and
plateau (Fig. 9). A non-linear increase of the observed rate constants
with increasing [HO2–] is observed for low concentrations ([HO2–] < 5
10–4 mol L–1). A clearly linear dependence of the observed rate con-
stants with [HO2–] is observed in the concentration range of 0.5 to 1.5
10–3 mol L–1, which could suggest the occurrence of SBC (Fig. 9 and
10). The bimolecular rate constant calculated from the slope of the
3. Discussion
The peroxyoxalate reaction with TCPO (0.1 mmol L–1, as limiting
reagent) leads to complex kinetic emission profiles when using low
concentrations of H2O2 (10 mmol L–1) and Lut (1.0 mmol L–1) (Fig. S5).
However, on using higher [Lut], above 50 mmol L–1, or [H2O2], above
100 mmol L–1, “normal” kinetic emission decay curves are obtained,
which can be fitted with monoexponential decay equations. The op-
timal standard conditions for obtaining reproducible emission decay
kinetics are TCPO (0.1 mmol L–1), H2O2 (100 mmol L–1) and Lut
linear correlation is kbim = 125
7 L mol–1 s–1, showing to be around
Fig. 4. Dependence of the observed rate constants with [Lut] at high concentrations, in the presence of H2O2 (100 mmol L–1), in emission (kEM) and absorption (kAB
experiments in DME (Tables S1 and S2). Linear fits are the same as in Figs. 1 and 3.
)
3

F.A. Augusto, et al.
JournalofPhotochemistry&PhotobiologyA:Chemistry382(2019)111967
Fig. 5. Dependence of the observed emission rate constants (kEM) with [H2O2], in experiments in DME and water (Table S3). Linear fit using kEM = a + b [H2O2]:
DME: a = 0.00 s–1 (fixed); b = 0.095 0.002 L mol–1 s–1; water: a = 0.08 0.05 s–1; b = 1.51 0.06 L mol–1 s–1
.
Fig. 6. Dependence of the chemiluminescence quantum yields (ΦCL) with [H2O2] in DME with [Lut] =50 mmol L–1 and in water with [Lut] =500 mmol L–1 (Tables
S3 and S4).
Fig. 8. Dependence of the observed emission rate constant (kEM) with [Lut], in
experiments in aqueous borate (20 mmol L−1) buffer at pH = 8.2 (Table S6).
Fig. 7. Dependence of the observed absorption rate constants (kAB
[H2O2], in experiments in DME (Table S5). Linear fit using kAB = a + b [H2O2]:
a = 0.00 s–1 (fixed); b = 0.0830 0.0008 L mol–1 s–1; excluding the value for
[H2O2] =0.85 mol L–1
) with
Linear fit using kEM
=
a
.
+
b
[Lut]: a = 0.0141
0.0006 s–1
;
b = 0.191
0.007 L mol–1 s–1
.
region, as the medium polarity, measured as ET(30) values, is sig-
nificantly lower for high molar fractions of Lut in DME (Table S10).
However, there is only a small medium polarity variation in the linear
region (0.5–1.5 10–3 mol L–1) as the empirical solvent parameter ET(30)
maintains essentially constant up to 10 mol% of Lut in DME and shows
only a small variation for 28 mol % (Table S10). Therefore, we believe
that solvent polarity effects are not important for the rate constants in
the initial and the linear region of the correlation, although these might
be operating in the plateau region (Fig. 9), where high molar fraction of
Lut lead to a significantly lower medium polarity [28].
three orders of magnitude larger than the ones obtained in the depen-
dence with [Lut] and [H2O2], illustrating the much higher nucleophi-
licity of the negative charged HO2–.
For high hydroperoxy anion concentrations ([HO2–] > 1.5 10–3 mol
L–1), the observed rate constants remain constant with increasing
[HO2–] (Fig. 9, plateau region). In a first instance this could indicate a
change in the rate limiting step, similar to the situation reported in the
literature for the imidazole catalyzed peroxyoxalate reaction with
TCPO for high peroxide concentrations; interestingly, the values of the
corresponding maximum rate constants are in the same order of mag-
nitude as the ones obtained here [10].
