Effect of metal ions on acetone dicarboxylic acid catalyzed peroxomonosulphate reactions

Article abstract of DOI:10.1016/j.molcata.2014.02.023

The oxidation of Fe(III), Ni(II) and Co(II) citrates by peroxomonosulphate (PMS) in the pH range 3.0-6.0 follows autocatalysis mechanism. The acetone dicarboxylic acid (ADC), the oxidative decarboxylation product from citrate, is found to catalyze the reaction. The added metal ions switch the reaction from the ADC catalyzed decomposition of PMS to the oxidation of citrates. Based on the results from alcohol quenching, a mechanism involving oxygen atom transfer is proposed.

Full text of DOI:10.1016/j.molcata.2014.02.023

G Model
MOLCAA-9024; No. of Pages7
ARTICLE IN PRESS
Journal of Molecular Catalysis A: Chemical xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Effect of metal ions on acetone dicarboxylic acid catalyzed
peroxomonosulphate reactions
∗
G. Ragukumar, P. Andal, M. Murugavelu, C. Lavanya, M.S. Ramachandran
School of Chemistry, Madurai Kamaraj University, Madurai 625021, India
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 31 July 2013
Received in revised form 20 February 2014
Accepted 22 February 2014
Available online xxx
The oxidation of Fe(III), Ni(II) and Co(II) citrates by peroxomonosulphate (PMS) in the pH range 3.0–6.0
follows autocatalysis mechanism. The acetone dicarboxylic acid (ADC), the oxidative decarboxylation
product from citrate, is found to catalyze the reaction. The added metal ions switch the reaction from
the ADC catalyzed decomposition of PMS to the oxidation of citrates. Based on the results from alcohol
quenching, a mechanism involving oxygen atom transfer is proposed.
©
2014 Elsevier B.V. All rights reserved.
Keywords:
Metal citrate
Peroxomonosulphate
Acetone dicarboxylic acid
Oxidation
Autocatalysis
1
. Introduction
of radical intermediates generated. Cobalt(II), nickel(II) and fer-
rous ions produce sulphate radical intermediate and Co(II) is more
efficient catalyst. The sulphate radical, with a redox potential of
2.5–3.1 V [13], is a better oxidizing agent than its precursor PMS.
Therefore, Co(II)–PMS system is one of the methodologies used in
advanced oxidation technology (AOT) for environmental technol-
ogy such as removal of organic pollutants in water remediation
[14–19] and degradation of dyes [20–25]. Wang and co-workers
[24,25] observed that the reaction Co(II)-PMS with chloride ion
proceed through both one electron transfer (resulting chlorine
radical through sulphate radical intermediate) and two electrons
transfer (producing chlorine). Results on the oxidation of cobalt(II)
malate complexes by PMS [26] show that in the pH range 4.0–5.9
the probable mechanism may be a molecular (oxygen atom trans-
fer) one. These observations suggest that the reactions Co(II)–PMS
system under favorable conditions can proceed through molecular
mechanism also.
The ␣-hydroxy carboxylates undergo oxidative decarboxylation
to corresponding aldehydes or ketones [27]. It has been shown
that the oxidative decarboxylation proceeds mainly through two-
electrons process if the carboxylate contains both alpha hydrogen
and alpha hydroxyl groups [28,29]. Citric acid/citrate has no hydro-
gen at hydroxyl carbon but the earlier reports [28,30–33] suggest
that the oxidation leads to carbon dioxide elimination giving
acetone dicarboxylic acid (ADC). Aliphatic ketones catalyze the
decomposition of PMS [34,35] through a stable intermediate oxi-
rane [36] which is also a powerful oxidizing agent [37–39]. Not
only the simple ketones, but also alpha keto acids/esters such as
Citric acid (2-hydroxy propane-1,2,3-tricarboxylic acid), a weak
organic acid, is an important intermediate in the tricarboxylic acid
cycle, a metabolic pathway involved in the conversion of carbo-
hydrates, fats and proteins to generate energy [1]. Its versatility
as ligand for transition metal ions in chemical reactions of bio-
logical and analytical interest is widely known. Citric acid is used
as a masking reagent for most metal ions [2] and buffering agent
in the universal buffer solution, Mcllavaine’s buffer [3]. The reac-
tions involving citric acid are used to determine the ultra trace
quantities of metal ions [4,5]. Citric acid plays an important role
in food and pharmaceutical industries as acidulent, flavoring and
preservative agent [6]. Citric acid and citrate are the reagents used
in the environmental friendly nickel electroplating baths [7–10].
The biological and industrial importance of citric acid results in an
increased attention on its oxidation and no report is available on
the reaction with peroxomonosulphate, an inorganic peroxide.
Peroxomonosulphate ion (PMS), commercially available as
®
OXONE is a powerful oxidant with a reduction potential of +1.8 V
[
11]. PMS in the presence of transition metal ions generates radical
intermediates of high oxidizing capacity. Anipsitakis and Dionys-
iou [12] studied the activation of PMS by various transition metal
ions with special reference to the catalytic efficiency and the nature
∗ Corresponding author. Tel.: +91 98942 23686; fax: +91 452 2459845.
E-mail address: ramachandran mku@yahoo.co.in (M.S. Ramachandran).
http://dx.doi.org/10.1016/j.molcata.2014.02.023
1
381-1169/© 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: G. Ragukumar, et al., J. Mol. Catal. A: Chem. (2014), http://dx.doi.org/10.1016/j.molcata.2014.02.023

G Model
MOLCAA-9024; No. of Pages7
ARTICLE IN PRESS
G. Ragukumar et al. / Journal of Molecular Catalysis A: Chemical xxx (2014) xxx–xxx
2
pyruvates [40,41] are also reported to give oxirane type inter-
mediates. These ketone catalyzed reactions were observed only
at neutral or weakly alkaline medium. Preliminary experiments
from our laboratory shows that the ADC catalyzes the decom-
position of PMS in moderate acidic pH (3.0) also. Results from
this laboratory on the oxidation of metal(II)-␣-hydroxy carboxyl-
ates by PMS [26,42] show that the reactions follow auto catalysis.
