Kinetics and mechanism of the oxidation of some α-hydroxy acids by tetrachloroaurate(III) in acetic acid-sodium acetate buffer medium

Article abstract of DOI:10.1002/(SICI)1097-4601(1999)31:12<873::AID-KIN6>3.0.CO;2-Z

The kinetics of oxidation of some neutralized α-hydroxy compounds such as glycolic (GA) lactic (LA), α-hydroxyisobutyric(IB), mandelic (MA), atrolactic (AL), and benzilic (BA) acids by tetrachloroaurate(III) have been studied. The substrates are oxidized to give formaldehyde, acetaldehyde, acetone, benzaldehyde, acetophenone, and benzophenone for the respective reactions. The rate of the reaction increases with increasing |substrate| and pH but decreases with increase in [Cl^-]. Temperature influence is quite marked in all these reactions. A mechanism involving the formation of an unstable complex, which decomposes to give the respective reaction products, is proposed. The reactivity of the α-hydroxy acids towards gold(III) are as follows: AL>MA>BA>IB>LA>GA.

Full text of DOI:10.1002/(SICI)1097-4601(1999)31:12<873::AID-KIN6>3.0.CO;2-Z

Kinetics and Mechanism
of the Oxidation of Some
␣-Hydroxy Acids by
Tetrachloroaurate(III)
in Acetic Acid-Sodium
Acetate Buffer Medium
KALYAN K. SEN GUPTA, BISWAJIT PAL, PRATIK K. SEN
Department of Chemistry, Jadavpur University, Calcutta-700 032, India
Received 3 December 1998; accepted 9 August 1999
ABSTRACT: The kinetics of oxidation of some neutralized ␣-hydroxy compounds such as gly-
colic (GA), lactic (LA), ␣-hydroxyisobutyric(IB), mandelic (MA), atrolactic (AL), and benzilic
(BA) acids by tetrachloroaurate(III) have been studied. The substrates are oxidized to give
formaldehyde, acetaldehyde, acetone, benzaldehyde, acetophenone, and benzophenone for
the respective reactions. The rate of the reaction increases with increasing [substrate] and pH
Ϫ
but decreases with increase in [Cl ]. Temperature influence is quite marked in all these re-
actions. A mechanism involving the formation of an unstable complex, which decomposes to
give the respective reaction products, is proposed. The reactivity of the ␣-hydroxy acids to-
wards gold(III) are as follows: AL Ͼ MA Ͼ BA Ͼ IB Ͼ LA Ͼ GA. ᭧ 1999 John Wiley & Sons, Inc.
Int J Chem Kinet 31: 873–882, 1999
INTRODUCTION
in the presence of a sodium acetate-acetic acid buffer
medium. Au(III) as an oxidant is of interest since it
may behave as a one equivalent or a two equivalent
oxidant in the reaction with ␣-hydroxy acids. The
present report deals with the kinetics and mechanism
of oxidation of some ␣-hydroxy carboxylates by te-
trachloroaurate(III) under different experimental con-
ditions.
The kinetics of oxidation of some ␣-hydroxy acids by
some oxidants have been published [1–12]. There are
no literature data involving the oxidation of ␣-hydroxy
carboxylates by gold(III) although the oxidations of a
number of inorganic reductants by gold(III) [13–15]
have been studied. Since halide of gold(III)
are partially hydrolysed [16] at lower acidities
7
3
(1.1–12.6)ϫ10Ϫ mol dmϪ and the reactions are
EXPERIMENTAL
Reagents
too slow to be monitored in a high acid medium
3
(Ͼ10Ϫ3 mol dmϪ ), the reactions have been carried out
Mandelic (BDH), benzilic (E. Merck, AG), ␣-hydroxy
Correspondence to: K. K. Sen Gupta
᭧ 1999 John Wiley & Sons, Inc.
CCC 0538-8066/99/120873-10
isobutyric (Fluka, AG), lactic (E. Merck), glycolic

874
SEN GUPTA, PAL, AND SEN
(E. Merck), and atrolactic (Aldrich Chem) acids were
used without further purification and their neutral salts
were prepared by the addition of the requisite amount
of NaOH. ␣-Deutero mandelic acid was prepared fol-
lowing the procedure of Jones, Waters and Littler [17]
by equilibrating three times of sodium mandelate in a
sealed tube at 150ЊC for 7 days with sodium hydroxide
(5%) in deuterium oxide (99.8%). Labile hydrogen
was then removed by recrystallizing the free acid twice
from water. (m.p. 121ЊC). Analysis showed that 98.5%
deuteration had been achieved. A stock of gold(III)
was prepared dissolving HAuCl4 (Johnson Matthey)
in 0.1 mol dmϪ3 HCl. The strength of the solution was
estimated gravimetrically [18]. To a known volume
(5 ml) of the standard gold solution in a 150 ml
beaker, was added dropwise with stirring, a 1N solu-
tion of KOH(A.R.) until the yellow color was just dis-
charged. A further amount of 2 ml KOH solution was
added, followed by dilution to about 75 ml with dis-
tilled water. 5–6 ml of 1N oxalic acid solution was
then added dropwise with stirring and the contents
were kept on a boiling water bath. The solution turned
purple with a violet tinge when gold separated out. The
beaker was kept on the water bath for about an hour
with occasional stirring. The solution was filtered
through a Whatman No. 42 (7 cm) filter paper. The
precipitate was washed with about 75 ml of distilled
water, dried, ignited in a silica crucible and weighed
as metal. The gold(III) solution after estimation was
stored in the dark and used under subdued lighting
condition and was diluted to desired concentrations
before kinetic experiments.