The dependence of the observed rate constants from the experi-
ments performed in water with the [HO2–] is similar to that discussed
above for experiments in DME, however, a much smaller Lut and H2O2
Solvent polarity effects might be of importance in the plateau
4

F.A. Augusto, et al.
JournalofPhotochemistry&PhotobiologyA:Chemistry382(2019)111967
Scheme 2. Mechanism for the peroxyoxalate reaction between TCPO and H2O2 with specific base catalysis (SBC) by Lut.
Fig. 9. Dependence of observed rate constants (kobs) in emission and absorption
experiments with [HO2–] (Table S7). Increase in [HO2–] is due to increasing
[Lut] at constant [H2O2] or increasing [H2O2] at constant [Lut]. Three distinct
regions are indicated as initial, linear, and plateau regions, for low, moderate,
and high [HO2–].
Fig. 10. Linear region in the variation of the observed rate constants with
[HO2–] in emission and absorption experiments, with increasing [Lut] or
[H2O2] (Table S7). Linear fit using kobs = a + b [HO2−]: a = –0.023
0.007
s–1, b = 125
7 L mol–1 s–1
.
3), termolecular rate constants can be calculated from the correlation of
the observed rate constants with the [Lut] and [H2O2] (Table 1).
The values for the termolecular rate constants (kter), calculated from
the [Lut] as well as [H2O2] dependences, are very similar for a given
solvent, regardless of the determination method (emission/absorption;
concentration range has been studied in water (Table S8). There is no a
significant variation in the rate constants kEM with the [HO2–] in the
low concentration range and an apparently linear correlation from 0.3
to 0.8 mmol L−1 (Figure S6). The plateau region is not observed here, as
the concentration range studied is much lower than in DME.
Lut/H2O2 variation), with average values of 1.8
0.2 and 26
4
The experimental facts outlined above are in agreement with the
occurrence of a SBC in the 2,6-lutidine catalyzed reaction of TCPO with
hydrogen peroxide, however they also do not exclude the occurrence of
a GBC, where a Lut molecule is assisting the nucleophilic H2O2 attack
on the carboxyl group of the oxalic ester, resulting in the formation of a
monoperoxalic acid derivative (Scheme 3) Therefore, a possible in-
volvement of general base catalysis has been investigated for the re-
action in aqueous medium, where the pH value can easily be kept
constant. The results obtained in aqueous borate buffer show that the
observed rate constants increase with increasing [Lut] with pH constant
at 8.2 (Fig. 8), clearly indicating the occurrence of general base cata-
lysis by Lut in these reaction conditions. Using the mechanistic scheme
of general base catalysis in the peroxyoxalate reaction, where both
proton transfer and cyclization occur in the rate-limiting step (Scheme
L2 mol–2 s–1 for DME and water, respectively. In water kter is one order
of magnitude higher than in DME, in agreement with the expected
development of charge separation in the transition state of the rate-
limiting step, which should be stabilized in the polar protic medium.
In the literature, the reaction of TCPO and H2O2 through GBC with
IMI-H in ethyl acetate as solvent has a termolecular rate constant of
kter = 160
50 L2 mol–2 s–1 [10]. Therefore, it appears that, in a si-
milar reaction step, Lut is two orders of magnitude less efficient than
IMI-H, even though their pKaH are very similar (IMI-H, pKaH = 6.95;
Lut, pKaH = 6.72) [26,29]. However, the mechanism of the main re-
action pathway in the imidazole catalyzed transformation occurs by
nucleophilic catalysis, whereas with Lut GBC is operating. For the so-
dium salicylate catalyzed peroxyoxalate reaction of TCPO, which
5

F.A. Augusto, et al.
JournalofPhotochemistry&PhotobiologyA:Chemistry382(2019)111967
Scheme 3. Mechanism for the peroxyoxalate reaction between TCPO and H2O2 in DME with general base catalysis by Lut.
4. Conclusion
Table 1
Bimolecular (kbim) and termolecular (kter) rate constant values obtained from
the dependence of kEM or kAB with [Lut] and [H2O2], for the peroxyoxalate
reaction in DME and water.