The intermediate between ␣-hydroxy carboxylate and its oxidative
decarboxylation product aldehyde is responsible for autocatalysis.
Similar to this, the product acetone dicarboxylic acid may catalyze
the oxidation of metal citrates or it can enhance the decomposi-
tion of PMS. Therefore, to explore the actual role of ADC it has been
studied that the oxidation of citric acid by PMS in the presence of
Fe(III), Ni(II) and Co(II) in the pH range 3.0–6.0 and the results are
discussed in this report.
4.5
4.0
3.5
A) [Citric acid] = 0.10 M
pH = 6.6
B) [Citric acid] = 0.05 M
-
3
[
Fe(III)] = 1.00 x10
pH = 3.0
C) [Citric acid] = 0.05 M
M
-
6
[
Co(II)] = 5.00x10 M
pH = 4.0
3.0
2.5
2.0
1.5
1.0
A
B
C
0
100
200
300
400
2
. Materials and methods
Time (Mins)
◦
Fig. 1. Plot of [PMS] vs. time at 31.0 C.
All the chemicals used were of the highest purity commercially
available and were used as received. The peroxide under the name
®
OXONE , monopersulphate compound was from Sigma–Aldrich
of PMS decomposed per mole of ADC, was greater than six. These
observations clearly show that in the absence of metal ions the
decomposition of PMS is observed as in Eq. (2).
GmbH (Germany). Ni(NO ) · 6H O (Extrapure, SD FINE-CHEM,
3
2
2
India), Co(OAC) ·4H O (Merck, India) and NH Fe(SO ) · 12H O
2
2
4
4
2
2
(
GR, Merck, India) were the source of metal ions. Acetone
−
C
i
t
r
i
c
a
c
i
d
+
A
D
C
+
H
S
O
→
C
i
t
r
i
c
a
c
i
d
+
A
D
C
+
O
↑
(2)
dicarboxylic acid (3-oxo glutaric acid) was 96% pure and from
Sigma–Aldrich GmbH (Germany). This compound was recrystal-
lized repeatedly from ethyl acetate to get a sharp melting point
5
2
However, in the presence of metal ions, the reaction reverted to
the oxidation of citric acid as confirmed by the formation of carbon
dioxide, absence of oxygen gas evolution and an increase in ADC
concentration equal to that of PMS concentration. Thus the reaction
in the presence of metal ions can be represented by Eq. (3).
◦
of 135 C [43]. The peroxide solution was freshly prepared daily
and standardized by iodometry. The pH of the reaction mixture
was adjusted with citric acid-phosphate buffer. The pH values were
adjusted to the predetermined values by adjusting the concentra-
tion of disodium hydrogen phosphate while keeping the citric acid
concentration at a predetermined value, usually at 0.05 M. The rates
of the reaction were calculated by following the concentration of
unreacted PMS iodometrically at various times.
M(II)/M(III)
Citric acid + ACD + HSO−5 −→ 2ADC + CO2
↑
(3)
3. Results and discussion
The stoichiometry of the reaction was determined at pH 4.8.
A large excess of PMS (0.05 M) over citric acid (0.01 M) and the
metal ion (0.002–0.0002 M) were allowed to stand for 6–8 h and
the unreacted PMS was estimated. The unreacted oxidant concen-
tration was very small in Ni(II) and Co(II) ions and this may be due to
the self decomposition of PMS catalyzed by the metal ions. There-
fore the stoichiometry was determined with equal concentration
The oxidation of citric acid was observed only at pH ≥6.0 and
that too at a slow speed. The conversion of [PMS] was ∼25% at
pH 6.0 and ∼50% at pH 6.6 (Fig. 1A) for 6 h. In the presence of
−5
−3
metal ions with the concentration range ∼10 M to 10 M, the
oxidation of citric acid/citrate proceeded smoothly at a measur-
able rate even at pH 3.0 (Fig. 1B and C). This clearly shows that the
metal ions catalyzed the oxidation of citric acid/citrate by PMS. The
(
0.01 M) of PMS and citric acid. The evolution of carbon dioxide
[
PMS]t–time profile for Fe(III) ion catalyzed reaction at pH 3.0 is
shown in Figs. 1B and 2A. The reaction shows an induction period,
usually 10–20 min after which the rate becomes fast. This feature
from this mixture was confirmed with freshly prepared lime water.
The oxidation product from citric acid gave 2,4-dinitro phenylhy-
drazone derivative which decomposed on heating. The formation
of acetone dicarboxylic acid (3-ketoglutaric acid) was confirmed
by the color test with sodium nitroprusside [44]. Acetone dicar-
boxylic acid was converted into acetone by the addition of aniline
4.5
[
(
45] and acetone was estimated as 2,4-dinitro phenylhydrazone
m.pt 126–127 C, lit. 128 C [46]). The quantitative estimation indi-
4.0
◦
◦
cated that ∼95–97% of the keto compound is produced per mole of
PMS. Therefore, the oxidation of citric acid can be represented as in
Eq. (1).