netics were followed in the visible region at ␭ ϭ
400 nm using higher gold(III) concentration, in order
to eliminate the possibility of the formation of col-
loidal gold. All of the kinetic investigations were car-
ried out in a Systronics (India) UV-visible spectro-
photometer using a thermostatted cell of 1 cm path
length. The water was circulated from the bath main-
tained at the required temperature (Ϯ0.1ЊC). At least
8–10 experimental readings were taken depending
upon the temperature of the experiment. The pseudo-
first-order rate constant (kobs) was determined from the
linear plots of log A (A ϭ absorbance) against time
and the linearity was observed up to two half-lives.
The kobs values were reproducible to within Ϯ4%.
Product Analysis
The product analysis of the oxidations of ␣-hydroxy
carboxylate by tetrachloroaurate(III) were carried
out with the reaction mixture: [␣-hydroxy
1
3
carboxylate] ϭ 3 ϫ 10Ϫ mol dmϪ , [AuIII] ϭ 2.2 ϫ
3
3
10Ϫ mol dmϪ and pH ϭ 4.05. The reaction mixture
(30 ml) for each substrate was kept at 30ЊC for 120
min in a stoppered bottle since the b.p.s. of formal-
dehyde and acetaldehyde are low. It was then filtered
to remove suspended particles, if any. In one part of
the filtrate the oxidation products were tested by the
formation of colors with different reagents in Table I
[21–23]. The other part of the filtrate was acidified
with dilute H2SO4 and added to 2,4-dinitrophenyl hy-
drazine hydrochloride solution, heated on a steam bath
for 15 mins and left at room temperature for an hour
when the yellow crystals of 2,4-dinitrophenyl hydra-
zone derivative of the product was obtained. The crys-
tals were filtered, washed with water and dried. The
crude derivative was chromatographed over neutral
alumina (Brockman) and eluted with dry benzene. The
purified 2,4-DNP derivative thus obtained, was crys-
tallized from ethanol, filtered and dried to record the
yield and m.p. of the sample. The melting points were
checked against the literature values [24]. The results
are furnished in Table I. The oxidation occurs accord-
ing to the equation.
Inorganic materials were of the highest available
purity. All solutions were made in doubly distilled wa-
ter. The oxidation studies were carried out in a
NaOAc-HOAc buffer. Buffer solutions were prepared
[19] from standard solution of sodium acetate and ace-
tic acid. The pH of the solution was checked against
standard buffer solution with a pH meter (Elico India
LI 120).
Kinetics and Measurements
The reaction rate was determined spectrophotometri-
cally using a sodium acetate-acetic acid buffer solution
under the condition where [Na-salt of ␣-hydroxy
acids] ϾϾ [AuIII]. Gold(III) absorbs maximum at ␭ ϭ
313 nm. However, according to a number of authors,
there is a possibility of formation of colloidal gold [20]
when aqueous solutions of AuCl3 or HAuCl4 are ex-
posed to UV light. The formation of colloidal gold
upon exposure to UV light may be enhanced in the
presence of reducing species. Consequently, the ki-
R
OH
Ϫ
C
R
ϩ AuCl4
RЈ
COOH
C"O ϩ CO2 ϩ AuCl2Ϫ ϩ 2Hϩ ϩ 2ClϪ (1)
RЈ

OXIDATION OF SOME ␣-HYDROXY ACIDS
875
Table I Identification of Products of Reactions
Lit. mp.(ЊC)[24]
of
% yield
mp(ЊC) of
(2,4-DNP 2,4-DNP
derivative) derivative
2,4-DNP
derivative
Substrate
Product
CH2O
Test
Color
Glycolic acid
Chromotropic acid
and conc.H2SO4
Equal volumes of 20% Blue [22]
piperidine and 5%
sodium nitroprus-
side
Pinkish violet [21]
88
89
163–165
147–148
166–167
148
Lactic acid
MeCHO
␣-Hydroxyisobutyric Me2CO
Equal volumes of 5%
Red [23]
80
127–128
126–128
acid
sodium nitroprus-
side and 30%
NaOH
—
Mandelic acid
Atrolactic acid
Benzilic acid
PhCHO
PhCOCH3
Ph2CO
—
—
—
94
85
91
233–235
239–240
237–239
235–237
238–240
238–239
—
—
where,
different experiments was 1.00 Ϯ 0.01. The reaction
stoichiometry may be expressed as [␣-hydroxy acid]:
[AuIII] ϭ 1 : 1.