It is shown in this work that 2,6-lutidine (Lut) acts as a non-nu-
cleophilic base catalyst in the peroxyoxalate system in two solvents,
1,2-dimethoxyethane and water. Similar bimolecular rate constants
were obtained from the linear relationships between the observed rate
constants with [Lut] and [H2O2], evidence that both molecules parti-
cipate in the same rate-determining step. The initial experimental re-
sults are in agreement with the occurrence of specific or general base
catalysis by Lut of the hydrogen peroxide attack to the oxalic ester. The
occurrence of general base catalysis in aqueous medium is evidenced by
kinetic experiments at constant pH, as the rate constants proved to be
dependent on the lutidine concentration. Therefore, the predominant
mechanism in DME should also be GBC, as the lower polarity of this
reaction medium, as compared to water, should not favor charge se-
paration occurring in SBC. In general, the results show that the per-
oxyoxalate chemiluminescence can be conducted in organic and aqu-
eous medium, using 2,6-lutidine as general base catalyst, leading to
reproducible results adequate for potential analytical and bioanalytical
applications. The rate constants are considerably higher in aqueous
medium than in organic solvent.
Solvent Method
[Lut] ×
[H2O2] ×
kbim (L
kter (L2 mol–2
a
10–3 (mol
10–3 (mol L–1
)
mol–1 s–1
)
s–1
)
L–1
)
DME
Emission
1.25 – 200b 100
50
25 – 1,000c
1.89 10–1
9.5 10–2
1.67 10–1
8.30 10–2
2.23
1.89
1.90
1.67
1.66
22.3
30.2
25
Absorption 6.25 – 200d 100
50
25 – 850e
Water
Emission
13 – 213b
100
50
50 – 1,000c
100
1.51
2.5
Absorption 13 – 113d
akter = kbim/[X], where X = Lut or H2O2. Data from Figures b1, c5, d3, and e7.
appears to occur by base catalysis, termolecular rate constants in the
order of 103 L2 mol–2 s–1 have been determined in ethyl acetate and a
variety of experimental conditions [13], also indicating a higher cata-
lytic efficiency of salicylate as compared to Lut. Additionally, it has
been shown in the literature that some pyridine and imidazole deri-
vatives and also aliphatic amines can be more efficient than imidazole
to catalyze the peroxyoxalate reaction of TCPO in acetonitrile as solvent
[14,30,31].
5. Experimental section
The monoperoxalic acid derivative, formed by Lut GBC of the hy-
drogen peroxide attack to the oxalic ester, is supposed to undergo base
catalyzed intramolecular cyclization to a cyclic peroxidic intermediate,
in a fast reaction step not observed kinetically [2–4]. The exact struc-
ture of this high-energy intermediate (HEI) is still not yet elucidated
[1,4,32], although the most probable structure appears to be 1,2-di-
oxetanedione [5–8], This peroxidic carbon dioxide dimmer appears to
have redox and structural properties [33–36] which allow the occur-
rence of efficient chemiexcitation by its interaction with an appropriate
activator, in a process known as chemically initiated electron exchange
luminescence (CIEEL) [25,37], leading to the formation of the elec-
tronically excited state of the activator which is responsible for the light
emission characteristic for the chemiluminescence of the peroxyoxalate
system (Scheme 3).
The chemiluminescence emission yields (ΦCL) decrease with in-
creasing [Lut] (Fig. 2) as already observed with imidazole [9]. This is
explained in the literature as the HEI decomposition by the catalyst,
which is feasible in the conditions used here [11]. On the other hand,
with increasing [H2O2] the ΦCL increase slightly in both solvents. In
DME, for instance, a 40-fold increase in [H2O2] led only to an increase
of around three times in ΦCL. Whereas, with a 50-fold increase in the
peroxide concentration in water, the quantum yields increase by a
factor of five (Fig. 6). The quantum yield increase in water can easily be
understood by the fact that higher peroxide concentrations will shift the
reaction pathway from hydrolysis (which occurs without light emis-
sion) to perhydrolysis (where light emission can occur) [23,24].