3
3
2
.5
.0
.5
A
Acetone dicarboxylic acid
M(II)/M(III)
Citric acid + HSO−5 −→
(ADC)
+ CO2
↑
(1)
B
The stoichiometric determinations were also made in the pres-
ence of acetone dicarboxylic acid also. One interesting observation
was that in the absence of metal ions, the evolution of oxygen
gas was observed and was confirmed by the color test with alka-
line sodium dithionite activated with indigo carmine [47]. Also
the evolution of carbon dioxide was inhibited and could not be
detected with lime water. The change in the concentration of
acetone dicarboxylic acid, estimated as acetone, was practically
negligible. Moreover, the turn over number, the number of moles
-3
[
Fe(III)] = 1.00 x 10
pH = 3.0
Temp = 31.0 C
M
o
2.0
A - [ADC] = 0.00 M
B- [ADC] = 5.00 x 10
-4
M
1.5
0
20
40
60
80
100
120
Time (Mins)
Fig. 2. PMS–time profile for Fe(III) ion catalyzed reaction.
Please cite this article in press as: G. Ragukumar, et al., J. Mol. Catal. A: Chem. (2014), http://dx.doi.org/10.1016/j.molcata.2014.02.023

G Model
MOLCAA-9024; No. of Pages7
ARTICLE IN PRESS
G. Ragukumar et al. / Journal of Molecular Catalysis A: Chemical xxx (2014) xxx–xxx
3
4
4
3
3
2
2
1
1
.5
.0
.5
.0
.5
.0
.5
.0
3.6
3.4
3.2
3.0
2.8
2.6
pH = 4.0
2.4
o
Temp = 31.0 C
2
.2
.0
A,B -[Co(II)] = 5.00 X 10-6
M
B
2
A - [ADC] = 0.00 M
B- [ADC] = 1.00 X 10
A
-3
M
1.8
.6
1.4
1
0
10
20
30
40
50
60
70
1.5
2.0
2.5
3.0
3.5
4.0
4.5
3
Time (Mins)
Fig. 3. PMS–time profile for Co(II) ion catalyzed reaction.
10 x [PMS] M
Fig. 4. Plot of (rate/[PMS]) vs. [PMS] for Co(II) ion catalyzed reaction in A.
is observed at all pH values (3.0–6.0) employed in this study. How-
ever no such induction period was observed in the Ni(II) and Co(II)
with the earlier reports [28,30–33]. Therefore, one would expect
that the catalyst ‘C’ in Eq. (4) may be acetone dicarboxylic acid.
The proposed kinetic scheme (Eqs. (4) and (5)) can be verified
by calculating kinetic constants k1 obs and k2 obs with the added
acetone dicarboxylic acid. Before that, the interaction of acetone
dicarboxylic acid with PMS and citric acid has to be studied. The
product analysis and stoichiometric determination (Eq. (2)) show
that only the self decomposition of PMS takes place and this reac-
tion is observed even at pH 3.0. One such result on the loss of PMS
in the absence of metal ions is shown in Fig. 5A. Perusal of the result
shows that the rate of [PMS] loss follows first order kinetics, as in
Eq. (9), over
(Figs. 1C and 3A) ions catalyzed reactions. The Co(II) catalyzed oxi-
dation at pH 3.0 can be represented by simple first order kinetics
with respect to [PMS] but at higher pH values and for Ni(II) (pH
3
.0–6.0) the rate can not be expressed by the simple rate equa-
tions. Analysis of the results suggests that the metal ion catalyzed
oxidation of citric acid can be explained by simple autocatalysis
mechanism shown in reactions (4) and (5).
k
1
O + S−→ C + P
(4)
(5)
k
2
S + C + O−→ 2C + P
−
d[PMS]
dt
= kobs[PMS] = kADC[ADC][PMS]
(9)
The oxidant PMS (O) reacts with the substrate (S) to yield the
products C and P. One of the products (C) accelerates the reaction.
The rate in terms of the decrease in [PMS] is given in Eq. (6).
the pH range 3.0–6.0 and the plots kobs vs. [ADC] (Fig. 6) are in
ADC
straight lines passing through the origin. The k
values calculated
−
[PMS]
dt
rate =
= k [S][PMS] + k [C] [PMS]
(6)
at different pH values are shown in Table 1.
1
t
2
t
t
The [PMS]–time profile in the presence of metal ion is shown in
Fig. 5(B). Comparison of the curves in Fig. 5 shows that the mecha-
nism of the reaction in the presence of the metal ion (Fig. 5B) may
be different from its absence (Fig. 5A). This is also supported by the
product analysis (Eqs. (2) and (3)). Careful analysis such as the plot
of (rate/[PMS] ) vs. [PMS] for the curve 5B shows that the reaction
If the catalyst (C) leakage is negligible, then [C]t can be replaced
by the term [C] + [PMS] − [PMS]t where [C] is the concentra-
0
0
0
tion of the catalyst at the start and ([PMS] − [PMS]t) represent
0
the catalyst produced in the reaction. Since [S] ꢀ [O] Eq. (6) can
be simplified as in Eq. (7) where k1
obs
= k · [S] and k
2obs
= k · [S].
1
2
t
t
with added ADC also follows autocatalysis. This can be verified by
rate = k1obs[PMS] + k
([C] + [PMS] − [PMS] )[PMS]
(7)
t
2obs
0
0
t
t
According to Eq. (7) the plot rate/[PMS]t vs. [PMS]t should be
in a straight line with a negative slope and positive intercept. This
is found to be true (Fig. 4) and this confirms that all the reactions
follow autocatalysis even though PMS-time profiles for Co(II) and
Ni(II) sans the initial induction period.