R ϭ RЈ ϭ 9H for glycolic acid
R ϭ 9CH3, RЈ ϭ 9H for lactic acid
R ϭ RЈ ϭ 9CH3 for ␣-hydroxyisobutyric acid
R ϭ 9C6H5, RЈ ϭ 9H for mandelic acid
R ϭ 9C6H5, RЈ ϭ 9CH3 for atrolactic acid
R ϭ RЈ ϭ 9C6H5 for benzilic acid
Effect of Reactant Concentrations
The reactions were studied at different concentrations
3
3
3
of Au(III) of 1.0 ϫ 10Ϫ , 2.0 ϫ 10Ϫ , 3.0 ϫ 10Ϫ ,
4.0 ϫ 10Ϫ , 5.0 ϫ 10Ϫ , and 6.2 ϫ 10Ϫ mol dmϪ
3
3
3
3
for all the substrates but at constant [␣-hydroxy acid],
1
5
[Hϩ], [ClϪ], and temperature of 2 ϫ 10Ϫ , 8.9 ϫ 10Ϫ ,
3 ϫ 10Ϫ mol dmϪ , and 313K, respectively. The
pseudo-first-order rate constant was found to be in-
dependent of the initial [AuIII] in each case (Table II).
The reactions were also studied at four different tem-
peratures with variation of [␣-hydroxy acid] from
2 ϫ 10Ϫ1 to 7 ϫ 10Ϫ1 mol dmϪ3 keeping [AuIII], [Hϩ],
2
3
Test for Free Radicals
Acrylonitrile [50% (v/v)] was added to the reaction
mixture during the course of the reactions. No precip-
itation of white polyacrylonitrile in the absence or
presence of methanol was formed (during the oxida-
tion of ␣-hydroxy carboxylate by gold(III)). The result
is in conformity with a one-step two-electron transfer
process with no free radical intermediate.
and [ClϪ] constant at 2.2 ϫ 10Ϫ , 8.9 ϫ 10Ϫ , and
3
5
Table II Effect of [AuIII] on the Rate of Oxidation at
Ϫ
Ϫ
313K. [AuIII] ϭ (1.0–6.2) ϫ 10 3 mol dm 3, [Substrate]
ϭ 2 ϫ 10 1 mol dm 3, [H ] ϭ 8.9 ϫ 10 5 mol dm
,
Ϫ
Ϫ
ϩ
Ϫ
Ϫ3
RESULT AND DISCUSSION
Stoichiometry
Ϫ
Ϫ
Ϫ3
[Cl ] ϭ 3 ϫ 10 2 mol dm
Ϫ
1
Substrate
kobs ϫ 103 (s
)
Glycolic acid
Lactic acid
␣-hydroxy isobutyric acid
Mandelic acid
Atrolactic acid
Benzilic acid
0.58 Ϯ 0.01
0.75 Ϯ 0.01
0.89 Ϯ 0.02
1.14 Ϯ 0.01
1.45 Ϯ 0.02
1.06 Ϯ 0.02
The stoichiometry of the reaction was studied
under different experimental conditions. The reaction
mixture containing excess [AuIII] as compared to
[␣-hydroxy acid], was kept for 24h and the uncon-
sumed [AuIII] was determined by the method men-
tioned above [18]. The average stoichiometry for three

876
SEN GUPTA, PAL, AND SEN
Table III Variation of Pseudo-First-Order Rate Constants for the Oxidation of Substrates by AuIII at Different
Ϫ
3
Ϫ
ϩ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ3
Temperatures. [AuIII] ϭ 2.2 ϫ 10 mol dm 3, [H ] ϭ 8.9 ϫ 10 5 mol dm 3, [Cl ] ϭ 3 ϫ 10 2 mol dm
Temp
(K)
[S] ϫ 102
(mol dm
kaobs ϫ 103
kbobs ϫ 103
kcobs ϫ 103
kdobs ϫ 103
keobs ϫ 103
kfobs ϫ 103
Ϫ
3
Ϫ
1
Ϫ
1
Ϫ
1
Ϫ
1
Ϫ
1
Ϫ1
)
(s
)
(s
)
(s
)
(s
)
(s
)
(s )
20
30
40
50
60
70
0.219
0.297
0.36
0.41
0.48
0.53
0.275
0.375
0.44
0.5
0.566
0.625
0.298
0.4
0.484
0.545
0.61
0.68
0.337
0.468
0.566
0.67
0.789
0.909
0.416
0.577
0.697
0.81
0.937
1.03
0.306
0.428
0.517
0.567
0.66
298
303
308
313
318
0.75
20
30
40
50
60
70
0.31
0.41
0.519
0.602
0.65
0.75
0.38
0.53
0.64
0.716
0.815
0.91
0.429
0.587
0.689
0.813
0.915
0.96
0.538
0.752
0.938
1.12
1.212
1.377
0.6
0.455
0.627
0.716
0.835
0.938
1.08
0.882
1.034
1.20
1.43
1.58
20
30
40
50
60
70
0.44
0.612
0.699
0.83
0.91
1.0
0.547
0.75
0.9
1.04
1.21
1.31
0.59
0.79
0.94
1.08
1.16
1.37
0.735
1.046
1.32
1.51
1.78
0.857
1.25
1.5
1.66
2.0
0.733
1.01
1.16
1.44
1.594
1.78
1.892
2.31
20
30
40
50
60
70
0.59
0.81
1.04
1.08
1.26
1.31
0.77
1.05
1.26
1.37
1.58
1.68
0.88
1.16
1.51
2.04
2.34
2.53
2.75
1.428
1.765
2.5
2.73
3.33
3.75
1.08
1.46
1.68
1.887
2.16
2.34
1.216
1.374
1.584
1.78
2.02
20
30
40
50
60
70
0.86
1.12
1.38
1.51
1.59
1.89
1.12
1.44
1.887
2.02
2.33
2.52
1.37
1.894
2.34
2.517
2.75
3.03
1.78
2.34
3.02
3.37
3.77
4.34
2.5
1.57
2.165
2.52
3.036
3.36
3.76
3.33
4.285
5.2
6.0
7.5
a ϭ GA, b ϭ LA, c ϭ IB, d ϭ MA, e ϭ AL, f ϭ BA
3
3 ϫ 10Ϫ2 mol dmϪ , respectively. The results indicate
mentioned in Table II have been compared. The values
3
3
3
that the rate increases with increasing [substrate]
(Table III). Plots of 1/kobs versus 1/[substrate] gave
straight lines with positive intercepts on the y-axis
and positive slopes at four different temperatures (Fig-
ure 1).