Light emission kinetic experiments were performed using fluores-
cence quartz cuvettes in a Varian Cary Eclipse spectrofluorometer,
while absorption kinetic experiments used absorbance quartz cuvettes
in a Varian Cary UV/Vis spectrophotometer. For washing out 2,6-luti-
dine (Lut) the cuvettes were washed with ethanol, HCl 10%, water,
ethanol again, and acetone. All experimental data were measured in
three replicates.
A typical essay was made at (25.0
0.5) oC adding the solvent
(either 1,2-dimethoxyethane, DME, or water), H2O2, and Lut to an
appropriate quartz cuvette. For emission experiments 2,5-diphenylox-
azol (PPO, 10 mmol L–1, in DME) or fluorescein (FLU, 0.1 mmol L–1, in
water) were used as well. After starting the measurement, the reaction
was initiated by adding bis(2,4,6-trichlorophenyl) oxalate (TCPO,
0.1 mmol L–1). The borate buffer (20 mmol L−1), used in experiments
where the constant pH was maintained, was prepared from a 0.1 mol
L
−1 H3BO3 solution and a 0.1 mol L−1 Na2B4O7·10H2O; for each [Lut],
the pH of the final solution was adjusted to 8.2 by addition of the ap-
propriate amount of H3BO3 solution (0.1 mol L−1).
The emission kinetic experiments were acquired for at least three
decay half-lives, while the absorption kinetics were registered until a
constant absorbance value was obtained. Chemiluminescence quantum
yields (ΦCL) were obtained using the luminol standard for photo-
multiplier calibration [2].
Data was linearly fitted (r > 0.99 in all cases) with Origin 2018 (64-
bit), version b9.5.0.193.
Acknowledgements
The authors would like to thank Solvay Peróxidos do Brasil for a
6

F.A. Augusto, et al.
JournalofPhotochemistry&PhotobiologyA:Chemistry382(2019)111967
generous donation of hydrogen peroxide. The authors also would like to
thank funding by Capes (FAA, MCC), CNPq (FAA, MCC), FAPESP
(WJB2014/22136-4, FHB2012/13807-7).
[17] A.R. Butler, I.H. Robertson, Nucleophilic and general base catalysis by pyridine and
methylpyridines in the hydrolysis of aryl acetates, J. Chem. Soc. Perkin Trans. I 2
(1975) 660–663.
[18] M. Hauröder, M. Schütz, K.J. Wannowius, H. Elias, Kinetics of the acid dissociation
of oligopeptide complexes of copper(II) and nickel(II): a study on general-acid
catalysis with ‘noncoordinating’ 2,6-lutidine type buffers, Inorg. Chem. 28 (1989)
736–742.
Appendix A. Supplementary data
[19] M. Khalid, S.P. Souza Jr., L.F.M.L. Ciscato, F.H. Bartoloni, W.J. Baader, Solvent
viscosity influence on the chemiexcitation efficiency of inter and intramolecular
chemiluminescence systems, Photochem. Photobiol. Sci. 14 (2015) 1296–1305.
[20] M. Khalid, M.A. Oliveira, S.P. Souza Jr., L.F.M.L. Ciscato, F.H. Bartoloni,
W.J. Baader, Efficiency of intermolecular chemiluminescence systems lacks sig-
nificant solvent cavity effect in binary toluene/diphenylmethane mixtures, J.
Photochem. Photobiol. A: Chem. 312 (2015) 81–87.
[21] S.P. Souza Jr., M. Khalid, F.A. Augusto, W.J. Baader, Peroxyoxalate chemilumi-
nescence efficiency in polar medium is moderately enhanced by solvent viscosity, J.
Photochem. Photobiol. A: Chem. 321 (2016) 143–150.
[22] M. Khalid, S.P. Souza Jr., F.H. Bartoloni, F.A. Augusto, W.J. Baader,
Chemiexcitation efficiency of intermolecular electron‐transfer catalyzed peroxide
decomposition shows low sensitivity to solvent‐cavity effects, Photochem.
Photobiol. 92 (2016) 537–545.
[23] G.A. Souza, M.M.M. Peixoto, A.P.F. Santos, W.J. Baader, General acid and base
catalysis by phosphate in peroxyoxalate chemiluminescence, ChemistrySelect 1
(2016) 2307–2315.
