Oxidation of threose-series pentoses and hexoses by sodium N-chloro-p- toluenesulfonamide

Article abstract of DOI:10.1016/S0008-6215(98)00046-9

The kinetics and mechanism of oxidation of threose-series hexoses and pentoses by chloramine-T in alkaline medium was investigated. Kinetic studies with D-galactose, D-sorbose, D-xylose, and D-lyxose showed that the rate of the reaction was first order with respect to sugar and chloramine-T, and second order with respect to hydroxide ion. p-Toluenesulfonamide and chloride ions, the reduced products of chloramine-T, have no effect on the reaction rate. The rate increases with increase in ionic strength of the medium, and the dielectric effect is negative. Proton inventory studies in H₂O-D₂O mixtures suggested a single transition state. Product analysis for D-gulose, Didose, L-sorbose, D-galactose, D-talose, D-tagatose, D-xylose, and D-lyxose revealed that all lyxoseseries hexoses gave mainly mixtures of lyxonic and threonic acids with minor proportions of hexonic, xylonic and glyceric acids, whereas all xylose-series hexoses gave mixtures of lyxonic, threonic and glyceric acids with minor amounts of xylonic and hexonic acids. Xylose and lyxose gave mixtures consisting mainly of lyxonic, threonic, and glyceric acids with minor proportions of xylonic acid. From the results of kinetic studies, reaction stoichiometry, and product analysis, a possible mechanism for the oxidation of threose-series sugars with chloramine-T is suggested.

Full text of DOI:10.1016/S0008-6215(98)00046-9

Carbohydrate Research 307 (1998) 253±262
Oxidation of threose-series pentoses and hexoses
by sodium N-chloro-p-toluenesulfonamide
Kanchugarakoppal S. Rangappaa, Huchappa Manjunathaswamya,
Manikanahally P. Raghavendraa, D. Channe Gowdab,*
a Department of Studies in Chemistry, University of Mysore, Manasagangothri, Mysore-570 006, India
b Department of Biochemistry and Molecular Biology, Georgetown University Medical Center,
Washington, DC 20007, USA
Received 17 November 1997; accepted 25 January 1998
Abstract
The kinetics and mechanism of oxidation of threose-series hexoses and pentoses by chloramine-
T in alkaline medium was investigated. Kinetic studies with d-galactose, d-sorbose, d-xylose, and
d-lyxose showed that the rate of the reaction was ®rst order with respect to sugar and chloramine-
T, and second order with respect to hydroxide ion. p-Toluenesulfonamide and chloride ions, the
reduced products of chloramine-T, have no e€ect on the reaction rate. The rate increases with
increase in ionic strength of the medium, and the dielectric e€ect is negative. Proton inventory
studies in H2O±D2O mixtures suggested a single transition state. Product analysis for d-gulose, d-
idose, l-sorbose, d-galactose, d-talose, d-tagatose, d-xylose, and d-lyxose revealed that all lyxose-
series hexoses gave mainly mixtures of lyxonic and threonic acids with minor proportions of
hexonic, xylonic and glyceric acids, whereas all xylose-series hexoses gave mixtures of lyxonic,
threonic and glyceric acids with minor amounts of xylonic and hexonic acids. Xylose and lyxose
gave mixtures consisting mainly of lyxonic, threonic, and glyceric acids with minor proportions of
xylonic acid. From the results of kinetic studies, reaction stoichiometry, and product analysis, a
possible mechanism for the oxidation of threose-series sugars with chloramine-T is suggested.
# 1998 Elsevier Science Ltd. All rights reserved
Keywords: Threose-series sugars; Oxidation with chloramine-T; Kinetics and mechanism
1. Introduction
d-glucose, d-mannose, d-fructose, d-arabinose and
d-ribose, with ꢀ chloramine-T (CAT) in alkaline
medium at 35 C [1]. The observed reaction stoi-
chiometry, 2±3 moles of CAT per mole of sugar,
was signi®cantly di€erent from the previously
reported sugar to oxidant stoichiometry of 1:1 for
aldoses and 1:2 for fructose [2±4]. We have also
shown by HPLC and GLC-MS analyses that the
products of oxidation for erythrose-series sugars
(both pentoses and hexoses) were mixtures of
aldonic acids consisting of arabinonic, ribonic,
erythronic, and glyceric acids [1]. These product
Recently, we reported on the kinetics and
mechanism of oxidation of erythrose-series sugars,
Abbreviations: CAT, RNClNa or chloramine-T, sodium
salt of N-chloro-p-toluenesulfonamide; S, sugar, D. dielectric,
T, absolute temperature; Ea, activation energy; n, atom fraction
of deuterium, TS, transition state; RS, reactant site; I, ionic
strength; dAB, size of the activated complex; k, Boltzmann con-
stant; HPLC, high-performance liquid chromatography; GLC-
MS, gas±liquid chromatography±mass spectrometry.