are 0.59 ϫ 10Ϫ , 0.76 ϫ 10Ϫ , 0.90 ϫ 10Ϫ , 1.14 ϫ
3
3
3
1
10Ϫ , 1.46 ϫ 10Ϫ and 1.08 ϫ 10Ϫ sϪ for the oxida-
tions of GA, LA, IB, MA, AL and BA, respectively.
The values of kobs was found to increase with an in-
crease in pH. The plot of 1/kobs vs {K1 ϩ [Hϩ]} gave
a straight line with a positive slope and a positive in-
tercept on the ordinate (Figure 2).
؉
Effect of [H ]
؊
Effect of [Cl ]
The rate of the reaction was studied at 313K but at
different pH values (4.05–5.89) using acetic acid-
acetate buffer while keeping [AuIII], [␣-hydroxy acid],
The reaction was studied at 313K but at different [ClϪ]
2
3
[(2–9) ϫ 10Ϫ mol dmϪ ] varied by the addition of
[ClϪ] and ionic strength constant at 2.2 ϫ 10Ϫ , 2 ϫ
NaCl but at constant [AuIII], [␣-hydroxy acid], [Hϩ]
3
1
2
3
3
1
10Ϫ , 3 ϫ 10Ϫ , and 0.3 mol dmϪ , respectively. The
pseudo-first-order rate constant at constant concentra-
tions of Au(III), substrate, Hϩ, ClϪ and temperature as
and ionic strength of 2.2 ϫ 10Ϫ , 2 ϫ 10Ϫ , 8.9 ϫ
10Ϫ , and 0.4 mol dmϪ , respectively. The value of kobs
5
3
was found to decrease with an increase of [ClϪ]. The

OXIDATION OF SOME ␣-HYDROXY ACIDS
877
Figure 1 Variation of pseudo-first-order rate constant with
Ϫ
1
Ϫ1
[glycolate]. Plots of kobs versus [glycolate] at different
temperatures. [AuIII] ϭ 2.2 ϫ 10 mol dm 3, [H ] ϭ 8.9
Ϫ
3
Ϫ
ϩ
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ3
ϫ 10 5 mol dm 3, [Cl ] ϭ 3 ϫ 10 2 mol dm
.
Figure 3 Variation of pseudo-first-order rate constant with
Ϫ
1
Ϫ
chloride ion concentration. Plots of kobs versus [Cl ] for
different substrates at 313K. [AuIII] ϭ 2.2 ϫ 10 3 mol dm
[substrate] ϭ 0.2 mol dm [H ] ϭ 8.9 ϫ 10 mol dm
␮ ϭ 0.4 mol dm 3. (⌬ ϭ GA; ᭡ ϭ LA; ᭹ ϭ IB; ٗ ϭ BA;
,
,
Ϫ
Ϫ
3
Ϫ
3
ϩ
Ϫ5
Ϫ
3
Ϫ
᭺ ϭ MA; ᭿ ϭ AL)
plot of 1/kobs against [ClϪ] is linear with a positive
slope and positive intercept on the ordinate (Figure 3).
The observation that ClϪ inhibits the rate is in keeping
with step (8) of the suggested mechanism.
Effect of Ionic Strength
The reaction was studied at different ionic strengths
maintained by the addition of NaClO4 [(0.05–0.3) mol
3
dmϪ ] but at constant [AuIII], [␣-hydroxy acid], [Hϩ],
4
5
[ClϪ] and temperature of 4 ϫ 10Ϫ , 0.1, 3.5 ϫ 10Ϫ ,
2 ϫ 10Ϫ mol dmϪ , and 308K, respectively. The
pseudo-first-order rate constant increases with increase
in salt concentrations (Figure 4). The observation is to
be expected if the reactions occur between ions of like
charge [25(a)].
