Kinetic study of the Ce(III)-, Mn(II)-, or ferroin-catalyzed Belousov-Zhabotinsky reaction with pyruvic acid

Article abstract of DOI:10.1002/(SICI)1097-4601(2000)32:7<408::AID-KIN3>3.0.CO;2-7

Ce(III)-, Mn(II)-, or ferroin (Fe(phen)₃²⁺)-catalyzed reaction of bromate ion and pyruvic acid (PA) or its dimer exhibits oscillatory behavior. Both the open-chain dimer (parapyruvic acid, γ-methyl-γ-hydroxyl-α-keto-glutaric acid, DP

Full text of DOI:10.1002/(SICI)1097-4601(2000)32:7<408::AID-KIN3>3.0.CO;2-7

Kinetic Study of the
Ce(III)-, Mn(II)-, or
Ferroin-Catalyzed
Belousov-Zhabotinsky
Reaction with Pyruvic Acid
HSING-LIEN LIN, YUEH-O YU, JING-JER JWO
Department of Chemistry, National Cheng Kung University Tainan, Taiwan 701, Republic of China
Received 20 September 1999; accepted 3 February 2000
2
ϩ
ABSTRACT: The Ce(III)-, Mn(II)-, or ferroin (Fe(phen)3 )-catalyzed reaction of bromate ion and
pyruvic acid (PA) or its dimer exhibits oscillatory behavior. Both the open-chain dimer (par-
apyruvic acid, ␥-methyl-␥-hydroxyl-␣-keto-glutaric acid, DPA1) and the cyclic-form dimer (␣-
keto-␥-valerolactone-␥-carboxylic acid, DPA2) show more sustained oscillations than PA
monomer. Ferroin behaves differently from Ce(III) or Mn(II) ion in catalyzing these oscillating
systems. The kinetics of reactions of PA, 3-brompyruvic acid (BrPA), DPA1, or DPA2 with Ce(IV),
3ϩ
Mn(III), Fe(phen)3 ion were investigated. The order of relative reactivity of pyruvic acids
3
ϩ
toward reaction with Ce(IV), Mn(III), or Fe(phen)3 ion is DPA2 Ͼ DPA1 Ͼ BrPA Ͼ PA and
3
ϩ
that of metal ions toward reaction with pyruvic acids is Mn(III) Ͼ Ce(IV) Ͼ Fe(phen)3 . The
rates of bromination reactions of pyruvic acids are independent of the concentration of bro-
mine and the order of reactivity toward bromination is (DPA1, DPA2) Ͼ BrPA Ͼ PA. Experi-
mental results are rationalized. ᭧ 2000 John Wiley & Sons, Inc. Int J Chem Kinet 32: 408–418, 2000
INTRODUCTION
idation of citric acid or malonic acid by acidic bro-
mate ion [1]. Zhabotinsky demonstrated that similar
oscillations are observed when the malonic acid is re-
placed by another organic substrate with an active
methylene hydrogen atom or when the cerium couple
is replaced by the Mn(III)/Mn(II) couple, or by the
Theoretical and experimental studies of oscillating re-
actions have been of interest to biologists and chemists
alike. In 1950, Belousov endeavored to model catal-
ysis in the Krebs cycle (citric acid cycle) using cerium
ion instead of the protein-bound metal ions common
in the enzymes of living cells and first observed sus-
tained, barely damped oscillations in the ratio of
[Ce(IV)]/[Ce(III)] during the cerium-ion-catalyzedox-
ϩ/2ϩ
Fe(phen)33
couple [2]. Field, Ko¨ro¨s, and Noyes
(FKN) [3] proposed a detailed mechanism to ration-
alize successfully the cerium-ion-catalyzed bromate
ion-malonic acid oscillating reaction. The FKN mech-
anism is strongly supported by the modeling compu-
tations of Edelson et al. [4] and of Field et al. [5].
Undoubtedly, Belousov-Zhabotinsky (BZ) reaction
is one of the most thoroughly studied and understood
chemical oscillating systems. In the previous work, we
studied the BZ reactions with saccharides, malonic
Correspondence to: Jing-Jer Jwo (jjjwo@mail.ncku.edu.tw)
Contract grant sponsor: National Science Council of the Repub-
lic of China
Contract grant number: NSC 81-0208-M-006-23, NSC 85-2113-
M-006-011, NSC 87-2113-M-006-004
᭧ 2000 John Wiley & Sons, Inc.

KINETIC STUDY OF THE BZ REACTION WITH PYRUVIC ACID
409
acid and its derivatives (methyl-, ethyl-, butyl-,
phenyl-, and dibrom-malonic acids) catalyzed by
Procedures
Kinetic Experiment. The oscillating reaction was fol-
lowed potentiometrically with a bromide-ion-selective
electrode (Orion 94-35) against a double junction ref-
erence electrode (Orion 90-02) [12]. The kinetics of
the oxidation reaction of pyruvic acids by Ce(IV),
2
ϩ
Ce(III), Mn(II), or Fe(phen)3 ion [6–10]. It is well
known that pyruvate is an important metabolite in the
catabolic pathways and involved in a number of en-
zyme-catalyzed intracellular phenomena. In stage I of
catabolism, polysaccharides are degraded to hexoses
or pentoses; lipids are degraded to fatty acids, glycerol,
and other products; and proteins are hydrolyzed to
amino acids. In stage II of catabolism, hexoses, pen-
toses, and glycerol fromstage I are degraded to a sin-
gle 3-carbon pyruvate intermediate, which is con-
verted to acetyl-coenzyme A and CO2 under aerobic
conditions, catalyzed by pyruvate dehydrogenase.
