Antioxidant activity of chalcones: The chemiluminescence determination of the reactivity and the quantum chemical calculation of the energies and structures of reagents and intermediates

Article abstract of DOI:10.1134/S0023158410040087

Six antioxidants from the class of chalcones (ArOH), compounds from which flavonoids are obtained in nature, were studied. The antiradical activity of chalcones and a number of related compounds was determined by a chemiluminescence method using the scavenging of peroxide radicals ROO ^? + ArOH → ROOH + OAr^? (with the rate constant k ₇) in a model reaction of diphenylmethane (RH) oxidation. The structures and energies of the reagents and intermediates were determined by semiempirical quantum chemical (PM3, PM6) calculations. 3,4-Dihydroxychalcone and caffeic acid, which have a catechol structure, that is, two neighboring OH groups in phenyl ring B, exhibited high antioxidant activity (k ₇ ≈ 10⁷ l mol^-1 s^-1); this is consistent with the lowest bond strengths D(ArO-H) of 79.2 and 76.6 kcal/mol, respectively. The abstraction of a hydrogen atom by the ROO^? radical is the main reaction path of these compounds; however, the low stoichiometric coefficients of inhibition (f = 0.3-0.7) suggest a contribution of secondary and/or side reactions of ArOH and OAr^?. In the other chalcones, the ArO-H bond is stronger (D(ArO-H) = 83-88 kcal/mol) and the antioxidant activity is lower (k₇ = 10⁴-10⁵ l mol^-1 s ^-1).

Full text of DOI:10.1134/S0023158410040087

ISSN 0023ꢀ1584, Kinetics and Catalysis, 2010, Vol. 51, No. 4, pp. 507–515. © Pleiades Publishing, Ltd., 2010.
Original Russian Text © R.F. Vasil’ev, V.D. Kancheva, G.F. Fedorova, D.I. Batovska, A.V. Trofimov, 2010, published in Kinetika i Kataliz, 2010, Vol. 51, No. 4, pp. 533–541.
Antioxidant Activity of Chalcones: The Chemiluminescence
Determination of the Reactivity and the Quantum Chemical
Calculation of the Energies and Structures
of Reagents and Intermediates
a
b
a
b
a
R. F. Vasil’ev , V. D. Kancheva , G. F. Fedorova , D. I. Batovska , and A. V. Trofimov
a
Emanuel Institute of Biochemical Physics, Russian Academy of Sciences, Moscow, 119991 Russia
b
Institute of Organic Chemistry with Centre of Phytochemistry, Bulgarian Academy of Sciences, Sofia, Bulgaria
eꢀmail: vasilev@sky.chph.ras.ru
Received August 26, 2009
Abstract—Six antioxidants from the class of chalcones (ArOH), compounds from which flavonoids are
obtained in nature, were studied. The antiradical activity of chalcones and a number of related compounds was
•
determined by a chemiluminescence method using the scavenging of peroxide radicals ROO + ArOH
→
•
ROOH + OAr (with the rate constant k7) in a model reaction of diphenylmethane (RH) oxidation. The
structures and energies of the reagents and intermediates were determined by semiempirical quantum chemꢀ
ical (PM3, PM6) calculations. 3,4ꢀDihydroxychalcone and caffeic acid, which have a catechol structure, that
7
–1 –1
is, two neighboring OH groups in phenyl ring B, exhibited high antioxidant activity (k7
is consistent with the lowest bond strengths (ArO–H) of 79.2 and 76.6 kcal/mol, respectively. The abstracꢀ
tion of a hydrogen atom by the ROO radical is the main reaction path of these compounds; however, the low
stoichiometric coefficients of inhibition ( = 0.3–0.7) suggest a contribution of secondary and/or side reacꢀ
tions of ArOH and OAr . In the other chalcones, the ArO–H bond is stronger ( (ArO–H) = 83–
≈ 10 l mol s ); this
D
•
f
•
D
4
5
–1 –1
8
8 kcal/mol) and the antioxidant activity is lower (k7 = 10 –10 l mol s ).
DOI: 10.1134/S0023158410040087
INTRODUCTION
either in a homogeneous anhydrous medium [7, 8] or
in micellar systems and in the presence of water [9].
These properties and the nontoxicity of chalcones
allow us to hope that they can be used as food stabilizꢀ
The oxidation of various organic materials (from
the simplest individual substances to macromolecular
compounds and complex biological systems in vivo) is
often a detrimental, undesirable process. To protect ers. However, systematic data on the antioxidant activꢀ
the materials from oxidation, an antioxidant inhibitor ity of these compounds are absent from the literature.
is added to trap peroxide radicals—oxidation chain
In this work, we studied six recently synthesized
chalcones (Fig. 1) with different substituents in a ring
min E and other tocopherols in plant oil)—were usually referred to as ring B. The antioxidant activity
carriers. In naturally occurring compositions, protecꢀ
tive antioxidants—bioantioxidants (for example, vitaꢀ
detected; the mechanism of formation of these bioanꢀ of chalcones was determined by trapping peroxide
•
tioxidants was developed in the course of evolution
1–3].
Substituted phenols (ArOH) [4], in particular,
chalcones—natural phenolic compounds, which are
considered the biological precursors of flavonoids [5,
radicals ROO —chain carriers in the model oxidation
reaction of a hydrocarbon (RH) initiated by the therꢀ
mal decomposition of an initiator (Y) (see the
scheme). The degree of quenching of chemiluminesꢀ
cence (CL), which was excited at the step of quadratic
chain termination (VI) (e.g., see [10]), by antioxidants
ArOH served as a measure of the antioxidant activity
[
6
], are practically important antioxidants. All of the
wellꢀknown chalcones possess biological and antioxiꢀ
dant activities, which manifest themselves to various
extents depending on the conditions of oxidation of ArOH:
507

508
VASIL’EV et al.