-5.2
-5.4
-5.6
-5.8
-6.0
-6.2
-6.4
-6.6
[
Citric Acid] = 0.05M
pH = 3.0
o
Temp = 31.0 C
-3
The better values of k1 obs and k2 obs can be obtained from Eq. (8),
the integrated form of Eq. (6).
A)[ADC] = 4.00*10
[Fe(III)] = 0.00M
B)[ADC] = 5.00*10
[Fe(III)] = 1.00*10
M
-
4
-3
M
M
[PMS]t
k1obs + k2obs([PMS]0 + [C] )
0
=
(8)
k2obs + ((k1obs + k2obs[C] )/[PMS] ) exp(t × (k1obs + k2obs([PMS] + [C] )))
0
0
0
0
B
The kinetic constants k1 obs and k2 obs can be obtained from Eq.
8) by non-linear regression analysis. The k1 obs and k2 obs values
A
(
calculated from Eqs. (7) and (8) are agreeable. However, as a rule
the k1 obs and k2 obs have been calculated from non-linear regression
analysis of Eq. (8) using the values from the derivative plots (Eq. (7))
as the approximate initial inputs for non-linear regression.
The product analysis (Eq. (1)) shows that the oxidation product
from citric acid is acetone dicarboxylic acid. This is in accordance
0
20
40
60
80
100
Time(Mins)
◦
Fig. 5. The ln([PMS])–time plot at 31.0 C.
Please cite this article in press as: G. Ragukumar, et al., J. Mol. Catal. A: Chem. (2014), http://dx.doi.org/10.1016/j.molcata.2014.02.023

G Model
MOLCAA-9024; No. of Pages7
ARTICLE IN PRESS
G. Ragukumar et al. / Journal of Molecular Catalysis A: Chemical xxx (2014) xxx–xxx
0.0005
4
5
4
3
2
1
0
0.0004
0.0003
0.0002
0.0001
0
.0000
0
.0000 0.0001 0.0002 0.0003 0.0004 0.0005 0.0006 0.0007 0.0008
0.000
0.002
0.004
0.006
0.008
0.010
0.012
[
Ni(II)] M
[ADC]M
◦
Fig. 7. Plot of k1obs vs. [Ni(II)] at pH = 6.0, temp. = 31.0 C.
◦
Fig. 6. Plot of kobs vs. [ADC] at 31.0 C.
−
4
the k1 obs values differ from k2 obs by a factor of ∼10 . The kinetic
Table 1
Kinetic constants k
−
3
ADC
studies were usually carried out with [PMS] ≈ 4.0 × 10 M and
for the decomposition of PMS.
0
hence k
ꢁ k
[PMS] . The k
values for Fe(III)–citric acid
◦
102 × kADC (M
−1 −1
1 obs
2 obs
0
1 obs
Temp. ( C)
s
)
system calculated by non-linear regression may have large fluctu-
ation and this may be the reason that the correlation between k1 obs
and [Fe(III)] could not be obtained.
At constant pH and citric acid, k2 obs values were independent
of Co(II) and Ni(II) concentration whereas in ferric citrate oxida-
pH
.0
3
4.0
4.8
13.14 ± 0.63
6.0
3
1.0
4.14 ± 0.23
9.92 ± 0.73
14.67 ± 1.34
4.0)a
a
a
a
(
–
(8.25)
(8.5)b
(15.0)
(11.7)b
(14.0)
(13.0)
b
tion k2 obs showed a linear correlation with [Fe(III)]. The plots k2 obs
vs. [Fe(III)] were straight lines passing through (approximately)
Values in the parenthesis from the metal ion catalyzed oxidation in the absence of
ADC.
the origin. The kinetic constant values were also calculated in the
presence of aliphatic alcohols such as ethanol and tert. butyl alco-
hol so as to ascertain the formation of radical intermediates such
as sulphate ion radical and or hydroxyl radical [12,13,48,49]. The
concentration of alcohols used were ∼100:1 molar ratio of the alco-
hol vs. the oxidant PMS. The alcohols did not affect the rate or
rate constants and this suggests that the possibility of the reac-
a
Ni(II) ion.
Co(II) ion.
b
calculating the kinetic constants k1 obs and k2 obs with and without
the added ADC. The kinetic constants for the Fe(III)–citric acid–PMS
system with no ADC at the start (Fig. 2A) calculated with Eq.
−
•
−
5
−1
−5 −1
) and
tion mechanism involving radical intermediates such as SO4 and
(
8) are: k1 obs = 1.01 ± 0.07 × 10
s
(3.77 ± 1.08 × 10
s
•
−2
−1 −1
−2
−1 −1
).
OH radicals can be neglected.
k2 obs = 17.56 ± 0.45 × 10
M
s
(13.22 ± 0.94 × 10
M
s
Peroxomonosulphate dissociates to give the dinegative anion as
in Eq. (11).
The values in the parenthesis correspond with ADC (Fig. 2B) at the
start. Similar values to the Co(II) ion catalyzed reaction (Fig. 3A
−5
−1
−5 −1
s
and B) are: k1 obs = 14.26 ± 0.17 × 10
s
(14.55 ± 1.37 × 10
)
Kd
−
2−
−
2
−1 −1
−2
−1 −1
HSO ꢀSO + H+
(11)
and k2 obs = 9.66 ± 0.02 × 10
M
s
(9.07 ± 0.73 × 10
M
s
).
5
5
The close agreement between the kinetic constant values from the
reactions in the absence and in the presence of ADC suggests that
the kinetic scheme proposed for the oxidation of metal citrate may
explain the experimental results.