* Corresponding author. Fax: (202) 687-7186.
0008-6215/98/$19.00 # 1998 Elsevier Science Ltd. All rights reserved
PII S0008-6215(98)00046-9

254
K.S. Rangappa et al./Carbohydrate Research 307 (1998) 253±262
pro®les were also di€erent from those reported
previously, the corresponding aldonic acids for
aldoses and arabinonic acid for fructose [2±4].
Our recent study on the oxidation of erythrose-
series sugars with CAT gave two interesting results
[1]. First, the sugars that can exist in the furanose
ring form in appreciable proportions reacted with
CAT much faster than those which exist almost
exclusively in the pyranose form. Thus, for hexo-
ses, rate of oxidation of fructose was higher than
glucose and mannose. Similarly, ribose was oxi-
dized faster than arabinose. Second, surprisingly
the products formed from both pentoses and
hexoses, including keto-hexoses, were strikingly
similar for all erythrose-series sugars studied.
Based on these results, we proposed a novel path-
way for the oxidation of erythrose-series sugars by
CAT [1]. In the present study, the mechanism of
oxidation of threose-series sugars by CAT was
investigated by kinetic studies and product analy-
sis. The results demonstrate that the kinetic and
thermodynamic properties, and mechanism of
oxidation of threose-sieres sugars are generally
similar to those observed for the erythrose-series
sugars.
reaction. The solvent isotopic studies were per-
formed with D2O.
Pseudo-®rst order rate constants (kobs) were cal-
culated from the plots of log[CAT]0 versus time,
and these were within ‹3%. Regression coecient
r and the standard deviation s, were determined by
regression analysis of the experimental data using
an EC-72 statistical calculator.
Stoichiometry and product analysis.ÐThe reac-
tion mixtures containing sugar (0.01 M), alkali
(0.1 M), and CAT (0.05 M) were kept for 24 h at 35
ꢀC. The unconsumed CAT was determined iodo-
metrically. From these data, the amount of the
oxidant consumed per mole of sugar was calcu-
lated.
The oxidation products were analyzed by a
Dionex HPLC system with pulsed amperometric
detection using a CarboPac PA1 high-pH anion-
exchange column (4Â250 mm) as reported pre-
viously [1,5]. Isocratic elution with 0.2 M NaOH
was used. The products were identi®ed by com-
parison of the retention times with retention times
of the standard aldonic acids, as reported pre-
viously [1].
3. Results
2. Experimental
The kinetics of oxidation of threose-series sugars
with CAT were generally similar to those observed
previously for the erythrose-series sugars [1]. When
sugars were used in excess, the plots of log[CAT]
versus time were linear (r> 0.9980, sꢁ 0.02), indi-
cating a ®rst-order dependence of reaction rate on
[CAT]0. The pseudo-®rst-order rate constants
(kobs) calculated from these plots are shown in
Table 1. The rate increased with increase in [S]0
(where S=sugar) and the plots of logkobs versus
log[S]0 were linear (r > 0.9998, sꢁ 0.01) with unit
slopes, indicating a ®rst-order dependence with
respect to sugars. Furthermore, the plots of kobs
versus [S]0 passed through the origin (r> 0.9855,
sꢁ 0.04), suggesting that the sugar±oxidant com-
plexes have only transient existence.
The rate of oxidation also increased with an
increase in alkali concentration (Table 2). The
plots of log kobs versus log[HO ] (r> 0.9990, sꢁ
0.01) indicated that the reactions follow second-
order dependence on [HO ].
Addition of p-toluenesulfonamide (0 to 0.008 M)
did not a€ect the reaction rate, suggesting that the
reduced product of CAT, p-toluenesulfonamide, is
Materials.ÐChloramine-T (E. Merck) was pur-
i®ed from dichloro contaminants by washing with
CCl4. d-galactose, d-xylose, d-lyxose, d-idose, d-
talose, d-tagatose, d-gulose, d-gluconic acid, d-
galactono-1,4-lactone, and d-ribono-1,4-lactone,
were purchased from Sigma Chemical Co. (St.