•
2
fcROO•
Y(+RH, O2)
→
2
f
c Y
r OO
→
(0)
(I)
(
rate of initiation Wi = 2fckY[Y]),
•
ROO•
rate constant k1),
R + O2
→
(
oxygen consumption
,
•
•
ROO + RH
(
→
R + ROOH
(
II)
hydroperoxide formation, k2),
•
•
ROO + ROO
ketone R =O + O + alcohol ROH + CL
→
(VI)
–
H
2
(quadratic chain termination, 2k6),
•
ROOH + OAr•
ROO + ArOH
→
(
VII)
(
linear termination on an inhibitor
,
k7),
k8),
•
•
→
ROO + OAr
products
(
VIII)
(
termination on an inhibitor radical
,
where rY is a primary radical, and fc is the probability of the escape of radical rY from a cage.
(The conventional notation of rate constants is used.)
Scheme.
The molecular structures of chalcones and interꢀ of AIBN (this fact suggested the presence of a contamꢀ
mediate products were determined using semiempiriꢀ inating impurity of a peroxide initiator), the hydrocarꢀ
cal quantum chemical calculations, and the enthalpies bon was additionally purified by passing through an
of dissociation of ArO–H bonds were calculated; alumina column. Chalcones are sparingly soluble in
these enthalpies were compared with activity characꢀ nonpolar solvents. Therefore, they were initially disꢀ
teristics. The following related phenols were also studꢀ solved in dimethyl sulfoxide; then, an aliquot of this
ied: caffeic acid (CA),
and chroman C1 (2,2,5,7,8ꢀpentamethylchromanꢀ6ꢀ concentration of dimethyl sulfoxide in the reaction
pꢀcoumaric acid, ferulic acid, solution was introduced into a reaction mixture. The
–5
–4
ol, CrC1), which is a synthetic analog of
α
ꢀtocopherol, vessel was insignificant (
3
×
10 –6 10 mol/l), and
×
as a model antioxidant.
the effect of polarity on the antioxidant activity could be
ignored. The role of reaction products was minimized
–10
–1 –1
using a low rate of initiation ( i
w
≈
10 mol l s ). This
EXPERIMENTAL
rate was measured from the kinetics of CL after adding
The chalcones were prepared by the Claisen– a known amount of a standard inhibitor—CrC (see
1
Schmidt reaction between acetophenone and a correꢀ Eq. (8) below;
f
= 2.0 and wi
=
2[CrC1]0/τ
0.5
for CrC1).
sponding arylaldehyde [11]. The average product yield
The oxidized mixture (5 ml) was placed in a therꢀ
after purification by recrystallization was 80%, and the mostated (50°C) cell of a chemiluminometer. The
structures were confirmed by UV, IR, NMR, and mass mixture was saturated with oxygen by bubbling air. The
1
spectra and elemental analysis. The H NMR spectra weak primary luminescence (the triplet–singlet emisꢀ
indicated that the synthesis products were only
mers.
Eꢀisoꢀ sion of the excited product—benzophenone) was
enhanced by energy transfer to an efficient luminoꢀ
3+
Caffeic acid from Merck and
ferulic acid from Fluka were used.
p
ꢀcoumaric acid and phor, the europium chelate Eu ꢀ1,10ꢀphenanthroꢀ
lineꢀtris(thenoyltrifluoroacetonate), which emits light
as a narrow band at 612 nm. The chemiluminometer
and the procedure of CL measurements were
described elsewhere [10].
Chroman C1 was kindly provided by Prof. V.A. Roꢀ
ginskii (Institute of Chemical Physics, Russian Acadꢀ
emy of Sciences).
The efficiency of an antioxidant was studied using
the inhibition of the liquidꢀphase oxidation of dipheꢀ
nylmethane (RH). Oxidation was initiated by the therꢀ
mal decomposition of azobisisobutyronitrile (Y =
RESULTS AND DISCUSSION
Energetics and Structure of Chalcones
AIBN); wi = 2f k [Y], where fc = 0.60, and the rate
Because the antioxidant activity of phenols is assoꢀ
c Y
15
constant kY = 1.58
×
10 exp(–30800/RT) is almost ciated with hydrogen atom abstraction from the pheꢀ
independent of the nature of the solvent [12]. AIBN, nolic OH group by a peroxide radical, it is reasonable
the chlorobenzene solvent, and diphenylmethane to relate the reactivity of a phenol antioxidant to the
were purified in accordance with standard procedures strength
D(ArO–H) of an OH bond, which is cleaved
[
10]. If diphenylmethane emitted CL with no addition in reaction (VII) of an inhibitor with a peroxide radical
KINETICS AND CATALYSIS Vol. 51
No. 4
2010

ANTIOXIDANT ACTIVITY OF CHALCONES: THE CHEMILUMINESCENCE
Chalcones
509
O
OH
O
O
OH
OH
2ꢀHydroxychalcone (Ch1)
3ꢀHydroxychalcone (Ch2)
4ꢀHydroxychalcone (Ch3)
O
OH
O
O
OCH3
OH
OH
OH
OCH3
2ꢀHydroxyꢀ3ꢀmethoxychalcone (Ch4) 3ꢀHydroxyꢀ4ꢀmethoxychalcone (Ch5) 3,4ꢀDihydroxychalcone (Ch6)
Cinnamic acid derivatives
HO
HO
H CO
HO
HO
OH
OH
OH
3
O
O
O
пꢀ
Coumaric acid (CoumA)
Ferulic acid (FA)
Caffeic acid (CA)
Model antioxidant
CH3
HO
CH3
CH3
H C
3
O
CH3
Chroman C (CrC1)
1
Fig. 1. Structures of the compounds examined.