Similarly, the fatty acids and the carbon skeletons of
most of the amino acids are also broken down to form
acetyl-CoA. In stage III of catabolism, acetyl-CoA en-
ters the citric acid cycle (Krebs cycle) and reacts with
oxaloacetate to yield citrate, catalyzed by citrate syn-
thase. The acetyl group of acetyl-CoA is oxidized
completely to CO2 and H2O in the citric cycle. Fur-
thermore, pyruvic acid is also a highly probable inter-
mediate in the BZ reaction with methylmalonic acid
[11]. Therefore, it is worthwhile to study the BZ re-
action with pyruvic acid as the organic substrate.
In this work, we demonstrate that the Ce(III)-,
3
ϩ
Mn(III), and Fe(phen)3 ions were studied spectro-
photometrically by following [Ce(IV)], [Mn(III)], and
2
ϩ
1
[Fe(phen)3
]
at 360 nm( ␧ ϭ 3.18 ϫ 103 MϪ
cmϪ ), 480 nm(110 M cm),
a4n8d0 nm
1
Ϫ1
Ϫ1
(1.05 ϫ 104 MϪ cmϪ), respectively with either a
stopped-flow spectrophotometer (Photol RA-401) or a
conventional spectrophotometer (Hitachi U-2000).
The organic substrate was in great excess stoichio-
metrically over the metal ion oxidant.
1
1
Stoichiometry. It was found that Ce(IV) ion reacts
rapidly with pyruvic acid (PA) to produce quantita-
tively acetic acid and carbon dioxide. The stoichiom-
etry was studied by measuring the concentration of
Ce(IV) ion spectrophotometrically and the amount of
CO2 gravimetrically. The CO2 produced in the reaction
was trapped in two connected flasks containing aque-
ous saturated solution of Ba(OH)2 and precipitated
as BaCO3 . The reliability of this gravimetric method
was supported by a parallel experiment of the reac-
tion of Ce(IV) ion and oxalic acid (OA), which
gives ⌬[CO2]/⌬[OA] ϭ 1.9, consistent with the stoi-
chiometric equation of 2 Ce(IV) ϩ (COOH)2 :
2 Ce(III) ϩ 2 Hϩ ϩ 2 CO2 . For (1.0, 1.5, 2.0, 2.5,
and 3.0) mmol PA and keeping [Ce(IV)]O/[PA]O ϭ
4, the values of ⌬[CO2]/⌬[PA] and ⌬[Ce(IV)]/⌬[PA]
are 1.1 Ϯ 0.2 and 2.2 Ϯ 0.2, respectively. This result
indicates that the stoichiometric equation of the reac-
tion of Ce(IV) ion and PA is consistent with reaction
(S1).
2
Mn(II)-, or Fe(phen)3 ϩ-catalyzed bromate-pyruvic
acid reaction in aqueous H2SO4 exhibits oscillations
in bromide ion concentration. It is observed that the
dimerized pyruvic acid is more reactive than pyruvic
acid monomer and also exhibits better oscillatory be-
havior. The kinetics of the reactions of pyruvic acid
and 3-bromopyruvic acid with Ce(IV), Mn(III), and
3
Fe(phen)3 ϩ ion is investigated. Rationalization of ex-
perimental results is also given.
EXPERIMENTAL
Materials
2 Ce(IV) ϩ CH3COCOOH !: 2 Ce(III)
ϩ 2 Hϩ ϩ CH3COOH ϩ CO2 (S1)
Ammonium cerium(IV) nitrate (99%), ammonium
iron(II) sulfate (99%), cerium(III) nitrate hexahydrate
(99%), and manganese(II) acetate tetrahydrate (99%)
(Merck); pyruvic acid sodiumsalt (99%, Sigma); py-
ruvic acid (98%, Ferak; 98%, Aldrich); 3-bromopy-
ruvic acid (97%, Aldrich) were used in this work.
Other reagents used were of the highest-grade chem-
icals commercially available. Deionized water from
Millipore Milli-RO 20 (reverse osmosis) was used.
Similar methods were applied to study the stoichi-
ometry of the reaction of Ce(IV) ion and 3-bromopy-
ruvic acid (BrPA). For (1.5, 2.0, 2.5, 3.0, 3.5, and 4.0)
mmol of BrPA and keeping [Ce(IV)]O/[BrPA]O ϭ 4,
the average values of ⌬[CO2]/⌬[BrPA] and
⌬[Ce(IV)]/⌬[BrPA] are 0.7 Ϯ 0.2 and 1.9 Ϯ 0.2, re-
spectively. Product analysis by NMR, MS, and poten-
tiometric methods shows that BrCH2COOH is the
main organic product and that side products include
bromide ion, succinic acid ((CH2COOH)2), 2,3-dibro-
mosuccinic acid ((CHBrCOOH)2) (trace amount), and
1,4-dibromo-2,3-butanedione ((COCH2Br)2) (trace
amount). The stoichiometric equation of the main re-
2ϩ
Solutions of Mn(III) [6], Fe(phen)3
[7], and
3
ϩ
Fe(phen)3 [7] ions, and sodiumpyruvate were
freshly prepared just before carrying out the kinetic
run.

410
LIN, YU, AND JWO
action of Ce(IV) ion and BrPA can be expressed by
reaction (S2).