2
3
Figure 2 Variation of pseudo-first-order rate constant with
Ϫ
1
ϩ
acid concentration. Plots of kobs versus {K1 ϩ [H ]} for
different substrates at 313K. [AuIII] ϭ 2.2 ϫ 10 3 mol dm
[substrate] ϭ 0.2 mol dm [Cl ] ϭ 3 ϫ 10 mol dm
␮ ϭ 0.3 mol dm 3. (⌬ ϭ GA; ᭡ ϭ LA; ᭹ ϭ IB; ٗ ϭ BA;
,
,
Ϫ
Ϫ
3
Activation Parameters
Ϫ
3
Ϫ
Ϫ2
Ϫ
3
Ϫ
The values of K2 (equilibrium constant for the for-
mation of the 1 : 1 complexbetween the reactants) and
᭺ ϭ MA)

878
SEN GUPTA, PAL, AND SEN
Figure 5 Variation of equilibrium constant (K2) with tem-
Ϫ
perature. Plots of log K2 versus T 1. (⌬ ϭ GA; ᭡ ϭ LA;
᭹ ϭ IB; ٗ ϭ BA; ᭺ ϭ MA)
Figure 4 Variation of pseudo-first-order rate constant with
ionic strength. Plots of log kobs versus ߛ␮ for different sub-
Ϫ
Ϫ
strates at 308K. [AuIII] ϭ 4 ϫ 10 4 mol dm 3, [substrate] ϭ
Ϫ
3
ϩ
Ϫ
5
Ϫ
Ϫ
0.1 mol dm [H ] ϭ 3.5 ϫ 10 mol dm 3, [Cl ] ϭ 2 ϫ
ꢀ
The values of ⌬H were calculated from the respective
10 mol dm 3. (⌬ ϭ GA; ᭡ ϭ LA; ᭹ ϭ IB; ٗ ϭ BA;
Ϫ2
Ϫ
slopes using relation (3), where R, N and h have their
usual significances.
᭺ ϭ MA; ᭿ ϭ AL)
RT
Nh
Hꢀ/RT
eϪ⌬
⌬
Sꢀ/R
k ϭ
e
(3)
k (the disproportionation constant of the intermediate
complex) as referred to in steps 8 and 9 for each sub-
strate at four different temperatures have been evalu-
ated from the plots of substrate effect. The results are
in Table IV. Plots of log K2 versus 1/T (Figure 5) are
linear and the enthalpy change (⌬HЊ) associated with
the equilibrium step was calculated followed by the
estimation of ⌬SЊ from relation (2).
The thermodynamic parameters associated with the
equilibrium step and the activation parameters asso-
ciated with the rate determining step are recorded in
ꢀ
Table V. The fairly large and negative ⌬S values may
indicate a cyclic complex. However, for ions of same
sign the transition state (T.S.) will be a more highly
charged ion which would be expected to be strongly
solvated so that more solvent molecules might be re-
quired than for the separate ions. This would lead to
a decrease in entropy in forming the T.S. [25(b)].
logK2 ϭ [⌬SЊ Ϫ (⌬HЊ/T)]/2.303R
(2)
Plots of log(k/T) versus 1/T (Figure 6) are also linear.
Table IV Equilibrium Constants for the Fast Step and Disproportionation Constants for the Slow Step for Different
Effects at 313K
ϩ
Ϫ
Effect of [S]
Effect of [H ]
Effect of [Cl ]
k ϫ 103
k ϫ 103
k ϫ 103
Ϫ
Ϫ
1
Ϫ
Ϫ
1
Ϫ
Ϫ1
Substrate
K2 ϫ 102
(dm3 mol 1 s
)
K2 ϫ 102
(dm3 mol 1 s
)
K2 ϫ 102
(dm3 mol 1 s
)
Glycolic acid
Lactic acid
␣-Hydroxy
8.759
8.225
7.578
2.33
3.03
3.37
9.452
8.374
6.497
2.16
3.03
3.78
9.541
8.228
6.883
2.02
3.00
3.5
isobutyric acid
Mandelic acid
Atrolactic acid
Benzilic acid
4.327
—
5.47
6.0
7.57
4.3
4.243
—
5.48
6.0
7.9
4.33
4.476
—
5.087
5.95
7.50
4.21

OXIDATION OF SOME ␣-HYDROXY ACIDS
879
Figure 6 Variation of disproportionation constant (k) with
Ϫ
temperature. Plots of log(k/T) versus T 1. (⌬ ϭ GA; ᭡ ϭ
LA; ᭹ ϭ IB; ٗ ϭ BA; ᭺ ϭ MA; ᭿ ϭ AL)
Figure 7 Isokinetic plots for the oxidations of ␣-hydroxy
ꢀ
carboxylates by AuIII in buffer medium. (a, aЈ): plot of ⌬S
ꢀ
vs ⌬H (b, bЈ): plot of log kЈ vs log k (1 ϭ GA; 2 ϭ LA;
3 ϭ IB; 4 ϭ MA; 5 ϭ BA; 6 ϭ AL)
The enthalpy of activation is linearly related to the
entropy of activation (r ϭ 0.99, Figure 7) and the iso-
kinetic temperature is 291K. This indicates that a sim-
ilar mechanism may be operative in all the reactions.