(
see the scheme). By definition, the strength (enthalpy ten standard sets of basis functions, the values of
or heat) of a bond is the difference of the heats of forꢀ
D(ArO–H) gradually (but not monotonically)
0
mation (ΔH ) of bond cleavage products and the heat increased on going from a minimum (6ꢀ31G) to a
f
of formation of the reagent. In the case under considꢀ maximum (6ꢀ311 + G(2d,2p)) basis set. However,
eration,
they were found 3–5 kcal/mol lower than experimenꢀ
•
•
ArO–H = OAr + H ;
tal values. Only nonstandard sets of basis functions
from 6–31G(,p) to 6ꢀ31 + G(,3pd)) made it possible
consequently,
(
0
i
D(ArO−H) = ΔH (OAr )
f
0
f
i
0
f
+
ΔH (H ) − ΔH (ArOH).
Table 1. Dissociation energy of the ArO–H bond in phenol
and catechol calculated by ab initio and semiempirical methꢀ
ods
The phenoxyl radical (the product of hydrogen
abstraction from ArOH by the radical ROO ) is often
referred to as ArO ; however, it has a quinoid structure
and an unpaired electron is distributed over ring C
atoms opposite to the C=O group resulting from O–H
bond cleavage [13]. We reflect this structure peculiarꢀ 6ꢀ31+G(,3pd)
ity by denoting this radical as OAr rather than ArO
It is well known that the heat of formation of the CCSD/6ꢀ31+G*
hydrogen atom is ΔHf (H•) = 52.10 kcal/mol. Bakalꢀ
bassis with coauthors (see [14, 15] and references
therein) performed the most thorough ab initio
DFT/B3LYP) calculations of the energetics of pheꢀ
nol, its hydroxyꢀ and orthoꢀmethoxyꢀsubstituted Experiment
derivatives, and corresponding phenoxyl radicals. In
•
•
D(ArO–H), kcal/mol
Refeꢀ
rence
Calculation method
phenol
catechol
88.5
–
–
–
87.3
81.7
80.6
81.6
81.5
[15]
[16]
•
•
.
CCSD/6ꢀ31G*
0
CCSD/6ꢀ31++G**
PM3
80.9 This
work
PM6
–
87.0; 88.3; 88.7
81.0
81.6
80.5
(
[15]
[17]
KINETICS AND CATALYSIS Vol. 51
No. 4
2010

510
VASIL’EV et al.
The main chalcone units (rings and C–C bonds)
(a)
lie in a plane (Fig. 2); this fact is responsible for a high
degree of their conjugation. Phenoxyl radicals also
have a planar structure. They can occur in conformaꢀ
tions ~180° different in the dihedral angle of the rotaꢀ
tion of the O–H bond about the C–O bond and, corꢀ
respondingly, in the heats of formation (Fig. 3). Analꢀ
ogous results were obtained for all of the compounds
examined, for example, caffeic acid (Fig. 4).
(b)
In some ArOH conformations, hydrogen bonds are
formed. Indeed, for example, the bond orders in the
O–H–O chain are 0.93 (i.e., smaller than unity) and
0
.004 (i.e., somewhat greater than zero) in the case of
caffeic acid. The hydrogen bond was also retained in
the O–H–O chain of a phenoxyl radical (bond orders
of 0.920 and 0.009, respectively). As a result of an
additional stabilization (by ~2 kcal/mol), these conꢀ
formations became the participants of reaction (VII)
0
(
k7), and the calculated values of ΔHf of the comꢀ
pounds examined pertain to them (see Table 5 below).
Fig. 2. Structures of (a) chalcone Ch6 and (b) its phenoxyl
•
radical OAr
.
Two conformers with different rotation angles of
the O–H bond about the O–C axis are also possible
for chroman C1; however, they are only slightly differꢀ
ent (by 0.5 kcal/mol) in the heat of formation and,
correspondingly, O–H bond strength.
to obtain the values of D(ArO–H) in good agreement
with experimental data (Table 1). This example indiꢀ
cates that it is difficult to choose a reliable version of ab
initio calculations, especially, for novel compounds
whose energetics has not been studied previously. On
the other hand, Table 1 indicates that the semiempiriꢀ
cal PM3 and PM6 methods give an adequate accuracy
Model Reaction of Hydrocarbon Oxidation
and the Chemiluminescence Monitoring
of Antioxidant Action
At moderate temperatures, hydroperoxide ROOH
is the final product of the radical chain oxidation of
hydrocarbon RH and the reaction is an unbranched
chain reaction (see the scheme). In the exothermal act
(
with consideration for experimental error). In this
work, we used these methods to determine the strucꢀ
tures of chalcones and related compounds and the
ArO–H bond strength.