◦
The pKa value of PMS is 9.4 at 25 C [50] and at the experimental
pH values (3.0–6.0) the equilibrium will be shifted towards left,
−
that is all the PMS will exist as HSO .i. The acetone dicarboxylic acid
5
catalyzed decomposition of PMS
The kinetic constants k1 obs, k2 obs and kobs were calculated at
different experimental conditions. The results show that sulphate
ion has no effect on the rate. However, all the kinetics studies were
carried out only in the presence of 0.05 M sulphate ion. The kinetic
constants k1 obs, k2 obs and kobs were found to be independent of
citric acid concentration in all the reactions.
The k1 obs values were found to increase with Co(II) and Ni(II)
ion concentration and the plots k1 obs vs. [M(II)] were in straight
lines passing through the origin (Fig. 7). Therefore the correla-
tion between k1 obs and [M(II)] can be represented as in Eq. (10).
The k1 values calculated at different pH values are shown in
Tables 2 and 3).
Acetone dicarboxylic acid is a weak acid with pKa [51] values
.23 and 4.27 at 25 C (Eqs. (12) and (13)). The experimental results
Eq. (2)) in the absence of metal
◦
3
(
K
HOOC CH COCH COOHꢁHOOC CH COCH COO + H+
a1
−
(12)
(13)
2
2
2
2
K
−
a2−
OOC CH COCH COO + H+
−
HOOC CH COCH COO
ꢁ
2
2
2
2
ions suggest that the loss of PMS can be due to the ADC cat-
alyzed decomposition of PMS. The results are similar to the ketone
catalyzed decomposition of caroate [34,35]. By analogy, some reac-
tive intermediate such as oxirane may be responsible for the ADC
catalyzed decomposition of PMS also. The detailed mechanistic
scheme is shown in Fig. 8.
Based on the experimental observations the following kinetic
scheme for the ADC catalyzed decomposition of PMS can be pro-
posed.
k1 obs = k · [M(II)]
(10)
1
The initial oxidation of Fe(III)–citric acid by PMS is very slow
and this may be the reason why the initial part of [PMS]–time pro-
file is almost parallel to time axis(Fig. 2A). Therefore k1 obs values of
Fe(III)–citrate may be associated with high uncertainty. Moreover
ADS
K
−
1
HOOC CH COCH COOH + HSO −→ Products + O
↑
(14)
2
2
5
2
Please cite this article in press as: G. Ragukumar, et al., J. Mol. Catal. A: Chem. (2014), http://dx.doi.org/10.1016/j.molcata.2014.02.023

G Model
MOLCAA-9024; No. of Pages7
ARTICLE IN PRESS
G. Ragukumar et al. / Journal of Molecular Catalysis A: Chemical xxx (2014) xxx–xxx
5
Table 2
Kinetic constants for the Co(II)–citrate oxidation.
◦
−1 −1
s )
Temp. ( C)
k1 (M
pH
3.0
4.0
4.8
6.0
2
3
5.0
1.0
2.28 (± 0.04)
4.89 (± 0.15)
17.23 (± 1.15)
28.58 (± 1.95)
59.27 (± 3.09)
95.31 (± 4.45)
289.83 (± 20.95)
471.52 (± 40.44)
Table 3
Kinetic constants for the Ni(II)–citrate oxidation.
◦
103 × k1 (M−1
−1
s )
Temp. ( C)
pH
3.0
4.0
4.8
6.0
2
3
5.0
1.0
1.10 (± 0.15)
3.06 (± 0.12)
11.76 (± 0.09)
27.0 (± 5.1)
75.24 (± 1.06)
101.0 (± 1.0)
362.1 (± 7.20)
548.0 (± 44.0)
ADS
◦
K
at 31 C. The observed difference can be explained via intra molec-
ular hydrogen bonding in ADC proposed by earlier researchers [51].
CH2
−
−
2
HOOC CH COCH COO + HSO −→ Products + O
↑
↑
(15)
2
2
5
2
ADS
3
K
O
CH
2
−
−
−
OOC CH COCH COO + HSO −→ Products + O
(16)
2
2
5
2
C
COOH
C
ADC
ADC
The k
values calculated from Eq. (9) can be written as k
=
ADC
1
ADC
2
ADC
3
˛H · k
+ ˛ · k
+ ˛ · k
where ˛ represents the fraction
A
HA
A
2
of the total acid present as corresponding ionized form. At pH
O
O
6
.0, ADC exists exclusively (>98%) as the doubly ionized form so
ADC
ADC
3
. Similarly from the values of kADC (Table 1) and
H
that k
= k
˛
values at different pH values the k values are calculated as
The studies on the self decomposition of PMS [50] showed that
the peroxide oxygen in SO 52 is a nucleophile while HSO5 is an elec-
−
−
ADC
1
−1 −1 ADC
S , k
−1 −1
ADC
−1 −1
k
= 0.02 M
= 0.07 M
S
and k3 = 0.15 M
S
2
trophile. Therefore, the first step in the interaction between ADC
and HSO5 is the electrophilic interaction of PMS with the carbonyl
oxygen (Fig. 8). The hydrogen bonding with the carbonyl group
would inhibit this reaction. Therefore, one would expect that the
reaction is more favorable in doubly ionized acetone dicarboxylic
acid and less in unionized ADC. Hence the expected order of rate
−
R
C
R1
_
O
+
HSO5
ADC
ADC
ADC
constant values would be k3 > k2 > k1
and the calculated
H
values are in accordance with this expectation.
R
ii. The metal ion catalyzed oxidation of citric acid:
_
C
O
O-O-SO3
Citric acid (H Cit) is an alpha hydroxy aliphatic acid and has
3
R1
four removable hydrogen atoms (as protons), three from carboxyl
groups and one from the hydroxyl group. Therefore, citrate ion can
act as a multidentate ligand towards transition metals. The metal
ion–citrate complex equilibria in solution have been investigated
extensively and the various complexes can be represented by the
following equations.