[24] M.C. Cabello, O.A.A. El Seoud, W.J. Baader, Effect of ionic liquids on the kinetics
and quantum efficiency of peroxyoxalate chemiluminescence in aqueous media, J.
Photochem. Photobiol. A: Chem. 367 (2018) 471–478.
[25] C.V. Stevani, S.M. Silva, W.J. Baader, Studies on the mechanism of the excitation
step in Peroxyoxalate Chemiluminescence, Eur. J. Org. Chem. (2000) 4037–4046.
[26] K. Clarke, K. Rothwell, A kinetic study of the effect of substituents on the rate of
formation of alkylpyridinium halides in nitromethane solution, J. Chem. Soc.
(1960) 1885–1895.
[27] P. Ulanski, C. Von Sonntag, The OH radical-induced chain reactions of methanol
with hydrogen peroxide and with peoxodisulfate, J. Chem. Soc. Perkin Trans. I 2
(1999) 165–168.
[28] J.P. Cerón-Carrasco, D. Jacquemin, C. Laurence, A. Planchat, C. Reichardt,
K. Sraïdi, Solvent polarity scales: determination of new ET(30) values for 84 organic
solvents, J. Phys. Org. Chem. 27 (2014) 512–518.
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.jphotochem.2019.
111967.
References
[1] M. Vacher, I. Fdez. Galván, B.-W. Ding, S. Schramm, R. Berraud-Pache, P. Naumov,
N. Ferré, Y.-J. Liu, I. Navizet, D. Roca-Sanjuán, W.J. Baader, R. Lindh, Chemi- and
bioluminescence of cyclic peroxides, Chem. Rev. 118 (2018) 6927–6974.
[2] F.A. Augusto, G.A. Souza, S.P. Souza Junior, M. Khalid, W.J. Baader, Efficiency of
electron transfer initiated chemiluminescence, Photochem. Photobiol. 89 (2013)
1299–1317.
[3] W.J. Baader, C.V. Stevani, E.L. Bastos, Chemiluminescence of organic peroxides,
Patai’s Chemistry of Functional Groups, The Chemistry of Peroxides vol. 2, John
Wiley & Sons, Ltd., 2006, pp. 1211–1278.
[4] L.F.M.L. Ciscato, F.A. Augusto, D. Weiss, F.H. Bartoloni, S. Albrecht, H. Brandl,
T. Zimmermann, W.J. Baader, The chemiluminescent peroxyoxalate system: state of
the art almost 50 years from its discovery, ARKIVOC (2012) 391–430.
[5] C.V. Stevani, I.P.A. Campos, W.J. Baader, Synthesis and characterisation of an in-
termediate in the peroxyoxalate chemiluminescence: 4-chlorophenyl-O,O-hy-
drogen-monoperoxyoxalate, J. Chem. Soc. Perkin Trans. I 2 (1996) 1645–1648.
[6] C.V. Stevani, W.J. Baader, Kinetic studies on the chemiluminescent decomposition
of an isolated intermediate in the peroxyoxalate reaction, J. Phys. Org. Chem. 10
(1997) 593–599.
[7] C.V. Stevani, W.J. Baader, Preparation and characterization of 2,2,2-triphenyl-2λ5-
1,3,2-dioxastilbolane-4,5-dione as standard for an attempt to trap 1,2-
Dioxetanedione, a possible high-energy intermediate in peroxyoxalate chemilumi-
nescence, J. Chem. Res. (2002) 430–432.
[8] S.A. Tonkin, R. Bos, G. Dyson, K.F. Lim, R.A. Russell, S.P. Watson, C.M. Hindson,
N.W. Barnett, Studies on the mechanism of the peroxyoxalate chemiluminescence
reaction: part 2. Further identification of intermediates using 2D EXSY 13C nuclear
magnetic resonance spectroscopy, Anal. Chim. Acta 614 (2008) 173–181.
[9] C.V. Stevani, D.F. Lima, V.G. Toscano, W.J. Baader, A kinetic study on the perox-
yoxalate reaction: imidazole as nucleofilic catalyst, J. Chem. Soc. Perkin Trans. I 2
(1996) 989–995.