(
(
Δ
H
≈
100 kcal/mol) of quadratic chain termination
VI), R =O and O2 molecules become electronically
–H
excited and emitting light (CL) [10]. The main set of
reactions given in the scheme (without considering
Heat of formation, kcal/mol
−55.0
−55.5
−56.0
−56.5
−57.0
−57.5
Heat of formation, kcal/mol
−
−
−
138
139
140
141
−
−
58.0
58.5
−
−
142
−
ꢀ60
0
60
Dihedral angle of O–H bond rotation
about the C–O bond, deg
120 180 240 300 360
60
0
60 120 180 240 300 360 420
Dihedral angle of O–H bond rotation
about the C–O bond, deg
Fig. 3. Dependence of the heat of formation (ΔH f0 ) of Ch6
Fig. 4. Dependence of the heat of formation (ΔH f0 ) of CA
on the dihedral angle of O–H bond rotation about the C–
O bond (PM3 calculation).
on the dihedral angle of O–H bond rotation about the
C⎯O bond (PM3 calculation).
KINETICS AND CATALYSIS Vol. 51
No. 4
2010

ANTIOXIDANT ACTIVITY OF CHALCONES: THE CHEMILUMINESCENCE
511
possible additional and secondary processes) correꢀ
sponds to the following set of differential equations:
i
+
ArOH
1
.00
•
• 2
d[ROO ]/d
k7[ArOH][ROO ] – k8[ROO ][OAr ],
t = wi – 2k6[ROO ]
•
(
1)
•
•
–
•
•
•
•
d[OAr ]/d
d[ArOH]/d
Under steadyꢀstate conditions with respect to radꢀ
t
=
k7[ArOH][ROO ] – k8[ROO ][OAr ],(2)
0.75
•
2
3
1
t
= – 7
k
[ROO ][ArOH].
(3)
0
.50
.25
0
•
•
τ
icals (d[ROO ]/d
t
= d[OAr ]/d
t
= 0), Eq. (1) is transꢀ
0.5
formed into a balance equation between chain initiaꢀ
tion and two types of chain termination, quadratic and
0
•
linear (with respect to the concentration of ROO radꢀ
T
icals):
•
2
wi = 2k6
[ROO ] + 2k7[ArOH][ROO•].
(4)
A particular manifestation of the effect of ArOH
depends on the experimental procedure. If the conꢀ
sumption of O2 or the buildup of a hydroperoxide is
0
200 400 600 800 1000 1200 1400
, s
t
Fig. 5. Kinetics of CL on the oxidation of diphenylꢀ
methane (0.58 mol/l, 10 vol % in chlorobenzene;
measured, a decrease in the reaction rate, which is
–
4
–10
–1 –1
•
[AIBN] = 1.0
Eu chelate] = 2.0
tion of antioxidants: (
×
10 mol/l; wi = 3.3
×
10 mol l
10 mol/l; 50°C) after the introducꢀ
CrС1, ( ) Ch6, and ( ) CA.
10 mol/l. For curve , the kinetic
parameters are shown: the reciprocal slope and the
induction period
s ;
proportional to [ROO ], is observed:
w
= –d[O ]/d
t
=
2
–4
[
×
•
d[ROOH]/dt = k2[ROO ][RH]; now, it is a fraction
1
)
2
3
•
•
•
⎯ 7
[
ROO ]/[ROO ] =
r
of the values of w0 or [ROO ]
[ArOH]0
=
2.0
×
1
0
0
for uninhibited reaction. If CL is measured, the lumiꢀ
T
•
2
τ
0.5.
nescence intensity
ciency of CL), which is proportional to [ROO ]
decreases; its fraction I0 is denoted as
. From Eq. (4) at [ArOH] = 0,
I
=
Φ
CLk6[ROO ]
(
Φ
is the effiꢀ
CL
•
2
,
I/
i: i =
•
2
•
2
2
values of
reliability and information content of CL monitoring.
The area over the curve of , that is, the integral
i, r
, [ROO•], and [ArOH] provides the high
[
ROO ] /[ROO ] = r
0
•
1/2
•
it follows that [ROO ] = (wi/2k6) ; [ROO ] =
[ROO ] = r(wi/2k6) ; and Eq. (4) can be written in
0
•
1/2
i(t)
r
0
∞
the form
(
1 − i)dt, is the light sum of CL quenched by the
∫0
r2
= r2
k7[ArOH](2k6w )–1/2
i
1
or 1 –
or 2k7[ArOH] = (2k6w ) (i–
–
antioxidant and its radical in equal fractions. That is,
i
=
i1/2
2
k7[ArOH](2k6w )–1/2
(5)
i
1/2
each molecule of ArOH eliminates two active ROO•
1/2
i1/2).
–
i
radicals and the stoichiometric coefficient f is 2.0:
Equation (5) relates the current relative intensity of
CL and the current concentration of ArOH.
The quenching effect of ArOH depends on the
contribution of linear termination, that is, on the
product k7[ArOH] (see Eq. (4)). Initially, let us conꢀ
sider an ideal case of the model antioxidant chroman
∞
f
[ArOH] = wi (1 − i)dt = w
τ
i 0.5
.