Louis, MO). l-sorbose, d-mannono-1,4-lactone,
d-xylono-1,4-lactone, and d-arabinono-1,4-lac-
tone, were from Pfanstiehl (Waukegan, IL). D2O
(99.4%) was from Bhabha Atomic Research Cen-
ter (Bombay, India).
Kinetic measurements.ÐThe reactions were car-
ried out in glass stoppered pyrex boiling tubes
coated black on the outside [1]. Pseudo-®rst order
conditions ([sugar]0>> [CAT]0) were maintained
for all kinetic studies. Stock solutions of alkali,
CAT, sugars, and NaClO4 were maintained at 35
ꢀC. From these stock solutions, the requisite mix-
tures of sugars, alkali, and NaClO4 were prepared.
The reaction was initiated by the addition of CAT
and monitored by iodometric determination of
unconsumed CAT at various time-intervals.
Sodium perchlorate was used to ``swamp'' the

K.S. Rangappa et al./Carbohydrate Research 307 (1998) 253±262
255
Table 1
E€ect of reactant concentrations on the rate of oxidation of
ꢀ
sugars by CAT at 35 C
1
103
102
104 kobs(s
)
[CAT]0 [S]0 (m)
(m)
d-galactose l-sorbose d-xylose d-lyxose
1.5
2.0
2.5
3.0
3.5
4.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
0.6
0.8
1.0
3.0
4.0
6.0
3.2
3.0
3.1
3.2
3.2
3.3
0.9
1.2
1.5
4.9
6.6
10.2
13.3
12.9
12.9
13.0
13.0
13.0
3.8
5.0
6.3
19.8
26.3
41.9
14.0
13.4
13.4
13.9
14.1
14.4
4.9
6.0
7.8
23.8
33.3
64.0
5.8
5.5
5.6
5.6
5.5
5.6
1.6
2.1
2.7
8.4
11.3
15.4
[HO ]=0.1 M, and [I]=0.4 M.
not involved in pre-equilibrium with the oxidant.
Addition of NaCl (0 to 0.02 M) to the reaction
mixtures had no e€ect on the rates, suggesting that
the free chloride ion was not formed before the
rate-limiting step.
Addition of NaClO4 (0 to 0.8 M) increased the
rate of reaction. The plots of log kobs versus I1/2
(I=the ionic strength of medium) were linear with
fractional slopes of 0.60±0.85 (not shown).
The rate decreased with an increase in methanol
content. The plots of log kobs versus 1/D (r>
0.9970, sꢁ 0.04; D=dielectric constant of the
medium) were linear with negative slopes (Fig. 1).
The Arrhenius plots of log kobs versus 1/T
(T=absolute temperature), for reactions studied
over a range of temperatures (303 to 318 K), were
found to be linear (r> 0.9991, sꢁ 0.01) (not
shown). The activation energies (Ea, Table 3) were
calculated from the slopes of the plots. From the
values of Ea and the thermodynamic parameters
ÁH#, ÁS#, and ÁG# (Table 3) were computed. Sor-
Fig. 1. Plots of logkobs versus 1/D. [CAT]0=0.002 M,
[S]0=0.002 M, [OH ]=0.1 M, [I]=0.4 M; temperature= 35
ꢀC.
bose and xylose have lower Ea and ÁS# values as
compared with other sugars studied. A low
enthalpy change for sorbose and a negative
enthalpy change for xylose suggest that the struc-
tures of sorbose and xylose are favorably disposed
for oxidation by CAT. Thus, a keto-enolic anion
appears to be the readily reacting structure for
hexoses and an aldo-enolic one for pentoses. Pre-
viously, it was observed, for erythrose-series
sugars, that a fructose-enolic anion is the readily
reacting structure for the hexoses and an aldo-
enolic form is that reacting for the pentoses [1].
The oxidation of sugars by CAT was studied in
H2O±D2O mixtures containing varying deuterium
atom fractions n. As in the case of erythrose-series
sugars [1], the oxidation was faster in D2O for the
threose-series sugars (Table 4). The solvent-isotope
e€ects, kobs(H2O)/kobs(D2O), were between 0.5 and
Table 2
E€ect of [NaOH] on the rate of oxidation of sugars by CAT at
ꢀ
Table 3
Thermodynꢀamic parameters for the oxidation of sugars by
35 C
1
102 [NaOH] (m)
104 kobs(s
)
CAT at 35 C
ÁH#
ÁG#
ÁS#
Log
A
d-galactose l-sorbose d-xylose d-lyxose
Sugars
Ea
(kJ mol 1) (kJ mol 1) (kJ mol 1) (JK 1 mol
1
)
2.0
4.0
6.0
8.0
0.2
0.6
1.5
2.2
3.0
6.8
0.5
2.1
5.0
8.0
12.9
36.2
1.1
3.4
7.1
11.3
13.4
38.4
0.3
1.0
2.4
4.1
5.5
17.8
d-Galactose
l-Sorbose
d-Xylose
121
98
119
96
96
93
93
95
72
9
21
17
16
19
91
111
89
109
12
44
10.0
15.0
d-Lyxose
[CAT]0=0.002 M,
[I]=0.4 M.