2 CH3COCOOϪ EF
CH3COH(COOϪ)CH2COCOOϪ (R2)
2 Ce(IV) ϩ BrCH2COCOOH !:
CH3COCOOH EF CH2 "COHCOOH (R3)
2 Ce(III) ϩ 2 Hϩ ϩ BrCH2COOH ϩ CO2 (S2)
1
The H NMR spectra of NaPA, DPA1 (Aldrich),
NMR Spectra
and DPA2 (Ferak) are shown in Fig. 2. The spectrum
of dimer-free NaPA in D2O shows two peaks at ␦ ϭ
1.83 and 2.71 ppmwith the area ratio of about 1 : 11,
which changes to 1 : 2.1 at 1.18 and 2.71 ppm, respec-
tively, in D2O containing 1 M D2SO4 (Fig. 2a). The
spectrumof DPA2 (Ferak) in D 2O shows three major
peaks at 1.08, 1.27, and 2.04 ppmwith the area ratios
of 1 : 1.1 : 3.5, which changes to 1 : 0.34 : 0.61 at 1.18,
1.31, and 2.07, respectively, in 1 M D2SO4 (Fig. 2b).
The spectrumof DPA1 (Aldrich) in D 2O shows three
major peaks at 1.31, 1.43, and 2.19 ppm with the area
ratios of 1 : 0.12 : 0.98, which change to 1 : 0.08 : 0.88
at 1.23, 1.36, and 2.12 ppm, respectively, in 1 M
D2SO4 (Fig. 2c). The results above suggest that the
colorless pyruvic acid of Aldrich contains mainly the
parapyruvic acid (DPA1), whereas the orange-colored
pyruvic acid of Ferak contains mainly the cyclic dimer
(DPA2). The 1H NMR spectrumof BrPA in D 2O con-
taining 1 M D2SO4 shows a singlet peak at 3.70 ppm,
which indicates that it is dimer-free and
that the amounts of the diol- and enol-BrPA are negli-
gible.
In aqueous solution, pyruvic acid is hydrated to yield
the geminate diol (␣,␣-dihydroxypropionic acid) (re-
action (R1)). Leussing et al. [13] reported that in 1.25
M PA the proportions were 46% keto-PA and 54%
diol-PA and in 1.0 M NaPA about 3% of the pyruvate
ion (PAϪ) were hydrated.
CH3COCOOH ϩ H2O EF
CH3C(OH)2COOH (R1)
It was reported that PA undergoes slow dimeriza-
tion reaction to yield the open-chain dimer, parapy-
ruvic acid (␥-methyl-␥-hydroxyl-␣-keto-glutaric acid,
DPA1) or the cyclic condensation dimer, ␣-keto-␥-
valerolactone-␥-carboxylic acid, DPA2) (Fig. 1). The
dimerization reaction is catalyzed by acid, base, or
metal ion (Zn(II), Mg(II), Cu(II), or Mn(II)) [13–18].
The equilibriumconstant of the dimerization reaction
of PAϪ ion to produce parapyruvate ion (reaction
(R2)) is 6.5 [15] and that of the keto-enol tautomeri-
zation reaction (reaction (R3)) is about 1 ϫ 10Ϫ3 [19–
20].
RESULTS AND DISCUSSION
The BZ Reaction with Dimer-free
Pyruvic Acid
In a stirred batch experiment, oscillatory behavior was
observed in the reaction of bromate ion with dimer-
free sodiumpyruvate (NaPA) in aqueous H SO4 cat-
2
2
alyzed by Ce(III), Mn(II), or Fe(phen)3 ϩ ion. Typical
oscillatory responses of a bromide-ion selective elec-
trode are shown in Fig. 3a–p. The Ce(III)- or Mn(II)-
catalyzed system may exhibit damped oscillations,
which depend somewhat on the aging of NaPA in
H2SO4 solution (Fig. 3a–c and d–f ). The condition
for the oscillations to occur is quite restricted. It was
observed that the oscillation reaction took place with
concomitant release of CO2(g). In contrast, the
2
Fe(phen)3 ϩ-catalyzed KBrO3-NaPA reaction may ex-
hibit quite sustained oscillations, which take place
without an induction period. The effects of N2(g),
2
NaPA, KBrO3, Fe(phen)3 ϩ ion, and H2SO4 on the os-
cillating pattern are shown in Fig. 3 (h, i–j, k–l, m–
n, and o–p).
Figure 1 Structures of pyruvic acids.

KINETIC STUDY OF THE BZ REACTION WITH PYRUVIC ACID
411
The BZ Reaction with Dimerized
Pyruvic Acid
In contrast to the dimer-free NaPA system, the reac-
tion of bromate ion with dimerized pyruvic acid
(DPA2, Ferak) catalyzed by Ce(III) or Mn(II) ion may
exhibit more sustained oscillations (Fig. 4a–i and Fig.
5a–g) and the condition for generating oscillations is
less restricted. Similar to the NaPA system, the oscil-
lation reaction occurred with concomitant release of
CO2(g) and the pattern of oscillations depends on the
aging of DPA2 in H2SO4 solution (Fig. 4a–c and Fig.
5a–c). Interesting two-stage dual-frequency oscilla-
tions were observed (Fig. 4d–e, in the presence of
BrPA or CH3CHO, and Fig. 5d), due to the presence
of different forms of pyruvic acid. The generation of
Ϫ
HOAc during the BrO3 -DPA2-Ce(III) or Mn(II) BZ
reaction (Fig. 4a and Fig. 5a) was followed by using
1
the H NMR technique. The result indicates that the
[HOAc]/[DPA2] ratio oscillates between 0.2 and 0.4
and between 0.04 and 0.06 for the Ce(III)- and Mn(II)-
catalyzed systems, respectively. The effects of the
concentrations of Ce(III) and Mn(II) ions are shown
in Fig. 4a, f–i) and Fig. 5e–g). The reaction of bro-
mate ion with dimerized pyruvic acid (DPA1, Aldrich)
2
ϩ
catalyzed by Fe(phen)3 or Ce(III) ion may also ex-
hibit oscillations with the characteristics of the oscil-
lation being between those of NaPA and DPA2
(Ferak) systems.