(8)
0
∫
0
From these equations and experimental data,
[ArOH]0 or the rate wi (i.e., one of these quantities if
the other is known), the rate constants ratio
f
(
CrC1; Fig. 5, curve
1). The introduction of this antiꢀ
oxidant into the reaction mixture completely inhibited
1
/2
•
k7/(2k6
)
, and the constant k7 (if
2
k6 is known) are
the reaction; that is, k6[ROO ]
intensities and , as well as the values of
ROO ], dramatically decreased to zero. Then, as
Ӷ
k7[ArOH]. The CL
calculated. The use of the CL semirecovery time 0
place of the integral and Eq. (8) simplifies the proceꢀ
dure but somewhat overestimates the value of
τ
in
.5
i
I
r
and
•
[
ArOH, was consumed, the steadyꢀstate concentration
•
f
[ArOH]0 (by no more than 2% [10]).
of ROO and, correspondingly, the initial rate of oxiꢀ
dation or CL intensity were restored. The Sꢀshaped
kinetic curve of CL is described by the equation (see
In Fig. 5, it can be seen that the effects of 3,4ꢀdihyꢀ
droxychalcone (Ch6) and CA (curves
tively; cf. the shape and slope of a CL recovery region)
are comparable with the effect of CrC1 (curve ) at the
same introduced concentration [ArOH] . The maxiꢀ
mum slopes (d /d )max (i.e., 1/ ) are similar; conseꢀ
2 and 3, respecꢀ
[
10])
F(t i
) = ln(1 + 1/2) – ln(1 – i1/2) – i–1/2
1
(6)
k7/(2k6) )wi1/2 + const,
1/2
0
=
(
i
t
Т
and a maximum slope of the curve (at an inflection
i(t)
quently, the rate constants k7 are also similar. However,
the quenched light sum is much smaller, and this
means that a considerable fraction of the antioxidant
point at
i
= 0.535) is
1/2
1
/2
(di
/d
t
)max = 1/Т = 0.237( 7
k
/(2k6) )wi
.
(7)
The continuous recording of the value of
rent in arbitrary units), that is, an infinite number of the scavenging of peroxide radicals. Therefore, the
taken samples, for the determination of the current stoichiometric coefficient should be calculated from
i
(photocurꢀ and/or its radical is consumed without participation in
f
KINETICS AND CATALYSIS Vol. 51
No. 4
2010

512
VASIL’EV et al.
i
Extended Scheme of Antioxidant Action,
Participation of Oxygen
+
ArOH
1
0
0
0
.00
.75
.50
.25
0
To interpret the experimental data, there were
attempts to extend the kinetic scheme. For example,
Belyakov et al. [18] formally introduced (without disꢀ
cussing the chemistry and structure of reactants) reacꢀ
tion (XVII) of phenoxyl radicals, which is additional
to reaction (VIII):
1
2
3
OAr•
→
product “r” (+O2)
→
rOO•
(XVII)
(
in an excess of oxygen, k17 is the apparent rate conꢀ
stant of firstꢀorder reaction).
Loshadkin et al. [19] also formally introduced the
bimolecular prooxidant reaction
0
400
800
1200
1600
2000
, s
•
t
OAr + O (O , RH)
2
2
→
quinone
•
–H 2 2
(
like O=Ar =O) + ROO + H O .
Vasil’ev and Trofimov [20] discussed the addition of
O2 to a phenoxyl radical OAr , that is, in fact, the reacꢀ
Fig. 6. Kinetics of CL on the oxidation of diphenylꢀ
•
methane after the addition of various concentrations of
7
tion between two radical centers (triplet O2 can be
considered as a biradical). The PM3 calculation for
the case of ArOH = catechol demonstrated that, along
CA. [ArOH]0
×
10 , mol/l: (
1
) 0.4, (
2
) 2.0, and (
3) 8.0.
Experimental conditions are specified in Fig. 5.
•
with the expected peroxide radical OArOO , the Cꢀ
∞
centered radicals of 1,2ꢀdioxetane can be formed
(Fig. 8).
the initial formula
f = wi (1 − i)dt/[ArOH]0 (cf.
∫0
Eq. (8)); it is 0.7–0.3 (see below).
The reaction is weakly endothermic or weakly exoꢀ
thermic (Table 2); the activation energy was estimated
at a few kcal/mol using the PM3 method (saddle
energy, calculation with the keyword SADDLE).
Dioxetane is a bicyclic compound; that is, its structure
is more ordered than that of the starting pair OAr +
O2. Thus, the expected negative entropy change does
At high initial concentrations of ArOH, CL was
incompletely weakened (Figs. 6 and 7, curves 3). The
calculated rate constants also depend on [ArOH]0. In
addition, after the complete consumption of the antiꢀ
oxidant, the rate of initiation was found higher (by 10–
•
2
0%) than that at the onset of the experiment. Eviꢀ
not imply an equilibrium shift toward dioxetane, and
dently, all of the above special features are related to
processes that are not taken into account in the
scheme. This is true in full measure for both chalcones
and a number of structurally related phenols [18].
(a)
i
+
ArOH
1.00
0.75
0.50
0.25
0
i
+
ArOH
1
0
0
.00
3
(b)
.75
.50
1
2
3
0.25
0
0
200
400
600
800
t, s
0
1000 2000 3000 4000 5000 6000 7000 8000
, s
t
i
Fig. 7. Kinetics of CL on the oxidation of diphenylꢀ
Fig. 8. Structure of the Cꢀradicals of 1,2ꢀdioxetane (a)
D
1
methane after the addition of various concentrations of
i
2
7
and (b)
D
, products of the addition of O2 to the radical
Ch6. [ArOH]0
×
10 , mol/l: (
1) 0.4, (
2
) 2.0, and (
3
) 8.0.
The insert shows the initial portions of kinetic curves 1–3
shows the result of a long (2ꢀh) experiment. Experimenꢀ
tal conditions are specified in Fig. 5.
;
OAr• of chalcone Ch6; ΔHf0
i
(
D
) = –19.9 kcal/mol;
1
3
ΔHf0
(
D
i
)
= –27.1 kcal/mol.
2
KINETICS AND CATALYSIS Vol. 51
No. 4
2010

ANTIOXIDANT ACTIVITY OF CHALCONES: THE CHEMILUMINESCENCE
513
0
f
.
k7[ArOH] = (2k6wi) (i–
1/2
1/2
i1/2
Table 2. Heats of formation (ΔH ) of dioxetane radical D
2
–
1/2 –1/2
)
(9)
.