#
OH
R
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
C
_
II
III
−
M
M
M
M
M
M
+ H Cit ꢁ [M H Cit]
(17)
(18)
(19)
(20)
(21)
(22)
(23)
2
2
R1
O-O-SO3
II
III
2
−
+ HCit ꢁ [M HCit]
+ Cit3− ꢁ [MCit]
ADC
II
III
k
II
III
3
−
O
O
+ 2Cit ꢁ [M(Cit)2]
R
2
-
+
+
SO4
+
H
C
II
III
+ 2Cit + H+ ꢁ [M(Cit)(HCit)]
3
−
R1
(
OXIRANE)
II
III
+ Cit3− ꢁ [M(H 1Cit)] + H2
−
O
R
_
+
2-
ꢀ
ꢁ
fast
1
II
III
HSO5
RR C=O
+
H
+ SO4
+ O2
+ 2Cit3− ꢁ [M (H Cit) + 2H ]
+
C
O
+
2M
2
−1
2
R1
The complexation of citric acid with Ni(II) and Co(II) ions
Fig. 8. Mechanism for acetone dicarboxylic acid catalyzed decomposition of PMS.
involves the bonding with carboxylate groups but no interaction
Please cite this article in press as: G. Ragukumar, et al., J. Mol. Catal. A: Chem. (2014), http://dx.doi.org/10.1016/j.molcata.2014.02.023

G Model
MOLCAA-9024; No. of Pages7
ARTICLE IN PRESS
G. Ragukumar et al. / Journal of Molecular Catalysis A: Chemical xxx (2014) xxx–xxx
6
Table 4
Kinetic constants for the Fe(III)–citrate oxidation.
◦
−1 −1
s )
Temp. ( C)
k3 (M
pH
3.0
4.0
4.8
6.0
2
3
5.0
1.0
71.22 (± 8.34)
70.59 (± 8.70)
123.33 (± 4.15)
53.69 (± 4.64)
85.17 (± 7.01)
56.64 (± 3.35)
95.63 (± 12.34)
102.78 (± 12.65)
with the hydroxyl group. The complexes given in Eqs. (17)–(21)
are observed in Ni(II)–citric acid system [52,53] and Co(II)–citrate
with respect to [M(II)]. Moreover, if intermediate formation is the
rate limiting step, then Eq. (25) is identical to the reactions as rep-
resented in Eqs. (14)–(16). The k2 values should be equal to k
ADC
[
53] complexes can be represented as in Eqs. (17)–(19). A simple
calculation using the reported equilibrium constant values [52,53]
shows that ∼20–30% of [M(II)]T exists as (un complexed) free ion
at pH 3.0 which decreases to an insignificant level (∼1–3%) as
the pH is increased to 4.0. The nature of the complex depends
calculated from Eq. (9). The k2 values (in the parentheses) along
ADC
with k
at various pHs are displayed in Table 1. Perusal of the
results shows that they are in close agreement with each other and
justifies this expectation.
upon the pH of the reaction mixture. At pH 3.0 Ni(II)–citrate exists
In the case ferric citrate oxidation, Eq. (26) may be the rate
limiting step. The concentration of M(III)–citrate is constant (equal
to [M(III)]), and hence Eqs. (25) and (26) can be considered
as the consecutive reactions with the second step as the slow-
est step. Therefore, the second term in Eq. (27) is replaced by
+
as NiHCit (∼50%) and NiH Cit (∼26%). At a higher pH (≥4), the
2
3
−
4−
Ni(II) ion exists mainly as Ni(Cit)(HCit) and as Ni(Cit) . When
the pH is increased, the concentration of Ni(Cit)(HCit) decreases
while that of Ni(Cit) increases. Similarly, Co(II)–citrate exists as
2
3−
4
−
2
+
CoH Cit (∼15%) and CoHCit (∼50%) at a pH 3.0; and above this
k · [ADC] · [PMS] · [Fe(III)] and this will explain why k
shows a
2
3
2 obs
−
pH, it is a mixture of CoHCit and CoCit . Fe(III)–citrate complexes
linear correlation with [Fe(III)]. The kinetic constants calculated are
shown in Table 4.
involve the interaction with deprotonated hydroxyl group of cit-
ric acid represented by H 1Cit. Perusal of literature [54,55] shows
Perusal of the k values shows that the rate of oxidation of M(II)
−
1
that the complex equilibria between Fe(III) and citric acid/citrate
can be represented as in Eqs. (19), (22) and (23). Ferric citrate
exists mainly (∼92%) as complexes of deprotonated hydroxyl group
or M(III)–citrate by PMS is Co(II) ꢀ Ni(II) ꢀ Fe(III). The complexa-
tion with transition metal ions enhances the ionization of citrate
hydroxyl group [56,57], thereby leading to a metal-oxide bond-
ing. The hydroxyl group is completely ionized and the tetra ionized
citrate complex predominates in Fe(III) citrate as in Eqs. (22) and
(23). Ni(II)–citrate also exhibits an appreciable degree of polar-
ization/ionization, and bonding by the hydroxyl group is reported
[57]. The oxidation of citrate to acetone dicarboxylic acid involves
the hydroxyl group and the carboxylate group (as carbon dioxide)
attached to the same carbon atom. Therefore, the ease of oxidation
may inversely depend upon the polarization of hydroxyl group, and
this will explain the observed order of oxidation.