Kinetics and Mechanism
The kinetics of the Ce(IV)-NaPA, Mn(III)-NaPA, and
Mn(III)-BrPA reactions in aqueous H2SO4 were stud-
ied by using a stopped-flow technique. A conventional
spectrophotometric method was applied to study the
3
kinetics of Fe(phen)3 ϩ-NaPA, Ce(IV)-BrPA, and
3
Fe(phen)3 ϩ-BrPA reactions. For the Ce(IV)-NaPA,
Mn(III)-NaPA, or Ce(IV)-BrPA reaction, the observed
pseudo-first-order rate constant (kobs) was calculated
fromthe linear-least-square (LLS) fit of the plot of
ln(At –Aϱ) vs time. For the Fe(phen)33ϩ-NaPA reac-
tion, kobs was obtained fromthe LLS fit of the plot of
ln(Atϩd –At) vs time, where the time interval d is about
half of the time needed to reach the maximum absorb-
ance. Typical plots are shown in Fig. 6A. For the
1
Ϫ1
Mn(III)-NaPA reaction, the plot of kϪobs vs [NaPA]O
is linear (Fig. 6B), which suggests that kobs can be ex-
pressed by kobs ϭ k [PA]/(1 ϩ Km [PA]) (Eq. (1),
shown below). The values of k and Km obtained at
various temperatures for the Mn(III)-NaPA reaction
are shown in Table I. For the Ce(IV)-NaPA, Ce(IV)-
Figure 2 1H NMR spectrumof pyruvic acid in D O con-
taining 1 M D2SO4. (a) NaPA; (b) PA (Ferak); (c) PA (Al-
drich).
2
3
BrPA, Fe(phen)3 ϩ-NaPA, and Fe(phen)33ϩϭ BrPA

412
LIN, YU, AND JWO
Ϫ
Figure 3 Potentiometric traces of log[Br ] vs time for the oscillating reaction of bromate ion and sodium pyruvate catalyzed
by Ce(III), Mn(II), or Fe(phen)3 ion at 25ЊC under aerobic conditions except (h). (a, b, c): [Ce(III)]O ϭ 1.00 ϫ 10 3 M,
2
ϩ
Ϫ
[NaPA]O ϭ 0.300 M, [BrO3 ]O ϭ 0.0300 M, [H2SO4]O ϭ 0.850 M; (d, e, f): [Mn(II)]O ϭ 1.00 ϫ 10 3 M, [NaPA]O ϭ
Ϫ
Ϫ
Ϫ
ϩ
Ϫ
0.300 M, [BrO3 ]O ϭ 0.0300 M, [H2SO4]O ϭ 0.850 M; (g, h): [Fe(phen)32 ]O ϭ 2.00 ϫ 10 3 M, [NaPA]O ϭ 0.300 M,
Ϫ
ϩ
Ϫ
Ϫ
[BrO3 ]O ϭ 0.0500 M, [H2SO4]O ϭ 0.525 M; (i, j): [Fe(phen)32 ]O ϭ 2.00 ϫ 10 3 M, [BrO3 ]O ϭ 0.0500 M, [H2SO4]O ϭ
ϩ
Ϫ
0.525 M, [NaPA]O ϭ (i) 0.200 M, (j) 0.350 M; (k, l): [Fe(phen)32 ]O ϭ 2.00 ϫ 10 3 M, [NaPA]O ϭ 0.300 M, [H2SO4]O ϭ
ϩ
Ϫ
0.525 M, [BrO3 ]O ϭ (k) 0.0400 M, (l) 0.0600 M; (m, n): [NaPA]O ϭ 0.300 M, [BrO3 ]O ϭ 0.0500 M, [H2SO4]O ϭ
2
ϩ
Ϫ
Ϫ
ϩ
Ϫ
0.525 M, Fe(phen)3 ]O ϭ (m) 1.00 ϫ 10 3 M, (n) 3.00 ϫ 10 3 M; (o, p) [Fe(phen)32 ]O ϭ 2.00 ϫ 10 3 M, [NaPA]O ϭ
Ϫ
ϩ
Ϫ
0.300 M, [BrO3 ]O ϭ 0.0500 M, [H2SO4]O ϭ (o) 0.500 M, (p) 0.600 M; (q): [Fe(phen)32 ]O ϭ 2.00 ϫ 10 3 M,
Ϫ
[PA (Aldrich)]O ϭ 0.300 M, [BrO3 ]O ϭ 0.0800 M, [H2SO4]O ϭ 0.550 M. (h): de-areated with N2. The time (tmix) for the aging
of pyruvic acid solution before mixing the reactant solutions: (a, d) 2 min; (c, f) 30 min; (b, e, g–q) 15 min.