The isokinetic behavior is also supported by the linear
plot of log k versus log kЈ (r ϭ 0.98, Figure 7) where
k and kЈ are the disproportionation constants of the
[AuIII], [Hϩ], [ClϪ] and temperature of 4.5 ϫ 10Ϫ , 2.2
1
3
5
2
3
ϫ 10Ϫ , 8.9 ϫ 10Ϫ , 3 ϫ 10Ϫ mol dmϪ , and 318K
respectively. The kH/kD value has been found to be
1.02. This indicates that C9H bond cleavage does
not occur.
complexat temperature 303K (T ) and 308K(T2), re-
1
spectively. The isokinetic temperature was calculated
from the relation ␤ ϭ T1T2(1 Ϫ f)/(T1 Ϫ T2f), where
f is the slope of the Exner plot [26]. The value of ␤
was found to be 276K which is lower than T, (mid
point of the experimentally used range of tempera-
tures). Although in most cases, ␤ is found to exceed
T, there is literature evidence [27] for a large number
of reactions for which ␤ is lower than T. The rates of
such reactions for which ␤ Ϫ T is negative, are be-
lieved to be entropy controlled.
In a solution of tetrachloroauric(III) acid the fol-
lowing equilibria exist [28–30]:
K3
Ϫ
HAuCl4 ER
F Hϩ ϩ AuCl4
(4)
K4
Ϫ
AuCl4 ϩ H2O ER
F AuCl3(OH2) ϩ ClϪ (5)
K5
AuCl3(OH2) E
R
F AuCl3(OH)Ϫ ϩ Hϩ
(6)
The kinetic isotope effect using mandelic and ␣-
deutero mandelic acids was carried out at [substrate],
6
where K3 ϭ 1.0, K4 ϭ 9.5 ϫ 10Ϫ and K5 ϭ 0.25 at
Table V Thermodynamic Data Associated with the Complex and the Activation Parameters Associated with the
Slowest Steps
ꢀ
ꢀ
⌬HЊ
(kJ mol
⌬SЊ
⌬H
⌬S
Ϫ
1
Ϫ
Ϫ
1
Ϫ
1
Ϫ
Ϫ1
Substrate
)
(JK 1 mol
)
(kJ mol
)
(JK 1 mol
)
Glycolic acid
Lactic acid
␣-Hydroxy isobutyric acid
Mandelic acid
Atrolactic acid
Benzilic acid
20 Ϯ 4
9 Ϯ 4
25 Ϯ 4
11 Ϯ 4
—
45 Ϯ 13
8 Ϯ 13
59 Ϯ 13
9 Ϯ 13
—
38 Ϯ 2
46 Ϯ 2
48 Ϯ 2
53 Ϯ 4
57 Ϯ 2
54 Ϯ 2
Ϫ173 Ϯ 7
Ϫ146 Ϯ 7
Ϫ139 Ϯ 7
Ϫ118 Ϯ 14
Ϫ101 Ϯ 7
Ϫ116 Ϯ 7
13 Ϯ 4
19 Ϯ 13

880
SEN GUPTA, PAL, AND SEN
298K. Since the reactions were studied at a very low
tially increases and then decreases gradually (Figure
8). The initial increase in absorbances indicates that
an intermediate complexis formed between the reac-
tants. The latter then decomposes to give products as
mentioned in steps (9) and (10), where R ϭ RЈ ϭ
9H for glycolic, RЈ ϭ 9H for lactic and mandelic
acid. The observed kinetic isotopic effect suggests that
the oxidation occurs via C9C bond cleavage and not
by C9H bond breaking.
3
[Hϩ] (ϳ10Ϫ5 mol dmϪ ), [HAuCl4] will be practically
Ϫ
insignificant compared to [AuCl4 ]. Again, in a solu-
2
3
tion where [ClϪ] ϳ 10Ϫ mol dmϪ , the value of K4
suggests that [AuCl3(OH)2] and hence [AuCl3(OH)Ϫ]
will be negligibly small compared to [AuCl4 ]. Con-
Ϫ
sequently, under the present experimental condition,
AuCl4Ϫ may only act as the oxidizing species.
Gold(III) is known [29] to behave as a one- or two-
electron transfer oxidant, depending upon the nature
of the substrate and the experimental conditions. In the
present investigation, the reaction mixture did not pro-
duce any polymeric suspension in presence of acry-
lonitrile. This suggests that gold(III) possibly behaves
as a two-electron transfer oxidant in the present reac-
tion. The observation is similar to that noticed in the
oxidation of glycolaldehyde by gold(III) [31].