×
[1 + (k17
/
k8)(2k6/ i
w
)
i
],
(
product of the addition of O to OAr ) and the heats of the reꢀ
2
.
.
and k17 is calculated from the dependence of
ArOH]:
i
on
action O + OAr
D (
Δ
H
) calculated by the PM3 method
2
[
0
(D.),
kcal/mol
∞
ΔHf
Δ
H
,
Reaction type:
ArOH
Ch6
CA
k8 (2k6wi)–1/2(2 –
−1/2
1/2
kcal/mol endo/exothermic
k17
=
f
)[ArOH]/ (i
− i )dt
.
∫
0
–27.1
–106.8
–99.3
–20.0
21.4
–20.2
–40.3
32.4
12.09
1.9
5.5
2.8
Endo
Endo
Endo
Exo
Exo
Exo
Table 3 illustrates the degree of coincidence
between calculated parameters upon the treatment of
the results of measurements by various procedures.
Note that the calculation of k7 from the maximum
slope of CL recovery, that is, by classical Eq. (7),
resulted in values close to those obtained from modiꢀ
fied refined Eq. (9); that is, the calculation was suffiꢀ
ciently reliable. Comparing the slopes at the end of an
induction period for various antioxidants (Fig. 5), we
FA
Ch4
Ch2
Ch5
Catechol
–2.4
–0.3
–2.5
–4.1
3.8
Exo
–
Endo
Endo
Exo
Phenol
Guaiacol
1.4
–29.44
24.34
–4.0
–1.1
can rapidly evaluate the rate constants k7
Unlike Ch6 and CA, the antioxidant effect of the
other chalcones was weak. For example, at chalcone
.
–
Exo
–7
concentrations of
negligible value of 1–2 to 5% (Table 4); thereafter, it
remained constant. For these antioxidants, k7[ArOH]
2 × 10 mol/l, CL was quenched by a
the rate constant of the reaction of OAr• with O2, of
course, is several orders of magnitude lower than that
of diffusion. However, we can expect that it can comꢀ
pete with the rapid reaction OAr + ROO
scheme) because the concentration of O2 in oxidized
Ӷ
•
2
k6[ROO ], and the relationship between the intensity
of CL and the concentration of ArOH is expressed by
the equation
•
•
→
(see the
solution is higher by several orders of magnitude than
the steadyꢀstate concentration of ROO radicals.
[ArOH](i– – 1)
1
–1
•
(
10)
1
/2
–1 1/2
= (k k6/k7k8) + (2k6wi) (2k7) i .
17
Measuring CL (i.e., i) upon sequentially adding
antioxidant portions, we can estimate k7 and k17
for example, by determining the intercept and slope
of a linear anamorphosis of Eq. (10) in the
Note that even the complete removal of phenoxyl
radicals from the antioxidant action would decrease
the stoichiometric coefficient
from 2 to 1. Indeed, is much lower than unity
Table 3); this fact suggests that the scheme of inhibꢀ
ited oxidation should be seriously revised (compliꢀ
cated).
,
f only by a factor of 2—
f
(
⎯
1
–1
1/2
[
ArOH](
i
– 1) –
i
coordinates.
On the Role of the Second O–H Bond
in a CatecholꢀType Phenol
Nevertheless, even the formal introduction of only
one side reaction, OAr + O2, allowed us to obtain k7
•
rate constant values that are more realistic. Belyakov The O–H bond order in the phenoxyl radicals of Ch6
et al. [18] found by calculations that, in this case, chalcone and CA is 0.92, that is, smaller than unity,
Eq. (5) is replaced by the equation
and we can expect that this bond is weakened. The
Table 3. Rate constants k and k of the reactions of Ch6 and CA .* Estimation from a maximum slope of CL reduction: (A)
7
17
from Eq. (7), (B) from classical Eq. (5), and (C) from modified Eq. (9)
–1
k7, 10–6 l mol–1 s **
107,
[
АrOH]
×
f
k17, s–1
mol/l
A
B
C
Ch6
CA
0.4
2.0
8.0
0.7
0.5
0.4
0.9
1.0
1.4
11.0
8.7
9.9
2.7
1.1
0.5
7.6
4.8
4.5
0.4
2.0
8.0
0.5
0.4
0.3
2.2
2.1
2.3
10.2
4.6
1.9
1.9
0.7
0.2
9.6
4.3
1.8
*
Experimental conditions are specified in Fig. 5.
8
–1 –1
; k = 5.0 × 10 l mol s–1
8
–1
** 2k6 = 1.5
×
10 l mol
s
.
8
KINETICS AND CATALYSIS Vol. 51
No. 4
2010

514
VASIL’EV et al.
Table 4. CL attenuation after the addition of an antioxidant
nol, it is stronger than the former, and its participation
in reaction with ROO is less probable. However, the
OH group plays the role of an electronꢀdonor substituꢀ
ent, which weakens the neighboring ArO–H bond, as
compared with the other chalcones (cf. the values of
•
CL intensity, % on an i basis
0
ArOH
[ArOH], mol/l
8.0
10–7
10–7
2.0
×
×
D
(ArO–H) for various phenols in Table 5). It is interꢀ
esting that the electronꢀwithdrawing substituent CN
increased (ArO–H) by 2.1 kcal/mol and thus
CrC1
Ch6
CA
Ch1
Ch2
Ch3
Ch4
Ch5
0
0
6
18
30
96
99
92
95
98
D
22
85
94
76
86
93
decreased the antioxidant activity of the model tocoꢀ
pherol [21].