KINETIC STUDY OF THE BZ REACTION WITH PYRUVIC ACID
413
Ϫ
Figure 4 Potentiometric traces of log[Br ] vs time for the oscillating reaction of bromate ion and dimerized pyruvic acid
Ϫ
Ϫ
catalyzed by Ce(III) ion in 1 M H2SO4 at 25ЊC. [Ce(III)]O ϭ (a–e, j, k, l) 3.00 ϫ 10 3 M, (f) 1.20 ϫ 10 3 M, (g) 1.90 ϫ
10 3 M, (h) 0.0250 M, (i) 0.0813 M; [BrO3 ]O ϭ (a–i) 0.0670 M, (j) 0.100 M, (k, l) 0.0720 M; [PA (Ferak)]O ϭ (a–i) 0.300
M, (j, k–l) 0 M; [BrPA]O ϭ (d) 0.0150 M, (j) 0.100 M, (a–c, e–i, k–l) 0 M; [PA (Aldrich)]O ϭ (k, l) 0.300 M, (a–j) 0 M;
[CH3CHO]O ϭ (e) 0.0100 M, (a–d, f–l) 0 M; tmix ϭ (a, d–l) 15 min, (b) 75 min, (c) 2 min.
Ϫ
Ϫ
reactions, the plots of kobs vs [PA]O or [BrPA]O are
linear (Fig. 6C–F), which indicates that these reac-
tions follow a second-order kinetics. The values of the
second-order rate constant (k) obtained at various tem-
peratures for these reactions are given in Table I. The
following mechanistic steps (M0–M6) are proposed
to rationalize the kinetic results of the oxidation re-
actions of PA:

414
LIN, YU, AND JWO
Ϫ
Figure 5 Potentiometric traces of log[Br ] vs. time for the oscillating reaction of bromate ion and dimerized pyruvic acid
Ϫ
Ϫ
catalyzed by Mn(II) ion in 1 M H2SO4 at 25ЊC. [Mn(II)]O ϭ (a–c) 5.00 ϫ 10 3 M, (d, f) 1.20 ϫ 10 3 M, (e) 1.20 ϫ
10 4 M, (g) 0.101 M; [BrO3 ]O ϭ (a–c) 0.0670 M, (d) 0.0820 M, (e–g) 0.0300 M; [PA (Ferak)]O ϭ (a–c) 0.300 M, (d)
Ϫ
Ϫ
0.450 M, (e–g) 0.0600 M; tmix ϭ (a, d–g) 15 min, (b) 75 min, (c) 2 min.
k2
ϩ1)ϩ
CH3COCO2H ϩ H2O EF
CH3C(OH)2CO2H
M(n
ϩ CH3C(OH)2CO2H ERF
kϪ
2
K
(M0)
ϩ1)ϩ
[M(n
CH3C(OH)2CO2H] (M2)
k1
ϩ1)ϩ
M(n
ϩ CH3COCO2H ERF
kϪ
1
k1Ј
ϩ1)ϩ
[M(n
ϩ CH3COCO2H] 99:
Mn ϩ CH3CB O ϩ CO2 ϩ Hϩ (M3)
ϩ
ϩ1)ϩ
[M(n
CH3COCO2H] (M1)

KINETIC STUDY OF THE BZ REACTION WITH PYRUVIC ACID
415
3
ϩ
Figure 6 (A) Plots of ln(At–Aϱ) (a, b) or ln(Atϩd–At) (c) vs time for Ce(IV)-NaPA (a), Mn(III)-NaPA (b), and Fe(phen)3
-
NaPA (c) reactions in 1 M H2SO4 and at 25ЊC. (a) [Ce(IV)]O ϭ 2.5 ϫ 10 4 M, [NaPA]O ϭ 0.0100 M, 360 nm; (b)
[Mn(III)]O ϭ 1.10 ϫ 10 3 M, [NaPA]O ϭ 0.0110 M, 480 nm; (c) [Fe(phen)33 ]O ϭ 1.00 ϫ 10 4 M, [NaPA]O ϭ 0.101 M,
480 nm, de-aerated with N2. (B) Plots of kobs vs [NaPA]O for the Mn(III)-NaPA reaction. (C–F) Plots of kobs vs [NaPA]O for
the Ce(IV)-NaPA reaction (C), the Ce(IV)-BrPA reaction (D), the Fe(phen)3 -NaPA reaction (E), and Fe(phen)33 -BrPA
Ϫ
Ϫ
ϩ
Ϫ
Ϫ1
Ϫ1
3
ϩ
ϩ
reaction (F).
(B, C): (a) 22ЊC, (b) 25ЊC, (c) 28ЊC, (d) 32ЊC;
(D): (a) 15ЊC, (b) 18ЊC, (c) 20ЊC, (d) 22ЊC, (e) 25ЊC;
(E): (a) 22ЊC, (b) 25ЊC, (c) 28ЊC, (d) 30ЊC;
(F): (a) 19ЊC, (b) 22ЊC, (c) 25ЊC, (d) 30ЊC.

416
LIN, YU, AND JWO
ϩ
1)
In Eq. (1), [M(n ϩ]t is the total concentration
Table I Rate Constants and Apparent Activation
Energies of the Reactions of Pyruvic Acid and
3-Bromopyruvic acid with Ce(IV), Mn(III), and
of M(n
[CH3C(OH)2CO2H], Km ϭ1k /(kϪ1 1 ϩ kЈ), Km
species, [PA]t ϭ [CH3COCO2H] ϩ
ϩ1)ϩ
ϭ
1
2
3
ϩ
Fe(phen)3 Ions
k2/(kϪ ϩ kЈ), k ϭ (k1ЈKm ϩ kЈK K )/(1 mϩ K), K
2
2
1
2
2
m
ϭ (Km1 ϩ K Km2)/(1 ϩ K), K ϭ [CH3C(OH)2CO2H]/
[CH3COCO2H], and kobs ϭ k [PA]t /(1 ϩ Km [PA]t).