[HO ]=0.1 M,
[S]0=0.05 M,
and
[CAT]0=0.002 M, [S]0=0.02 M, and [I]=0.4 M.

256
K.S. Rangappa et al./Carbohydrate Research 307 (1998) 253±262
Table 4
Proton inventory studies for the oxidation of sugars by CAT in
H2O±D2O mixtures at 35 C
and tagatose) gave threonic and lyxonic acids as
predominant products with small amounts of gly-
ceric and xylonic acids (Fig. 2a). All hexoses except
sorbose gave minor amounts of hexonic acids
(Fig. 2a and b). Furthermore, all threose-series
sugars except lyxose and galactose were oxidized
almost quantitatively by CAT; after 24 h incuba-
tion with CAT at 35 ꢀC, 20±25% of lyxose and 5%
of galactose remained unoxidized (Fig. 2a±c).
The oxidation products were analyzed at 0.5, 1,
2, 4, 8, 20, and 24 h for all the sugars. The relative
proportions of various aldonic acids formed were
similar at all time-points analyzed. Approximately
95% of sorꢀbose and idose were oxidized by CAT in
4 h at 35 C, whereas 90% oxidation of gulose
required about 8 h. When gulose and idose were
treated with alkali in the absence of CAT, a gra-
dual build up of sorbose was observed for both
sugars; for example, about 30% of gulose was iso-
merized to sorbose in 2 h. However, when gulose
and idose were treated with both CAT and alkali,
only a minor proportion of sorbose was detected
from each reaction mixture at all time-points ana-
lyzed, suggesting that sorbose formed by the alkali-
catalyzed isomerization readily reacts with CAT.
Sorbose treated with alkali alone was not iso-
merized to gulose and idose to detectable levels.
These data indicate that the keto-isomer is the
reactive species.
ꢀ
1
Atom fraction
of deuterium (n)
104 kobs(s
)
d-galactose l-sorbose d-xylose d-lyxose
0.00
0.25
0.50
0.75
0.93
3.2
3.4
4.0
5.0
5.8
12.9
15.4
18.3
22.0
25.5
13.4
16.1
18.6
22.3
25.9
5.5
6.4
7.6
8.9
10.4
[CAT]0=0.002 M, [HO ]=0.1 M, [S]0=0.02 M, and [I]=
0.4 M.
0.6 for all sugars (Table 4). Proton-inventory plots
(not shown), kobs(H2O)/kobs(D2O) versus n , were
linear and similar to those obtained previously for
erythrose-series sugars (1).
The oxidant to sugar stoichiometry of nearly 3
was observed for all sugars except for lyxose
(Table 5).
Although kinetic studies could not be carried out
for gulose, idose, talose and tagatose due to the
limited availability of the sugars, the oxidation
products were analyzed for all threose-series hexo-
ses (Fig. 2). HPLC analysis indicated that lyxonic,
xylonic, threonic, and glyceric acids are the pro-
ducts of oxidation for all threose-series hexoses
and pentoses (Fig. 2a±c, and Table 5). Xylose and
lyxose gave major proportions of lyxonic, threonic
and glyceric acids and minor proportions of xylo-
nic acid (Fig. 2c). The xylose-series hexoses (gulose,
idose, and sorbose) gave predominantly threonic
and glyceric, acids with minor proportions of
xylonic and lyxonic acids (Fig. 2b). On the other
hand, all lyxose-series hexoses (galactose, talose,
About ꢀ95% of tagatose was oxidized by CAT in
2 h at 35 C, whereas oxidation of similar amounts
of galactose and talose required ꢂ20 h. When
galactose and talose were treated with alkali alone,
signi®cant amounts of tagatose formed by the iso-
merization of these sugars could be detected. For
example, 17% of talose and 23% of galactose were
Table 5
HPLC analysis of the products formed by the oxidation of sugars by CAT in alkaline medium
Sugar
Mol of CAT consumed
per mol of sugar
Products (approximate percentage) a
Glyceric acid Threonic acid, erythronic acid b Xylonic acid Lyxonic acid Hexonic acid
d-Galactose
d-Talose
d-Tagatose
d-Gulose
d-Idose
l-Sorbose
d-Xylose
d-Lyxose
2.8
n.d.
n.d.