CONCLUSIONS
Note: Experimental conditions are specified in Fig. 5.
Thus, we studied a number of chalcones, the bioꢀ
logical precursors of flavonoids. We estimated the rate
constants (k7) of the main reaction path, hydrogen
bond cleavage product is о
ꢀquinone. In the case of atom abstraction from ArOH by peroxide radical
•
•
Ch6 chalcone, the PM3 estimation gave ROO with the formation of aroxyl radical OAr
0
(Table 5), and the efficiency of inhibition (number
f
of
ΔHf (quinone) = –22.09 kcal/mol. Consequently,
•
ROO radicals trapped by a molecule of ArOH).
considering the hypothetical reaction of O–H bond
•
cleavage in OAr
In addition to the antioxidant effect, chalcones
exert a prooxidant effect because of the participation
of an oxygen molecule at the steps following the forꢀ
mation of radical OAr . The structure and energetics
of possible products of the reaction of phenoxyl radiꢀ
cals OAr with О2 were calculated using the semiemꢀ
pirical PM3/PM6 method. It was found that adducts
OAr• = quinone + H,
0
0
•
we obtain
D(O–H) = ΔHf (quinone) + ΔHf (H) –
0
•
ΔHf (OAr ) = –22.09 + 52.10 + 29.03 = 59.04 kcal/mol.
An analogous evaluation for (O–H) in CA phenoxyl
•
D
0
0
(
ΔHf (quinone) = –105.1 kcal/mol; ΔHf (OAr•)
=
•
can be not only peroxide radicals OArOO but also
⎯
112.3 kcal/mol) resulted in D(O–H) = 59.3 kcal/mol. dioxetane radicals—the products of cyclization, that
is, the addition of the terminal atom of OArOO• to
That is, indeed, the bond is weak and, in principle, it
can be cleaved in both the reaction of phenoxyl with a
carbon of the Ar ring. The reaction is almost thermally
•
•
peroxide radical (OAr + ROO ) and, likely, the reacꢀ neutral; consequently, it is highly probable and can be
tion with O2. As for the second O–H bond in Ch6 pheꢀ a path involving O2
.
0
f
Table 5. Heats of formation (ΔH ) of phenols and phenoxyl radicals calculated by the PM6 method, O–H bond energies
D(O–
*
*
H)*, and rate constants k7 and k17
0
ΔHf (ArOH)
ΔHf (O )
Ar.
0
D(O–H)*
**
k ,
7
Phenols
k17, s–1
l mol–1 s–1
kcal/mol
Ch6
CA
CrC1
FA
Ch5
Ch3
Ch4
CoumA
Ch1
Ch2
–54.8
–134.7
–109.6
–134.7
–50.8
–14.4
–57.1
–98.2
–13.8
–13.5
–27.7
–110.2
–83.7
–103.0
–18.3
21.4
–20.8
–61.6
22.8
79.2
76.6
78.0
83.9
84.5
87.9
88.5
88.7
88.7
89.1
8.7
4.6
7.6
3.6
1.2
1.1
2.4
2.2
3.9
1.1
×
×
×
×
×
×
×
×
×
×
106
1.0
2.1
–
0.6
0.1
2.5
0.3
0.1
1.9
1.7
6
10
6
10
4
10
4
10
5
10
10
4
4
10
4
10
4
23.5
10
0
0
0
0
*
D(ArO–H) = ΔHf (OAr ) + ΔHf (H) – ΔHf (ArOH); ΔHf
(
H
)
= 52.1 kcal/mol.
10–7 mol/l using Eq. (7) and, for the other substances,
*
* Estimated from experimental data for the first three substances at [ArOH] = 2.0
×
0
from the dependence of i on [ArOH] using modified Eq. (10) at
2k6 = 1.5
×
108 l mol
–1 –1
s
and k8 = 5.0
×
108 l mol
–1 –1
s .
KINETICS AND CATALYSIS Vol. 51
No. 4
2010

ANTIOXIDANT ACTIVITY OF CHALCONES: THE CHEMILUMINESCENCE
515
The compounds examined can be subdivided into
two groups. 3,4ꢀDihydroxychalcone (Ch6), which has
a catechol structure, that is, two neighboring OH
groups in phenyl ring B, exhibits high antioxidant
activity. This is consistent with the lowest bond
REFERENCES
1
. Denisov, E.T. and Afanas’ev, I.B., Oxidation and Antioxꢀ
idants in Organic Chemistry and Biology, Boca Raton,
Fla.: CRC, Taylor & Francis Group, 2005.
2. Roginskii, V.A., Fenol’nye antioksidanty: reaktsionnaya
aktivnost' i effektivnost’ (Phenol Antioxidants: Reactivꢀ
ity and Efficiency), Moscow: Nauka, 1988.
. Burlakova, E.B., Ross. Khim. Zh., 2007, vol. 51, no. 1,
p. 3.
4. Kancheva, V.D., Eur. J. Lipid Sci. Technol., 2009,
vol. 111, p. 1072.
. Shahidi, F. and Wanasundara, P.K.J.P.D., CRC Critical
Reviews in Food Science and Nutrition, 1992, vol. 32,
p. 67.
strength
D(O⎯H) = 79.2 kcal/mol. In this case, the
•
second OH group does not react with the ROO radiꢀ
cal, but it serves as an electronꢀdonor substituent. For
Ch6, the reaction of hydrogen abstraction by the
3
•
ROO radical is the main reaction path, and the rate
constant k7 (~107 l mol s ) has the same order of
magnitude as that in efficient antioxidants—chroman
С1 and caffeic acid. With consideration for similarity
to a naturally occurring material and nontoxicity, this
allows us to hope that chalcone Ch6 and/or other
related compounds with a catechol structure can be
used as stabilizers for food products and pharmaceutiꢀ
cals.