This rate equation is similar to the Michaelis-Menten-
like kinetics as suggested by Kasperek and Bruice
[21]. The kinetics of the Mn(III)-NaPA reaction is
consistent with rate equation of Eq. (1). At sufficiently
Ce(IV)-PA reaction
T/ЊC
Ce(IV)-BrPA reaction
T/ЊC
Ϫ
Ϫ
1
Ϫ
Ϫ1
k/M 1 s
k/M 1 s
22
25
28
32
3.79
4.11
5.29
7.75
15
18
20
22
25
3.88
5.28
6.39
8.08
10.7
low [PA] (Km [PA]t ϽϽ 1), kobs reduces to kobs
ϭ
Ϫ
1
Ea/kJ mol
55.1 Ϯ 8.7
72.1 Ϯ 1.3
k [PA]t and a second-order kinetics is obtained as ob-
3
served in the Ce(IV)-NaPA and Fe(phen)3 ϩ-NaPA re-
3
ϩ
ϩ
Fe(phen)3 -PA reaction
Fe(phen)33 -BrPA reaction
actions. As reported by Zuman et al. [22], the pH-
independent rate constants of the solvent-catalyzed
Ϫ
Ϫ
1
Ϫ
Ϫ1
T/ЊC
k/M 1 s
T/ЊC
k/M 1 s
1
hydration and dehydration at 25ЊC are 0.27 sϪ and
22
25
28
4.99
5.85
9.10
12.8
19
22
25
30
4.62
6.22
7.55
19.1
1
0.092 sϪ , respectively. The equilibriumof step (M0)
can be maintained under the present reaction condi-
ϩ
1)
tions, since [PA]O ϾϾ [M(n ϩ]O .
30
Ϫ
1
Ea/kJ mol
88.5 Ϯ 14
72.3 Ϯ 0.8
Since the diol formof BrPA is negligible, the
mechanism of the oxidation reactions of BrPA consists
of steps similar to (M1), (M3), and (M5), which is
consistent with the stoichiometric equation (S2) for the
Ce(IV)-BrPA reaction. The derived rate law is shown
in Eq. (2), where kobs ϭ kЈ1 Km1 [BrPA]/(1 ϩ Km1
[BrPA]).
Mn(III)-PA reaction
Mn(III)-BrPA
reaction
Ϫ
Ϫ
1
Ϫ
1
Ϫ1
T/ЊC
k/M 1 s
Km/M
T/ЊC
k1Ј/s
22
25
28
49.4
63.4
83.2
150
20.1
15.6
14.4
19.6
22
25
27
30
0.478
0.683
0.818
32
1.08
ϩ
1)
Ϫd[M(n ϩ]/dt
Ϫ
1
Ea/kJ mol
82.7 Ϯ 9.3
75.1 Ϯ 3.3
ϩ1)ϩ
ϭ (k1Ј Km [BrPA]/(1 ϩ K [Br1PA])) [M(n
]
t
(2)
1
m
ϩ1)ϩ
ϭ kobs [M(n
]
t
k2Ј
At sufficiently low [BrPA] (Km1 [BrPA] ϽϽ 1), kobs re-
duces to kobs ϭ k[BrPA] (k ϭ k1Ј Km1) and a second-
order kinetics is obtained, as observed in the Ce(IV)-
ϩ1)ϩ
[M(n
ϩ CH3C(OH)2CO2H] 99:
Mn ϩ CH3CB (OH)2 ϩ CO2 ϩ Hϩ (M4)
ϩ
3
BrPA and Fe(phen)3 ϩ-BrPA reactions. However, the
ϩ1)ϩ
M(n
M(n
ϩ CH3CB O ϩ H2O !:
1
values of (kobs/sϪ ) of the Mn(III)-BrPA reaction are
Mn ϩ CH3CO2H ϩ Hϩ (M5)
ϩ
(0.474, 0.476, 0.478, 0.478, and 0.485) at 22ЊC;
(0.689, 0.685, 0.681, 0.682, and 0.679) at 25ЊC;
(0.811, 0.822, 0.822, 0.818, —) at 27ЊC; and (1.08,
1.08, 1.07, 1.09, —) at 30ЊC for [BrPA]O ϭ (2.20,
4.41, 6.60, 8.79, and 11.0) ϫ 10Ϫ2 M, respectively. It
is interesting to observe that for this reaction the value
of kobs is independent of the concentration of BrPA.
This result can be rationalized by invoking Eq. (2). At
sufficiently high [BrPA] (Km1 [BrPA] ϾϾ 1), kobs be-
ϩ1)ϩ
ϩ CH3CB (OH)2 !:
Mn ϩ CH3CO2H ϩ Hϩ (M6)
ϩ
ϩ
1)ϩ
ϩ
(M(n
ϭ Ce(IV), Mn(III), or Fe(phen)33
)
This mechanism is consistent with the stoichio-
metric equation (S1) for the Ce(IV)-PA reaction. If
initially [PA]O ϾϾ [M(n ϩ]O , the rate law (Eq. (1))
can be derived by applying the steady-state approxi-
mation to two metal-substrate complexes.
ϩ
1)
comes k1Ј. The values of kЈ at various temperatures are
1
given in Table I. The apparent activation energies ob-
tained fromthe LLS fits of the Arrhenius plots of the
preceding reactions are also shown in Table I.