R
OH
R
OH
K1
C
C
ϩ Hϩ (7)
RЈ
COOH
RЈ
COOϪ
(HA)
(AϪ)
R
OH
Cl
Cl
Cl
K2
C
ϩ
AuϪ
The acid dissociation constants [32] of some ␣-hy-
droxy acids at 298K are known, the values are 1.47 ϫ
fast
RЈ
COOϪ Cl
4
4
4
4
10Ϫ , 1.38 ϫ 10Ϫ , 2.2 ϫ 10Ϫ , 3.89 ϫ 10Ϫ , and
R
C
O9H
9.1 ϫ 10Ϫ for glycolic, lactic, ␣-hydroxy/isobutyric,
4
Cl
C99O99AuϪ
Cl
Cl ϩ ClϪ (8)
mandelic and benzilic acids, respectively. Since the
reactions were carried out at low acidities, the disso-
ciation constant (K1) values [32] of the ␣-hydroxy ac-
ids suggest that they will remain dissociated to a large
extent (Eq. 7). The substrate anion enter into the co-
ordination sphere of square planar gold(III) complex
and replaces a chloride ligand to give an intermediate
complex(X). This is further supported by the spectro-
photometric evidence for the intermediate complex(X)
formation. The absorbance of the reaction mixture ini-
RЈ
O
(X)
R
O9H
Cl
C
Cl
k
RЈ
C99O99AuϪ
slow
O
Cl
(X)
R
O9H
Cϩ
ϩ CO2 ϩ AuCl2Ϫ ϩ ClϪ
(9)
RЈ
R
O9H
R
fast
Cϩ
C"O ϩ Hϩ (10)
RЈ
RЈ
The rate of disappearance of AuIII may be expressed
as
Ϫd[AuIII]/dt ϭ k[X]
K2[AϪ][AuIII]f
(11)
(12)
Now [X] ϭ
Figure 8 Spectrophotometric evidence for intermediate
complexformation between mandelic acid and gold(III) at
[ClϪ]
K1
where [AϪ] ϭ
[S]
Ϫ
Ϫ
ϩ
Ϫ5
400 nm. [AuIII] ϭ 2.2 ϫ 10 3 mol dm 3, [H ] ϭ 8.9 ϫ 10
K1 ϩ [Hϩ]
mol dm 3, [Cl ] ϭ 3 ϫ 10 mol dm 3, Temp ϭ 318K,
Ϫ
Ϫ
Ϫ
2
Ϫ
K2K1 [S] [AuIII]f
{K1 ϩ [Hϩ]}[ClϪ]
[MA] are ٗ, 3 ϫ 10 1; ᭡, 4 ϫ 10 1; ᭝, 5 ϫ 10 1; ᭹,
Ϫ
Ϫ
Ϫ
І [X] ϭ
6 ϫ 10 1 and 0, 7 ϫ 10 1 mol dm
.
Ϫ
Ϫ
Ϫ3

OXIDATION OF SOME ␣-HYDROXY ACIDS
881
K2K1 [S] [AuIII]f
{K1 ϩ [Hϩ]}[ClϪ]
glycolic acid stabilizes the anion through stronger in-
tramolecular H-bonding and the value of K2 dimin-
ishes in the order GA Ͼ LA Ͼ IB (Table IV). In-
and [AuIII]t ϭ [AuIII]f ϩ
so that [AuIII]f ϭ
troduction of
a
phenyl group decreases the
[AuIII]t
nucleophilicity of the anion thereby further diminish-
ing the value of K2. Again, introduction of a methyl
group at the ␣-carbon of glycolic acid increases the
electron density on the ␣-carbon, thus rendering the
C9C bond breaking easier in the order GA Ͻ LA Ͻ
IB. Also, introduction of bulky methyl groups at the
␣-carbon facilitates the C9C bond breaking in the
slow step due to release of steric strain in the product
carbonyl compounds with sp2 hybridization at that ␣-
carbon atom. From the same reasoning, introduction
of a phenyl group should decrease the electron density
on the ␣-carbon and the decrease in k (step 9) is ex-
pected. On the contrary, the value of k is higher for
mandelic acid than for ␣-hydroxy isobutyric acid
which can be explained only by the fact that carbo-
nium ion formed on the ␣-carbon after removal of CO2
is stabilized through resonance in the benzene ring.
The value, however, slightly diminishes after intro-
duction of a second phenyl group (viz, benzilic acid)
possibly due to the fact that the benzene rings go
slightly out of plane (that is, twisted) making reso-
nance stabilization of the positively charged interme-
diate slightly difficult. In case of atrolactic acid, the ␣-
carbon of the 9OH group is attached to one 9Ph
group and one 9Me group. The 9Ph group attached
to the ␣-carbon stabilizes the carbonium ion formed
in the intermediate through resonance while the 9Me
group stabilizes the system through hyperconjugation.
The k values which are in the order GA Ͻ LA Ͻ IB
Ͻ BA Ͻ MA Ͻ AL are in conformity with the sug-
gested mechanism.
(13)
K1K2 [S]
1 ϩ
(K1 ϩ [Hϩ])[ClϪ]
Equation (10) can be transformed into
Ϫd [AuIII]t/dt
kK1K2 [S]
{K1 ϩ [Hϩ]}[ClϪ]
[AuIII]t
K2K1 [S]
ϭ
1 ϩ
ͭ ͮ
(K1 ϩ [Hϩ])[ClϪ]
(14)
so that the pseudo-first-order rate constant
1
d[AuIII]t
dt
kobs ϭ Ϫ
[AuIII]t
kK1K2[S]
{K1 ϩ [Hϩ]}[ClϪ] ϩ K2K1[S]
ϭ
(15)
Rearranging equation (15), we obtain
{K1 ϩ [Hϩ]}[ClϪ]
1/kobs
ϭ
ϩ 1/k
(16)
kK1K2[S]
Thus equation (16) can explain the linear plot (Fig-
ure 1) of 1/kobs versus 1/[S] and from the intercept of
the plot at each temperature the value of the dispro-
portionation constant (k) at that temperature was eval-
uated and using the literature value of K1, the value of
K2 at that temperature was calculated from the slope
(Table IV). We are unable to calculate the value of K2
in case of atrolactic acid, since the dissociation con-
stant at that temperature is not available in the litera-
ture. Figure 2 shows an inverse dependence of the rate
on [Hϩ] which conforms to the rate law (16). The val-
ues of k and K2 at 313K evaluated from the plot of
1/kobs against {K1 ϩ [Hϩ]} at that temperature (Figure
2) are found to be in agreement with those obtained
from the substrate effect as well as effect of [ClϪ] at
that temperature (Table IV).