−
[Fe(H Cit)] , 19% and [Fe (H Cit) ] , 73%) even at pH 3.0 and
−1 2 −1
2
2−
(
only a minor fraction as [FeCit]. When the pH ≥4.0 [Fe (H Cit)2]2
−
2
−1
−
increases to ∼80%, [Fe(H Cit)] remains at the same level and
−
1
[
FeCit] can be neglected. Therefore, it can be inferred from the above
discussion that almost all the metal ions exist in the complexed
state even at the lowest pH (3.0) used in this study.
The quenching studies with aliphatic alcohols such as ethanol
and tert-butanol show that the formation of radical intermediates
•
−•
such as OH and SO4 [12,13,48,49] in the present investigation
can be excluded. Therefore, the oxidation of metal–citrate by PMS
proceeds through two electron processes, probably by oxygen atom
transfer, and the product is ADC. The catalyst ((C) in Eq. (2)) ADC
enhances the rate of oxidation through an intermediate with PMS.
The detailed kinetic scheme for the oxidation of metal citrate by
PMS can be written as in the following equations:
4. Conclusion
The oxidation of citric acid by peroxomonosulphate in the pH
range 3.0–6.0 (citric acid buffer) occurs only in the presence of
a metal ion. The Fe(III), Co(II) and Ni(II) citrates follow auto cat-
alyzed mechanism. Acetone dicarboxylic acid is the product which
catalyzes the oxidation of metal citrate. In the absence of metal
ions, acetone dicarboxylic acid catalyzes the decomposition of PMS
at pH 3.0–6.0 as in ketone catalyzed one at neutral or weak alka-
line pH. However, the complexation with metal ion changes the
mechanism to the acetone dicarboxylic acid catalyzed oxidation of
metal–citrate by PMS.
ꢀ
ꢁ
Acetone dicarboxylic acid
II
III
− Citrate + HSO−5 −k→1
M
(ADC)
+ CO2 ↑ +products
(24)
k
−
2
ADC + HSO −→ Intermediate + products
(25)
(26)
5
ꢀ
ꢁ
II
III
k3
Intermediate + M
− Citrate−→ 2 ADC + products
The experimental results suggest that Eq. (25) is the rate limiting
step for the oxidation of Ni(II) and Co(II) citrates and Eq. (26) is in
Fe(III)–citrate. The rate equation for Co(II) and Ni(II) can be written
as in Eq. (27).
Acknowledgements
The financial assistances from MKU, Madurai (USRF to Raguku-
mar) and UGC, New Delhi (Meritorious fellowships to Andal and
Lavanya, RGNF to Murugavelu) are gratefully acknowledged.
−
d[PMS]
dt
=
k .[M − Citrate][PMS] + k [ADC][PMS]
(27)
1
2
References
If there is no side reaction, the concentration of the catalyst
ADC can be approximated to the [PMS] reacted. Eq. (27) is iden-
[
1] J.E. Baldwin, H.A. Krebs, Nature 291 (1981) 381–382.
tical with Eq. (7) if k1 obs is replaced by k [M − Citrate] and k
by
1
2 obs
[2] D.D. Perrin, Masking and Demasking of Chemical Reactions, Wiley-Interscience,
New York, 1970.
k . Under the experimental conditions, all the metal ions will exist
2
[
[
[
3] T.C. Mcllvaine, J. Biol. Chem. 49 (1921) 183–186.
4] J. Kato, O. Ohno, S. Igarashi, Anal. Sci. 21 (2005) 705–707.
5] S. Prasad, Anal. Chim. Acta 540 (2005) 173–180.
as metal–citrate, and hence, the term [M − Citrate] in Eq. (27) can be
replaced by the metal ion concentration as [M(II)] itself. Therefore,
Eq. (27) explains all the experimental observations, namely the lin-
ear correlation between k1 obs vs. [M(II)] and independent of k2 obs
[6] R.E. Kirk, D.F. Othmer, Kirk-Othmer Encyclopedia of Chemical Technology, John
Wiley, New York, 2004.
Please cite this article in press as: G. Ragukumar, et al., J. Mol. Catal. A: Chem. (2014), http://dx.doi.org/10.1016/j.molcata.2014.02.023

G Model
MOLCAA-9024; No. of Pages7
ARTICLE IN PRESS
G. Ragukumar et al. / Journal of Molecular Catalysis A: Chemical xxx (2014) xxx–xxx
7
[
[
[
7] R. Tarozaite, O. Gyliene, G. Stalnious, Surf. Coat. Technol. 200 (2005) 2208–2213.
8] M. Ghorbani, A.G. Dolati, A. Afshar, Russ. J. Electrochem. 8 (2004) 1173–1177.
9] O. Younes, E. Gileadi, J. Electrochem. Soc. 149 (2002) 100–111.
[33] S.P. Datta, A.K. Grzybows, S.S. Tate, Nature 207 (1965) 1047–1049.
[34] R.E. Montgomery, J. Am. Chem. Soc. 96 (1974) 7820–7821.
[35] J.O. Edwards, R.H. Pater, R. Curci, F. DiFuria, Photochem. Photobiol. 30 (1979)
63–70.
[
[
[
[
[
[
10] R.Y. Ying, J. Electrochem. Soc. 135 (1988) 2957–2964.
11] M. Spiro, Electrochimca. Acta 24 (1979) 313–314.
[36] R.W. Murray, R. Jeyaraman, J. Org. Chem. 50 (1985) 2847–2853.
[37] W. Adam, R. Curci, J.O. Edwards, Acc. Chem. Res. 22 (1989) 205–211.
[38] R.W. Murray, Chem. Rev. 89 (1989) 1187–1201.
[39] R. Curci, A. Dinoi, M.F. Rubino, Pure. Appl. Chem. 67 (1995) 811–822.