The stopped-flow technique is applied to study the
kinetics of the reactions of the dimerized pyruvic acid
(DPA1 and DPA2) with Ce(IV) and Mn(III) ions,
whereas the conventional spectrophotometric method
is used for the reaction of DPA1 and DPA2 with
ϩ
1)
Ϫd[M(n ϩ]/dt
((kЈ Km ϩ kЈ K K )/(1 2ϩ K)) [PA]t
1
1
2
m
ϩ1)ϩ
ϭ
[M(n
]
t
1 ϩ ((Km ϩ K K )/(1 ϩ K)) [PA]
2 t
ϭ (k [PA]t /(1 ϩ Km [PA]t)) [M(n
1
m
ϩ1)ϩ
]
t
ϭ kobs [M(n
]
t
(1)
ϩ1)ϩ

KINETIC STUDY OF THE BZ REACTION WITH PYRUVIC ACID
417
3
ϩ
Fe(phen)3 ion. It was observed that the kinetics of
these reactions were complicated by the compositions
of the solutions of DPA1 and DPA2. However, a def-
inite conclusion can be deduced fromparallel experi-
ments. The results indicate that the order of relative
reactivity of pyruvic acids toward reaction with
Ϯ 27.6 J/(mol K ) for the enolization reactions of PA
and BrPA, respectively. The negative value of ⌬Sꢀ
can be rationalized by invoking the solvent reorgani-
zation due to the more ionic transition state of the en-
olization reaction. The bromination reaction of PA
produces BrPA first and then Br2PA (Br2CHCO-
COOH). No significant amount of Br3PA (CBr3
COCOOH) was observed. When a sufficient amount
of bromine is present, Br2CHCOOH is also produced.
A similar result was observed for the bromination re-
action of BrPA. A parallel study indicates that the rate
of the bromination reaction of the dimerized pyruvic
acid (DPA1 and DPA2) is considerably faster than that
of PA, consistent with the result mentioned before that
DPA2 and DPA1 are much more reactive than PA
3
ϩ
Ce(IV), Mn(III), or Fe(phen)3
ion is DPA2 Ͼ
DPA1 Ͼ BrPA Ͼ PA and that of metal ions toward
reaction with these pyruvic acids is Mn(III) Ͼ
Ce(IV) Ͼ Fe(phen)3 ϩ. These reactions are facilitated
by the complex formation of metal-ion oxidant and
pyruvic acids. As suggested by Leussing et al. [15],
the predominant mode of chelation involves the hy-
droxyl and carbonyl groups attached to the asymmetric
carbon of pyruvic acid.
3
3
ϩ
toward reaction with Ce(IV), Mn(III), or Fe(phen)3
ion.
Bromination Reactions of Pyruvic and
Bromopyruvic Acids
Implication
The kinetics of the bromination reactions of PA and
BrPA was studied by following [Br2] spectrophoto-
According to a classification by Noyes [26], the “clas-
sical” BZ reaction is a metal-ion catalyzed oxidation
and bromination of an organic substrate by acidic bro-
mate. In the prototype BZ reaction, malonic acid is
reactive toward oxidation by the oxidized formof the
metal-ion catalyst and also reactive toward bromina-
tion. The brominated product of malonic acid, bro-
momalonic acid, is capable of reacting with the metal-
ion catalyst and liberating the bromide ion. In this
work, we demonstrate that the BZ system of the di-
merized pyruvic acids exhibits more sustained oscil-
lations than that of the pyruvic acid monomer. This
observation can be rationalized by invoking that the
dimerized pyruvic acids are more reactive toward bro-
mination or oxidation by Ce(IV), Mn(III), or
1
metrically at 400 nm( ␧ ϭ 167 MϪ cmϪ) 1 with PA
and BrPA in great excess. It was found that these re-
actions are zero-order with respect to bromine
4
((1.2–7.2) ϫ 10Ϫ M) and first-order with respect to
PA or BrPA. This result is consistent with the well-
known observation that halogenation of acetone in
aqueous solution is zero-order with respect to halogen
and the enolization of acetone is the rate-determining
step when the concentration of halogen is sufficiently
high [23–25]. The values of the first-order rate con-
stant (ken) of the bromination or enolization reaction
of PA in 1 M H2SO4 are (1.55 Ϯ 0.07, 2.55 Ϯ 0.17,
1
3.10 Ϯ 0.10, and 4.11 Ϯ 0.21) ϫ 10Ϫ6 sϪ at 22, 25,
28, and 31ЊC, respectively. The value of ken is con-
3
ϩ
sistent with that reported by Burgner et al. (about
Fe(phen)3 ion, as mentioned earlier. The higher re-
activity of dimerized pyruvic acids also supports that
parapyruvate blocks the citric acid cycle by specifi-
cally inhibiting the oxidative decarboxylation of ␣-
ketoglutarate to succinate [27–28]. Similar to the BZ
1
4 ϫ 10Ϫ6 sϪ ) [19]. The values of ken of BrPA in
1 M H2SO4 are (0.70 Ϯ 0.03, 1.02 Ϯ 0.02, 1.46 Ϯ
1
0.02, and 1.71 Ϯ 0.05) ϫ 10Ϫ5 sϪ at 20, 23, 25, and
27ЊC, respectively. The kinetics of the bromination re-
action of BrPA was also studied by following the pro-
duction of bromide ion potentiometrically. It was
found that the rate of the bromide ion production is
also zero-order with respect to bromine and is first-
order with respect to BrPA. The value of ken at 25ЊC
2ϩ
reactions of malonic acids [7,9–10], Fe(phen)3 ion
behaves differently fromCe(III) or Mn(II) ion as a
catalyst for the BZ reactions of pyruvic acids. This
result can be rationalized by invoking the differ-
3ϩ/2ϩ
ent reactivity of Fe(phen)3
couple fromCe(IV)/
1
is (1.78 Ϯ 0.13) ϫ 10Ϫ5 sϪ , which is consistent with
Ce(III) or Mn(III)/Mn(II) couple toward reaction with
5
1
Ϫ
that of 1.46 ϫ 10Ϫ sϪ of the spectrophotometric
method mentioned above. The activation energies ob-
tained fromthe LLS fits of the Arrhenius plots of ken
are 77 Ϯ 11 and 89.4 Ϯ 9.6 kJ/mol for the enolization
reactions of PA and BrPA, respectively. The activation
parameters (⌬Hꢀ and ⌬Sꢀ) obtained from the LLS
fits of the Eyring plots are 75.3 Ϯ 10.6 kJ/mol,
Ϫ100 Ϯ 35 J/(mol K) and 90.1 Ϯ 8.2 kJ/mol, Ϫ35.8
pyruvic acids and with bromine species (BrO3 ,
HBrO2 , BrO2 · , HOBr, and Br2) [29].