–1
–1
5
6
7
8
. Barlow, S.M., in Food Antioxidants, Hudson, B.J.F.,
Ed., Amsterdam: Elsevier, 1990, p. 253.
. Dziedzic, S.Z. and Hudson, B.J.F., Food Chem., 1983,
vol. 12, p. 205.
. Popova, M., Bankova, V., Spassov, S., Tsvetkova, I.,
Naydenski, Ch., Silva, M.V., and Tsartsarova, M.,
Z. Naturforsch., A: Phys. Sci., 2001, vol. 56c, p. 593.
. Kancheva, V.D. and Bankova, V.S., Bulg. Chem. Comꢀ
mun., 2008, vol. 40, p. 546.
The other compounds are characterized by lower
antioxidant activity and, correspondingly, stronger
9
ArO–H bond:
D(O–H) = 83–88 kcal/mol. The
4
5
–1 –1
uncorrelated variation of k7 from 10 to 10 l mol s
and a low stoichiometric coefficient suggest that the
reaction paths other than a classical path are considerꢀ
able for these molecules; consequently, they hardly
can serve as efficient antioxidants.
1
0. Belyakova, V.A., Vasil’ev, R.F., and Fedorova, G.F.,
Kinet. Katal., 2004, vol. 45, no. 5, p. 695 [Kinet. Catal.
Engl. Transl.), vol. 45, no. 5, p. 655].
1. Batovska, D., Parushev, S., Slavova, A., Tsvetkova, I.,
(
1
Ninova, M., and Najdenski, H., Eur. J. Med. Chem.
007, vol. 42, p. 87.
2. Van Hook, J.P. and Tobolsky, A.V., J. Am. Chem. Soc.
958, vol. 80, p. 779.
3. Buchachenko, A.L. and Vasserman, A.M., Stabil’nye
radikaly (Stable Radicals), Moscow: Khimiya, 1973.
4. Lithoxoidou, A.T. and Bakalbassis, E.G., J. Phys.
Chem. A, 2005, vol. 109, p. 366.
5. Bakalbassis, E.G., Lithoxoidou, A.T., and Vafiadis, A.P.,
J. Phys. Chem. A, 2006, vol. 110, p. 11 151.
6. Lucarini, M., Pedulli, GF., and Guerra, M., Chem. Eur.
J., 2004, vol. 10, p. 933.
7. Suryan, M.M., Kafafi, S.A., and Stein, S.E., J. Am.
Chem. Soc, 1989, vol. 111, p. 4594.
8. Belyakov, V.A., Roginsky, V.A., and Bors, W., J. Chem.
Soc., Perkin Trans. 2, 1995, p. 2319.
9. Loshadkin, D., Roginsky, V., and Pliss, E., Int. J. Chem.
Kinet., 2002, vol. 34, p. 162.
0. Vasil’ev, R.F. and Trofimov, A.V., Kinet. Katal., 2009,
vol. 49, no. 4, p. 562 [Kinet. Catal. (Engl. Transl.),
vol. 49, no. 4, p. 518].
,
,
2
The correlation of antioxidant activity with the
energy of the O–H bond was repeatedly discussed (see
1
1
1
1
[
14–16, 22–25] and references therein). In spite of the
scatter of experimental data and theoretical evaluaꢀ
tions, it is indubitable that the bond energy plays an
important role in the reactivity of antioxidants, as eviꢀ
denced, in particular, by the results of this work. In real
systems, the correlation can be masked by other facꢀ
tors (the role of a medium, redox properties, structure
peculiarities, side reactions, solubility, etc.).
1
1
1
1
1
2
The semiempirical calculation of the ArO–H bond
strength is a rapid and convenient test for the applicaꢀ
bility of one or another structure (at least among strucꢀ
turally similar compounds) as a potential antioxidant
and, likely, as a test for the reasonability of the syntheꢀ
sis of the given molecule.
ACKNOWLEDGMENTS
21. Singh, N.K., O’Malley, P.J., and Popelier, P.L.A.,
J. Mol. Struct. (THEOCHEM), 2007, vol. 811, p. 249.
2. Kovtun, G.A., Katal. Neftekhim., 2000, no. 4, p. 1.
3. Wright, J.S., Johnson, E.R., and DiLabio, G.A., J. Am.
Chem. Soc., 2001, vol. 123, p. 1173.
4. Hussain, H.H., Babic, G., Durst, T., Wright, J.S., Fluꢀ
eraru, M., Chichirau, A., and Chepelev, L.L., J. Org.
Chem., 2003, vol. 68, p. 7023.
This work was performed within the framework of a
cooperation program of the Russian and Bulgarian
Academies of Sciences (theme no. 1); it was supported
by the Russian Academy of Sciences (program no. 1 of
the Division of Chemistry and New Materials), and
the National Scientific Foundation of the Ministry of
Education and Science of Bulgaria (contract nos.
Xꢀ1513 and BIn4/04).
2
2
2
2
5. Tikhonov, I., Roginsky, V., and Pliss, E., Int. J. Chem.
Kinet., 2009, vol. 41, no. 2, p. 92.
KINETICS AND CATALYSIS Vol. 51
No. 4
2010