2.9
n.d.
n.d.
2.8
9
9
44
45
47
38
40
44
37
32
3
2
2
12
8
9
32
32
27
5
5
3
12
11
12
7
12
38
40
44
20
25
7
Ð
Ð
Ð
7
6
36
37
1.9
a
Based on the peak areas.
Peak 2 in Fig. 2 represents 95±96% threonic acid and 4±5% threonic acid, which can be separated by using a low ¯ow rate.
n.d., Not determined.
b

K.S. Rangappa et al./Carbohydrate Research 307 (1998) 253±262
257
Fig. 2. HPLC analysis of the products formed by the oxidation of sugars (0.01 M) by CAT (0.05 M) in the presence of alkali
ꢀ
(0.1 M) at 35 C. 1, glyceric acid; 2, threonic acid (ꢂ95%) plus erythronic acid (ꢂ5%); 3, xylonic acid; 4, lyxonic acid; 5, hexonic
acids. Gal, Tal, Tag, Gul, Ido, Sor, Xyl and Lyx, respectively represent HPLC chromatograms of the reaction of CAT with d-
galactose, d-talose, d-tagatose, d-gulose, d-idose, l-sorbose, d-xylose, and d-lyxose. Peak 2 in (a)±(c) represents 95±96% threonic
acid and 4±5% threonic acid which can be separated by using a low ¯ow rate. Note: about 20±25% of lyxose (retention time 4.6 min
in (c)) and ꢂ5% galactose (retention time 5.0 min in (a)), and 1±2% of xylose (retention time 5.0 min in (c) were not oxidized by
CAT; all other sugars were almost quantitatively oxidized.

258
K.S. Rangappa et al./Carbohydrate Research 307 (1998) 253±262
isomerized to tagatose in 4 h. However, when
talose was treated with both CAT and alkali,
unoxidized talose was accounted by 1% tagatose,
1% galactose, and 98% talose. Similarly, when
galactose was treated with both CAT and alkali,
tagatose was barely detectable. Upon treatment
and aldo-enolic anions of pentoses) react with
RNCl to form an intermediate (X), which under-
goes cleavage to form products (Fig. 3). From the
results of the kinetic study, the oxidation of sugars
is predicted to proceed through the reaction
sequences shown in Scheme 1.
ꢀ
with alkali at 35 C, only 1±2% of tagatose was
Under steady state conditions for E , the rate of
isomerized into galactose and talose. Together,
these data indicated that for lyxose-series hexoses,
as in the case of the xylo-series hexoses, the keto-
isomer (tagatose) is the reactive species.
All of the aldohexoses studied here were oxi-
dized mainly to pentonic and tetronic acids rather
than to hexonic acids. This ®nding is in agreement
with the foregoing conclusion that hexoses are
oxidized by CAT in the keto-enolic form. More-
over, we have previously shown, based on the ease
of aldose±ketose isomerization, and relative rates
of oxidation and product pro®les, that glucose and
mannose undergo oxidation by CAT pre-
dominantly in the fructose-enolic form [1].
Scheme 1.
disappearance of CAT is given by
For all the hexoses studied here, threonic and
pentonic acids were formed within 30 min; hexonic
acids were detectable in minor proportions only
after signi®cant amounts of the former products
are formed. This demonstrate that the lower-car-
bon aldonic acids were not derived from the initi-
ally formed six-carbon aldonic acids. We have
previously shown that d-gluconic, d-mannonic, d-
galactonic, d-ribonic, and d-arabinonic acids were
not oxidized by CAT [1].
Formation of high proportions of pentonic acids
from xylose and lyxose suggest that pentoses react
with CAT predominantly in the aldo-enolic form.
The formation of glyceric acid also in high pro-
portion suggest that appreciable amounts of xylose
and lyxose react in the keto-enolic form to give
glyceric and threonic acids.
2
d‰CATŠ
dt
k2k1‰SŠ‰HO Š ‰CATŠ
Rate ˆ
ˆ
1
k ‰H2†Š ‡ k2‰HO Š‰CATŠ
1
ꢀ1†
Since k ‰H2OŠ > k2‰HO Š‰CATŠ, rate law (1) is
1
reduced to
2
d‰CATŠ k2k1‰SŠ‰HO Š ‰CATŠ
Rate ˆ
ˆ
ꢀ2†
dt
k ‰H2OŠ
1
which agrees with the observed rate ˆ kobs‰SŠ
2
‰OH Š ‰CATŠ.