SUMMARY
In a stirred batch experiment, the reaction of bromate
ion with pyruvic acid monomer (PA) or dimer (DPA)

418
LIN, YU, AND JWO
2ϩ
8. Sun, S. S.; Lin, H.-P.; Chen, Y.-F.; Jwo, J.-J. J Chinese
ChemSoc 1994, 41, 651.
catalyzed by Ce(III), Mn(II), or Fe(phen)3 ion in
aqueous H2SO4 exhibits damped oscillations. Both the
open-chain (DPA1) and the cyclic-form(DPA2) di-
mers show more sustained oscillations than PA mono-
mer. The characteristics of the oscillations depend on
the concentrations of organic substrate, bromate ion,
9. Lin, H.-P. Jwo, J.-J. J Phys Chem1995, 99, 6897.
10. Chen, Y.-F.; Lin, H.-P.; Sun, S. S.; Jwo, J.-J. Int J Chem
Kinet 1996, 28, 345.
11. Ruoff, P.; Nevdal, G. J Phys Chem1989, 93, 7802.
12. Jwo, J.-J.; Chang, E.-F. J Phys Chem1989, 93, 2388.
13. Leussing, D. L.; Stanfield, C. K. J AmChemSoc 1964,
86, 2805.
14. Wolff, L. Ann Chem1899, 305, 154.
15. Tallman, D. E.; Leussing, D. L. J Am Chem Soc 1969,
91, 6253.
16. Tallman, D. E.; Leussing, D. L. J Am Chem Soc 1969,
91, 6256.
17. Raghavan, N. V.; Leussing, D. L. J Indian ChemSoc
1977, LIV, 68.
18. Cheong, M.; Leussing, D. L. J AmChemSoc 1989, 111,
2541.
19. Burgner, J. W.; Ray, W. J., Jr. Biochemistry 1978, 17,
1654.
20. Kuo, D. J.; O’Connell, E. L.; Rose, I. A. J AmSoc Chem
1979, 101, 5025.
2
ϩ
metal-ion catalyst, and sulfuric acid. Fe(phen)3 ion
behaves differently fromCe(III) and Mn(II) ions in
catalyzing these oscillating reactions. The results of
the kinetic study indicate that the order of relative re-
activity of pyruvic acids toward reaction with Ce(IV),
3
ϩ
Mn(III), or Fe(phen)3 ion is DPA2 Ͼ DPA1 Ͼ
3-bromopyruvic acid (BrPA) Ͼ PA and that of metal
ions toward reaction with pyruvic acids in Mn(III) Ͼ
3
Ce(IV) Ͼ Fe(phen)3 ϩ. The rates of bromination re-
actions of pyruvic acids are independent of the con-
centration of bromine and the order of reactivity
toward bromination is (DPA1, DPA2) Ͼ BrPA Ͼ
PA.
21. Kasperek, G. J.; Bruice, T. C. Inorg Chem1971, 10,
382.
BIBLIOGRAPHY
22. Kozlowski, J.; Zuman, P. J Electroanal Chem 1987,
226, 69.
23. Bell, R. P.; Yates, K. J ChemSoc 1962, 1927.
24. Bell, R. P.; Davis, G. G. J ChemSoc 1964, 902.
25. Albery, W. J.; Gells, J. S. J ChemSoc, Faraday Trans
1 1982, 78, 1569.
1. Belousov, B. P. Sb Ref Radiat Med 1958, 145.
2. Zhabotinsky, A. M. Dokl Akad Nauk SSSR 1964, 157,
392.
3. Field, R. J.; Ko¨ro¨s, E.; Noyes, R. M. J AmChemSoc
1972, 94, 8649.
4. Edelson, D.; Noyes, R. M.; Field, R. J. Int J ChemKinet
1979, 11, 155.
5. Gyo¨rgyi, L.; Tura´ny, T.; Field, R. J. J Phys Chem1990,
94, 7162.
6. Ou, C.-C.; Jwo, J.-J. Int J ChemKinet 1991, 23, 137.
7. Chou, Y.-C.; Lin, H.-P.; Sun, S. S.; Jwo, J.-J. J Phys
Chem1993, 97, 8450.
26. Noyes, R. M. J AmChemSoc 1980, 102, 4644.
27. Montgomery, C. M.; Wedd, J. L. J Biol Chem 1956,
221, 359.
28. Montgomery, C. M.; Fairhurst, A. S.; Webb, J. L. J Biol
Chem1956, 221, 369.
29. Ke´ki, S.; Magyar, I.; Beck, M. T.; Ga´spa´r, V. J Phys
Chem1992, 96, 1725.