[40] M.-K. Wong, N.-W. Chung, L. He, X.-C. Wang, Z. Yan, Y.-C. Tang, D. Yang, J. Org.
Chem. 68 (2003) 6321–6328.
12] G.P. Anipsitakis, D.D. Dionysiou, Environ. Sci. Technol. 38 (2004) 3705–3712.
13] P. Neta, R.E. Huie, A.B. Ross, J. Phys. Chem. Ref. Data 17 (1988) 1027–1284.
14] G.P. Anipsitakis, D.D. Dionysiou, Environ. Sci. Technol. 37 (2003) 4790–4797.
15] F.J. Rivas, F.J. Beltron, F. Carvalho, P.M. Alvarez, Ind. Eng. Chem. Res. 44 (2005)
7
49–758.
[
[
[
[
[
16] G.P. Anipsitakis, D.D. Dionysiou, Appl. Catal. B: Environ. 54 (2004) 155–163.
17] G.P. Anipsitakis, T.P. Tufano, D.D. Dionysiou, Water. Res. 42 (2008) 2899–2910.
18] P. Sukla, H. Sun, S. Wang, H.M. Ang, M.O. Tade, Catal. Today 174 (2011) 380–385.
19] Y.-F. Huang, Y.-H. Huang, J. Hazard. Mater. 167 (2009) 418–426.
[41] M.-K. Wong, N.-W. Chung, L. He, D. Yang, J. Am. Chem. Soc. 125 (2003) 158–162.
[42] S. Shailaja, M.S. Ramachandran, Int. J. Chem. Kinet. 43 (2011) 620–630.
[43] E.O. Wiig, J. Phys. Chem. 32 (1928) 961–981.
[44] F. Feigl, Spot Tests in Organic Analysis, Elsevier, Amsterdam, 1956, pp. 224.
[45] G. Hilgetag, A. Martini (Eds.), Preparative Organic Chemistry, John-Wiley, New
York, 1972, p. 1013.
20] J. Fernandez, P. Maruthamuthu, J. Kiwi, J. Photochem. Photobiol A: Chem. 161
(
2004) 185–192.
21] J. Fernandez, P. Maruthamuthu, A. Renken, J. Kiwi, J. Appl. Catal. B: Environ. 49
2004) 207–215.
[
[46] J. Buckingham (Ed.), Dictionary of Organic Compounds, Chapman and Hall,
London, 1982, p. 17.
(
[
[
22] X. Chen, X. Qiao, D. Wang, J. Lin, J. Chen, Chemosphere 67 (2007) 802–808.
23] Y.-H. Huang, Y.-F. Huang, C.-I. Huang, C.-Y. Chen, J. Hazard. Mater. 170 (2009)
[47] A.I. Vogel, A Text Book of Quantitative Inorganic Analysis, ELBS Longmans,
London, 1964, pp. 1081.
1
110–1118.
[48] E. Hayon, A. Treinin, J. Wilf, J. Am. Chem. Soc. 94 (1972) 47–57.
[49] G.V. Buxton, C.L. Greenstock, W.P. Helman, A.B. Ross, J. Phys. Chem. Ref. Data
17 (1988) 513–886.
[50] D.L. Ball, J.O. Edwards, J. Am. Chem. Soc. 78 (1956) 1125–1129.
[51] R.W. Hay, K.N. Leong, J. Chem. Soc. A (1971) 3639–3647.
[52] P.G. Daniele, G. Ostacoli, C. Rigano, S. Sammartano, Transit. Met. Chem. 9 (1984)
385–390.
[
[
[
[
24] Z. Wang, R. Yuan, Y. Guo, X. Lei, J. Liu, J. Hazard. Mater. 190 (2011) 1083–1087.
25] R. Yuan, S.N. Ramjuan, Z. Wang, J. Liu, J. Hazard. Mater. 196 (2011) 173–179.
26] M. Murugavelu, M.S. Ramachandran, Transit. Met. Chem. 37 (2012) 687–700.
27] J. March, Advanced Organic Chemistry: Reactions, Mechanisms and Structure,
Wiley-Interscience, Singapore, 1992, pp. 1175.
[
[
28] H.B. Abrahamson, A.B. Rezvani, J.G. Brushmiller, Inorg. Chim. Acta 226 (1994)
1
17–127.
[53] E. Campi, G. Ostacoli, M. Meirone, G. Saini, J. Inorg. Nucl. Chem. 26 (1964)
553–564.
[54] C.F. Timberlake, J. Chem. Soc. (1964) 5078–5085.
[55] T.B. Field, J.L. McCourt, W.A.E. McBryde, Can. J. Chem. 52 (1974) 3119–3124.
[56] A.K. Grzybowski, S.S. Tate, S.P. Datta, J. Soc. Chem. A (1970) 241–245.
[57] J. Strouse, J. Am. Chem. Soc. 99 (1977) 572–580.
29] A.T. Stone, K.L. Godtfredson, B.L. Deng, in: G. Bidoglio, W. Stumm (Eds.), Chem-
istry of Aquatic Systems: Local and Global Perspectives, Kluwer, Dordrecht, The
Netherlands, 1994, pp. 337–374.
30] A.T. Stone, Y. Wang, Geochim. Cosmochim. Acta 70 (2006) 4463–4476.
31] Z. Khan, M. Akram, Ud.D. Kabir, Oxid. Commun. 24 (2001) 257–267.
32] R.A. Larson, A.L. Rockwell, Environ. Sci. Technol. 13 (1979) 325–329.
[
[
[
Please cite this article in press as: G. Ragukumar, et al., J. Mol. Catal. A: Chem. (2014), http://dx.doi.org/10.1016/j.molcata.2014.02.023