The observed ®rst-order dependence of rate on
‰CATŠ and ‰SŠ and second-order dependence on
0
0
‰HO Š agree with eq (2).
Since DO is a stronger base than HO by a
factor of 2, the reaction rate is expected to be
doubled in D2O for reactions involving a fast pre-
equilibrium H+ or HO ion transfer [12]. In
agreement with this, the observed values of the
inverse solvent isotope e€ect kobs(D2O)/kobs(H2O)
were approximately 2 for the sugars (Table 4).
The results of proton inventory plots (not shown)
are also in accordance with Scheme 1. The
dependence of rate constant ꢃknobs† on n (n=the
atom fraction of deuterium in a solvent mixture
of H2O and D2O) [13,14] is given by the Gross±
Butler eq (3).
4. Discussion
In aqueous solutions, sodium salts of aryl-N-
halosulfonamides ionize into several species in a
pH dependent manner [6±8]. In acidic solutions,
the oxidizing species of CAT are RNClH, RNCl2,
and HOCl, while in alkaline solution RNCl is the
active oxidant [6±10].
In alkaline solutions, sugars undergo isomeriza-
tion to an equilibrium mixture of aldoses and
ketoses which exist as enediol anions (E ) [11]. The
enolic-anions (E , keto-enolic anions of hexoses

K.S. Rangappa et al./Carbohydrate Research 307 (1998) 253±262
259
Q
TS
charged ions (eq (ii) requiring a very high activa-
tion energy. The observed high activation ener-
gies (Table 3) agree with this prediction. Nearly
constant ÁG# values (Table 3) suggest that a com-
mon mechanism is operative for the oxidation of
sugars.
ꢀ1 n ‡ nꢀi†
ꢀ1 n ‡ nꢀj†
knobs=koobs
ˆ
ꢀ3†
Q
RS
where ꢀi and ꢀj are isotopic fractionation factors
for the isotopically exchangeable hydrogen sites in
the transition state (TS) and reactant site (RS),
respectively. If the reaction proceeds through a
single transition state, then eq (3) becomes eq (4):
In a previous study on erythrose-series hexoses
[1], it was observed that each of the sugars studied
was oxidized by CAT predominantly to pentonic
and threonic acids. However, for the threose-series
hexoses studied here, the product pro®les can be
grouped into three categories, pro®les consisting
of: (a) predominant proportions of lyxonic and
threonic acid formed from lyxose-series sugars
(Fig. 2a), (b) threonic and glyceric acids as major
products formed from xylose-series hexoses
(Fig. 2b), and (c) high proportions of pentonic,
threonic, and glyceric acids formed from xylose
and lyxose (Fig. 2c). Based on these results and on
the observed isomerization of sugars in alkaline
solutions, the following reaction pathway (Scheme
2) can be suggested:
The observed reaction stoichiometry (Table 5)
agrees with the formation of mixtures of pentonic,
tetronic and trionic acids.
For lyxose-series hexoses, the major products are
formed by the loss of one or two carbon atoms
with the cleavage of C-1±C-2 and C-2±C-3 bonds,
respectively from the keto-enolic anion inter-
mediates. The predominance of lyxonic acid and
Á
knobs=kobs ˆ 1 ‡ nꢀꢀj 1†
ꢀ4†
A comparison of the plots of kobs versus n (not
shown) with the standard curves [15], suggested a
single proton exchange in the transition state. Fur-
thermore, the plots of knobs=kobs versus n (Fig. 1)
were linear with slopes ꢃꢀj 1†. The ꢀj for the oxi-
dation of sugars by CAT is about 0.6. This value
resembles the fractionation factor of HO ion.
Since the rate-determining step in Scheme 1
involves three negative ions, the reaction rate is
expected to increase with an increase in the ionic
strength (I) of reaction medium. The plots of log-
kobs versus I1/2 were linear with slopes between 0.5
and 0.85, even though the ionic strengths employed
were beyond the Debye±Huckel range. The theo-
retical slope of unity has not been realized, possibly
due to the formation of Bjerrum ion pairs in con-
centrated solutions [16].