It is evident that in the slow step (9), the transition
state will be much more polar than the intermediate
complex(X) and the former will be more stabilized
by the presence of ion atmosphere than the latter. Thus
an increase in ionic strength will favor the transition
state, thereby increasing the value of k (step 9) and
consequently an increase in kobs is expected as ob-
served in the present investigation (Figure 4).
Thanks are due to C.S.I.R. for financial assistance to
K.K.S.G.
BIBLIOGRAPHY
1. McAuley, A. J Chem Soc 1965, 4054.
2. Prasad, S.; Choudhury, J. Ind J Chem 1979, 17A(2),
167.
3. Krishna, B.; Tewari, K. C. J Chem Soc 1961, 3097.
4. Hanna, S. B.; Sarac, S. A. J Org Chem 1977, 42, 2063.
5. Hanna, S. B.; Sarac, S. A. J Org Chem 1977, 42, 2069.
6. Levesley, P.; Waters, W. A. J Chem Soc 1955, 217.
7. Jones, J. R.; Waters, W. A.; Littler, J. S. J Chem Soc
1961, 630.
8. Sen Gupta, K. K.; Maiti, S.; Chatterjee, U.; Samanta, T.
Transition Met Chem 1982, 7, 89.
Introduction of a methyl group at the ␣-carbon of

882
SEN GUPTA, PAL, AND SEN
9. Natarajan, R.; Venkatasubramanian, N. Tet Lett 1969,
57, 5021.
21. Feigl, F. Spot Tests in Organic Analysis; Elsevier, Am-
sterdam, 1966, p. 434.
10. Natarajan, R.; Venkatasubramanian, N. Tetrahedron
1974, 30, 2785.
11. Natarajan, R.; Venkatasubramanian, N. Int J Chem Ki-
net 1975, 8, 205.
12. Sen Gupta, K. K.; Banerjee, A.; Chatterjee, H. Tetra-
hedron 1992, 48, 5323.
13. Sen Gupta, K. K.; Basu, B. Polyhedron 1984, 3, 805–
810.
14. Sen Gupta, K. K.; Basu, B.; Sen Gupta, S.; Nandi, S.
Polyhedron 1983, 2, 983–986.
15. Sen, P. K.; Sanyal, A.; Sen Gupta, K. K. Bull Chem Soc
Jpn 1996, 69, 1543–1548.
16. (a) Fry, F. H.; Hamilton, G. A.; Turkevich, J Inorg
Chem 1966, 5, 1943.
22. Huntress, E. H.; Mulliken, S. P. Identification of Pure
Organic Compounds; John Wiley and Sons, New York,
1953, p. 43.
23. Clarke, H. T. Handbook of Organic Analysis; Edward
Arnold, London, 1956, p. 130.
24. Huntress, E. H.; Mulliken, S. P. Identification of Pure
Organic Compounds; John Wiley and Sons, (New
York), Fourth Printing, 1953, 50, 44, 374, 388, 60, 363,
610.
25. Frost, A. A.; Pearson, R. G. Kinetics and Mechanism;
Wiley Eastern (New Delhi), 1970, a) p. 150, b) p. 143.
26. Exner, O. Nature (London) 1964, 201, 488.
27. Leffler, J. E. J Org Chem 1955, 20, 1202.
28. Van, P.; Bekker, Z.; Louw, W. J.; Robb, W. Inorg Nucl
Chem Lett 1972, 8, 849.
(b) Rich, R. L.; Taube, H. J Phys Chem 1954, 58, 1.
(c) Robb, W.; Inorg Chem 1967, 6, 382.
17. Kemp, T. J.; Waters, W. A. J Chem Soc 1964, 1194.
18. Sounder Rajan, S. C.; Appala Raju, N. J Indian Chem
Soc 1973, 50, 110.
29. Moodley, K. G.; Nicol, M. J. J Chem Soc, Dalton Trans
1977, 993.
30. Forsberg, H. G.; Widdel, H.; Erwall, L. G. J Chem Educ
1960, 37, 44.
19. Vogel, A. I. Textbook of Quantitative Inorganic Anal-
ysis; Fourth Edition, ELBS, U.K., 1978, p. 45.
20. Balzani, V.; Carassiti, V. Photochemistry of Coordina-
tion Compounds; Academic Press, London, 1970, 273.
31. Sen, P. K.; Bilkis, A. B.; Sen Gupta, K. K. Int J Chem
Kinet 1998, 30, 613.
32. St. Nikolov, G. Inorg Chim Acta 1971, 5, 559–562.