[29] T.C. Bruice, G.L. Schmir, Imidazole catalysis. II. The reaction of substituted imi-
dazoles with phenyl acetates in aqueous solution, J. Am. Chem. Soc. 80 (1958)
148–156.
[30] T. Jonsson, M. Entemborg, K. Irgum, Heterocyclic compounds as catalysts in the
peroxyoxalate chemiluminescence reaction of bis(2,4,6-trichlorophenyl)oxalate,
Anal. Chim. Acta 361 (1998) 205–215.
[31] T. Jonsson, K. Irgum, Very fast peroxyoxalate chemiluminescence, Anal. Chim. Acta
[10] S.M. Silva, F. Casallanovo, K.H. Oyamaguchi, L.F.M.L. Ciscato, C.V. Stevani,
W.J. Baader, Kinetic studies on the peroxyoxalate chemiluminescence reaction:
determination of the cyclization rate constant, Luminescence 17 (2002) 313–320.
[11] A. Boaro, F.H. Bartoloni, Peroxyoxalate high-energy intermediate is efficiently de-
composed by the catalyst imidazole, Photochem. Photobiol. 92 (2016) 546–551.
[12] M.M. Rauhut, Chemiluminescence from concerted peroxide decomposition reac-
tions, Acc. Chem. Res. 2 (1969) 80–87.
[13] G.A. Souza, A.P. Lang, W.J. Baader, Mechanistic studies on the peroxyoxalate
chemiluminescence using sodium salicylate as base catalyst, Photochem. Photobiol.
93 (2017) 1423–1429.
[14] T. Jonsson, K. Irgum, New nucleophilic catalysts for bright and fast peroxyoxalate
chemiluminescence, Anal. Chem. 72 (2000) 1373–1380.
[15] C. Kuyper, K. Denham, J. Dickson, J. Murray, R. Milofsky, Chemiluminescence
detection in capillary electrophoresis using an ultra-fast co-catalyzed peroxyoxalate
chemiluminescent reaction and electrokinetic reagent delivery, Chromatographia
53 (2001) 173–178.
400 (1999) 257–264.
[32] F.H. Bartoloni, L.F.M.L. Ciscato, M.M.M. Peixoto, A.P. Fileno dos Santos,
C.S.S. Santos, S. Oliveira, F.A. Augusto, A.P.E. Pagano, W.J. Baader, Light: a rare
reaction product, Química Nova 34 (2011) 544–554.
[33] E.L. Bastos, P. Farahani, E.J.H. Bechara, W.J. Baader, Four‐membered cyclic per-
oxides: carriers of chemical energy, J. Phys. Org. Chem. 30 (2017) 1–18 e3725.
[34] P. Farahani, W.J. Baader, Unimolecular decomposition mechanism of 1,2-
Dioxetanedione: concerted or biradical? That is the question!, J. Phys. Chem. A 121
(2017) 1189–1194.
[35] F.A. Augusto, A. Francés-Monerris, I.F. Galván, D. Roca-Sanjuán, E.L. Bastos,
W.J. Baader, R. Lindh, Mechanism of activated chemiluminescence of cyclic per-
oxides: 1,2-dioxetane and 1,2-dioxetanone, Phys. Chem. Chem. Phys. 19 (2017)
3955–3962.
[36] F.H. Bartoloni, M.A. Oliveira, L.F.M.L. Ciscato, F.A. Augusto, E.L. Bastos,
W.J. Baader, Chemiluminescence efficiency of catalyzed 1,2-dioxetanone decom-
position determined by steric effects, J. Org. Chem. 80 (2015) 3745–3751.
[37] S.M. Silva, K. Wagner, D. Weiss, R. Beckert, W.J. Baader, Studies on the che-
miexcitation step in peroxyoxalate chemiluminescence using steroid-substituted
activators, Luminescence 17 (2002) 362–369.
[16] H. Neuvonen, Kinetics and mechanisms of reactions of pyridines and imidazoles
with phenyl acetates and trifluoroacetates in aqueous acetonitrile with low content
of water: nucleophilic and general base catalysis in ester hydrolysis, J. Chem. Soc.
Perkin Trans. I 2 (1987) 159–167.
7