The rate decreased with decrease in the dielectric
constant (D) of the reaction medium. The plots of
logkobs versus 1/D (D=dielectric constant) were
linear with negative slopes (Fig. 1). The e€ect of
solvent composition on the rate of a reaction
involving two negative ions is given by eq (5) [16]:
log k ˆ log ko ZAZBe2=DkTdAB
ꢀ5†
where ko is the rate constant in a medium of in®-
nite dielectric constant, ZAe, and ZBe are the char-
ges on ions, dAB is the size of the activated
complex, k is the Boltzmann constant, and T is the
absolute temperature. From the slopes of the
straight lines in Fig. 1 (slope= ZAZBe2/kT,dAB),
dAB was calculated. The derived dAB values for d-
galactose, l-sorbose, d-xylose, and d-lyxose are
2.70, 2.20, 2.02, and 2.40 A, respectively. These
values are comparable with those obtained for
similar reactions [16].
According to Scheme 1, the rate-determining
step should involve interactions between similarly
Scheme 2.

260
K.S. Rangappa et al./Carbohydrate Research 307 (1998) 253±262
minor amounts of xylonic acid from all lyxose-series
sugars suggest that, for these sugars, the cleavage of
the C-1±C-2 bond occurs without appreciable epi-
merization at C-3 (Fig. 2a). The formation of sig-
ni®cant amounts of hexonic acid from these sugars,
suggest that appreciable proportions of the sugars
react with CAT in the aldo-enolic forms, which up
on cleavage of the C-1±H bond form hexonic acids.
Minor amounts of pentonic acids are formed by
the cleavage of the C-1±C-2 bond from the aldo-
enolic anions of hexoses. The formation of hexonic
acid from tagatose is presumably due to iso-
merization to the aldo-enolic form.
The xylose-series hexoses, gulose, idose and sor-
bose, gave mainly threonic and glyceric acids and
minor proportions of pentonic acids (xylonic and
negligible amounts of lyxonic acid) (Fig. 2b). The
predominance of threonic and glyceric acids indi-
cate the preferential cleavage of C-2±C-3 and C-3±
C-4 bonds compared with C-1±C-2 bonds. Since
xylonic acid was formed in higher proportion
compared with lyxonic acid, the cleavage of the C-
1±C-2 bond in xylose-series hexoses must occur
without appreciable epimerization at C-3, as in the
case of lyxose-series hexoses (cf. Fig. 2b with
Fig. 2a). Formation of small amounts of hexonic
acids from gulose and idose was presumably due to
slow oxidation of the aldo-enolic form. Formation
of only a trace amount of hexonic acid from sor-
bose is due to negligible isomerization of sorbose
to gulose and idose.
In view of the foregoing considerations, a plau-
sible mechanism for the oxidation of sugars by
CAT is proposed in Scheme 3. This mechanism
accounts for the observed kinetics, reaction stoi-
chiometry, and products formed.
In the proposed mechanism (Scheme 3), the
anions (E ) of sugars react with CAT to form
intermediates X1±X3. For threose-series hexoses,
the anions (E ) intermediates are predominantly
the keto-enolic forms and minor proportions of
aldo-enolic forms. However, for pentoses, the
major reacting species are the aldo-enolic anions;
probably minor proportions of keto-isomer may
also be involved. In the case of anions (E ) from
hexoses, the loss of hydrogen can occur at either C-
1 or C-3 to form C-1±C-2 or C-2±C-3 enediols
containing a hypochlorite group at C-2. Since epi-
merization at C-3 was limited, as evidenced by the
formation of only very minor proportions of epi-
meric pentonic acids from hexoses, it can be con-
cluded that cleavage of the C-1±H bond occurs
preferentially as compared with cleavage of the C-
3±H bond to form C-1±C-2 enediols. The ene-diols
thus formed contain polarized double bonds to
which hydroxide ion can add at C-2 to form inter-
mediates X1 (major) and X2 (minor). X1 and X2
then can undergo cleavage of C±C bonds between
C-1 and C-2, the former giving lyxonic acid and the
latter forming a mixture of lyxonic and xylonic
acids.
In the case of aldo-enolic anions from pentoses,
hydrogen can be removed only from C-2 to form
the C-1±C-2 enediol-anion, which in the presence
of CAT and alkali forms intermediate X3 with
epimerization at C-2. The cleavage of C-1±H bonds
from X3 gives a mixture of lyxonic and xylonic
acids. The cleavage of C±C bonds between C-2 and
C-3 in X1 and X2, and the breaking of C±C bonds
between C-1 and C-2 in X3 yield aldo-tetrose
without epimerization at C-4 (hexoses) or at C-3
(pentoses). The aldo-tetrose further oxidizes to
yield threonic acid and a minor proportion of ery-
thronic acid (Table 5). The reaction can proceed
further, with the cleavage of C±C bonds between
C-3 and C-4 of hexoses and the breaking of C±C
bonds between C-2 and C-3 of pentoses, to form
glyceric acid. Minor proportions of threonic and
glyceric acids could also be formed by the cleavage
of C-1±C-2 and C-2±C-3 bonds, respectively, from
the keto-enolic form of pentoses through the reac-
tions sequences similar to those outlined for keto-
hexoses in Scheme 3.
In contrast to hexoses, xylose and lyxose gave
high proportions of pentonic acids (Fig. 2c).
Clearly, these major products are formed by the
cleavage of the C-1±H bond. Both xylose and lyx-
ose also gave high proportions of threonic and
glyceric acids, which are formed by cleavage of the
C-1±C-2 and C-2±C-3 bonds, respectively. Since
the latter type of bond-cleavage is facilitated
through the involvement of the keto-enolic form, it
is possible that portions of pentoses react in the
keto-enolic form. However, lyxose and xylose were
not isomerized to xylulose to a signi®cant level.
Together, these data suggest that pentoses undergo
oxidation by CAT mainly through aldo-enolic
intermediates, and only minor proportions may be
oxidized via the keto-enolic form. In accordance
with the observed three-times faster reaction rate
of xylose compared with lyxose, a large amount of
lyxose was not oxidized by CAT, even after 24 h at
ꢀ
35 C (Fig. 2c). Under similar conditions, xylose
was almost completely oxidized.

K.S. Rangappa et al./Carbohydrate Research 307 (1998) 253±262
261
Scheme 3.

262
K.S. Rangappa et al./Carbohydrate Research 307 (1998) 253±262
[6] J.C. Morris, J.R. Salazar, and M.A. Winemann, J.
Acknowledgements
Am. Chem. Soc., 70 (1948) 2036±2041.
[7] E. Bishop and V.J. Jennings, Talanta, 1 (1958)
197±212.
[8] F.F. Hardy and J.P. Johnston, J. Chem. Soc. Per-
kin Trans II, (1973) 642±647.
[9] T. Higuchi and A. Hussain, J. Chem. Soc., (1967)
549±553.
[10] D.S. Mahadevappa, K.S. Rangappa, N.M.M.
Gowda, and B.T. Gowda, J. Phys. Chem., 85
(1981) 3651±3658.
The proposed mechanism for the reaction is
based on suggestions given by Professor Derek
Horton for the oxidation of erythrose-series sugars
by CAT [1], and we thank Professor Horton for his
help. H.M and M.P.R. are grateful to the Uni-
versity of Mysore for the University Grants Com-
mission, India, research fellowships. K.S.R. thanks
the CSIR, India, for the ®nancial support.
[11] W. Pigman and E.F.L.J. Anet, Mutarotations and
Actions of Acids and Bases, in W. Pigman, and D.
Horton (Eds.), The Carbohydrates: Chemistry and
Biochemistry, 2nd ed., Vol. 1A, Academic Press,
New York, 1972, pp 165±194.
References
[1] K.S. Rangappa, M.P. Raghavendra, D.S. Maha-
devappa, and D. Channe Gowda, Carbohydr. Res.,
306 (1998) 57±67.
[12] C.J. Collins and N.S. Bowman, Isotope E€ects in
Chemical Reactions, Van-Nostrand±Reinhold,
New York, 1970, p 267.
[2] T.A. Iyengar, Puttaswamy, and D.S. Mahade-
vappa, Carbohydr. Res., 197 (1990) 119±130.
[3] T.A. Iyengar, Puttaswamy, and D.S. Mahade-
vappa, Carbohydr. Res., 204 (1990) 197±206.
[4] T.A. Iyengar and D.S. Mahadevappa, J. Carbo-
hydr. Chem., 11 (1992) 37±58.
[13] W.J. Albery and M.H. Davies, J. Chem. Soc.
Faraday Trans., 68 (1972) 167±181.
[14] G. Gopalakrishnan, and J.L. Hogg, J. Org. Chem.,
50 (1985) 1206±1212.
[15] N.S. Isaacs, Physical Organic Chemistry, Long-
man, New York, 1987, p 276.
[5] M.R. Hardy and R.R. Townsend, Methods Enzy-
mol., 230 (1994) 208±225.
[16] K.J. Laidler, Chemical Kinetics, Tata McGraw±
Hill, Bombay, 1965, p 214.